Eu.-

... NOM...Emhhohhh..h...Ehoh...hhh..I

Al.

1.

j~OR.R-81 -0559 LEEL'

II

SLIPID P93Ifl~hT, OID I & UWFUN FILTRATMcivil

A *flnginewgingresgorch

Fna Paort h The

I MRini-Grant Progr

Air Force office of Scientific R~esearch

Bolling Air Force Baset DC

By

'Tuma E. Higgins Ph.D., P.E.Jc Assistant Profesbor

I Grat NO ~fQA~44N~ L D T II ELECTE

Ify? 19RI JUL 28 1981

D

center for research

college of engineering & applied sciences

arizona state university 6istributIOU UuI~dtempe, arizona 0526124 048 12N--11

,t

AIR FORCE OFFICS 0 9CIOTIFIC URCIA (AFS)IOTICE OF mAMSMITTAL TO DDC

This teohnlcal report has beou roviovwd &M Lapproved for publc rhloa'se LAW AYR 19O-12 (Tb).

DistrIbution ia wiliwited.A. D. BUSTochnioal Ieformetlon Officer

R J1 DOCUmENTATION~ PAGE ___'OR

*~ii .. 810559i'66 _

9 Treatmrent of Plating Wastewaters by Ferrous arl~ewt ,-

Reduction, Sulfi.de Precipitaticn., Cogulaticn --.-

,Im j gins0 AFOSR-81--0044

P ;A AT N A E A~ N- Z; E SS

Departirent of Civil D~gineerina 2VD*IArizona State University 3,AZ 85287________ ___

it :O -q N5 -I C NAmE AN' A20 -5 ? .r m- ::Ah1

Air Force Of fice of Scientific Research May.W81 ____Bolling AFB, DC 20332 _ -

"NCLASSIED

0 S R E A 8 ~ $-----

EAVY =EALS, ,SMLTER TP-ZXTT -, CIULM RDJI~ A.

-- C~rrent cractice Ln treat-eant of :Jixed -etal .:,asteaters moflcainL-IChe~aetc-=T-- consists of -Acidiir 3) crori= t-duczi-cn fowed

bv alkaline, -x-tal-h-vdroxide t-reciz;:itation; as tw-stag7e pr-ocess. \Silf1-stage pro-cess could reduc-2 capital and operation. zosts. Synthetic cr~ircadrurru, andi nickel ;*:asterwaters wecre successfully treated by rous sulfsteand sodi,-n. sulfidc tn a sirncle cell reactor at: al]<alinle 7oj -1) , thusclarif-/,ing ccnzlicti-ng reports i thne litf7rat'.Lre. Batch tests were prc~to stjdvy this 2rocess. _Sol.'ds A*-re romcved b'. zmr-brane filtration. 'MtalA

DD , A%"v3 1473 - ~ s:~

S~±,t~~-N ~ A, E A--

residuals were iaalzed by atcrtic absorption spectrophotcretry. Chrcimiz" wasrapidly and stoichirtric,3Lvy reduced by ferrcus sulfate. Scdiun salf'.de byI itself negligibly reduced chmr iun, and was no better than hydrcxide inprecipitatino cadxniun and nickel. However, in caration wlth ferrous sjI f.ate,sulfide iriproved chrcrni.r reduction and me~tal r v Al. AkaIir.itv, ME-TA andJI cyPanide interfereri wit-h c&irrui' reduction and rveta 1 remv.aI; cal21.ia Kardnc-sscc' iteracted these interferences.

IF -

A. I" - ONS SP 5

i .

T A B L E O F C O N T T S P GJ PAGE

COVER LETTER . I.

REPORT WCUNTATON PAGE . ....................

TABLE OF CONTENTS ......................... iv

LIST OF FIGURES . . .. ... .. . . . . . .. .. . .. . . .. . v

LIST OF TABLES ........... ........................... viii

INRrDUCTION ............. ............................ 1

LITERATURE ............. ............................. 3

RESEARCH OBJECTrIVES & RATIONALE ....... .................. .. 11

METHODS & MATERIALS .......... ........................ .13

1. Waste Conposition .......... ..................... 13

2. Batch Treatment ........ ...................... ... 13

3. Chemical Analyses .......... .................... 13

4. Continuous Flow Aparatus ...... .................. ... 14

RESULTS ............. ............................. .. 16

1. Chromium Reduction & Renoval ........ .............. 16

2. Effects of Hardness, Alkalinity & Complexing Agents . . . 25

3. Mixed Metal Treatment ....... .................. ... 45

4. Treatment of Plating Rinsewaters .... .............. ... 64

5. Continous Flow Studies ...... ................... ... 77

CONCLUSIONS ........... ............................ . 80

ONGOING RESEARCM .......... .......................... ... 82

1. Objectives ......... ......................... .. 82

2. Proposed Process Development ..... ................ ... 82

3. Instrumentation and Control ...... ................ ... 83

4. Pilot Plant ........... ....................... 84

BIBLIOGkAPHY ............ ........................... ... 85

"MM PA

v

LIST OF FIGURES

Figure Page

1. Solubilities of Cd, Cr and Ni as a function of pH. 8

2. Solubilities of Cd, Cr and Ni with 10-6M total sulfides. 9

3. Distribution of sulfide species with pH. 10

4. Upflow filtration apparatus. 15

5. Chromium reduction by sulfide. 18

6. Chromium reduction by ferrous sulfate and sodium sulfideat pl 7. 19

7. Chromium reduction by ferrous sulfate and sodium sulfideat pH 8.5. 20

8. Chromium reduction by ferrous sulfate and sodin sulfideat pH 10. 21

9. Isoequivalent chromium residuals (meq/l) for ferrous andsulfide doses at p- 7 (initial chromium = 0.58 meqlJ. 22

10. Isoequivalent chromium residuals (meq/l) for ferrous andsulfide doses at pH 8.5 (initial chromium = 0.58 meq/l) . 23

11. Isoequivalent chromium residuals (mecl) for ferrous andsulfide doses at pH 10. (initial chromium = 0.58 mecql). 24

12. Chromium reduction by ferrous sulfate and sodium sulfideat pH 8.5 in distilled water. 28

13. Effects of alkalinity on chromium reducticn by ferroussulfate pH 8.5. 29

14. Effects of alkalinity on chromium reduction by ferroussulfate and sodium sulfide (0.29 meq/l) at pH 8.5. 30

15. Effects of calcium hardness on chromium reduction by ferroussulfate at pH 8.5. 31

16. Effects of hardness on chromium reduction by ferrous sulfateand sodium sulfide (0.29 meq/1) at pH 8.5. 32

17. Effects of hardness and alkalinity (200 mg/l as CaC%3 ) onchromium reduction by ferrous sulfate at pH 8.5. 33

18. Effects of hardness and alkalinity (200 mg/l as Ca C ) onchromium reduction by ferrou sulfate and sodium sulfide(0.29 meq/1) at pH 8.5. 34

vi

19. Chromium reduction oy ferrous sulfate and sodium sulfide atpH 8.5 in a hard tap water. 35

20. Chromium (100 mg/l reduction by ferrous sulfate and sodiumsulfide at pH 8.5 in a hard tap water. 36

21. Effects of EDTA on chroniuLn reduction by ferrous sulfate atpH = 8.5. 37

22. Effects of EDTA on chromiim reduction by ferrous sulfate andsodium sulfide (0.29 .eq/l) at pH 8.5. 38

23. Effects of EDTA on chromium reduction by ferrous sulfate inhard tap water at pH 8.5. 39

24. Effects of EDTA on chromium reduction by ferrous sulfate andsodium sulfide (0.29 meq,'l) in hard tap water at pU 8.5. 40

25. Effects of cyanide on chromium reduction by ferrous sulfateat pH 8.5. 41

26. Effects of cyanide on chromium reduction by ferrous .ulfateand sodium sulfide (0.29 zues/l) at pH 3.5. 42

27. Effects of cyanide on chromium reduction by ferrous sulfatein hard tap water at pH 8.5. 43

28. Effects of cyanide on chromium reduction by ferrous sulfateand soidum sulfide (0.29 meq") in hard tap water at cH 6.5. 44

29. Cadmium and chromium removal by sodium sulfide in softenedtap water at pi4 7. 48

30. Cadmium and chromium removal by sodium sulfide in softenedtap water at pH 8.5. 49

31. Cadmium removal by sodium sulfide in softened tzp water atpH 10. 5o

32. Isoequivalent chromium residuals (meo Il) for treatment cfmixed metals in softened tap water at pH 8.5. 51

33. Isoequivalent caCziu-n residuals (meq/l) for treatment ofmixed metals in softened tap water at pH 7. 52

34. Isoequivalent cadmim residuals for treatment of mixed metalsin softened tap water at pH 8.5. 53

35. Isoequivalent cacmium residuals for treat:.ent of mixed metalsin softened tap water at p1 10. 54

36. Chromium (C.58 meaqjl) removal from a miixed -a2 waste at PH8.5. 55

vii

37. Cadmium (0.18 meq/l) removal from mixed metal waste at pH8.5. 56

38. Nickel (0.34 meq/1) removal from a mixed metal waste at pH 8.5. 57

39. Chromium (0.58 meq/1) removal from a mixed metal waste in hardtap water at pH 8.5. 58

40. Cadmium (0.18 meq/l) removal from a mixed metal waste in hardtap water at pH 8.5. 59

41. Nickel (0.39 meq/l) removal from a mixed medal waste in hardtap water at pH 8.5. 60

42. Chromium (0.58 meq/l) removal from a mixed metal waste in hardtap water (Ca(OH)2) with pH adjusted to 8.5 with lime. 61

