EPA-670/2-74-008
January 1974
                         Environmental  Protection Technology Series
Metallic Recovery
from Waste  Waters
Utilizing Cementation
                                    Office of Research and Development
                                    U.S. Environmental Protection Agency
                                    Washington, O.C. 20460

-------
             RESEARCH REPORTING SERIES
 Research reports of the  Office  of  Research  and
 Monitoring,  Environmental  Protection Agency, have
 been grouped into five series.   These  five  broad
 categories  were established  to facilitate further
 development  and  application   of   environmental
 technology.   Elimination   of traditional grouping
 was   consciously  planned   to  foster   technology
 transfer   and  a  maximum  interface  in  related
 fields.   The five series are:

   1.  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   4.  Environmental Monitoring
   5.  Socioeconomic Environmental Studies

 This report has been assigned to the ENVIRONMENTAL
 PROTECTION   TECHNOLOGY   series.     This   series
 describes   research   performed  to  develop  and
 demonstrate   instrumentation,     equipment    and
 methodology  to  repair  or  prevent environmental
 degradation from point and  non-point  sources  of
 pollution.   This work provides  the new or improved
 technology  required for the control and treatment
 of pollution sources to meet environmental quality
standards.
                  EPA REVIEW NOTICE


Thli r«port has been reviewed by the Offlee of Research and
Develop**!*!, IPA, and approved for publication.  Approval does
not signify that the contents necessarily reflect the views and
policies ol the Environmental Protection Agency, nor does
mention of trade naves or commercial products constitute
endorsement or recommendation for use. .

-------
                                     EPA-670/2-74-008
                                     January 1974
         METALLIC RECOVERY FROM WASTE
         WATERS UTILIZING CEMENTATION
                       By


                 Oliver P. Case
               Grant No. S-802254
             Program Element  1B8036

                 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
For «al« by the Superintendent of Document*, U.S. Government Printing Office, WMhlngtoD, D.C. 20*08

-------
                                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 *-n
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



                                                                 Page




Abstract                                                          ij-




List of Figures                                                   iv



List of Tables                                                     v



Acknowledgements                                                  vl




Sections




I       Conclusions                                                1




II      Recommendations                                            2




III     Introduction                                               3



IV      Use of Soft Iron Shot as Reductant                         9



V       Use of Particulate Iron as Reductant                      23




VI      Use of Silicon Alloys as Reductant                        26




VII     References                                                31



VIII    Appendices                                                32
                                    iii

-------
                                 FIGURES

82..

1     Reactor Utilized for Reduction Experiments With Iron Shot    10

2     Effect of pH of Feed Solution on Copper Cementation With     12
      Iron Shot

3     Effect of Oxygen on Copper Cementation With Iron Shot        14

4     Effect of Surface Area of Reductant on Copper Cementation    15
      With Iron Shot

5     Effect of Previously Deposited Copper on Copper              16
      Cementation With Iron Shot

6     Reduction of Hexavalent Chromium With Iron Shot              ^

7      Simultaneous Cementation of Copper and Reduction of         19
       Hexavalent Chromium With Iron Shot

8      Effect of Speed of Mixing  on Simultaneous Copper             20
       Cementation and Reduction  of Hexavalent Chromium With Iron
       Shot

 9     Continuous  Simultaneous Cementation  of  Copper and            21
       Reduction of  Hexavalent Chromium With Iron Shot

                                                                   24
10     Cementation of Copper With Particulate  Iron
                                     iv

-------
                                 TABLES

No-                                                              Page

1     Metal Removal By Mg-Sl  Alloy Reacted By Magnetic Stirring    19

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

3     Precipitated Metals Recovered After Reaction of 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

-------
                             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 commercial systems utilizing the reaction.

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

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 much more rapid than the copper cementation.

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

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

The efficiency of cementation is not directly proportional to 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.

Farticulate 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.

-------
                               SECTION II

                             RECOMMENDATIONS

Experiments with soft iron shot were limited to a bench scale by the
scope of the project.  However, results were sufficiently promising
to warrent larger scale experiments, say on the order of flow rates
of 1 to 2 liters per minute (16 to 32 gallons per hour).

An important advantage of larger scale experiments is the better
ability to study the efficiency of copper recovery.  Small scale
experiments such as were carried out in this project do not generate
sufficient quantities of particulate copper to collect efficiently,
since much of the copper remains on the surface of the shot.

To our knowledge the kinetics of the reduction of hexavalent chromium
by metallic iron have not been studied in depth.  This is a
potentially very useful reaction since reduction of hexavalent chromium
by the usual processes is expensive and difficult to control.  A
detailed study such as Nadkarni et al4»5 undertook for the cementation
of copper with metallic iron and Strickland and Lawson? for the
cementation of copper with metallic zinc would seem worth while.

Extensive study of the reaction of silicon alloys with heavy metal
ions was outside the scope of this project since the reaction is not
primarily a reducing one.  However, the possibilities of utilizing
this reaction for removal of heavy metal ions from industrial effluents
is intriguing and it is recommended that such a study be carried out.
It should be noted that the process has been patented.10

-------
                               SECTION III

                               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  Strictly speaking the term
"cementation" is restricted to cases where the displaced metal is
precipitated from solution, e.g. 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.  In 1969, Case and Jones2 applied the reaction to
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
means for separating the precipitated copper in a recoverable form was
not developed.  Subsequent work done at the Research and Technical Center
of the Anaconda American Brass Co. resulted in the development of a
rotating chamber for carrying out the reaction.3

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
solutions, 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.  Alloys of
certain of the alkaline earth metals, especially calcium and barium, and
alloys of magnesium have also been utilized.9» 10

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.  Case and Jones?
compared the cost of treating the discharge from a medium-sized brass

-------
 mill  by a conventional  system utilizing 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*  Including capital costs for
 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.

 WORK  BY OTHERS

 A  detailed study of the kinetics of copper precipitation has been
 carried out by Nadkarni et al^»5 wfco W0rked with metal plates, wire and
 granulated cast iron suspended in a stirred solution of copper sulfate
 containing approximately 1 gram per liter of copper.

