United States
Environmental Protection
Agency
Industrial Environmental Research EPA-600 2-79-1 30
Laboratory July 1979
Cincinnati OH 45268
Research and Development
Activated Carbon
Process for
Treatment of
Wastewaters
Containing
Hexavalent
Chromium
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental 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.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-130
July 1979
ACTIVATED CARBON PROCESS FOR
TREATMENT OF WASTEWATERS CONTAINING HEXAVALENT CHROMIUM
by
C. P. Huang
and
Alan R. Bowers
University of Delaware
Newark, Delaware 19711
Grant No. R-804656
Project Officer
Mary K. Stinson
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
Edison, New Jersey 08817
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory - Cincinnati, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection Agency,
nor does mention of trade names or commercial products constitute endorsement
or recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory -
Cincinnati (IERL-CI) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
These studies were undertaken to develop a laboratory scale activated
carbon process for treatment of hexavalent chromium containing wastewater.
The mechanisms for removal of hexavalent chromium by activated carbon and
the effects of various aqueous solution parameters on removal were examined,
as well as different methods of regenerating the activated carbon.
i
Such information will be of value both to EPA's regulatory program
(Effluent Guidelines Division) and to the industry itself in arriving at
meaningful and achievable discharge levels. Within EPA's R & D program the
information will be used as part of the continuing program to develop and
evaluate improved and less costly technology to minimize industrial waste
discharges. Besides its direct application to effluents from metal finishing
industry, this technology may find application to treat metal-containing
wastes generated by a host of other industries.
For further information concerning this subject the Industrial Pollution
Control Division should be contacted.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
The removal of hexavalent chromium, Cr(VI), from dilute aqueous solu-
tion by an activated carbon process has been investigated. Two removal
mechanisms were observed: hexavalent chromium species were removed by ad-
sorption onto the interior carbon surface and/or through reduction to the
trivalent state at the external carbon surface.
The effects of Cr(VI) concentrations, pH, carbon dosage and extent of
mixing in the reaction vessel were studied in the batch mode and in contin-
uous flow packed column experiments in the laboratory. The adsorptive
capacity of the carbon and the rates of Cr(VI) adsorption and reduction
have been determined.
Thermal regeneration of the exhausted carbon was examined, along with
caustic or acid stripping solutions and a combined caustic-thermal process.
A case study was presented and the experimental data and rate expres-
sions obtained from the data were used to evaluate the design variables
(i.e., pH, carbon dose, Cr(VI) concentration and mixing in the reaction
vessel). Several Cr(VI) treatment schemes were proposed, together with an
economic analysis of each scheme.
This report was submitted in fulfillment of Grant No. R804656
by the University of Delaware, Department of Civil Engineering, under the
sponsorship of the U. S. Environmental Protection Agency. This report
covers the period from October 18, 1976, to May 1, 1978, and work was
completed as of May 17, 1978.
iv
-------
CONTENTS
Foreword .- ill
Abstract iv
Figures vi
Tables ix
Acknowledgment. x
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. Activated Carbon - Background Information 7
General 7
Chromium(VI) and Chromium(III) 7
5. Experimental Procedure 10
Batch Mode Experiments 10
Packed Column Experiments 10
6. Results and Discussion 13
Activated Carbon-Selection, Physical and
Chemical Properties 13
Adsorption of Cr(VI> 28
7. System Design 51
System 1 -r Treatment by Adsorption in a
Completely Mixed Reactor (CMR) 52
System 2 - Column Operation 60
References ( 64
Appendix - Aqueous Chemistry of Chromium 67
-------
FIGURES
Number Page
1 Experimental Apparatus: a. batch reactor; b. packed column
experiments ........................... 9
2 The effect of Filtrasorb 400 on solution pH ........... 15
3 The electrophoretic mobility of various commercial activated
carbons (after Huang and Ostovic) ................ 16
4 A simplified view of the chemical and physical interactions
of chromium with activated carbon in dilute acidic medium. ... 17
5 Reduction potentials of several common oxidizing agents
(after Latimer) ................. ........ 18
6 Ratio of observed reduction rate to the maximum rate as a
function of HCrOit" Cr(VI) concentration ............. 19
7 a. Batch removal of Cr(VI), and b. production of Cr(III) ... 21
8 Inverse plot of initial reduction rate and Cr(VI) concentration
(data were taken from Figure 7) ................. 22
9 The removal of Cr(VI) and production of Cr(lII) at different
mixing intensities ....................... 23
10 The effect of velocity gradient on the reduction rate constant
11 Inhibition of the surface reduction reaction due to Cr(III)
in the bulk solution ...................... 26
12 The increase in pH with increasing initial Cr(VI) concentration. 27
13 The effect of pH on the reduction rate constant ......... 29
14 Cr(VI) adsorption density versus reaction time ......... 30
15 Adsorptive equilibrium constant as a function of pH and Cr(Vl)
concentration in the bulk phase ................. 31
16 The removal of total chromium at various pH values in solution . 35
vi
-------
Number Page
17 The removal of Cr(VI) at various stirring speeds; the
effect on adsorption 36
18 The effect of carbon bed size on: a. pH; b. residual Cr(VI);
c. Cr(III) produced; d. the Cr(VI) adsorption density 38
19 The effect of carbon bed size on: a. pH; b. residual Cr(VI);
c. Cr(lII) produced; d. the Cr(VI) adsorption density 39
20 The effluent characteristics of a 5 x 10~k M Na2CrOit (26 ppm
Cr(VI) ) wastewater after treatment with a prewashed packed
column 41
21 The effluent characteristics of a 1 x 10~3 M Na2CrO£t (52 ppm
Cr(VI) ) wastewater after treatment with a prewashed packed
column 41
22 A comparison of Cr(VI) removal capacities following various
regeneration techniques 43
23 A comparison of regeneration techniques after four adsorption
cycles and successive regeneration cycles 44
24 Measure of chromium remaining on the carbon surface by x-ray
emission, following various regeneration techniques 46
25 Cummulative percent desorption of chromium from the carbon
surface by caustic solution after successive repetitions .... 47
26 The concentration of Cr(VI) desorbed in caustic solution
as a function of time 48
27 Maximum Cr(VI) concentration in various caustic regenerant
solutions 49
28 The desorption of Cr(VI) from the carbon surface in acidic
solution 50
29 The rates of reduction and adsorption (mole/Jl-min) as
functions of pH, G and ir, for He] = 2 g/£, [HCrOlp = 1.0 x
10~5 M and [Cr(III)] = 0 52
30 The ratio of Rre(j to Rads as a function of pH 53
31 ir vs. pH over equal values of Rred/Rads ratios 54
32 The carbon dose required for 99% treatment of a 1 x 10~3 M
Na2CrC\ solution as a function of pH and Na2CrOit in the head
reactor 55
vii
-------
Number Page
33 Treatment costs in a single complete-mix contact vessel at
various pH's as a function of ir 59
34 Comparison of treatment costs for multiple reactors in series. . 60
35 Bed volumes to breakthrough as a function of carbon bed depth
for 1 x 10~3 M Na2CrOlf at pH 2.50 62
viii
-------
TABLES
Number Page
1 Physical Properties and Specifications of Activated
Carbon, Calgon Filtrasorb 400 .................. 13
2 Calculated Values for Power Input and Mean Velocity
Gradients in a 1 Liter Reaction Vessel and a Standard
1 Inch Stirring Bar ....................... 24
3 Reduction Reaction Velocities and Inhibition Constants
with Cr(III) in the Bulk Solution ................ 26
4 A Summary of the Half Saturation Constant, K,, as the pH in
Solution Increases ....................... 28
5 Equilibrium Constants for Formation of H2CrOtf in the Carbon
Micropores ........................... 32
6 Comparison of Calculated T& Versus Observed Te ......... 33
7 Evalution of k* at Known Initial Conditions for pH, CcH and
............................ 35
8 Average Loss of Carbon During Batch Adsorption and Regeneration. 45
9 Comparison of Wastewater Concentrations and Volume Reduction
with Strength of Caustic Regenerant Stream ........... 48
10 Reactor Size and Estimated Costs Associated with the Reactor
Volume and Applied Carbon Dose for a Single CMR ......... 58
11 Column Depth Versus Number of Bed Volumes to Cr(VI) Breakthrough 61
ix
-------
ACKNOWLEDGMENT
We would like to thank Ms. Mary Stinson, the U.S. EPA Project Officer,
for her assistance and suggestions on many occasions during the course of
this research project.
-------
SECTION 1
INTRODUCTION
Recent concern over the presence and effects of heavy metals in the
aquatic environment has prompted a significant national pollution control
effort. Much '.emphasis has bee'ii on. the re-evaluation of available removal
techniques and the development of new processes for heavy metal removal from
industrial and municipal wastewaters. Chromium has been a heavy metal of
particular concern, due to its widespread industrial use and its chemical
complexity. Chromium is a common pollutant found in a variety of industrial
wastewaters,including those from the textile, leather tanning, electroplating
and metal finishing industries. Chromium may be found in solution in various
chemical forms. Textile and tanning wastes may contain either the hexava-
lent Cr(VI), or the trivalent Cr(III), chromium species; while electropla-
ting and metal finishing wastes contain, primarily, the Cr(VI) species (1,2).
The toxicity and deleterious effects of chromium in natural waters are
well documented. High concentrations of Cr(VI) species are lethal to marine
organisms; while sub-lethal levels may retard fish reproduction and accumu-
late in fish tissues (3). The toxicity of Cr(III) is reportedly much less
than Cr(VI); however, this may be due to the decreased solubility of the
Cr(III) species compared to Cr(VI) (3,4).
Current practice in treatment of chromium laden wastewater has been
oriented toward large scale operations. However, in the case of small
plants, where economics prohibit extensive treatment, the wastewater often
is simply discharged into the municipal sewage system (5). Much recent atten-
tion has been focused on the accumulation of heavy metals, including chromi-
um, in municipal sludges. The extent of heavy metal accumulation in munici-
pal sludge has provided a strong ground for challenging land application of
municipal sludges on agricultural lands and thereby made sludge disposal an
increasingly difficult problem (6). It is quite obvious that pretreatment of
industrial wastewater before discharge to municipal sewers is necessary, if
not currently mandatory, to alleviate this problem.
Conventionally, the most common means of removing heavy metals from
wastewater is by chemical precipitation. In the case of hexavalent chromi-
um which constitutes many anionic species, the Cr(VI) species must first be
reduced to the trivalent form by reacting with sodium metabisulphite, sul-
phur dioxide, ferrous iron, or another common reducing agent at a pH of 2 to
3. The trivalent chromium may then be precipitated as the hydroxide, Cr(OH)3,
a-t optimumpH values of 8 to 9 (7). The precipitation process has not proven
effective, because hydroxides form light precipitates which are difficult to
-------
settle and dewater. Further difficulties arise when other heavy metals are
present in solution with chromium and are to be removed simultaneously by
co-precipitation. These difficulties result from the differences in the pH
of minimum solubility for various metals. For instance, zinc has a minimum
solubility at pH 9 to 10, nickel at pH 9.5 to 10.6, and cadmium has a mini-
mun solubility at pH ranges from 11 to 12 (8,9,10).
Ion exchange has received limited application to industrial wastewater
systems due to the destructive nature of most of these wastewaters on ex-
change resins, the poor selectivity for discrete ions in solution, poor
loading capacities, high cost of exchange resins and high operating costs
compared to other treatment systems (.9,11).
Other methods, such as reverse osmosis, ion-flotation, electrodeposi—
tion, and solvent extraction have shown some experimental promise but have
not fully demonstrated their effectiveness nor economic advantage for large
scale operations (5,7,11),
Recently, activated carbon has shown promise for removal of inorganic
heavy metals from wastewater. It is the intent of this investigation to
develop and model the operating parameters which can be applied to a full-
scale process for the removal of hexavalent chromium from a metal finishing
related wastewater.
-------
SECTION 2
CONCLUSIONS
Activated carbon is an effective adsorbent for removal of hexavalent
chromium species from wastewater.
The maximum Cr(VI) adsorptive capacity of the carbon occurs at pH 2.5
and decreases rapidly between pH 2.5 and 7.1, primarily due to the decreas-
ing electrostatic attraction between the positively charged carbon surface
and the anionic Cr(VI) species in solution. The Cr(VI) adsorptive capacity
decreases at pH < 2.5 due to the rapid reduction of the Cr(VI) species and
the subsequent dominance of the cationic Cr(III) species at low pH.
The individual rates of removal due to reduction and adsorption were
derived from laboratory experiments. The kinetic equations were:
2.4 [HCrO.-Xc] G [H+]
dt red 1.2 + 4.8 x 104 [Cr(III)] + 2.4 x 105 G CH+XCr(VI)]
which is the rate of Cr(VI) reduction and:
is the rate of Cr(VI) adsorption. The carbon dosage, Cc], is reported in
(g/&); [HCrO^H, [Cr(VI)], and CH+] are all reported in (M) and the velocity
gradient in the reaction vessel, G, is reported in (sec"1).
