EPA-670/2-74-008 January 1974

METALLIC RECOVERY FRCM WASTE WATERS UTILIZING CEMENTATION

BY

Oliver P. Case

Grant No. S-802254 Program Element 1BB036

Project Officer

Richard Tabakin Environmental Protection Agency

Edison Water Quality Research Laboratory National Environmental Research Center

Edison, New Jersey 08817

Prepared for OFFICE OF RESEARCH AND MONITORING

U.S. ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460

ABSTRACT

This report presents the results of a series of bench-scale experiments utilizing the so-called "cementation" reaction (i.e. electrochemical reduction by contact with a metal of higher oxidation potential) for the precipitation of copper and the reduction of hexavalent chromium in industrial waste streams such as brass mill and metal finishing dis- charges . Reductants studied included soft iron shot approximately 4.37 mm dia (0.172 in.), particulate iron (-325 mesh and -400 mesh), and silicon alloys in granular form.

The bulk of the work was done utilizing the iron shot as a reductant. Studies were carried out by both a batch process and a continuous process (back-mix reactor). Results were evaluated in terms of per- cent reduction, dissolution of excess iron and change in pH vs time.

The particulate Iron was studied utilizing batch processes only.

The silicon alloy reductant was studied utilizing both a batch process and a continous process (plug-flow reactor).

This report was submitted in fulfillment of Grant No. S-802254 by the Engineered Environments Division of the Anaconda American Brass Co., under the (partial) sponsorship of the Environmental Protection Agency. Work was completed as of October 1973.

, ii

CONTENTS

Abs trace

L i s t of Figures

L i s t of Tables

Acknowledgements

Sections

I

I1

I11

IV

V

VI

VI1

VI11

Conclusions

Recommendations

Introduction

Use of Soft Iron Shot as Reductant

Use of Particulate Iron as Reductant

Use of Silicon Alloys as Reductant

References

Appendices

iii

..

i

v i

1

2

3

9

23

26

31

32

No. 1

2

3

4

5

6

7

8

9

10

FIGURES

Reactor Util ized f o r Reduction Experiments With Iron Shot

Effect of pH of Feed Solution on Copper Cementation With Iron Shot

Effect of Oxygen on Copper Cementation With Iron Shot

Effect of Surface Area of Reductant on Copper Cementation With Iron Shot

Effect of Previously Deposited Copper on Copper Cementation With Iron Shot

a

Reduction of Hexavalent Chromium With Iron Shot

Simultaneous Cementation of Copper and Reduction of Hexavalent Chromium With Iron Shot

Effect of Speed of Mixing on Simultaneous Copper Cementation and Reduction of Hexavalent Chromium With Iron Shot

Continuous Sirmiltaneous Cementation of Copper and Reduction of Hexavalent Chromium With Iron Shot

Cementation of Copper With Part iculate Iron

Pane

10

12

14

15

16

17

19

20

2 1

24

TABLES

- NO Page

Metal Removal By Mg-Si Alloy Reacted By Magnetic S t i r r ing 1 19 ._I

2 Metal Removal By Mg-Si Alloy Reacted By Tumbling 20

3 Precipitated Metals Recovered After React ionpf Feed Solution With Mg-Si Alloy 20

4 Metal Removal By Ca-Si Alloy Reacted By Tumbling 21

5 Metal Removal By Mg-Si Alloy Reacted By Continuous Flow 21

-. - - .

V

ACKNOWLEDGMENTS

Dr. Lawrence C. Kenausis, Assistant Investigator, was helpful in providing information relating to the electro-chemical theory of the cementation reaction and in interpreting experimental results.

Henry A. Agonis, P.E., did a patent search relating to cementation and provided information on comnercial systems utilizing the reaction.

David E. Bingham carried out all of the experimental work and provided valuable suggestions on equipment desifgn.

Robert J. Klancko, P.E., provided helpful information on the design of experiments and on the theory of continuous chemical reactions.

vi

SECTION I

CONCLUSIONS

Small diameter iron shot is an effective medium for cementation of copper and reduction of hexavalent chromium, either separately.or in a mixture of the two.

Both reactions proceed readily between pH 1.5 and 3.0, but the chromium reduction is puch more rapid than the copper cementation.

The chromium reduction is quantitative and the coppdr cementation nearly so, given enough time.

Exclusion of air makes both reactions more efficient, especially cementation.

The efficiency of cementation is not directly proportional t o the available iron surface area.

Increased speed of mixing increases the copper cementation efficiency.

Previously deposited copper does not seriously impede further copper deposition.

Copper apparently catalyzes the reduction of hexavalent chromium and, to a lesser degree, the dissolution of excess iron.

Hexavalent chromium suppresses both copper cementation and dissolution of excess iron to a marked degree.

Particulate iron is effective for cementation of copper, but is difficult to disperse in the feed solution because of "clumping" as soon as it becomes coated with copper.

Silicon alloy reductant is reasonably effective for removing copper as well as zinc and trivalent chromium from an industrial effluent containing these ions, but is not reliable for reducing hexavalent chromium since the reaction is not primarily a reducing one.

1

SECTION I11

INTRODUCTION

GENERAL

"Cementation", i.e. the displacement of a metal from solution by a metal higher in the electromotive series, has been utilized in hydro- metallurgical processes for many years.1 "cementation" is restricted to cases where the displaced metal is precipitated from solution, e.%. the precipitation of metallic gold by replacement with zinc supplied from zinc dust. For our purposes we are considering the term to cover reduction of any ionic species by contact with a metal of higher oxidation potential whether the reduced species is precipitated or not, i.e. the reduction of ferric ion to the ferrous form by contact with metallic iron or zinc.,

Generally the application of the cementation reaction has been for the purpose of winning metals from ores or for the recovery of precious metals from scrap. treatment of brass mill effluents to accomplish reduction of hexavalent chromium and precipitation of a significant portion of the copper. Their work demonstrated the feasibility of the application, but they did not study the reactions in depth. Further, an efficient chamber for continuous processing of feed solution which would also provide a m e a n s for separating the precipitated copper in a recoverable form was not developed. of the Anaconda American Brass Co. resulted in the development of a rotating chamber for carrying out the reaction.3

Strictly speaking the term

I

In 1969, Case and Jones2 applied the reaction to

Subsequent work done at the Research and Technical Center

While the present work is largely concerned with removing toxic elements from waste streams typical of the Brass Industry, cementation offers an attractive possibility for treating any waste stream containing reducible metallic ions which must be removed from solution. Common examples are the precipitation of silver from photographic processing t~olutiona, the precipitation of copper from printed circuit etching solutions and the reduction of hexavalent chromium in waste streams from chromium plating, iriditing and chromate-inhibited cooling water blow- down. Zinc, aluminum and other metals having relatively high oxidation potentials may be utilized as reductants as well as iron. certain of the alkaline earth metals, especially calcium and barium, and alloys of magnesium have also been utilized.9, 10

Alloys of

Since the reductant material is usually scrap metal which is low in cost and the process offers the possibility of recovering valuable metals such as copper and silver, the use of cementation for treating waste streams presents significant economic advantages. compared the cost of treating the discharge from a medium-sized brass

Case and Jones2

~ . - , ~ ._.. I . . . . " . . . - . ..