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

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

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

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

 4.  The back reaction due to ferrous ions is negligible.

 5.  The direct reaction of ferric ions does not normally affect kinetics
    of  cementation as much as the reaction between ferric ions and the
    deposited  copper.

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

7.  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.  At  lower copper ion concentrations
    excess iron consumption increased markedly and was roughly propor-
    tional to the hydrogen ion concentration.

Rickard and Fuerstenau^ 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

-------
present a simplified model.   Other conclusions  include:

1.  The specific rate constant for cupric ion reduction  is  about  twice
    as large as for ferric ion reduction.

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

3.  An increase in agitation will produce an equivalent  increase  in
    reduction rates of both ferric and  cupric  ions up to a limiting
    value.

A comprehensive study of the cementation of copper with  zinc  has  been
undertaken by Strickland and Lawson? who present an excellent summation
of the cementation reaction mechanism.  They used a rotating  disc tech-
nique developed by Levich^, using discs of the  reductant metal (iron
as well as zinc) immersed in a sulfate solution containing  10 ppm of
copper.  Conclusions include:

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

2.  When the specific mass of deposited copper is less than about 0.35
    mg/cm^, the absolute cementation rate per unit initial  zinc area
    increases with:

    a.  increasing reactant concentration
    b.  increasing relative velocity between reactant surface and solu-
        tion
    c.  increasing temperature
    d.  decreasing solution viscosity

3.  For specific deposit masses in excess of the critical,  the reaction
    rate is enhanced by increased roughness.

4.  The deposit is sufficiently porous that, normally, the  diffusion
    of product ions to the solution-deposit interface does  not become
    controlling.

Recently McKavney et al^*1^ have described experimental  work utilizing
silicon alloys for removal of heavy metals from water and brine.   It
was at first thought that this reaction involved metal "exchange" and
was, therefore, analagous to cementation, however, our own work showed
the primary mechanism to be otherwise.  This work will be further
discussed In the body of the report.

Hoover and Masselli1* proposed the use of scrap iron for the reduction
of hexavalent chromium in waste liquors from chromium plating in 1941.
Since that time the only published work discussing this reaction we

-------
                                      o
 have  found is  that  of Case and  Jones.

 PURPOSE  OF PRESENT  WORK

 Most  previous  investigations of the  cementation reaction have been
 oriented toward a hydrometallurgical application.  The present work
 differs  in that it  is oriented  toward an industrial waste, anti-
 pollution,  application.  This difference is important in at least two
 fundamental respects:

 1.  The  concentration of reducible ions in industrial waste streams is
    much lower than in most metallurgical process streams and the lower
    concentrations  are more difficult to reduce efficiently.

 2.  The  metal  ion concentrations in  the waste ultimately discharged
    must be extremely low, usually 1 ppm, or less, thus necessitating
    very efficient  separations.

 Because  of  the orientation of this work we tried to use experimental
 systems  which might be scaled directly to an industrial application.

 THEORY OF CEMENTATION

 Strickland  and Lawson  have expounded the theory of cementation in
 considerable detail.  As explained previously in this work we are
 concerned with the reduction of any ionic species by contact with a
metal of higher oxidation potential.   In particular we worked principally
with the copper-iron and the hexavalent chromium-iron systems.

Copper-iron system

The copper-iron system differs  from some similar systems  because of
the side reactions involving oxygen.   The reactions are as follows:

     CuSC-4 + Fe -»-FeS04 + Cu                                     (1)

     H2S04 + Fe —*• FeS04 + ^                                     (2)

     4FeS04 + 2H2S04 + 02 ->> 2Fe2 (S04> 3 + 2H20                    (3)

     2Cu + 2H2S04  + 02 -> 2CuS04 + 2H20                           (4)

     2Fe + 2H2S04  + 02 ->2FeS04 + 21^0                           (5)

     Fe2(S04)3  + Fe  ->3FeS04                                     (6)

                Cu  ->-CuS04  +  2FeS04                            (7)

                                 + 3HS0                          (8)

-------
Equation (1) is the basic cementation reaction which it  is  desired  to
complete.  It will be seen that the reaction does not consume acid  and
is thus not pH dependent (except, as will be explained later, pH values
above approximately 3,5 lead to precipitation of hydrous oxides  and
basic salts of iron and copper).  It is also apparent that  for the
precipitation of one unit of copper, theoretically 0.8789 units  of  iron
are required.

All of the other reactions are competing side-reactions  which it is
desirable to suppress.  Nadkarni and Wadsworth^ have discussed the
kinetics of all these reactions in considerable depth.

Equation (2) defines the dissolution of iron with acid.   It proceeds
more slowly than (1) but cannot be completely suppressed.  In our work
it is the reaction which we tried to minimize.

Equations (3), (4), and (5) show the influence of oxygen on the
reduction system.  It is obvious that oxidation of any of the components
of the system will increase the iron consumption by providing reducible
ionic species.  The effects of oxygen on the system can be eliminated
by carrying out the reaction in a reducing or an inert atmosphere.   It
is of especial advantage to carry out the cementation reaction under
hydrogen, which in any event is generated according to equation   (2),
since the hydrogen itself has some reducing action on the cupric ion.
Back^^ has described a commercial system which takes advantage of this
fact.

Equations (6) and (7) define the reactions between ferric salts  and
metallic iron and copper respectively.  Obviously this reaction can be
prevented by eliminating oxygen from the system  (unless ferric ion is
already present in the feed solution, as is usually the case in leach
liquors).

Equation (8) illustrates the precipitation of ferric hydroxide by
hydrolysis which occurs between pH 3 to 4.  This is undesirable  since
the ferric hydroxide coats the surface of the metallic iron and also
may induce coprecipitation of basic copper sulfate.