The rate equations indicate that the ratio of reduction to adsorption
of Cr(VI) species can be controlled by the pH and the mixing intensity in
the reaction vessel.
For optimum removal, the carbon should be pre-washed with a solution at
a pH approximating the pH of the wastewater to be treated.
Regeneration of the exhausted activated carbon is best accomplished by
a caustic stripping solution. This also makes possible the recovery of
Cr(VI) for reuse in a plating process.
Hexavalent chromium may be exclusively removed from solution through
adsorption in a completely-mixed reaction vessel. By controlling the pH and
-------
mixing in the vessel to inhibit the Cr(Vl) reduction reaction at the carbon
surface, the production of Cr(III) can be eliminated. However, the cost of
operating such a system is somewhat higher than operating a packed carbon
column.
Reduction of the Cr(Vl) species cannot be eliminated from a packed car-
bon column, but a packed column is the most economical and simplest treat-
ment scheme to operate. Therefore, if a separate carbon system can be
devised to remove the Cr(III) produced, packed carbon columns would be an
efficient, simple, economical, and environmentally compatible treatment
process for the removal of hexavalent chromium species from wastewater.
-------
SECTION 3
RECOMMENDATIONS
In this study, the parameters affecting the kinetics and extent of re-
moval from synthetic Cr(VI)-containing wastewater were thoroughly examined.
An actual wastewater will generally adhere to all of the kinetic equations,
theories and treatment trends established in this report.
The ionic strength, Cr(VI) concentration, heavy metal constituents and
various complexing agents found in an actual wastewater, may differ signifi-
cantly from the laboratory prepared wastewater. These differences in solu-
tion components may alter the rate constants and extent of Cr(VI) removal
for an actual wastewater. Therefore, a separate evaluation of each indi-
vidual case is necessary.
The following procedures are recommended for designing an actual full
scale activated carbon treatment process:
1. Obtain a typical sample of the wastewater to be treated (.ca 50 gallons).
2. Perform several batch laboratory experiments at various adjusted solu-
tion pH's (one-liter samples with 2 grams carbon at initial pH 2.0 to
6.0 are recommended).
3. Adjust the wastewater to the desired pH and run several laboratory scale
packed columns with different depths of carbon in each Cat least 3 dif-
ferent depths) until breakthrough. The carbon should be pre-washed with
clean water at the same pH as the adjusted wastewater. A flow rate of
2 gal/min-ft2 is recommended.
4. The Cr(VI)-free effluent from the column experiments should be saved and
batch experiments performed for removal of Cr(III), either by Cr(OH)3
precipitation or carbon adsorption.
5. Exhausted carbon should be regenerated in batch experiments with various
strengths of caustic solutioni to determine the percent concentration of
the wastewater per strength of the caustic stripping solution. At least
10 grams of exhausted carbon per 200 m£ of caustic solution should be
used.
The experimental data obtained from the procedures outlined above can
be evaluated and used in the graphs and equations presented in this report
to determine:
-------
a. the rate of reduction and adsorption of the Cr(VI) species;
b. the Cr(VI) adsorptive capacity of the activated carbon;
c. the amount of carbon needed for treatment;
d. the amount of Cr(III) which will be produced;
e. how often the carbon must be regenerated;
f. the volume of regenerant solution needed;
g. the cost of operating the system.
To improve this effort and establish a more complete set of criteria
for operating an activated carbon system to remove total chromium from solu-
tion, the following are recommended:
1. A full scale system should be established in a small plant and opera-
tional parameters applied.
2. Further work should be done to examine various means of inhibiting the
reduction of hexavalent chromium to trivalent chromium at the carbon
surface.
3. More experimental work must be done on the adsorption and regeneration
of an activated carbon process for the removal of trivalent chromium.
-------
SECTION 4
ACTIVATED CARBON - BACKGROUND INFORMATION
GENERAL
Activated carbon has been widely used for the removal of taste and odor
producing compounds and a variety of other organic pollutants, such as
phenol or laundry surfactants, from water and wastewater (12).
Recently, the removal of inorganic pollutants and heavy metals by acti-
vated carbon adsorption has received considerable attention. Much of this
work has been done in Japan, where heavy metals are a pressing concern.
Kawashima and others reported significant removal of heavy metals from syn-
thetic wastewaters using activated charcoal (13). Saito showed that the re-
moval of heavy metals such as copper, cadmium, and ferric iron could be im-
proved by treating activated carbon with sulfonate (14). In this country,
Huang and Ostovic found a variety of commercial activated carbons to effec-
tively adsorb cadmium, Cd(II), from dilute aqueous solution largely as a
result of charge potential on the carbon surface (15).
CHROMIUM(VI) AND CHROMIUM(III)
Use of activated carbon to remove chromium(VI) from water is a recent
endeavor. Toyokichi reported that chromates are effectively removed from
wastewater by passing wastewater containing chromates through a column pack-
ed with platinum black catalyst-impregnated activated carbon. One Kg of
activated carbon was mixed with 1 m£ of platinum black colloid, containing
0.001 mg Pt/& and 1 gm pure E^SOi^. The platinum black catalyst-impregnated
activated carbon (50 £) was packed in a column and then wastewater contain-
ing 100 ppm chromate was passed through the column at 1 m3/hr. The result-
ing wastewater contained less than 0.1 ppm chromate (16)« A similar research
work was conducted by Tagashira, et al., who found that by mixing 200 mi
K2Cr207 solution (534 ppm Cr) with 5 g powdered coconut shell charcoal (100-
200 mesh 15%, 200-325 mesh 15%, and < 325 mesh 70%) and heating in an auto-
clave at 200°C for 30 minutes, can reduce the Cr(VI) concentration to 0.01
ppro (17),
Huang and Wu studied the removal of chromium(VI) by calcinated charcoal
and found that removal was most significant at low pH and low initial Cr(VI)
concentrations (18). They also postulated that HCrOtj" ions are the major
species being removed, as shown by the following reactions:
-------
C 0 + HCrOit + H20 = C O^Cr + 2 OH
X X
or
C 02 + HCrOi^" + H20 = C 05Cr+ + 2 OH~
X X
Seto and Tsuda reported that by mixing a 50 mH Na2CrOit (10%) solution, with
initial pH being adjusted to < 3, with 5 g activated carbon in a flask for 2
hrs at 25°C, the Cr03 adsorption by the activated carbon was 39.7% and 3.3%,
respectively, for final pH values of 3 and 7 (19) .
Nagasaki, by heating 1-f.gnite with 14% HNC>3 acid for 13 minutes, demon-
strated that chromic acid ions were effectively removed (20), By passing a
wastewater containing chromic acid (100 ppm), with pH being adjusted to
equal to or lower than that of chromic acid, through an activated carbon
column for 100 hrs, Nagasaki and Terada reported that the effluent contained
neither Cr(VI) nor Cr(IIl) (21)0 After passing 1350 & of water the effluent
pH went up to 7 and contained 0.5 ppm Cr(VI). Ten liters of 25% HC1 solu-
tion was then passed through the column to regenerate the column by dis-
solving the reduced Cr(III). The column was reused for another 100 hrs
without breakthrough.
An EPA supported work, conducted by Landrigan and Hallowell, also
demonstrated that activated carbon can be used by many small plating plants
to relieve the chromium burden on municipal sewage systems (22)*
Very recently, Yoshida, ^t al., studied the adsorption of Cr(VI) and
Cr(IIl) onto activated carbon as a function of pH and the amount of total
Cr and Cr(VI) eluted from activated carbon at pH 4-6.5 (23), They reported
that Cr(VI) is readily adsorbed on activated carbon as anionic species, such
as HCrO^" and CrO^"2, while Cr+^ ion is scarcely adsorbed on activated car-
bon. They also observed that in acidic solution, Cr(VI) is easily reduced
to Cr(III) in the presence of activated carbon. The adsorbed Cr(VI) species
was elutable with NaOH (> 0.1 N) or with 1 N HC1 solution.
Huang and Wu studied the effect of pH on Cr(VI) and Cr(III) under acidic
conditions and in the presence of activated carbon. C24). Kim reported that
reducing reaction can be suppressed by adjusting the proton concentration to
become equal to that of the hexavalent chromate, or to maintain a Cr(VI)
system predominated by HCrOif~ species (25). Although a similar statement has
been made by Nagasaki, e£ _al.,(21) no such finding was observed in this
study.
-------
Effluent
vo
I.S I Pyrex Beaker
~ ^r
i I3-6'1"!
I.I cm I I
(a
T
Magnetic Stirrer
O
Speed Control
33cm
2.S4
cm
Glass Wool
— (b)
Influent
O
Influent
Reservoir
^
d
Magnetic
j
b
Stirrer
Peristaltic
Pump
Figure 1. Experimental Apparatus: a. batch reactor; b. packed column experiments.
-------
SECTION 5
EXPERIMENTAL PROCEDURE
BATCH MODE EXPERIMENTS
All experiments were conducted with batch reactors (1500 mH pyrex
beakers) containing one liter of Cr(VI) solution prepared from a stock
10-2 M Na2CrOit solution. The range of Cr(VI) concentrations was 5 x 10~5 M
to 5 x 10~3 M (2.6 to 260 ppm) and the ionic strength was 10"1 M NaCl. The
initial pH values were adjusted with 0.1 M HC1 or NaOH solution and recorded
with a pH-meter (Beckman, pHasar-I) equipped with a combination electrode
(Beckman, Cat. No. 37501). The beakers were well mixed by a magnetic stirrer
and the stirring speed was monitored with a strobotac (General Radio, Type
1531). At the end of predetermined intervals, 10 mH of samples was pipetted
from the reactor and centrifuged at 15,000 g for 10 minutes with a centri-
fuge (International Model HT) to separate the residual carbon particles from
solution prior to chromium analysis.
Torque on the stirring magnet to determine power input to the vessels
was monitored with a Motomatic-Motor-Generator controlled speed system
(Electro-Craft Corp. Model No. E-650-M).
PACKED COLUMN EXPERIMENTS
All experiments were conducted in stainless steel or plexiglass columns,
having an inside diameter of one inch. The amount of carbon used was deter-
mined by mass and was placed loosely in columns with glass wool to insure
even distribution of flow at both ends. All solutions were prepared the
same as in the batch mode. Variable speed peristaltic pumps (Cole-Parmer
Masterflex) were used to maintain a consistant flow rate throughout the ex-
periments and avoid contamination in the pump head. Samples were collected
in 100 m£ bottles at predetermined time intervals and covered prior to
chromium analysis.
The experimental apparatus configurations are shown in Figure 1 for
batch and packed column reactors.
Regeneration of Exhausted Carbon
Adsorption-Desorption Cycles—
The experimental parameters and procedure for each adsorption cycle
were: Cr(VI) = 127.8 mg/£, pH = 2.0, carbon dose = 10 g/£, adsorption
period = 1 hour, and total sample volume = 500 m£. At the end of the
10
-------
reaction period the treated solution was replaced by another 500 mSt of the
same chromium solution while the carbon particles remained in the bottle.
Two procedures were carried out for the study of the regeneration process.
In the first procedure, carbon particles were regenerated only after nine
consecutive adsorption cycles. In the second procedure, regenerations were
made three times after every four consecutive cycles of adsorption. No com-
pensation for the loss of carbon was made.
Caustic Regeneration—
In the caustic desorption method, chromium attached on the carbon sur-
face was desorbed with 250 m£ of NaOH with concentrations ranging from 0.25%
to 20%. The reaction period was 16 hours. After this desorption period,
the particles were rinsed twice (for five minutes each) with 100 mH dis-
tilled water.
Thermal Regeneration—
In the thermal activation method, the chromium-laden carbon particles
were first dried in an oven at 103°C for 24 hours. The dried particles were
then heated at an elevated temperature for reactivation. Application of a
muffle furnace at a maximum temperature of 550°C was first tried. The re-
sult showed only partial recovery of the adsorption capacity. However, by
employing a specially designed heating device at 950°C, a distinct improve-
ment in the recovery of the adsorption capacity was observed. This device
was finally used for the rest of the thermal regeneration study.
A steady flow of carbon dioxide (about 30 mJl/min) was passed through
the holding tube during the 30 minute heating period. The carbon particles
were cooled in air at room temperature (23 ± 1°C) for 24 hours.
Combined Caustic-Thermal Regeneration—
In the combined caustic-thermal process, the chromium-laden carbon was
first regenerated by caustic desorption (1% NaOH), and then the carbon par-
ticles were driei
mentioned above.
tides were dried at 103°C and thermal activated at 950°C in C02 gas as
Desorption Capacity—
Carbon was exhausted in a packed column, 10 grams carbon, by treating
approximately 16 liters of 1 x 10~3 M Cr(VI) or 52 ppm. The effluent was
saved and the amount of chromium on the carbon surface was determined by
analysis of the effluent. The carbon was then regenerated by a caustic or
acid solution to determine the maximum desorptive capacity of each solution
and the desorptive cycle was repeated if all of the Cr(VI) was not removed.