3

. . . .,. .a. ., . _. . . -. . -. . . . . -.

mill by a conventional system utilieing sulfur dioxide for reduction of hexavalent chromium with the cost of treating the stream with scrap iron for precipitation of a significant portion of the copper, as well as reduction of the hexavalent chromium. equipment (15-year life), operating costs and the value of the re- covered copper, the cementation system showed a cost advantage of nearly $40,000 per year.

Including capital costs for

WORK BY OTHERS

A detailed study of the kinetics of copper precipitation has been carried out by Nadkarni et a 1 4 ~ 5 who worked with metal plates, wire and granulated cagt iron suspended in a stirred solution of copper sulfate containing approximately 1 gram per liter of copper.

Experimental data were obtained by measuring the &unt of copper and iron ions in solution at successive time intervals. conclusions reached were:

Among the

Copper cementation is a first order reaction rate process. specific rate is only slightly dependent upon initial copper concentration and pH.

The

At lower stirring speeds the rate-controlling process is solution diffusion,

At high stirring speeds the rate-controlling process is bulk solution diffusion through a limiting boundary film.

The back reaction due to ferrous ions is negligible.

The direct reaction of ferric ions does not normally affect kinetics of cementation as m c h as the reaction between ferric ions and the deposited copper.

Oxygen in solution causes &cess iron consumption due to rapid reaction with the deposited copper, the metallic iron and ferrous ions.

In a nitrogen atmosphere excess iron consumption only varied slightly with pH in the pH range of 1.5 to 3.0 as long as the copper ion concentration was over 150 ppm. excess iron consumption increased markedly and was roughly propor- tional to the hydrogen ion concentration.

At lower copper ion concentrations

Rickard and Fuerstenau6 carried out an extensive electrochemical Investigation of the cementation of copper by iron using galvanostatic electrode measurements. Solutions of copper sulfate up to approximately 0.05 M were used as electrolyte and Ferrovac E (high-purity iron), steel and cast iron as anodic materials. The authors conclude that the cementation process can be described as a galvanic corrosion cell and

4

presenE a simplified model. Other conclusions include:

1. The spec i f ic rate constant f o r cupric ion reduction is about twice as large as fo r ferric ion reduction.

2. Hydrogen ion reduction i n solutions containing oxygen is dependent on the amount of oxygen dissolved i n the solution. rate of hydrogen ion reduction i s much slower than that of copper.

Normally the

3. An increase i n ag i ta t ion w i l l produce an equivalent increase i n reduction rates of both f e r r i c and cupric ions up t o a l imit ing value .

A comprehensive study of the cementation of copper with zinc has been undertaken by Strickland and Lawsonf who present an excellent sunnnation of the cementation reaction mechanism. They usedr'a rotat ing disc tech- nique developed by Levich8, using discs of the reductant m e t a l (iron as w e l l as zinc) imnersed i n a su l f a t e solut ion containing 10 ppm of copper. Conclusions include:

1. Cementation i s a diffusion controlled reaction with the reaction rate being highly dependent on the presence of previously deposited copper.

2. When the spec i f ic mass of deposited copper is less than about 0.35 mg/cm2, the absolute cementation r a t e per uni t i n i t i a l zinc area increases with:

a. increasing reactant concentration b.

C. increasing temperature d. decreasing solution viscosi ty

increasing r e l a t ive veloci ty between reactant surface and solu- t i on

3. For spec i f ic deposit masses i n excess of the c r i t i c a l , the reaction rate is enhanced by increased roughness.

4. The deposit i s suf f ic ien t ly porous that, normally, the diffusion of product ions t o the solution-deposit interface does not become controlling.

Recently McKavney et a19s lo have described experimental work u t i l i z ing s i l i c o n al loys f o r rewval of heavy metals from water and brine. was a t first thought that t h i s reaction involved metal "exchange" and was, therefore, analagous t o cementation, however, our own work showed the primary mechanism t o be otherwise. discussed i n the body of the report.

It

This work w i l l be fur ther

Hoover and Massellill proposed the use of scrap i ron fo r the reduction of hexavalent chromium i n waste l iquors from chromium plat ing i n 1941. Since that t i m e the only published work discussing t h i s reaction w e

5

2 have found is that of Case and Jones.

PURPOSE OF PRESENT WORK

Most previous investigations of the cementation react ion have been oriented toward a hydrometallurgical application., The present work d i f f e r s i n that it is oriented toward an indus t r i a l waste, a n t i - pollution, application. fundamental respects:

This difference i s important i n a t least two

1. The concentration of reducible ions i n indus t r i a l waste streams i s much lower than i n most metallurgical process streams and the lower concentrations are more d i f f i c u l t t o reduce e f f ic ien t ly .

2. The metal ion concentrations i n the waste ul t imately discharged must be extremely low, usually 1 ppm, o r less{ thus necessitating very e f f i c i en t separations.

Because of t he or ientat ion of t h i s work w e t r i e d t o use experimental systems which might be scaled d i r ec t ly t o an indus t r i a l application.

THEORY OF CEMENTATION

Strickland and Lawson7 have expounded the theory of cementation i n considerable de ta i l . concerned with the reduction of any ionic species by contact with a - metal of higher oxidation potential . with the copper-iron and the hexavalent chromium-iron systems.

As explained previously i n t h i s work w e are

In par t icu lar w e worked pr incipal ly

Copper-iron system

The copper-iron system di f fe rs from some similar systems because of the s ide reactions involving oxygen. The reactions are as follows:

CuSOc, + Fe d F e S 0 4 + Cu

H2S04 + Fe +FeS04 + H2

(1)

(2)

4FeS04 + 2H2S04 + O2 --F 2Fe2 ( S 0 4 ) 3 + 2%0 (3)

2Cu + 2H2S04 + 02+2CuS04 + 2H20

2Fe + 2H2S04 + 02+2FeS04 + 2H20 ( 4 )

( 5 )

Fe 33FeSO4

Cu +-CuS04 + 2FeS04

6H20 z2Fe(OH)3 + 3H2S04

6

Equation (1) is the basic cementation reaction which it is desired t o complete. is thus not pH dependent (except, as w i l l be explained later, pH values above approximately 3.5 lead to precipi ta t ion of hydrous oxides and basic salts of i ron and copper). p rec ip i ta t ion of one uni t of copper, theore t ica l ly 0.8789 un i t s of i ron a r e required.

It w i l l be seen that the reaction does not consume acid and

It i s a l so apparent that f o r the

A l l of the other reactions are competing side-reactions which it i s desirable t o suppress. k ine t ics of a l l these reactions i n considerable depth,

Nadkarni and Wadsworth5 have discussed the

Equation (2) defines the dissolution of i ron with acid. It-.proceeds more slowly than (1) but cannot be completely suppressed. it i s the reaction which w e t r i ed t o minimize.