Hexavalent chromium * iron system

As previously mentioned, not much work on this system has appeared in
the literature.  The reactions are:
Na Cr 0  + 7H SO,  + 6FeSO, —>Na SO, + Cr (SO )  + 3Fe (SO,), + 7H_ 0   (10)
  227     24         4      24     243      243     2
Equation (9) is the basic reaction.  It will be noted that the reaction
requires acid and is thus pH dependent, that the iron is oxidized to
the ferric state and that for each unit of hexavalent chromium reduced,
1.0741 units of iron are required.

-------
In the presence of ferrous Ion (such as would result from cementation
of copper in the same solution) the reaction defined by (10) takes
place.

Under the proper conditions of pH, diffusion, and ratio of iron surface
to reducible ion the reaction is quantitative and extremely rapid.

The following reaction is also possible in the presence of cemented
copper:
Na2Cr20? + 7H2S<>4 + 3Cu -». Na^O^ + Cr2(S04)3 + 3CuS04 + 7H20       (11)

However, reactions (9) and (10) are so rapid it is doubtful that there
is ever sufficient residual hexavalent chromium present for reaction
(11) to proceed.

In a solution of pH 2.0 or less containing both copper and hexavalent
chromium in the absence of air the probable sequence of reactions is:
(9), (1), (2), (6), (7).
                                   8

-------
                               SECTION IV

                      USE OF IRON SHOT AS  REDUCTANT

Small diameter shot appeared to us to have several advantages  from  the
standpoint of geometry,  including:

1.  Relatively large surface area per unit volume.

2.  Point-to-point contact between pieces  rather than surface-to-surface
    contact.

3.  Good mixing action.

4.  Easily controlled surface area.

REDUCTANT

The shot utilized was obtained from Grossman Arms Co. Inc.,  Freeport,
N.Y..  Analysis showed it to have a carbon content of approximately
0.057. and only traces of metallic contaminants.  The shot had an average
diameter of 4.37 mm, an average weight of  0.34 grams and a packing
density of 642 shot per 50 ml.  Thus 50 ml of shot had a surface area
of approximately 385 cm2.  The shot as received was covered with a  film
of lubricant which was removed by a hot water and soap solution followed
by an acetone rinse.  Normally the shot was cleaned by pickling in  a
mixture of 257. NH4OH (sp. gr. 0.90) plus 57. H202 (30%) immediately
before use and between runs.

When a volume of less chan 50 cc of shoe was used in order to reduce
the iron surface, the difference in volume was made up by adding chem-
ically-resistant glass beads of approximately the same diameter.  It
was thought that this would keep the shear rate between the solids  and
the solution relatively constant.

REACTOR

The reactor was a 500-ml Erlenmeyer flask  mounted about 30° from the
horizontal on a circular plate capable of  revolving at speeds up to
60 RPM.  In most experiments nitrogen was  fed through a tube reaching
nearly to the bottom of the flask at the rate of 2 liters per rain.

When the reactor was operated on a continuous basis the feed solution
was pumped into the flask through a tube reaching nearly to the bottom
and the overflow allowed to drip off a collar around the mouth of the
flask into a collection beaker.  Fig. 1, shows the arrangement.

-------

Figure 1.  Reactor Utilized for Reduction Experiments with Iron Shot

-------
FEED SOLUTIONS

Feed solutions were normally made up to contain 100  pptn of  the
reducible ion species (copper and hexavalent chromium)  which  is  the
highest level likely to be found in brass  mill discharges.  Reagent
grade CuSO^.Sl^O was used as a source of cupric ion  and ^20^07.2H20
of similar purity was used as a source of  hexavalent chromium ion.   The
pH of the solutions was adjusted with ^SO^ or NaOH.

EXPERIMENTAL PROCEDURES

Experiments were carried out by both batch and continuous  reactions.
The procedures will be formed in Appendix  A, page 33.

EXPERIMENTAL RESULTS (BATCH REACTIONS)

Copper only in feed solution

Effect of pH- The pH of feed solutions was adjusted  to 1.50,  1.90,  and
3.00 and the solutions reacted with 50 cc  of shot in air.   Curves
illustrating che results are plotted on Fig. 2.

It will be noted that in all instances 99% of the copper was  removed
in 10 min .  The rates of removal were quite similar, although  the  rate
appeared to be slightly higher at pH 3.0.   However,  since the initial
pH of the solution quickly rose to nearly 5, ferric  iron precipitated
as ferric hydroxide and probably carried some copper in the form of
basic sulfate with it.  The fact that the precipitates from runs at
pH 1.5 and 1.9 carried a negligible amount of copper, whereas preci-
tates from runs at pH 3.00 averaged nearly 1 mg of copper lends
credence to this probability.  Hence, there is probably no real differ-
ence in the rates of the reaction.

The fact that a higher level of excess iron dissolved in a solution
having an initial pH of 1.90 than in one having an initial pH of 1.5
is puzzling.  A possible explanation may be the increased oxidation of
ferrous ion at higher pH.  Data for excess iron dissolved in a solution
having an initial pH of 3.00 is not available.  As previously mentioned,
a significant part of the dissolved iron precipitated as ferric hydrox-
ide.

Only at the median initial pH was the rise in pH gradual.  At both
higher and lower acid concentrations the pH leveled out rapidly.  At
the higher pH this results from the equilibrium by the precipitation
of ferric hydroxide.  At the lower pH it results from the fact that
relatively large changes in hydrogen ion concentration result in only
small pH changes in this area.  It is interesting to note that at the
median initial pH the curves for pH rise and excess iron dissolution
roughly parallel each other.

Effect of oxygen- To assess  the effect of eliminating oxygen from the
                                   11

-------
  100
1
o
e/j
5
   75 _
50



25



 0

35 —

30

25



15

10

 5

 0
100 ml SOLUTION
100 ppm Cu
50 cc SHOT
TUMBLED 50 RPM
ATMOSPHERE - AIR
O pH 1.50
A pH 1.90
D pH 3.00
                                 6         8
                             REACTION TIME, min.
10
                                                           12
      Figure 2.   Effect of pH of Feed Solution on Copper Cementation
                 with Iron Shot
                                    12

-------
system, nitrogen was bubbled through solutions at 2 liters/min.  during
a series of runs.  Also included in this series of experiments were
runs made without any copper in the feed solution.  Curves illustrating
the results are plotted on Fig. 3.