43. Cadmium (0.18 meq/l) removal from a mixed metal waste in hardtap water with pH adjusted to 8.5 with lime. 62

44. Nickel (0.39 meq/l) removal from a mixed metal waste in hardtap water with pH adjusted to 8.5 with lime. 63

45. Chromium (0.58 meq/1) removal from a chromium plating waste indistilled water at pH 8.5. 66

46. Chromium (0.58 meq/l) removal from a chromium plating waste inhard tap water at pH 8.5. 67

47. Chromium (0.58 meq/!) removal from a mixed metal plating wastein distilled water at pH 8.5. 68

48. Cadmium (0.18 Teq/1) removal from a mixed metal plating wastein distilled water at pH 8.5. 69

49. Nickel (0.39 meq/i) removal from a mixed metal plating wastein distilled water at pH 8.5. 70

50. Chromium (0.58 -q/'l) removal from a mixed metal plating wastein hard tap water at pH 8.5. 71

51. Cadmium (0.18 meq/l) removal from a mixed metal plating wastein hard tap water at pH 8.5. 72

52. Nickel (0.39 meq/l) removal from a mixed metal plating wastein hard tap water at pH 8.5. 73

53. Chromium (0.58 meq/l) removal from a mixed metal plating wastein hard tap water with pH adjusted to 8.5 with lime. 74

54. Cadmium (0.16 meq/l) removal from a mixed metal plating wastein hard tap water with p-i adjusted to 8.5 with lire. 75

viii

55. Nickel (0.39 meq/1) removal from a mixed metal plating waste inhard tap water with pH adjusted to 8.5 with lime. 76

56. Treatment of mixed metal wastewater by ferrous and sulfide andupflow filtration. 79

LIST OF TABLES

Table Page

1. Comparison of Pre-Treatment Effluent Guidelines, Water QualityCriteria, and NPDES Discharge Limits for Selected Metals. 1

2. Typical Analysis of Tap Water, Tyndall Air Force Base, Florida. 16

3. Analysis of Tap Water, Tempe, Arizona. 26

4. Operating Conditions and Results for Continuous Flow TreatnentSystem. 78

7Il

INTRODUCTION

Recent pre-treat ent standards promulgated by the Environmental

Protection Agency (EPA) are aimed at the reduction of significant

discharges of heavy metals into domestic sewage systems. Small

electroplating operations have limited facilities for treatment. Current

Air Force practice in small electroplating shops is to discharge

rinsewaters directly into sanitary sewers without pre-treatment. A

treatment system is needed which has application at small installations to

meet the new pre-treatment standards.

TABLE 1

Comparison of Pre-Treatment Effluent Guidelines,Water Quality Criteria, and NPDES Discharge LLmits

for Selected Metals (References 1, 2, and 3).

Pollutant Cd Cr(VI) Cr(T) Cu Ni Pb

EPA Pre-treatment Guidelines1 Day Max. (a) 1.2 - - 0.64 Day Ave. (a) 0.7 - - - - 0.4

. Day Max. (b) 1.2 - 7.0 4.5 4.1 0.64 Day Ave. (b) 0.7 - 4.0 2.7 2.7 0.4

EPA Drinking WaterQuality Criteria 0.01 0.05 0.05 1.0 - 0.05

Tinker AFB 4PDESPermit (c) 0.019 0.05 0.11 0.01 0.22 0.06

Electroplating Waste (d)Max. 0.23 55 156 0.42 170 0.259 Day Ave. 0.14 28 43 0.09 37 0.06

N('OES: a. Flows less than 10,000 gallons per day.b. Flows greater than 10,000 gallons per day.c. Based on 0.7 '.GID flow from IlTP and 0.4 Kl) stream flow.d. Based on daily samples taken Nov. 10-18, 1977.e. Concentrations in mg/l.

2

Minimum pre-treatment standards for selected metals are shown in

Table 1. National Pollution Discharge Elimination System (NPDES) permit

requirements can result in more stringent effluent standards (example

shown). Principal contaminants in the wastewater (principally rinsewater)

are chromium and nickel which, along with cadmium, are the principal metals

used for electroplating by the Air Force.

This project is a continuation of a study initiated by the principal

investigator as an AFOSR Summer Faculty Research Associate at the HQ AFESC,

Tyndall AFB, FL. The initial objective of that study was to investigate

the feasibility of sulfide precipitation, coagulation by :ron salts and

upflow filtration for removal of heavy metals from electroplating rinse

waters at small Air Force facilities.

Literature indicated that chromium reduction by ferrous iron was

essentially non-existent at a pH above th-ree. Nevertheless, there were

reports by Air Force Treatment personnel of successful reduction of

hexavalent chromium by ferrous addition at alkaline pH.

The present study was undertaken to determine I f reducticn of

* hexavalent chromium L, ferrous iron is reasonably rapid at alkaline pH,.

The coagulant properties of the resulting ferric iron were noted. The

effectiveness of sodium sulfide as a re-ducing agent and as a precipitant

for mixed metals was investigated by itself and in combination with ferrous

sulfate. The effects of competing ions, normally present in plating

wastewaters (hardness, alkalinity, EDTA, cyanide), on chromium reduction

and precipitation were also investigated.

3

LITERATUtRE

Conventional treatment of mixed metal wastewaters containing hexavalent

chromium is a multi-stage system (24,34a). Acidic (pH~3) chromium

reduction, utilizing either a ferrous salt or a reduced sulfur compound

(S2, H2SO3 etc.) is followed by hydroxide precipitation at alkaline ph

(8-10). Solids removal is normally accomplished by sedimentation with

improved removal efficiencies achieved by filtration.

The use of acidic and alkaline processes complicates the system and can

result in excess chemical consumption. Alkaline reduction of chromium

would result in combining the reduction and precipitation operations, and

thus reduce capital and operatiorz.I expenses.

Problems with hydroxide precipitatin result from the relatively high

solubility of metal hydroxides. The for-iaton of soluble metal hydroxide

complexes at high pH's prevents optimun-. rcmcval of a mixed metal solution

at a common pH. Figure 1 shows t-hat minimnm hydroxide solubilities of

cadmium, chromium and nickel are not achieved at a common pH, and that safe

drinKing water standards fcr these metals cannot be achieved by hydroxide

precipitation.

It has been retorted (4,25) that, since metal sulfide solubilities are

considerably less than corresponding hydroxides, impr oved removal

efficiencies can be achieved by sulfide precipitation. Figure 2 shows the

solubility of chromium I), nickel (II) and cadmium (II) in the presence

of 10-6 M total sulfide. A total sulfide concentration of 10-6 M was

i4

has been assumed that solubility of this metal is

Minimum solubilities for these three metals of int

pH between 8 and 9. Increasing the pH beyond

solubility due to formation of soluble metal hyd

strong reducing agent and could participate in chror

The distribution of sulfide species with pH use

Figure 2 is shown as Figure 3. Below pH 7 virtual]

present as H2 S, a toxic gas. An alkaline pH fi

process helps to minimize the evolution of this toxi

Competing systems for sulfide treatment ar

insoluble sulfide precipitation (SSP and ISP reE

consists of adding a soluble sulfide compod (Na2S

reactor at a controlled pH. Proble-s associated wi

sulfide (H2 S) odor problers and removal of the fI

metal sulfide precipitates. Solutions of the odor

of reactors, close control of sulfide dose with a

and oxidation c residual sulfide with peroxide

connonly used to flocculate the colloidal precipitat

ISP (Sulfex Process, consists cf the addition

FeS to the reactor (29,3,31). Since ferrous

soluble, it is a zource of sulfide for precipitation

flow steam and yet controls sulfide solubility

sulfide problas associated with SSP overdosing.

A third system of sulfide precipitation for the

5

Flocculation and solids removal are accomplished directly in ai

filter.

It is fairly well established that reduction of hexavalent chr,

reduced sulfur compounds is reasonably rapid at pH below 3 and t

rate of this reaction slows logarithmically with increased pH.

There is some disagreement, however, concerning the effects of p-

rates of chromium reduction by ferrous iron. Using reactio

determined by Espenson (15,16,, Thomas (34a) estimated the redu'ti

typical chromi-n concentration to be 99% completed in approxim

seconds at pH 2, and 90 minutes at pH 3, with the rate decreasin

orders of magnitude for each additional unit increase in pH.

authors (9,20,24,36) have stated that a pH of less than 3 is requi

reasonably rapid reduction of chromium by ferrous sulfide. Germaine

(18) claimed that ferrous sulfate was effective regardless of pH.

are reports of effective chromium reduction without pH control. (5)

Precipitated ferrous sulfide (ISP) is reported to effectivf

rapidly reduce nexavalent chromium at alkaline pH (28) . Addition of

sulfide was ineffective for chromium reduction at pH above 7, (35) a

when ferrous sulfate was added in combination with sodium s

reduction was effected (34a). It was postulated (but not demons

that ferrous iron is tne effEctive reducing agent in the sulfex

(35) and that ferrous iron acts as a catlyst for chromium reducl

sulfide at alkaline p 4-'.

Following precipitation, rnioval of the solid metal hydroxi

sulfides is normally accomplished by flocculation with polyne

sedimentatrin. ?urther addition of polymer and filtration is n

required to insure consistently hich metal removal efficiencies.