In our work

Equations (3), (4), and ( 5 ) show the influence of'bxygen on the reduction system. of the system w i l l increase the i ron consumption by providing reducible ionic species. by carrying out the reaction i n a reducing o r an i n e r t atmosphere, is of especial advantage t o carry out the cementation reaction under hydrogen, which i n any event i s generated according to equation (2), since the hydrogen i tself has some reducing act ion on the cupric ion. Back12 has described a commercial system which takes advantage of t h i s fact .

It i s obvious that oxidation of any of the components

The ef fec ts of oxygen on the system can be eliminated It

Equations (6) and (7) define the reactions between f e r r i c salts and metallic i ron and copper respectively. Obviously t h i s reaction can be prevented by eliminafing oxygen from the system (unless f e r r i c ion is already present i n the feed solution, a s is usually the case i n leach liquors).

Equation ( 8 ) i l l u s t r a t e s the prec ip i ta t ion of f e r r i c hydroxide by hydrolysis which occurs between pH 3 t o 4. the ferric hydroxide coats the surface of the metal l ic i ron and also may induce coprecipitation of basic copper sulfate .

This is undesirable since

Hexavalent chromium - i ron system

As previously mentioned, not much work on t h i s system has appeared i n the l i t e r a tu re . The reactions are:

Na2Cr207 + 7H2S04 + 2Fe +Na 2 4 SO + Cr2(S04)3 + Fe2(S04)3 + 7 H 2 0 ( 9 )

Na2Cr207 + 7H SO + 6FeSO + N a SO + 2 4 4 2 4

Equation ( 9 ) is the basic reaction. It requires ac id and is thus pH dependent, the f e r r i c state and tha t fo r each un i t 1.0741 un i t s of i ron are required.

C r (SO ) + 3Fe (SO ) + 7H20 (10) 2 4 3 2 4 3

w i l l be noted that the reaction tha t the i ron is oxidized t o of hexavalent chromium reduced,

7

%. . . . . - . -. i -' ,

In the presence of ferrous ion (such as would r e su l t from cementation of copper i n the same solution) the react ion defined by (10) takes place.

Under the proper conditions of pH,diffusion, and r a t i o of i ron surface t o reducible ion the reaction is quant i ta t ive and extremely rapid.

The following reaction i s a l so possible i n the presence of cemented copper:

N a C r 0 + 7H2S04 + 3Cu Na2S04 + C r 2 (so4)3 + 3cUSo4 + 7H20 2 2 7

However, reactions (9) and (10) are so rapid it is doubtful t ha t a the re is ever su f f i c i en t res idual hexavalent chromium present f o r reaction (11) t o proceed.

In a solut ion of pH 2.0 o r less containing both coppen’and he.xavalent chromium i n the absence of a i r the probable sequence of reactions is: (91, (11, (2), (61, (7).

8

SECTION IV-

USE OF IRON SHOT AS REDUCTANT

_* Small, diameter shot appeared t o us t o have several advantages from the standpoint of geometry, including:

1.

2.

Relatively large surface area per un i t volume.

Point-to-point contact between pieces ra ther than surface-to-surface contact . - >

3. Good mixing action.

4. Easily controlled surface area. i

The shot u t i l i zed was obtained from Crossman Arme Co. Xnc., Freeport, N.Y.. Analysis showed it t o have a carbon content of approximately 0.052 and only t races of metall ic contaminants. diameter of 4.37 me an average weight of 0.34 grams and a packing density of 642 shot per 50 ml. of approximately 385 cm2. of lubricant which was removed by a hot water and soap solution followed by an acetone rinse. mixture of 25% NH4OH (sp. gr. 0.90) plus 5% H202 (30%) immediately before use and between runs.

The shot had an average

Thus 50 m l of shot had a surface area The shot as received was covered with a fi lm

Normally the shot was cleaned by pickling i n a

When a volume of l e s s chan 50 cc of shot was used i n order t o reduce che i ron surface, the difference i n volume was made up by adding chem- ica l ly- res i s tan t glass beads of approximately the same diameter. It wa8 thought that th i e would keep the shear rate between the sol ids and the solut ion r e l a t ive ly constant.

REACTOR

The reactor was a 500m1 Erlenmeyer f l a sk mounted about 30" from the horizontal on a c i r cu la r p l a t e capable of revolving a t speeds up to 60 RR4. nearly t o the bottom of the f lask a t the r a t e of 2 l i ters per min.

In most experiments nitrogen was fed through a tube reaching

When the reactor was operated on a continuous basis the feed solut ion was pumped in to rhe f l a sk through a tube reaching nearly t o the bottom and the overflow allowed t o drip of f a c o l l a r around the mouth of the f l a sk in to a col lect ion beaker. Fig. 1, shows the arrangement.

9

FEED SOLUTIONS

Feed solut ions were normally made up t o contain 100 ppm of the reducible ion species (copper and hexavalent chromium) which is the highest level l i k e l y t o be found i n brass m i l l discharges. Reagent grade CuS04.5H20 was used as a source of cupric ion and Na2Cr2q.2H20 of similar pur i ty was used as a source of hexavalent chromium ion. pH of the solutions was adjusted with H2SO4 o r NaOH.

!

The

EXPERIMENTAL PROCEDURES

Experiments were carr ied out by both batch and continuous reactions. The procedures w i l l be formed i n Appendix A, page 33.

EXPERMENTAL RESULTS (BATCH REACTIONS)

Copper only i n feed solut ion i

Effect of PH- The pH of feed solutions was adjusted t o 1.50, 1.90, and 3,OO and the solutions reacted with 50 cc of shot i n air. Curves i l l u s t r a t i n g the r e su l t s a r e plotted on Fig. 2. . \

It w i l l be noted that i n a l l instances 99% of the copper was removed i n 10 min . appeared t o be s l i g h t l y higher a t pH 3.0. However, since the i n i t i a l pH of the solut ion quickly rose t o nearly 5 , f e r r i c i ron precipitated as f e r r i c hydroxide and probably carr ied some copper i n the form of basic s u l f a t e with it. The f a c t tha t the prec ip i ta tes from runs a t pH 1.5 and 1.9 car r ied a negligible amount of copper, whereas preci- t a t e s from runs a t pH 3.00 averaged nearly 1 mg of copper lends credence t o t h i s probability. Hence, there i s probably no r ea l d i f fe r - ence i n the rates of the reaction.

The r a t e s of removal were qui te similar, although the r a t e

The f a c t that a higher level of excess i ron dissolved i n a solution having an i n i t i a l pH of 1.90 than i n one having an i n i t i a l pH of 1.5 is puzzling. ferrous ion a t higher pH. Data fo r excess i ron dissolved i n a solution having an i n i t i a l pll of 3.00 is not available. a s igni f icant p a r t of the dissolved i ron precipi ta ted as f e r r i c hydrox- ide,

Only a t the median initial pH was the rise in pH gradual. higher and lower ac id concentrations the pH leveled out rapidly. A t the higher pH t h i s r e su l t s from the equilibrium by the precipi ta t ion of f e r r i c hydroxide. A t the lower pH it resu l t s from the fac t t ha t r e l a t ive ly large changes i n hydrogen ion concentration resu l t i n only small pH changes i n t h i s area. It i s in te res t ing t o note that a t the median in i t ia l pH the curves f o r pH rise and excess i ron dissolution roughly p a r a l l e l each other.