The rate of copper removal was better in the nitrogen atmosphere.   This
may be due to the fact that re-solution of copper by oxidation was
prevented*

Interestingly the level of excess dissolved iron remained fairly constant
in the nitrogen atmosphere.  This seems to confirm the observations of
Nadkarni and Wadsworth5 on the increased iron consumption resulting from
oxidation.  It is however, surprising to note that iron dissolution is
at a lower level when there is no copper present in solution; quite
possibly the metallic copper deposited on the surface of the iron
catalyzes the dissolution.

In the absence of copper the pH levels out at a slightly higher  value.
This is at variance with the fact that slightly less iron is dissolved
under the same conditions.

Effect of surface area- Runs were made using 50 cc, 25 cc, and 12.5 cc
of shot.  Where less than 50 cc of shot was used the volume was  made
up to 50 cc with glass beads.  Data illustrating the results are shown
on Fig. 4.

It is evident that the efficiency of the reaction is not directly pro-
portional to the surface area of the iron.  For instance, by reducing
the surface of the iron from 385 cm2 to 193 cm2, the amount of copper
removed from the -solution in lOmin. is reduced only 8%, although the
excess iron dissolved is halved and the pH rise is reduced by more
than 907..

By reducing the surface of the iron from 385 cm2 to 96 cm2* the amount
of copper removed from the solution in 10 minutes is reduced only 367.,
whereas the excess iron dissolved is only about one fifth and the pH
rise is reduced by about 957..

It seems obvious that in practice it is advantageous to use the
minimum iron surface which will accomplish satisfactory copper removal
in a reasonable time.  From the data shown on Fig. 4, at the 100 pptn
copper level a minimum of about 200 cm2 of iron surface is required to
remove 90% of the copper from  100 ml of solution during a contact
period of 10 min. (equivalent  to approximately 0.045 mg Cu/cm2Fe).
During this period about 13 mg of excess iron will be dissolved (equiv-
alent to approximately 1.4 mg  excess Fe/mg Cu cemented).

Effect of previously deposited cooper- To determine the effect of
previously deposited copper on the efficiency of the iron reductant
six consecutive  10-min. runs were made using the same charge of shot.
                                   13

-------
w
w
w
O
100



 75i__



 50



 25



  0

 35

 30

 25

 20

 15

 10

   5

   0

   4
§
I
                                                    100 ml  SOLUTION
                                                    100 ppm Cu
                                                    pH 2.0
                                                    50 cc SHOT
                                                    TUMBLED 50  RIM
                                                    O RUN  IN AIR
                                                    A RUN  UNDER  N0
                                                    D  NO  Cu(N2 ATMOS.)
                                 6         8
                             REACTION TIME, min.
                                                  10
12
     Figure 3.  Effect of Oxygen on Copper Cementation With Iron Shot
                                   14

-------
9
3
100



 75j_



 50



 25



  0

 35

 30

 25

 20

 15
S 10
M
U
 a   *
 B
                                            I
100 ml SOLUTION
100 ppm Cu
pH 2.0
TUMBLED 60 RBI
ATMOSPHERE - N2

 O 50 cc SHOT
 A 25 cc SHOT
 D 12.5 cc SHOT
  I
                                                    10
           12
    02468
                            REACTION TIME, rain.
    Figure 4.   Effect  of  Surface Area of Reductant on Copper Cementation
               With Iron  Shot
                                    15

-------
  100
   75
§
                                               100 ppm Cu
H
s
Q
50
  25
   4
5.
A  3
o  ,
w  1
                                             100 ml  SOLUTION
pH 1.90
50 cc SHOT
TUMBLED 50 RPM
ATMOSPHERE - AIR
CONTACT TIME - 10 mln.
NO CLEANING
BETWEEN RUNS
   0
             o-
             12         3456
                                   RUNS

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

 Results are  plotted on Fig.  5.   Since  the  runs  were made  in  air
 substantial  amounts of ferric ion were precipitated as hydroxide.  The
 data  show chat better than 99% of the  copper was  removed  during each
 run and that there was no  deterioration of efficiency in  succeeding
 runs.   Evidently any film  which forms  on the surface  of the  iron is
 porous  enough so that the  feed solution readily penetrates to the
 metal.   This confirms the  observation  of Strickland and Lawson?.

 Hexavalent chromium only in feed  solution

 In experimenting with hexavalent  chromium  solutions it quickly became
 apparent that the reduction proceeded  rapidly to  completion.  Typical
 data  are plotted on Fig. 6.

 When  these data are compared with those for copper under  the same
                                   16

-------
  100,






  75
H
 ft




| 50

9
rt



  25






   0



  35



 e 30



I 25

d
« o
tn ^
M
Q
 oi 15
fe


310


^5


   0
§
M  o

I
   x
    I—       e-
                                                  100 ml SOLUTION

                                                  100 ppm Cr*"6

                                                  pH 2.0

                                                  12.5 cc SHOT

                                                  TUMBLED 30 RPM

                                                  ATMOSPHERE - N*
                                6         8

                            REACTION TIME, tnin.
                                                   10
12
    Figure 6.   Reduction of Hexavalent  Chromium With Iron Shot
                                  17

-------
conditions which are plotted on Fig. 4 it is apparent that:

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

2.  The chromium reduction results in dissolution of slightly more
    excess iron than the copper cementation, probably due to presence
    of ferric ion.

3.  The pH rise during the chromium reduction is a little greater than
    that which occurs during copper cementation.  This is to be expected
    since the reaction consumes acid.

Copper and hexavalent chromium in feed solution

Experiments were run to study the effect of equal amounts of copper and
hexavalent chromium upon the reduction of each other.  Data are plotted
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-
    tion.  For a contact time of 8 min. the cementation of copper is
    about 16% less.

