AEAT-1981

Protocol for the Testing of Bioremed-ation Products

A report produced for MAFF and NER Inc

Richard PJ Swannell DavidJ Mitchell Anthony J Flemming

Report No AEAT-1981

July 1997

-jETCEN the rndmg Ilm 01 he ~n~ur13l EnvlToJ1m~maI TeLhnalogmiddot LtnUt

NETCEN - lied bEA T~chll Igy pic Na[lOnal EnYlmnmentJI Technology Cemr(

AEAT-1981

Jbull Protocol for the Testing of Bioremediation Products

A report produced for MAFF and NER Inc

Richard PJ Swannell David] Mitchell Anthony] Flemming

Report No AEAT-1981

--

July 1997

-------------------

AEAT-1981

Title Protocol for the Testing ofBioremediation Products

Customer IMAFF and NER Inc

Customer reference I OD 13

Confidentiality copyright and reproduction

File reference IHDOCUMENTBIOTESTTESTREP3DOC I Reference number IAEAT-1981

Report Manager

This document has been prepared by ABA Technology pIc in connection with a contract to supply goods andor services and is submitted onI on the basis ofstrict confidentiality The contents must not be disclosed to third parties other than in accordance with the terms of the contract shy

-Richard PJ Swannell ABA Technology pIc National Environmental Technology Centre Culham Abingdon OX14 3DB Oxfordshire Telephone +44 (0)1235463974 Facsimile +4 (0)1235463030

shy

AEA Technology is the trading name of AEA Technology pIc

AEA Technology is certified to I 09001

Name IRichard PJ Swannell

Approved by ame

Signatme

Date

AEA Technology ii

AEAT-19S1

J Contents ~

1 Introduction 1

2 Flask Test 2 -

21 OVERVIEW OF TEST 2 22 OIL 2 23 ARTIFICIAL SEAWATER 2 24 STANDARD MICROBIAL INOCULUM 2 2S STANDARD NUTRIENT PREPARATION 3 26 STERILISATION PROCEDURE 3 27 TEST DESIGN AND OPERATION 3 28 CHEMICAL ANALYSIS 4

281 Introduction 4 282 Materials S

J 283 Methods 6 2831 Extraction 6 2832 Fractionation 7

284 Quantification 7 2841 GC Quantification 7 2842 GC-MS Quantification 7 2843 Correcting of Recovery data 8

29 ANALYSIS OF RESULTS 8

3 Microcosm Test 9

31 OVERVIEW OF TEST 9 32 MICROCOSM DESIGN 9 33 OIL 11 34 ARTIFICIAL SEAWATER 11 35 STANDARD NUTRIENT PREPARATION 11

36 TEST DESIGN AND OPERATION 11 37 MICROBIOLOGICAL ANALYSIS 12

371 Microbial Activity 12 372 Enumeration of Hydrocarbon-Degraders 13

38 CHEMICAL ANALYSIS 13 39 ANALYSIS OF RESULTS 13

AEA TKhnology iii

AEAT-1981 cshy

shy

4 Reporting of Results 14 ~ Ii

5 References 15 ~bullshybullr

-bull

bull

I

bullrbull bull bull -

~ bull bull

1

bull

bull

AEA TKhnology iv

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

Jbull Protocol for the Testing of Bioremediation Products

A report produced for MAFF and NER Inc

Richard PJ Swannell David] Mitchell Anthony] Flemming

Report No AEAT-1981

--

July 1997

-------------------

AEAT-1981

Title Protocol for the Testing ofBioremediation Products

Customer IMAFF and NER Inc

Customer reference I OD 13

Confidentiality copyright and reproduction

File reference IHDOCUMENTBIOTESTTESTREP3DOC I Reference number IAEAT-1981

Report Manager

This document has been prepared by ABA Technology pIc in connection with a contract to supply goods andor services and is submitted onI on the basis ofstrict confidentiality The contents must not be disclosed to third parties other than in accordance with the terms of the contract shy

-Richard PJ Swannell ABA Technology pIc National Environmental Technology Centre Culham Abingdon OX14 3DB Oxfordshire Telephone +44 (0)1235463974 Facsimile +4 (0)1235463030

shy

AEA Technology is the trading name of AEA Technology pIc

AEA Technology is certified to I 09001

Name IRichard PJ Swannell

Approved by ame

Signatme

Date

AEA Technology ii

AEAT-19S1

J Contents ~

1 Introduction 1

2 Flask Test 2 -

21 OVERVIEW OF TEST 2 22 OIL 2 23 ARTIFICIAL SEAWATER 2 24 STANDARD MICROBIAL INOCULUM 2 2S STANDARD NUTRIENT PREPARATION 3 26 STERILISATION PROCEDURE 3 27 TEST DESIGN AND OPERATION 3 28 CHEMICAL ANALYSIS 4

