Odor Related Issues in Commercial Composting › ref › 12 › 11500.pdf2.1 Gas chromatography...

Odor Related Issues in Commercial Comoostinq

Workshop presented at the Y2K Cornposting in the Southeast

Conference & Expo October 9 - 11, 2000

Charlottesville's Omni Hotel Charlottesville, VA

K.C. Das University of Georgia

Bioconversion Research and Education Center Athens, Georgia

Composting Odor Workshop

Odor Related Issues In Commercial Composting Workshop presented at the

Y2I< Composting in the Southeast Conference & Expo, October 9,2000

TABLE OF CONTENTS

1. INTRODUCTORY TOPICS Page # 1.1 What is an odor: human perception mechanism 1.2 Definitions and descriptions relating to odor 1.3 Types of odors released in cornposting 1.4 Release of odors from hfferent parts of the cornposting process 1.5 Biologcal generation of odorous compounds 1.6 VOCs - why are they regulated and releases during composting 1.7 Transport and dspersion of odors

2. MEASUREMENT TECHNIQUES 2.1 Gas chromatography methods 2.2 Odor panels - triangle forced choice method 2.3 Field measurement of odor concentration - Scentometer D/T 2.4 Odor - compound detection using tubes 2.5 Dehcated detectors - Total hydrocarbon, hydrogen sulfide,

ammonia, etc.

3. ODOR MANAGEMENT I N COMPOSTING 3.1 Treatment options for odor control 3.1.1 Biofdtration

AdvantuJes and disadvantages o f biojlters Applimtions o f biojlters Design Of a biojlter Maintenance and operational miteria Cost infomation

3.1.2 Chemical scrubbing 3.1.3 Cost comparison

3.2 Odor mevention and control throuph Drocess manapement 3.2.1 Nutrient balance 3.2.2 Ash as an odor-control amendment 3.2.3 Other amendments 3.2.4 Use of biologcal odor control adhtives 3.2.5 Influence of the size and shape of composting piles

4 5 7 10 11 12 14

19 20 23 23

24

25 25 26 27 27 30 31 31 32

32 33 35 35 36


3.2.6 Influence of tuming on the rate of composting and

3.2.7 Workmg with hgh moisture feedstocks odor production

3.3 Other oDtions and toDics in odor manapement 3.3.1 Site selection for odor management success 3.3.2 Barriers to prevent transport 3.3.3 Operational and contingency plans

4. REFERENCES

5. APPENDICES 5.1 Conversion of gas concentration from mass basis to volume basis 5.2 Calculation of biodegradable C/N ratio

36 37

37 39 39

40

44 45

DISCLAIMER T h s booklet contains information gathered from the scienttfic and industry literature in composting, from the personal experience of the author and from ongoing research studes at the author’s laboratory. The contents are provided for educational purposes only. The accuracy of information presented has not been edited or evaluated through a peer-review; therefore the use of the information requires careful judgment on your part. Use of brand names and trade names are for illustration purposes only and do not imply endorsement.

When developing an odor management plan for an operation, consultation with a professional engmeer or scientist is strongly recommended.

Cornposting Odor Workshop

Odor Related Issues In Commercial ComDosting Workshop presented at the

Y2K Composting in the Southeast Conference & Expo, October 9,2000

1. INTRODUCTORY TOPICS

1.1 What is an odor: Human DerceDtion mechanism

An odor is a sensation that results from specific compounds depositing on the olfactory system in the human nose. Air is inha 9 ed in through the nose and passes through the regons of the olfactory system where initial contact occurs. In a simplified view, the olfactory system Figure 1) consists of the mucous membrane, the olfactory epithelium [sensory cells], olfactory hairs and the trigeminal nerve located inside the nose.

C L m i J

The evolutionary purpose of smell is to serve as a guide to humans through unsafe and safe environments. The abhty to identify flavor and tastes is closely linked with smell. As a safety mechanism, the trigeminal nerve acts to ensures the trigger response [e.g. sneezing] when an irritant k e ammonia has been inhaled.

Figure system


4 0

9

1.2 Definitions and descrbtions relatinp to odor

Odor is perceived by dtfferent people lfferently and is therefore measured in dtfferent ways. In order to quantify odor several descriptors are commonly used. Odor intensity and soncentration are the most commonly used descriptors. Odor intensity is a measure of the strength of the odor. It is measured using the 1-butanol reference standard [ASTM E544 - Summary in Appendur]. A human compares the odor intensity of the unknown sample with the intensity of a sample of 1-butanol at dtfferent known concentrations. The results are reported as equal or greater than X ppm of butanol.

The odor concentration of a compound in an air stream is typically measured as ppm [parts per d o n v/v]. Thls value can be measured using a gas chromatograph or a dedrcated analyzer that can detect the target compound. The Steven's Law relates concentration of odor with perceived intensity, thus definmg pervasive and dutable [non-pervasive] odors (Figure 2).

Equation 1. Steven's law relating concentration (C) and intensity (I) of an odorous compound. Where k, n = constants based on type of compound

I = kcn

Figure 2. Relationshp between concentration of an air pollutant and perceived intensity [Steven's law].

0.1 0

0.01 0.1 1 10 100 loo0 loo00

Concentration

The value of "n" in the above equation varies fiom 0.2 to 0.8 in most compounds (Dravnieks, 1979). The hgher the value of n, the quicker the perceived intensity

pervasive" or dutable. A good example is ammonia, whch can be very quickly . reduces with reduction in concentration. Therefore the odor is termed "non-

c

I


reduced to low levels of intensity by dtluting with odor free air. In contrast, hydrogen sulfide and many of the amines are considered “pervasive” odors. They have a very low value of “nYy and even after several ddutions wdl inchcate little change in perceived intensity.

Odor detection threshold D e s h o l d odor concentration] is the minimum concentration of a specific compound in a gas stream at whch the odor can be detected as being present. The recognition threshold is the concentration at whch a person can actually identify the odor. Dravnieks (1980) reports that for most odorous gas contaminants, the recognttion threshold is 2-10 times hrgher than the detection threshold. A representative set of commonly found gases in composting is shown in Table 1. You can see that some compounds hke skatole and methyl mercaptans are detected at very low concentrations compared to others lrke butanone [methyl ethyl ketone], whch are detected only at hgher concentrations.

Table 1. Characteristics of a representative set of odorous

I Skatole 131 95 7.3 x 0.03 J ’ Detection threshold recalculated from Epstein (1997) & WEF & ASCE (1995); assuming T 20°C. Thresholds depend on the sensitivity of the odor panel and therefore have a range. The numbers reported in the table are the lowest values in the range. Odor index values obtained from Haug (1993). Odor index > 1X106 represent compounds that have “ a hugh odor potential.

2

During composting, gaseous compounds are generated as byproducts of metabolism. They are released to the atmosphere based on their physical properties [volatlltty]. The potential of a compound to generate an odor is affected by both its volatlltty and its detection threshold. In order to take into account both these properties, H e h a n and Small (1973) defined the Odor Index (Equation 2) of a compound as the ratio between the vapor pressure [ppm] and the odor recognttion threshold [ppm]. T h ~ s non-dimensional value is hrgher when (1) the vapor pressure is hgh, i.e. the gas volaalrzes easily; and (2) the recopt ion threshold is lower, i.e. a human can easily

- r e c o p e the odor (Table 1).

a


Equation 2: Odor Index of a compound indicating its potential for causing odors.

Vapor pressure, ppm Odor recognition threshold, ppm

Odor Index =

Generally the concentration of a mixed stream of odorous compounds is measured using a standard odor panel [ASTM E-6791. The odor concentration then is reported as Dilutions to Threshold (D/T) also called Odor unit (ou) or ED50 [which are dimensionless numbers]. When an odor panel of eight or more people tests the sample, D/T represents the number of llutions with odor-free air required to ddute a volume of the test sample dl 50% of the panel is able to detect the odor. For example, if 1 m3 of test sample requires 3 m3 of odor-free air to reach a point where 50% of the panel can detect the odor, then the concentration is reported as 4 D/T [ou or EDSO]. [Note: In literature from the European Union, the odor unit is sometimes referred to as ou/m3. The d e h t i o n here is that the odor concentration bf an air sample at the detection threshold is equal to 1 0u/m3. Thls value is used to calculate the total emission by multiplying O U / ~ with the airflow m3/h.r resulting in ou/hr whch is numerically equal to 0u-m3/hr = concentration X flow rate].

When workmg with windrows or invessel systems of composting [especially in forced aerated systems], the release of odor at the surface is called a surface odor emission rate [SOER, m3/hr]. The value is calculated from a measured value of D/T and air velocity rate at the surface of compost and the total surface area present (Equation 3). For example, if the surface area sampled using a hood is 2 m2 and has a volumetric flow rate was 10 m3/hr releasing gases at a D/T of 30. Then the total SOER is calculated using equation 3 to be 150 m3/hr-m2. Thls implies that for a composting total surface area of 100 m2, you would require 15,000 m3 of odor-free air per hour to ddute the odors to their threshold value.

Equation 3: Surface odor emission rate (SOER) of a area source of odors.

9 m3 hr c+ [Concentration in D / TI x [Volumetric Flow in -1

Surface Area of Sampling in m2 obo' COER =

/J 9.7 p@-#d T(,$&d 5 1.3 TvDes of odors released in compostin3

> Composting is the biologcal decomposition and stabhation of organics with the goal of reusing the organics beneficially. Some typically composted organics include yard trimmings, biosolids, farm manures, food wastes and industrial wastes &e paper mtll sludges. All of these organics are made up of one or all of the following:

. carbohydrates, proteins, lipids and nucleic acids. The biologcal breakdown of these


feedstocks results in compounds &e ammonia, amines, acids, sulfides, alcohols and aldehydes (Table 2). Dependmg on the type of feedstock being composted, the type of aeration system and the level of control of the process, specific emissions and concentrations vary.

Table 2. Typical types of odorous volatde intermebtes generated during decomposition of organics.

Source Alcohols Carbohydrates

Odorous intermediates or h a 1 products

Volatile organic acids - low molecular weights Aldehydes Ketones

Proteins Ammonia + Ammes Volaale organic acids - low molecular weights Hydrogen sulfide [from S-containing amino acids] Mercaptans

. Lipids Volade organic acids - low molecular weights Alcohols Hydrogen sulfide and Mercaptans

Odorous compound Mean concentration, ppmv I Ammonia

0.7 Hvdropen sulfide 171

Dimethyl sulfide

784 ED,, komDosite odor concentration1 0.3 Dimethyl disulfide 0.5

1 Values in the table are recalculated from measured values using ambient T = 2OOC. 2

3

4

- Values are averages of three samples over time. Sources : Van Durme et al. (1990); Epstein (1997) Although at a much hgher concentration, ammonia represented only 30 D / T compared to dunethyl sulfide w h c h represented 493 D/T.

Gaseous odor concentrations at the forced air exhaust in an aerated static pile system composting dgested biosolids were reported by Van Durme et al. (1990). Selected values are recalculated to ppmv units assuming an ambient temperature of 2OoC and summarized in Table 3. It can be seen that ammonia was released at a very hgh concentrations. Thls is a result of the hgh rate composting where ammonia volathzation is enhanced with hgher temperatures and hgher pH values. Although, the concentration of ammonia was much hgher than the sulfur compounds detected, its D/T was much lower relative to dunethyl slufide, whch has an odor index 16.5 times hrgher than ammonia. Van Durme et al. (1990) reported that measured

. ammonia concentrations represented only 30 D/T whde hydrogen sulfide, dunethyl


sulfide, and dunethyl &sulfide represented 61,493 and 208 D/T, respectively. Total mass emissions of odorous VOC from an invessel biosolids composting operation were measured to be equal to 7.3 pgrotal hydrocarbons/kgDV cornposting miu (von Fahenstock, 1995). The maximum mass releases were in the form of acetone and methyl ethyl ketone (2-butanone). T h s provides an estimate that for a 100 tpd composting fachty, the total hydrocarbon emissions would be well below 1 kg per year.

Food wastes are typically hgh in moisture content and hghly decomposable sugars, starches, proteins and fats. Therefore, composting is usually rapid resulting in oxygen depletion and anaerobic condrtions, enhancing the release of odorous compounds. Typically, short chain fatty acids [formic and acetic acids], organic sulfur compounds, ketones and ammonia are released. Epstein (1997) reported odor samplmg at a food waste composting site revealed fatty acids concentrations of 2.5-25.3 ppm, sulfur compounds of 0.32 ppm and butanone [methyl ethyl ketone] concentrations of 600 ppm. Ammonia was not found in the samphg, whch could be a result of the timing of samplmg more than the lack of its presence.

Overall volatde organic emissions were reported to be 590 g/ton of biowaste during aerobic composting. Total inorganic ammonia release was 152 g/ton (Smet et al., 1999). The hghest single component were short chain alcohols contributing to 48% of the total mass of volatde organics released. Table 4 summarizes the results reported in thts study.

Table 4. Total emission of odorous volatde organic and inorganic compounds during the aerobic composting of biowaste.

Compound groups Total emission

Alcohols: 285 Propanol, ethanol, butanol . . . Carbonyl compounds: Acetone, butanone, heptanone, methylbutanal

158

Terepenes: 82 Limonene. Dinene.. . .

g/ ton

Esters: I 53 Ethyl acetate, methyl acetate, methyl propionate . . . Sulfur compounds: Dimethvl sulfide. dunethvl &sulfide. carbon &sulfide . . .

9.2

Ethers: I 2.7 Ethryl fixane, methyl furane, &ethyl ether, . . .

Sub total 590 Ammonia

742 Grand to tal 152

;ource: Smet et al. (1999).


1.4 Release of odors from different Darts of the comDostinP Drocess:

A comprehensive measure of odors released from Qfferent parts of a mixed waste (MSW) windrow composting process showed that the hghest odors were released from the windrow areas (Homans and Fischer, 1992). They report odor concentrations of 600-1000 ou in the composting area compared to 50-500 in other areas (Table 5). T h s study also reported that total ambient odors were hghest in the first week of composting, however, surface odors from cornposting piles [not includmg turning times] peaked at the fourth week of composting.

Table 5. Odor concentrations measured at Qfferent parts of an MSW composting operation.

Source locations where odor is generated Typical concentration ranges, ou D/TI

Composting areas

5,000 - 25,000 Aerated windrow exhaust 20,000 - 80,000 Invessel drum exhaust

600 - 1,000 Ducted air

Source: Homans and Fischer, 1992. . I

In food waste composting, odors were found to be hgher with recieving area air having concentrations of 4,800 ou with a range of 256 to 8450 ou (Bidhgmaier, 1992). In the second and fourth week, the ducted air from invessel composting ranged from 16,700 to 20,400 ou. Turned windrows had surface release of odor at concentrations of 355 ou [range 30 to 1150 0.1 before turning and 4,455 ou [range 90 to 15,940 ou] ImmeQately after turning.

**

Odor release from the composting process is well known to be hgher in the initial stages of composting. T h ~ s is a result of two phenomena acting in conjunction, the hgher biologcal activity and its resulting lack of oxygen avadabhty. Because of the low oxygen levels and hgh temperatures, composters typically agitate the piles more during the initial stages of composting thus contributing to the release of odors. Iacoboni et al. (1984) reported surface odor emission rates [SOER] of 10-20 m3/min- m2 in the first seven days of cornposting biosolids. Thereafter, SOER was below 5 m3/min-m' and reduced to below 2 m3/min-m2 after 15 days of composting. Thls study also showed that the release of odor was 25 times hgher immediately after turning a cornposting pile [550 m3/min-m2] and reduced to pre-turning levels [22 m3/min-m?] only after about four hours.

Some of the odorous compounds, e.g. short chain fatty acids (FA) are rapidly degraded during the composting process. They are only released during the first 24

" .


hours because of the increase in temperature of the cornposting pile whch causes volathzation. If agtation and mixing of the pile is avoided during t h rs time, some or most of the FA could be biologcally broken down. Kuroda et al. (1996) reported measurement showing that almost 100% of the FA was rapidly degraded withm the first 24 hours of composting of swine manure.

1.5 Biolodcal peneration of odorous compounds - S, N metabolism

Of the several compounds and their mixtures the nitrogenous and sulfurous compounds deserve specific attention. The breakdown of nitrogenous compounds results first in ammonia. Ammonia is fairly soluble in water and remains in solution as ammonium ions (NH4+, Equation 4) when the pH is low. At pH greater than 7 , ammonium salts shf t to ammonia and volaalrze. The transformation of ammonium to ammonia is also accelerated with increasing temperature.

Equation 4. Ammonia-Ammonium equrlrbrium.

NH4+ @ NH3 + H+

Biologcally, ammonia is broken down into nitrites m02-1 and nitrates [ N 0 3 - ] by nitrosomonas and nitrobacter, the nitrifying bacteria species. Once in the form of nitrates, denitrifiers convert the nitrates to atmospheric nitrogen. Ammes [rotten fish odor], amides and nitnles are further end products of the breakdown of nitrogenous compounds.

Cysteine and methtonine are two amino acids in proteins that contain sulfur. Decomposition of organic sulfur under oxic (presence of oxygen) condtions results in mecaptans (Organic group - SH) and anoxic condtions (absence of oxygen) results in hydrogen sulfide. Both these products are generated when decomposition occurs in soil. In addtion dunethyl sulfide is generated under condtions &e marine environments where anoxic condtions predominate. Under anoxic condtions hydrogen sulfide is the intermedate, finally leadmg to elemental sulfur, whde under oxic condtions, sulfate [SO4*-] is the intermedate finally leadmg to elemental sulfur (Atlas and Bartha, 1998).

Hydrogen sulfide is in equrlrbrium with hydrogen ions and sulfide ions as shown by equations 5 below. The sulfide ion is very water-soluble, however, hydrogen sulfide is not water soluble and volattllzes. Volathzation of the sulfur occurs only when the equations shtft to forming more hydrogen sulfide, whtch occurs when the pH reduces (Figure 3).

h


Equation 5. Eqdbr ium of Sulfide forms.

H2S e HS- + H+ HS- S-2 + H+

A common analysis performed is measuring total reduced sulfur compounds FRS) in an air stream. Thts analysis measures the total of hydrogen sulfide, methyl mercaptan, dunethyl sulfide and dtmethyl &sulfide.

Figure 3. Sulfide eqdbr ium at T=28C.

4 6 8 10 12 14

pH

1.6 VOCs - Whv are thev remlated and releases durinp comoosting

The US EPA regulates VOCs (volatde organic compounds, also referred to as hydrocarbons) as primary air pollutants. Since, in non-attainment air shed areas VOCs are regulated, it is of interest to us to understand VOCs and their potential release from composting. By defuution a VOC is an organic compound (hydrocarbon) that at 25°C has a vapor pressure in the range of 0.1 to 380 mm of Hg.

