WATER POLLUTION CONTROL RESEARCH SERIES
17O1OEEX1O/7O
DEVELOPMENT OF A CHEMICAL
DENITRIFICATION PROCESS
ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFICE
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Reports describe
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Environmental
Protection Agency, through inhouse research and grants
and contracts with Federal, State, and local agencies,
research institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Planning and Resources Office, Office of Research
and Development, Environmental Protection Agency, Room 1108,
Washington, D. C. 20242
-------
DEVELOPMENT OP A CHEMICAL DENITRIPICATION PROCESS
Rocketdyne Research
North American Rockwell Corporation
Canoga Park, California 91304
for the
ENVIRONMENTAL PROTECTION AGENCY
Program #17010 EEX
Contract #14-12-5^6
October, 1970
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price 65 cents
Stock Number 5501-0136
-------
EPA REVIEW NOTICE
This report has been reviewed by the
Environmental Protection Agency and
approved for publication. Approval
does not signify that the contents
necessarily reflect the views and
policies of the Environmental Pro-
tection,, nor does mention of trade
names or commercial products con-
stitute endorsement or recommendation
for use.
ii
-------
ABSTRACT
Chemical denitrification of dilute (lO ppra NO-^ -N) nitrate solutions
has been achieved with high conversion of the nitrate to a mixture of
innocuous gaseous products. Laboratory development studies of the deni-
trification process based on the copper catalyzed ferrous iron reduction
of nitrate ion in basic media were conducted to determine the effects of
process variables on the extent of reduction and on product distribution.
In selected experiments the nitrogen mass balance was found to be 95 -
98$ comprised of NH^, N2, N20, N(>2~ and unreacted N0^~. In the majority
of experiments the gaseous-N products (Nfe + NgO) were determined by dif-
ference, a valid method since hydroxylamine was shown not to be a major
reduction product.
The effects of initial pH (or better the OH~/Fe++ mole ratio), Fe
mole ratio, lime vs. NaOH for pH adjustment, catalyst concentration,
catalyst recycle, anions introduced with the iron and copper, and other
anions present in the water were studied at constant 10 ppm NO^ -N con-
centration and constant 85° F temperature. Means for removal of inhib-
itory species such as phosphate and carbonate were evaluated. Limited
studies were conducted on the uncatalyzed reduction of nitrite as an aid
to elucidation of the chemical reactions involved in the overall process.
The influence of selected parameters on the reaction time was studied
and the results were applied to continuous flow reactions.
Preliminary evaluation of the denitrification process was made using tap
water and secondary effluent as reaction media.
This report was submitted in fulfillment of Contract No. 14-12-546,
Program No. 17010 EEX, between the Federal Water Quality Administration
and Rocketdyne, a Division of North American Rockwell Corporation.
KeyVords: Denitrification, nitrates, reduction, ferrous ion, catalysis,
potable water, wastewater
111
-------
CONTENTS
Page
Abstract - - " iii
Conclusions 1
Recommendations """ 2
Introduction 3
Experimental ~~ 5
Discussion 9
Identification of the Gaseous-N Species 9
The Effect of Initial pH --- 10
Effect of the Base - NaOH vs. Ca(OH)2 11
Effect of Fe*+/N0," Mole Ratio 11
Catalyst Evaluation 12
Effect of Anions - - 15
Reduction of Nitrit^ 26
Denitrification Reaction Time and Its
Application to a Dynamic System 2?
Preliminary Evaluation of the Denitrification Process
Using Both Tap Water and Secondary Effluent 29
Analytical Support 30
Acknowledgements 35
References 36
Appendices 38
IV
-------
FIGURES
Page
1. Cross Section of Flow Reactor 7
2. Potentiometric Titration Curves for
FeSO^, FeS04-CuS04, FeClg, and FeClg-CuClg 18
3. Potentiometric Titration Curves for
FeSO^-CuClg, FeCl2-CuS04, and FeSC^-CuSO^-CaSO^ 19
4. Potentiometric Titration Curves for
FeSO,-Na0HPO. , FeSO.-CuSO,-Na0IIPO. - 21
424' 4 424
5. Potentiometric Titration Curves for
FeS04-CuS04-NaHCO 25
-------
TABLES
No. Page
I. Reduction of Nitrogenous Species
to Ammonia 32
II. Determination of Hydroxylamine and
Ammonia in Known Mixtures Jk
VI
-------
CONCLUSIONS
1. Nitrate ion can be reduced by ferrous hydroxide under anaerobic
conditions in the presence of a copper catalyst to a mixture of
nitrogen and nitrous oxide with lesser amounts of nitrite ion and
ammonia.