Chromium Analysis—
The residual Cr(VI) concentrations were determined by the purple (540
nm) s-diphenyl carbizide-chromate complex with a UV-visible spectrophoto-
meter (Hitachi-Perkin Elmer). The Cr(III) concentrations were determined
by, first oxidizing the samples with potassium permanganate, then determin-
ing the total chromium as Cr(VI) and obtaining the Cr(III) concentration by
subtracting the Cr(VI) from the total. Detailed analytical procedures were
described in Standard Methods,(26) except that boiling-oxidation time was
11
-------
extended from a recommended 3 minutes to 10 minutes with glass beads to
assure total oxidation. This modification was found necessary because there
were occasions when Cr(II) appeared in solution as results of further Cr(III)
reduction by activated carbon.
12
-------
SECTION 6
RESULTS AND DISCUSSION
ACTIVATED CARBON-SELECTION, PHYSICAL AND CHEMICAL PROPERTIES
The activated carbon selected for this study was Filtrasorb 400, pro-
vided by Calgon Co., Pittsburgh, Pennsylvania. The selection of this carbon
was based on the available physical and chemical data for the carbon, as
well as the potential of this particular carbon for chromium removal.
Filtrasorb 400 is a granular activated carbon having a relatively large
surface area and high resistance to abrasion. Table 1 lists the various
physical properties of the carbon, according to the manufacturer and our
laboratory.
Filtrasorb 400 is apparently an H-type activated carbon according to
the classification of Steenburg (27)» An H-type carbon is characteristic of
activated carbons produced under anoxic conditions at high temperatures,
greater than 900°C, and which in contact with water raise the pH of solution
through hydrolytic reactions and adsorb only acids (28). The effects of
Filtrasorb 400 on blank solutions containing 0.1 M NaCl for ionic strength
for both batch and packed column are exhibited in Figure 2 (a, b and c).
As shown, the pH of solution always increases upon contact with Filtra-
sorb 400. In the batch reaction vessels the magnitude of increase is propor-
tional to the carbon dose, while the pH increases to about 10 and gradually
decreases, approaching equilibrium at the influent pH after successive bed
volumes of the influent pass through a column.
The electrokinetic surface charge characteristics of Filtrasorb 400 in
comparison to a variety of other commercial carbons is shown in Figure 3 (15)s
The electrophoretic mobilities of the carbons are shown as functions of pH.
The pH at which zero electrophoretic mobility occurs is known as the pH of
zero point of charge, pHzpc (29). High pH is typical for H-type carbons.
The pHzpc for Filtrasorb 400 compared to other carbons tested may partial-
ly explain the increased affinity of Filtrasorb 400 toward anionic species
in solution, such as Cr(VI) (15).
Filtrasorb 400 has been proven effective for Cr(VI) removal by previous
investigators. Huang and Wu,(24) along with Kim and Zolteck,(25) obtained
good Cr(VI) removal efficiencies with Filtrasorb 400 activated carbon.
13
-------
TABLE 1. PHYSICAL PROPERTIES AND SPECIFICATIONS OF ACTIVATED CARBON,
CALGON FILTRASORB 400*
Physical Properties
Total surface area, m2/g
(N2, BET method)
Bulk density, lbs/ft3
Partical density wetted in
water, g/cc
Pore volume, cc/g
Effective size, mm
Uniformity coefficient
1050 - 1200
25
1.3 - 1.4
0.94
0.55 - 0.165
1.9 or less
(941')
Specifications
U.S. Standard Series
Sieve Size
Larger than No. 12
Smaller than No. 40
Mean particle diameter, mm
Iodine number, minimum
Abrasion number, minimum
Moisture
Ash content
Specifications
Max. 5%
Max. 5%
0.9 - 1.1
1050
75
Max. 2.0%
8%f
Typical
Analysis
3%
1%
1.0
1100
80
0.5%
w
Information was provided by the manufacturer unless otherwise indicated.
Determined in our laboratory.
14
-------
(a) Batch
(b) Batch
2.9 -
2.7
2.5
8.0
0 100 200 300
Time (minutes)
0 IOO 200 300
Time (minutes)
(c) Packed Column
IO.O
8.0
6.0
4.O
0 200 400 600 80O 1000
No. Of Bed Volumes
Figure 2. The effect of Filtrasorb 400 on solution pH:
a. in batch reactor with a given initial pH = 2.5 and
varying carbon dosages.
b. in batch reactor with a given carbon dose = 2 g/£
and varying initial pR.
c. in packed columns with varying initial pH in the
influent.
15
-------
o Filtrosorb 400
Nuchar 722
Darco 12 x40
Pac
Nuchar C-190'N
= O.OI M NaCI
UJ
-4
6
14
PH
Figure 3. The electrophoretic mobility of various
commercial activated carbons(after Huang
and Ostovic) (.15).
Cr(VI) Removal from Solution
The removal of CrCVI) from solution occurs through several steps of
interfacial reactions; 1) the direct adsorption of Cr(VI) onto the carbon
surface; 2) the reduction of CrCVI) species to Cr(III) by carbon on the sur-
face; and 3) adsorption of the Cr(lII) species produced, which occurs to a
much lesser extent than the adsorption of the Cr(VI) species. The rate of
each reaction depends on the following mechanisms: a) the transport of
CrCVI) anions, HCrOi*", by molecular or eddy diffusion, toward the carbon
surface; b) chemical reactions, reduction and/or adsorption, which take
place on the external carbon surface; c) desorption and back transport of
the Cr(VI) and Cr(.III) species from the external surface into the bulk
phase; d) inner transport of the Cr(VI) and Cr(III) species into the inter-
nal surfaces bounding the micropores and capillaries of the carbon; e) chem-
ical reactions, reduction and/or adsorption, taking place at the internal
16
-------
surfaces; and f) back transport of the Cr(VI) and Cr(III) species across the
internal surface and the external interface into the bulk phase. Figure 4
shows schematically the interactions of the most common Cr(VI) and Cr(III)
species with the external and internal carbon surfaces which are indicated
by the appropriate numbers and letters on the figure.
Bulk Phase
Cr
+ 3
HCr O
Carbon Surface
Figure 4. A simplified view of the chemical and physical
interactions of chromium with activated carbon
in dilute acidic medium.
Reduction Reaction at the Surface—
Hexavalent chromium species are powerful oxidizing agents. The reduc-
tion potentials for the various Cr(VI) species, in aqueous solution, com-
pared to other common oxidants, are shown in Figure 5, calculated from
Latimer (30).
It should be noted that HCrO^~ is the most predominant species for
dilute acidic solutions, (Appendix A) and is not only a stronger oxidizing
agent, but also has a much lower activation energy than the C^Oy"2 species,
which is a slow acting oxidant even in 1 M H+ solution (30). It is the HCrO~
species which is the most active and takes part in the surface reduction
reaction.
17
-------
H202.
Cl
C (Graphite)
-Cr+l
•2.0
.HUO
MnO,
c
H,
••1.0
NH+
H20
co
•-0.0
c
9
O
a.
•o
0)
(T
-1.0
Figure 5. Reduction potentials of several common
oxidizing agents (after Latimer) (30).
Surface Reaction Kinetics—
Normally, reactions which occur on surfaces, such as granular carbon
particles, do not follow simple reaction order kinetics, due to the finite
number of surface sites available for the reaction to occur. Therefore,
the reduction rate, —-jr or v ,, will reach a maximum reaction velocity,
Vjnax, as the concentration of the reactant, Cr(VI)(mole-A"1), increases or
the reaction rate will become zero order as the concentration of Cr(VI) be-
comes sufficiently large (.31,32). The expected results for this type of re-
action are shown in Figure 6, where the ordinate is quantity vre(j/vmax and
the abiscissa represents the increasing reactant or Cr(VI) concentration as
HCrOjj".
18
-------
i.o -
E
>
0.5
[HCrO~] (M)
Figure 6. Ratio of observed reduction rate to the maximum rate
as a function of HCrOij" Cr(VI) concentration.
The quantity Kg shown in Figure 6 represents the concentration of Cr(VI)
at which vre(j is equal to one half of Vmax- This reaction is similar to
first order Michaelis-Menten enzyme reactions, where the reaction rate is
limited by the number of active reaction sites available or the carbon dose,
C, in this case.
It was noted that both adsorption and reduction reactions occurred
simultaneously. The reduction reaction may be obtained by considering the
following stoichiometry of the reaction:
HCrOij + 7 H + 3e f- Cr"™ + 4 H20
.+3
(1)
This implies that the rate of Cr(VI) removal due to reduction must also equal
the rate of production of Cr(III). Therefore the rate of reduction may be
written as:
= dCr(III) =
7red dt
dCr(VI)
dt red
+ [Cr(VI)]
(2)
where v , = rate of Cr(VI) reduction (mole-£~1-min~1)
v = maximum rate of Cr(VI) reduction per unit mass of
carbon (mole-g~1-min~1)
DlCrOi,."] = concentration of bichromate in solution (mole-5,"1)
19
-------
= mass of carbon applied per unit volume (g-A"1)
CCrVlD = concentration of total hexavalent chromium in solution
(mole-JT1)
K- = Cr(VI) concentration at which the rate of reduction equals
one half of v (mole-Jl'1)
max
Figure 7 shows typical batch reaction data for the removal of Cr(VI)
(Figure 7a) and production of Cr(III) (Figure 7b). The rate of reduction at
any time may be determined by the instantaneous slope of a line struck tan-
gent to the Cr(III) production curve. By taking the slope at t =_0, the
initial conditions are all known. Therefore, knowing that CHCrO^ H = MI x
HCr(VI)H,where ai is the fraction of the total Cr(VI) species in the HCr04~
form and when Cr(VI) < 1 x 10~3 M, then a! - 1.0 (Appendix A), Equation (2)
can be inverted to yield:
1 K6 1+1
Vred VmaxM CCr(VI): v^M (3a)
Therefore, plotting 1 * initial vred vs 1 ^ initial CCr(VI)H results in a
linear plot having an intercept = 1 * {vmax Ed]} and a slope = Kg
as shown in Figure 8. Since the initial conditions are known, pH = 2.50,
CcH = 2 g/H, and a constant mixing intensity, the reduction rate equation
may now be evaluated from the results in Figure 7 and 8 as:
dCr(IIl) = rdCr(V!K = 1.0 x 10"5
"
2.4 x 10-* + CCrCVI)]
The rate of the reduction reaction is also influenced by the rate of
mass transport or mixing intensity in the system, the pH in the bulk solu
tion, and the amount of Cr(III) which has been produced and is present in
the bulk solution.
20
-------
2 g/l F-400
2.50
10 15 20
Time (minutes)
(a)
25
30
10 15 2O
Time (minutes)
25
30
Figure 7. a) Batch removal of Cr(VIl and
b) production of Cr(III).
21
-------
Figure 8. Inverse plot of initial reduction rate and Cr(VI)
concentration (data were taken from Figure 7).
Effect of Mixing Intensity on Cr(Vl) Reduction
The rate of mass transport, or mixing in the reaction vessel, effects
the rate of Cr(VI) reduction and therefore becomes the rate limiting step at
low levels of mixing in the system. Figure 9 shows the removal of Cr(VI) and
the corresponding production of Cr(III) at increasing mixing intensities,
i.e., angular velocity of the magnetic stirring bar, a)(sec-I).
The mass transport in a mixed vessel is a function of the velocity grad-
ient in the system, which can be determined by the power input to the vessel
through the stirring bar or paddle. By measuring the torque on the stirring
motor at known values of u, the power input to the reaction vessel can be
determined by:
P = T x GJ
where P = power input to the system (in-lb/sec)
T = torque applied to the mixing shaft (in-lb)
u = angular velocity of the mixing shaft (sec"1)
(4)
22
-------
200 rpm
• 500 rpm
O lOOOrpm
O 2000rpm
10 15 2O
Time (minutes)
30
Figure 9. The removal of Cr(VI) and production of Cr(III)
at different mixing intensities.
and the mean velocity gradient, G (sec~*) under turbulent conditions may then
be estimated by the Saffman-Turner equation:
G - ^ * «>
where V = the volume of the reaction vessel (in3)
y. = the dynamic viscosity of the bulk solution (lb-sec/in2)
A summary of these results for various mixing speeds and 1-liter of 0.1 M
NaCl solution are shown in Table 2.