VOCs are considred pollutants.because they contribute to ozone formation in photochemical smog. Normally, NO in the atmosphere combine with oxygen to form N02. NO2 reacts with ultraviolet rahation causing the production of N O and 0. The 0 combines with oxygen forming 0 3 , whch in turn reacts with NO to form NO2 and oxygen. Therefore these four NO,-03 reactions stay in balance. However,

- when hydrocarbons are present in the atmospheres, they react with N O thus takmg


them out of the loop and resulting in formation of more N02. The result is an imbalance in the NO,-03 reactions and accumulation of 0 3 in the atmosphere. Ozone when inhaled results in lower efficiency of the lungs and severely affects indwiduals with asthma. It also affects plants, destroying vegetation and results in the formation of photochemical smog.

Table 6. VOCs measured in the blower exhaust during the active phase of biosolids composting. , pg/m3

Compound detected 60 Acetaldehvde

Concentration

I 25 I I Acetic acid I 2.574 I I Benzene I 104 I Carbon &sulfide

442 Cyclopentane 13 Cyclohexanone

327 Cyclohexane 9 Chlorobenzene

224

Dichlorbenzene 9 2-Ethoxyethanol 9 Ethvlbenzene 16

I Fluorotrichloromethane I 1.493 I

Heptanone 46 Methanol 153 Methylacetate

16 Methylchloride 144

Methyl ethyl ketone 974 Nonane 19 Octane 15

I -i%ntane I 884 I I Phenol I 13 I

Pyridme

29 Xylene 27 l71,2-Trich1oroethane 488 Toluene 26 Styrene 47

Source: Van Durme et al. (1992).


Table 6 shows measured vo1at.de organics released in biosolids compoting (van Durme, 1992). The hghest concentration is of the short chain fatty acid, acetic acid. The survey of eight MSW composting facrlrties by Eitzer (1995) showed that average VOC concentration in various parts of the cornposting operation ranged from 8.9 mg/mg3 [in curing piles] to 104.1 mg/m3 [in freshly developed composting piles]. The hghest total VOC emissions were found in the fresh piles, tipping floor and the lgester. The concentrations in these samples were 8-10 times hgher in concentration than the curing piles. From the data reported by Eitzer (1995), it appears that the concentrations of most VOCs are hgher in MSW composting than in biosolids composting. Thls could be a result of the greater amount of carbonaceous fraction in MSW compared to biosolids.

1.7 TransDort and dispersion of odors

Odorous gaseous emissions from composting operations are inevitable. Process management could reduce the amount of emissions and emissions can be contained ‘and treated using state of the art air pollution control technology. However, finally, effective ddution is the recourse for all gaseous releases to the atmosphere. With sufficient lspersion in the receiving atmosphere, the air quality wik the neighboring areas can be maintained wik acceptable levels.

When emissions are released to the atmosphere they are reduced in concentration by several mechanisms: ddution, s e t h g of particulates, washmg action of rain, photochemical reactions, and absorption by plants and soil. The amount of lspersion [and ddution] that occurs in atmosphere is dnven by two phenomena, namely, surface wind speeds and vertical movement. Vertical movement is a result of temperature lfferences as you move vertically away from the surface of the earth. Normally, as you move away from the surface of the earth, the air temperature decreases. Thts decrease is called the [actual measured] lapse rate and varies from location to location, local cloud cover, sunlight etc. The adiabatic lapse rate is a theoretical change in temperature of a parcel of dry air as it is lsplaced to a higher position in the atmosphere. Thu reduction in temperature is a result of reduction in pressure as you move up into the atmosphere and is -lO°C/km.

Meteorologcally, during daytime hours there are at least three condttions that occur relating to lapse rates (Figure 4). These are: 1. Unstable: When actual lapse rate decreases with altitude faster than the adlabatic

lapse rate. When a parcel of air moves up, it cools at the adlabatic lapse rate, if the actual surroundmg temperatures are lower, then the air parcel d always be warmer than its surroundmg, thus getting accelerated further up. Thls conltion results in h g h turbulence and the parcel of air that is lsplaced vertically wdl accelerate because of the added push of buoyancy.

http://vo1at.de


2. Neutral: When the measured lapse rate decreases with altitude at the same rate as the adrabatic lapse rate. Here buoyancy is zero and a drsplaced air parcel remains in its new location without further acceleration.

3. Stable: When the measured lapse rate decreases with altitude slower than the adrabatic lapse rate. This condrtion is accompanied by very low turbulence and a lsplaced parcel of air is restored to its origmal location by buoyancy forces. Vertical transport is hmdered.

If temperature increases as you go up into the atmosphere, &IS is called a temperature inversion (Figure 5). Under these conltions, a rising parcel of air d always be cooler than its surroundmgs, therefore heavier and d be pushed back to its origmal position.

Atmospheric stabdtty is a function of solar radration, degree of cloud cover and wind speed. Early in the day when the sun rises and drrect solar radration is hgh, the soil heats up and the air around it heats up. Thls leads to very active vertical movements And unstable condrtions. Late in the day and m d y during the night, the temperature profile changes as the earth’s surface cools. ThIs leads to the neutral to stable condrtions untd late into the q h t the profile shfts to complete inversions. Fqure 5 shows a condrtion where the inversion is bounded to a certain height. In such cases there is good mixing above the boundmg altitude, but little or no vertical mixing below it. Release of odorous emissions during inversion condrtions is one of the most important causes of odor complaints from neighbors.

Dispersion models use meteorologcal data along with mathematical equations that include plume transport and lspersion. These then predrct the concentration of odor downwind under various topographcal situations and meteorologcal conltions. Knowledge of the worst-case condrtion can be useful in both siting the facdtty [for new fachties] and for planning operations [in existing facdtties].

Among various models evaluated for modelmg odor Qspersion from composting fachties, the EPA ISCST3 was found to be most appropriate (Alpert and Wu, 1997). They indrcate that target odor concentrations at a receptor should be at 5 D/T to avoid odor complaints. With &IS target, the model could be run under various metrologcal conltions and known values of emissions from the source. If the fachty is in planning and drrect estimates of odor emissions are not avadable, the following emission concentrations can be used (Giggey et al., 1995):

Maximum odor release at composting area = 2000 D/T [Receiving/mixing area

0 Wet scrubber outlet = 60-120 D/T Biofilter outlet = 20-30 D/T [max 50 D / q

next hlghest and Curing and storing areas are lowest]


Figure 4. Relationshp between lapse rate and atmospheric stabhty. The three graphs inchcate unstable, stable and neutral conddons.

500

400

p 300 I

al 0 L

3 200

100

0

500

400

p 300 I

al s L

9 200

100

0

500

400

T 300 - m 0 c

$7 200

100

0

..... Adiabatic lapse rate

0 5 10 15 20

Temperature [“C]

T

‘r

0 5 10 15 20

Temperature [“C]

,t,.LlC

0 5 10 15 20

Temperature [“C]


Fgure 5. Inversion temperature profile. Notice the inversion occurs only to an altitude of 250 m, above whch atmospheric stabhty exits.

300- I

0

s .- c Boundmg altitude for inversion a 2 0 0 -

100 -

0 - 0 5 10 15 20

Temperature [“C]

Modehg results could tell you if the site is feasible at all or if a completely &fferent location needs to be considered. Alpert and Wu (1997) show an example of a composting site with a biofdters where the odor emissions were slightly above the target 5 D/T at receptor. In such cases minor mo&fications &e moving the biofdters 50 m south and addmg a cover on the biofdters brought the emissions to wih the targets. Modehg would also tell if there is a need for addrtional odor treatment at a particular site. Modehg results tell you when during the day or during a season you could expect your worst case conditions. With knowledge of &us, operators could ensure that turning of piles or delivery of fresh materials would not coincide with these condrtions.

Correlation of actual odor complaints and dspersion modehg results has shown three identifiable rules of thumb [ind~cators] listed below (Valentin and North, 1980). These are usehl to develop quantitative control points for indrvidual fachties using local condrtions and modehg results.

1. Odor complaints largely occur when wind speeds are low, both under very stable and very unstable condrtions.

2. The &stance from the site at whch neighbors perceive odors and complain is a function of the odor emission rate.

3. Odor complaints generally occur when the neighbors are a &stance where the odor concentration is at a level equal to about 5 times the detection threshold (5 D/T) or greater. Therefore t h ~ s target value at a receptor could be used to determine the maximum release allowable under any specified meteorologcal condtion.


From the frrst observation we r e c o p e that transport of odors occurs more effectively when wind speeds are low and minimal mixing occurs. When combined with the second observation, we can model the minimum emission rate at whch the nearest nelghbor can perceive the odor under worst-case local condttions pow wind speeds]. The thud observation provides a target concentration [5 D/T] at the point where the neighbor is located. Using these three observations [or rules] and avdable modelmg tools, we can develop target release concentration for dtfferent times of the day or year that would ensure minimal odor irnpact on neighbors.

Smalley (1998) indtcates how the use of dtspersion modehg results was incorporated into their biosolids cornposting operating plan. They required the establishment of a field weather station whch records temperature, dew point, wind velocity and wind hections. The operating plan called for windrow turning only when the wind velocity was greater than 4 mph and the dtfference between ambient temperature and dew point to be greater than 10'F. These condttions ensured a maximization of dtspersion.


2. MEASUREMENT TECHNIOUES

2.1 Gas Chromatocraphy Methods

Chromatography is a technique of takmg a composite gas or liquid sample of multiple compounds, separating it in a column and identifymg indrvidual comounds. In its simplest explanation, the drfferent compounds in the sample introduced to the column are captured at drfferent parts along the length of the column [usually 15-30 m long], then the column is forced to release [elute] indrvidual compounds at drfferent times by con t rohg carrier gas flow and column temperatures. As each compound is released from the column, a detector that provides a response proportional to the amount of compound detects it. The identification of the compound is based on the time it takes to release from the column, and the quantification of the amount of compound is made using the response of the detector.

'Known standards of target compounds are presented to the Chromatograph to determine the time at whch it is eluted and the amount of response for known concentrations of standards. This calibration is then used when unknown compounds are tested.

Figure 6 shows an example of a chromatogram generated when a sample containing 13 drfferent compounds is being analyzed. The horizontal axis of the chromatogram refers to time [in minutes or number of scans] at whch a compound is released from the column [i.e. when each peak occurs]. The vertical axis is a numerical value of 0- 100% of the maximum response obtained from the detector. The basehe is the response of the detector to the carrier gas [typically helium]. Every time a compound is detected, a peak appears where the area under the peak is proportional to the amount of compound. The chromatogram also shows an internal standard peak, which is a standard gas of known concentration that is introduced with each sample to ensure that the chromatograph is operating wih specifications.

Different types of detectors are avadable that are suited for drfferent groups of compounds. For example, the flame ionization detector PID] whch is typically used for all hydrocarbons works on the principles of qpt ing the compound released and providmg a response proportional to the amount of combustible carbon. S d a r l y drfferent types of columns are avdable that provide separation of drfferent groups of compounds. The selection of the column and detector is based on the type of gases being measured. The US EPA has provided gudehes for using gas chromatography for various VOC testing. Portions of Method 14-A and Method 17 are provided in the Appendm for reference.

t


Figure 6. Illustrative chromatogram from the output of gas chromatograph that was iniected with a multiple compound pas sample.

.... Tll

Compound Peak

Baseline response Of detector

2.2 Odor Danels - triancle forced choice method

The dynamic olfactometry (ASTM E-679) is a standard method of assessing the odor concentration (ddutions to threshold) of a sample. Although not a perfect method, t h l s is the best we have and is used world wide as minor modrfications of the triangle- forced choice method of dynamic olfactometry using an odor panel. The results of &us testing method are influenced by the panelist's judgment, anticipation and adaptation. The standard method minimizes false positives by using a combination of forced-choice multiple stimult, by presenting the samples using a method of systematically increasing concentrations and by using a panel of eight or more people.

The appendlx provides a summary of the scope of the method and provides the contact information for purchasing the full method. Here I have briefly o u h e d the key steps of sample collection, presentation, data recordmg, analysis and reporting (ASTM E-679-91; McGinely and McGinely, 1997; O'Brien, 1992).


SamDle collection: 1.

2.

3.

4. 5.

6.

7.

L

Sample is collected using non-reactive samphg pump, tube and a collection bag (typically Teflon@ and/or Tedlar?. When collecting samples from a duct, it is recommended to sample at a location where the pressure in the duct is positive [not vacuum]. It is recommended to measure and record the air temperature, flow rate [in ducts] and pressure at the time and place that the sample is collected. Typical samphg flow rates are 1-5 liter per minute. The sample bag should be condrtioned by f h g with the odorous air dl the bag is ‘/z full, hold one minute and then empty. T h ~ s condtioned bag is now ready for test sample collection. It is strongly recommended to store the samples at room temperature 25°C away from drrect sunlight. Samples should ideally be analyzed withrn 8-hours and never more than 24-hours after collection.

SamDle mesentation and data collection method: 1.

2.

3. 4.

5.

6.

Three samples [one true test sample and two blanks] are presented to each individual and the indwidual identifies whxh sample has a detectable [or recognrzable] odor. Generally, the frrst concentration presented would be at a level that most indrviduals in the panel are unable to detect the odor, and the sample presentation series is terminated when all panelists can detect the odor sample correctly. Samples are presented starting at a low concentration and successively increasing. Each step of presentation is at a concentration that is a constant multiple of the previous, e.g. D/4, D/2, D, 2D, 4D, 8D, . . . [where D is a du t ion level]. Each indrvidual is “forced” to select one of the three samples as an odorous sample and two as odor-free. If the panelist gets the odor correctly, + is recorded, else a 0 is recorded. See Table 7 below for a illustrative sample data set.

L I

Data analysis. 1. The series of indrvidual 0 and + are used to calculate the indrviduals threshold as

shown in the sample calculation below. 2. The group threshold is calculated using a geometric average.


SamDle calculation of D/T

Using Table 7 , for panelist # 1, the point of true detection is between drlution factors 135 and 45. Therefore the geometric mean is calculated as :

Individual threshold D / T = [135 X 45]0.5 = 78

ExceDtion cases: 1. In panelist 4, the lowest d u t i o n presented (hlghest concentration) is regstered

incorrectly. Therefore it is assumed that the next level would be correctly identified, which is a dilution of 5. Hence the calculation would be:

Individual threshold D/T = [15 X 5]0.5 = 9

2. Panelist 6 registered a correct identification in all cases; therefore it is assumed that their limit lies outside of the tested range. The next highest drlution power concentration] would be 3645 X 3 = 10,935. Hence the calculation would be:

Indwidual threshold D/T = [10,935 X 3,645]0.5 = 6313

Log-mean calculation: The last column of Table 7 shows the log of each value D/T for each panelist. Thrs value is averaged to be 2.25 [Sum/Number of samples, 18/81. The final reported D/T is the inverse log of 2.25 = 177.8 D/T. T h s number is often referred [or reported] as Best Estimate Threshold [BET].

Repeatabhty and precision are not very hgh in h s subjective measurement method. When 14 dfferent laboratories determined the detection threshold of hydrogen sulfide, the ratio of the hghest reported threshold to the lowest was 20. When four


panels of 23-25 members assessed a sample containing butanol, the ratio of hghest to lowest reported measure was 2.7; when a single panel repeated a measurement for four days, the ratio was 2.4:l. These results provide us with a feel for how much of variabhty is expected in measurements because of the subjective nature of b s test. Several of the laboratories train their panel to work with a specific type of odors, e.g. hog farm odor, biosolids cornposting odors, etc. These trained panels are reported to have a hgher consistency in their measurement.

2.3 Field measurement of odor concentration - Scentometer D/T

A simple field-measuring instrument called the “Scentometery’ is manufactured and sold by Barnebey and Sutcliffe of Columbus Oho. The instrument has two inlets of air that are dlverted to a sniff port that the operator uses to sniff through. One path of air comes duectly through from ambient air, and the second goes through a carbon filter whch removes all odorous components. A series of holes are present on the carbon filtered air path that regulates the amount of filtered air that is mixed ‘with the odorous air. Dependmg on whch hole is selected to be open affects the volumetric mixing of odorous and odor-free air, thereby providmg dlfferent ddutions. The operator changes the hole size dl they are unable to detect the odor. T h s hole specifies the D/T. Haug (1993) reports that the Scentometer is particularly suited for low-end (D/T < 20) concentration measurements, whde dynamic olfactometry is more suited for hgher concentrations.

A search of the literature inlcated that the Scentometer is not used in rigorous comparison. The instrument may be commonly used among fachty operators to estimate odor concentration on-site, however results are rarely published. The only published data were from Feldmann (1998) who used the Scentometer in evaluating odors released from hog houses. The paper reports values of air lscharge from the fan to be 5.4 to 27 D/T whch are in the range inlcated by Haug (1993) as suitable for measurement using a Scentometer.

2.4 Odor - comDound detection usinc tubes

Detector tubes are colorimetric indlctors of the presence of specific compounds in a gas stream. They are usually glass tubes, 4-6 inches long and 0.25 to 0.5 inches in lameter, filled with a reactive material. The ends of the tubes are broken off and a specified amount of gas is introduced into the tube at a defrned flow rate. When the target compound is present, it reacts with the material in the tube providmg a color change. T h s change is visually observed on a scale.

Although these detector tubes do provide a quantitative estimate, they are not very rehble because of their spot-check nature. In adltion, most detector tubes have

- h t a t i o n s in that they would react to false positives when certain other compounds are present in the gas stream. Nevertheless, they are a quick, inexpensive and simple

Y


method to check the presence of odorous compounds. When using detector tubes, read the duections carefully and look for interacting compounds and other h t a t i o n s of tubes.

2.5 Specific detectors - Total hydrocarbon. hydroEen sulfide? ammonia, etc.

Several dedicated detectors are avadable from various manufacturers. Most of these are for internal monitoring at the fachty or for research purposes when a single compound is being monitored.

The total hydrocarbon [flame ionization detector, FID] meter however, is a general- purpose meter that provides a measure of the total volatde organics in a gas stream. The US EPA allows the use of portable FIDs in monitoring total volatde organic emissions in stacks and other field condtions. Method 25-A, a partial copy of whch is provided in the Appendm, describes in detad how t h l s method is to be used.


3. ODOR MANAGEMENT IN COMPOSTING

3.1 Treatment options for odor control

There are several dfferent odor treatment options in composting. The use of chemical scrubbers, incinerators, catalytic oxidners, adsorption systems and biofilters has been reported in the literature. It is the authors belief that biofdters are a very cost competitive technology and very suited for composting odors because of their abhty to treat large amounts of air with low concentrations of multiple odorous comounds. In addtion biofdters are more environmentally friendly, a concept that goes well with composting. For these reasons, the description on biofdters is comprehensive in this document.

3.1.1 Biofiltration

Biofdtration is the process of using packed beds of porous meda (the biofilter) such 6s soil, bark, peat or compost, to capture and biologcally treat organic odorous compounds from a process air stream. h4icroorganisms grow in a b i o f h on the surface of the particles of the bed. Compounds from the air stream are either adsorbed to the surface of particles in the biofilter or absorbed into the b i o f h . Once the compounds are in solution in the b i o f h , they are avadable to the microorganisms for use as an organic carbon source for microbial metabolism (Figure 7). The end products of the metabolism are microbial biomass, CO2, H20 and heat. When the compounds in solution have been biologcally oxidrzed, the sorption capacity of the biofdter is restored. Hence, at a steady state the microbial population reaches a maximum based on the amount of organics in the air stream. Provided appropriate condtions are maintained, biofilters are known to operate effectively for long periods of time rangmg from months to years (Leson and Winer, 1991).