2. The yield of gaseous-N products (N2 + N20) under current optimum
conditions is ~70 percent with 70 percent conversion of nitrate
ion. Yield of gaseous-N products from nitrite ion is about 90
percent with 90 percent conversion.
3« Current optimum conditions include an initial pH near 8 (mole ratio
OH~/Fe "*•*•= 1.5) mole ratio Fe++/NO-j~ = 8, and catalyst concentration
1-5 ppm Cu++ with the reaction carried out under anaerobic condi-
tions, the Fe'H- ion derived from the sulfate, NaOH used for pH
adjustment, and the solution pretreated to remove phosphate and
decrease carbonate content.
4. Lime can be substituted for sodium hydroxide for pH adjustment with
only a small decrease in the conversion of nitrate ion to gaseous-N
products.
5. The fraction of the sulfate precipitated with the iron hydroxides
is small (10 -
6. Sulfate ion (150 ppm maximum) is necessary to effect formation of
N2 and NgO and thereby decrease formation of ammonia.
7. Phosphate ion inhibits the reduction of both nitrate and nitrite
ions but the effect can be largely eliminated by pretreatment to
remove the phosphate.
8. Carbonate ion is a mild inhibitor, the effect of which can be nulli-
fied by lime pretreatment.
9. Preliminary examination of potable water and secondary effluent
indicates the presence of unknown inhibitory substances, the effect
of which is lessened by the iron salts used for denitrif ication. A
given water would therefore need to be examined for inhibitory
substances and appropriately treated to enable effective application
of the denitrif ication process.
-------
RECOMMENDATIONS
This laboratory process development program was limited largely to
defining the limits of the process parameters and finding the optimum
combination of these parameters to effect maximum conversion of nitrate
ion to nitrogen and nitrous oxide. To a lesser extent means of accessing
and overcoming the inhibitory effects of dissolved substances (i.e. ,
phosphates, carbonates, silicates) were studied. Only in a preliminary
manner were the effects of continuous flow reactors and of actual water
sources investigated. It is recommended that selected nitrate-contami-
nated waters such as stainless steel waste pickle liquor or dyestuff
manufacturing wastewater be used as the reaction medium for a bench
scale continuous flow denitrification process evaluation.
The larger scale investigation of the process, which would include pre-
treatment to nullify inhibitory substances if required, would enable
evaluation of both the technical and particularly the economic factors to
be applied in an actual situation. Utilization of crude copperas (ferrous
sulfate heptahydrate) as an iron(ll) source and spent ferric chloride
etchant or the copper cement derived therefrom as a catalyst source
should be evaluated in such a program.
-------
INTRODUCTION
The purpose of the work described in this report was to evaluate the
parameters which affect both the extent of the dilute solution reduction
of nitrate ion and the product distribution in the copper catalyzed
reaction with ferrous iron under basic conditions. Included in the
study were the mole ratio of Fe++/^0j~, initial pH, buffering agents,
role of the accompanying anions, catalyst effectiveness, and reaction
time. Means of limiting the amounts of soluble materials introduced in
the process and optimization of the process were investigated as well.
Preliminary attempts were made to apply the process to removal of nitrate
ion from both a potable water supply and municipal wastewater.
At present, the only practical method for removal of nitrate ion from
dilute aqueous solution is biological denitrification such as is found
in trickling filters and activated carbon columns. Ion exchange is an
efficient process but its applicability is seriously limited because of
the relatively high cost and fouling of the resins. A chemical treat-
ment process for removal of nitrate ion would be most advantageous
because of its flexibility and inherent reliability. To be practical,
such a chemical treatment would have to convert the nitrate nitrogen
into an easily removed or biologically near-inert form while intro-
ducing a minimum amount of less objectionable chemical species into the
treated water. Of equal importance in making chemical treatment a
practical process is that the cost be low.