The main disadvantage of conventional downflow sand filters is that the

finest material is encountered first, and the greatest portion of the

solids removal occurs in the top layer of the bed. As a result, high head

losses develop in a short period of time, and the bottom portion of the bed

is poorly utilized. The coarse to fine gradation in an upflow filter would

indicate that good re-ocvals can be obtained without the head loss problems

of conventional sand filters. Diaper and Ives (13) conducted a study on

upflow filtration and concluded that: for a given headloss, upflow

filtration allowed longer runs due to a more even distribution of solids;

and upflow filtration has some disadvantages which include (a) a tendency

for the bed to expand as pressure differences overcome the weight of the

media, (b) difficulty in backwashing layers of material which tend to

accumulate, (c) the possibility of deposits :n the underdrain which may be

difficult to remove, and (d) size graded media are best used with high

flows and high concentration of solids.

HamaaT. and McKinmev (18a) reported similar advantages of upflow

filtration. They also describe additional advantages: (a) a reduction in

the size of the physical plant since flocculation and sedimentation

facilities are not required; and kb) a reduction in the amount of chemical

coagulant is required tecause a large, easy to settle floc is not needed.

They found that effectiveness improved with bed depth.

Upflow filters have been found to perform as well or better than

downflow filters for removal of turbidity (18a). Average reductions i-

bacteria have also teen found to be approximately equivalent to those in a

convertiona plan t (19)

If there is a more .niform distribution of solids with upflow

filtration than with downflow, upflow should have less tendency to cake or

nudball and be easier to clean (17). However, the studies by Diaper and

Ives (14,19) and Hamaam and McKirney (18) have found difficulties with

backwash. Bands of deposit tend to form within the bed (14,19). During

backwash the bands of deposit are pushed into finer sand where the clogged

media formea an impervious barrier. The barriers caused an arching effect

which lifted the bed in layers (6,11). Compressed air was found necessary

to break up the bands of deposit and adequately clean the entire bed

(6,7,11).

" -' " - '". . =* , ..'

-,- i I : i : - -" ; " - " - . ' . .*- - , .. .. .. . . '

eI'Z1 ,

-

p H

Ficure I. SclubLiit;.es c' Cd, Cr ad Ni as 3a function of cH.

i.~[-

C0

NI Ss

-J

a24 6310 12 14

p HF'iglre 2. Solubilities off d Cr andi N:. wit-h 10-6% total sulfides.

JL-

10

C-

6-

62

p HFiaure 3. Distrib2t~cr ci' s-ulfide spocies with pH.

RESEARCH OBJECTIVES & RATIONALE

Three metals, chromium, cadmium and nickel, were selected for

experimentation due to their usc in Air Force electroplating shops. Batch

studies were performed to test process steps and optimize chemical dosages.

Tests were designed to determine whether sulfide or ferrous iron or a

combination of the two is responsible for the reported reduction of

chromium (VI) at alkaline pH. A pH range of 7 to 10 was selected to

minimize hydrogen sulfide evolution and heavy metals solubility.

It is reported that freshly prepared ferrous sulfide precipitate was

more effective than aged precipitate in metal removal. Tests were

performed to see if improvements in chemical usage efficiencies could be

accomplished by separate addition of ferrous and sulfide to the

precipitation reactors and to determine if combinations of iron to sulfide

other than one to one were effect:ve in metal removal. It is postulated

that addition of soluble iron to the reactor would elimirnate the need for

additional coagulant chemicals.

The processes selected for testing in continuous flow experiments were

chromium reduction, pH adjustment, metal precipitation and coagulation by

the addition of ferrous sulfate, sodium sulfide and sodium hydroxide in one

reactor followed by solids removal in an upflow filter. These processes

were selected to provide maximum process control in a compact package.

Separate addition of ferrous and sulfide in optimal combinations provides

highly reactive chemicals for precipitation and coagulation.

Filtration is normally used in the sulfide precipitation process.

Between sedimentation and clltraticn units there is usually a pumping unit

employed to provide tr-e required filtration head. Pmping creates

turbulence and tends to break up flocculated solids. To keep solids

flocculated, additional coagulants are used after pumnping to facilitate

good filtration. Direct filtration would preclude the need for double use

of coagulants.

Separate control of sulfide and ferrous iron allows use of the ferrous

iron for coagulation. Excess sulfide not precipitated by heavy metals can

be removed by iron precipitation. Iron trapped in the filter would be

available if needed during fluctuations in influent metal concentrations.

Upf low filtration was selected because of previous experience in metal

sulfide precipitation which showed longer filter runs due to com~plete use

of the filter (17).

13

METHODS AND MATERIALS

waste C Chromium (VI), cadmium and nickel were selected for

testing due to their conmon use in Air Force electroplating shops.

Synthetic wastewaters were prepared using reagent grade salts and distilled

or tap water. Metals were added as potassium dichromate (K2Cr2O7), cadmium

chloride (CdCl 2 ) and nickel chloride (NiCl 2 ). The effects of competing

ions was determined by the addition of hardness as calcium chloride

(CaCI 2 ), alkalinity as sodium bicarbonate (NaHO3), and complexing agents

as disodium EDTA (Na2H2EDTA) or sodium cyanide (NaCN).

Batc Treatment All batch treatnent experiments were conducted in

borosilicace glass beakers stirred at a moderate speed by a teflon coated

stirring bar on a magnetic stirrer. The pH was monitored with a glass

electrode and double junction reference electrode with a pH meter. The pH

was adjusted manually by the addition of dilute sodium hydroxide (NaOH) or

nitric acid (Hm4 3 ) as necessary. Stock solutions of ferrous sulfate

(FeSO4 ), acidified with 5 drops (0.25 ml) of nitric acid ( /NO3 ) per liter

to prevent oxidaticn, and sodium sulfide (Na2 S) were prepared fresh daily.

Batch synthetic waste was prepared from stock solutions. An aliquot

was transferred to a stirred beaker, and pH adjusted to the selected set

point. Doses of ferrous sulfate and sodium sulfide were added by pipet, as

the pH was manually maintained at the set point. Doses were based on

stoichiometric requirements for reduction of chromium (VI) to chromium

(III), assuming oxidation of ferrous (II) iron to ferric (III) and sulfide

(-II) to elemental sulfur (0).

Cei las An atomic absorption spectrophotometer was used for

metals analysis. Chromium analysis was performed by direct aspiration into

14

a reducing nitrous oxide-acetylene flame; cadmiuar into an oxidizing

air-acetylene flame; nickel into an oxidizing air-acetylene flame with

background correction for non-atomic atsorption. Other cne~rical analyses

were performed in accordance with "Standard Methods" (8).

CFow Aparat, s A schenatic flow diagram Cf the Vreatment

apparatus is shown in Figure 4. A 13-foot length of clear plastic tubing

is used for the filter colun. The column contains a 5-foot layer of sand

supported by a 6-inch laver of gravel.

Manometer ports are located at the top and bottom of the filter bed and

at one-foot intervals. Manometers are water-filled to allow a direct

reading of the head loss. Due to height iLitations mercury -nometers are

used for the bottcm head readings.

Waste streams are prepared by dilutinc stock metal solutions prepared

from metal salts with distilled water in a waste tank. Waste and treatment

chemicals are fed to a mixer using variable flow pumps. Ferrous sulfate

and sodium sulfide feed solutions are made fresh daily. Ferrous sulfate

solutions are preserved with nitric acid (0.25 ml/l).

Tap water is used for backwashing through the same inlet used for the

influent, compressed air injection is provided to aid in breaking up filter

sand balls. The backwash effluent line is located five feet above the

normal filter bed, allowing for sand expansion during backwash.

15

MANOMETER TU BE S

8A W AS H

FILTER EFFLUENT7

SODIUMP U MR PO

M ET AL S AL T

IN TAP WATER

STATICMIE

PUM~tT ~ eGRAVEL SUPPORT

FERROUS

SULFATE -BACKWASH NATER

S U L F0E A AIR

P UJ M P

Figure 4. glow filtration arzparatus.

16

RESULTS

Chromiu Reduction ad Removal Studies were designed to determine if

chromium (VI) reduction was accomplished by sulfide, ferrous iron or a

combination of the two at alkaline pH. Synthetic wastewater with a

hexavalent chromium concentration of 10 mn/l (0.58 mecll) , was prepared in

Tyndall Air Force Base tap water (Table 2).

TABLE 2 Typical Analysis of Tap Water, I ndall Air Force Base, Florida.

Constituent Concentration

pH 8.1Alkalinity (mg/l as CaCO) 34Total Hardness (ng/l as CaC%3) 78Calcium Hardness (mq/l as CaCO3) 70Cl (mgil) 16TDS 'mg,/l) 134

Negligible chromi'a removal was effected bv sulfide alone at pH 7 throuch

10 (Figure 5). This supported findings by Walden (35) that sulfide alone

is ineffective for chromium reduczicn at alkaline pti. Reduction and

removal of chrcmium was accomplished with a stoichiometric dose of ferrous

iron, compared to a 3:1 dosage for the precipitated ferrous sulfide

(sulfexi process (4) This removal by ferrous iron occured independently

of pH between 7 and 10. Addition cf sodium sulfide enhanced reduction and

removal of chrom-nn by ferrcus iron at oH 7 ,Ficure 6). To a decree,

sulfide replaced ferrous iron on a stoichiometric 'asis. A minimum

concentration of ferrous iron was nceded for sulfide to be effective. This

seems to substantatt ;.alden's (35) cor.ectre that ferrous iron acts as a

catalyst for sulfide reduction of chromi,, at alkaline pH. At pH 8.5

17

(Figure 7) the enhancement of reduction by sulfide diminished and

disappeared at pH 10 (Figure 8).