Biofdtration is widely used as an air pollution control technology in Europe. In the USA, the technology is just bepn ing to be accepted. The first patent for a soil biofdtration system was obtained in 1957 and research and testing has been in progress since then. Most commercially used biofdters are b d t as open single-bed systems, although other variations such as completely enclosed containers, layers etc. are in design and testing.

Bacteria, actinomycetes and fung are known to be involved in the biodegradation of pollutants in the biofdter. Microbial growth is dependent on, (1)dssolved oxygen present in the b i o f h , (2)absence of toxic by products of metabolism, (3)sufficient nutrients for microbial growth, (4)moisture, (5)temperature, and (6)pH suitable for microbes.

\


Figure 7. A schematic of a biofdtration system

Advantages and disadvantages of bioHters As biofdtration is a biologcal process performed by a consortium of microbes, a variety of contaminant compounds from the air stream can be simultaneously treated. Thls is a major advantage over chemical scrubbers whch are typically designed in stages that remove one group of compounds each. Biofdters are also particularly well

1

r\L suited for large air streams with a low concentration of contaminants.

Biofdters are low in cost to install and operate compared to other air pollution control technologes (see section on costs). Energy requirements to operate are low and dependent on the type of meda selected. Once in place and operating, they are

%. simple to maintain and produce no treatment residuals that need further dsposal.

Biofdters adapt easily to air streams that have a time varying level of contaminants. Wcroorganisms in the filter can use residual compounds adsorbed to the filter and hence the filter can survive periods of at least two weeks without reduction in microbial activity. With nutrient enhancement of the filter, these periods of survival without an inlet stream can be extended to hvo months (Leson and Winer, 1991). Spent filter material can be reused dtrectly as a compost, thereby producing no solid waste that needs to be dsposed.


Since a biofdter is a living system, a period of acclunatization is required before effective operation begms. Methods such as inoculation with a live population of microbes can decrease h s time r eqhed

Applications of bioflters Biofdters have been largely used in odor control at such fadt ies as rendering plants, composting operations and waste handlrng sites. The applications of biofilters are much wider than have been commercially tested. Table 8 gves a representative list of compounds that are removed in biofilters. l h s is not a comprehensive list and research stu&es have shown several other compounds can be detected. If toxic byproducts of metabolism can be removed, biofiltration has no known h t s and can conceivably be used to treat any contaminant.

Desip of a bioflter The basic biofdter system consists of four components, namely, (1)the vendation system, whch carries the gas stream to the filter, (2)blowers that control the air

'movement withrn the filter, (3)the gas drstribution system to the filter, and (4)the fdter bed (Kuter, 1990). Duct work for the vendation system carryrng air to the biofdter and blowers that can provide the appropriate flow rates agalnst the total back pressures are sized based on standard design procedure. Kuter (1990) recommends the use of a perforated network of air dstribution pipes embedded in (12.7 mm, 0.5 inch) gravel as the gas &stxibution system to the biofilter mela . Deslgn of the biofdter should manage the balance between static pressure, air flow rate and bed mecla porosity.

Table 8. Some compounds that can be removed effectively using: biofdters Kuter. 1990).

Compounds removed in biofiltration Removal rate, Yo Aldehydes 92.0-99.9 Amines, amides 92.0-99.9 Ammonia 92.0-95.0 Benzene > 92.0 Hydrogen sulfide 98.0-100.0 Mercaptans 92.0-95.0 Organic acids 99.9 Organic sulfides and &sulfides 90.0-99.9 Sulfur &oxide 97.0-99.0 Terepenes > 98.0

High particulate content, resins and grease in the air stream can lead to cloggmg of the bed and formation of sludge in the humidfication step. In such cases accessory filters need to placed before the biofiltration step and these fdters need frequent

. replenishment.


Table 9. Parameters used in the design of a biofilter (Kuter, 1990). Critical design parameters Typical recommendation

Residence time (sec) 30-60 Contaminant gas Loadmg rate (m3/min-m2) 0.9-1.5 Height of filter bed (m) 0.6-1.0 Porosity (?/o at field capacity) 40-60 Specific weght (kg/m3) 400-800 Moisture content (%) 40-60 Temperature ("C) 2-40 PH 6.0-8.5

Sizing of bed Degradation rates of commonly found air contaminants are 10-140 g(contaminant)/m3(air)/hr in an active biofilter (Leson and Winer, 1991; Tang et al., 1996). Due to the potential of producing a toxic environment for microorganism in the fdter by overloadmg, a maximum range of VOC in the influent gas should be less than 3000 to 5000mg/m3 (Leson and Winer, 1991).

Typical airflow loadmg rates of 300 m3(air)/hr-m2(filter surface area) is used without overly h g h back pressures. With good choice of medra 500 m3(air)/hr-m*(fdter surface area) flow rates can be used with good contaminant removal rates and low back pressure (Leson and Winer, 1991). However, if the d e t gas stream is not sufficiently condrtioned these excess loadmg rates can lead to excess heat loss and c o o h g of the bed, and dehydration of the bed.

As adsorption and absorption are the critical processes that h t the capacity of the filter, sufficient time has to be provided for effective transfer. Residence times of 15 seconds and greater is required when odor concentrations are variable. Kuter (1990) recommends the use of 30 seconds or greater residence times in the biofdter Fable 9). The required volume of the bed is based on Equation 6 with the typical loadmg rates of

Equation 6. Residence time calculation for a biofilter bed.

Volume of the filterbed (H X A) Specific loading rate (Q)

Re sidence time (t ) =

Medra The choice of material for the bed affects the air permeabhty of the bed, porosity and the specific surface area, and abhty to provide a good environment for growth of microorganisms. Medra should have pH between 7 and 8, pore volume greater than

the gradual compaction of the bed and biodegradation, the pressure drop over the ~ 80%, and total organic content greater than 55% (Leson and Winer, 1991). Due to


biofdter increases with time. The surface of the biofilter should be covered with bark or a s d a r mulch to prevent compaction from heavy rains and the growth of weeds on the surface.

Typical power consumption wdl be 1.8-2.5 kWh/lOOO m3. Thrs low energy consumption is one of the most attractive features of biofiltration (Leson and Winer, 1991). Table 10 below shows some examples of use of m e l a in biofiltration and their characteristics.

Table 10. Some typically used medta for biofdtration applications. M e l a Reference Comments Compost+bark+soil Kuter, 1990 Provides h g h adsorption, low pressure

Compost derived Leson and Sufficient nutrient base provides good

municipal waste microorganism Addrtion of Ottengraf, Polystyrene reduces pressure drop polystyrene spheres 1986 through bed and increase durabhty of

Addrtion of Ottengraf, Improves the buffering capacity of the Activated carbon 1986 biofilter thus reducing volume of filter

drop and sufficient organic matter

. from leaf, bark or Winer, 1991 environment for growth of

bed preventing se t thg and degradation

Desip steps Step 1 : Determine flow rate (Q) and contaminant concentration (C) of the gas stream that has to be treated. For example a g a s stream 5100 m3/hr containing a total contaminant concentration of 19 g(contaminant)/m3(air).

Step 2: Determine minimum time required for oxidation of the compound. Ottengraf (1986) has shown that typically thts oxidation rate (R) is in the range of 10 to 100 g(contaminant)/m3(air)-hr. Choosing a number in thls range, e.g. 100 g(contaminant)/m3(air)-hrY the minimum time required wdl be :

c 19 t, =-=- R 100

= 0.1 9 hr

Step3: As the biofdter operates on adsorption and oxidation, a correction factor is included to account for adsorption capabhty. "us correction factor (k) ranges from 0.2 to 0.3


and decreases the minimum time required. Thus, the true detention time required d b e :

t, = k * t, = 0.20 * 0.19 = 0.038hr

Step 4: Typically filter materials have a porosity (E) rangmg from 0.7 to 0.8. Filter volume (v> required is then calculated as :

Q*t , 5100*0.038 V=- - - = 276.8 m3 E 0.7

Step 5 : From table 9 recommendations, the height of the bed (H) is chosen to be 1.0 m. Hence, the surface area of the bed (A) should be at least 276.8 m2.

Step 6: A final check for the l o a h g rate pressure drop condrtion can be made using:

Q 5100 m3 A 276.8 hr - m2 - - - - = 185

T h ~ s is well withm the range of 300 m3(air)/hr-m2(filter surface area) recommended earlier.

Maintenance and operational cxitena As microbial oxidation is the primary removal mechanism of the odorous compounds, an environment for the healthy growth of microorganism is required for maintaining the effectiveness of the filter. The filter has to be kept aerobic, with sufficient moisture, optimal temperature, even flow of contaminant gas through the medra, sufficient supplemental nutrients for microbd growth and good structural properties of the medra for h g h performance over long periods of time.

The following parameters need frequent monitoring ('Kuter, 1990,2Leson and Winer, 1991): 1. Bed temperature - ddy' 2. Off gas temperature and humidrty - ddy' 3. Moisture content of influent gas - d d y 4. Moisture content of bed - monthly 5. pH - monthly' 6. Organic matter content- semi annually1 7. Pressure drop across bed and condttion of filter material - monthly'


Biofdters are slow at respondmg to changes in loadmg. Therefore grab samples are not recommended to test removal efficiency. Continuous monitoring of exit gas for total organic carbon using flame ionization detector or photoionization detector is recommended.

Maintaining moisture in the bed prevents craclung the bed thus preventing channehg of the air through the cracks and reduced contaminant filtration. Au coming in has to be saturated with moisture to prevent dqmg of the bed. Humihfication of the influent air and sprinkhng of the surface of the bed with water is required to achleve &IS. Water use in a biofilter is low compared to other treatment options. Leson and Winer (1991) recommend the rule of thumb of 5-10 gallons per 100,000 ft3(of air) as an estimate of water replenishment need. Some reuse of drainage in humihfying the inlet air is recommended, as the drainage can be rich in nutrients with a BOD of several thousand mg/L. Excess recychg of drainage, however, can result in the blockage of pores.

*Bed pH should be maintained as close to neutral as possible, but is dependent on the gas stream characteristics. If the pH drops, lune or wood ash should be added to the filter meha. Raw gas should be in the temperature range of 20-40°C for optimum filter operation. Preheating or c o o h g of influent gas may be required dependmg on the d e t air temperature and ambient temperature (winter vs. summer conhtions).

Typical labor requirement for maintenance are in the order of 0.8 to 1.0 person- hour/m2 (fdter area)-year (L,eson and Winer, 1991).

Cost information Typically open bed filters are low in capital cost and are estimated at approximately $5/CFM of gas treated (Mycock et al., 1995). Capital cost estimates for a biofilter handhg a flow of 10,000 cfm are reported to be around $75,000 compared to $195,000 for incineration systems, $212,000 for wet scrubbing systems and well over $500,000 for carbon adsorption systems (Bohn and Bohn,l987). Dependmg on the size of the filter, operational costs for biofiltration are $0.6-1.5/100,000ft3 of gas treated (Leson and Winer, 1991). Capital costs of open filters range $55-90/ft2 and for closed fdters are in the range of $90-500/ft2.

3.1.2 Chemical scrubbin3

Packed towers using chemicals and water to absorb and neutrahe odorous compounds are popular among large composting operations, especially invessel fachties. One of the main reasons is that chemical scrubbing relative to biofiltration is more prelctable and consistent over the long run. The down side of chemical scrubbers is that they are expensive to install and to operate. In adltion they use a

. large amount of water for the scrubbing process. Many of the scrubbers in operation around the country typically use from 15 to 90 gal-water/lOOO ft3-ai.r (Haug, 1993).


The most common type of system is the wet scrubber with two stages of reactions. In the first stage, ammonia is removed using sulfunc acid and water. Ammonia is removed first because it would interfere with oxidation agents in the second phase of organics removal. Once ammonia is removed, the air enters the second phase where a strong oxiduing agent &e sodum hypochlorite or chlorine &oxide is used. Both these oxiduing agents leave residual chlorine in the gas stream along with other partially chlorinated organics. Therefore in some cases a thlrd stage scrubbing with a reducing agent is performed to remove residual chlorine.

3.1.3 Cost comparison

Comparing various scrubbing technologes for removal of VOCs from an air stream, Ottengraf and Drks (1992) reported that capital costs for chemical absorption scrubbers were comparable to biofdters. However, operating costs were two to seven times hgher than biofdters. Thermal incineration systems cost twice as in capital

‘costs compared to biofdters, but operating costs were three to fifteen times hrgher than biofdters.

3.2 Odor mevention and control through Drocess manapement

3.2.1 Nutrient balance

During biologcal degradation, the avadabhty of the substrate to microbes dctates the rate at whrch the material is composted. Thts avadabhty is not well established because it is a function of the type of material, particle size, pre-processing and amendment. Typically we use C/N ratio using total C and total N measured in the laboratory as an indlcator of nutrient balance. Thls does not tell the full story because in many cases, a sipficant part of the C is not avadable and in some cases much of the N is not avadable. One example is the work reported by Nakasah et al. (1992) who looked at a mixture of rice residues, garbage and recycled compost at different C/N ratios. They found that 22.4 was the best when considering the amount of degradation during composting. Temperature profiles of the dfferent treatments were not dfferent, however, the C/N of 22.4 whch provided the most mass reduction also had maximum ammonia release in the fust 2-days. All other treatments, had less ammonia release and at a later time. These indlcate the effect of a healthy biologcal activitjr.

One approach to determining the biodegradable fractions is to use the lip content method (Kayhanian and Tchobanoglous, 1992). In t h s estimate of biodegradable C/N, it is assumed that all of the nitrogen is avadable for composting and the amount of carbon avadable is a function of the l i p content [see Appendu for calculation

value of total C/N. Therefore by operating the process to acheve total C/N close to I . information]. The biologcally avadable C/N ratio is less than the commonly used


30, the actual C/N ratio can be much lower thus contributing to ammonia and other odor emissions.

Many times as described by Smalley (1998), there is too much biosolids in a mix of biosolids-wood clups. T h ~ s results in a very h g h oxygen consumption rate dnving the composting anaerobic. W e troubleshooting their system, a change from a ratio of 1:2 to1:5 changed the odor characteristic in a remarkable way. Smalley (1998) inhcates how t h l s change along with dady oxygen measurement was incorporated into their fachty operating plan to change their process from a fermentation process to true aerobic composting.

3.2.2 Ash as an odor-control amendment

One low-cost amendment for odor control is wood ash, whch contains un-burnt carbon potentially acting as an adsorbent. Goldstein (1999) reports the work at Montgomery County composting fachty that evaluated the adlt ion of ash in biosolids composting to reduce odors and VOCs. The study showed that ammonia and total reduced sulfur compounds (TRS includmg dunethyl sulfide and dunethyl &sulfide) was captured very effectively either by using an ash filter or by covering the composting piles with a layer of wood ash. In freshly made piles covered with a layer of ash, the reduction in ammonia release was 34% and TRS was 64%. When the air from a pile of compost with no cover was passed through a filter column of ash only, the removal rates of TRS was 99.5%, suggesting that effective odor control could be aclveved using an ash fdter Fable 11).

Table 11. Effect of using wood ash as an amendment for con t rohg odorous compounds in biosolids composting.

Ammonia, ppm Total reduced sulfur compounds, ppm

Freshly mixed compost pile Pile with no cover 0.233 760 Pile with ash cover

14-day active compost pile (no cover) 63.9 34.2 Percentage reduction, Yo

0.084 500

Sample collected duectly

99.5 - Percentage reduction, Yo 0.017 - Sampled through ash filter 3.400 1300

Source: Goldstein, 1999.

Presently, our group at the University of Georga is evaluating the use of lfferent types of coal fly ash in aclveving s d a r results. The ashes vary in particle size, pH and un-burnt carbon content. The effect of these on the adsorption of lfferent odorous sulfur compounds is being measured. Prellmlnary results inlcate that amendmg 30% by dry weight with ash results in a reduction of odorous emissions as

V

1


measured using dunethyl Qsulfide [DMDS] as a model gas. Odor releases typically were hghest after 24-36 hours (Figure 8) then reducing. There is a clearly identifiable secondary peak of odor release that occurs around 72-hours. These results are consistently repeatable, however since the condttions are under ideal laboratory condrtions, the drrect translation of times of release may not occur in field conditions. Peak releases under field conQtions may occur at a delayed point.

Another example (Figure 9) shows the effect of a temporary anaerobic conchtion of 6-hours at the end of first day and at the end of the h r d day. Results inQcate that emissions double after the anaerobic phase with a greater increase at the h r d day compared to the first. The odor-reducing effect of ash is also more pronounced on the h r d day compared to the first. It should be noted that these are p r e h a r y results and the behavior seen is very specific to the type of ash used. Ash characteristics vary sipficantly in carbon content, particle size, surface area, pH and salt content. It is very hkely that other ashes may behave in a Qfferent way. The characterization of these behaviors is presently under way.

AdQtion of ash may have some undesirable effects because of its h g h pH and salt contents. One study showed that ash Qd not sipficantly decrease degradation rates during composting, however the frnal product had hgher pH and salt content (EC 8 dS/m in ash amended treatment compared to unamended EC = 5 dS/m). There was a reduction in maturity as measured by seed germination in a greenhouse study (Campbell et al., 1997) possibly a result of the hgher salt content.

Figure 8. Release of dunethyl &sulfide (DMDS) a model odor compound during the initial stages of biosolids composting. Effect of ash amendment on odor release.

10.0

ao

40

20

ao


Figure 9. Impact of 6-hours of complete anaerobic condtions on the increase in odor release from composting. Effect of ash amendment on odor release.

Y.""

20.00

E 15.00 :: - H 10.00 n

5.00

0.00 24 78

3.2.3 Other amendments

A bench-scale research on using h e , peat moss, sulfimc acid, elemental sulfur and other amendments in con t rohg odors from hog manure was reported by Al-Kanani et al. (1992a; 1992b). They found that among options evaluated, addmg 4% by weight of peat moss reduced the emission of odorous sulfur compounds. The mechanisms were not explored, but it is most lrkely a result of absorption by peat, whch has a very hlgh surface area and water holdmg capacity.

S d a r results were seen in ammonia releases. The 4%-peat moss addtion reduced cumulative ammonia emissions over 15 days from a level of 627 down to 6 m g ~ ~ 3 / k g l u l ~ ~ ~ ~ , a 99.4% reduction. S d a r amounts of reduction were acheved by addtion of mono-calciumphosphate monohydrate or phosphoric acid both of which reduced pH down to 5 or below. Amendment of pH fachtates sipficant NH3 conservation, however, with peat moss, reduction of over 90% was acheved even at a 6.2 pH (AI-Kanani et al., 1992b).