281 Introduction 4 282 Materials S

J 283 Methods 6 2831 Extraction 6 2832 Fractionation 7

284 Quantification 7 2841 GC Quantification 7 2842 GC-MS Quantification 7 2843 Correcting of Recovery data 8

29 ANALYSIS OF RESULTS 8

3 Microcosm Test 9

31 OVERVIEW OF TEST 9 32 MICROCOSM DESIGN 9 33 OIL 11 34 ARTIFICIAL SEAWATER 11 35 STANDARD NUTRIENT PREPARATION 11

36 TEST DESIGN AND OPERATION 11 37 MICROBIOLOGICAL ANALYSIS 12

371 Microbial Activity 12 372 Enumeration of Hydrocarbon-Degraders 13

38 CHEMICAL ANALYSIS 13 39 ANALYSIS OF RESULTS 13

AEA TKhnology iii

AEAT-1981 cshy

shy

4 Reporting of Results 14 ~ Ii

5 References 15 ~bullshybullr

-bull

bull

I

bullrbull bull bull -

~ bull bull

1

bull

bull

AEA TKhnology iv

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

-------------------

AEAT-1981

Title Protocol for the Testing ofBioremediation Products

Customer IMAFF and NER Inc

Customer reference I OD 13

Confidentiality copyright and reproduction

File reference IHDOCUMENTBIOTESTTESTREP3DOC I Reference number IAEAT-1981

Report Manager

This document has been prepared by ABA Technology pIc in connection with a contract to supply goods andor services and is submitted onI on the basis ofstrict confidentiality The contents must not be disclosed to third parties other than in accordance with the terms of the contract shy

-Richard PJ Swannell ABA Technology pIc National Environmental Technology Centre Culham Abingdon OX14 3DB Oxfordshire Telephone +44 (0)1235463974 Facsimile +4 (0)1235463030

shy

AEA Technology is the trading name of AEA Technology pIc

AEA Technology is certified to I 09001

Name IRichard PJ Swannell

Approved by ame

Signatme

Date

AEA Technology ii

AEAT-19S1

J Contents ~

1 Introduction 1

2 Flask Test 2 -

21 OVERVIEW OF TEST 2 22 OIL 2 23 ARTIFICIAL SEAWATER 2 24 STANDARD MICROBIAL INOCULUM 2 2S STANDARD NUTRIENT PREPARATION 3 26 STERILISATION PROCEDURE 3 27 TEST DESIGN AND OPERATION 3 28 CHEMICAL ANALYSIS 4

281 Introduction 4 282 Materials S

J 283 Methods 6 2831 Extraction 6 2832 Fractionation 7

284 Quantification 7 2841 GC Quantification 7 2842 GC-MS Quantification 7 2843 Correcting of Recovery data 8

29 ANALYSIS OF RESULTS 8

3 Microcosm Test 9

31 OVERVIEW OF TEST 9 32 MICROCOSM DESIGN 9 33 OIL 11 34 ARTIFICIAL SEAWATER 11 35 STANDARD NUTRIENT PREPARATION 11

36 TEST DESIGN AND OPERATION 11 37 MICROBIOLOGICAL ANALYSIS 12

371 Microbial Activity 12 372 Enumeration of Hydrocarbon-Degraders 13

38 CHEMICAL ANALYSIS 13 39 ANALYSIS OF RESULTS 13

AEA TKhnology iii

AEAT-1981 cshy

shy

4 Reporting of Results 14 ~ Ii

5 References 15 ~bullshybullr

-bull

bull

I

bullrbull bull bull -

~ bull bull

1

bull

bull

AEA TKhnology iv

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-19S1

J Contents ~

1 Introduction 1

2 Flask Test 2 -

21 OVERVIEW OF TEST 2 22 OIL 2 23 ARTIFICIAL SEAWATER 2 24 STANDARD MICROBIAL INOCULUM 2 2S STANDARD NUTRIENT PREPARATION 3 26 STERILISATION PROCEDURE 3 27 TEST DESIGN AND OPERATION 3 28 CHEMICAL ANALYSIS 4

281 Introduction 4 282 Materials S

J 283 Methods 6 2831 Extraction 6 2832 Fractionation 7

284 Quantification 7 2841 GC Quantification 7 2842 GC-MS Quantification 7 2843 Correcting of Recovery data 8

29 ANALYSIS OF RESULTS 8

3 Microcosm Test 9

31 OVERVIEW OF TEST 9 32 MICROCOSM DESIGN 9 33 OIL 11 34 ARTIFICIAL SEAWATER 11 35 STANDARD NUTRIENT PREPARATION 11

36 TEST DESIGN AND OPERATION 11 37 MICROBIOLOGICAL ANALYSIS 12

371 Microbial Activity 12 372 Enumeration of Hydrocarbon-Degraders 13

38 CHEMICAL ANALYSIS 13 39 ANALYSIS OF RESULTS 13

AEA TKhnology iii

AEAT-1981 cshy

shy

4 Reporting of Results 14 ~ Ii

5 References 15 ~bullshybullr

-bull

bull

I

bullrbull bull bull -

~ bull bull

1

bull

bull

AEA TKhnology iv

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981 cshy

shy

4 Reporting of Results 14 ~ Ii

5 References 15 ~bullshybullr

-bull

bull

I

bullrbull bull bull -

~ bull bull

1

bull

bull

AEA TKhnology iv

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

1

AEAT-1981

Introduction

This protocol is a test for assessing the efficacy ofbioremediation produers designed to treat oil on contaminated shorelines For the purposes orchis test a bioremediation product is defined as that which can stimulate the biodegradation or co-metabolism ofcontaminants (see Swannell et al 1996a) The target contaminant in this protocol is weathered crude oil

The protocol consists ofcwo experimental phases

1 A flask test designed to detennine whether the product can stimulate the metabolism of oil under optimal conditions This test is based on the NETAC Tier II test (NETAC 1993)used by the USEPA

2 A microcosm test designed to cletennine whether the product can stimulate the metabolism of oil in a simulated beach enviromnent This test uses microcosms designed by AEA Teclmology to simulate the marine environment (Croft et aI 1995)