Another way of showing the combined effects of ferrous and sulfice

doses on equivalent chromium (VI) removal efficiencies is presented aE

Figures 9, 10, and 11. At pH 7 ferrous iron addition was sufficient to

reduce and precipitate the chromium (VI), on a stoichiometric basis. To a

degree, sulfide replaced iron on a stoichiometric basis. A minimum

concentration of ferrous iron is needed for sulfide to be effective.

Increasing the pH of reaction to 8.5 (Figure 10) has negligible Lmpact

on the effectiveness of ferrous iron in renmoval of chromium (VI).

Increased pH, however, reduced the effectiveness of sulfide addition, even

in combination with ferrous iron (Ficure 11).

Cis

08 0

0 2

00

S- DOS E ~meq/I

0 6 . . . . I °-I

SYMBOL SUL FDfE mIR, I

0 0

0 0 145

0 029

0 0.2 0.4

Fe++ OOSE {meq/UFjcrare 6. cnroru= reduction by, ferrous sulfate and socl'.

sulfide at 3H 7.

20

SYMBOL SULFIDE me q

06 0 0

D0 145

0 029

0435

""- 04 t -J

~02

OQ M02 a4 06

Fe ++ DOSE imeq/I]Figure 7. Cnrc~raumr reduction b ferrous s-,i-lfztc and sodiur,

sulfide at pH 8.5.

21

SIMB0L SULFIDE mep

0 0

S0 145

0 029

04 0 435

_= -1

c..02 ('-

0I

02 04 06

Fe++ DOSE [meq/I)Figure 8. Chra mitr. reduction by ferrous sulfate and sodium

sulfide at pH 10.

22

05

I 0,

++ 0

0 3

0 0 2 05 4 06

S DOSE [meq/I]Figure 9. Isoequivalent chrcrnitrn residuals (rre-q,") f or

ferrous and sulfide doses at pH 7 (initial chrcirm=0.58 rreq/l).

23

I' 7 -

O..j 4

S= OS E ineqFigure 10. isoc-quivalent chrcraii. residuals (rreqjD for .>v'rrOus

and sulfide doses at pH 9.5 (initial chrc-dar, 0.58

MEL

24

0 6 1 I I I

0 6

CoLL

~02

Lon

4

_ _ 22I I1

C02 04 06

S= DOSE meq,/11

Figure 11. Isoequivalent chromia, residuals (neq/1) forferrous and sulfide doses at pH 10. (initialchromia = 0.58 ,q/i).

25

Ef fts 9f Hardness A2ini ad Al " agent Initial studies

were performed on a synthetic wastewater prepared from a soft, low-alkaline

tap water (Tyndall AFB). To investigate interferences of ions normally

found in electroplating wastewaters, studies were performed on synthetic

wastewaters prepared from reagent-grade salts of these ions and distilled

water. These studies were performed at pH 8.5.

Studies of reduction and removal of chromium from synthetic wastewater

prepared from distilled water were identical to those performed with

Tyndall tap water (Figure 12). Ferrous iron rapidly and stoichiametrically

reduced the chromium. Sulfide alone was ineffective. In combination with

ferrous iron, sulfide aided in chromium reduction.

Alkalinity (NaHOD3 ) was found to slightly interfere with chromium

reduction and removal at low and intermediate ferrous sulfate doses

(Figure 13). The interference by alkalinity is more pronounced when

sulfide is used along with ferrous iron to reduce chromium (Figure 14). In

both of these examples, floc size was greatly reduced with increasing

alkalinity. This effect is not apparent in the figure due to the use of a

membrane filter for solids removal. Reduced floc size would have a

significant effect on removal efficiencies in an actual treatment system.

In spite of this, interference by alkalinity would be classified as slight

and its effect is eliminated by stoichicmetric doses of ferrous sulfate.

Figure 15 shows that the addition of calcium hardness has little effect

on chromium reduction and removal by ferrous sulfate alone. However, the

addition of calcium hardness to the colrbination of ferrous sultate and

sodium sulfide enhanced chromium reduction and removal (Figure 16). In

slightly alkaline water hardness had no effect on treatment with ferrous

sulfate alone (Figure 17), but slightly improved the effectiveness of

I! •~-~

f 26

sulfide (Figure 18), eliminating the alkaline interferences. In addition,

the presence of hardness resulted in improved floc characteristics,

partially eliminating the adverse effects of alkalinity.

To determine the coibined effects of hardness and alkalinity in

treatment of hexavalent chromium in a typical hard water, synthetic wastes

were prepared in "Tempe Tap" Water (Table 3).

TABLE 3

Analysis of Tap Water, Tempe, Arizona.

Constituent Concentration

pH 6.2Alkalinity (n/l as CaCD3) 100Ca Hardness (nx/l as CaC03) 150Total Hardness (mg/l as CaCO3 ) 165904= (rg/1) 60C- (mg/l) 44Na+ (Rn/l) 24

Figure 19 shows the removal of chromium by ferrous sulfate and sodium

sulfide for synthetic wastewater prepared fran "Tempe Tap" water. There

are no significant differences from similar runs with waste prepared from

distilled water or "Tyndall Tap" water (softened low alkaline). Improved

floc characteristics are noted with increased hardness. Apparently in the

hard alkaline water (Tempe Tap) the opposing effects of alkalinity and

hardness cancel each other out. It is also noted that no significant

interferences to chromium reduction and removal are found in either of the

tap waters tested.

Treatment of 100 mg/l of hexavalent chromium in "Tempe Tap" water by

ferrous and sulfide (Figure 20) demonstrates the stoichiometric nature of

the treatment process. It also shows that the rate of chromium reduction

* -. ~( ~ 7~ 2 -!

27

by ferrous iron at alkaline pH goes rapidly to completion even at such a

high concentration of chromium.

Complexing agents (EDTA and cyanide) are used extensively in

electroplating bath formulations. EDTA strongly interfered with chromium

reduction and removal by ferrous sulfate and sodium sulfide (Figures 21,22)

in wastewaters prepared using distilled water. Floc size was significantly

reduced. When wastewater was prepared with Tempe tap water the effect of

EDTA was greatly reduced and floc characteristics improved (Figyures 23,24).

Cyanide is a strong complexing agent cormonly used in plating bath

formulations. Since cyanides form strong complexes with ferrous iron, it

was anticipated that the presence of cyanide would interfere with ferrous

reduction of hexavalent chromium. The molar ratio of CN to Fe in the

ferrocyanide complex is 6 to 1. Therefore concentrations of cyanide up to

6 times maximum ferrous dose were selected to determine cyanide's effect on

the process.

Cyanide interfered with chrominn reduction by ferrous iron for

synthetic wastes prepared in distilled water (Figure 25), with the adverse

effect proportional to cyanide concentration. Even at the highest

concentration of cyanide, some reduction was accomplished. The addition of

0.29 meq/l of sulfide (Figure 26) improved chromium reduction, but did nct

eliminate the cyanide interference. Cyanide was less of an interferent in

reduction of synthetic chromium waste prepared with Tempe tap water.

(Figures 27 and 28).

Cyanides interfere with chromium reduction by ferrous sulfate and

sulfide; probably by forming ferrocyanide complexes. These complexes are

hard to break and would interfere with cyanide treatment. It would be

advisable to segregate cyanide containing wastes for cyanide removal prior

to treatment with chromium containing wastes by this process.

I

"- .i ' - " - r - : - - - ,- -,..- - --- _. - -- - - _ _ : . ., . - . . . . . . , ,.-, '

28

06 r

SYMeC- SJLFi:E rreq,

3 290

or 045

E 0.4 M.inrQ'e Symncls

z0

z

0 02-0

0ijr 0 ?2fd at 04~

Fe + DOSE (meq /HFgue12. r o(:z rncuctiz,-n by :errous ---u, fatc anr-

sodi~nsllie tp 8.5 in distille6cwater.

29

SYMBCL AIK (mg/I aOO3)

050

06. o00200

__ 400

ZZ Multiple Symbols

Cr

zo 0.4-:

z

zO0.2-

00

0 0.2 04 06

Fe DOSE (meq/1)Figure 13. Effects of alkalirnitv on chranin reduction

by ferrous sulfate pH 8.5.

j 30

SYMBCL A; K (rng/ os C~-

00

04 0 u')lesm c

;-0 200.

ze OE(e/

-Fe -~ DOSE________/I

31

I I! I

SYMBOL HARDNESS rm(:/t CaC0 3 )

50

0.6 _ 0_200

0 400

t Multiple Symbols

E

z

z

0

0

0.2-

0 0,2 04 0.6

Fe DOSE (meq/I)Figure 15. Effects of calcian hardness en clironitur

reduction bv ferrous sulfate at pH 8.5.

32

SYMBOL HARDNESS (mg/I0 0 CaCOW

0.6 0 500 100A 200

-% 400" C Multiple Symbols

z0 0.4

z0 .2 o\

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 16. Effects of hardness cn craui

reductilon '% ferrous sulfate andsodi zn sulfide (0.29 at pH 9.5.

4 nec , tsH85

33

06SYMBOL HARDNESS (mg/i G

0© 00 50

Laon 200C> 400

4 - Multiple SymbolsE

04z0

X

zLiiU

~02-

0 1 I I

0 02 04

Fe* DOSE (meq/l)Figire 17. Effects of hardncss and alkalinity

(200 c.'l as kCa0,D3 ) on chr-iTz reauclfv ferrous sulfate at pH 8.5.