3.2.4 Use of biological odor control additives

Several products are avdable in the market that claim the abhty to suppress the production or release of odors in composting. The author has not tested these products and do not know of their effectiveness. However, d u s is not to say that they are not effective. Goldstein (1994) reported two cases: First, a micronutrient supplement that enhanced biologcal activity thereby reducing odors. Scott's Composting in Ontario used the product whde worlung with a hghly anaerobic

c


stockpile of food wastes. They report qualitative results stating that the product when sprayed per the manufacturers drrections resulted in the suppression of odors. Second, an enzyme based deodorizing product was used by two lfferent biosolids composting operations in Connecticut and Massachusetts. Both reported benefits of using the product based on qualitative evaluations.

Several chemical and biologcal adltives to reduce odors in manure pits have been tested by researchers at NC State University and Iowa State University (Bundy and Hoff, 1998). Their standard test method simulates a lagoon that is loaded with liquid manure dady. The test is performed for 35 days whde ammonia, hydrogen sulfide, and odor concentration (TI/") are monitored once a week. Their results show that several of the adltives tested provide odor reduction. In some cases as hgh as 89% of the odor was reduced. S d a r test results for composting odors are not ready avadable. Without evidence from controlled s tules in composting, it is lfficult to evaluate whether these products are effective in composting or if the testimonial evidence heard from operators is largely from maskmg or placebo effects.

3.2.5 Influence of the size and shaDe of comDostinP Diles

Windrows for composting can be made in lfferent sizes and configuration based on the type of turning equipment used. Dependmg on the type of windrow, lfferent levels of natural aeration occurs P c h e l , 1999). In windrows that are small [4-5 ft tall, 14 ft base] natural vendation is hgh and uniform providmg interstitial oxygen concentrations to be uniformly in the range of two to five percent. In large windrows [lo ft tall, 30 ft base], natural vendation was very non-uniform providmg interstitial oxygen concentrations range from one in some places to 18% in others. In trapezoidal composting systems [slab of continuous compost, 10 ft hgh], natural vendation was very poor with interstitial oxygen concentrations measured to be as low as 0.1% or below. Ahchel(l996, 1999) found that providmg oxygen in the level required for microbial activity resulted in hgh level of decomposition of substrate and a compost of good quality pow volade fatty acids and h g h maturity]. However, if oxygen was h t i n g , fatty acid generation was hgh leadmg to odorous compost. Another comparison of two fachties one using 4-m h g h piles and the second using 3-m h g h piles showed that measured odors from the composting area were 4.5 times k h e r in the larger pile system (Homans and Fisher, 1992).

These inlcate that piles should be sized based on the activity of the feedstock and care should be taken to ensure that sufficient porosity is present to provide natural aeration matchmg the needs of the system.

3.2.6 Influence of turnin? on the rate of compostin? and odor production

Comparing identical feedstocks in windrows where one treatment was turned seven times a month and a second treatment was turned once a month, no dtfference was


seen in the amount of decomposition or the quality of the final compost. If sufficient aeration was provided, by ensuring porosity in the piles, both frequencies of turning resulted in s d a r quality compost with s d a r odor potential (Mchel, 1996).

The main benefits of turning are to (1) homogenize the materials, (2) reduce moisture, (3) reduce particle size and (4) ensure pathogen reduction. If these objectives are met with turning once a month, then thls is the recommended approach. Excessive turning can result in sigruficant particle reduction thereby reducing porosity and &biting natural aeration. In addtion turning releases odors and the cost of turning cannot be justified if compost quality cannot be improved.

Since turning of pile results in a sigruficant part of the release of odors, Bidhgmaier (1992) concludes that it is best to start the first turning after two weeks of composting. Ths practice can result in reduction of total emissions by over 60%. It is also extremely critical to turn so that the pile maintains sufficient porosity after turning. Therefore use turning equipment that do not pulverize the material and

‘ensure that there is sufficient b u h g agent to maintain porosity.

[Note: the exception to these recommendation is when composting biosolids, the EPA 503 regulations requires turned windrows to be tumed 5 times during a window of 15 days when the temperature is above 55OCI.

3.2.7 WorkinP with hiPh moisture feedstocks

Food wastes typically have very hgh moisture content and most of h s water is in the form of temporanly bound water. Thrs means that once the material b e p s cornposting and breaksdown (1-2 days), the water is released and fills up the pore spaces resulting in anaerobic condtions. A practical strategy that worked for Scott’s Composting Farm, Ontario (Goldstein, 1994) was the establishment of a bed of hghly absorbing amendments into whch the food wastes are placed before mixing. These absorbents helped keep the water out of the pore spaces thus reducing anaerobic condtions.

3.3 Other options and topics in odor manapement

3.3.1 Site selection for odor manapement success

In my experience, siting of a fachty has been one of cornposter’s best friend or worst enemy. It is clear that the more “out of sight” and out of path the faclltty is, the less llkely it is to have odor complaints. However, choice is most often not dnven by the need to ensure minimal impact on neighbors. Choice of a site to establish a facllrty is usually driven by other issues, such as: Convenient access, Already owned property,

. Ease of permitting, Location relative to feedstocks, etc.


Time and a p , designers of fachties dunk that emissions can be reduced to “zero” with engmeered air handlmg systems and emission treatment systems. In reality, no engmeered system of such complexity works correctly 100% of the time. Odor complaints b e p , leadmg to loss of public confidence and eventually closure of the fachty.

Giggey et al. (1995) recommends the following steps whde at the site selection stage of establishmg a composting operation: 1. Have [2-31 alternate sites to choose from. 2. Develop a prehmmary design to determine approximate odor emission. This

3. Evaluate dtfferent containment and treatment technology options. 4. Conduct lspersion modelmg under &fferent worst-case weather condtions for

design would specify feedstocks, technology, throughput etc.

each potenual site.

Dispersion modelmg usually wdl allow you to pick an odor concentration at the ‘release point on site, whch would be required to keep the impact on neighbors minimal. In order to reduce the impact on neighbors, sufficient buffer distance is required to allow for the emissions to dsperse under the worst case situations. Thrs buffer &stance d depend on the type of neiphbors, tvDe of feedstock beins comoosted, volume of air &scharg.e. other odor sources in the area, Drevahnp weather conltions, type of containment structures. and t p e of odor treatment avadable.

Many states have defined setbacks as part of their permit requirements. For example, New York requires 500 ft buffer for biosolids composting fachies and Maine uses 500 ft for all composting operations (Epstein, 1997). However, case stu&es reported by Ginney et al. (1995) show that sites with minimal odor complaints had buffers in the range of 2000-3500 ft. The Lewiston-Auburn fachty in Maine composting biosolids had a buffer of 3000 ft ra lus where there were no residences. The plant is operating with no odor complaints. In contrast, the Yarmouth-Massachusetts fachty composting biosolids had 20+ residences in the 2000 ft and over 50 residences in the 3000 ft r a lus buffer. They found that release from the fachty biofilter was at 50D/T and under some weather condttions thrs reached neighbors at a concentration of 7D/T, whch resulted in complaints. There were on average about one complaint a month. The Tri-town facllrty in Massachusetts described had the smallest buffer with 37 nelghbors in the 1000 ft buffer and over 100 in the 2000 ft buffer. In addttion the closest neighbor was less than 200 ft away from the fence-lme. T h ~ s fachty composting biosolids operated less than one year after whch it was closed because of odor complaints.


3.3.2 Barriers to prevent transDort

The presence of barriers changes the wind flow patters around fachties. In addrtion, the adage of “out of srght, out of mind” may occur when a good tree cover is present around the fachty. Smalley (1998) reports how their composting site operated two years with no odor complaints dl a major road construction project removed much of the tree cover around the area. Thereafter odors became a sipficant part of their fachty problems.

Although there is not much quantitative data to prove the effect of a sipficant tree cover, it is strongly recommended to have a cover that keeps the site from view of neighbors. Trees also have the abhty to capture some of the dust particulates that are carriers of odors for long &stances. In addrtion, trees provide noise suppression and an aesthetic appeal that would benefit the operation.

3.3.3 ODerational and ContinPencv Dlans

Every fachty should have a written plan that describes how to operate the process, what to monitor, how often and when the product is ready. In addrtion to an operating plan, each fachty should have a written document that outlrnes what to do when a contingency occurs. Smalley (1998) describes how during regular operations, one day a pile drd not reach target temperatures. Without much thought, the operators blended it with the neighboring pile at a hgher temperature, anticipating that the ddution effect would make both piles reach temperature. What resulted was that both piles now had too much biosolids in them and began to turn anaerobic resulting in a large amount of materials causing odors. It took the fachty three days to deal with the piles, by whch time the neighboring community was greatly angered. Such incidents result in major loss of public confidence and great amount of resources being spent in worlung with the community. In order to avoid such situations, a contingency plan was put in place, whch ensured that all potentially “out of specification” piles were transported off-site in a period of eight hours. T h ~ s does not occur often, however when it does, operators are now in a position to respond to the contingency quickly and effectively.

Operational plans outlrne storage pile [raw materials and product] management practices, delivery times for new feedstocks, mixing times and methods, etc. Using weather monitoring and pile oxygen/temperature monitoring, the fachty operator can decide on whch part of the day is the most suitable for creating new piles or turning composting piles.

1


5 . REFERENCES

Al-Kanani, T., E. Akoch, A.F. MacKenzie, I. A h and S. Barrington. 1992a. Odor control in liquid hog manure by added amendments and aeration. Journal of Environmental Quality, 21,704-708.

Al-Kanani, T., E. Akochl, A.F. MacKenzie, I. A h and S. Barrington. 1992b. Organic and inorganic amendments to reduce ammonia losses from liquid hog manure. Journal of Environmental Quality, 21,709-715.

Alpert, J.E. and N.T. Wu. 1997. Odor modehg as a tool in site planning. BioCycle, 38(10):75-80.

ASTM E-679-91. Standard practice for determination of odor and taste threshold by a forced-choice ascendmg concentration series method of h t s (1997). American Society for Testing and Materials. 1916, Race St., Phdadelpha, PA 19103.

Atlas, R.M. and R. Bartha. 1998. Microbial ecology: Fundamentals and applications, 4th edttion. Addtson Wesley Longman, Inc., Menlo Park, California.

Bibgmaier , W. 1992. Emissions from composting plants. Acta Horticulturae, 302, 19-27.

Bohn, H. and R. B o b . 1987. Biofiltration of odors from food and waste processing. Proceedmgs of food processing waste conference, Georga Tech Research Institute, 1-2 Sept., 1987.

Bundy, D.S. and S.J. Hoff. 1998. The testing procedures and results of pit addtives tested at Iowa State University. In: Animal production systems and the environment. Proceedmgs of the international conference on odor, water quality, nutrient management and socioeconomic issues. Des Moines, Iowa, July 19,1998

Campbell, A.G., R.L. Folk and R.R. Tripepi. 1997. Wood ash as an amendment in municipal sewage sludge and yard waste composting processes. Compost Science and Utthzation, 5(1):62-73.

Dravnieks, A. 1980. Odor threshold measurement by dynamic olfactometry: Sipficant operational variables. Journal of h Pollution Control Association, 30(12):1284.

Dravnieks, A. 1979. Odors from stationary and mobile sources. National Academy of Science Publications, Washmgton DC.


Eitzer, B.D. 1995. Emissions of volattle orgamc chemicals from municipal solid waste composting fachties. Environmental Science and Technology, 29(4):896-902.

Feldmann, T.L. 1998. Evaluation of microsource S and mictrotreat luckoff in reducing odor from an anaerobic lagoon. In: Animal production systems and the environment. Proceedmgs of the international conference on odor, water quality, nutrient management and socioeconomic issues. Des Moines, Iowa, July 19,1998.

Glggey, M., J.R. Pinnette and C.A. Dwinal. 1995. Odor control factors in compost site selection. BioCycle, 36(2).

Goldstein, J. 1999. Reducing odor and VOC emission: Composting research update. BioCycle, 40(3):68-71.

Goldstein, N. 1994. Odor control progress at composting sites. BioCycle, 35(2):64- 67.

Haug, R.T. 1993. The practical handbook of compost engmeering. Lewis Publishers, CRC Press, Boca Raton, Florida.

Hellman, T.M. and F.H. Small. 1973. Characterization of odor properties of 101 petrochemicals using sensory methods. Chemical Engmeering Progress, 69,9.

Hentz, L.H., C.M. Murray, J.L. Thompson, L.L. Gasner and J.B. Dunson. 1992. Odor control research at the Montgomery County regonal cornposting facllrty. Water Environment Research, 64(1):13-18.

Homans, W.J. and K. Fischer. 1992. A cornposting plant as an odor source. Acta Horticulturae, 302,37-44.

Iacoboni, M.D., J.R. Livingston, and T.J. LeBrun. 1984. Windrow and static pile composting of municipal sewage sludges. US EPA Report number EPA-600/2-84- 122.

Kayhanian and Tchobanoglous. 1992. Computation of C/N ratios of various organic fractions. BioCycle, 33(5):58-60.

Kuroda, K., T. Osada, M. Yonaga, A. Kanematu, T. Nitta, S. Mouri, and T. Kojima. 1996. Emissions of malodorous compounds and greenhouse gases from composting swine feces. Bioresource Technology, 56,265-271.

Kuter, G.A., 1990. Odor control - completing the composting process. International - process systems, Inc., 655, Windmg Brook Dr., Glastonbury, CT 06033.

Y

1


Leson, G. and A.M. Winer. 1991. Biofiltration: An innovative air pollution control technology for VOC emissions. J. Au Waste Manage. Assoc., 41:8, 1045-1054

McGinley, D.L. and M.A. McGmley. 1997. Understandmg odor panels and odor evalution. Paper presented at the Au and Waste Management Association annual meeting, Toronto, Canada. June 8-13, 1997.

hhchel, F.C., L.J. Forney, A.J.F. Huang, S. Drew, M. Czuprensh, J.D. Lindeberg and C.A. Reddy. 1996. Effects of turning frequency, leaves to grass mix ratio and windrow vs. Pile configuration on the compostng of yard trimmings. Compost Science and Uthation, 4(1):26-43.

hhchel, F.C. 1999. M a n a p g compost piles to maximize natural aeraiton. BioCycle, 40(3):56-58.

Mycock, J.C., J.D. McKenna, and L. Theodore. 1995. Handbook of air pollution control enpeering and technology. CRC Press, Lewis Publishers, Boca Raton, Florida.

Nakasah, K., H. Yaguch, Y. Sasah, and H. Kubota. 1992. Effect of C/N ratio on thermophhc composting of garbage. Journal of Fermentation and Bioengmeering, 73(1):43-45

O’Brien, M.A. 1992. Guideltnes for odor sampltng and measurement by dynamic dut ion olfactometry. Draft method: AWMA EE-6 Subcommittee on the Standardnation of Odor measurement.

Ottengraf, S.P.P and R.M.M. Drks. 1992. Process technology of biotechmques. In: Biotechniques for air pollution abatement and odor control policies. Dragt, AJ. and J. van Ham (Eds.). Elsevier Science Publishers, The Netherlands

Ottengraf, S.P.P. 1986. Exhaust gas purification. In Biotechnology, Vol8, Rehm, H.J. and G. Reed (Eds). VCH Verlagsgesellsch., Wemheim, Germany.

Smalley, C. 1998. Hard earned lessons on odor management. BioCycle, 39(1):58-61

Smet, E., H. Van Langenhove, and In. De Bo. 1999. The emission of volatde compounds during the aerobic and the combined anaerobic/aerobic composting of biowaste. Atmospheric Environment, 33, 1295-1303.

Tang, H-M., S-J. Hwang, and S-C. Hwang. 1996. Waste gas treatment in biofilters. Joumal of air and waste management association, 46, 349-354.


Valentin, F.H.H. and A.A. North. 1980. Odor control: a concise gurde. Warren Spring Laboratory, Department of Industry, Hertfordshtre, UK.

Van Durme, G.P., B.F. McNamara, and C.M. McGdey. 1992. Bench-scale removal of odor and vo1at.de organic compounds at a composting fachty. Water Environment Research, 64(1):19-27.

Van Durme, G.P., B.F. McNamara, and C.M. McGdey. 1990. Characterization and treatment of composting emissions at Hampton Roads Sanitation District. Paper presented at the 63rd Annual Water Pollution Control Federation Conference, Washmgton, DC.

Von Fahnestock, M., 1995. Personal communications.

Water Environment Federation and American Society of Civil Enpeering. 1995. * Odor control in wastewater treatment plants. WEF Manual of practice No. 22.

The following are web addresses for adhtional infonnation and resources:

Source for EPA Methods: htto://wnw.eDa.pov/ttn/amtic/airtox.html httD://www.eDa.Pov/ttn/emc/Drommte.html

Source for ASTM Methods: httD: / /astm.micronexx.com/ [follow your way to ASTM store and Standard methods]

c

http://vo1at.de

http://astm.micronexx.com


6. APPENDICES

6.1 Conversion of pas concentration from mass basis to volume basis

Sometimes the mass basis [mg/m3 or pg/mfl concentration is provided. In order to convert this value to ppm[v], we need to know the density of the gas. It is common practice to perform the calculation at a standard condttion of pressure and temperature. The pressure chosen is latm and temperature is 20C.

1 mg/m3 = 1 ppmv X Density [mg/m3, g/L]

The density of a gas in mg/m3 or g/L is calculated as: Density = Molecular weight/ 24 [at 1 atm and 20 C] Density = Molecular weight/ 22.4 [at 1 atm and 0 C]

Examole:

If the concentration of Hydrogen sulfide in an air stream is measured to be 10 ppmv, h s is equal to 10 parts hydrogen sulfide per 1 d o n parts of air:

IO ppmv = IO m3/1,000,000 m3 = IO c m 3 ~ 2 ~ ] / m 3 [ ~ ]

Density of hydrogen sulfide at 20 C = 34 /24 = 1.4167 gv,S]/L[H,S] = 1.4167 mgv2Sl/cm3v,Sl

Therefore,

mg/m3 = 10 ppmv X 1.4167 mg[H,S]/cm3[H,S] = 14.167 mg/m3

pg/m3 = 14.167 X 1000 pg/mg = 14,167

6.2 Calculation of biodemadable C/N ratio.

True C/N ratio is based on biodegradable-C and N whch are dlfficult parameters to estimate. One good method suggested by Kayhanian and Tchobanoglous (1992) is the use of lip content. T h ~ s calculation dustrates that true C/N is lower than total C/N [the commonly measured ratio]. If h s is not taken into account, too much nitrogen could be added to the process thus contributing to odors.

Normal method of calculating C/N ratio, e.g. Food waste:

C Total Carbon 50.00 N Total Nitrogen 3.20 " -" - - - 15.6


In calculating C/N ratio based on biodegradable fractions, it is assumed that in most waste streams, the nitrogen is completely avdable, but the carbon is not. The avadabhty of carbon is a function of the amount of l i p in the feedstock.

Table 12. Selected properties of two common feedstocks'.