23
-------
TABLE 2. CALCULATED VALUES FOR POWER INPUT AND MEAN VELOCITY GRADIENTS IN A
1 LITER REACTION VESSEL AND A STANDARD 1 INCH STIRRING BAR
0)
(sec"1 x 10"1)
0.17
0.33
0.83
1.67
2.50
3.33
T measured
(in-lb)
0.01
0.01
0.02
0.03
0.04
0.05
P
(in-lb/sec x 10~2)
0.06
0.12
0.60
1.80
3.20
6.00
G
(sec'1)
1.69
2.39
5.35
9.27
13.10
16.90
The mass transfer rate in solution will effect the half-velocity con-
stant, K,, in equation 3, while vmax will remain unchanged. Therefore, Kg
may be calculated from the initial rates of Cr(III) production in Figure 8,
and the effect of G on Kg under these conditions is shown in Figure 10. The
slope of the log-log plot for Kg versus G indicates that Kg is a function of
G, or
K . 2.1 x 10-3 (6)
0 tr
Effect of Cr(III) on Reduction,
The production of Cr(III) at the carbon surface results in an accumula-
tion of Cr(III) in the bulk solution and a decline in the Cr(III) concentra-
tion gradient between the surface and bulk phases. This results in a decrea-
sed driving force to remove the Cr(III) from the surface and subsequently the
surface sites are tied up longer by the Cr(III) produced. Therefore, in-
creasing Cr(lII) concentration in the bulk solution will competitively inhib-
it the reduction reaction and the rate of reduction will be decreased or:
dCr(HI) fdCr(VI). ^ K3
dt ^ dt "red ~ K3K6 + K6L~Cr(IIl)J + K3LCr(VI)J v
where K3 = the constant of trivalent inhibition (mole-Jl"1)
and CCr(lII)H = the concentration of Cr(III) in the bulk phase imole-JZT1) .
The rate of reaction at any Cr(III) concentration may be taken from the slope
of the line tangent to that point on the Cr(III) production curve. Therefore,
by striking tangents to points of equal Cr(IIl) concentrations, 5 x 10~5 and
2.5 x 10~5 M in Figure 7b, and knowing the Cr(VI) concentration in the bulk
solution at that time, from Figure 7a, the inverse of equation 7 may be plot-
ted as shown in Figure 11, where all of the lines intersect the ordinate at
1/v CcH and the slope of the line is:
max
24
-------
1.5
•
X
2
0.5-
0.2 0.6 1.0 1.4
10
G (sec1)
20
Figure 10. The effect of velocity gradient on the
reduction rate constant (K,).
K3
max
(8)
where K6, vmax and [c] are known values from Figure 8 (2.41 x 10~tf M,
1.0 x 10~5 moles-min-1-g~1, and 2.0 g/Jl) and CCr(III)H was constant at
5 x 10~5 and 2.5 x 10~5 M (2.6 and 1.3 ppm). K3 can be evaluated from
equation 8 and the slope of the lines in Figure 11. The values of K3 are
listed in Tahle 3. The average value of K3 was 2.4 M x 10~5 M which was used
for all subsequent calculations.
25
-------
TABLE 3. REDUCTION REACTION VELOCITIES AND INHIBITION CONSTANTS WITH Cr(III)
IN THE BULK SOLUTION
(M x 105)
Slope of v~l vs CCr(VI)]"1
(min x IO-1)
(M x 105)
0
2.5
5.0
1.21
1.37
2.52
* Calculated from equation 8.
2.49
2.37
K3 = 2.43 x 10~5
X
c
-"S 2
• Initial Condition* - No Cr (HI)
a 2.5 xlO"S M Cr(m)
O 5.0 xlO~5 M Cr(in)
I 2
[CrtVnf (M x IO"4)
•Figure 11. Inhibition of the surface reduction reaction
due to Cr(III) in the bulk solution.
Effect of pH on Reduction
The reduction of Cr(VI) to Cr(III) involves the consumption of H ions
from solution, as shown by equation 1, for the reduction half reaction. A
proposed overall reaction for Cr(VI) reduction at the carbon surface is:
+ 3 C (amorphous) + 16 H+ £ 4 Cr+3 + 3 C02 + 10 H20
C9)
26
-------
which implies the consumption of H+ ions, per Or (III) ions produced, occurs at
a 4:1 ratio. Therefore, the H+ concentration will limit the extent of Cr(IIl)
poroduce'd and should strongly effect the rate of the reduction reaction.
2.8
2.7
x
Q.
2.6
2.5
pH = 2.5
2 g/l F-400
A Distilled Water O.I M NaCI
• 5 x I0~5 M Na2Cr04
I X
5 x
10
10"
M
M
10 20
Time (minutes)
30
Figure 12. The increase in pH with increasing initial
Cr(VI) concentration.
The change in pH of solution is shown in Figure 12 for increasing initial
concentrations of Cr(VI), i.e., increasing magnitudes of Cr(lII) produced,
and compared to the change of pH for the carbon in solution without the
presence of Cr(VI). As expected, the rate and magnitude of H+ consumption
increases proportionally with the amount of Cr(III) production.
Decreasing the H"*" concentration in solution effects the rate of reduc-
tion by increasing the concentration of Cr(VI) needed to reach the maximum
reaction velocity, vmax. Therefore, the half velocity constant, Kg, will
increase with decreasing IT*" concentration. The actual reaction rates at
various pH's were determined by the instantaneous slope of the Cr(lII) pro-
duction curve shown previously in Figure 6b, at 5 x 10-lf M initial Cr(VI) con-
centration, 2 g/A carbon dose, and having an initial pH of 2.5. The pH cor-
responding to each point in time was read from Figure 12, and the rate was
corrected for the instantaneous concentration of Cr(III) in the bulk solution.
27
-------
Therefore, Kg may be expressed quantitatively by:
red max
and the values of K, are summarized in Table 4.
TABLE 4. A SUMMARY OF THE HALF SATURATION CONSTANT, K,, AS THE pH IN
SOLUTION INCREASES.
Time
(min)
0
5
10
15
20
30
CCr(VI):
(M x 101*)
5.0
4.24
3.41
2.84
2.48
1.79
[Cr(III):
(M x 104)
0
0..61
0.85
1.03
1.16
1.29
pH vred
(M-min~ x 106)
2.50
2.62
2.67
2.71
2.73
2.76
13.3
6.80
4.15
2.93
2.12
1.30
V
(M x 104)
2.41
3.20
3.66
3.87
4.34
4.81
*
Calculated from equation 10.
The results of K, versus pH are shown in Figure 13. From the figure,
the reduction rate constant, Kg, exhibits an approximate 1st order dependence
in solution. Therefore, K., is written as:
o
K, * 4.8 x 10~6 G CH+] (11)
D
and the overall reduction rate equation becomes:
,dCr(VI). _ 2.4 CHCrO.-][c] G
'*• dt ' . ~ 1.2 + 4 x 104 [Cr(III)D + 2.4 x 105 G Cff*l[Cr (VI) H
red
ADSORPTION OF Cr(VI)
Equilibrium Adsorption Capacity
The adsorptive capacity of the carbon for Cr(VI) depends upon the ionic
strength of the bulk solution (0.1 M NaCl for all experiments performed in
this study), the pH of the bulk solution, and the concentration of Cr(VI) in
the bulk solution at equilibrium.
Batch experiments were run to determine the equilibrium adsorptive capa-
city, T eq. (ym/g), as a function of bulk pH and Cr(VI) concentration. Typi-
28
-------
•g
x
5
-3.3 -
2.5
2.6
2.7
pH
figure 13. The effect of pH on the reduction rate constant.
cal batch results are shown in Figure 14 for an initial Cr(VI) concentration
of 5 x 10~3 M and increasing initial pH values. Equilibrium is observed to
occur at the end of 5 hours contact time. Initial and final pH's are indica-
ted on the graph and little reduction of Cr(VI) was observed to occur due to
elevated pH's observed (greater than 3.5 after 1 hour).
>
Adsorption of the Cr(VI) species due to the condensation of I^CrO^ within
the carbon micropores, and the adsorption of each HCrO^ molecule consumes one
H+ from solution. An adsorption equation can then be written as an equili-
brium between H^CrO^, which accumulates in the porous phase, D^CrO^] p, the
bulk proton concentration, Dr*~Ilb, and Cr(VI) in the bulk phase, CHCrO^"!^, or:
H + HCrOif £ H2CrOt(;
K
(13a)
where
K =
e
Cl3b)
29
-------
1000
E
o>
•N.
E
500 -
Figure 14. Cr(VI) adsorption density versus reaction time.
Values of Ke have been calculated from Figure 14 and the results are
shown in Table 5.
The results of Table 5 show that Ke is not a constant, but is itself de-
pendent upon pH and Cr(VI) concentration in the bulk phase. This result is
expected, due to the effect of pH on the surface charge of the carbon which
acts as an electrostatic driving force for adsorption of Cr(VI) species and
due to the effect of the Cr(VI) concentration gradient as a driving force,
along with the capillary forces involved in the carbon micropores.
An empirical equation for Ke may be derived by assuming Ke will have the
same order dependence on pH regardless of Cr(VI) concentration, or that a plot
of log Ke against pH will result in a straight line with an intercept, b,
which will decrease as the concentration of Cr(VI) is decreased. Figure 15
shows the results of such an analysis of the data from Table 5, The adsorp-
tion constant Ke can be expressed as:
log K = 0.8 pH + b
(14a)
where
30
-------
-1
log b = ~ log [Cr(VI)2b + 0.41
(14b)
10
*
o
-E.3
-2.5
6
PH
Figure 15. Adsorptive equilibrium constant as a function
of pH and Cr(VI) concentration in the bulk phase.
It should be noted, however, that these equations are strictly empirical and
have no theoretical basis for their prediction. Equations I4a and 14b are
accurate for a pH range of 2.5 to 7.0, since higher values would be above the
oH and would not be expected to conform. The lower pH values would in-
r zpc
31
-------
TABLE 5. EQUILIBRIUM CONSTANTS FOR FORMATION OF H2Cr04 IN THE CARBON
MICROPORES.
iii §
(g/A) (ym/g) (M) (M x 103) (M'1 x 10~8) log KS
2.5
3.5
4.5
5.5
6.5
4.85
5.94
6.09
6.32
7.15
2.0
2.0
2.0
2.0
2.0
870
590
560
495
180
0.93
0.63
0.60
0.53
0.27
3.02
3.62
3.56
3.75
4.28
0.22
1.51
2.06
2.96
8.78
7.34
8.18
8.31
8.47
8.94
Initial pH in bulk solution.
pH in bulk solution at equilibrium.
Calculat
Table 1.
' Calculated from Te using an internal pore volume of 0.94 ml/g from
§
Calculated from equation 13b.
crease the acidity inside the micropores to the extent that reduction would
occur faster than H2CrOtt condensation and all of the Cr(VI) would be reduced
to Cr(III) and released back into solution.
A comparison of observed capacities and calculated capacities are shown
in Table 6. The results show good agreement between actual and predicted
values. The calculated values are shown to be slightly conservative, which
would be desirable for actual system design.
Kinetics of Cr(VI) Adsorption
The transfer of Cr(VI) from the bulk solution into the surface pores of
the activated carbon is driven by the charge forces between the carbon parti-
cles and the Cr(VI) species, the capillary forces which exist between the car-
bon micropores and the bulk solution, and the concentration gradient which
exists across the surface-bulk interface. According to Weber and Morris (33),
the rate of steady state transfer across this interface, (dCr(VI)/dt)a
-------
TABLE 6. COMPARISON OF CALCULATED T VERSUS OBSERVED T
e e
u>
w
[crVlD
(M) °
(x 103)
< 5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
1.0
1,0
1.0
CcrVl]
(M),
(x 103)
3.02
3.62
3.56
3.75
2.17
2.98
3.08
3.29
1.92
2.77
U96
1.0*
1.0*
1.0*
PH0
2.50
3.50
4.50
5.50
2.50
3.50
4.50
5.50
2.50
3.50
2.50
2.50
2.75
3.00
*.
4.85
5.94
6.09
6.32
5.83
6.50
6.54
6.78
6.45
7.00
6.83
2.61
2.86
3.27
[c]
g/t
2.0
2.0
2.0
2.0
4.0
4.0
4.0
4.0
6.0
6.0
8.0
observed
(ym/g)
870
590
560
495
688
495
475
363
508
355
378
1280
1130
1070
predicted
(ym/g)
830
560
518
481
632
385
385
359
302
292
257
1267
1129
935
Error
(ym/g)
-40
-30
-42
-14
-55
-110
-90
-4
-206
-63
.-120
-13
-1
-135
% Error
-4.8
-5.4
-8.1
-2.9
-8.1
-28.6
-18.9
-1.1
-40.6
-17.7
-31.7
-1.0
-0.0
-12.6
Data obtained from continuous flow packed column experiments run to equilibrium.
-------
C = mass of carbon (g-£-1)
A Cr(VI) = concentration difference across the surface bulk interface (M)
For the Cr(VI)-Filtrasorb system, the carbon inner-surface has an ex-
tremely high affinity for Cr(VI) and a concentration gradient will exist in
favor of the bulk phase, and the driving force due to the concentration gra-
dient will be:
A Cr(VI) = (Cr(VI), - ir Cr(VI), } (16a)
D D
where T is a partition factor between the surface and bulk phase Cr(VI) con-
centrations , or :
where T = amount of Cr(Vl) adsorbed into the carbon (y mole/g) ym/g
T = amount of Cr(VI) adsorbed into the carbon at equilibrium with the
bulk concentration (ym/g)
From equation 15, the transfer coefficient, k, and the effective trans-
fer per unit mass of carbon, ac, may be lumped into a single constant,
k*(Jl-g~1-min~1) , which will depend upon the pH of solution due to the change
in the surface charge density and the subsequent effect on the electro-
kinetic driving force of Cr(Vl) adsorption. Figure 16 shows the residual
C*total» which ^s equal to the amount of total chromium remaining, Cr(Vl) -t-
Cr(III), as a result of Cr(VI) adsorption and is exclusive of Cr(VI) removal
by reduction at constant initial conditions, except for increasing pH values.