Ifno evidence ofenhanced oil metabolism is noted in the flask experiment then the product will be deemed to have failed the test and it will not be tested further A product will only be deemed effective if it significantly (pltOOS) enhances oil biodegradation above that recorded in untreated controls under the conditions ofeach test A microcosm test should then be conducted to assess efficacy under simulated marine conditions The product passes the test ifit significantly (pltOOS) enhances oil biodegradation and substantially reduces oil concentrations in comparison to negative controls

rf a product is found to pass the efficacy test as described herein this does not mean that it will work successfully for any oil type on any shoreline type at any time of the year The aim ofthe efficacy test is merely to establish whether the product can stimulate oil biodegradation under controlled conditions and therefore can be justifiably used as a bioremediation product in the shoreline envirorunent

AEA Technology

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

2 Flask Test -

21 OVERVIEW OF TEST

The aim of the test is to determine whether a bioremediation product can stimubte the biodegradation ofoil under optimal conditions This is done by determining oil biodegradation in Erleruneyer flasks created with product in comparison to untreated sterile controls Produces shycontaining nutrients alone are tested in flasks containing a standard oil-biodegrading microbial inocula A positive control containing a standard inocula of oil-biodegrading bacteria and a sranwd unoont ofnutrient is also conducted The positive control biodegrades oil efficiently and is used to confirm that the tcst is operating satisfactorily All flasks are treated with an

identical quantity ofoil and all are conducted in a standard artificial SC2water or marine shyBushnell-Haas medium Bushnell-Hus medium is specially designed to stimulate microbial hydrocarbon biodeldation (Brown amp Braddock 1990)

22 QIL -LWeathered Forties crude oil is used for the (cst procedureas it is biodegradable and has been

spilled in UK coastal waters The oil is weathered to remove vobtile compoWlds which rapidly eV2porate after a spill at sea or on shorelines This involves distilling the oil to a V2pOur temperature of250degC in a fume hood using seandard distillation apparatus heated by an electrical mantle This results in tbe oil being weathered to the same degree as it would over a

24 to 48 hour period at sea The mass ofthe oil is recorded prior to distillation and on completion once the oil has cooled to ambient temperature

The precise composition of Forties is variable however since the test is based on comparison between the concomitantly nul product and concrolflasks this is unimportant as long as the same batch ofoil is used for the product being tested and the controls

23 ARTIFICIAL SEAWATER

Arti6cial seawater is made by preparing [meant Ocean (Aquarium Systems Inc) or an equivalent in accordance with the manufacturers instructions

24 STANDARD MICROBIAL INOCULUM

The standard microbial inocula consists of six hydrocarbon-degrading micro-organisms

I Mi lu NClME 13267 2 Pseudomonasfluoresans NCIMB 9046 3 Pseudomonas pulida NCIMB 9571

bull4 Pseudomonas nautica 5 Alteromonas sp bull

AEA hchnology 2

bull

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

I

1

l

bull

AEAT-1981

6 Plarwcoaus akarwdastus

All the above will he fully classified and lodgedat the National Collection ofIndustrial and Marine Bacteria (NClMB)

All six micro-organisms grow in the presence of marine concentrations ofsak The latter three micro-organisms were isolated from the intertidal zone ofStert Flats (Somerset UK) All the organisms are storedin glycerol at _25degC (Gerhardt et al 1981) Before each experiment the organisms are removed from the freezer and allowed to defrost at room temperature They are then separatdy added aseptically to 100 m1 of sterile Ill2Iine broth in 250 mJ Erlenmeyer flasks and incubated at 2011degC for 4 days to revive the bacteria An aliquot volume (1 ml) of each micro-organism is then trmsferred separately co fresh sterile marine broth under sterile conditions and incubated again at 2011 degC for 4 days Agar plates are inoCllhted with samples from each of these flasks to monitor the-colony colour and morphology and to check for the presence ofconuminants Ifcontamination is found chen a fresh culture is removed from the freezer and the above process repeated Before use the number ofbaeteria contained within the culture is established by measuring the optical density of the culture This is then converted to a microbial number in Colony Forming Units (CFU) per ml using a previously deterrniJied calibration cwve Each culture is grown CO an optical density of1 at 570 run prior to use

25 STANDARD NUTRIENT PREPARATION

The positive control (scmdard microbial preparation oil and standard nutrients) and the negative control (standard nutrient oil and filter-sterilised microbial preparation) are prepared in sterilised tap water mixed with marine Bushnell Haas medium This mediwn is well known to provide all the required nutrients for oil biodegradation (Brown amp Braddock 1990)

26 STERILISATION PROCEDURE

Sterilisation is achieved either by autoclaving at 121degC and 15 psi for 30 min Or by passage through a sterile 02 JLrn filter into a sterile container under aseptic conditions The success of autoclaving is confirmed using autoclave tape and Browns tubes St~tion will only be accepted if the strips on the autoclave tape change colour and the Browns tubes change from red to green All aseptic procedures are conducted in a laminar flow hood All surfaces are cleaned with 70 ethanol before and mer use and they are cleansed regularly with an antiseptic cleaner (Vircon or equivalent)

27 TEST DESIGN AND OPERATION

The experiment is conducted in sterile 250 ml Erlenmeyer flasks at 2oplusmnldegC in a rotary shaker at 200 rpm (plusmn20 rpm) The flasks are plugged with sterile foam bungs to allow free passage of filtered air and the shaker speed is selected to maintain aerobic conditions during the test The flasks are placed in the middle of the sh2ker plate as placing them at the edge can cause oil contamination of the foam bung Temperature is monitored throughout the experiment to