1 All units are Yo of dry matter with the exception of LC LC is lignm content, Yo of volade solids; 3 Volade solids, O/O = 100- Ash

Step 1: Calculation of biodegradable fraction

BF = 0.83 - (0.028)~ LC BF = 0.83 - (0.028) X 0.40 = 0.81 18

Step 2: Calculate the volade solids fraction

3.16 100

VS,,,,, = 1 - - - - 0.9684

Step 3: Calculate biodegradable carbon content

Biodegradable - C = Total Carbon x BF x VSf,,,, Biodegradable - C = 50.00 x 0.8188 x 0.9684 = 39.65

Step 4 Biodegradable-C/N

39.65 3.20

TrueC/N = - - - 12.40

Recognizing that the true C/N is lower than total C/N will prevent operators from adding too much N into the composting process. This adjustment of nutrient balance can reduce ammonia and other odor emissions. The table below summarizes and compares calculated values of total and True-C/N for food waste and yard waste.

Table 13. Comparison of total C/N and biodegradable C/N ratios. Feedstock [C/r\rlBiodegradable [C/r\rlTonl BFhction VSfnction

- 12.4 15.6 0.8188 0.9684 Food waste I Yard waste I 0.8880 I 0.7152 I 22.9 I 14.5 I

I E544-99 Standard Practices for Referencing Suprathreshold Odor I n t e ~ l ~tm.micronexx.com/cgi-bin/SoftCa...E.CART/PAGES/E544.htm?L+mystore+wykd30.

E544-99 Standard Practices for Referencing Suprathreshold Odor Intensity Copyright 2000 AMERICAN SOCIETY FOR TESTING AND MATERIALS, West Conshohocken, PA. All rights reserved.

1. Scope 1.1 These practices are designed to outline a preferred means for referencing the odor intensities of a material in the suprathreshold region.

1.2 The general objective is to reference the odor intensity rather than other odor properties of a sample.

1.3 These practices are designed to reference the odor intensity on the ASTM Odor Intensity Referencing Scale ofany odorous material. This is done by a comparison of the odor intensity of the sample to the odor intensities of a series of concentrations of the reference odorant, which is 1-butanol ( -butanol).

1.4 The method by which the reference odorant vapors are to be presented for evaluation by the panelists is specified. The manner by which the test sample is presented will depend on the nature of the sample, and is not defined herein.

1.5 Test sample presentation should be consistent with good standard practice (1) and should be explicitly documented in the test report.

L

1.6 This standard may involve hazardous materials, operations, and equipment. This standard does not purport to address all of the safety problems associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.

Adopted by: Developed by ASTM Subcommittee: E18.04

Ordering Information

Price: $35.00 Printed Pages: 11

The information above is only a s u m m q of the ASTM standard. Order the complete standard in three ways:

1 of2 9/29/00 8:50 PM

E679-91( 1997) Standard Practice for De te... ng Concentration Series M ~ ~ I . o r g l c g i - b i n l S o f t C a r t . e x e ... E.CARTIPAGESIE679.htm?l+mystore+bfoq70: I

E679-91(1997) Standard Practice for Determination of Odor and Taste Thresholds By a Forced-Choice Ascending Concentration Series Method of Limits Copyright 2000 AMERICAN SOCIETY FOR TESTING AND MATERIALS, West Conshohocken, PA. All rights reserved.

1. Scope 1.1 This practice describes a rapid test for determining sensory thresholds of any substance in any medium.

1.2 It prescribes an overall design of sample preparation and a procedure for calculating the results.

1.3 The threshold may be characterized as being either (a) only detection (awareness) that a very small amount of added substance is present but not necessarily recognizable, or (b) recognition of the nature of the added substance.

.

1.4 The medium may be a gas, such as air, a liquid, such as water or some beverage, or a solid form of matter. The medium may be odorless or tasteless, or may exhibit a characteristic odor or taste per se.

1.5 This practice describes the use of a multiple forced-choice sample presentation method in an ascending concentration series, similar to the method of limits.

1.6 Physical methods of sample presentation for threshold determination are not a part of this practice, and will depend on the physical state, size, shape, availability, and other properties of the samples.

1.7 It is recognized that the degree of training received by a panel with a particular substance may have a profound influence on the threshold obtained with that substance (1).

1.8 Thresholds determined by using one physical method of presentation are not necessarily equivalent to values obtained by another method.

Adopted by: Developed by ASTM Subcommittee: E18.04

Ordering Information

Price: $30.00 Printed Pages: 5

The information above is only a summary of the ASTM standard. Order the complete standard in three ways:

1 of2 -

10/7/00 1O:Ol AM

I,

Compendium of Methods for the Determination of Toxic

Organic Compounds in Ambient Air

Second Edition

Compendium Method TO-14A

Determination Of Volatile Organic Compounds (VOCs) In Ambient Air Using

Specially Prepared Canisters With Subsequent Analysis By Gas

Chromatography

Center for Environmental Research Information Office of Research and Development

U.S. Environmental Protection Agency Cincinnati, OH 45268

January 1999

Method TO- 14A Acknowledgements

This Method was prepared for publication in the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, Second Edition, (EPA/625/R-96/01Ob), which was prepared under Contract No. 68-C3-03 15, WA No. 3-10, by Midwest Research Institute (MRI), as a subcontractor to Eastern Research Group, Inc. (ERG), and under the sponsorship of the U.S. Environmental Protection Agency (EPA). Justice A. Manning, John 0. Burckle, and Scott Hedges, Center for Environmental Research Information (CERI), and Frank F. McElroy, National Exposure Research Laboratory (NERL), all in the EPA Office of Research and Development, were responsible for overseeing the preparation of this method. Additional support was provided by other members of the Compenda Workgroup, whch include:

. .

. . . . * .

John 0. Burckle, U.S. EPA, ORD, Cincinnati, OH James L. Cheney, Corps of Engineers, Omaha, NB Michael Davis, US. EPA, Region 7, KC, KS Joseph B. Ellcins Jr., U.S. EPA, OAQPS, RTP, NC Robert G. Lewis, U.S. EPA, NERL, RTP, NC Justice A. Manning, U.S. EPA, ORD, Cincinnati, OH William A. McClenny, US. EPA, NERL, RTP, NC Frank F. McElroy, U.S. EPA, NERL, RTP, NC Heidi Schultz, ERG, Lexington, MA William T. "Jerry" Winberry, Jr., EnviroTech Solutions, Cary, NC

Method TO-14 was originally published in March of 1989 as one of a series of peer reviewed methods in the second supplement to "Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, I' EPA 600/4-89-018. Method TO-14 has been revised and updated as Method TO-14A in this Compendium to eliminate time sensitivity material and correct a small number of errors.

Peer Reviewer

Lauren Drees, U.S. EPA, NRMRL, Cincinnati, OH

Finally, recognition is gven to Frances Beyer, Lynn Kaufman, Debbie Bond, Cathy Whitaker, and Kathy Johnson of Midwest Research Institute's Administrative Services staff whose dedcation and persistence during the development of this manuscript has enabled it's production.

DISCLAIMER

This Compendium has been subjected to the Agency's peer and administrative review, and it has been approved f i r publication as an €PA document Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

11

METHOD TO-14A

Determination Of Volatile Organic Compounds (VOCs) In Ambient Air Using Specially Prepared Canisters With Subsequent Analysis By Gas Chromatography

TABLE OF CONTENTS

1 . Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-1

2 . SummaryofMethod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-1

3 . Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-4

4 . Applicable Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-5 4.1 ASTMStandards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-5 4.2 EPADocuments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-5 4.3 OtherDocuments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-6

5 . Defintions 14A-6

6 . Interferences and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-7

7 . Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-7 'b. 7.1 Sample Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-7

7.2 Sample Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-9 7.3 Canister Cleaning System .................................................. 14A- 1 1 7.4 Calibration System and Manifold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 1 1

8 . Reagents and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 Gas Cylinders of Helium, Hydrogen, Nitrogen. and Zero Air ....................... 8.2 Gas Calibration Standards .................................................. 8.3 Cryogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4 Gas Purifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 DeionizedWater . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.6 CBromofluorobenzene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 Hexane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.8 Methanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

14A- 1 1 14A- 1 1 14A- 1 1 14A-12 14A-12 14A-12 14A-12 14A-12 14A-12

9 . Samplingsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-12

9.1.1 Subatmospheric Pressure Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 12 9.1 System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 12

9.1.2 Pressurized Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 13 9.1.3 All Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-13

9.2 Sampling Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 14

. iii

3

TABLE OF CONTENTS (continued)

10 . AnalyhcalSystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-15 10.1 System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 16 10.2 GC/MS/SCAN/SIM System Performance Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A- 19 10.3 GC/FID/ECD System Performance Criteria (With Optional PID System) . . . . . . . . . . . 14A-20 10.4 Analytical Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-22

11 . Cleaning and Certification Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-25 11.1 Canister Cleaning and Certification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-25 11.2 Sampling System Cleaning and Certification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-26

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 . Performance Criteria and Quality Assurance 14A-27 12.1 Standard Operating Procedures (SOPS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-27 12.2 Method Relative Accuracy and Linearity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-27 12.3 Method Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-28 12.4 Method Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-29 12.5 Quality Assurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-29

. Acknowledgements 14A-30 13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

14 . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14A-32

APPENDIX A . Availability d V 0 C Standards From U . S . Environmental Protection Agency (USEPA) APPENDIX B . Operating Procedures for a Portable Gas Chromatograph Equipped with a

APPENDIX C . Installation and Operating Procedures for Alternative Air Toxics Samplers Photoionization Detector

iv

METHOD TO-14A

Determination Of Volatile Organic Compounds (VOCs) In Ambient Air Using Specially Prepared Canisters With Subsequent Analysis By Gas Chromatography

1. Scope

1.1 This document describes a procedure for sampling and analysis of volatile organic compounds (VOCs) in ambient air. The method was originally based on collection of whole air samples in SUMMA@ passivated stainless steel canisters, but has now been generalized to other specially prepared canisters (see Section 7.1.1.2). The VOCs are separated by gas chromatography and measured by a mass spectrometer or by multidetector techniques. This method presents procedures for sampling into canisters to final pressures both above and below atmospheric pressure (respectively referred to as pressurized and subatmospheric pressure sampling).

1.2 This method is applicable to specific VOCs that have been tested and determined to be stable when stored in pressurized and sub-atmospheric pressure canisters. Numerous compounds, many of wluch are chlorinated VOCs, have been successfully tested for storage stability in pressurized canisters (1 -3). However, minimal documentation is currently available demonstrating stability of VOCs in subatmospheric pressure canisters.

13 The Compendium Method TO-14A target list is shown in Table 1. These compounds have been successfully stored in canisters and measured at the parts per billion by volume (ppbv) level. This method applies under most conditions encountered in sampling of ambient air into canisters. However, the composition of a gas mixture in a canister, under unique or unusual conditions, will change so that the sample is known not to be a true

' representation of the ambient air from which it was taken. For example, low humidity conditions in the sample may lead to losses of certain VOCs on the canister walls, losses that would not happen if the humidity were

fractional losses of water-soluble compounds. Since the canister surface area is limited, all gases are in competition for the available active sites. Hence an absolute storage stability cannot be assigned to a specific gas. Fortunately, under conditions of normal usage for sampling ambient air, most VOCs can be recovered from canisters near their original concentrations after storage times of up to h t y days.

L higher. If the canister is pressurized, then condensation of water from high humidity samples may cause

2. Summary of Method

2.1 Both subatmospheric pressure and pressurized sampling modes typically use an initially evacuated canister and pump-ventilated sample line during sample collection. Pressurized sampling requires an additional pump to provide positive pressure to the sample canister. A sample of ambient air is drawn through a sampling train comprised of components that regulate the rate and duration of sampling into a pre-evacuated specially prepared passivated canister.

2.2 After the air sample is collected, the canister valve is closed, an identification tag is attached to the canister, a chain-of-custody (COC) form completed, and the canister is transported to a predetermined laboratory for analysis.

2.3 Upon receipt at the laboratory, the canister tag data is recorded, the COC completed, and the canister is attached to the analytical system. During analysis, water vapor is reduced in the gas stream by a Nafion@ dryer (if applicable), and the VOCs are then concentrated by collection in a cryogenically-cooled trap. The cryogen is then removed and the temperature of the trap is raised. The VOCs originally collected in the trap are

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 14A-1

Method TO-14A vocs revolatilized, separated on a GC column, then detected by one or more detectors for identification and quantitation.

2.4 The analytical strategy for Compendium Method TO-14A involves using a high-resolution gas chromatograph (GC) coupled to one or more appropriate GC detectors. Historically, detectors for a GC have been divided into two groups: non-specific detectors and specific detectors. The non-specific detectors include, but are not limited to, the nitrogen-phosphorus detector (NPD), the flame ionization detector (FID), the electron capture detector (ECD) and the photo-ionization detector (PID). The specific detectors include the linear quadrupole mass spectrometer (MS) operating in either the select ion monitoring (SIM) mode or the SCAN mode, or the ion trap detector (see Compendium Method TO-15). The use of these detectors or a combination of these detectors as part of the analyhcal scheme is determined by the required specificity and sensitivity of the application. While the non-specific detectors are less expensive per analysis and in some cases far more sensitive than the specific detectors, they vary in specificity and sensitivity for a specific class of compounds. For instance, if multiple halogenated compounds are targeted, an ECD is usually chosen; if only Compounds containing nitrogen or phosphorus are of interest, a NPD can be used; or, if a variety of hydrocarbon compounds are sought, the broad response of the FID or PID is appropriate. In each of these cases, however, the specific identification of the compound within the class is determined only by its retention time, which can be subject to shifts or to interfimce from other non-targeted compounds. When misidentification occurs, the error is generally a &ult ofa cluttered chromatogtam, making peak assignment diecult. ~n particular, the more volatile organics (chloroethanes, ethyltoluenes, dichlorobenzenes, and various freons) exhibit less well defined chromatographic peaks, leading to possible misidentification when using nonspecific detectors. Quantitative comparisons indicate that the FID is more subject to error than the ECD because the ECD is a much more selective detector and exhibits a stronger response. Identification errors, however, can be reduced by: (a) employing simultaneous detection by different detectors or (b) correlating retention times from different GC columns for confirmation. In either case, interferences on the non-specific detectors can still cause error in identifjmg compounds of a complex sample. The non-specific detector system (GCMPDIFIDECDIPID), however, has been used for approximate quantitation of relatively clean samples. The non-specific detector system can provide a "snapshot" of the constituents in the sample, allowing determination of:

- Extent of misidentification due to overlapping peaks. - Determination of whether VOCs are within or not within concentration range, thus requiring further

analysis by specific detectors (GC/MS/SCAN/SIM) (i.e., if too concentrated, the sample is further diluted).

- Provide data as to the existence of unexpected peaks which require identification by specific detectors.

On the other hand, the use of specific detectors (MS coupled to a GC) allows positive compound identification, thus lending itself to more specificity than the multidetector GC. Operating in the SIM mode, the MS can readily approach the same sensitivity as the multidetector system, but its flexibility is limited. For SIM operation the MS is programmed to acquire data for a limited number of targeted compounds. In the SCAN mode, however, the MS becomes a universal detector, often detecting compounds which are not detected by the multidetector approach. The GS/MS/SCAN will provide positive identification, while the GC/MS/SIM procedure provides quantitation of a restricted list of VOCs, on a preselected target compound list (TCL).

Ifthe MS is based upon a standard ion trap design, only a scanning mode is used (note however, that the Select Ion Storage (SIS) mode of the ion trap has features of the SIM mode). See Compendium Method TO-15 for M e r explanation and applicability of the ion-trap to the analysis of VOCs from specially prepared canisters.

Page 14A-2 Compendium of Methods for Toxic Organic Air Pollutants January 1999

vocs - Method TO-14A

The analyst often must decide whether to use specific or non-specific detectors by considering such factors as project objectives, desired detection limits, equipment availability, cost and personnel capability in developing an analyhc strategy. A list of some of the advantages and disadvantages associated with non-specific and specific detectors may assist the analyst in the decision-making process.

0

0

0

'L

0

0

0

0

0

Advantanes

Somewhat lower equipment cost than GC/MS 0 Multiple detectors to calibrate Less sample volume required for analysis 0 Compound identification not positive More sensitive 0 Lengthy data interpretation' (1 hour each for - ECD may be 1000 times more sensitive than analysis and data reduction)

G C / M S 0 Interference(s) from co-eluting compound(s) 0 Cannot identify unknown compounds

- outside range of calibration - without standards

interfering compounds 0 Does not differentiate targeted compounds from

lcal Svstem

GC/MS/SIM

Advantages DlsadvantaPes -

positive compound identification 0 cannot identify nonspecified compounds (ions) greater sensitivity than GC/MS/SCAN 0 somewhat greater equipment cost than less operator interpretation than for multidetector GC multidetector GC 0 greater sample volume required than for can resolve co-eluting peaks multidetector GC more specific than the multidetector GC 0 universality of detector sacrified to achieve

enhancement in sensitivity

GC/MS/SCAN

positive compound identification lower sensitivity than GC/MS/SIM can identify all compounds 0 greater sample volume required than for less operator interpretation multidetector GC can resolve co-eluting peaks 0 somewhat greater equipment cost than

multidetector GC

The analytical finish for the measurement chosen by the analyst should provide a definitive identification and a precise quantitation of volatile organics. In a large part, the actual approach to these two objectives is subject to equipment availability. Figure 1 indicates some of the favorite options that are used in Compendium Method TO- 14A. The GC/MS/SCAN option uses a capillary column GC coupled to a MS operated in a scanning mode and supported by spectral library search routines. This option offers the nearest approximation to


Method TO-14A VOCs

unambiguous identification and covers a wide range of compounds as defined by the completeness of the spectral library. GCiMS/SIM mode is limited to a set of target compounds which are user defined and is more sensitive than GC/MS/SCAN by virtue of the longer dwell times at the restricted number of m/z values. Both these techniques, but especially the GC/MS/SIM option, can use a supplemental general nonspecific detector to venfjdidentify the presence of VOCs. Finally the option labelled GC-multidetector system uses a combination of retention time and multiple general detector verification to identify compounds. However, interference due to nearly identical retention times can affect system quantitation when using this option.

Due to low conmixations of toxic VOCs encountered in urban air (typically less than 25 ppbv and the majority below 10 ppbv) along with their complicated chromatographs, Compendium Method TO-14A strongly recommends the specific detectors (GC/MS/SCAN/SIM) for positive identification and for primary quantitation to ensure that high-quality ambient data is acquired.

For the experienced analyst whose analykal system is limited to the non-specific detectors, Section 10.3 does provide pdehes and example chromatograms showing typical retention times and calibration response factors, and utilizing the nonspecific detectors (GC/FID/ECD/PID) analyhcal system as the primary quantitative technique.

Cbmpendium Method TO-15 is now available as a guidance document containing additional advice on the monitoring of VOCs. Method TO- 15 contains information on alternative water management systems, has a more complete quality control section, shows performance criteria that any monitoring technique must achieve for acceptance, and provides guidance specifically directed at compound identification by mass spectrometry.