2.  The presence of hexavalent chromium also suppresses the dissolution
    of excess iron.

3.  The presence of hexavalent chromium causes an increase in pH rise,

Effect of copper on the reduction of hexavalent chromium- The following
effects are apparent:

1.  The presence of copper increases the rate of reduction of hexavalent
    chromium slightly.  This effect was also noted by Hoover and
    Masselli10.

2.  The copper suppresses the dissolution of excess iron.  This is
    contrary to the effect noted during a continuous run where, as will
    be seen later,  the dissolution of excess iron was increased in the
    presence of copper.

3.  The copper has  little effect on the pH rise during the reaction.

Effect of mixing speed- Since reduction by a solid material is a diffu-
sion controlled reaction,  the speed of mixing necessarily influences
the rate of reduction.   Data obtained by increasing the tumbling rate
from 30 to 60 RPM are  plotted on Fig.  8.  It will be noted that a slight
increase in the copper cementation rate occurred.  Reduction of
                                   18

-------
 100,
  75
y so
a
05
  25



   0

  35

»30

325
1
VI >
w '
H
 0,
   15
                          "Cr
                                    Cu
                                O
                                                   100 ml SOLUTION
                                                   pH 2.00
                                                   12.5 cc SHOT
                                                   TUMBLED 30  RPM
                                                   ATMOSPHERE  - N2

                                                    O 99 ppm Cu
                                                   A 96 ppm Cr+6
                                                    D 100 ppm  Cu +
                                                      107 ppm  Cr*6
                                 6         8
                             REACTION TIME,  min.
                                                   10
12
     Figure 7.   Simultaneous Cementation of Copper and Reduction of
                Hexavalent Chromium With Iron Shot
                                   19

-------
  100
   75
§
M
H

§50
  25
   0
§
en
                       100% Cr+6 REDUCTION BOTH RUNS

                                         A
 100 ml SOLUTION
 100 ppm Cu
 25 ppm Cr+6
 pH 2.00
 25 cc SHOT
 ATMOSPHERE - N2
 O TUMBLED 30 RIM
A TUMBLED 60 RIM
                             IDENTICAL BOTH RUNS
                                6        8
                            REACTION TIME, min.
  10
12
    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 instances.  There
 was no effect on pH rise.  It is probable that  the  results would have
 been more pronounced  if the change in rotational speed had been  greater
 or the ratio  of iron  shot volume to liquid  volume had been greater.

 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
     not change greatly during the  course of the experiment.  The amount
     of copper cemented averaged a  little over 797..

 2.  The amount of excess iron dissolved  increased very gradually and
     finally stabilized at about 320 ppm.
                                    20

-------
   100

    75

    50|—

    25

     0
o

x   50

a   40
 •t
w   30
M
°   20
w  in
CO  iU
w
§   0
I    2
M    1
                                           100% Cr+6 Reduction
50 cc SHOT
TUMBLED 30 RBI
ATMOSPHERE -
FLOW - 25 ml/min
pH 2.05
                                                                                Z
                                                                   O 95 ppm Cu
                                                                  A 99 ppm Crf6
                                                                   D 101 ppm Cu + 115 ppm Cr+°
                                                                          -©•
                                                                                   ^tŁ
                                                            I      I      I	I	L
      0    0.5    1.0   1.5   2.0    2.5    3.0    3.5    4.0    4.5   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 immediately
    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 of iron
    surface (in terms of the requirements for quantitative reduction
    of the hexavalent chromium) the amount of excess iron dissolved was
    very low.  This may indicate that under stabilized dynamic condi-
    tions the presence of chromium inhibits dissolution of iron by
    acid.

 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.  Again the reduction of hexavalent chromium was quantitative during
    the course of the experiment.  However, the cementation of copper
    was slightly depressed, averaging only 71.297..

 2.  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.  It is significant that the
    curves for excess iron dissolution by chromium alone and by chromium
    plus copper parallel each other closely, indicating the dominating
    effect of the chromium.

 3.  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 dis-
 lodged from the surface of the shot.  While some of these were swept
out of the reactor by the flow of solution, most collected in the bottom
of the reactor.  It seems likely that at higher flow rates coupled with
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,  907. or more of the copper adhered to the surface of the
shot.
                                   22

-------
                                SECTION V

                  USE OF PARTICULATE IRON  AS  REDUCTANT


Back 2 has described the advantages  of particulate iron as  a medium  for
the commercial cementation of copper.  This material was explored  only
briefly in the present work because  the use of iron shot seemed much
more suitable for a small-scale laboratory system.

REDUCTANT

Two different sizes of iron particles were used:   approximately -400
mesh (Leco Low-Sulphur  Iron Powder  Accelerator No. 501-78, Laboratory
Equipment Corp., St. Joseph, Mich.)  and -325  mesh (ACF  Powder,
Hoeganaes Corp., Riverton, N.J.).

FEED SOLUTION

The feed solution contained 100 ppm  of Cu as  CuSO^ and  had a  pH of 2.0.

EXPERIMENTAL PROCEDURES

The system was tested only with copper and only in batch reactions.
The procedures will be found in Appendix B,  page 34.

EXPERIMENTAL RESULTS

It was observed that initially iron powder of either type was uniformly
dispersed throughout the feed solution, however, as copper was cemented
on the surface the individual particles began to clump together until
finally most of the metal remained on the bottom of the reaction vessel.
Data for both runs are plotted on Fig. 10.  The data for the two runs
are not directly comparable with each other or with any of the data
obtained with iron shot.

Several facts may be observed from the data:

1.  The removal of copper in the mechanically stirred system gradually
    deteriorated with time.  This is no doubt due to the "clumping" of
    the iron particles.

2.  The removal of copper in the nitrogen-dispersed system gradually
    increased with time, probably because the "clumping" effect was not
    as serious when a fresh charge of  iron was used for each run.