The slopes of lines tangent to the initial points yield the initial adsorp-
tion rate, {dCr(VI)a(|s/dt}o, when T/re = 0, which implies that:
(dCrJ(Vl)) = k* \J^T CCr(VI): CGI] (17)
dt ads
and the insert in Figure 16 shows the dependence of k* on the one-fifth power
of DT*"]]. A summary of the initial conditions in Figure 16 and the resulting
values of k* are calculated in Table 7.
34
-------
TABLE 7. EVALUATION OF k* AT KNOWN INITIAL CONDITIONS FOR pH, [C] AND
CCr(VI):.
CCr(VI):
(M x 103)°
5.0
5.0
5.0
5.0
Cc]
(g/A)
2.0
2.0
2.0
2.0
PH0
2.5
4.5
5.5
6.5
,dCr(VlK
*• dt ;ads
(M/min x 105)
9.52
3.67
2.35
1.00
CH : k* calculated
(x 10)
3.16
1.26
0.79
0.50
U-g-l-min-1 x
3.01
2.91
2.97
2^00+
k* = 2.93
102)
Excluded from average calculation due to the closeness of pH
zpc
2 5
X
o
'o
-4
1/5
O pH • 6.5
A pH »5.5
x PH'4.5
• pH-2.5
20 40
Time (minutes)
60
Figure 16. The removal of total chromium at various pH values
in solution.
35
-------
Therefore, the rate of Cr(VI) from solution may be written as:
CdCr(VI) = - 2 9 x ID'2
L dt 'ads />y x iU
CCr(VI)] {1 - -}
(18)
where T^ may be calculated using equations 13 (b) and 14 (a and b).
The Effect of Mixing on the Adsorption of Cr(VI)
The reduction reaction at the carbon surface is controlled by the rate of
mixing or velocity gradient in the reaction vessel. However, the adsorption
of Cr(VI) species onto the carbon is controlled by the diffusion of the,
species into the micropores on the surface. Therefore, the mixing in solu-
tion would not be expected to influence the adsorption reaction as long as the
rate of Cr(VI) transport to the surface is larger than the adsorption rate.
Figure 17 showns the removal of Cr(VI) from solution at various stirring
speeds.
O 500 rpm
A 750 rpm
D1000 rpm
• 1500 rpm
2 4
Reaction Time (hr)
Figure 17. The removal of Cr(VI) at various stirring speeds;
the effect on adsorption.
36
-------
By using a much larger Cr(VI) concentration in these experiments, com-
pared to the reduction experiments, 5 x 10~3 M vs 5 x 10~5 M, the reduction
reaction velocity should be close to the maximum and mixing should effect ad-
sorption rather than reduction. Therefore, as expected, Figure 17 shows
little effect on Cr(VI) removal as long as the stirring speed in the reaction
vessel is greater than 500 rpm (G = 5.4 sec"1), or G can be neglected in the
rate equation for adsorption in a well-mixed vessel.
Packed Column Experiments
Loosely packed carbon columns were run to determine the importance of
the various operational parameters in maintaining an efficient and effective
system for removal of Cr(VI) and to minimize the amount of Cr(III) produced.
The duration time of these experiments was reported in number of bed
volumes of influent passed through the column, where a single bed volume,
vjj (£), was the volume of the voids in the column. For Filtrasorb 400 packed
loosely in one inch diameter columns, one bed volume was defined as:
Vb = {1.26 x ID"3 J>/g} x MC (19)
where M = mass of carbon in the column (g).
c
The effects of carbon bed depth, influent Cr(VI) concentration, and pH
on the removal efficiency were studied in these experiments. Pre-washing of
the carbon before contact with Cr(VI) was also investigated.
The influence of bed size on the removal efficiency is indicated in
Figure 18 (a through d) for 10, 30, and 50 gram carbon beds, all receiving a
constant influent of 2 gal/min/ft2 or 44 m£/min, at pH 2.50, 10~3 M Na2CrOi+
and 0.1 M NaCl. The inability of the carbon to remove all of the Cr(VI) over
the first few bed volumes (Figure 18a) is due to the high initial pH observed,
primarily due to the amount of H~*" ions needed to hydrolyze the carbon surface,
as shown previously in Figure 2. Since there is no Cr(VI) present after 100
bed volumes for 30 or 50 gram carbon beds, the equality of the Cr(III) pro-
duced by the 30 and 50 gram carbon beds (Figure 18b) indicates that adsorp-
tion of the trivalent species does not occur and may be neglected.
The influent Cr(VI) concentration was varied from 1 x 10"^ M to 5 x 10~3
M Na2CrOif, 5.2 to 260 ppm as Cr, while the carbon bed size, influent pH, and
flow rate remained constant, 50 g, 2.50, and 2 gal/min/ft2, respectively.
The results of these experiments are shown in Figure 19 (a through d). Fig-
ure 19 (a) shows a retardant effect of increased Cr(VI) concentration on the
time required for the pH of the column to equilibrate as the surface hydroly-
sis reactions go to completion. This is due to the increased demand imposed
on the H concentration as the influent Cr(VI) concentration is increased and
more Cr(III) is subsequently produced (Figure 19c) and more Cr(VI) is also
adsorbed (Figure 19d).
The removal of Cr(VI) is incomplete for the initial bed volumes, (Figure
19 b) due to the high pH which is maintained for the first 100 bed volumes as
a result of the H+ demand for the hydrolysis reactions. The complete removal
37
-------
• BlQnk -0- Cr(ZJ)
o lOg F-400
10
(o)
200 400 600
No. Bed Volumes
• 30g F-400
o 50g F-400
200 400 600
No. Bed Volumes
soo
400
I 300
(-, 200
100
(d)
200 400 600
No. Bed Volumes
200 400 600
No. Bed Volumes
Figure 18. The effect of Carbon Bed size on:
a. pH
b. residual Cr(VI)
c. Cr(III) produced
d. the Cr(VI) adsorption density.
38
-------
• 0-Cr(S3)
° 5 » IO"3 M
• 5 x I0"4 M
* I . IO"4 M
1.4
0.2
100 200 300
No. Bed Volumes
100 ZOO 300
No. Bed Volumes
•o
o
b
a
(c)
800
600
400
200
100 200 300
No. Bed Volume*
100 200 300
No. Bed Volumes
Figure 19. The effect of carbon bed size on:
a. pH
b. residual Cr(VI)
c. Cr(III) produced
d. the Cr(VI) adsorption density.
39
-------
of 5 x 10~3 M Cr(VI) was never achieved during the entire experiment since
the pH remained high (greater than 4 over the entire 300 bed volumes). The
H+ concentration was insufficient to obtain complete removal. Also, the re-
sults of Figure 19 (c) show that Cr(III) will still be produced when the in-
fluent Cr(VI) to H+ ratio is greater than one (1.58 for 5 x 10~3 M Cr(VI) to
pH 2.50). The 1:1 ratio is generally valid when the Cr(VI) concentration is
less than 1 x 10"^ M, or a 1:1 ratio implies the pH is greater than 4.0, which
is not a strong reducing condition regardless of the Cr(VI) concentration.
Equation 12 shows the reduction rate to depend on the first power of the H"*"
concentration. This finding did not agree with what was reported by Kim and
Zoltek, who claimed a 1:1 total Cr(VI) to H+ for minimum reduction and maxi-
mum adsorption reactions.
To eliminate the initial Or(VI) removal deficiency and smooth out the
pH in the system, the carbon must be prewashed with an acidic solution before
contact with Cr(Vl). It is not recommended that a strong acid solution be
used for this wash cycle, since strong acids are corrosive to the carbon and
may result in a degree of hydrolysis which overshoots the equilibrium that can
be obtained by the H concentration in the subsequent wastewater influent.
Therefore, the carbon must be hydrolyzed by a wash solution which closely
approximates the pH of the influent to be treated. In a column configuration,
the wash cycle may be accomplished in approximately 150 bed volumes or less,
by a pH of 2.50 or less, as shown previously in Figure 1 (c), while at pH
3.00 or greater, the wash cycle requires too much time and becomes a cumber-
some operation. In this case, it may be more convenient to eliminate the
concentration gradients which occur in a packed column and hydrolyze the car-
bon granules by titrating them with acid in a well-mixed reactor until an
equilibrium pH is obtained, before placing the carbon into the column.
The results of prewashing with 150 bed volumes of pH 2.50, 0.1 M NaCl
washwater before contacting the carbon with Cr(VI) are shown in Figures 20 and
21 for 50 grams of carbon, 5 x lO"4 M and 1 x 10~3 M Na2CrOtt, respectively,
in the influent. The pH is observed to be much smoother over the course of
operation and no Cr(VI) was detected in the effluent for 600 bed volumes.
The production of Cr(III) was also consistent and increased linearly as the
adsorption equilibrium and the progressive increase in Cr(VI) concentration
propagated up the column.
40
-------
_ 4
o
o
o
o
O
5 x id"4 M No2Cr04 ; O.I M NoCI
SO grams Carbon - Prewashed- pH« 2.5
x pH o Cr(EH)
• Cr(3I) O Cr- adsorption
density
5 3
3 pH
250
200
ISO £
100
50
200 400
No. Bed Volumes
600
Figure 20. The effluent characteristics of a 5 x ICT4 M Na2Cr04
(26 ppm Cr(VI) ) wastewater after treatment with a
prewashed packed column.
1.0
10
'o
x
S
o
c
0.5
Ixl0~3 M No2Cr04; O.I M NoCI
50 grams Carbon- Prewashed-pH r 2.5
x pH O Cr(m)
CrfiZI) O Cr - adsorption
density
-x-
3 PH -
6OO £
400
ZOO
2OO 400
No. Bed Volumes
600
Figure 21. The .effluent characteristics of a 1 x 10~3 M Na2CrOit
(52 ppm Cr(VI) ) wastewater after treatment with a
prewashed packed column.
41
-------
Regeneration of Exhausted Activated Carbon
If a full scale system for the removal of Cr(VI) by activated carbon is
to be practical or economically feasible, a sound regeneration procedure must
be established. Regeneration must perform several functions and meet the
following criteria:
1. The removal capacity must not be destroyed by the regeneration process.
2. Chromium must be removed from the carbon surface and concentrated in such
a way that the resulting regenerant stream is much more concentrated than
the original wastewater stream which was treated.
3. The carbon loss during regeneration must be minimal.
4. The process must be economically feasible.
5. The process should yield no secondary contaminants.
Four methods of regenerating the chromium-laden activated carbon were
examined:
1. Thermal regeneration (drying at 103°C in air, 550°C in air or 950°C in
C02).
2. Caustic regeneration (with NaOH).
3. Combined caustic-thermal regeneration (caustic and 950°C in C02)•
4. Acid regeneration.
Restoration of Removal Capacity
In the first set of experiments, the carbon was exhaused by subjecting
it to nine successive batch adsorption cycles, as described in Section 5,
after which the carbon was regenerated and the adsorption cycles repeated.
Figure 22 shows the results for the removal of Cr(VI) after regeneration by
treating with 1% NaOH, heating in air at 103°C for 24 hours or 550°C for one
half hour and heating at 950°C in C02 for one half hour, compared with the
virgin carbon and a virgin carbon pre-heated at 950°C in C02 for one half
hour. Thermal regeneration at 950°C, or the combined caustic-thermal techni-
ques, show the same removal capacity and appear to be the most effective,
while 1% NaOH is next best and drying in air at 103°C or 550°C appear to
create little readsorption capacity. A comparison of the caustic, thermal
(950°C in C02) and combined caustic-thermal regeneration techniques over sev-
eral regeneration cycles is shown in Figure 23. Here the combined caustic-
thermal technique acquires an advantage over the thermal regeneration alone,
and the thermal regeneration approaches the same removal capacity as caustic
regeneration as subsequent regeneration cycles are performed.
42
-------
100
Pre-heated virgin
Carbon
4 6
Adsorption Cycles
Figure 22. A comparison of Cr(VI) removal capacities following
various regeneration techniques.
43
-------
100
Comblntd
Caustic
Tharmal
Proem
o
<^
o
1234 1234 1234
Adsorption Cycles
Figure 23. A comparison of regeneration techniques after four
adsorption cycles and successive regeneration cycles.
Loss of Carbon
The loss of original carbon due to various physical chemical reactions
is shown in Table 8. The average loss of 0.17% carbon by weight during the
first adsorption cycle, and another 0,.06% loss per additional adsorption
cycle, was due mostly to carbon ash. A 1.5% loss of carbon was found when
it was treated with 1% NaOH solution. Heating the used carbon in air at
550°C caused the greatest loss of carbon, apparently due to combustion.
Thermal activation at 950°C in a CQ% atmosphere gave a 5% loss of carbon.