A possible explanation may be the increased oxidation of

As previously mentioned,

A t both

Effect of oxygen- To assess the e f f ec t of eliminating oxygen from the

11

.)

0 B B 50- 3 u

25

35

8

f 6 2 n c(

0) Lr

v) m ba

ii

= B

6 z v)

100 m l SOLUTION

50 cc SHOT TUMBLED 50 RPM ATMOSPHERE - A I R

100 ppm cu

. O pH 1.50 J A pH 1.90

0 pH 3.00 -

0-

- 30-

25-

20-

15-

10-

5-

0-

4-

3-

2 -

1-

I I I I I I

12

system, nitrogen was bubbled through solutions a t 2 liters/min. during a series of runs. runs made without any copper i n the feed solution. Curves i l l u s t r a t i n g the resu l t s are plot ted on Fig. 3.

Also included i n t h i s series of experiments were

The rate of copper removal w a s b e t t e r i n the nitrogen atmosphere. may be due t o the fact t h a t re-solution of copper by oxidation was prevented. *

This

Interest ingly the level. of excess dissolved i ron remained f a i r l y constant i n the nitrogen atmosphere. Nadkarni and Wadsworth5 on the increased i ron consumption resu l t ing from oxidation, a t a laser level when there i s no copper present i n solutfon; qu i te possibly the metallic copper deposited on the surface of the i ron ca ta 1 yzes the d ie solut ion . In the absence of copper the pH levels out a t a s l i g h t l y higher value. This i s at variance with the fact t h a t s l i g h t l y less i ron i s dissolved under the same conditions.

This seems t o confirm the observations of

It is however, surprising t o note t h a t i ron dissolution is

Effect of surface area- Runs were made using 50 cc, 25'cc, and 12.5 cc of shot. up t o 50 cc with glass beads. Data i l l u s t r a t i n g the resu l t s a r e shown on Fig. 4.

Where l e s s ' t h a n 50 cc of shot was used the volume w a s made

It i s evident t h a t the eff ic iency of t he reaction i s not d i r e c t l y pro- portional to the surface area of the iron. For instance, by reducing the surface of the i ron from 385 cm2 t o 193 cm2, the amount of copper removed from the so lu t ion i n 10min. is reduced only 8%, although the excess i ron dissolved i s halved and the pH rise i s reduced by more than 90%.

By reducing the surface of the i ron from 385 cm2 t o 96 cm2, the amount of copper removed from the solut ion i n 10 minutes is reduced only 36%, whereas the excess i ron dissolved i s only about one f i f t h and the pH rise i s reduced by about 95%.

It seems obvious that i n prac t ice i t i s advantageous t o use the minimum i ron surface which w i l l accomplish sa t i s fac tory copper removal i n a reasonable t i m e . copper level a minimum of about 200 an2 of i ron surface i s required t o raaove 90% of the copper from 100 m l of solut ion during a contact period of 10 min. (equivalent t o approximately 0.045 mg Cu/cm2Fe). During t h i s period about 13 mg of excess i ron w i l l be dissolved (equiv- a l e n t t o approximately 1.4 mg excess Fe/mg Cu cemented),

From the data shown on Fig. 4, a t the 100 ppm

Effect of previously deposited copper- To determine the e f f ec t of previously deposited copper on the eff ic iency of the i ron reductant six consecutive 10-min. runs were made using the same charge of shot.

13

100 m l SOLUTION ,100 ppm cu pH 2.0 50 cc SHO?! TUMBLED 50 RPM 0 RUN I N A I R

;A RUN UNDER N2 0 NO Cu(N2 ARIOS.)

I I I I I I 4 6 8 10 12 1 2 REACTION TIME, min.

Figure 3. Effect of Oxygen on Copper Cementation With Iron Shot

14

0

35

30

25

20

15

10

5

0

4

3

2

1

100 ml SOLUTION 100 ppm cu pH 2.0 TUMBLED 60 RPM

,‘A’IMOSPHERE N2

0 50 cc SHOT A 25 cc SHOT 0 12.5 cc SHOT

1

I I I I I I 2 4 6 8 10 12

REACTION TIME, min. Figure 4. Effect of Surface Area of Reductant on Copper Cementation

With Iron Shot

15

-. .

H a

9z H

i 9

25

2;

I4 g

100 m l SOLUTION

pH 1.90 50 cc SHOT TUMBLED 50 RPM ATMOSPHERE - A I R CONTACT TPiE - 10 min. NO CLEANING BETWEEN RUNS

100 ppm cu

75-

50-

-

0-

4-

3 -

2-

1-

I f I I I I I 1 2 3 4 5 6

RUNS

Figure 5. Effect of Previously Deposited Copper on Copper Cementation With Iron Shot

Results are p lo t ted on Fig. 5. subs tan t ia l amounts of f e r r i c ion were precipi ta ted a s hydroxide. data show that betcer than 99% of the copper was removed during each run and that there was no deter iorat ion of eff ic iency i n succeeding runs. Evidently any film which fonns on the eurface of the i ron i s porous enough 80 t h a t the feed solut ion readi ly penetrates t o the metal.

Since the runs were made i n a i r The

This confirms the observation of Strickland and Lawson7.

Hexavalent chromium only i n feed solut ion

In experimenting with hexavalent chromium solutions it quickly became apparent t h a t the reduction proceeded rapidly t o completion. Typical data a r e plot ted on Fig. 6.

When these data a r e compared with those f o r copper under the same

16

75- #

I4

50-

a d

25

35

6 30

6 2 5 3 z rn 20 n 0) 15,-

10

bo

H

Er

w

8

X a 8 2 g

100 ppm c+6 pH 2.0 _,

12.5 cc SHOT TUMBLED 30 R R l ARiOSPHERE - N2

?d -

0 -

- - - -

-

5 -

0 . -

4.-

3.-

2.-

1 . -

0 REACTION TIME, min.

I 2 4 6 a 10 12

J I I 1 I I

Figure 6. Reduction of Hexavalent Chromium With Iron Shot

17

conditions which are plot ted on Fig. 4 it is apparent that:

1. The chromium reduction is complete i n 8 min. whereas the copper removal i s only about 60% complete i n 8 mfn. and never reaches much above 70% completion.

2. The chromium reduction resu l t s i n dissolution of s l i g h t l y more excess i ron than the copper cementation, probably due t o presence of f e r r i c ion.

3. The pH rise during the chromium reduction i s a l i t t l e greater than tha t which occurs during copper cementation. This is t o be expected since the reaction consumes acid.

Copper and hexavalent chromium i n feed solution

Experiments were run t o study the e f f ec t of equal&mounts of copper and hexavalent chromium upon the reduction of each other. Data a r e plot ted on Fig. 7.