AEA Technology )

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

confirm it is maintained within the limits mentioned above The experiment is conducted in f

triplicate and the addition to each flask is given in Table 1

The artificial sea~ter used in the test is filter-sterilised as autoclaving causes a precipitate to form The oil is added using a graduated glass syringe and the density of the oil is meltlSured in order to calculate the volume to be added to each flask Details of the amount of biorememation product to be added to each flask are supplied by the manufacrurer

The test is conducted for 28 days At the end afthe experiment the oil is extracted from each flask and analysed for total oil content The change in composition ofche oil is also detennined using the technique outlined below

Table 1 Design ofFlask Experiment

Flask Aqueous Phase Oil Addition Inoculum Conunents Nuntbers

123 Marine 05 g 1 ml of each ofthe Negative Control f

Bushnell-Haas 6 standard microshy To determine the abiotic medium orgarusms fate of oil under test

Filter sterilised conditions

456 Marine Bushnell-Haas

05 g 1 ml ofeach ofthe 6 standard microshy

Positive Control To confinn that oil is

medium orgamsms being biodegraded under the test conditions bullI

~

789 AJtificill Seawater

05 g 1 ml ofeach of the 6 standard micro-

Test ofa fertiliser product

+ test fertiliser ~~~

bull101112 Artificial 05 g Product (Microshy Test of a product

Seawater otganisms and containing nutrients and fertiliser) bacteria

28 CHEMICAL ANALYSIS

281 Introduction

In order to confinn the efficacy of the test product the composition of the test oil is studied in detail at the end of the experiment The most effective method of detemrining the degree to which oil has been biodegraded is to calculate the ratio ofselected petroleum compounds against an oil component know to be resist biodegradation (a biomarker) The C3Q hopane 17a(H) 21f3(H)-hopane was selected as the conserved biomarker for the test because of its abw1Ciance in the test oil and because it has not been found in significant quantities in sediments

r AEA Technology 4

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

1

AEAT-1981

uncontaminated by pecrolewn hydrocarbons Hopanes have a low water solubility andJ volatility and tend to remain with the oil while other petroleum components are lost by other

processes such as evaponltion dissolution and biodegradation (Btlgget aI 1994) Thus changes in the relative concentration of the hopanes provide a quantitative assessment of these1 weathering processes

Although evaporation and biodegradation may occur in each test vessel evaporation will be reduced because of the prior weathering of the oil (Swannell et of 1994a) Hence by contrasting the ratios ofhopane to more biodegradable petroleum compounds in the negative control with those in the treated sediments the effect of bioremediation products can be detennined

Oiled samples are extracted fractionated and individual compounds quantified by the method described below The compounds that are specifically quantified ate given in Table 3 Quality

assurance procedures are in place to ensure that all samples are traceable All samples are stored

at _20degC for 12 months after analysis should the measurement require confirmation

282 Materials

The solvents used are dichloromethane methanol and light petroleum spirit (petroleum ether BP 40-60degC) all of which are Distol grade Granular sodi1Ull sulphate is analytical grade silica gel 60 is chromatographic grade with a particle size of02-05 nun alumina is Brockmann grade 1 for column chromatography all are extracted with dichloromethane (DCM) in the

soxhlet apparatus prior to use and stored at 120degC

All glassware is soaked for 24 hours in a 5 Decon solution rinsed in distilled water and oven dried Prior to use glassware is rinsed with oeM In the extraction procedure stainless steel centrifuge tubes (100 ml) and pear shaped separating funnels with a 1 litre capacity are used

Both internal and surrogate standards are used to quantify oil recovery and the response factors for the target hydrocarbons The I11dSS spectrometer is tuned in accordance with EPA Method 8270 (US EPA 1992)

-

AEA Technology 5

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

Table 3 List ofCompounds Quantified

ALIPHATIC COMPOUNDS

0-lt11

n_el2

-Cil

0-lt1lt

0-lt

0-lt16

-C17 0-lt18 shyn-C19

--= 0-01

--= n-C23

0-0 -as 0-06 bull (Z7

-C2ll

-09

-00

-CJ1

_CJ2

n-03

-CW n_OS

17a(H21 PH)-Hopanc

283 Methods

2831 Extraction

AROMATIC COMPOUNDS iI NoptIWcnc (JaIl 128) shy2-~(m112)

1~ 142)

26-n74M~(JaI1 156)

1J-117~(hL56)

16-diMhl_r lS6)

l5-Jl~bdWonpound(-II 156)

1S-l14~(mIII56) shy124Maup (II 156)

1J1-uiMcNopldamp (0amp11170)

1J~(mil 170)

13S-146-criMdUpbda (mi 170)

236-criMcNapbdl (mil 170) shy1)7IA7(I~ lt-11170) bull 124-oiMeNpldL (1QI 170)

12S-oiMcNpbU (-11170)

Igtibcmoxhiapho- (ala 184)

4-M~opbeno(mil 198)

2-I3-M~ophav(rAJ 198)

I-Mcdibmrotbiopbcnc ~h 198)

DiM~opbcn_bullbull 212)

P~(=I1171)

3-McpbcNnthronc (mill 192)

2-Mcp~(lnI1 192) bull 9-Mcpbnuntlvcne (mlr 192)

I-Mcpbcrumhronc (mil 192)

DiMcpbcnaNhnna(mtr 206)

Chrylcnc (mlr 228)

PayJene (m1z252) bull00 Me Triarommc Iteroid (mh 231)

C2 Me T riuouatic Ilcroid (mh 231) ~ C26 20S MOl TrUrouuuc Itid (avID1)