44
-------
TABLE 8. AVERAGE LOSS OF CARBON DURING BATCH ADSORPTION AND REGENERATION
Type of Operation Average Loss of
Carbon per Cycle, % by Weight
First Adsorption Cycle 0.17
Further adsorption cycles 0.06
Caustic regeneration cycle (1% NaOH) 1.50
Thermal regeneration cycle at 7.40
550°C in air
Thermal regeneration cycle at 5.00
950°C in C02
Disadvantage of Thermal Regeneration
Even though the thermal regeneration technique appears to be more effec-
tive in restoring the Cr(VI) removal capacity, the disadvantages of a thermal
regeneration system are significant.
1. Thermal regeneration at 950°C in C02 results in a 5% loss of carbon, which
translates into significant carbon replacement costs.
2. The costs of building and operating a multiple hearth or fluidized bed
furnace at 950°C for regeneration are prohibitive except on a very large
scale.
3. The resulting air pollutants, chrome carbonyl or other organo-chromium
compounds, which are formed at high temperatures, would require a great
deal of additional study and undoubtedly require stringent controls which
could be economically and technologically unfeasible.
Therefore, thermal techniques for regeneration of carbons laden with inorganic
compounds cannot be an environmentally compatible or economically sound pro-
cedure.
Removal of Chromium from the Carbon Surface
In order to evaluate the amount of chromium removed from the carbon sur-
face by regeneration, the original adsorption density and amount of chromium
in the regenerate stream must be known. Acid or caustic stripping solutions
may be analyzed by standard procedures. However, the analysis of the chromium
in the gas flow from thermal regeneration is difficult. Therefore, a proced-
ure for analyzing the amount of chromium directly on the carbon surface may
be used. By bombarding elements heavier than sodium with protons, the ele-
ment can be identified by the wavelength of the resulting x-ray emission and
the amount can be determined relative to a reference sample. Since carbon is
lighter than sodium, this procedure is ideal for an investigation of heavy
45
-------
metals on the carbon surface.* Figure 24 shows the x-ray intensity for the
chromium emission from the carbon surface after different regeneration tech-
niques were employed.
in 2
'o
•o
o
(E
0 - Unueed Carbon
I - Dried At I03°C
2-1% NaOH
3 - Heated At 990°C In C02
4 - 1% NaOH Then Heated .
At 950°C In C02
S - Heated At 950°C In C02
Then 1% NaOH
6 - Heated At 950°C In C02
Then I N HCI
7 - Heated At 950°C In C02
Then Dlttilled H,0
2468
Regeneration Technique
Figure 24. Measure of chromium remaining on the carbon surface
by x-ray emission, following various regeneration
techniques.
This figure indicates qualitatively that little removal of the chromium
adsorbed occurs during thermal regeneration (technique no. 3) compared to
caustic regeneration (technique no. 2), and acid or caustic desorption after
thermal regeneration (techniques no. 6 and 7) is unsuccessful in removing the
adsorbed chromium. Therefore, there is some irreversible chemical reaction
which occurs during thermal regeneration which locks the chromium onto the
carbon surface.
Caustic Desorption of Adsorbed Chromium
The desorption of chromium from the carbon surface by treatment with
caustic solution can be measured directly from chromium analysis of the waste-
water treated and the regenerant solution used. Figure 25 shows the percent
of chromium desorbed from the carbon surface by various NaOH solutions after
successive 24 hour regeneration contact periods with the carbon. Ten grams of
carbon were used, which was brought close to equilibrium in a packed column
after 1300 bed volumes of influent (1 x 10~3 M Na2CrOtt, pH 2.50 and 0.1 M
NaCl), which was equivalent to an adsorption density of oa 940 ym/g. The
* This procedure was developed by Dr. C. M. Fou, University Physics Depart-
ment, University of Delaware, Newark, Delaware 19711.
46
-------
carbon was then regenerated in 500 ml of NaOH solution at the indicated
strength. If an insufficient amount of Cr(VI) was stripped off after the
first contact period, the regenerant was filtered off and placed in a second
identical regenerant solution; this was repeated for a third cycle if neces-
sary.
100 -
Figure 25.
No. Of Regenerations
Cummulative percent desorption of chromium from the
carbon surface by caustic solution after successive
repetitions.
From Figure 25, chromium is more effectively desorbed from the carbon
surface as the strength of the caustic solution is increased, but the weaker
solutions approach the efficiency of the stronger solutions as the number of
regeneration cycles are increased. Therefore, the concentration ratio between
chromium in the bulk solution and on the carbon surface increases with the
caustic strength. Figure 26 shows the Cr(VI) concentrations reached in the
bulk solution for the various caustic solutions during the regeneration
cycles. Notice that close to 100% regeneration is achieved by 1.0 M NaOH
after 1 cycle for 500 and 250 ml volumes of caustic solution, but the concen-
tration of Cr(VI) is doubled for the solution of lesser volume.
The primary goal of any waste treatment system is to concentrate large
volumes of wastewater into a small volume of waste which is easy to handle
and dispose of. Therefore, the purpose of regeneration is not only to remove
the adsorbed chromium from the surface, but to concentrate that chromium into
the smallest regenerate stream possible. Table 9 evaluates the equilibrium
Cr(VI) concentrations reached in Figure 26 and shows the percent volume of
the regenerant stream compared to the influent stream.
47
-------
10
20 10 2O
Time (hours)
10
20
Figure 26. The concentration of Cr(VI) desorbed in caustic solution
as a function of time.
TABLE 9. COMPARISON OF WASTEWATER CONCENTRATIONS AND VOLUME
REDUCTION WITH STRENGTH OF CAUSTIC REGENERANT STREAM.
(M)
0.01
0.10
1.00
[NaOHH
(% by weight)
0.04
0.40
4.00
CCr(VI)]e
(M x 10~2) (ppm)
0.37 193
1.44 749
3.78 1966
*
Amount Influent
was concentrated
(X)
3.7
14.4
37.8
*
% of Influent
Volume
27.0
6.9
2.6
* Based on 1 x 10~3 M Na2CrOit.
The concentration of Cr(VI) desorbed as a function of NaOH concentration
is shown in Figure 27 and can be expressed by the following equation:
Aft
-------
0.6
-2
- I
log [NoOH]
Figure 27. Maximum Cr(VI) concentration in various caustic
regenerant solutions.
log [Cr(VI)He = 0.6 log [NaOH] - 1.3
(20)
and the volume of regenerant solution needed for approximately 100% regener-
ation of the carbon surface is:
c x r x io~6
LCr(VI)J_VR
(21)
where C = mass of carbon (g)
T = Cr(VI) adsorption density (um/g)
[Cr(VI)l = equilibrium concentration of Cr(VI) in the bulk NaOH
e solution calculated from equation 20 (M)
V = volume of regenerant solution needed for oa 100%
Cr(VI) removal (£)
49
-------
Acidic Desorption of CrCVI)
Regeneration of the exhausted carbon with strong acid solution appears
attractive in an economical sense because acid is generally less expensive
than caustic. Figure 28 shows the reduction and desorption of adsorbed Cr(VI)
from the carbon surface in 0. 1 M HC1 solution. The desorption in the acid
solution is much slower than that observed in the caustic regenerate solu-
tions, however, stronger acid solutions might be used to speed up the re-
action and increase the extent of desorption. Regeneration with strong acids
does have many disadvantages compared to caustic regeneration:
1. The chromium is recovered only in the Cr(III) form; Cr(VI) is not
recoverable.
2. The acid regenerant stream must be neutralized with caustic in order to
precipitate the Cr(OH)3 species. Therefore, any cost advantage is lost
and more caustic must be used than acid.
3. The resultant sludge must be dewatered and disposed of.
40 -
B
in
<
10 grams F- 400
500 ml O.I M HCI
10
20 30 40
Reaction Time (hr)
50
60
Figure 28. The desorption of Cr(VI) from the carbon surface in
acidic solution.
50
-------
SECTION 7
SYSTEM DESIGN
The removal of Cr(VI) from wastewater by an activated carbon process in-
volves reduction and adsorption of the Cr(VI) species occurring simultaneous-
ly. Based upon the results obtained in this research, two types of treatment
schemes may be suggested:
1. That the reaction parameters are manipulated to favor the adsorption of
Cr(VI) to the extent that no significant amount of Cr(III) will be pro-
duced .
2. That the reduction of Cr(VI) is allowed to occur and a subsequent treat-
ment scheme is utilized to remove the Cr(III) which has been produced.
From the previous kinetic studies, the major reaction parameters for re-
duction and adsorption were found to be pH, degree of mixing or velocity gra-
dient (G), carbon dose (DO), and total Cr(VI) concentration. By comparing
the kinetic expressions for reduction and adsorption (Equations 11 and 17),
the most effective parameters governing the ratio of reduction rate to adsorp-
tion rate (Rred:^ads) are P*1 anc* **• From Equation 11, it is seen that Rred
a EH4"] and G; while Equation 17 shows Rads <* D^H0'2 and is independent of G
over a normal range of mixing.
Figure 29 shows the rates of reduction and adsorption (M-min"1). calcu-
lated from Equations 11 and 17, assuming constant values for CHCrO^I], CcH,
and CCr(III)!] = 0, at various values of G (for the reduction rate) and ir (for
the adsorption reaction) as a function of solution pH.
From Figure 29, the adsorptive removal becomes dominant as the pH in the
bulk solution is increased. However, the adsorption rate decreases as the
carbon approaches maximum capacity (IT •*• 1.0) and the reduction reaction will
dominate when i: becomes sufficiently close to 1.0. Therefore, if the treat-
ment process is to be operated in an adsorption-controlled mode, with little
Cr(III) produced, i.e., Rred:Rads - 0» then the wastewater must be contacted
with the carbon in a completely mixed type reactor; then ir can be controlled
by the residence time of the carbon in the reactor. In contrast to this, the
initial layers of carbon approach equilibrium rather quickly in a packed
column reactor. The resulting concentration gradient along the column is
utilized to optimize the adsorption density and thereby fully utilize the
carbon capacity to the fullest extent.
51
-------
-6
-7
-8
c
o
o
S -9
-10
- II
2.5
3.5
4.5
PH
TT'O
5.5
6.5
Figure 29. The rates of reduction and adsorption (mole/£-min) as
functions of pH, G and IT* for [cG = 2 g/£, [HCrO£] =
1.0 x 10-5 M and [Cr(III)I] = 0.
* r
TT = — =» 0 initially and approaches 1 as total adsorption
re
capacity is reached.
SYSTEM 1 - TREATMENT BY ADSORPTION IN A COMPLETELY MIXED REACTOR CCMR)
A CMR offers the advantage of simplifying the control on all the reac-
tion parameters for reduction and adsorption. The variables which may be
•selected for the design of a CMR system are:
1. The velocity gradient in the reactor, G (sec"1).
2. The solution pH in the reactor.
3. The carbon dose in the reactor, CcJ (g/£) .
4. The extent of the carbon adsorptive capacity maintained, ir.
5. The ratio of reduction rate to adsorption rate, Rred/Rads-
The simplest way to restrict Rre^/Rads ls by controlling the velocity
gradient in the reactor. Since Rre(j a G, the lowest possible value of G is
52
-------
desirable as long as Ra(js is not effected. Therefore, G = 5 sec"1 has been
selected for all design calculations in this system.
Figure 30 shows Rred/Rads at several values of TT and G = 5.0 sec'1 as a
function of pH. Figure 30 also indicates the need to operate at an increas-
ingly high pH value to maintain a small ratio of Rreds to Rads as T increases.
The dependence of IT on pH to maintain a constant Rred/Rads *-s shown in Figure
31. Therefore, from Figure 31, the minimum pH or maximum IT value for any
Rred to Rads ratio may be determined. To insure that the production of
Cr(III) is insignificant, a value of Rred/Rads of at most 1:5° is necessary.
To meet this criteria, the pH in the reaction vessel must be greater than oa
4.8 (from Figure 31). Many combinations of pH and IT may be chosen to satisfy
the recommended value of Rred/Rads» however, several constraints on choosing
the proper pH and ir values must be considered:
1. Is the Cr(VI) capacity of the desired carbon dosage, at the pH and TT
value, large enough to accommodate the Cr(VI) which must be removed
from the wastewater?
2. Are the kinetics of adsorption fast enough to insure the desired removal
in a reasonable reactor size?
7 -
Z.5
3.5
4.5
PH
5.5
6.5
Figure 30. The ratio of Rred to Rads as a function of pH
(calculated from Figure 29).
53
-------
The Cr(VI) adsorptive capacity of the carbon can be determined from
Equations 13 and 14 and the adsorption density on the carbon in a reaction
vessel at any pH and ir can be evaluated as:
r = ir Te (22)
where r = adsorption density in the reactor (y mole/g).
IT = fraction of adsorptive capacity chosen from Figure 31.
Te = maximum adsorptive capacity calculated from Equations 13 and 14
(y mole/g).