Effect of hexavalent chromium on copper cementation- Examination of the data shows:

1. The presence of hexavalent chromium suppresses the copper cementa- t ion. about 16% less.

For a contact t i m e of 8 min. the cementation of copper i s

2. The presence of hexavalent chromium a l so suppresses the dissolution of excess iron.

3. The presence of hexavalent chromium causes an increase i n pH rise.

Effect of copper on the reduction of hexavalent chromium- The following e f f ec t s are apparent:

1. The presence of copper increases the rate of reduction of hexavalent chromium sl ight ly . ass el 1 i10.

This e f fec t was a l so noted by Hoover and

2. The copper suppresses the dissolut ion of excess iron. This i s contrary t o the e f fec t noted during a continuous run where, as w i l l be seen l a t e r , the dissolution of excess i ron w a s increased i n the presence of copper.

3. The copper has l i t t l e e f fec t on the pH rise during the reaction.

Effect of mixing speed- Since reduction by a so l id material is a diffu- s ion controlled reaction, the speed of mixing necessarily influences the r a t e of reduction. Data obtained by increasing the tumbling rate from 30 t o 60 RPM are plot ted on Fig. 8. increase i n the copper cementation r a t e occurred.

It w i l l be noted that a s l i gh t Reduction of

f I

18

\

2L 0

pH 2.00 12.5 cc SHOT

r' TUMBLED 30 RPM ATMOSPHERE - N2 0 99 ppm cu

;A 96 ppm Cr+6

J I 1 1 I I I 0 2 4 6 8 10 12

REACTION TIME, min.

Figure 7. Simultaneous Cementation of Copper and Reduction of Hexavalent Chromium With Iron Shot

19

.) z I3 u OH

8

2

; 6 VI

75-

50-

25-

0-

4-

3-

2-

1-

A //

I I I I I I I

100 ml SOLUTION

25 ppm Cr6 pH 2.00 25 cc SHOT ATMOSPHERE -. N2 0 TUMBLED 30 RPM ATUMBLED 60 RPM

100 ppm cu

€

A A L A a M

IDENTICAL BCXH RUNS

Figure 8. Effect of Speed of Mixing on Simultaneous Copper Cementation and Reduction of Hexavalent Chromium With Iron Shot

hexavalent chromium was almost instantaneous in both instgnces. There was no effect on pH rise. been more pronounced if the change in rotatidnal speed had been greater or the ratio of iron shot volume to liquid volume had been greater.

It is probable that the results would have

EXPERIMENTAL RESULTS (CONTINUOUS REACTIONS)

Data are plotted on Fig. 9. The following conclusions may be drawn:

Copper only in feed solution

1. The rate of copper cementation stabilized quite rapidly and did The amount not change greatly during the course of the experiment.

of copper cemented averaged a little over 79%.

The amount of excess iron dissolved increased very gradually and finally stabilized at about 320 ppm.

2.

i /

. . . . . . - . .- ,

20

10

0

4 i 3

2k 1

100% C& Reducti

Ei cu 50 cc SHUI

TUMBLED 30 RPM ATMOSPHERE - N FLU4 - 25 ml/mfn pH 2.05

0 95 ppm cu a 99 ppm Cr+6 0 101 ppm Cu + 115 p p CrM

.,

I I I 1 I I I I 1 1 I I I I i I 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 405 5.0 5.5 6.0 6.5 7.0 7.5

REACTION TIME, hours

Figure 9. Continuous Cementation of Copper and Reduction of Hexavalent Chromium With Iron Shot

3. The pH also increased very gradually and stabilized at about 2.75. It is interesting to note that the curves for excess iron dissolution and the pH rise closely parallel each other.

Hexavalent chromium only in feed solution

1. Hexavalent chromium was quantitatively reduced almost imnediately and continued to be so during the course of the experiment.

2. The amount of excess iron dissolved stabilized fairly quickly at about 129 ppm; less than 20% of the amount dissolved by the copper cementation under the same conditions. It is interesting to note that even though the system contained a substantial excess o f iron surface (in terms of the requirements for quantitative reduction of the hexavalent chromium) the amount of excess iron dissolved was very low. tions the presence of chromium inhibits dissolution of iron by acid.

This may indicate that uder stabilized dynamic condi-

3. The pH remained constant at about 2.55, slightly lower than for the copper cementation, probably because less acid was expended in the dissolution of iron.

Copper and hexavalent chromium in feed solution

1.

2.

3.

Again the reduction of hexavalent chromium was quantitative during the course of the experiment. However, the Cementation of copper was slightly depressed, averaging only 71.29%.

The dissolution of excess iron was considerably less than when copper alone was present in the solution, but slightly greater than when chromium alone was present. curves for excess iron dissolution by chromium alone and by chromium plus copper parallel each other closely, indicating the dominating effect of the chromium.

It is significant that the

The pH remained constant at a very slightly lower level than where chromium alone was present in the feed solution.

Not apparent from the data is the observed fact that during the continuous runs fairly large particles of copper were continually dio- lodged from the surface of the shot. out of the reactor by the flow of solution, most collected in the bottom of the reactor. a lesser angle of inclination of the reactor most of the precipitated copper would eventually be swept from the reactor and could be settled in a collection chamber. During batch runs where only 10 mg of copper was present, 90% or more of the copper adhered to the surface of the shot.

While some of these were swept

It seems likely that at higher flow rates coupled with

22

SECTION V

USE OF PARTICULATE IRON AS REDUCTANT

Back12 has described the advantages of pa r t i cu la t e i ron as a medium f o r the comnercial cementation of copper. b r i e f l y i n the present work because the use of i ron shot seemed much more su i tab le f o r a small-scale laboratory system.

REDUCTANT

This material was explored only

Two di f fe ren t s izes of i ron pa r t i c l e s were used: approximately -400 mesh (Leco Low-Sulphur Equipment Corp., St. Joseph, Mich.) and -325 mesh (ACF Powder, Hoeganaes Corp., Riverton, N.J.) .

Iron Powder Accelerator' No. 501-78, Laboratory

FEED SOLUTION

The feed solut ion contained 100 ppm of Cu as CuSO4 and had a pH of 2.0.

EXPERIMENTAL PROCEDURES

The system was tes ted only with copper and only i n batch reactions. The procedures w i l l be found i n Appendix B, page 34.

EXPERIMENTAL RESULTS

It was observed tha t i n i t i a l l y i ron powder of e i t h e r type was uniformly dispersed throughout t he feed solut ion, however, as copper was cemented on the surface the individual pa r t i c l e s began t o clump together u n t i l f i n a l l y most of the metal remained on the bottom of the reaction vessel. Data f o r both runs are plot ted on Fig. 10. The data f o r the two runs are not d i rec t ly comparable with each other o r with any of the data obtained with i ron shot.

Several f a c t s may be observed from the data:

1. The removal of copper i n the mechanically s t i r r e d system gradually deter iorated with t i m e . This is no doubt due t o t h e "clumping" of t h e iron part ic les .