3.  The amount of excess  iron dissolved in the mechanically stirred
    system gradually  decreased after an initial  sharp  drop, whereas the
                                   23

-------
  100
8-4
a
3
U
W
Cfl
(d
EC
 O.
75|_



50



25



  0

35

30

25

20

15



  5

  0

  k

  3

  2
                                                    1OO ml SOLUTION
                                                    100 ppm Cu
                                                    pH 2.0

                                                   O  1 g LEGO POWDER
                                                      MECHANICALLY
                                                      STIRRED

                                                   A  0.05 g ACF POWDER
                                                      DISPERSED WITH N2
                                  6        8
                              REACTION TIME, min.
                                                  10
12
     Figure 10.  Cementation of Copper With  Particulate  Iron

-------
    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 particles greatly limits
the usefulness of this 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
                                                    q
As previously noted, McKavney, Fassinger and Stivers7 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.  They
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 mm x 30 mm 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.  The advantage over conventional removal tech-
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.

Because of the presumed analogy to the "cementation" reaction we
investigated the sytem both on a batch and a continuous (plug-flow
reactor) basis.

REACTANTS

The silicon alloys used were:

1.  Magnesium Ferrosilicon (Si 44.50%, Mg 8.907., 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 32.47., Si 63.17., Fe 4.07., Ba 0.4%) 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.  The
composition was:

             Copper                    45 ppm
             Zinc                      43 ppm
             Chromium (+3)             43 ppm
             Chromium (+6)             7.5 ppm
             pH                        2.05 ppm


                                   26

-------
All cations were present as sulfates.

EXPERIMENTAL PROCEDURES

Contact between feed solution and reactants was achieved by:

1.  Magnetic Stirring  (Mg-Si alloy only)*

2.  Tumbling.

3.  Continuous flow through a vertical column.

The procedures are described in Appendix C page 35.

EXPERIMENTAL RESULTS

Magnetic Stirring

Results are shown in Table 1.
Table 1. METAL
Run
1
2
3
4
5
6
ppm
Copper
Eff. %Rem.
<.5 98.9+
<.5 98.9+
<.5 98.9+
0.6 98.7
0.5 98.9
0.9 98
REMOVAL BY Ma
ppm
Zinc
Eff. %Rem.
<.l 98.8+
<,1 98.8+
0.2 99.5
0.6 98.6
0.5 98.8
1.1 97.4
•Si ALLOY REA
ppm
Chromium^
Eff. %Rera.
<1.0 98+
<1.0 98+
<1.0 98+
<1.0 98+
<1.0 98+
1.2 97.6
CTED BY MAGNETIC STIRRING
ppm
Chromium4"^
Eff. 7«Rem.
0 100
0 100
<.l 98.7+
0 100
0 100
0 100
ppm
Mg
Eff.
160
190
200
202
204
202
pH
Eff.
10
10
9.8
9.8
9.85
9.75
 Some readily apparent  conclusions are:
1.
2.
     In 6 successive  5-min.  runs,  removal of all heavy metals exceeded
     97%, although there was a very  slight  deterioration after the 5th
     run.
     The pH of the solution rapidly stabilized at  10 which approaches
     that of a saturated magnesium hydroxide  solution.

 3.   After the first run the amount of magnesium which  entered the  solu-
     tion remained fairly  constant at  about 200  ppm which is much higher
     than would be expected for a saturated magnesium hydroxide  solution,
     indicating that some  of the magnesium was present  as sulfate.

 A run made with 100 ml of distilled water under the same conditions as
 the runs with feed solution containing  metal ions and  acid  resulted in
 a pH of 10.5 and a magnesium concentration of 10  ppm.
                                    27

-------
Tumbling

Mg-Si alloy- Results are shown in Table 2.
Table 2. M
Run
1
2
3
4
5
6
ppm
Copper
Eff. %Rem.
<.5 99
<.5 99
0.8 98
<.5 99
<.5 99
3.3 93
ETAL REMOVAL BY ME- Si ALLOY REACTED BY TUMBLI
ppm
Zinc
Eff. 7.Rein.
<.l 100
<.l 100
0.2 100
<. 1 100
<.l 100
3.0 93
ppm
Chromium™
Eff. %Rem.
6.8 87
8.8 83
8.8 83
6.8 87
5.7 89
12.6 75
ppm
Chromium™
Eff. 7.Rem.
5.0 33
5.0 33
5.5 27
5.0 33
4.0 47
5.0 33
ppm
Mg
Eff.
80
60
57
49
40
54
NG
PH
Eff.
10.25
10.25
10.2
10.1
10.0
10.0
 It is evident that:

 1.  Removal of copper and zinc is satisfactory for the first 5 runs,
    but efficiency drops off after Che 5th run.

 2.  Removal of trivalent chromium is fairly satisfactory, though less
    complete than for copper and zinc.

 3.  Removal of hexavalent chromium is unsatisfactory.

 4.  The amount of magnesium taken into solution is only about one-
    third the amount dissolved by the more vigorously stirred solutions.

 5.  There is a marked deterioration in efficiency of removal of all
    metals after the 5th run.  The teason for this is not clear since
    the pH remained constant at about 10.

The precipitates from runs 4 and 5 were analyzed for copper, zinc, and
 chromium.  Results are tabulated in Table 3.

 Table 3.  PRECIPITATED METALS RECOVERED AFTER REACTION OF FEED SOLUTION
                            WITH Mg-Si ALLOY


Run
4
5


Rem.
4.5
4.5
mg
Copper
Rec.
4.0
3.2


%Rec.
89
71


Rem.
4.3
4.3
mg
Zinc
Rec.
2.8
2.4


%Rec.
65
56

mg

Chromium
Rem.
4.4
4.5
Rec.
3.3
2.7
7.Rec.
75
60
We believe the results to indicate that a significant amount of the
precipitated metals remains on the surface of the reactant metal.
                                   28

-------
Ca-Si Alloy- Resulcs are shown in Table 4.
Table 4.