3. Significance

3.1 The availability of reliable, accurate and precise monitoring methods for toxic VOCs is a primary need for state and local agencies addresing daily monitoring requirements related to odor complaints, fugitive emissions, and trend monitoring. VOCs enter the atmosphere from a variety of sources, including petroleum refineries, synthetic organic chemical plants, natulal gas pn>cessing plants, biogenic sources, and automobile exhaust. Many of these VOCs are toxic so that their determination in ambient air is necessary to assess human health impacts.

3.2 The canister-based monitoring method for VOCs has proven to be a viable and widely used approach that is based on research and evaluation performed since the early 1980s. This activity has involved the testing of sample stability of VOCs in canisters and the design of time-integrative samplers. the development of procedures for analysis of samples in canisters, including the procedure for VOC preconcentration from whole air, the treatment of water vapor in the sample, and the selection of an appropriate analyhcal finish has been accomplished. The canister-based method was initially summarized by EPA as Method TO-14 in the First Supplement to the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air. The present document updates the original Compendium Method TO- 14 with correction of time-sensitive information and other minor changes as deemed appropriate.

3.3 The canister-based method is now a widely used alternative to the solid sorbent-based methods. The method has sub-ppbv detection limits for samples of typically 300-500 mL of whole air and duplicate and replicate precisions under 20 percent as determined in field tests. Audit bias values average within the range of *lO percent. These performance parameters are generally adequate for monitoring at the lo-’ lifetime exposure risk levels for many VOCs.


vocs - Method TO-14A

3.4 Collection of ambient air samples in canisters provides a number of advantages: (1) convenient integration of ambient samples over a specific time period (e.g., 24 hours); (2) remote sampling and central analysis; (3) ease of storing and shipping samples; (4) unattended sample collection; ( 5 ) analysis of samples from multiple sites with one analmcal system; (6) collection of sufficient sample volume to allow assessment of measurement precision andor analysis of samples by several analykal systems; and (7) storage stability for many VOCs over periods of up to 30 days. To realize these advantages, care must be exercised in selection, cleaning, and handling sample canisters and sampling apparatus to avoid losses or contamination.

3.5 Interior surfaces of canisters are treated by any of a number of passivation processes, one of which is S U M M A polishing as identified in the on@ Compendium Method TO- 14. Other specially prepared canisters are also available (see Section 7.1.1.2).

3.6 The canister-based method for monitoring VOCs is the alternative to the solid sorbent-based method described in conventional methods such as the Compendium Methods TO-1 and TO-2, and in the new Compendium Method TO-17 that describes the use multisorbent packings including the use of new carbon-based sorbents. It also is an alternative to on-site analysis in those cases where integrity of samples during storage and transport has been established.

4. Applicable Documents

4.1 ASTM Standards

Method D 1356 Definition of Terms Relating to Atmospheric Sampling and Analysis Method E260 Recommended Practice for General Gas Chromatography Method E355 Practice for Gas Chromatography Terms and Relationships Method D3 1357 Practice for Planning and Sampling of Ambient Atmospheres Method D5466-93 Determination of Volatile Organic Chemicals in Atmospheres (Canister Sampling Methodology)

4.2 EPA Documents

Technical Assistance Document for Sampling and Analysis Toxic Organic Compounds in Ambient Air, U. S. Environmental Protection Agency, EPA-60014-83-027, June 1983. Quality Assurance Handbook for Air Pollution Measurement Systems, U. S. Environmental Protection Agency, EPA-600/R-94-038b, May 1994. Compendium ofMethodr for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-14, Second Supplement, U. S. Environmental Protection Agency, EPA 60014-89-018, March 1989. Compendium ofMethodr for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-15, Second Edition, U. S. Environmental Protection Agency, EPA 625/R-96-010b, January 1997. Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, First Supplement, U. S. Environmental-Protection Agency, Research Triangle Park, NC, EPA-60014-87-006, September 1997. Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-I, U. S. Environmental Protection Agency, Research Triangle Park, NC, EPA-60014-84-041,1986.


Method TO-14A v o c s

4.3 Other Documents

U. S. Environmental Protection Agency Technical Assistance Document (3). Laboratory and Ambient Air Studies (4-17).

5. Definitions

m: Definitions used in this document and any user-prepared Standard Operating Procedures (SOPSs) should be consistent with those used in ASTM Dl356 All abbreviations and symbols are defined within this document at the point offirst use.]

5.1 Absolute Canister Pressure (Pg+Pabgauge pressure in the canister (@a, psi) and Pa = barometric pressure (see Section 5.2).

5.2 Absolute Pressurwmure measured with ref-ce to absolute zero pressure (as opposed to atmospheric pressure), usually expressed as kPa, mm Hg or psia.

5 j C r y o g e m refigexant used to obtain very low temperatures in the cryogenic trap of the analytical system. A typical cryogen is liquid nitrogen (bp -1958°C) or liquid argon (bp -1857°C).

5.4 Dynamic Calibration-calibration of an analytical system using calibration gas standard concentrations in a fom identical or very similar to the samples to be analyzed and by introducing such standards into the inlet of the sampling or analyhcal system in a manner very similar to the normal sampling or analytical process.

5.5 Gauge Pressurqressure measured above ambient atmospheric pressure (as opposed to absolute pressure). Zero gauge pressure is equal to ambient atmospheric (barometric) pressure.

5.6 Ms/SCAN-the GC is coupled to a MS programmed in the SCAN mode to scan all ions repeatedly during the GC run. As used in the current context, this procedure serves as a qualitative identification and characterization of the sample.

5.7 MS/SI"the GC is coupled to a MS programmed to acquire data for only specified ions and to disregard all others. This is performed using SIM coupled to retention time discriminators. The GC/SIM analysis provides quantitative results for selected constituents of the sample gas as programmed by the user.

5.8 Megabore@ Column-chromatographic column having an internal diameter (I.D.) greater than 0.50-mm. The Megabor& column is a trademark of the J&W Scientific Co. For purposes of this method, Megabor& refers to chromatographic columns with 0.53-mm I.D.

5.9 Pressurized Sampling-collection of an air sample in a canister with a (final) canister pressure,above atmospheric pressure, using a sample pump.

5.10 Qualitative Accuracy-the ability of an analyhcal system to correctly identify compounds.

5.11 Quantitative Accuracy-the ability of an analytxal system to correctly measure the Concentration Of an identified compound.


v o c s - Method TO-14A

5.12 Static Calibration-calibration of an analytical system using standards in a form different from the samples to be analyzed. An example of a static calibration would be injecting a small volume of a high concentration standard directly onto a GC column, bypassing the sample extraction and preconcentration portion of the analytical system.

5.13 Subatmospheric Sampling-collection of an air sample in an evacuated canister at a (final) canister pressure below atmospheric pressure, without the assistance of a sampling pump. The canister is filled as the internal canister pressure increases to ambient or near ambient pressure. An auxiliary vacuum pump may be used as part of the sampling system to flush the inlet tubing prior to or during sample collection.

6. Interferences and Limitations

6.1 Interferences can occur in sample analysis if moisture accumulates in the dryer (see Section 10.1.1.2). An automated cleanup procedure that periodically heats the dryer to about 100°C whle purging with zero air eliminates any moisture buildup. This procedure does not degrade sample integrity for Compendium Method TO-14A target compound list (TCL) but can affect some organic compounds.

6.2 Contamination may occur in the sampling system if canisters are not properly cleaned before use. Additionally, all other sampling equipment (e.g., pump and flow controllers) should be thoroughly cleaned to ensure that the filling apparatus will not contaminate samples. Instructions for cleaning the canisters and certifymg the field sampling system are described in Sections 1 1.1 and 1 1.2, respectively.

'I." 6.3 The Compendium Method TO-14A analytical system employs a N&on@ permeable membrane dryer to remove water vapor from the sample stream. Polar organic compounds permeate this membrane in a manner similar to water vapor and rearrangements can occur in some hydrocarbons due to the acid nature of the dryer. Compendium Method TO-15 provides guidance associated with alternative water management systems applicable to the analysis of a large group of VOCs in specially-treated canisters.

7. Apparatus

/&&: Equipment manufacturers identifed in this section were originally published in Compendium Method TO-14 as possible sources of equipment. They are repeated in Compendium Method TO-14A as reference only. Other manufacturers'equipment should work as well, as long as the equipment is equivalent. Modifications to these procedures may be necessary if using other manufacturers' equipment.]

7.1 Sample Collection

/&&: Subatmospheric pressure andpressurized canister sampling systems are commercially available and have been used as part of US. Enviroqmental Protection Agency's Toxic Air Monitoring Stations (TAMS). Urban Air Toxic Monitoring Program (UATW), the non-methane organic compound (NMOC) Sampling and Analysis Program, and in the Photochemical Assessment Monitoring Stations (PAMS).]

7.1.1 Subatmospheric Pressure (see Figure 2 Without Metal Bellows Type Pump). 7.1.1.1 Sampling Inlet Line. Stainless steel tubing to connect the sampler to the sample inlet.


Method TO-14A v o c s

7.1.1.2 Specially-Treated Sample Canister. Leak-free stadess steel pressure vessels of desired volume (e.g., 6 L), with valve and passivated interior surfaces. Major manufacturers and re-suppliers are:

BRC/Ramussen 17010 NW Skyline Blvd. Portland, OR 9732 1

1790 Potrero Drive San Jose, CA 95 124 Restec Corporation 1 10 Benner Circle Bellefonte, PA 16823-8812

Meriter

XonTech Inc. 6862 Hayenhurst Avenue Van Nuys, CA 9 1406

P.O. Box 8941 Moscow, ID 83843

500 Technology Ct. Smyma, GA 30832

Scientific Instrumentation Specialists

Graseby

7.1.1.3 Stainless Steel Vacuum/Pressure Gauge. Capable of measuring vacuum (-100 to 0 kPa or 0 to 30 in. Hg) and pressure (0-206 kPa or 0-30 psig) in the sampling system, Matheson, P.O. Box 136, MOITOW, GA 30200, Model 63-3704, or equivalent. Gauges should be tested clean and leak tight.

7.1.1.4 Electronic Mass Flow Controller. Capable of maintaining a constant flow rate (zk 10%) over a .sampling period of up to 24 hours and under conditions of changing temperature (2040°C) and humidity, Tylan Corp., 19220 S. Normandie Ave., Tomme, CA 90502, Model FC-260, or equivalent.

7.1.1.5 Particulate Matter Filter. 2-pm sintered stainless steel in-line filter, Nupro Co., 4800 E. 345th St., Willoughby, OH 44094, Model SS-2F-K4-2, or equivalent.

7.1.1.6 Electronic Timer. For unattended sample collection, Paragon Elect. Co., 606 Parkway Blvd., P.O. Box 28, Twin Rivers, WI 54201, Model 7008-00, or equivalent.

7.1.1.7 Solenoid Valve. Electrically-operated, bi-stable solenoid valve, Skinner Magnelatch Valve, New Britain, CT, Model V5RAM49710, with VitonB seat and O-rings. A Skinner Magnelatch valve is used for purposes of illustmtion only in Figures 2 and 3.

7.1.1.8 Chromatographic Grade Stainless Steel Tubing and Fittings. For interconnections, Alltech Associates, 205 1 Waukegan Rd., Deerfield, IL 60015, Cat. #8125, or equivalent. All such materials in contact with sample, analyte, and support gases prior to analysis should be chromatographic grade stainless steel.

7.1.1.9 Thermostatically Controlled Heater. To maintain temperature inside insulated sampler enclosure above ambient temperature, Watlow Co., Pfafftown, NC, Part 04010080, or equivalent.

7.1.1.10 Heater Thermostat. Automatically regulates heater temperature, Elmwood Sensors, Inc., 500 Narragansett Park Dr., Pawtucket, RI 0286 1, Model 3455-RC-0 100-0222, or equivalent.

7.1.1.11 Fan. For cooling sample system, EG&G Rotron, Woodstock, N Y , Model SUZAI, or equivalent. 7.1.1.12 Fan Thermostat. Automatically regulates fim operation, Elmwood Sensors, Inc., Pawtucket, RI,

7.1.1.13 Maxbnum-Minimum Thermometer. Records highest and lowest temperatures during sampling

7.1.1.14 Stainless Steel Shut-Off Valve. Leak free, for vacuum/pressure gauge. 7.1.1.15 Auxiliary Vacuum Pump. Continuously draws ambient air through the inlet manifold at 10

Umin. or higher flow rate. Sample is extracted from the manifold at a lower rate, and excess air is exhausted.

Model 3455-RC-0 100-0244, or equivalent.

period, Thomas Scientific, Brooklyn Thermometer Co., Inc., P/N 9327H30, or equivalent.

[&&: The use of higher inletjlow rates dilutes any contamination present in the inlet and reduces rhe possibility of sample contamination as a result of contact with active adsorption sites on inlet walls.]


Compendium of Methods for the Determination of Toxic

Organic Compounds in Ambient Air

Second Edition

Compendium Method TO47

Determination of Volatile Organic Compounds in Ambient Air Using Active

Sampling Onto Sorbent Tubes

Center for Environmental Research Information Office of Research and Development

U.S. Environmental Protection Agency Cincinnati, OH 45268

January 1999

Method TO- 17 Acknowledgements

Ths Method was prepared for publication in the Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, Second Edition (EPA/625/R-96/01Ob), which was prepared under Contract No. 68-C3-03 15, WA No. 3-10, by Midwest Research Institute (MRI), as a subcontractor to Eastern Research Group, Inc. (ERG), and under the sponsorship of the U. S. Environmental Protection Agency (EPA). Justice A. Manning, John 0. Burckle, and Scott Hedges, Center for Environmental Research Information (CERI), and Frank F. McElroy, National Exposure Research Laboratory (NERL), all in the EPA Office of Research and Development, were responsible for overseeing the preparation of this method. Additional support was provided by other members of the Compendia Workgroup, which include:

John 0. Burckle, U.S. EPA, ORD, Cincinnati, OH James L. Cheney, Corps of Engineers, Omaha, NB Michael Davis, U.S. EPA, Region 7, KC, KS Joseph B. Elkins Jr., U.S. EPA, OAQPS, RTP, NC Robert G. Lewis, U.S. EPA, NERL, RTP, NC Justice A. Manning, U.S. EPA, ORD, Cincinnati, OH William A. McClenny, U.S. EPA, NERL, RTP, NC Frank F. McElroy, U.S. EPA, NERL, RTP, NC Heidi Schultz, ERG, Lexington, MA William T. "Jerry" Winberry, Jr., EnviroTech Solutions, Cary, NC

Ths Method is the result of the efforts of many individuals. Gratitude goes to each person involved in the preparation and review of this methodology.

Author(s) Elizabeth A. Woolfenden, Perkin Elmer Corp., Wilton, CT William A. McClenny, U.S. EPA, NERL, RTP, NC

Peer Reviewers Joan T. Bursey, ERG, Morrisville, NC Martin Harper, SKC Inc., Eighty-Four, PA Irene D. DeGraff, Supelco, Inc., Bellefonte, PA Joseph E. Bumgarner, U. S. EPA, NERL, RTP, NC Lauren Drees, U.S. EPA, NRMRL, Cincinnati, OH

Finally, recognition is given to Frances Beyer, Lynn Kaufman, Debbie Bond, Cathy Whitaker, and Kathy Johnson of Midwest Research Institute's Administrative Services staff whose dedication and persistence during the development of this manuscript has enabled it's production.

DISCLAIMER

This Compendium has been subjected to the Agency's peer and administrative review, and it has been approved for publication as an EPA document. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

11

I

METHOD TO-17

'L

Determination of Volatile Organic Compounds in Ambient Air Using Active Sampling Onto Sorbent Tubes

TABLE OF CONTENTS

1 . Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1

2 . S u m m a r y of Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-2

3 . Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-3

4 . Applicable Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-4 4.1 ASTM Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-4 4.2EPADocumen ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-4 . 4.3OtherDocuments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-4

5 . Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-5

6 . Overview of Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-6 6 . I Selection of Tube and Sorbent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7 6.2ConditioningtheTube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7 6.3 Sampling Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7 6.4SamplingRates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7 6.5 Preparing for Sample Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-8 6.6SettheFlowRates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-8 6.7 Sample and Recheck Flow Rates .............................................. 17-8 6.8 Reseal the Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-8 6.9 Selection of Thermal Desorption System ....................................... 17-8 6.10 Dry Purge the Tubes and Prepare for Thermal Desorption ...................... 17-9 6.11 Check for System Integrity ............................................... 17-9 6.12 Repurge of Tube on the Thermal DesorbedAddition of Internal Standard . . . . . . . . . . 17-9 6.13 Thermally Desorb the Packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-9 6.14 Trap Desorption and GC/MS Analysis ...................................... 17-9 6.15 Restoring the Tubes and Determine Compliance with Performance Standards . . . . . . . 17-9 6.16 Record and Store Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-10

7 . Interferences and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 10 7.1 Interference from Sorbent Artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 10 7.2 Minimizing Interference from Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 1 1 7.3 Atmospheric Pollutants not Suikble for Analysis by thls Method .................... 17- 12

7.5 Suitable Atmospheric Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 12 7.4 Detection Limits and Maximum Quantifiable Concentrations of Air Pollutants . . . . . . . . . 17- 12

... ll1

TABLE OF CONTENTS (continued)

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 . Apparatus Selection and Preparation 17- 13 8.1 Sample Collection 17- 13 8.2Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-14

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8.3 Tube Conditioning Apparatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 15

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 . Reagents and Materials 17- 16

9.2 Gas Phase Standards 17-17 9.1 Sorbent Selection Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 16

9.3 Liquid Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 17 9.4 Gas Phase Internal Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 19 9.5 Commercial, Preloaded Standard Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17- 19

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.6 Carrier Gases 17-19

10 . Guidance on Sampling and Related Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-20 10.1 Paclung Sorbent Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-20 10.2 Condtioning and Storage of Blank Sorbent Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-21 . 10.3 Record Keeping Procedures for Sorbent Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-2 1 10.4 Pump Calibration and Tube Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-22 10.5 Locating and Protecting the Sample Tube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-22 10.6 Selection of Pump Flow Rates and Air Sample Volumes ........................ 17-22

Backup Tubes. and Distributed Volume Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-23 10.8 Determining and Validating Safe Sampling Volumes (SSV) ..................... 17-24 10.9 Resealing Sorbent Tubes After Sample Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-25 10.10 Sample Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-25

1 1.1 Preparation for Sample Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-25 1 1.2 Predesorption System Checks and Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-25 1 1.3 Analytical Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-26

10.7 Sampling Procedure Verification - Use of Blanks, Distributed Volume Pairs,

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 . Analytical Procedure 17-25

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 . Calibration of Response 17-27

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 . Quality Assurance 17-27 13.1 Validating the Sample Collection Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-27 13.2 Performance Criteria for the Monitoring Pump . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-28

14 . Performance Criteria for the Solid Adsorbent Sampling of Ambient Air . . . . . . . . . . . . . . . . . . 17-28

14.2 Method Detection Limit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-28 14.3 Analytical Precision of Duplicate Pairs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-29 14.4 Precision for the Distributed Volume Pair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-29

14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-28

14.5 AuditAccuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-29

15 . References 17-30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iv

METHOD TO-17

Determination of Volatile Organic Compounds in Ambient Air Using Active Sampling Onto Sorbent Tubes

1. Scope

1.1 This document describes a sorbent tube/thermal desorptioxdgas chromatographic-based monitoring method for volatile organic compounds (VOCs) in ambient air at 0.5 to 25 parts per billion @pbv) concentration levels. Performance criteria are provided as part of the method in Section 14. EPA has previously published Compendium Method TO-1 describing the use of the porous polymer Tenax@ GC for sampling nonpolar VOCs and Compendium Method TO-2 describing the use of carbon molecular sieve for lughly volatile, nonpolar organics (1). Since these methods were developed, a new generation of thermal desorption systems as well as new types of solid adsorbents have become available commercially. These sorbents are used singly or in multisorbent packings. Tubes with more than one sorbent, packed in order of increasing sorbent strength are used to facilitate quantitative retention and desorption of VOCs over a wide volatility range. The higher molecular weight compounds are retained on the front, least retentive sorbent; the more volatile compounds are retained farther into the packing on a stronger adsorbent. The higher molecular weight compounds never encounter the stronger adsorbents, thereby improving the efficiency of the thermal desorption process.