2. The removal of copper i n the nitrogen-dispersed system gradually increased with t i m e , probably because the "clumping" e f f ec t was not as serious when a fresh charge of i ron was used f o r each run.

3. The amount of excess i ron dissolved i n the mechanically s t i r r e d system gradually decreased a f t e r an i n i t i a l sharp drop, whereas the

23

3 u

25

0-

35,-

2 30- f 25- I4 % 20- 3 CI

k m w

Q) 15-

10-

Ii 5-

0 -

ID0 ml SOLUTION NO ppm cu pH 2.0

0 1 g LECO POWDER MECHANICALLY ST fRRED

A 0.05 g ACF POWDER DISPERSED WITH N2

A A A A PIP u

4-

3 -

2-

1.-

I I I I I 12 I 6 8 10 0 2 4 REACTION TIME, min.

Figure 10. Cementation of Copper With Particulate Iron

24

- . . . ..

amount very gradually increased in the nitrogen-dispersed system; probably for the same reason as the variation in the behavior of the copper cementation.

4. There was little rise in pH with increase in contact time in either system.

We believe the tendency of the cemented copper to adhere to the iron particles and the resulting "clumping" of the particlee greatly limits the usefulness of thls medium. Possibly these effects can be overcome by using a high flow rate to keep particles in suspension.

25

SECTION VI

USE OF SILICON ALLOYS AS REDUCTANT

BACKGROUND

As previously noted, McKavney, Fassinger and Stivers' have described a system which utilizes silicon alloys (especially alloys with calcium and magnesium) for the removal of heavy metals, including copper, zinc and chromium, in concentrations of 5 to 100 ppm from waste water. carried out batch reactions by shaking a solution containing heavy metal ions with the granulated silicon alloy in a flask and semi-continuous reactions by passing the solution containing the heavy metals through a 150 nun x 30 rum I . D . column packed with the alloys:'

Removal of copper, zinc, trivalent chromium and other heavy metals was impressive, however the reduction (and removal) of hexavalent chromium was not satisfactory. niques claimed for the process is the fact that the only ion going into solution is calcium or magnesium,neither of which is generally considered objectionable in a waste stream. The authors explained the mechanism of reaction as being primarily electrochemical with possible hydroxide formation through hydrolysis.

They

The advantage over conventional removal tech-

Because of the presumed analogy to the "cementation" reaction we investigated the sytem both on a batch and a continuous (plug-flow reacto r) bas is . REACTANTS

The silicon alloys used were:

1. Magnesium Ferrosilicon (Si 44.50%, Mg 8.90%, Ca 1.00%, Fe 43.00%) supplied by Globe Metallurgical Div. of Interlake, Inc. , Beverly, Ohio which was sieved to 10/20 mesh before use.

2. Calcium Silicon (Ca 32.4%, Si 63.1%, Fe 4.0%, Ba 0.49.) supplied by Union Carbide, Ferro Alloys Div., Marietta, Ohio, which was sieved to 28/65 mesh before use.

FEED SOLUTION

The feed solution used was typical of a brass mill effluent. composition was:

The

Copper Zinc Chromium (+3) Chromium (+6) PH

26

45 Ppm 43 Ppm 43 PPm 7.5 Ppm 2.05 ppm

A l l catlons were present as sulfates.

EXPERIMENTAL PROCEDURES

Contact between feed solution and reaatantr was achieved by:

1. Magnetic St i r r ing "Si a l loy only).

2. Tumbling.

3.

The procedures are described in Appendix C page 35.

Coatinuous flow t h r a s h a ver t ical colurnn.

160 190

202 204 202

,200

EXPERIMENTAL RESULTS

Magnetic S t i r r i n g

Results ate shown i n Table 1.

Table 1.

Copper

<e5 980% <a5 98.9+ <.5 98.W 0.6 98.7 0.5 98.9 0.9 98

LEMOVAL BY M Ppm

Zinc Effo %Re50

< e 1 98.W C.1 98.8+ 0.2 99.5 0.6 98.6 0.5 98.8 1.1 97.4

41 ALLOY'REACTED BY MAGNETIC STIRRING Ppm

ChromiumT Effo %R em.

4 . 0 98+ Cl.0 98+ < L O 98+ C l e o 9- a . 0 98+

102 97.6

PPh Chromium*

Effr %R aa,

0 100 0 100 <.1 98,7+ 0 100 0 100 0 100

+ f f . E f f

LO 10

9.8 9.8 9.85 9.75

Some readily apparent concluliioas are:

1. In 6 successive 5 d n . runs, removal of a l l heavy metals exceeded 972, although there was a vary s l igh t deterioration after the 5th nm.

2. The pH of the solution rapidly s tabi l ized a t 10 which approachea that of a saturated magnesium hydroxide solution,

3. After the f i r s t run the amount of magnesium which entered the solu- t ion remained f a i r l y constant a t about 200 ppm which i s much higher than would be expected fo r a saturated magnerium hydroxide solution, indicating that some of tho magnesium w a s present as sulfate.

A run made with 100 m l of d i r t i l l e d water under the same condltions as the mns with feed solution containing metal ions and acid resulted i n a pH of 10.5 and a magnesium concentration of 10 ppm.

27

Tumblim

PPm Ppm PPm Copper Zinc ChromiumT

Run Eff. 7aem. Eff. %Rem. Eff. %Rem.

Mg-Si alloy- Results are shown i n Table 2.

P P Ppm Chromium* Mg

Eff. %Rem. E f f .

I

c.5 99 c.5 99 0.8 98 <.5 99 c.5 99 3.3 93

I I I I

<.l 100 <.l 100 0.2 100 c.1 100 c.1 100 3.0 93

mg mg Copper Zinc

Run Rem. Rec. %Rec. Rem. Rec. %Rec.

4 4.5 4.0 89 4 . 3 2.8 65 5 4.5 3.2 71 4.3 2.4 56

6.8 87

8.8 83 6.8 87 5.7 89

12.6 75

8.8 83

mg Chromium

Rem. Rec. %Ret.

4.4 3.3 75 4.5 2.7 60

5.0 33 5.0 33 5.5 27 5.0 33 4.0 47 5.0 33

80 60 57 49 40 54

NG - PH

Eff.

10.25 10.25 10.2 10.1 10.0 10.0

It

1.

2.

3.

4.

5.

is evident t ha t :

Removal of copper and zinc is sa t i s f ac to ry f o r the f i r s t 5 runs, but eff ic iency drops off a f t e r the 5th run.

Removal of t r iva len t chromium is f a i r l y sa t i s fac tory , though less complete than fo r copper and zinc.

Removal of hexavalent chromium i s unsatisfactory.

The amount of magnesium taken in to solut ion is only about one- t h i r d the amount dissolved by the more vigorously s t i r r e d solutions.

There is a marked deter iorat ion i n eff ic iency of removal of a l l metals a f t e r the 5th run. the pH remained constant a t about 10.

I

The reason f o r t h i s is not c l ea r since

The prec ip i ta tes from runs 4 and 5 were analyzed fo r copper, zinc, and chromium. Results a r e tabulated i n Table 3.