0.5-
Figure 31. ir vs. pH over equal values of Rred/^ads
The minimum carbon dosage required for treatment in one or more CMR is:
Cc]
CCr(VI)3 x P x
o
min
(23)
where
.
mm
minimum carbon dosage required
CCr(VI)H = Cr(VI) concentration in the wastewater before treatment (M) .
o
P = percentage of Cr(VI) removal desired (%) .
54
-------
r = desired adsorption density in the head reactor calculated
from Equation 22 (y mole/g).
The minimum carbon dose required to maintain a 1:50 Rred/Rads rati° f°r 99%
Cr(Vl) removal from a 1 x 10~3 M Na2CrOtt (52 ppm Cr(VI) ) wastewater solution
as a function of pH and Cr(VI) concentration in the head reactor are shown in
Figure 32.
100
o
Q
o
o
50
I xI0"a M
Ixio"4
5x10
4.5
5.5
pH
6.5
Figure 32. The carbon dose required for 99% treatment of a 1 x 10~3
M Na2CrOit solution as a function of pH and Na2CrOit in the
head reactor.
The carbon dose is high at the lower pH range due to the lower ir values
necessary to maintain Rred/Rads = 1:50. The absolute minimum dosage occurs
over a fairly flat region between pH 5.2 to 6.5.
The carbon dosage also decreases as the concentration of Cr(VI) in the
reactor increases, due to the high adsorptive capacities (re) at high concen-
trations. This in turn indicates that the carbon may be used more efficient-
ly if more than one reactor in series is used (i.e., continuous counter-
current operation).
55
-------
Case Study - Continuous Completely-Mixed Reaction System
A great number of system designs may be used, all of which are capable
of providing the same degree of treatment, however, it is the cost of these
individual designs which will determine the best or optimal solution to the
wastewater treatment. In order to evaluate the economics of operating at
different pH, carbon dose or TT values, the cost associated with the size of
the reactor needed and the cost of the carbon used by the system, must be
determined.
In order to optimize these operational variables, a hypothetical case
was considered as follows:
Wastewater flow = 10,000 gal/day
CCr(VI)] = 1 x 10~3 M (52 ppm)
Removal percentage = 99%
Rred:Rads = 1:5°
System operation = 8 hrs/day and 300 days/year
Capital costs are amortized over 10 years at an interest rate of 8%
(i.e., capital recovery factor = 0.149)
Filtrasorb 400 = 61/lb in bulk rates
Considering the corrosiveness of the Cr(VI) wastewater, the best mater-
ial for the reaction vessel would be fiberglass. A rough survey of costs for
large fiberglass tanks indicates that:
log (C.C.) - 0.8 log (V)*0.63 (24)
where:
C.C. = capital cost of a fiberglass tank ($)
V = volume of the reaction vessel (gal)
By neglecting the reduction of Cr(VI) species and assuming that all
Cr(VI) removal will occur through adsorption, the reactor volume needed for
a single reactor, or each vessel in a series of equal sized reactors, can be
found by applying a mass balance to the system at a steady state and using
Equation 18 for the rate of adsorption, or:
56
-------
y.ii x *
[H+]0.2 (_JL_ 1}
where:
V = volume of each reactor (gal)
Te = maximum adsorptive capacity at given pH and Cr(VI) concentration
in the reactor, calculated from Equations 13 and 14 (y mole/g)
EH H = hydrogen ion concentration (M)
IT = fraction of adsorptive capacity
P = percentage of Cr(VI) removal in each reactor (%)
and P may be calculated for any number of equal sized vessels in series:
1
100 - PT ~
P - 10° - < <—TorT^ x 10° ) (26)
where:
f
-
PT = total removal efficiency of the entire treatment train (%)
n = number of vessels in series (n = 1, 2, 3...)
Evaluation of pH for Cost-Effective Adsorption of Cr(VI)
It has been determined previously that an effective treatment system
must operate with at least 1:50 Rred'^ads ratio, which restricts the maximum
ir value (iTinax) at different pH's as shown by Figure 31. In turn, the requir-
ed carbon dosage is affected by irmax as shown in Equation 23 and the reactor
volume is fixed by the pH, carbon dose and IT value. Therefore, the trends
indicate an increase in capital costs due to the increased vessel size as the
pH and ir values.increase, while the operating costs due to carbon usage de-
crease to a minimum with the carbon dosage as shown in Figure 32. A compari-
son of reaction vessel sizes and costs at various allowable pH's and ir < ifmax
are shown in Table 10.
Derived by substituting Cc] = *'"uu into Equation 18 (applies for 1 x 10~3
M Cr(VI) only).
57
-------
TABLE 10. REACTOR SIZE AND ESTIMATED COSTS ASSOCIATED WITH THE
REACTOR VOLUME AND APPLIED CARBON DOSE FOR A SINGLE CMR.
pH
4.88
5.0
5.25
5.50
6.0
IT
0.05
0.10
0.15
0.1
0.2
0.3
0.1
0.3
0.5
0.1
0.4
0.6
0.7
0.1
0.4
0.6
0.9
*
V
(gal)
1,960
4,140
6,570
4,140
9,300
15,970
4,130
15,940
37,200
4,130
24,810
55,810
86,860
4,140
24,840
55,890
335,210
cf
(g/4)
357
179
119
189
95
63
212
70
43
238
60
40
34
299
75
50
33
C.C.
($)
1,840
3,340
4,830
3,340
6,380
9,830
3,340
9,830
19,340
3,340
13,983
26,750
38,110
3,340
13,980
26,754
112,266
§
Cost of Carbon
(C/gal)
3.58
1.80
1.20
1.90
0.95
0.64
2.13
0.70
0.43
2.38
0.60
0.41
0.35
3.00
0.75
0.50
0.33
Volume calculated from Equation 25 assuming P = 99%.
Carbon dose calculated from Equation 23.
T Capital cost calculated from Equation 24.
• Based on 2% carbon loss during use and regeneration at 61$/lb Filtrasorb
400.
The assumed carbon cost was based on a 2% loss of carbon after each
cycle of use. The complete summary of total treatment cost due to amortized
capital costs over 10 years at 8% interest rate and operating expense due to
the carbon loss, is shown in Figure 33. The graph shows that the most econ-
omical design for a single reaction vessel occurs at pH 5,25 and TT = 0.55.
The total treatment cost is 0.52(?/gal or $52/day ($15,600/year) for a 10,000
gal/day operation.
58
-------
3.0
.- 2.0
o
o
c
«
6
o
o
1.0
nnnhnum _co»t
0.52*/gal
0.2
0,4
0.6
0.8
TT
Figure 33. Treatment costs in a single complete-mix contact
vessel at various pH's as a function of ir.
Equal Sized Reactors in Series
From Table 10, the most significant portion of the treatment cost re-
sults from replenishing the carbon supply. This cost accounts for 77% of op-
timal treatment cost at pH 5.25. In a series of CMR's, the steady state con-
centration of Cr(VI) in each reactor is larger than the Cr(VI) concentration
of the following reactor in the series. Therefore, by using a series of CMR's
and cycling the used carbon from the tail reactor to the head end of the sys-
tem before regenerating, the step-ups in Cr(VI) concentration can be utilized
to increase the carbon adsorptive capacity (re) and thereby decrease the car-
bon dosage needed for-treatment. As the number of CMR's in series increases,
the system will gradually approach the carbon efficiency of a packed column
and the treatment cost associated with the carbon dose will approach a mini-
mum. The series of reactors also reduces the reactor volume needed to achieve
59
-------
the same degree of treatment. Capital costs for 1, 2, 3, and 4 reactors in
series were estimated using Equations 25 and 26 for the volume of each reactor
and Equation 24 (less 10% for more than one vessel) plus an additional $3,000
for extra equipment, such as pumps, pH controls, carbon separators and mixers,
on each additional reactor. Carbon costs were estimated by assuming 2% loss
per cycle. Figure 34 compares the total treatment costs and indicates that
the cost can be decreased to oa 0.20<:/gal at IT = 5.5 for 4 reactors in a
series. However, the difference between 3 and 4 reactors appears insignifi-
cant 0.21c/gal vs. 0.20c/gal, and the increased complexity of the system for
each additional reactor would indicate that 3 reactors would be the optimal
design at a cost of 0.2l£/gal or $21/day.
i.o
o
Oi
o
o
0.5
pH -5.25
I reactor
O.I
0.3
7T
0.5
Figure 34. Comparison of treatment costs for multiple reactors in
series.
SYSTEM 2 - COLUMN OPERATION
Column operation is more efficient than a completely mixed reactor, due
to the fact that a column fully utilizes the carbon adsorption Capacity due to
a large concentration gradient existing opposite to the direction of flow in
the column. Therefore, Cr(VI) adsorption may be less expensive in a column if
a subsequent treatment system is employed to remove the Cr(III) which has been
produced.
Column operation is sensitive to the pH in the wastewater influent
stream, since there is a staichiometric requirement of one mole of H* per mole
Cr(VI) adsorbed and 4 moles of H+ per mole Cr(VI) reduced. In order to
60
-------
completely remove all of the Cr(VI) from solution, a 1:1 molar ratio of H"1" to
Cr(VI) would be the absolute minimum ratio of H+rCrCVI), if Cr(VI) was removed
exclusively by adsorption. In fact, an excess of H+ is required to satisfy
the stoichiometric demand of H"*~ for reduction and to prevent the decreased H"^
concentration in the latter portions of the column from limiting the removal
rates and diminishing the Cr(VI) adsorptive capacity of the carbon.
For a case study similar to the one presented previously, i.e., 1 x 10~3
M Cr(VI), the maximum pH in the influent would be 3.0. However, an excess of
H* is desired and the maximum Cr(VI) adsorptive capacity of the carbon occurs
at ea pH 2.50 (1,280 y moles Cr(VI)/g). Bench scale experiments have also in-
dicated excellent Cr(VI) removal performance at this pH value (Figure 19).
Therefore, pH 2.50 appears to be the optimum condition for complete removal of
Cr(VI) by adsorption and reduction.
The surface loading rate of the carbon columns in this study has been
2 gal/min/ft2. Therefore, at a wastewater flow rate of 10,000 gal/day oper-
ating 8 hr/day, the column would require 10.4 ft2 of surface area or be 3.64
ft in diameter.
The.depth of the column can be estimated from experimental data. Table
11 shows the experimental bed size and the number of influent bed volumes
passed through each column before breakthrough occurred.
TABLE 11. COLUMN DEPTH VERSUS NUMBER OF BED VOLUMES TO Cr(VI) BREAKTHROUGH
Bed Size
(grams)
10
30
50
Bed Depth
(ft)
0.14
0.42
0.70
Number of Bed Volumes
to Breakthrough
125
400
600
* Flow rate = 2 gal/min/ft2; pH =2.5; total Cr(VI) = 1 x 10~3 M
o o
A log plot of bed volumes versus bed depth is shown in Figure 35. The
number of bed volumes may then be written as a function of bed depth or:
log (BVb) = 0.92 log (dp + 2.94 (27)
where:
BV, = number of bed volumes to breakthrough
d, = carbon bed depth (ft)
61
-------
The depth of the bed needed for a system can be evaluated by choosing
the desired time interval between regenerations, which implies:
b 7.48 (SA)d,
(28)
where:
Q = wastewater flow rate (gal/day)
t = time interval between regenerations (days of actual system opera-
tion)
SA = Column surface area perpendicular to the flow (ft2)
Q, t, and SA should all be known, from which d, can be derived by solv-
ing Equation 27. By assuming 10,000 gal/day with regeneration cycles once a
month (22 days of actual operation), d, = 2.4 ft. Therefore, a 4 ft diameter
by 3 ft depth carbon bed would be a conservative design and require 1,082 Ib
of carbon. Therefore, regenerating once a month at a 2% carbon loss means
the replacement cost for the column would be O.OlC/gal. The column operation
offers a significant advantage over the continuous flow system where the car-
bon operating cost was 0.12<:/gal at pH 5.25 and IT = 5.5 using three reaction
vessels.
2.8
• 2.4
o
>
e»
o
2.0
Inflow • 2 gol/mln/ft
pH -2.5
Ixio"3 M No2Cr04
0.92
0.2 0.6
-log [bed depth (ftjj
1.0
Figure 35. Bed volumes to breakthrough as a function of carbon bed
depth for 1 x 10~3 M Na2CrOit at pH 2.50.
62
-------
The initial column must be followed by a Cr(III) treatment system capable
of treating 10,000 gal/day with 2.0 to 3.0 x I0~k M Cr(III) (see Figure 19).
Although no data on the adsorption of Cr(III) by activated carbon has been
presented in this report, it is believed that a comparable system for its re-
moval might be developed and operated at about the same cost as the Cr(VI)
adsorption column and the total carbon cost might be 0.02 to 0.03c/gal for
complete adsorption of all chromium species from solution.
It is also relevant to compare the cost of using activated carbon for
Cr(Vl) reduction to the cost of conventional reduction with sulfur dioxide
(802). The following unit material costs can be attributed to each:
Filtrasorb 400 costs 61/lb or $61/100 Ib
S02 costs $8.51/100 Ib
From the stoichiometry of the reduction reactions:
1. for Cr(VI) reduction by carbon, Equation 9 implies that three moles of
carbon is required to reduce four moles of Cr(VI)
2. for Cr(VI) reduction by S02 the following stoichiometry applies:
+ 3 S02 J Cr2 (80^)3 + 2 H20
or three moles of S02 is required per two moles of Cr(Vl) reduced.