1.2 A large amount of data on solid adsorbents is available through the efforts of the Health and Safety Laboratory, Health and Safety Executive (HSE), Sheffield, United Kingdon (X). This group has provided written methods for use of solid adsorbent packings in monitoring workplace air. Some of their documents on the subject are referenced in Section 2.2. Also, a table of information on safe sampling volumes from their research is provided in Appendu 1.

13 EPA has developed data on the use of solid sorbents in multisorbent tubes for concentration of VOCs from the ambient air as part of its program for methods development of automated gas chromatographs. The experiments required to validate the use of these sorbent traps include capture and release efficiency studies for given sampling volumes. These studies establish the validity of using solid adsorbents for target sets of VOCs with minimal (at most one hour) storage time. Although questions related to handling, transport and storage of samples between the times of sampling and analysis are not addressed, these studes provide information on safe sampling volumes. Appendix 2 delineates the results of sampling a mixture of humidified zero air and the target VOCs specified in the Compendium Method TO-14 (2) using a specific multisorbent.

1.4 An EPA workshop was convened in November of 1995 to determine if a consensus could be reached on the use of solid sorbent tubes for ambient air analysis. The draft method available at the workshop has evolved through several reviews and modifications into the current document. The method is supported by data reported in the scientific literature as cited in the text, and by recent experimental tests performed as a consequence of the workshop (see Table 1).

1.5 The analytical approach using gas chromatography/mass spectroscopy (GUMS) is identical to that mentioned in Compendium Method TO- 15 and, as noted later, is adapted for this method once the sample has been thermally desorbed from the adsorption tube onto the focusing trap of the analytical system.

January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 17-1

Method TO-17 vocs

1.6 Performance criteria are given in Section 14 to allow acceptance of data obtained with any of the many variations of sampling and analytical approaches.

2. Summary of Method

2.1 The monitoring procedure involves pulling a volume of air through a sorbent packing to collect VOCs followed by a thermal desorption-capillary G C M S analytical procedure.

2.2 Conventional detectors are considered alternatives for analysis subject to the performance criteria listed in Section 14 but are not covered specifically in this method text.

2.3 Key steps of tlus method are listed below.

2.3.1 Selection of a sorbent or sorbent mix tailored for a target compound list, data quality objectives and

2.3.2 Screening the sampling location for VOCs by taking single tube samples to allow estimates of the

2.3.3 Initial sampling sequences with two tubes at nominally 1 and 4 liter total sample volumes (or

2.3.4 Analysis of the samples and comparison to performance criteria. 2.3.5 Acceptance or rejection of the data. 23.6 If rejection, then review of the experimental arrangement including repeat analysis or repeat analysis

sampling environment.

nqure and amount of sample gases.

appropriate proportional scaling of these volumes to fit the target list and monitoring objectives).

with backup tubes and/or other QC features.

M: EPA requires the use of distributed volume pairs (see Sectionl4.4) for monitoring to insure high quality data. However, in situations where acceptable data have been routinely obtained through use of distributed volume pairs and the ambient air is considered well characterized, cost considerations may warrant single tube sampling. Any attendant risk to data quality objectives is the responsibility of the project’s decision maker.]

2.4 Key steps in sample analysis are listed below.

2.4.1 Dry purge of the sorbent tube with dry, inert gas before analysis to remove water vapor and air. The sorbent tube can be held at temperatures above ambient for the dry purge.

2.4.2 Thermal desorption of the sorbent tube (primary desorption). 2.4.3 Analyte refocusing on a secondary trap. 2.4.4 Rapid desorption of the trap and injectiodtransfer of target analytes into the gas chromatograph

2.4.5 Separation of compounds by hgh resolution capillary gas chromatography (GC). 2.4.6 Measurement by mass spectrometry ( M S ) or conventional GC detectors (only the MS approach is

explicitly referred to in Compendium Method TO- 17; an FIDECD detector combination or other GC detector can be used if Section 14 criteria are met. However, no explicit QA guidelines are given here for those alternatives).

(secondary desorption).

Page 17-2 Compendium of Methods for Toxic Organic Air Pollutants January 1999

vocs- Method TO-17

2.5 The target compound list (TCL) is the same as listed in Compendium Method TO-15 (i.e., subsets of the 97 VOCs listed as hazardous pollutants in Title III of the Clean Air Act Amendments of 1990). Only a portion of these compounds has been monitored by the use of solid adsorbents. This method provides performance criteria to demonstrate acceptable performance of the method (or modifications of the method) for monitoring a given compound or set of compounds.

3. Significance

3.1 This method is an alternative to the canister-based sampling and analysis methods that are presented in Compendium Methods TO- 14 and TO- 15 and to the previous sorbent-based methods that were formalized as Compendlum Methods TO-1 and TO-2. All of these methods are of the type that include sampling at one location, storage and transport of the sample, and analysis at another, typically more favorable site.

3.2 The collection of VOCs in ambient air samples by passage through solid sorbent pachngs is generally recognized to have a number of advantages for monitoring. These include the following:

. The small size and light weight of the sorbent packing and attendant equipment.

The placement of the sorbent packing as the first element (with the possible exception of a filter or chemical scrubber for ozone) in the sampling train so as to reduce the possibility of contamination from upstream elements.

The avdabdity of a large selection of sorbents to match the target set of compounds including polar VOC.

The commercial availability of thermal desorption systems to release the sample from the sorbent and into the analytical system.

The possibility of water management using a combination of hydrophobic sorbents (to cause water breakthrough while sampling); dry gas purge of water from the sorbent after sampling; and splitting of the sample during analysis.

The large amount of literature on the use of sorbent sampling and thermal desorption for monitoring of workplace air, particularly the literature from the Health and Safety Executive in the United Kingdom.

3.3 Accurate risk assessment of human and ecological exposure to toxic VOCs is an important goal of the U. S . Environmental Protection Agency (EPA) with increased emphasis on their role as endocrine disrupters. Accurate data is fundamental to reaching this goal. The portability and small size of typical sampling packages for sorbent-based sampling and the wide range of sorbent choices make this monitoring approach appealing for special monitoring studies of human exposure to toxic gases and to use in network monitoring to establish prevalence and trends of toxic gases. Microenvironmental and human subject studies are typical of applications for Compendium Method TO- 17.

3.4 Sorbent-based monitoring can be combined with canister-based monitoring methods, on-site autoGC systems, open path instrumentation, and other specialized point monitoring instruments to address most monitoring needs for volatile organic gases. More than one of these approaches can be used simultaneously as a means to check and insure the quality of the data being produced.

~~


Method TO-17 vocs

3.5 In the form specified in Compendium Method TO- 17, sorbent sampling incorporates the distributed volume pair approach that provides inherently defensible data to counter questions of sample integrity,, operator performance, equipment malfunction during sampling, and any other characteristic of sample collection that is not linear with sampling volume.

3.6 In keeping with the consensus of EPA scientists and science advisors, the method is performance-based such that performance criteria are provided. Any modification of the sorbent approach to monitoring for VOCs can be used provided these criteria are met.

4. Applicable Documents

4.1 ASTM Standards

Method D 1356 Definition of Terms Relating to Atmospheric Sampling and Analysis Method E260 Recommended Practicefor General Gas Chromatography . Method E355 Practice for Gas Chromatography Terms and Relationships

4.2 EPA Documents

Technical Assistance Document for Sampling and Analysis Toxic Organic Compounds in Ambient Air, U. S. Environmental Protection Agency, EPA-60014-83-027, June 1983. Quality Assurance Handbook for Air Pollution Measurement Systems, U. S. Environmental Protection Agency, EPA-600/R-94-038b, May 1994. Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air: Methods TO-I and TO-2, U. S. Environmental Protection Agency, EPA 60014-84-041, April 1984. Compendium ofMethods for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-14, Second Supplement, U. S. Environmental Protection Agency, EPA 60014-89-018, March 1989. Compendium ofMethods for the Determination of Toxic Organic Compounds in Ambient Air: Method TO-15, U. S. Environmental Protection Agency, EPA 625/R-96-010b, January 1997.

4.3 Other Documents

MDHS 3 - Generation of Test Atmospheres of Organic Vapors by the Syringe Injection Technique, Methods for the Determination of Hazardous Substances (MDHS), Health and Safety Laboratory, Health and Safety Executive, Sheffield, UK.

MDHS 4 - Generation of Test Atmospheres of Organic Vapors by the Permeation Tube Method, Methods for the Determination of Hazardous Substances (MDHS), Health and Safety Laboratory, Health and Safety Executive, Sheffield, UK.

MDHS 72 - Volatile Organic Compounds in Air, Methods for the Determination of Hazardous Substances (MDHS), Health and Safety Laboratory, Health and Safety Executive, Sheffield, UK.

TAD - Technical Assistance Document (TAD) on the Use of Solid Sorbent-based Systems for Ambient Air Monitoring, Perkin Elmer Corp., 50 Danbury Rd., Wilton, CT 06897, USA.


vocs - Method TO-17

5. Definitions

m: Definitions used in this document and any user-prepared Standard Operating Procedures (SOPS) should be consistent with those used in ASTM D l 356. All abbreviations and symbols are defined within this document at the point offirst use.]

5.1 Thermal Desorption-the use of heat and a flow of inert (carrier) gas to extract volatiles from a solid or liquid matrix directly into the carrier gas and transfer them to downstream system elements such as the analytical column of a GC. No solvent is required.

5.2 Two-stage Thermal Desorption-the process of thermally desorbing analytes from a solid or liquid matrix, reconcentrating them on a focusing tube and then rapidly heating the tube to “inject” the concentrated compounds into the GC system in a narrow band of vapor compatible with high resolution capillary gas chromatography.

53 Sorbent Tube (Also referred to as ‘tube‘ and ‘sample tube’)-stainless steel, glass or glass lined (or fused silica lined) stamless steel tube, typically 114 inch (6 mm) O.D. and of various lengths, with the central portion packed with greater than 200 mg of solid adsorbent mated, depending on density and packing bed length. Used to concentrate VOCs from air.

5.4 Focusing Tube-narrow (typically <3mm I.D.) tube containing a small bed of sorbent, whch is maintained near or below ambient temperature and used to refocus analytes thermally desorbed from the sorbent tube. Once all the VOCs have been transferred from the sorbent tube to the focusing tube, the focusing tube is heated very

- II rapidly to transfer the analytes into the capillary GC analytical column in a narrow band of vapor.

5.5 Cryogen (Also referred to as ‘cryogenic fluid’)-typically liquid nitrogen, liquid argon, or liquid carbon dioxide. In the present context, cryogens are used in some thermal desorption systems to cool the focusing tube.

5.6 High Resolution Capillary Column Chromatography-conventionally describes fused silica capillary columns with an internal diameter of 320 pn or below and with a stationary phase film thickness of 5 pn or less.

5.7 Breakthrough Volume (BV)-volume of air containing a constant concentration of analyte which may be passed through a sorbent tube before a detectable level (typically 5%) of the analyte concentration elutes from the nonsampling end. Alternatively, the volume sampled when the amount of analyte collected in a back-up sorbent tube reaches a certain percentage (typically 5%) of the total amount collected by both sorbent tubes. These methods do not give identical results. For purposes in the document the former definition will be used.

5.8 Retention Volume (RV)-the volume of carrier gas required to move an analyte vapor plug through the short packed column which is the sorbent tube. The volume is determined by measuring the carrier gas volume necessary to elute the vapor plug through the tube, normally measured at the peak response as the plug exits the tube. The retention volume of methane-is subtracted to account for dead volume in the tube.

L


Method TO-17 v o c s

5.9 Safe Sampling Volume (SSV)-usually calculated by halving the retention volume (indirect method) or hkmg two-thirds of the breakthrough volume (direct method), although these two approaches do not necessarily give identical results. The latter definition is used in this document.

5.10 Sorbent Strength-tern used to describe the affinity of sorbents for VOC analytes. A stronger sorbent is one which offers greater safe sampling volumes for most/all VOC analytes relative to another, weaker sorbent. Generally spealang, sorbent strength is related to surface area, though there are exceptions to this. The SSVs of most, if not all, VOCs will be greater on a sorbent with surface area " lon" than on one with a surface area of "n". As a general rule, sorbents are described as "weak" if their surface area is less than 50 m'g" (includes Tenax@, CarbopackTM/trap C, and Anasorb@ GCB2), "medium strength" if the surface area is in the range 100-500 m'g" (includes Carbopackm/trap B, Anasorb0 GCBI and all the Porapaks and Chromosorbs listed in Tables 1 and 2) and "strong" if the surface area is around lo00 m'g" (includes SpherocarbB, Carbosievem S-111, Carboxenm 1000, and Anasorb@ CMS series sorbents.)

5.11 Total Ion Chromatogram (TIC)-chromatogram produced fkom a mass spectrometer detector operating in 1 1 1 scan mode.

5.12 MS-SCAN-mode of operation of a GC mass spectrometer detector such that all mass ions over a given mass range are swept over a given period of time.

5.13 MS -SIM-mode of operation of a GC mass spectrometer detector such that only a single mass ion or a selected number of discrete mass ions are monitored.

5.14 Standard Sorbent (Sample) Tube-stamless steel, glass or glass lined (or fused silica lined) stainless steel tube, 114 inch (6 mm) O.D. and of various lengths, with the central portion packed with 2200 mg of solid adsorbent material depending on sorbent density. Tubes should be individually numbered and show the dxection of flow.

5.15 Time Weighted Average (TWA) Monitoring-if air is sampled over a fixed time period - typically 1,3, 8 or 24 hours, the time weighted average atmospheric concentration over the monitoring period may be calculated from the total mass of analyte retained and the specific air volume sampled. Constraints on breakthrough volumes make certain combinations of sampling time and flow rates mutually exclusive.

6. Overview of Methodology

&: The following is intended to provide a simple and straightforward method description including the example of a specific samplingproblem. Although specific equipment is listed, the document is intended only as an example and equipment mentioned in the text is usually only one of a number of equally suitable components that can be used. Hence trade names are not meant to imply exclusive endorsement for sampling and analysis using solid sorbents. Later sections in the text give guidance as to what considerations should be made for a number of VOC monitoring applications.]


vocs - Method TO-17

6.1 Selection of Tube and Sorbent

6.1.1 Select a tube and sorbent packing for the sampling application using guidance from Tables 1 and 2 on sorbent characteristics as well as guidance from Appendlx 1 and Table 3 on safe sampling volumes and breakthrough characteristics of sorbents.

6.1.2 As an example, assume the TCL includes a subset of the compounds shown in Table 3. In this case, the multisorbent tube chosen consists of two sorbents packed in a 1/4 inch O.D., 3.5" long glass tube in the following order and amounts: 160 mg of Carbopackm graphitized carbon black (60/80 mesh) and 70 mg of Carboxenm-1000 type carbon molecular sieve (60/80 mesh). This is an example of Tube Style 2 discussed Section 9.1.3.2.

6.1.3 Pack the tube with the adsorbent by using the guidance provided in Section 10.1 or buy a prepacked tube from a supplier. In the example, tubes were purchased from Supelco Inc., Supelco Park, Bellefonte, PA 16823-0048.

6.2 Conditioning the Tube

6.2.1 Condition newly packed tubes for at least 2 hours (30 mins for preconditioned, purchased tubes) at 358°C while passing at least 50 mL/min of pure helium carrier gas through them.

w: Other sorbents may require drfferent conditioning temperatures - see Table 2 for guidance. J

Once conditioned, seal the tube with brass, 114 inch SwagelokB -type fittings and PTFE ferrules. Wrap the sealed tubes in uncoated aluminum foil and place the tubes in a clean, airtight, opaque container.

6.2.2 A package of clean sorbent material, e.g. activated charcoal or activated charcoal/silica gel mixture, may be added to the container to ensure clean storage conditions.

6.2.3 Store in a refrigerator (organic solvent-free) at 4°C if not to be used within a day. On second and subsequent uses, the tubes will generally not require further conditioning as above. However, tubes with an immediate prior use indicating high levels of pollutant trace gases should be reconditioned prior to continued usage.

',L

#

6.3 Sampling Apparatus

6.3.1 Select a sampling apparatus with accommodations for two sampling tubes capable of independent control of sampling rate at a settable value in the range 10 to 200 mL/min. Laboratory and field blanks must also be included in the monitoring exercise.

63.2 Backup tubes may be required to determine the cause of any problem if performance criteria, outlined in Section 14, are not met.

6.4 Sampling Rates

6.4.1 Select sampling rates compatible with the collection of 1 and 4 liter total sample volume (or of

6.4.2 Air samples are collected over 1 hour with a sampling rate of 16.7 mL/min and 66.7 mL/min, proportionally lowerhigher sampling volumes).

respectively.


Methad TO-17 vocs

6.5 Preparing for Sample Collection

6.5.1 At the monitoring location, keep the tubes in their storage and transportation container to equilibrate with ambient temperature.

6.5.2 Using clean gloves, remove the sample tubes from the container, take off their caps and attach them to the sampling lines with non-outgassing flexible tubing. Uncap and immediately reseal the required number of field blank tubes.

6.53 Place the field blank tubes back in the storage container. If back-up tubes are being used, attach them to the sampling tubes using clean, metal Swagelok@ type unions and combined PTFE ferrules.

6.6 Set the Flow Rates

6.6.1 Set the flow rates of the pump using a mass flow monitor. 6.6.2 The sampling train includes, from front to back, an in-line particulate filter (optional), an ozone

scrubber (optional), a sampling tube, a back-up tube if any is being used, and a flow controller/pump combination.

6.6.3 Place the mass flow monitor in line after the tube. Turn the pump on and wait for one minute. 5stablish the approximate sampling flow rate using a dummy tube of identical construction and packing as the sampling tube to be used. Record on Field Test Data Sheet (FTDS), as illustrated in Figure 1.