C2620RI 07 20S Me TriaroowiencZOid (mh23l) C28 20S Me TriarollUluc IIcroid (mit nl) C27 lOR Me TrWomatic lIeroW (mlamp231)

C28 201 Me T rUromatic JtcrOid (mlamp 231)

bull

After removal from the freezer and thawing the sample is mixed well and the full flask contents are placed into a centrifuge tube together wich an aliquot ofche surrogate st3ndard solution A further aliquoc ofche surrogate standard is removed Rored for analysis wich che sample fractions

AEA Technology 6

bull

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AfAT-1981

Methanol (40 ml) is added to the sample in separating funnel The supernatant liquid is decanted into ltI second separating fiumd containing 200 m1 ofdistilled water The sample is further extrlcted and centrifuged with methanoldichloromethane (11 40 ml) and the liquid decanted off into the separating funnel the procedure is then carried out twice more with 40 ml portions ofdich1oromedwte The flask is rinsed with dich1oromethane three times to remove oil attached to the glass The DeM is added to the second separating funnel

The sepanting funnel is vigorolll1y shaken allowed to settle for approximately one hour and the organic layer Oower) removed into a rOWld bottom flask The aqueous layer is then extncted three more times with DeM (50 ml) each time allowing the layen to separate before removing the organic layer in to the round bottom flask The extract is rorary evaporlted to CQ

S ml before anhydrous sodium sulphate is added to remove any water After rinsing the round bottom flask the extract is accurately made up to a knovvn volume and a knovvn aliquot is transferred to a pre-weighed round bottom flask and rotary evaporated to near dryness After any traces ofsolvent are removed using a stream ofdry nitrogen it is allowed to dry overnight (approx 40degC) to constant weight in order to detennine the Extract2ble Organic Matter (EOM)

2832 Fractionation

An aliquot cf the remaining EOM is adsorbed onto alUInina using a stream ofdry nitrogen to remove tbe solvent Samples of the added Forties crude oil are similarly adsorbed onto alumina for fractionation Liquid chromatographic sepantion colurtUlS are made with activated silica and activated alumina The extract is trmsferred to the top of the column and the saturated hydrocarbons are eluted with light petroleum (100 ml) and the aromatic hydrocarbons duted with DCMlight petrolewn (100 mI) with botb eluates collected together in the same round bottom flask to give the total petroleum hydrocarbon ([PH) fraction Excess solvent is removed from this fraction by rotary evaporation An aliquot of this TPH fraction is placed in a auotsampler vial for gas chromatography (GC) and a further aliquot transferred to a vial for gas chromatography-mass spectrometry (GC-MS) To the TPH aliquot for GC an internal standard solution is added To the TPH aliquots for GC-MS an internal standard solution is also added GC and GC-MS are conducted under standard conditions appropriate for oil samples

284 Quantification

2841 GC Quantification

For the TPH fractions the Tocal GC area is measured The Resolved GC area is the quantified peak areas reponed by the GC data system The Blank area refen to the area for a DCM bLmk run The T ota Resolved and Unresolved GC areas are calculated rdative to EOM The nshyalhnes (ClI to ~ Pristane and Phytane were quantified agrinst the internal standards

2842 GC-MS Quantification

The Steranes (mz 217) and T riterpanes (mz 191) are quantified as for the GC data using the appropriate internal stand2rd The aromatics are quantified in the same way

AEA Technology 7

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-19S1

2843 Correcting of Recovery data

All values for individual compoWlds are corrected for losses during extraction and fractionation and the triterpanes are further corrected for their response on the GC-MS The surrogate -standards were analysed on GC and quantified against the internal standards

f 29 ANALYSIS OF RESULTS

The success of bioremediation products is assessed by comparing sutistically the l1USS ofoil recovered from the test flasks with that recovered from the negative controls and through analysis of the oil chemistry data The oil chemistry analysis yidds 4 measures ofoil biodegradation EOMhopane Total GC detectable aliphaticshopane Total GC resolvable aliphatics hopane and Total GC resolvable aronuQcs hopane For the purposes of the approval procedure only the latter two measurements are used as pan ofthe passfail criteru Both these measurements are detennined by reference to appropriate analytical standards (see Section 38) and are therefore prone to a minimum oferror EOM me2Surements can be influenced by non-petroleum sources of organic carbon which I1llly be present in some produces The measurement ofTotal GC detectable aliphatics requires the accurate quantification ofthe wrresolved complex mixture (UCM) seen on gas chromatographs ofpetroleum hydrocarbons which may also be prone to error This additional information will be reported as it may be of value to oil spill responders and may help them distinguish between similar products However the ratios ofEOMlhopane and Total GC detectable Aliphaticslhopane are not analysed statistically and not included 2S part of the passfail criteria

The ratios ofTotal GC resolVdble aromatiahopane and Toul GC detectable aliphaticslhopane are analysed statistically The variance of the data from each microcosm is r detennined and tbe largest and smallest variance from the test data are compared to check whether there is a significant difference using the Fuwc test (parker 1981) If there is a significant difference (pltOOS) then a non-parametric test is employed to detennine statistical differences bull (ie Kruskal-Wallis test) If the variances are not fOtuld to be statistically different (pgtOOS) then k the statistical analysis is conducted using a one-way Analysis of Variance (Sakal amp Rolfe 1981) Differences between treatments are identified using the DutU1ett procedure (Montgomery rbull 1991) which compares many treatments with data from the negative control