34

06SYMBOL H,4RDNES(rng'' CC

0 500 o OC" 200C 400

C::) Mui:pie Symbols

0.4

0 -

zw0 02

U

k-1

0

0 004 0.6

Fe " DOSE (meq/I)Figure 18. Effects of hardness and alkalinitv

(200 mg/1 as CaCO3) on chromi-Lim reductionby ferrous sulfate and sodium sulfide(0.29 meq/1) at pH 8.5.

P~

Ii~

35

S YMBOL SULFIDE (meqit)CQ 0

0.6 0 .145A .29

cr .58

04

z0.

z0

00.2

0

*0 0.2 0.4 0.6

Fe DOSE (meq/I)F~cue 19 r~niinreduction by ferrous sulfnte

and sodiun sulfide at pH 8.5 in a hardtap water.

36

! - L I NI ! u ,

SYMBOL MJLFVE NmeWO0 0

0* .

E 4M3

o PW*Ab~ Symb~ob

0

zeLU

0 2.0 4.0 6.0

Fe DOSE (meq/l)Fion e 20. Chrcnium (100 - /l/) reduction by ferrous

sulfate and sodiin sulfide at pH 8.5 in ahard tap water.

- . ... .. . .' * " I . .. " . .. . . . - . . . . . ..- , '- .. .|'- " T O"W Nl -

37

I

I II

SYMBOL EDTA (meq/i)0 00c 0 145

0.6 0 0 29058

04-

z0

z

w

Fe** DOSE (meq/I)Figure 21. Effects of WA on-chrcrndwi reduction by

ferrous sulfate at pH = 8.3.

I-.

38

I

t 0 6 ,.

SYMBOL EDTA (MM)0 0.0C1 0.145a 0.290 0.58

C Multiple Symboli-

wi 0.4-E

z0

I-C

zwU.) 0.2-zo

0

0 I 1 I I , I

0 0.2 0.4 0.6Fe DOSE (meq/I)

Figure 22. Effects of EDTA on chrorimn reduction byferrous sulfate and sodiurv sulfide(0.29 Teq/l) at pH 8.5.

T-lu

39

SYMBOL EDTA (meg/I)

0 00

0.6 C 0.58C= Multiple Symbol

zo 0.4

Cr_

zLii

z0

0.2-

01

0 02 0.4 0.6

Fe" DOSE (meq /I)Figure 23. Effects of EUFA on chraniun reduction by

ferrous sulfate in hard tap water at pH 8.5.

40

SYMBOL IEDTA (meq/1)

0 0.0ln 029

0.6 0 0.58

E

zo 0.4 0

w0z

0.2

00

0 02 0.4 0.6

Fe*4 DOSE (meq/I)Figure 24. Effects of =~A or. chraiui reduction by

ferrous sulfate and sodiumn sulfide(0.29 meq/1) in hard tap water at pH 8.5.

j 41

SYMBOL NaCN (mM)

0 0

0 3.4

z0 0.

0

0 0.2 0.4 0.8

Fe DOSE (meq/1)Figure 25. Effects of cyanide on chrcmtn reduction bv

ferrous sulfate at pH 8.5.

42

- YMBOL NaCN (mtv0 0

.580.8 1.16

0 &4

z0w .

0-z0

00.2

00 0.2 0.4 0.8

Fe DOSE (meq/I)Figure 26. Effects of cyaieo Iiai.n reductinb

ferrous sulfate and sodiz sulfide (0.29 reqll)at pH. 85

43

I ISYMBOL NaCN (mfMO

0 00E .58./ 1.16

00.E

z00

z

a02

0

0 0.2 0.4O.

Fe DOSE (meq/I)Figure 27. Effects of cyanide on chrclnium reduction by

ferrous sulfate in hard tap water at pH 8.5.

., I

21\

44

I I

SYMBOL NaCN (mM), o 00.6 .580.6 \ A 1.16

0 48

0, Multile Symbols

EC

z 0o 0.4 0

F

I.

Z !w0

0Q 0.2

0C

o 0

00

0 0.2 0.4 0.1

Fe DOSE (meq/i)Figure 28. Effects of cyanide on chxrrumr reduction by

ferrous sulfate and sodizcn sulfide (0.29 71a/lin hard tap 'water at pH 8.5.

:1 45

Rixed Metal Treatzen Treatment of a mixed metal waste was

investi.ated. A synthetic mixed metal wastewater was prepared in Tyndall

Air Force Base tap water using 5 mg/l (0.09 meq/l) of cadmium and 10 mg/l

of chromium (0.58 meq/1).

Removal of chromium by sulfide alone was ineffective. At pH 7

(Figure 29), removal of over half of the cadmium was accomplished without

sulfide addition, probably due to precipitation of cadmium hydroxide.

Addition of sulfide further increased cadmium removal, although a sulfide

dose of over 4 times stoichiometric was required to produce an effluent

with a cadmium concentration below 0.01 mg/l. At pH 8.5 (Figure 30),

additional cadmium removal is effected (by hydroxide precipitation) without

sulfide addition. Sulfide is less effective at removal of additional

cadmium except at doses greater than 4 times stoichiometric. Hydroxide

precipitation of cadmium is further improved at pH 10 (Figure 31). At this

pH, the addition of sulfide at doses above stoichiometric (.09 meq/!)

resulted in resolubilization of cadmium. Results are for multiple runs of

this experiment, both on cadmiz-i only wastes and mixtures with chromium.

Cadmium had no noticeable effect on chromium removal by ferrous sulfate

and sodium sulfide. Figure 32 shows lines of equal chromium residuals for

combinations of ferrous sulfate and sodium sulfide doses at pH 8.5. This

figure is identical to one prepared for treatment of a chromium only

wastewater (Figure 10). Again this demonstrates that ferrous iron alone

stoichiometrically removes chromium and that sulfide alone is ineffective,

except in combination with ferrous iron.

Conversely, cadmium is significantly removed by sulfide addition at pH

7 (Figure 33). Treatment by ferrous iron alone is ineffective and removal

of cadmium by sulfide is not significantly enhanced by its addition. At

46

pH 8.5, (Figure 34) ferrous iron addition greatly enhances the removal of

cadmium. In combination with ferrous iron, the dose of sulfide for

complete removal is reduced to stoichiometric (.09 meq/l). At pH 10

(Figure 35), the cadmium resolubilizing effect of sulfide overdosing was

eliminated by minimal addition of ferrous iron.

Synthetic mixtures of 10 mqil each of chromium (0.58 meq/1) cadmium

(0.18 meq/l) and nickel (0.34 meq/l) prepared in distilled water and Tempe

tap water were treated with ferrous sulfate and sodium sulfide at pH 8.5.

Treatment of this synthetic waste prepared with distilled water showed that

chromium was removed stoichiometrically by ferrous sulfate (Figure 36).

Addition of stoichiometric dose (0.58 meqil) resulted in a chromium

residual of less than 0.05 ,/I. Sulfide does not appreciably improve

chromium removal in this mixture, probably due to its use in precipitation

of tne other metals.

Cadmium was essentially completely removed (0.02 Tg/l residual) ty PH

adjustment to 8.5 (Figure 37). The addition of ferrous sulfate or sodium

sulfide had either no effect or increased the cadmium solubility. Likewise

nickel was removed to 0.15 mg/l by pH adjustment (Figure 38). Little

benefit was effected by ferrous sulfate and sodium sulfide addition.

Similarly for synthetic mixed metal waste in Tempe tap water, chromium

was removed stoichiometrically by ferrous sulfate (Figure 39). Sulfide was

of little benefit. The residual chromium concentration for a

stoichiometric dose of ferrous sulfate was 0.4 my/l, considerably more than

for treatment of the mixed metal waste prepared in distilled water. This

increased in concentration could be due to carbonate complexation.

Likewise cadmium is considerably more soluble in tap water at pH 8.5

(Figure 40) than in distilled water, also probably due to carbonate

I

• - I , - ' - - .- - a+ +" ' - " + - " --.-- ----- -.

I47

ccmplexation. Sulfide stoichiometrically reduces the cadmium

concentration. Ferrous sulfate also improves cadmium removal. Minimum

cadmium residual was 0.05 Nm/l.

Nickel remains very soluble in this tap water at pH 8.5 (Figure 41).

Ferrous sulfate and sodium sulfide improve removal, but the minimum

residual nickel concentration is still over 1 mg/l. It is noted

(Figures 1 & 2) that nickel sulfide and nickel hydroxide are considerably

more soluble than cadmium sulfide, and hence nickel is more likely to be

poorly removed by sulfide precipitation.

On the previous batch treatment tests the pH was adjusted by addition

of sodium hydroxide or nitric acid. There are reports (34a) that Lnproved

metal removal can be accomplished by the use of lime (Ca(OH) 2 ) for pH

adjustment, due to the improved coagulant properties of calcium over

sodium. A mixed metal waste was prepared in Te-mpe tap water and pH

adjusted to 8.5 with a dilute solution of lime. Removal of chromium was

stoichiometric as usual (Ficure 42). The chromiu m concentrations for

stoichiometric dose of ferrous sulfate were reduced to less than 0.05 rg/l,

as compared with 0.4 Nr/l for the test when pH adjustment was with sodium

hydroxide (Figure 43). Little effect is noted on cadmium removal. Minimum

residual cadmium was 0.03 rag!!. No benefit is seen for nickel removal

(Figure 44).

a.;am

II

I

48

Crx

= 4 L

I~ Cd

EF

0 2 o04

S DOSE meq/IFigure 29. Cadxmiun and chrcoium removal by sodiu.0 sulfide in

softeed tap water at pH 7.