6.6.4 Place the sampling tubes to be used on the sampling train and make final adjustments to the flow controller as quickly as possible to avoid significant errors in the sample volume.

6.6.5 Adjust the flow rate of one tube to sample at 16.7 ml/min. Repeat the procedure for the second tube and set the flow rate to 66.7 mL/min. Record on FTDS.

6.7 Sample and Recheck Flow Rates

6.7.1 Sample over the selected sampllng period (i.e., l-hour). Recheck all the sampling flow rates at the end

6.7.2 Make notes of all relevant monitoring parameters including locations, tube identification numbers, of the monitoring exercise just before switching off each pump and record on FTDS.

pump flow rates, dates, times, sampled volumes, ambient conditions etc. on FTDS.

6.8 Reseal the Tubes

6.8.1 Immediately remove the sampling tubes with clean gloves, recap the tubes with SwagelokB fittings using PTFE ferrules, rewrap the tubes with uncoated A1 foil, and place the tubes in a clean, opaque, airtight container.

6.8.2 If not to be analyzed during the same day, place the container in a clean, cool (<4"C), organic solvent- free environment and leave there until time for analysis.

6.9 Selection of Thermal Desorption System

6.9.1 Select a thermal desorption system using the guidance provided in Section 8. 6.9.2 Place the thermal unit in a ready operational status.

I


vocs Method TO-17

6.10 Dry Purge the Tubes and Prepare for Thermal Desorption

6.10.1 Remove the sampling tubes, any backup tubes being used, and blanks from the storage area. and allow the tubes to come to room temperature. Using clean gloves, remove the Swagelok@-type fittings and dry purge the tubes with a forward (sampling direction) flow of, for example, 50 mL/min of dry helium for 4 minutes (see Section 7.2 concerning dry purging).

/&&: Do not dry purge the laboratory blanks. J

6.10.2 Reseal the tubes with Teflon@ (or other) caps compatible with the thermal desorber operation. Place the sealed tubes on the thermal desorber (e.g., Perkin Elmer Model ATD 400 Automated System or equivalent). Other thermal desorben may have different arrangements for automation. Alternatively, use equivalent manual desorption.

6.11 Check for System Integrity

6.11.1 Check the air tightness of the seals and the integrity of the flow path. . 6.11.2 Guidance is provided in Section 11.2 of this document.

6.12 Repurge of Tube on the Thermal DesorbedAddition of Internal Standard

6.12.1 Because of tube handling after dry purge, it may be necessary to repurge each of the tubes with pure,

6.12.2 If the initial dry purge can be perfomed on the thermal desorber so as to prevent any further exposure dry helium (He) before analysis in order to eliminate any oxygen.

- of the sorbent to air, then this step is not necessary. Proceed with the addition of an internal standard to the sorbent tube or the focusing tube.

6.13 Thermally Desorb the Packing

6.13.1 Reverse the flow direction of He gas, set the flow rate to at least 30 mL/min, and heat the tube to 325 "C (in this case) to achieve a transfer of VOCs onto a focusing tube at a temperature of 27°C. Thermal desorption continues until all target species are transferred to the focusing trap. The focusing trap is typically packed with 20 mg of Carbopackm B (60/80 mesh) and 50 mg of a Carboxenm 1000-type sorbent (60/80 mesh).

6.14 Trap Desorption and GC/MS Analysis

6.14.1 After each tube is desorbed, rapidly heat the focusing trap (to 325°C in this example) and apply a rev- flow of at least 3 mL,/min of pure helium carrier gas. Sample splitting is necessary to accommodate the capillary column. Analytes are transferred to the column in a narrow band of vapor.

6.14.2 The GC run is initiated based on a time delay after the start of thermal desorption. The remaining part of the analyhcal cycle is described in Section 3 of Compendium Method TO-15.

6.15 Restoring the Tubes and Determine Compliance with Performance Standards

- January 1999 Compendium of Methods for Toxic Organic Air Pollutants Page 17-9

MethobTO-17 vocs

6.15.1 When tube analysis is completed, m o v e the tubes fiom the thermal desorber and, using clean gloves, replace the Teflon@ caps with Swagelok fittings and PTFE ferrules, rewrap with aluminum foil, replace in the clean, airtight container, and re-store the tubes in a cool environment (<4"C) until the next use.

6.152 Using previously prepared identification and quantification subroutines, identify the target compounds and document the amount ofeach measured compound (refer to the Section 3 of Compendium Method TO-15). Compare the results of analysis for the distributed volume pair taken during each sampling run and use the comparison to determine whether or not the performance criteria for individual sampling events have been met. Also examine the results of any laboratory blanks, field blanks, and any backup tube being used. Accept or reject the data based on the performance criteria (see Section 14).

6.16 Record and Store Data

6.16.1 Accurately retrieve field data (including the tube identification number) from the FTDS. The data should include a samphg site identifier, time of sample initiation, duration of sampling, air pump identification, flow rate, and other information as appropriate.

6.16.2 Store G C M S data in a permanent form both in hard copy in a notebook and in digital form on a disk. Also store the data sheet with the hard copy.

[&: Sections 7 through 14 below elaborate on the method by providing important information and guidance appropriate to explain the method as outlined in Section 4 and also to generalize the method for many applications. Section 14 gives the performance criteria for the method.]

7. Interferences and Limitations

7.1 Interference from Sorbent Artifacts

7.1.1 Minimizing Artifact Interference. 7.1.1.1 Stringent tube conditioning (see Section 10.2.1) and carefbl tube capping and storage procedures

(see Section 10.2.2) are essential for minimizing artifacts. System and sorbent tube conditioning must be carried out using more stringent conditions of temperature, gas flow and time than those required for sample analysis.

7.1.1.2 A reasonable objective is to reduce artifacts to 10% or less of individual analyte masses retained during sampling. A s u m m a r y of VOC levels present in a range of different atmospheric environments and the masses of individual components collected from 1 , 2 or 10 L samples of air in each case is presented in Table 4.

7.1.13 Given that most ambient air monitoring is carried out in areas of poor air quality, for example in urban, indoor and factory fenceline environments where VOC concentrations are typically above 1 ppb, Table 4 demonstrates that the mass of each analyte retained will, therefore, range from -5 ng to -10 pg in most monitoring situations. Even when monitoring 'ultraclean' environments, analyte masses retained will usually exceed 0.1 ng (3).

7.1.1.4 Typical artifact levels for 1/4 inch O.D. tubes of 3.5" length range from 0.01 ng and 0.1 ng for cdnaceous sorbents and Tenax@ respectively. These levels compare well with the masses of analytes collected - even from sub-ppb atmospheric concentrations (see Table 4). Artifact levels are around 10 ng for Chromosoh@ Century series and other porous polymer sorbents. However, these types of sorbents can still be used for air monitoring at low ppb levels if selective or mass spectrometer detectors are used or if the blank profile of the tube demonstrates that none of the sorbent artifacts interfere analyhcally with the compounds of interest.


1250

METHOD 25A - DETERMINATION OF TOTAL GASEOUS ORGANIC CONCENTRATION USING A FLAME IONIZATION ANALYZER

'L

L

1.0 Scope and Application.

1.1 Analytes.

Analyte CAS No. Sensitivity

Total Organic N/A c 2% of span ComDounds

1.2 Applicability. This method is applicable for the

determination of total gaseous organic concentration of

vapors consisting primarily of alkanes, alkenes, and/or

arenes (aromatic hydrocarbons). The concentration is

expressed in terms of propane (or other appropriate organic

calibration gas) or in terms of carbon.

1.3 Data Quality Objectives. Adherence to the

requirements of this method will enhance the quality of the

data obtained from air pollutant sampling methods.

2.0 Summary of Method.

2.1 A gas sample is extracted from the source through

a heated sample line and glass fiber filter to a flame

ionization analyzer (FIA). Results are reported as volume

concentration equivalents of the calibration gas or as

carbon equivalents.

3.0 Definitions.

3.1 Calibration drift means the difference in the

measurement system response to a mid-level calibration gas

1251

before and after a stated period of operation during which

no unscheduled maintenance, repair, or adjustment took

place.

3.2 Calibration error means the difference between

the gas concentration indicated by the measurement system

and the know concentration of the calibration gas.

3.3 Calibration gas means a known concentration of a

gas in an appropriate diluent gas.

3.4 Measurement system means the total equipment

required for the determination of the gas concentration.

The system consists of the following major subsystems:

3.4.1 Sample interface means that portion of a system

used for one or more of the following: sample acquisition,

sample transportation, sample conditioning, or protection of

the analyzer(s) from the effects of the stack effluent.

3.4.2 Organic analyzer means that portion of the

measurement system that senses the gas to be measured and

generates an output proportional to its concentration.

3.5 Response time means the time interval from a step

change in pollutant concentration at the inlet to the

emission measurement system to the time at which 95 percent

of the corresponding final value is reached as displayed on

the recorder.

L.

1252

3 . 6 Span Value means the upper limit of a gas

concentration measurement range that is specified for

affected source categories in the applicable part of the

regulations. The span value is established in the

applicable regulation and is usually 1.5 to 2 . 5 times the

applicable emission limit. If no span value is provided,

use a span value equivalent to 1 . 5 to 2 . 5 times the expected

concentration. For convenience, the span value should

correspond to 100 percent of the recorder scale.

3 . 7 Zero drift means the difference in the

measurement system response to a zero level calibration gas

before or after a stated period of operation during which no

unscheduled maintenance, repair, or adjustment took place.

4.0 Interferences. [Reserved]

5 . 0 Safety.

5.1 Disclaimer. This method may involve hazardous

materials, operations, and equipment. This test method may

not address all of the safety problems associated with its

use. It is the responsibility of the user of this test

method to establish appropriate safety and health practices

and determine the applicability of regulatory limitations

prior to performing this test method. The analyzer users

manual should be consulted for specific precautions to be

taken with regard to the analytical procedure. L;

1253

5.2 Explosive Atmosphere. This method is often

applied in highly explosive areas. Caution and care should

be exercised in choice of equipment and installation.

6.0 Equipment and Supplies

6.1 Measurement System. Any measurement system for

total organic concentration that meets the specifications of

this method. A schematic of an acceptable measurement

system is shown in Figure 25A-1. All sampling components

leading to the analyzer shall be heated 2 110°C (220°F)

throughout the sampling period, unless safety reasons are

cited (Section 5.2) The essential components of the

measurement system are described below:

6.1.1 Organic Concentration Analyzer. A flame

ionization analyzer (FIA) capable of meeting or exceeding

the specifications of this method. The flame ionization

detector block shall be heated >120"C (250°F).

6.1.2 Sample Probe. Stainless steel, or equivalent,

three-hole rake type. Sample holes shall be 4 mm (0.16-in.)

in diameter or smaller and located at 16.7, 50, and 83.3

percent of the equivalent stack diameter. Alternatively, a

single opening probe may be used so that a gas sample is

collected from the centrally located 10 percent area of the

stack cross-section.

I.",.

1254

6.1.3 Heated Sample Line. Stainless steel or Teflon@

tubing to transport the sample gas to the analyzer. The

sample line should be heated (2110 "C) to prevent any

condensation.

6.1.4 Calibration Valve Assembly. A three-way valve

assembly to direct the zero and calibration gases to the

analyzers is recommended. Other methods, such as quick-

connect lines, to route calibration gas to the analyzers are

. applicable.

6.1.5 Particulate Filter. An in-stack or an out-of-

stack glass fiber filter is recommended if exhaust gas

particulate loading is significant. An out-of-stack filter

should be heated to prevent any condensation.

6.1.6 Recorder. A strip-chart recorder, analog

computer, or digital recorder for recording measurement

data. The minimum data recording requirement is one

measurement value per minute.

7.0 Reagents and Standards.

7.1 Calibration Gases. The calibration gases for the

gas analyzer shall be propane in air or propane in nitrogen.

Alternatively, organic compounds other than propane can be

used; the appropriate corrections for response factor must

be made. Calibration gases shall be prepared in accordance

with the procedure listed in Citation 2 of Section 16. L.

1 2 5 5

Additionally, the manufacturer of the cylinder should

provide a recommended shelf life for each calibration gas

cylinder over which the concentration does not change more

than 2 percent from the certified value. For calibration

gas values not generally available ( i . e . , organics between 1

and 10 percent by volume), alternative methods for preparing

calibration gas mixtures, such as dilution systems (Test

Method 2 0 5 , 4 0 CFR Part 51, Appendix M), may be used with

. prior approval of the Administrator.

7 . 1 . 1 Fuel. A 40 percent H,/60 percent N, or He gas

mixture is recommended to avoid an oxygen synergism effect

that reportedly occurs when oxygen concentration varies

significantly from a mean value.

7 . 1 . 2 Zero Gas. High purity air with less than 0.1

part per million by volume (ppmv) of organic material

(propane or carbon equivalent) or less than 0.1 percent of

the span value, whichever is greater.

7 . 1 . 3 Low-level Calibration Gas. An organic

calibration gas with a concentration equivalent to 25 to 35

percent of the applicable span value.

7 . 1 . 4 Mid-level Calibration Gas. An organic



1256

7.1.5 High-level Calibration Gas. An organic



8.0 Sample Collection, Preservation, Storage, and

Transport.

8.1 Selection of Sampling Site. The location of the

sampling site is generally specified by the applicable

regulation or purpose of the test (i.e., exhaust stack,

4 inlet line, etc.). The sample port shall be located to meet

the testing requirements of Method 1.

8.2 Location of Sample Probe. Install the sample

probe so that the probe is centrally located in the stack,

pipe, or duct and is sealed tightly at the stack port

connection.

8.3 Measurement System Preparation. Prior to the

emission test, assemble the measurement system by following

the manufacturer's written instructions for preparing sample

interface and the organic analyzer. Make the system

operable (Section 10.1) .

8.4 Calibration Error Test. Immediately prior to the

test series (within 2 hours of the start of the test),

introduce zero gas and high-level calibration gas at the

calibration valve assembly. Adjust the analyzer output to

the appropriate levels, if necessary. Calculate the

1257

predicted response for the low-level and mid-level gases

based on a linear response line between the zero and high-

level response. Then introduce low-level and mid-level

calibration gases successively to the measurement system.

Record the analyzer responses for low-level and mid-level

calibration gases and determine the differences between the

measurement system responses and the predicted responses.

These differences must be less than 5 percent of the

respective calibration gas value. If not, the measurement

system is not acceptable and must be replaced or repaired

prior to testing. No adjustments to the measurement system

shall be conducted after the calibration and before the

drift check (Section 8.6.2). If adjustments are necessary

before the completion of the test series, perform the drift

checks prior to the required adjustments and repeat the

calibration following the adjustments. If multiple

electronic ranges are to be used, each additional range must

be checked with a mid-level calibration gas to verify the

multiplication factor.

8 . 5 Response Time Test. Introduce zero gas into the

measurement system at the calibration valve assembly. When

the system output has stabilized, switch quickly to the

high-level calibration gas. Record the time from the

concentration change to the measurement system response

i

1258

equivalent to 95 percent of the step change. Repeat the

test three times and average the results.

8.6 Emission Measurement Test Procedure.

8.6.1 Organic Measurement. Begin sampling at the

start of the test period, recording time and any required

process information as appropriate. In particulate, note on

the recording chart, periods of process interruption or

cyclic operation.

8.6.2 Drift Determination. Immediately following the

completion of the test period and hourly during the test

period, reintroduce the zero and mid-level calibration

gases, one at a time, to the measurement system at the

calibration valve assembly. (Make no adjustments to the

measurement system until both the zero and calibration drift

checks are made.) Record the analyzer response. If the

drift values exceed the specified limits, invalidate the

test results preceding the check and repeat the test

following corrections to the measurement system.

Alternatively, recalibrate the test measurement system as in

Section 8.4 and report the results using both sets of

calibration data (i.e., data determined prior to the test

period and data determined following the test period).

NOTE: Note on the recording chart periods of process

interruption or cyclic operation.

1259

9.0 Quality Control

Met hod Quality Control Effect Section Measure

8.4 Zero and calibration Ensures that bias drift tests. introduced by drift

in the measurement system output during the run is no greater than 3 percent of span. -

10.0 Calibration and Standardization.

10.1 FIA equipment can be calibrated for almost any

range of total organic concentrations. For high

concentrations of organics ( > 1.0 percent by volume as

propane), modifications to most commonly available analyzers

are necessary. One accepted method of equipment

modification is to decrease the size of the sample to the

analyzer through the use of a smaller diameter sample

capillary. Direct and continuous measurement of organic

concentration is a necessary consideration when determining

any modification design.

11.0 Analytical Procedure.

The sample collection and analysis are concurrent for

this method (see Section 8 . 0 ) .

12.0 Calculations and Data Analysis.

12.1 Determine the average organic concentration in

terms of ppmv as propane or other calibration gas. The

1260

average shall be determined by integration of the output

recording over the period specified in the applicable

regulation. If results are required in terms of ppmv as

carbon, adjust measured concentrations using Equation 25A-1.

c, = K cm,,, Eq. 25A-1

where :

cc = Organic concentration as carbon, ppmv.

C,,,, = Organic concentration as measured, ppmv.

K = Carbon equivalent correction factor.

= 2 for ethane.

= 3 for propane.

= 4 for butane.

= Appropriate response factor for other organic

calibration gases.

13.0 Method Performance.

13.1 Measurement System Performance Specifications.

13.1.1 Zero Drift. Less than k3 percent of the span

value.

13.1.2 Calibration Drift. Less than +3 percent of

span value.

13.1.3 Calibration Error. Less than +5 percent of

the calibration gas value.

14.0 Pollution Prevention. [Reserved]

15.0 Waste Management. [Reserved] L.

1261

16 .0 References.

1. Measurement of Volatile Organic Compounds-

Guideline Series. U.S. Environmental Protection Agency.

Research Triangle Park, NC. Publication No. EPA-450/2-78-

041. June 1978. p. 46-54.

2 . EPA Traceability Protocol for Assay and

Certification of Gaseous Calibration Standards. U.S.

Environmental Protection Agency, Quality Assurance and

Technical Support Division. Research Triangle Park, N.C.

September 1993.

3 . Gasoline Vapor Emission Laboratory Evaluation-Part

2 . U.S. Environmental Protection Agency, Office of Air

Quality Planning and Standards. Research Triangle Park, NC.

EMB Report No. 75-GAS-6. August 1975.

17.0 T a b l e s , D i a g r a m s , F l o w c h a r t s , a n d V a l i d a t i o n D a t a .

1262

J

L

8 a 8 - C

e Y

W

c u w 4 PI 6 b

L

P

. . -

€ E 8 a u

0 -

33

r P

n z a

8

c c ..I c)

v

L

Odor Related Issues in Commercial Composting › ref › 12 › 11500.pdf2.1 Gas chromatography...

Documents

Transcript of Odor Related Issues in Commercial Composting › ref › 12 › 11500.pdf2.1 Gas chromatography...