We believe the resu l t s t o indicate that a s ign i f icant amount of the precipi ta ted metals remains on the surface of the reactant metal.

28

.- .~. . . . . . . . . .

Ca-Si Alloy- ResulKs .. are - shown i n Table 4.

Ppm Ppm Ppm Copper Zinc ChromiumT

Bun E f f . "* Ef f. %Rem. E f f . %R em.

Ppm Ppm Chromium+6 Ca Eff. %Rem. Eff.

<. 5 99 <. 5 99 c. 5 99 1.1 98 1.1 98 7.2 84

Fe Mg si in ef- i n ef- i n ef- fluent fluent fluent * * * <. 5 58 5 c. 5 61 6 <. 5 62 6 <. 5 60 6 <. 5 63 7 <. 5 65 7

0.2 9% <.l 100 0.5 99 0.8 98 8.7 80

39 9

PH

9.70 10 . 10 10.05 10.0 10.0 9.95

2.0 96 1 1.8 96

1.3 97 3.4 93 2.4 95

12.6 75

C r %Rem.

99 91 89 86 78 77

1.2 84 1.2 84 1.0 87 1.4 81 1.0 87 1.2 84

Fe %R=*

86 78 74 7 1 65 63

100 96 84

100 74 48

YG PH 4 Eff

9.95 9.35 7.95 7.95 6.35 5.05

Significant differences from the behavior of the magnesium-silicon a 1 loy include :

1. Removal of zinc decreases rapidly a f t e r the 4th run, probably because of the drop in pH.

2. Reduction of hexavalenc chromium remains f a i r ly constant a t 85% through the f irst 5 runs. with the magnesium alloy, probably because of the higher oxidation potential of calcium.

This is significantly be t te r than achieved

3. The pH drops progressively as runs a re continued.

4. The amount of calcium dissolved drops progressively as runs a re continued. This may be because a l l of the calcium has been leached from the surface of the al loy o r because the surface has been coated with a film of reaction products which inhibi t further dissolution.

Continuous (Plug-Flow) Reaction

Resul t s are shown i n Table 5.

Tab' Elapsed

Time, Min

0-15 15-30 30-45 45-60 60-7 5 7 5- 90

*Pm

I 5. B

Zn %Rem 4

98 92 91 88 82 80

-

I

29

30

..~..~..... I -. . . . . . . ...-_ . .

It w i l l be noted t h a t a l l of the metals were removed less ef fec t ive ly with the passage of time.

During the course of the run it was noted that the eff luent exhibited increasing turbidity. When the column was back flushed a f t e r the run a large quantity of finely-divided, black, amorphous material was re- moved. but only small amounts of the metals present i n the feed solution.

This was found t o contain a high concentration of magnesium,

GENERAL CONCLUSIONS

Our study of the s i l i con a l loy system h d t o the following conclusions:

1. The primary reaction appears t o be formation of hydroxide by reac- t i on of the alkali component of the a l loy with water, thus:

r

Mg + HzO-MgO $. H2

In turn the magnesium hydroxide reacts with metal ions Uo form hydrous oxides or basic salts,

2. A reduction reaction a l so appears t o take place, thus:

3Mg + 2Cr03 + 6H2S04+-3+4gS04 + Cr2(S04)3 + 6H20

This reaction i s probably inhibited by the rapid neutral izat ion of the system.

3. There 1s no evidence that any consti tuent of the a l loy other than the calcium o r magnesium enters in to the reaction.

4. The effectiveness of the alloys deter iorates before the bulk of the materia1 is s igni f icant ly diminished, We believe t h i s t o be a surface phenomenon, but have not determined whether it i s due primarily t o depletion of the ac t ive m e t a l on the surface of the a l l o y granules o r t o formation of an inhibi t ing layer of reaction products. Vigorous s t i r r i n g of the mixture of a l l o y granules and feed solution helps to prolong the effectiveness of the m e t a l removal.

I n s p i t e of t he drawbacks of the system w e believe alkali earth metal a l loys and alloys of other metals with high oxidation poten- t i a l o f f e r an interest ing area fo r fur ther study.

1.

2.

3.

4.

5.

60

7.

8.

9.

10 . 11.

12.

SECTION VI1

REFERENCES

Habashi, F., Principles of Extractive Metallurgy, Vol. 2 Hydrometallurgy, p 235, Gordon and Breach, New York, N.Y. 1970.

Case, 0. P. , and Jones, R. B. L., "Treatment of Brass Mill Effluents", Conn. Research Commission RSA-68-34, Sept. lo, 1969.

Case, 0. P., et al, "Method For Simultaneous Reduction of Hexavalent Chromium and Cementation of Copper", U.S. patent 3,748,124, Jd. 24, 1973.

Nadkarni, R. M.; Jelden, C. E.; Bowles, K. C.; Flanders, H. E.; and Wadsworth, M. E.; "A Kinetic Study of Copper Precipitation on Iron-Part I", Trans. Met. SOC. A M , Vol. 239, Apr. 1967, p. 581.

;

Nadkarni, R. M., and Wadsworth, M. E., "A Kinetic Study of Copper Precipitation on Iron-Part II", Trans. Met. SOC. AIUE, Vol. 239, Jul. 1967, p. 1066.

Rickard, R. S., and Fuerstenau, M. C., "An Electrochemical Investigation of Copper Cementation by Iron", Trans, Met. SOC.

Vol. 242, Aug. 1968, p. 1487.

Strickland, P.H., and Lawson, F., '%ementation of Copper With Zinc From Dilute Aqueous Solutions", Proc, Aust. Inst. Min. Met. No. 236, Dec. 1970, p. 25.

Levich, V. G., Physiochemical Hydrodynamics, Prentice Hall, Englewood Cliffs, N.J. 1962.

McKaveney, J. P.; Fassinger, W. P., and Stivers, D. A. "Removal of Heavy Metals from Water and Brine U s i n g Silicon Alloys", Environ. Sci. and Tech., Vob. 6, No. 13, Dec. 1972, p. 1109.

McKaveney, J. P., "Process For the Removal of Ionic Metallic Impurities From Water", U.S. patent 3,766,036, Oct. 16, 1973.

Hoover, C. R., and Masselli. J. W., "Disposal of Waste Liquors from Chromium-Plating", Ind: Eng. Chem., Vol. 33, No. 1, Jan. 1941, p. 131.

Back, A. E., "Precipitation of Copper from Dilute Solutions U s i n g Particulate Iron," Jour. Met., May 1967, p. 27.

31

SECTION VI11

APPENDICES

Page

33 A. Experimental Procedure Ut i l ized With Iron Shot as Reductant

Experimental Procedures Utilized With Particulate Iron as Reductant

34 B.

C. Eyperimental Procedures Utilized With Alkalin; Earth- 35 Silicon Alloys as Reactant

I

APPENDIX A

EXPERIMENTAL PROCEDURE UTILIZED WITH IRON SHOT AS REDUCTANT

BATCH REACTION

1.

2.

3.

4.

5.