Therefore, assuming 100% efficiency of reaction, for 10,000 gal of 1 x
10~3 M Na2CrOtt per day, to completely reduce all of the Cr(VI), the chemical
costs would be 51/day for S02 and 46/day for Filtrasorb 400. Therefore,
it is favorable to use activated carbon for reduction of Cr(Vl) in solution
compared to currently used reducing agents.
63
-------
REFERENCES
1. Udy, M. J. Chromium. Vol. I. Rheinhold, New York, New York, 1956.
433 pp.
2. Thackston, E. L. Secondary Waste Treatment for a Small Diversified
Tannery. EPA-R2-73-209, U. S. Environmental Protection Agency,
Cincinnati, Ohio, 1973* 75 pp.
3. Mearns, A. J., P. S. Oshida, M. J. Sherwood, 0. R. Young and D. J. Reish.
Chromium Effects on Coastal Organisms. J.W.P.C.F., 48(8):1929-1939,
1976.
4. Atchinson, G. J. Uptake and Distribution of Trace Metals in Fish.
Report No. 73. Purdue University Water Resources Research Center, West
Lafayette, Indiana, 1975.
5. Smithson, G. R., Jr. An Investigation of Removal Techniques for Removal
of Chromium from Electroplating Wastes. Proj. No. 12010 EIE. Environ-
mental Protection Agency, Cincinnati, Ohio, 1971. 91 pp.
6. Page, A. L. Fate and Effects of Trace Elements in Sewage Sludge When
Applied to Agricultural Lands: A Literature Review Study. EPA-670/2-
74-005, U. S. Environmental Protection Agency, Cincinnati, Ohio, 1974.
7. Lanouette, K. H. Heavy Metals Removal. Chemical Engineering, 84(22):
73-80, 1977.
8. Risley, C., Jr. Waste Water Treatment and Reuse in a Metal Finishing
Job Shop. EPA-670/2-74-042, U. S. Environmental Protection Agency,
Cincinnati, Ohio, 1974. 59 pp.
9. Tabakin, R. B., and J. Ciancia. An Ion Exchange Process for Recovery of
Chromate from Pigment Manufacturing. EPA-670/2-74-044, U. S. Environ-
mental Protection Agency, Cincinnati, Ohio, 1974. 92 pp.
10. Thomas, M. J., and T. L. Theis. Effects of Selected Ions on the Removal
of Chrome (III) Hydroxide. J.W.P.C.F., 48(8);2032-2045, |976.
11. Dean, J. G., and K. H. Lanouette. Removing Heavy Metals from Wastewater.
Environmental Science and Technology, 6(6):518-522, 1972.
12. Weber, W. J., Jr. and J. C. Morris. Adsorption in Heterogeneous Aqueous
Systems. American Water Works Association J., 56(4):447-456, 1964.
64
-------
13. Kawashimat, et al. Treatment of Wastewater Containing Heavy Metal Ions
Using Activated Charcoal. Mitzushou Gijutsu (Japan), 14(4):379, Chemical
Abstracts 79, 57376, 1973.
14. Saito, I. Removal of Heavy Metals from Aqueous Solutions Using Sulfon-
ated Coal and Activated Carbon. Kogai Shigen Kenkyusho Iho (Japan),
5(2):57-64, 1976.
15. Huang, C. P., F. B. Ostovic. The Removal of Cadmium (II) From Dilute
Aqueous Solution by Activated Carbon Adsorption. Presented at: 173rd
Nat. Meeting A. C. S., New Orleans, Louisianna, 1977.
16. Abe, T. Purification of Chromate-Containing Waste Water. Kokai (Japan)
740717, 1974.
17. Tagashira, Y., et a}.. Removal of Chromium Ions from Waste Waters with
Activated Carbon. Kokai (Japan) 750820, 1975.
18. Huang, C. P., and M. H. Wu. Chromium Removal by Carbon Adsorption.
J.W.P.C.F., 47(10):2437-2446, 1975.
19. Miyagawa, T-., S. Ikeda, K. Koyama. Removal of Heavy Metals from Waste
Water. Kokai (Japan) 760417, 1976.
20. Nagasaki, Y. Removal of Metal Ions from Wastewater with Coal Adsorbent.
Kokai (Japan) 122, 477, 1974.
21. Nagasaki, Y., A. Terada. Chromium-Containing Waste-Water Treatment with
Activated Carbon. Kokai (Japan) 750721, 1975.
22. Landrigan, R. B., and J. B. Hallowell. Removal of Chromium From Plating
Rinse Water Using Activated Carbon. EPA-670/2-75-055, U. S. Environ-
mental Protection Agency, Cincinnati, Ohio. 44 pp.
23. Yoshida, H., K. Komegawa, S. Arita. Adsorption of Heavy Metal Ions on
Activated Carbon. Nippon Kagaku Kaishi (Japan), Issue 3, 387-390, 1977.
24. Huang, C. P. and M. H. Wu. The Removal of Chromium (VI) from Dilute
Aqueous Solutions by Activated Carbon. Water Research, 11(8):673-679,
1977.
25. Kim, J. I. Adsorption of Chromium on Activated Carbon. Ph.D. Thesis,
University of Florida, Gainesville, Florida, 1976. 196 pp.
26. APHA - AWWA - WPCF. Standard Methods for the Examination of Water and
Wastewater. 14th Edition, 1975. 1193 pp.
27. Steenburg, B. Adsorption and Exchange of Ions on an Activated Carbon.
Almquist and Wilsells, Uppsala, Sweden, 1944.
65
-------
28. Mattson, J. S., and H. B. Mark, Jr. Activated Carbon. Marcell Dekker,
New York, New York, 1971.
29. Parks, G. A. Aqueous Surface Chemistry of Oxides and Complex Oxide Min-
erals. In: Equilibrium Concepts in "Natural Water Systems, R. F. Gould,
ed. ACS Publications, Washington, D. C., 1967- pp. 121-160.
30. Latimer, W. M. Oxidation Potentials, 2nd ed., Prentice-Hall, Englewood
Cliffs, New Jersey, 1952. 392 pp.
31. Masterson, W. L. and E. J. Slowinski. Chemical Principles. 3rd ed.,
W. B. Saunders, Philadelphia, Pensylvania, 1973. 707 pp.
32. Levenspiel, 0. Chemical Reaction Engineering. 2nd ed-., John Wiley &
Sons, New York, New York, 1972. 578 pp.
33. Weber, W. J., Jr. and J. C. Morris. Kinetics of Adsorption in Columns
of Fluidized Media. J.W.P.C.F., 37(4):425-443, 1965.
34. Pourbaix, M. Atlas d'Equilibres Electrochemiques. Gauthiers-Villars,
Paris, France, 1963. 644 pp.
35. Chamberlain, N. S. Technology of Chrome Reduction with Sulfur Dioxide.
In: Proceedings of the llth Industrial Waste Conference, Purdue Univer-
sity, West Lafayette, Indiana, 1956. p. 129.
36. Stumm, W. and J. J. Morgan. Aquatic Chemistry. John Wiley & Sons, New
York, New York, 1970. 583 pp.
66
-------
APPENDIX
AQUEOUS CHEMISTRY OF CHROMIUM
Chromium is a highly active transition metal which exists in several
oxidation states from the divalent to hexavalent form. However, the major
stable aqueous species are either in the hexavalent or trivalent state (1).
The stability of the various chromium species is dependent upon the various
reducing, oxidation and pH conditions. Figure A-l shows the stable domains
for various chromium species in aqueous systems as affected by the oxidation
potential (Eh) and pH (34). Under the pH (< 3), temperature (20-30°C) and
reducing-oxidizing conditions commonly found in industrial wastewaters, the
predominant species are bichromate, HCrO^, dichromate, Cr207 2, and Cr+3- It
is interesting to note that the divalent chromium ions, Cr+2, may be found in
extremely reducing environments.
The constitution of hexavalent chromium in the wastewater includes sev-
eral common hexavalent chromium anions; namely, (a) the hydrochromate or bi-
chromate monovalent anion, HCrO^; (b) the anhydride of the acid chromate ion,
the divalent dichromate anion, Cr207 ; and (c) the divalent chromate anion,
CrOit" (35). The equilibria of various Cr(VI) species are shown as follows
(36):
+ H+ £ HCrO^ ; log kx = 6.5
+ H+ £ H2Cr04 ; log k2 = -0.8
£ Cr207~2 + H20 ; log k3 = 1.52
Cr207~2 + H+ J HCr207 ; log k^ = 0.07
Based upon the above equilibria, the distribution of various Cr(VI)
species as a function of pH and concentration can be computed and the fraction
of the total Cr(VI) species which is in the HCrO^ form under most conditions
can be written quantitatively as:
lO-6'5 + I 10-6'5 2 2
l + ^rW Id + ^r^F-r)2 + 1-33 x io2
(29)
66,3
67
-------
UI
Figure A-l.
The stability diagram of chromium species in aqueous
solution at 20°C, after Pourbaix (34), as function of
E. and pH. The values of 0, -2, -4 designate
- log (CrT - molar).
where:
al •
the fraction of Cr(VI) species which are
= 10~pH (M)
= total Cr(VI) in solution (M)
The concentration distribution between HCrO^ and C^Oy' is largely governed
by the total Cr(VI) present. As indicated in Figure A-2 (36), the fraction
of C^Oy"2 only becomes significant at high concentration of total Cr(VI).
Figure A-2 also shows the distribution of Cr(VI) species as function of pH for
total Cr(VI) covering from ID"4 M to 6 x 10~3 M (or 5.2 to 312 ppm as Cr), an
average concentration range which generally occurs in most industrial waste-
waters (5).
68
-------
80
40
HCrO~,J
6 xlO3 M
80
« 40
o
2
HCrO~4
I x IO"3 M
80
40
HCrO"4
I x IO4 M
4 5
PH
Figure A-2.
The molar distribution of the Cr(VI) species as
function of pH and total Cr(VI). (Recalculated
from Stumm and Morgan)(36).
When present in a trivalent state, the dominant solid form is Cr(OH)3(s)
• nH20 which is not very soluble. The solubility of trivalent chromium
hydroxide is minimum at pH 8.5 (= 0.02 ppm) and increases to 520 ppm at pH 5.
Figure A-3 shows the effect of pH on the solubility of Cr(OH)3(s) • nH20.
69
-------
-i 4
§
o
i_i
9
o
10 12
PH
14
Figure A-3. The concentration distribution of various Cr(III)
species as governed by the solubility of solid
Cr(OH)3 • nH20, after Pourbaix (34).
70
-------
3. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Delaware
Newark, Delaware 19711
TECHNICAL REPORT DATA
//'/< air rciiJ lnMriiilii»i!i nn the r.'i rnc he/an- coint>lcling)
HI.POH F NO.
_ EPA-600/2-79_-130
i. TITLL AND SUBIITLC
Activated Carbon Process for Treatment of Wastewaters
Containing Hexavalent Chromium
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
July 1979 issuing date
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
C. P- Huang and Alan R. Bowers
8. PERFORMING ORGANIZATION REPORT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab.
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati. Ohio 45268
- Cinn, OH
10. PROGRAM ELEMENT NO.
1BB610
11. CONTRACT/GRANT NO.
R-804656
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The removal of hexavalent chromium, Cr(VI), from dilute aqueous solution by an acti-
vated carbon process has been investigated. Two removal mechanisms were observed: hexa-
valent chromium species were removed by adsorption onto the interior carbon surface and/
or through reduction to the trivalent state at the external carbon surface.
The effects of Cr(VI) concentrations, pH, carbon dosage and extent of mixing in the
reaction vessel were studied in the batch mode and in continuous flow packed column ex-
periments in the laboratory. The adsorptive capacity of the carbon and the rates of
Cr(VI) adsorption and reduction have been determined.
Thermal regeneration of the exhausted carbon was examined, along with caustic or
acid stripping solutions and a combined caustic-thermal process.
A case study was presented and the experimental data and rate expressions obtained
from the data were used to evaluate the design variables (i.e., pH, carbon dose, Cr(VI)
concentration and mixing in the reaction vessel). Several Cr(VI) treatment schemes were
proposed, together with an economic analysis of each scheme.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
jb. IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Adsorption
Reduction
Activated Carbon
Batch mode adsorption
Packed Carbon Column
Regeneration
Hexavalent Chromium
13B
•<--t. DISTRIBUTION STATEMEN r
Release to Public
; 19. SECURITY CLASS (ThisReport)
| Unclassified
r20" SECURITY CLASS~(Tliispage)
: Unclassified
21. NO. OF PAGES
81
22. PRICE
Form 2220-1 (9-73)
71
ft U.S. GOVEMOISn reudlKCOfFlCt 1975 -657-060/5356
------- |