Run
1
2
3
4
5
6
Ppro
Copper
Eff. 7oRem.
<.5 99
<.5 99
<.5 99
1.1 98
1.1 98
7.2 84
4ETAL REMOVAL BY Ca-Si ALLOY REACTED BY TUMBLI
ppm
Zinc
Eff. 7.Rem.
0.2 99+
<,1 100
0.5 99
0.8 98
8.7 80
39 9
ppm
Chromiunr
Eff. 7Jlem.
2.0 96
1.8 96
1.3 97
3.4 93
2.4 95
12.6 75
PPm
Chromium1"0
Eff. %Rem.
1.2 84
1.2 84
1.0 87
1.4 81
1.0 87
1.2 84
ppm
Ca
Eff.
100
96
84
100
74
48
NG
PH
Eff.
9.95
9.35
7.95
7.95
6.35
5.05
Significant differences from the behavior of the magnesium-silicon
alloy include:

1.  Removal of zinc decreases rapidly after the 4th run,  probably
    because of the drop in pH.

2.  Reduction of hexavalenc chromium remains fairly constant at 85%
    through the first 5 runs.  This is significantly better than achieved
    with the magnesium alloy, probably because of the higher oxidation
    potential of calcium.

3.  The pH drops progressively as runs are continued.

4.  The amount of calcium dissolved drops progressively as runs are
    continued.  This may be because all of the calcium has been leached
    from the surface of the alloy or because the surface has been coated
    with a film of reaction products which inhibit further dissolution.

Continuous (Plug-Flow) Reaction

Results are shown in Table 5.
Tablj
Elapsed
Time,
Min.

0-15
15-30
30-45
45-60
60-75
75-90
*ppm
i 5. ME
Zn
%Rem.

98
92
91
88
82
80

TAL REMOVAL BY M
Cr
%Rem*

99
91
89
86
78
77

Fe
%Retn»

86
78
74
71
65
63

K-Si ALLOY REACTED BY CONTINUOUS FLOW
Fe
in ef-
fluent
*
<.5
<.5
<.5
<.5
<.5
<.5

Mg
in ef-
fluent
*
58
61
62
60
63
65

Si
in ef-
fluent
*
5
6
6
6
7
7

PH

9.70
10.10
10.05
10.0
10.0
9.95

                                   29

-------
It will be noted that all of the metals were removed less effectively
with the passage of time.

During the course of the run it was noted that the effluent exhibited
increasing turbidity.  When the column was back flushed after the run
a large quantity of finely-divided, black, amorphous material was re-
moved.  This was found to contain a high concentration of magnesium,
but only small amounts of the metals present in the feed solution.

GENERAL CONCLUSIONS

Our study of the silicon alloy system led to the following conclusions;

1.  The primary reaction appears to be formation of hydroxide by reac-
    tion of the alkali component of the alloy with water, thus:
    Mg + H20-^-MgO + H2

    MgO + H
In turn the magnesium hydroxide reacts with metal ions  to form hydrous
oxides or basic salts.

2.  A reduction reaction also appears to take place,  thus:

    3Mg + 2Cr03 + 6H2S04-s*-3MgS04 + Cr2(S04)3 + 6H20
This reaction is probably inhibited by the rapid neutralization of the
system.

3.  There is no evidence that any constituent of the alloy other than
    the calcium or magnesium enters into the reaction.

4.  The effectiveness of the alloys deteriorates before the bulk of
    the material is significantly diminished.  We believe this  to be
    a surface phenomenon, but have not determined whether it is due
    primarily to depletion of the active metal on the surface of the
    alloy granules or to formation of an inhibiting layer of reaction
    products.  Vigorous stirring of the mixture of alloy granules and
    feed solution helps to prolong the effectiveness of the metal
    removal.

    In spite of the drawbacks of the system we believe  alkali earth
    metal alloys and alloys of other metals with high oxidation poten-
    tial offer an interesting area for further study.
                                   30

-------
                              SECTION VII

                              REFERENCES

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

2.  Case, 0. P., and Jones, R. B. L,, "Treatment of Brass Mill
    Effluents", Conn. Research Commission RSA-68-34, Sept.  10, 1969.

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

4.  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. AIME, Vol. 239, Apr. 1967, p. 581,

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

6.  Rickard, R. S., and Fuerstenau, M. C., "An Electrochemical
    Investigation of Copper Cementation by Iron", Trans, Met. Soc,
    AIME. Vol.  242, Aug. 1968,  p. 1487.

7.  Strickland, P.H.,  and Lawson, F.,  "Cementation  of Copper  With
    Zinc From Dilute Aqueous  Solutions",  Proc. Aust.  InstA Min. Met.
    No.  236, Dec.  1970, p.  25.

8.  Levich, V.  G., Physiochemica1 Hydrodynamic s,  Prentice Hall,
    Englewood Cliffs,  N.J.  1962.

 9.  McKaveney,  J.  P.;  Fassinger,  W.  P.,  and  Stivers, D.  A.  "Removal
    of Heavy Metals  from Water and  Brine Using  Silicon  Alloys",
    Environ.  Sci.  and Tech..  Vol.  6,  No.  13, Dec.  1972,  p.  1109.

10.  McKaveney,  J.  P.,  "Process For  the Removal  of Ionic Metallic
     Impurities  From Water",  U.S. patent 3.766,036.  Oct.  16,  1973.

11.   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.

12.   Back,  A.  E.,  "Precipitation of Copper from Dilute Solutions
     Using Particulate Iron," Jour.  Met.. May 1967, p. 27.
                                    31

-------
                              SECTION VIII

                               APPENDICES

                                                                  Page

A.  Experimental Procedure Utilized With Iron Shot as              33
    Reductant

B.  Experimental Procedures Utilized With Particulate Iron         34
    as Reductant

C.  Experimental Procedures Utilized With Alkaline Earth-          35
    Silicon Alloys as Reactant
                                   32

-------
                               APPENDIX A

EXPERIMENTAL PROCEDURE UTILIZED WITH IRON SHOT AS REDUCTANT

BATCH REACTION

1.  Place 50 cc of freshly-cleaned shot, or combination of shot and
    glass beads, in reactor shown in Fig. 1.
                            a
2,  Start flow of nitrogen at 2 liters/min. (initial experiments were
    carried out in air).