I

06 JC1

C=,

C.C.d

*~ 0 2 0 4

S- DOSE meq/flj* Figure 30. Cadrtizn and chromim~ removal by sodi-n sulfide in

softened tap water at pP 8.3.

50

0 06

cr0

o0 CD

LA.J

C..'0

0

0 320 4 0 6

S DOSE 1meq;I) 1Figure 3 1. cadrdum rer-val by sodim.n sulfide in softened tap

water at pH 10.

51

0 6

04

+ 02+

@33

034

0 92 0 4 0 6

S DOSE HIeq/I)Ficgure 32. Isoecraivalent chrarii~t residuals (rreq/1) for treat-

trent of mixed -etals in~ softened tap water at pH 8.5.

f 52

06

0.4

5C,

NJP

0-02

00.2 0 4 0 6

S= DOSE Imeq/1)Figure 33. Isoequivalent cadmiu residuals (meq/1) for

treatient of mixed metals in softened tap waterat pH 7.

I

.. .. ... 1 .. . .- --.- --,,,, I -

53

I

II0 6

0 4

'I

E

002

00

02 04 0

S= DOSE 'meq/IjFicure 34. Isoequivalent cadmiun residuals for treatmrent of

mixed rtals in softened tap water at pH 8.5.

I

S54

06

.2

- - 0 5

U

0 0.2 0 .4 0 6

S- BOSE Imeq/1)Figure 35. Isoequivalent cadmiu residuals for treatment of

mixed retals in softened tap water at pH 10.

*1

I

55

SYMBOL SULFDE (MeQ/00 0o .146

0.6 & 290' .4350 .58

MLAOW Symbol

0*

z A-

0

z0

II

!z

z0 G0

0 0.2 0.4 0.6Fe DOSE (meq/I)

Figure 36. chranitrn (0.58 np-q/1) remtoval from amixed metal waste at pH 8.5.

56

I

I

SYMBOL SULFDE (meq/M0 0o .145A .29

1 350*0 .58CL Mukt* Symbels

-40

0 0.2 0.4 0.0

Fe DOSE (meq/I)Figure 37. Cadmium~ (0.18 meq/1) removal from a mixed

metal waste at pH 8.5.

'~i

I i i lII I 1 " " ' i. . . .

5 7

III II

SYMBOL SULF1DE 0neq/D0 00 .146

0Z 0M 29

E202 AM.20 .5a

M" Synb~iez0

c-

0.10

z0

C.,)

00 0.2 OA 0.5

Fe DOSE (meq/1)Figure 38. Nickel (0.34 nreq/1) removal from a

mixed mretal waste at pH 8.5.

58

SYMBOL SULFUE (meq/

0 0o .145

0. a .29

o "00 .

CYC

z00.4

zoAI-z

0

002

0 0.2 0.4 0.6Fe DOSE (meq/I)

Figure 39. Chrcmium (0.58 rrq/l) remval from a mixedmetal waste Li hard tap -dater at pH 8.5.

I!

I

59

0.06 , I I ,SYMBOL SULFIDE (meq/O

0 0o[ .145

A 29.435

0 58Q)

0.0

z0

c-z0Z 00

00

0A

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 40. Cadmilzn (o.18 meq/l) remval from a mixed

metal waste in hard tap water at pH 8.5.

I ! -

60

SYMBOL SULFIDE (meq/O0 o0 0 .145

0 .58[t :: Muftie Symb-ois

z0

LLI 0.10_ C A

0

0 0.2 0.4 0.6

Fe r.OSE (meq/I)Figure 41. Nickel (0.39 n~q/l) removal frm a mixed

metal waste Ln hard tap water at DH 8.5.

I"

C .i

61

SYMBOL SULFIE (meq/O0 0

0.6 290 .58-- MWtk* Symbob

E

% Vz

0 02

zw

0

0 0.2 0.4 0.6

Fc DOSE (meq/I)Figure 42. Chrondn (0.58 mrq/1) removal from a .-uxed

metal waste in hard tar) water (Ca(OH)2)with pH adjusted to 8.5 with lime.

1.7" - 1--

62

I

I

015 ,

SYMBOL SULFIDE (meq/Oo 0

A .29o .58M Mukk~e Symboi

00.10-

00

Z 0.02

0

ojt t

F0

00 0.2 OA 0.8

Fe DOSE (me qII)Figure 43. Cadmiumn (0.18 neq/1) rentval frcr a n.Lx.ed

Ital waste in hard tap water with piadjusted to 8.5 with lire.

MONONA ....I

63

II

0.4

SSYMBOL SULFE (mneq/0* 0 0

EA 290 .58z MuMple Symbob

0

020

zw 00z 0

0 p .0 0.2 0.4 0.6

Fe DOSE (meq/1)Figure 44. Nickel (0.39 meq/l) renval from a

mixed metal waste in hard tap waterwith pH adjusted to 8.5 with lime.

II

r - -.

64

Taten 2L Platin B Plating bath rinsewaters were

prepared by diluting plating solutions (from operating plating baths, Kelly

Air Force Base, Texas), in distilled water and Tempe tap water (Figures

45,46). Ferrous iron effectively reduced and removed the hexavalent

chromium. Sulfide addition benefitted treatment of the rinsewater prepared

from distilled water, although it was more beneficial in tap water. There

was no significant interference from anything in the plating solution to

alkaline chromium reduction and removal by ferrous sulfate and sodium

sulfide.

A mixture of plating baths, 10 mg/l each of chromium, cadmium and

nickel, were prepared in distilled water to simulate plating rinsewaters

that might be found in Air Force shops.

Chromium was removed stoichiometrically by ferrous sulfate (Figure 47).

The chromium residual for stoichiometric dose of ferrous sulfate was less

thza '. 5 mg/l. The complexing agents normally found in these plating

baths did not interfere with chromium reduction or removal.

Cadmium was considerably more soluble in the plating bath wastewater

(Figure 48) than in synthe.ic wastes produced fron metal salts (Figure 37).

This is probably due to the prescnce of complexing agents in the plating

bath formulations. Addition of ferrous sulfate alone resulted in some

cadmium removal, probably due to competitionj by the iron for the complexing

agents in the plating baths releasing cadmium to precipitate as a

hydroxide. In combination, the sulfide probably broke the complex,

precipitated the cadmium and the iron coagulated the resulting colloidal

precipitates. Residual cadmium was reduced to less than 0.05 mu/l by

treatment. Nickel was essentially unaffected by treatment (Figure 49).

Nickel sulfide is considerably more soluble than cadmium sulfide. Unlike

-AN-

65

j cadmium, sulfide is unable to break the nickel complex in the plating

formulation.

Treatment of mixed plating solutions diluted in tap water resulted in

the familiar chromium removal by ferrous sulfate with sulfide providing

some improvement (Figure 50). Residual concentrations of chromium for

stoichiometric doses of ferrous sulfate are less than 0.05 mg/l. Cadmium

remains soluble at this pH, due probably to the compl 'ing agents in the

plating baths (Figure 51). Ferrous sulfate provides some removal, probably

due to the competition mechanism mentioned previously. Sulfide reduces the

concentration to less than 0.01 mg/l. The removal pathway is probably

precipitation of cadmium sulfide followed by coagulation by the hardness

present in the tap water.

Nickel removal was negligible (Figure 52).

Adjustment of pH by lime provided slight improvement in chromium

removal from a mixed plating waste diluted in tap water (Figure 53).

Cadmium and nickel removal were unaffected by use of lime rather than

sodium hydroxide for pH adjustment (Figures 54 and 55).

iI

_ 1 w - .-- - ~-----

66

SYMBOL SULFIDE (meq/0o 0' .145

0.6 .290 .435o .58

Muftple Symbols

0 8,E% 0.4 .\

Zo

zw0

C)

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 45. Chraniun (0.58 meq/l) rermval from a

chromium plating waste in distilledwater at pH 8.5.1

I

f 67

SYMBOL SULFIDE (meq/0o 0o .145

0. .29C .4350 .58

~]Multple Symbols

c*

00

Z 0.4 0

0

z0 .20

00

0 0.2 0.4 0.6

Fe DOSE (meq/1)Figure 46. hLuaim (0.58 Tieqj'1) rerroval frcan a

chrcmixn piating waste in hard tap waterat pH 8.5.

68

II

SYMBOL SULMIE(Moq/Do 0a .29

OAo .580 MilM SyniboOIB

0 OA

I-

I-

zz

00

Z

00o.

0 02 0.4 O.6

Fe DOSE (meq/I)Figure 47. Chratiun (0.58 meq/1) remval fron a

mixed metal plating waste in distilledwater at pH 8.5.

1

1 69

0 0

0)A

z0

z0

0

00

0 0.2 0.4 .

Fe DOSE (meq/I)Figure 48. Cadmiumn (0.18 meq/l) rewrva1 frcrn a

mixed neta1 plating waste in distilledwater at pH- 8.5.

70

S 0.4

E0z

0

S0.2 0 .58z C= Mukk~~b Syntbob

z0

0

0 0.2 04 0.8

Fe DOSE (meq/I)Figure 49. Nickel (0. 39 mecij') rermval Fraxn a m'ixed

mretal plating waste in distilled waterat pH 8.5.

71

SYMBOL SULFIDE (meq/Ilo 0

0.8 .29o .58L~ Multiple Symbols

Z 0.40

z

z 0 .20

0

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 50. Chraniin (0.58 ireq,') removval fran. a mixed

nretaJ. plating waste in hard tap water atpH 8. 5.