The test is only regarded as valid if the positive control is found CO be scatisrially different (pltOOS) from the negttive concrol If no significant difference is found between these results then the experiment must be regarded as being invalid and should be repeated

A product passes the flask test if the oil levels in me test flasks are significantly lower (pltOOS) than those in the negative control and if the oil in the test flask is significantly more biodegraded (pltOOS) than that recorded in the negative control Ifa product passes the flask test it is then tested in beach microcosms (Section 3) 1

I

r AEA Technology 8

bull

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

3 Microcosm Test

31 OVERVIEW OF TEST

The aim of the experiment is to detemUne whether a bioremediation product em stimulate me biodegradation of oil in beach microcosms (Figure 1) The most important features aCme microcosm design are that the sediment is subjected to tidal cycle characteristic of the intertidal zone of3 natllIdl beach and char fresh artificial seawater is used for each tide This tests the ability of the bioremediation product to resist tidal removal and persist within the beach

The ability ofbioremediation products to stimulate oil biodegradation will be studied in comparison to an untreated oiled microcosm (negative control) and an oiled microcosm treated wich 3 product known to stimulate oil biodegradation (positive control) The product will pass the test if it stimulates oil biodegradation to a significantly greater extent (pltOOS) chan that recorded in the negative control over me course aCthe experiment and results in 3 substantial decre2Sc in oil concenmtion Oil biodegradation is assessed by studying changes in the ratios of biodegradable petroleum components to the recalcitrant biomarker 17a 21f-hopane (Bragg ~t

aI 1994) This latter compound is not thought to be present in uncontaminated sediment and is know to be broken down only slowly in natural sediments Oones et aI 1986 Ries-Kautt amp

Albrecht 1989 Bragg et aI J 1994) This technique is capable ofdistinguishing oil biodegradation from processes which merdy enhance oil removal from the microcosms (Bragg et al 1994 Swannell et al 1995 Swannell et al 1997)The oil chemistry is analysed wing the same methodology as chat wed for the flask experiment However changes in microbial number and activity are aho measured to confinn that the bioremediation product is stimulating hydrocarbon-degrading micro-orgmisms and promoting microbial decomposition processes Stimulation of oil biodegradation in the positive control above that recorded in the negative control serves as confirnution that the test is operating satisfactorily

32 MICROCOSM DESIGN

Beach sediment is collected by coring from the inter-tidal zone and from the aerobic zone ofa beach The sediment should be a fine sand according to a stmclard classification (Klute 1986) containing 3 mud and with a mean total orgmic matter content of 06 It must contain at least 10

3 hydrocarbon - degrading bacteria per grun ofsediment and should be free of

hydrocarbon contamination Such a sediInent can be obtained at Sten FIltlts (Somerset UK) On arrival at the laboratory the cores are sieved washed with arti6cial sea water and placed in 3 columns (glass 032 m high 01 m internal diameter containing approximately 13 kg ofsand (Figure 1raquo These columns are equilibrated for 7 days (ie at le2St 14 tidal cycles) before the addition of oil Peristaltic pumps arc wed to simulate a tidal cycle Fresh artificial seawater is pumped into each core at a nuximum bulk flow velocity of 033 mh-I The seawater is retained for 50 h and is then pumped from C2cb microcosm into plastic 20 litre bottles for storage prior to disposal No seawater is recycled to the microcosms The tidal cycle is repeated every 12 h ensuring that each core is covered with fresh seawater for 10 h in every 24 h (the filling and emptying ofeach microcosm takes approximately 2 hday~I) The tests are conducted in the dark except when sampling

AEA Technology 9

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

1

-14 em

I-4 em

-14 em

10 em ID

AEAT-1981

Figure 1 Design ofLaboratory Microcosms

Tofrom C02 meter ~

i-C==J ~===H-1======------- Seawater Outlet

~i Non-return Valve Taps

AEA Technology 11

32 em

Seawater Inlet TIdal Volume -660 ml In 5 min

-

Overflow

-r1~ -

Headspaee -1400 ml

Water (at high tide)

Crude Oil (75 (vv) in seawater)

Sediment Total Volume -1400 ml Pore Space -100 ml

Sintered Glass Support

Plastic Diffusion Rings

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981 bull

33 OIL

Weathered Forties erode oil is used for the test procedure This UK crude oil was selected because it is biodegradable and it has been spilled in UK coastal waters The oil is weathered to remove volatile compounds which rapidly evaporate after a spill at sea or on shorelines (see Section 22)

34 ARTIFICIAL SEAWATER

Artificial seawater is made by preparing Instant Ocean (Aquarium Systems mc) or an equivalent in accordance with the manufacturers instructions

35 STANDARD NUTRIENT PREPARATION

In order to confinn that the test is operating successfully a positive control is incorporated as part ofche test design This consists ofadding a treatment which is known to promote the biodegradation of oil under the conditions ofche test Previous research has shown that regular addition ofsodiwn nitrate and potlSSiwn dihydrogen orthophosphate promote oil biodegradation both in laboratory mesocosms and in the field (Swannell et al 1995 Venosa et a1 1996) Hence the positive control for the microcosm test consists of the weekly addition of sodium nitrate (305 g) and potassium dihydrogen orthophosphate (020 g fertiliser grade) dissolved in 50 ml of tap water The negative control is treated at the same time with 50 ml water Three additions of the standard nutrient preparation are made over the course of the experiment There is no need to add the standard microbial inoculum to the positive control as there is an active hydrocarbon-degrading microbial population in Stert beach sediment (as there is in most shoreline environments (Swannell et al 1996b)) The presence ofa hydrocarbonshydegrading population is confinned by enumerating the population capable ofgrowth on Forties crude oil in all microcosms at the beginning of the experiment and periodically as the experiment proceeds (Section 37)