3,  Add 100 ml of feed solution to reactor.

4.  Rotate reactor at 30 to 60 RPM for 2 min.

5.  Pour solution through a No. 5 filter paper.

6.  Test the filtrate qualitatively for ferric iron.  Determine pH
    and concentrations of iron and residual reducible ions (copper
    and hexavalent chromium).

7.  Analyze the precipitate for copper and  iron  (in some cases).

8.  Repeat using contact times of 4, 6, 8,  10, and 12 min..

CONTINUOUS  (BACK-MIX) REACTION

1.  Place 50 cc of freshly-cleaned shot, or combination of shot and
    glass beads in reactor  shown in Fig. 1.

2.  Start flow of nitrogen  at 2 liters/min..

3.  Fill the reactor  to over flowing with  feed solution  (480 ml).

4.  Rotate  reactor at 30 RPM.

5.  Pump feed  solution through  reactor at  the rate of  25 ml/min.  for
    8 hours.

6.  Following  an  initial period of  30  min., sample  effluent over  a
    15-min.  period at 30-min.  intervals.

7.  Analyze effluent  for  iron,  residual copper and  hexavalent  chromium
    and determine pH.
                                    33

-------
                               APPENDIX B

            EXPERIMENTAL PROCEDURES UTILIZED WITH PARTICIPATE

                            IRON AS REDUCTANT

LEGO POWDER

1.  Place 1 g of powder in a 250-ml beaker.

2.  Add 100 ml of feed solution.

3.  Stir with a mechanical stirrer at 500/700 RPM (rough estimate) for
    2 min..

4.  Decant solution and analyze for iron and copper and determine pH.

5.  Rinse powder with distilled water and repeat steps 2 thru 4
    using contact times of 4, 6, 8, 10, and 12 min..

HOEGANAES ACF POWDER

1.  Add 100 ml of feed solution to a 250-ml separatory funnel connected
    to a source of nitrogen through the stem.

2.  Add 0.05 g of powder (in the form of a well-mixed 17. suspension)
    and disperse the particles throughout the feed solution with a
    stream of nitrogen (0.8 liters/min.).

3,  After a 2-min. contact time, pour the solution through a filter
    paper to remove iron particles.

4.  Analyze the filtrate for copper and iron and determine pH.

5.  Repeat steps 1 through 4, using contact tiroes of 4, 6, 8, 10, and
    12 min..
                                   34

-------
                               APPENDIX C

             EXPERIMENTAL PROCEDURES UTILIZED WITH SILICON

                           ALLOYS AS REACTANT

BATCH REACTIONS

Mixing By Stirring Pig-Si Alloy)

1.  Add 50 cc (78 g) of granules to a 300-ml Erlenraeyer flask.

2.  Add 100 ml of feed solution to flask.

3.  Stir for 5 min. with a 4.45 cm magnetic stirrer at approximately
    200 RPM (rough estimate).

4.  Filter solution and analyze filtrate for copper, zinc, chromium (T),
    chromium  (+3) and magnesium and determine pH.

5.  Using the same metal charge, add another 100 ml of feed solution
    and repeat steps 3 and 4.

6.  Continue runs as in step 5 until 6 runs have been made.

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

1.  Add 50 cc of granules to reactor shown in Fig.  1.

2.  Add 100 ml of feed solution to reactor.

3.  Rotate reactor at 45 RPM for 5 min.

4.  Filter solution and analyze filtrate for copper, zinc, chromium (T),
    chromium  (+3), and magnesium and determine  pH.

5.  Using the same metal charge, add another 100 ml of feed solution
    and repeat steps 3 and 4.

6.  Continue  runs as in step 5 until 6 runs have been made.

CONTINUOUS  (PLUG-FLOW) REACTION

1.  Place 50  cc  (81.5 g) of  10/20 mesh Mg-Si granules in  a vertical
    glass column  12.7 mm I.D. x 419 cm  long.

2.  Pump  feed solution up  through the packed column at a  flow rate of
    10 ml/min. for  90 min..
                                    35

-------
3.  Collect effluent, replacing receiver every 15 min..

4.  Analyze each portion of effluent for copper, zinc, chromium (T)
    and determine pH.
                                   36

-------
SELECTED WATER
RESOURCES ABSTRACTS

INPUT TRANSACTION FORM	
4.  Til If
   METALLIC RECOVERY FROM WASTE WATERS UTILIZING CEMENTATION
7.  Authorfs)
  Case, Oliver P.
9.  Organization
               Engineered Environments
                          Div.
              Anaconda American Brass Co.
      11.  Contract I Grant No.
         S-802254
   Environmental Protection Agency report number,
   EPA-670/2-74-008,  January 1974.
 !'j.  A hi it act
 Bench-scale experiments utilizing the "cementation"  reaction (ie 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 were
 performed.  Reductants included:  soft iron shot 4.37 mm dia,  particulate iron
 approximately -400 mesh and alkali earth metal-silicon  alloys  in granular form.

 Soft iron shot was found to be an effective reductant.   Below  pH 3.0 chromium reduction
 is rapid and quantitative.  Copper cementation  is slower and is influenced more etrongl;
 by oxygen, speed of mixing, and available iron  surface.  Previously deposited copper
 does not significantly impede further copper  deposition.  Copper catalyses the chromium
 reduction.

 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, zinc and  trivalent chromium, but .
 not reliable for hexavalent chromium since the  reaction is not primarily a reducing
 one.
 J7a. Descriptors
 Water Pollution Abatement, Chemical Manufacturing 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.
 J~c. CO WRR Field & Group
 IS. Availahilitv
          Oliver P. Case
                                                     Send To:
                                                     WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                                     US DEPARTMENT OF THE INTERIOR
                                                     WASHINGTON. D C. 2O24O
Emri rnrpxmf..
                                                      « U. S. OOV«IW1«NT PRINTINC OFFICE : 1»74 731-939/346

-------