72

0.15

E

o 0.10-

SYMBOL SULFIDE (meq/I)

I- 0 0Z .29WJ 0 .58

005M Multiple Symbolsz 0.o50

0

0-

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 51. Cadmium (0.18 req/1) remval frcr a mixed

metal plating waste in hard tap water atpH 8.5.

t

73

c" 0.4

E

zo0SYMBOL SULFIDE (meq/I)

X 0.2 0 0I- A.29Z .58Lw =Multiple Symbols0z00

0 0.2 0.4 0.6

Fe DOSE (meq/I)Fi.,-ure 2. .,ic.,I D.39 a,-/'!) remval from a :izxed

,retal clatL-c waste in hard tari water atpH 3,5.

It

74

SYMBOL SULFIDE (m0

A .290.8 .58

SMultiple Symb

0*

E

zo 0.4

I-z

Z 0.2

0

0

00.2 0.4

Fe DOSE (mec/I)Figu-re 53. Chroruim (0.-58 trc, 1 ) roe-cval fror

metal olatinc -wast:: n hard tap wat.714 adjustec toc 3.; with Iijr-.

75

I

Ia

0.15

00

EZ 0.100 SYMBOL SULFIDE (meq/i)

cc .29

. 0 .58

zt u-JZ Multiple Symbols

0.05

00 -I ::: ' " '-'

0 0.2 0.4 0.6

Fe DOSE (meq/I)Figure 54. Cadmiun (0.18 neq/l) renoval from a

mixed retal plating waste in hard tapwater with pH adjusted to 8.5 with 1L,e.

76

J% 0.4 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

0*0 0 0I0

SYMBOL SULFIDE (meq/O<0 0

cc 0.2- 2~.29z 0 .58W Multiple Symbols

z00

0 0.2 0.4 0.6

Fe DOSE (meq/I)Fic ure 55. Nickel (0.39 req/l) renval fram a nixed

retal plating waste in hard tap water withpH adjusted to 8.5 with lim.

I

77

Contl~inuou Z1QW Studies Limited studies were performed on a continuous

f low treatment system. Synthetic wastewaters were prepared in Tyndall Tap

water.

The results of selected continuous f low runs are presented in Table 4.

Sodium sulfide and ferric sulfate effectively removed cadmium at pH 9.4

(run 1). Head loss is low with a resulting long filter run. Chromium (VI)

reduction and removal was not accomplished with ferric iron and sulfide

addition at pH's at or above 7.

Runs 2 through 5 consisted of treatment of chromium (VI) and cadmium

containing wastewater with ferrous sulfate and sodium sulfide. Figure 56

depicts variations in effluent concentration6 and head loss during filter

run 4. Manual pH adjustment of the low alkaline water resulted in large

fluctuations in the beginning. Metal removal was constant despite pH

fluctuations. Head loss climbed steadily to breakthrough. In other runs

where high doses of ferrous iron was used relative to sulfide (runs 2,3),

floc sizes were increased, removal took place near the entrance of the

filter, better removal was effected, but head losses were higher and time

to breakthrough was reduced. Breakthrough was by lifting of the filter

sand from the bottom. Increasing the sulfide and lowering the ferrous dose

(run 5) resulted in lower head losses, but increased floc penetration in

the filter and chromium carry over to the effluent. Filter run times were

not greatly increased because breakthrough resulted from lifting of the

plugged fine sands at the surface of the bed.

Backwashing was accomplished by increasing the flow up through the bed,

producing a concentrated sludge with minimal water usage.

f 78

TABLE 4

Operating Conditions and Results for ContinuousFlow Treatment System

Run Number 1 2 3 4 5

Initial Cd (ag/l) 4.0 - 4.0 10.0 5.0(meq/l) 0.07 - 0.07 0.18 0.09

Initial Cr (rg/i) - 10.0 10.0 10.0 10.0(meq/1) - 0.58 0.58 0.58 0.58

Na2 S Dose (ag/1) 5.5 5.5 10.5 13.6 19.5(meq/l) 0.14 0.14 0.27 0.35 0.50

Fe SO4 Dose (mgFe/l) - 53 40 20 14(meq/1) - 0.95 0.72 0.35 0.25

Fe 2 (SO 4 ) 3 Dose (mg/Fe/l) 5 - - -

(meq/1) 0.24 ....

Filter Rate (gal/ft 2-min) 4 4 4 4 4

Effluent at 2 Hours

Cd (mg/l) 0.04 - <0.01 <0.02 <0.04(aweq/1) 0.0007 - <0.0002 <0.0004 <0.0007

Cr (ag/i) - <0.05 (0.05 1.9 <0.05(meq/i) - <0.003 <0.003 0.11 <0.003

pH 9.4 9.1 8.5 8.5 7.5

Head Loss (ft) at 2 hrs 1.5 6.5 5.4 2.5 2.4at breakthrough 3.6 6.7 6.5 4.0 2.7

Time to Breakthrough(hours) 9 2.1 2.5 3.5 2.5

II

* *~ A -

79

I

6 0.69

-4 0 4

I--

0 0 26

0 I 2 3 4

TIME jHOURSFigure 56. reatment of mixed metal wastewater by ferrous and

sulfide and upflow filtration.

80

CONCLUSIONS

1 Hexavalent chromium can be rapidly and stoichiometrically reduced byferrous sulfate at alkaline pH (7-10).

2 Sodium sulfide alone was ineffective as a reducing agent for hexavalentchrcmium at alkaline pH.

3 In combination with ferrous sulfate, sodium sulfide aided in chromiumreduction at alkaline pH., replacing ferrous sulfate. Ferrous ironseemed to act a,, a necessary catalyst for this reduction.

4 Alkalinity, EDTA, and cyanide interfered with chromium reduction, flocformation, and removal by ferrous sulfate and sodium sulfide.

5 Calcium hardness counteracted these interferences. Use of lime for pHadjustment would probably be preferable to soda ash.

6 Chromium reduction and removal by ferrous sulfate and sodium sulfidewas virtually identical for wastes prepared in distilled water, in asoftened, low-alkaline tap water and a hard, alkaline tap water;however treatment of hard water resulted in larger, better settlingfloc.

7 Cadmium was effectively precipitated by sulfide at pH 7-10, althoughdoses greater than 4 times stoichiometric were required. Overdosing atpH 10 increased cadmium solubility. These problems were eliminated bythe addition of ferrous sulfate.

8 Treatment in a mixture of cadmium and nickel had no noticeable effecton chromium removal. In distilled water, cadmium and nickel wereeffectively removed by hydroxide precipitation at pH 8.5. In a hard,alkaline tap water, cadmium and nickel were largely soluble at pH 8.5,probably due to inorganic complexation. Ferrous addition aided incadmium removal, probably due to competition for the effectivecomplexing agent. Sulfide effectively precipitated the cadmium, butnot the nickel, probably due to the appreciably lower solubility ofcadmium sulfide compared to nickel sulfide. Adjustment of pH with lime(rather than sodium hydroxide) improved floc size and settling, but hadlittle effect on metal removal.

9 Treatment of plating rinsewaters prepared from plating bath solutionsin both distilled and tap waters demonstrated a lack of significantinterferences to alkaline chromium reduction by ferrous sulfate andsodium sulfide. Cadmium was not removed by hydroxide precipitation,even in distilled water, probably due to complexation with constituentsin the plating baths. Ferrous sulfate improved cadmium removal,probably due to competition for the complexing agents. Sulfideprecipitation was sufficient to break these complexes. Ferrous sulfatewas required to coagulate the sulfide precipitate in distilled water.In hard tap water, hardness cations provided sufficient coagulation.Nickel was not removed by treatment of the mixed plating bath waste,apparently due to the high solubilities of nickel hydroxide andsulfide.

81

10 Upflow filtration in conjunction with direct addition of ferroussulfate and sodium sulfide effectively removed a mixture of hexavalentchromium and cacnium fram solution. The system provides a single step(unit process) treatment of mixed metal wastewater. However,additional work is needed to optimize chemical addition and filtermedia to increase length of filter runs.

82

ONGOIN RESEARCH

Ojecie The objective of future research is to develop and test a

pilot plant incorporating alkaline ferrous reduction of hexavalent

chromium, metal sulfide precipitation and solids separation processes.

Individual work areas have been identified as:

A. Investigate effects of chemical dosages on solids removalcharacteristics (floc size) of metal precipitates.

B. Improve run times in upflow filtration of metal precipitatesby:

1. including pre-sedimentation without intermediate pumpingprior to filtration (can cause floc breakup);

2. improve depth and size characteristics of media used inupflow filters.

C. Develop control methods to pace chemical addition to variations

in waste flow rate, metal concentrations pH, and alkalinity.

D. Plan, design and construct pilot plant.

E. Test pilot plant on synthetic wastewaters, making modificationsas necessary.

Proosed Process De These findings are substantially based on

batch studies using membrane filtration for solids separation. While these

results compared favorably with results using upflow filtration, it is

proposed that a pilot plant be developed, incorporating alkaline ferrous

reduction of hexavalent chromium, sulfide precipitation of metals and iron

coagulation for possible use by small Air Force electroplating facilities.

hI treatment objective would be to produce an effluent of higher quality

than conventional metal hydroxide precipitation and to do so with a compact

treatment process (elimination of the separate acidic reduction step, and

possible use of upflow filtration). Direct upflow filtration produced an

-I

83

effluent with low concentrations of metals, but resulted in short filter

runs. Work needs to be done to extend filter runs by such modifications as

increasing filter depth, varying filter sand effective size and uniformity,

or by providing sedimentation prior to filtration.