36 TEST DESIGN AND OPERATION

After the sediment cores have equilibrated in the microcosms (Section 32) oil is added to each core at 37 kgm-2 Previous experiments have demonstrated that this application ofoil will penetrate rapidly into the beach sediment simulating concentrations that could be achieved after a spill incident (Croft et al 1995) The oil is applied at high tide just before the ebb tidal cycle begins The ebb tide draws the oil evenly into the surface of the beach sediment as the seawater is pumped out of the microcosms The microcosms are left for two tidal cycles (24 h) to allow equilibration with the oil before the bioremediation treatments are applied (Table 2) The test bioremediation product(s) are applied to microcosms exactly as directed by the manufacturer Ifthe product is a fertiliser then its ability to stimulate the indigenous hydrocarbon-degrading population is assessed No additional micro-organisms are added As such a fertiliser product will be tested in an identical manner as that described for the positive concrol The treatments are allocated to the microcosms at random The experiment is conducted for 28 days

AEA Technology 11

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981

Table 2 Design ofMicrocosm Experitnent

Treatment otrient Addition

Inoculum Connnents

egative Control

50 ml water on

one To do determine the background biodegradation rate ofoil under test

conditions

Positive Standard None To confirm that a standard Control Nutrient bioremediation treatment causes a fl

u Addition significant stimulation of oil

(Section 35) biodegradation above that noted in the control

Test As directed by As directed by Test ofbioremediation product manuampcrurer manufacturer

~___

37 M CROBIOLOGICA A A YSIS rshy

371 Microbial Activity

As indigenous micro-organisms biodegrade oil they produce carbon dioxide water and some breakdown products such as fatty acids and catechols (prince 1993 Swannell et al 1994a amp b Swannell et ai 1995) Therefore monitoring carbon dioxide evolution indicates whether any treatment is stimulating microbial decomposition processes Ifthe bioremediation treatment is inorganic then any stimulation of CO2 evolution is likely to be the result ofoil biodegradation Thus in the positive control a stimulation ofCO2 production is likely to indicate a stimulation ofoil biodegradation (Swannell et ai 1995 SwarmeD et aI 1997) However ifa product contains organic molecules then a stimulation of CO2 evolution may indicate merely the biodegradation of components of the product (Croft et al 1995) Therefore CO2 evolution is used for indicative pwpo only and not as part of the passfail criteria for the produer

The evolution rate ofcarbon dioxide is determined in each microcosm before oil addition and on each orlcing day thereafter at the same time in the tidal cycle A graph of CO2 evolution against time is plotted for each microcosm at each sample time and the linearity of the relationship is confirmed using regression analysis The results of the regression analysis are then used to calculate the rate ofCO2 production in vpmmin-J

bull

The device is calIbrated each working day prior to analysis using CO2 -free air (to zero the instrument) and a standard gas containing a known amount of CO2 mixed with nitrogen The subility of the analyser over the duration of the measurements must be confinned prior to conducting the test

AEA Technology 12

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

AEAT-1981bull

372 Enumeration of Hydrocarbon-Degraders

When oil is added to marine sediments the indigenous bydrocarbon-degrlCiing community proliferates to utilise the new source ofcarbon and energy However the proliferation is usually limited by the concentrations ofspecific nutrients such as nitrogen and phosphorus Thw adding these limiting nutrients may well stimulate the concentration ofhydrocarbon-degraders (prince et al 1993 Swannell el aI 1996 Swannel d aI 1997) The aim ofmonitoring the number ofhydrocarbon-degraders is therefore to confinn the presence ofan active hydrocarbon-degrading microbial community in the Sterr sediment and to study in outline the effect of the bioremediation treatments on the community

Three samples (1-2 g) are taken at random from each microcosm for microbiologiCl1 analysis directly before addition ofehe oil and 7 and 28 days after oil addition Samples are collected within an 8 em diameter circle in the centre ofeach cace to avoid edge effects The nwnbers of hydrocarbon-degrading micro-organisms are detennined using a Most Probable Number (MPN) technique based on the sheen screen technique (Brown and Braddock 1990) and described by Swannell et al (1995)

38 CHEMICAL ANALYSIS

This is conducted in the same maruter as for the flask test except that 10 grams ofthe sediment are extracted rather than the volume of the Erlenmyer fbsk The same quality assurance procedw-es are used for these analysis

39 ANALYSIS OF RESULTS

The test will only be regarded as valid if the positive controls demonstrate significant stimulation (pltOOS) of the biodegradation of oil in comparison to the negative control Once this has been confinned then the efficacy of the product(s) may be assessed

The oil chemistry data is analysed in the same way as for the flask test The product passes the test if the oil in the test microcosm is significantly (pltOOS) more biodegraded and at a much lower concentration than that in the negative control

AEA Technology 13

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

4

AEAT-1981

Reporting of Resultsbull

Based on the results of the tests statistical analysis is used to detennine whether the product has significantly (pltOOS) enhanced the biodegradation ofoil in comparison to untreated controls in both the flask and the microcosm test (Section 29 amp 39) The following documentation will be generated at this stage

bull data listings ofeach analyte (eg oil chemistry CO2 evolution and MPN data where appropriate)

bull a table ofSlll11ll1aI) statistics for each analyte including mean standard deviation and vanance

bull th mean reduction (plusmn standard deviation) in oil content in each test

bull an ANOVA table for each analyte tested statistically (ie oil chemistry results from flask and microcosm studies)

bull a clear sununary of the statistical findings as produced by the appropriate computer software bull details of the computer software used to conduct the statistical analyses