6 .

7.

8.

Place 50 cc of freshly-cleaned shot, o r combination of shot and glass beads, i n reactor shown i n Fig. 1.

Start flow of nitrogen a t 2 l i ters/min. ( i n i t i a l experiments were carr ied out i n a i r ) .

Add 100 m l of feed solution t o reactor.

Rotate reactor a t 30 t o 60 RPM for 2 min.

Pour solution through a No. 5 f i l t e r paper.

T e s t the f i l t r a t e qua l i ta t ive ly fo r f e r r i c iron. Determine pH and concentrations of i ron and residual reducible ions (copper and hexavalent chromium).

Analyze the prec ip i ta te for copper and i ron ( in some cases).

Repeat using contact t i m e s of 4, 6, 8 , 10, and 12 min..

I

CONTINUOUS (BACK-MIX) REACTION

1.

2.

3.

4.

5.

6.

7.

Place 50 cc of freshly-cleaned shot, o r combination of shot and glass beads i n reactor shown i n Fig. 1,

S t a r t flow of nitrogen a t 2 l i ters/min,.

F i l l the reactor t o over flowing with feed solution (480 m l ) .

Rotate reactor a t 30 RPM.

Pump feed solut ion through reactor a t the r a t e of 25 mllmin. f o r 8 hours.

Following an i n i t i a l period of 30 min., sample eff luent over a 15-min. period a t 30-min. intervals .

Analyze eff luent for i ron, residual copper and hexavalent chromium and determine pH.

33

APPENDIX B

EXPERIMENTAt PROCEDURES UTILIZED WITH PARTICULATE

IRON AS REDUCTANT

LECO PWDER

1.

2,

3.

4.

5.

Place 1 g of powder i n a 250-ml beaker.

Add 100 m l of feed solution.

S t i r with a mechanical stirrer a t 500/700 RW(rough estimate) f o r 2 min..

Decant solution and-analyze f o r i ron and copper and detennine pH.

Ridse powder with d i s t i l l e d wattr and repeat steps 2 thru 4 using contact times of 4, 6, 8 , 10, and 12 min..

HOEGANAES ACF P O E R

1.

2.

3.

4.

5.

Add 100 m l of feed solut ion t o a 250-ml separatory funnel connected t o a source of nitrogen through the stem.

Add 0.05 g of powder ( in the form of a well-mixed 1% suspension) and disperse t h e par t ic les throughout the feed solut ion with a stream of nitrogen (0.8 literr/min.)

After a 2-min. contact t ima , pour the solution through a f i l t e r paper t o remove i ron par t ic lea .

Analyze the f i l t r a t e for capper and i ron and determine pH.

Repeat s teps 1 through 4, using contact times of 4, 6 , 8, 10, and 12 min..

--. . ” ., . - . .. .I_.. . . . ”. * ..-

APPENDIX C

EXPERIMENTAL PROCEDURES UTILIZED WITH SILICON

ALLOYS AS REACTANT

BATCH REACT IONS

pixim By S t i r r i n g (hg -Si Alloy)

1. Add 50 cc (78 g) of granules t o a 300-ml Erlenmeyer flask.

2. Add 100 m l of feed solut ion t o flask. I

3.

' *

S t i r f o r 5 min. with a 4.45 cm magnetic stirrer a t approximately 200 RPM (rough estimate).

4. F i l t e r so lu t ion and analyze f i l t r a t e f o r copper, zinc, chromium (T), chromium (+3) and magnesium and determine pH.

5. Using the same metal charge, add another 100 m l of feed solution and repeat s teps 3 and 4.

6. Continue runs as i n s tep 5 u n t i l 6 runs have been made.

Mixing By Tumbling (Mg -Si Alloy and Ca-Si Alloy)

1.

2.

3.

4.

5.

6.

Add 50 cc of granules t o reac tor shown i n Fig. 1.

Add 100 m l of feed solut ion t o reactor.

Rotate reactor a t 45 RPM fo r 5 min.

F i l t e r so lu t ion and analyze f i l t r a t e f o r copper, zinc, chromium (T), chromium (+3), and magnesium and determine pH.

U s i n g the same m e t a l charge, add another 100 m l of feed solut ion and repeat s teps 3 and 4.

Continue runs as i n s tep 5 u n t i l 6 runs have been made.

CONTINUOUS (PLUG-FLOW) REACT ION

1. Place 50 cc (81.5 g) of 10/20 mesh Mg-Si granules i n a ve r t i ca l g lass column 12.7 mm I.D. x 419 c m long.

2. Pump feed solut ion up through the packed column a t a flow rate of 10 ml/min. fo r 90 min..

35

3.

4.

Collect effluent, replacing receiver every 15 min..

Analyee each portion of effluent for copper, zinc, chromium (T) and detemine pH.

. - . . . . ... . . , .. . . ._ ..

36

- - ..*IT.,- - --.-,--....

-.._-

SELECTED W A T E R RESOURCES ABSTRACTS INPUT TRANSACTION FORM 4. Title

METALLIC RECWERY FR(M WASTE WATERS

7 . Author(s) Case, Oliver P.

UTILIZING CEMENTATION

9. Organization Engineered Environments 1 . ContractIGrant N o . Dive I S-802254

Anaconda American Brass Co.

15. Supplementary N o r m " '

Environmental Protection Agency report number, EPA-670/2-74-008, January 1974.

16. Abstract Bench-scale experiments utilizing the "cementation" reaction (Le electrochemical reduction by contact with a metal of higher oxidation potentiat) for the precipitatio of copper and the reduction of hexavalent chromium in industrial waste stream were performed. Reductants included: soft iron shot 4.37 uan dia, particulate iron approximately -400 mesh and alkali earth metal-silicon alloys in granular fonn.

Soft iron shot was found to be an effective reductant. is rapid and quantitative. by oxygen, speed of mixing, and available iron surface. Previously deposited copper does not significantly impede further copper deposition. reduction.

Below pH 3.0 chromium reducti Copper cementation is slower and is influenced m r e stron

Copper catalyzes the chromi

Particulate iron was not as satisfactory as the iron shot due to "clumping" of the particles when they became coated with copper.

Silicon alloys were effective for removal of copper, cinc and trivalent chromium, but not reliable for hexavalent chromium since the reaction is not primarily a reducing one.

I ?a. Descriptors

Water Pollution Abatement, Chemical Marrufacturing Wastes, Industrial Wastes, Metal Finishing Wastes, Material Recovery from Wastewaters, Metal Recovery.

17b. Identifiers

Copper Recovery, Hexavalent Chromium Reduction, Electrochemical Reduction, Electro- chemical Replacement, Cementation, Iron Reductant, Particulate Iron, Alkali Earth Metal-Silicon Alloys.

17c. CO W R R Field & Group ~

18. Availability

WATER RESOURCES SCIENTIFIC INFORMATION CLNTE U 6 DEPARTMCNTOF THE INTERIOR WASHINGTON. 0 C 20240

t U. 8. OWERNUPIT PRWTUiG OFFICI : 1914 731-939/346