Intrmnain Md Coto Because of the importance of pH1 and

concentrations of the sulfide and ferrous ions on the proper operation of

the process, these variables should be constantly monitored and controlled.

The initial phase of this portion of the research will be a study of

the currently available instrumentation which would permit continuous

monitoring of the variables mentioned above. (A suitable reference is the

"Survey of the Instrumentation for Environmental Monitoring" put out

annually by the Lawrence Berkeley Laboratory) . Cost, sensitivity,

suitability for use in a production environment, and response time would

all be factors to be considered.

The next phase will be directed to a study of the process itself in

terms of developing models based on expected operating parameters such as

waste stream flow rates, levels of ions, etc. Such a model would lead to a

decision as to whether the process scheme should be continuous or perhaps

run as a batch system. For low volume effluents, the batch process would

generally be preferred while a continuous process is usually optimal for

processing large volumes from continuous plating operations.

Once the processing scheme is established for the laboratory scale

prototype, the attention will be directed toward the design arnd

implementation of a control system which will allow setting and maintaining

desired process conditions such as the feed flow rate, pH, etc. Two

different control strategies, feedforward and feedback, will be examined.

In feedback control, the most coimmonly used, the end result is measured and

j 841

comnpared to a desired value and then some action is taken to modify an

input to the process to bring this output to the desired level. An example

would be the monitoring of the sulfide ion concentration in the reactor and

based on the difference between the actual value and the desired, the flow

rate of sulfide solution added to the reactor would be adjusted.

Feedforward control makes use of a measurement of an input to the process;

for example, the concentration of the nickel ion in the feed stream to

calculate what amount of sulfide ion solution to add to the reactor to

maintain a desired sulfide level. This strategy requires a good model or

description of the reactions in order to predict the amount of sulfide to

add. If properly used this strategy will prevent upsets in the reactor

output that could occur using feedback control. It may' be that the final

scheme would involve a combination of the two strategies.

Because of the number of control loops on the system (waste stream

feed, pH, sulfide and ferrous ion) and the number of pramreters to be

monitored, it may be advantageous to use a microcomputer-based real-time

data acquisition and control system rather than conventional analog

controllers and recorders. The pros and cons of these alternatives will

also be studied both with respect to the laboratory research system and use

* with a final plant-scale unit.

Pilo Plant~ Proposed pilot plant would be built and tested using

synthetic plating rinsewater during the first year of the study. If the

* process developed proves promising, the pilot plant would be moved to an

appropriate electroplating facility for field testing during a follow-on

year of the project.

mail, .i

85

BIBLIOGRAPHY

1. "EPA to Propose Relaxed Effluent Guidelines", P.agng d SurfaceFi nhing, 28 (April, 1980).

2. al Cr i f Water, USEPA, Office of Wate' and HazardousMaterial (July, 1977).

3. E In Wastewater Treatment At Tinker AM, OX, USAFOccupational and Environmental Health Laboratory, AFSC, Brooks, AFB,Texas.

4. Control and Treatment hn fr he M l Finishing nu -sulfide Preciitatign, U.S. EPA Technology Transfer Summary Report,EPA 625/8-80-003 (1980).

5. Industrial Operating Treatment Logs, Kelly AFB, Texas.

6. "Chloralkali Industry Has Hg Headache," Canadian Checal Processing,54/10, 44, (Oct. 1970).

7. "Extraction cuts Mercury From Wastewater, Solids", C News,Chemical Manufacture Assoc. (Mar, 1973).

8. Standard Methods o Eamination f Water and Wastewater, 14thEdition, American Public He, -Lh Association, New York, (1976).

9. Bennet, J.R. "The Treatment of Effluents from Metal Cleaning andFinishing Processes", Metal Fini n Journal, 1, 272 (1972).

10. Bhattacharyya, D., Jumawan, A.B., & Grieves, R.B., "Separation ofToxic Heavy Metals by Sulfide Precipitation", Separation Science _&Technol , 1A, 441 (1979).

11. Conley, W.R. Jr., Integration of the Clarification Process", Journal&MA, 5, 1333 (1965).

12. Conley, W.R. Jr., and Pittman, R.W., "Test Program for FilterEvaluation at Hanford", Journal &AW, 52, 205 (1960).

13. Diaper, E.W.J., and Ives, K.J., Filteration Through Size GradedMedia", Jo.= n. jgr. Div., Amr. QSo. Civi ., 9, SA 3, 89(1965).

14. Diaper, E.W.J., and Ives K.J., "Discussion, Filteration Through SizeGraded Media", Zur, San. LaU. Div., ProQ. A=nr. S&. Civil .2Z, SA 4, 73 (1966).

15. Espenson, J.H. and King, E.L., "Kinetics and Mechanisms of Reactionsof Chromium (VI) and Iron (II) In Acidic Solution", Journal ofAmerican Chemical , 85, 3328 (1963).I

86

16. Espenson, J.H., "Rate Studies on the Primary Step of the Reduction ofChromium (VI) by Iron (II)," Journal of he American Chemical

2, 1880 (1970).

17. Fox, G.T.J., "Discussion, Filtration Through Size Graded Media", Jour.n. Z=. Divr.A= -w.. Q zil =., 2Z, SA 1, 326 (1966).

18. Germain, J., Vath, C. & Griffin, C. "Solving Complex Waste DisposalProblems in the Metal Finishing Industry", Paper presented at GeorgiaWater Pollution Control Assn. Mtg. (1968).

18a. Hamaan, C.L., and McKinney, R.E., "Upflow Filtration Processes",Journal AWWA, 60, 1023 (1968).

19. Ives, K.J., "Discussion, Russian Water Supply and TreatmentPractices," Jour;. San. . Div., Ae. Civil .Ligjr., 8, SA 4, 101(1960).

20. Kunz, R.G., Hess, T.C. & Yen, A.F. "Kinetic Model for ChromateReduction in Cooling Tower Blowdown", JWPCF 52, 2327 (1980).

20a. Lantz, J.B., "Evaluation of a Developmental Heavy Metals WasteTreatment System", Tech. Report to Civil Engineering Laboratory, NavalConst. Bat. Center, Port Huenene, CA, May (1979).

21. Latimer, W.M., The Oxidation States at the Elements and TheirPotentials jn Aqueous Solution, Prentice Hall, 1952.

22. Ling, J.T., "Progress in Technology of Water Filtration", Water andS Works, J0, 315 (1962).

23. Mracek, W.A., "Control & Automation of Chromate Waste ReducticnPlants", Engineers Society of Western Pennsylvania Proceedings (1969).

24. Patterson, J.W., Wastewater Treatment Technoloc-, Ann Arbor Science(1975).

25. Robinson, A.K., "Sulfide vs. Hydroxide Precipitation of Heavy Metalsfrom Industrial Wa.tewater", First Conference Advanced PolltionControl fr he Metal Finishing Industry, p. 59, EPA-600/8-78-010(May, 1978).

26. Saam, R.D., Hearst, P.J., Schwab, L.K., & Chan, D.B., "Drydock WasteTreatment Study", TN no. N-1490, Civil Eng. Lab., Naval ConstructionBatallion Center, Port Huenene, CA (1977).

27. Saltzman, R.S., & Bonam, W.L., "Continuous Monitoring of Low ChromiumVI Concentrations with a Photometric Analyzer System", ISA AD 74418,95 (1974).

28. Schlauch, R.M., and Epstein, A.C., "Treatment of Metal Finishingwastes by Sulfide Precipitation", p. 267-284, EPA-600/2-77-049 (1977).

87

28a. Schroeder, D.C., & Lee, G.F., "Potential Transformations of Chromiumin Natural Waters", Water, Air, and Soig P u , 4, 355 (1975).

29. Scott, M.C., "An EPA Demonstration Plant for Heavy Metals Removal bySulfide Precipitation", S Conferene n Advanced PoUlutioControl for l F inishing Industry, p. 106-113, EPA-600/8-70-014(1979).

30. Scott, M.C., "Sulfide Process Removes Metals, Produces DisposableSludge", nduil W e, July-August (1979).

31. Scott, M.C., "Sulfex - A New Process Technology for Removal of Heavy

Metals From Waste Streams", Proceedings, 32nd Industrial Wat

Conference Pr University, 622 (1977).

32. Sillen, L.G., and Martell, A.E., Stabilit Constants ol Meta Ion_o~lexe$, The Chemical Society, London (1964).

33. Smith and Martel, Critical Stabilit Constants, Vol 4 InorganicQomplexes, Phenum (1976).

34. TerMaath, S.G., Hea Metals Removal by Sulfide Precipitation,Coaoulation ad Filtration, M.S. Thesis, University of Texas atAustin, (1974).

34a. Thomas, M.J., Colloi Properties of Chroggic Hydroxide Applied to 'eta

Finishinq Wastes, Ph.D. Dissertation, University of Notre Dame, SouthBend, IN (1975).

35. Walden, Enineerin alysis and Feasibility St ;d for Treat -OftPlating Shop Rinsewaters at Tobyharna Ar Depot (1979).

36. Watson, M.R., Pollution Control in Metal Fini.hinc, ..cyes "Park Ridge, NJ (1973).

37. West, C.M., Bonner, W.P., and Bust1Tante,......From Wastewater Treatr'ent Plant "y Ch-:..1 : -"-

Purdue Id. Waste 11 7