[[the product fails to stimulate oil biodegradation in the flask test then the result will be reported to the Marine Environmental Protection Division of MAFF and the testing will cease If a significmt difference is recorded in both the flask and microcosm experiments it is reported to the Marine Environmental Protection Division of MAFF that the product does significantly enhance the biodegradation of oil under the test conditions Based on the results of the toxicity assessment and the efficacy test MAFF will decide whether to approve the product for use on shorelines in the UK

r

AEA Technology 14

bull

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

5

bulllbull

I

[

AEAT-1981

References

BraggJ- R Prince RC Harner EJ amp R M AtUs (1994) Effectiveness of bioremedi2tion for the Exxon Valdez oil spill Nature 368413-418

Brown EoJ- ampJ- F Braddock (1990) Sheen Screen a Miniaturized Most Probable Number for Enumeration crOil-Degrading Micro-organisms App Environ Microbial 563895-3896

Croft BC Swannell RPj Grant AL amp K Lee (1995) The effect ofbioremediation agents on oil biodegradation in medium-fine sand In Applied Bioremediation ofHydrocarbons (Hinchee RE bull Reisinger HJ amp Kinel J-A (Eds) Banelle Press Columbus OH pp423-434

Jones nM Rowland S] amp Douglas AG (1986) Steranesas indicatOIS ofpetroleum-like hydroorbons in marine surl2ce sediments Mar Poll Bull 17 24-27

Klute A (Ed) (1986) Methods ofsoil analysis Pan 1 Physical and mineralogical methods No 9 Part 1 Sccond edition American Society ofAgronomy inc Soil science Society ofAmerica inc pp 383 - 409

NETAC (National EnvirorunenuJ Technology Applications Corporation) (1993) EVltl1uation Methods Manual Oil Spill Response Bioremediation Agents Published by NETAC Univ~rsity

ofPittsburgh Applied Res=ch Cenrer Pittsburgh PA USA

Montgomery DC (1991) Design and Analysis ofExperiments Third Edition John Wiley amp Sons NYUSA

Park~r RE (1980) Introductory Statistics for Biology S~cond Edition Edward Arnold Ltd London

Princ~ RC (1993) P~trol~tun Spill Bioremediation in Marine Environm~nts Crit Rev MicrobioL 19 217-242

Prince R c Bare RE George GN Haith CE Grossnun MJ Lute JR Elmendorf DL Minak-Bemero V Seniw JD Keirn LG Chiandli RR Hinton SM amp A R Teal (1993) The effect ofbioremediation on the microbial populations of oiled b~aches in Prince William Sound Alaska In Proceedings of the 1993 Oil Spill Conference American Petroleum Institute Washington DC USA pp 469-475

Ptitchard P H bull MuellerJGbull RogenJ-C Kremer FV amp J- A Glaser (1992) Oil spill bioremedi2tion experiences lessons and results from the Exxon Valdez oil spill in Alaska Biudegradation 3315-335

Ries-Kaurr M amp Albrecht P (1989) Hopane-dcrived triterpenoids in soils Chem Geo 76 143-151

AEA Technology 15

l4 LJAEAT-1981

Sakal RR amp Rohlf FJ (1981) Biometry Second Edition WH Freeman amp Co San Francisco USA

Swannell RPJ Tookey D amp M McDonagh (1994a) The Bioremediation of Oil Spills A State afthe Art Review Warren Spring Laboratory Report LRlO08 Obtainable from the Library National Environmental Technology Centre Culham Abingdon Oxon OX14 3BB UK

SwanneIl RPJ Basseres A Lee K Merlin F X (1994b) A direct respirometric method for the in situ detennination ofbioremediation efficacy Environment Canada Onawa Canada K1A OH3 pp 1273-1286

Swannell RPJ Lepa JE Lee K Pritchard PH amp DM Jones (1995) Bioremediation ofoil contaminated fine-grained sediments in laboratory microcosms Published in Proceedings of the Second International Oil Spill Research and Development Forum 23-26 May 1995 Intcmational Maritime Organisation 4 Albert Embankment London UK pp 45-55

Swannell RP] Mitchell D] amp Engelhardt MA (1996a) Natue 379 14

Swannell RP] Lee K amp M McDonagh (1996b) Field Evaluotions ofMarine Oil Spill Bioremediation Microbiology Reviews 60 342-365

Swannell RPjjone DM Wilfu AL Lee K LepojE (1997) An evaluation of Oiled Sediments Buried within a Mudflat Envirornnent Proceedings of the Twentieth Arctic and Marine Oihpill Programme Technical Seminar pp703 - 713

US Environmental Protection Agency (1990) Semivolati1e organic compounds by Gas ChromatographyMass Spectrometry (GCMS) capillary colwnn technique Standard Method 8270 As published in Electronic EPA methods by ChemSoft Inc CA USA

Venosa AD Suidan MT HainesJR WrelUl BA Stroruneier KL Eberhardt BL Kadkhodayan M Holder E King D amp Anderson B (1995) Field Bioremediation Study Spilled crude oil on Fowler beach Delaware In Bioaugmentationfor Site Remediation Hinchee RE Fredrickson] Alleman BC Eds Batelle Press Columbus pp 49-56

i 1

t

bull bull

AEA Technology 16

bull