The recent feasibility study (Gunderloy, 1968) conducted in the Rocket-
dyne laboratories for IWQA indicated that the copper catalyzed ferrous
iron reduction of nitrate in dilute basic solution was potentially
capable of development into both an efficient and economic chemical
denitrification process. Indicated in the feasibility study were the
following points: (l) complete reduction of nitrate could be achieved,
(2) nearly half of the nitrogen was neither N0-j~, N02~ nor NH-j in solu-
tion (the missing-N was believed to be N2 or N20), (3) under anaerobic
conditions the extent of nitrate reduction and the product distribution
were pH dependent, (4) the pH of the medium changed markedly during the
course of reaction, (5) reduction occurred under aerobic conditions only
at high pH, (6) catalyst type and concentration were important variables,
(7) phosphate inhibited the reduction of nitrate, (8) carbonate altered
the nitrogenous product distribution, and (9) lime could be used in the
process for pH adjustment without difficulty. At the end of the feasi-
bility study it was apparent that considerable process oriented research
would be required to evaluate the potential of this denitrification
process. Possible areas of application include (l) reclaiming municipal
waste waters to permit storage without excessive growth of algae,
(2) reclaiming industrial waste waters where co-pollutant species are
toxic in biological denitrification, and (3) purifying potable water
sources contaminated with nitrate and/or nitrite to eliminate diseases
-------
associated with these ions such as methemoglobinemia in human infants
or goiter and bloating in livestock.
The investigation reported herein was divided into two phases. The first
and larger of the two phases was a stepwise study of the effects of each
of the variables on the reaction conducted essentially as jar tests
(although somewhat more sophisticated as required). In the second phase,
the process was carried out in a flow system in an effort directed
toward optimization of variables and to gain an insight into problems
unique to the dynamic process. As a practical matter, work on the second
phase was initiated before completion of the first phase. The individual
tasks of Phase 1 outlined below as discreet areas of study are in many
instances inter-related.
1. Identify the "missing"N" and obtain a nitrogen material balance in
gaseous, liquid, and solid phases of the reaction mixture.
2. Study the effect of varying Fe++/1lQ-i~ ratio at constant NO-j -N
concentration.
J, Compare the effect of the anions on the process when the ferrous
source is either FeSO/j or FeClg.
4. Study the effect of initial pH and in addition the effect of buffers.
5. Determine the distribution of sulfate and chloride between the solid
and liquid phases of the reaction mixture.
6. Seek more effective and more economical catalyst and study catalyst
recycle.
7. Seek means of overcoming the deleterious effects of phosphate, car-
bonate, or other species present in waters to be treated.
8. Determine minimum reaction time.
9. Evaluate in preliminary fashion the effects of naturally occurring
(and possibly unknown) species in both potable and municipal waste
waters.
-------
EXPERIMENTAL
Reagent solutions were prepared from A.C.S. Reagent grade chemicals with
helium deaerated distilled water and either used immediately or stored
under 0-25 psig of helium in borosilicate glass pressure burets (stone,
1948) fitted with Teflon needle valves. During the course of the work
it became evident that the sodium hydroxide stock solution was attacking
the glass container and a stainless steel and polypropylene system was
substituted. A calcium hydroxide slurry was maintained in a magnetically
stirred polyethylene bottle under nitrogen. The concentrations of the
reagent stock solutions used in the reported study are listed in Appendix
I.
A typical experiment is described below. A 200 ml volumetric flask was
placed under a bell jar which was swept with a stream of helium, a
portion of which was directed to the bottom of the volumetric flask.
The helium was saturated with water vapor by passing it through a fritted
tube in a 45-liter bottle which also served as the supply reservoir for
deaerated water. Approximately 190 ml. of the deaerated water was added
to the volumetric flask through a hole in the top of the bell jar where
the helium stream exited. If required an aliquot of the Na2HP04 stock
solution (and/or other inhibitory substances) was added to the reactor
from either a syringe filled under helium in a glove bag or from a
pressure buret (5 ml graduated in 0.01 ml) without contact with atmos-
pheric oxygen by inserting the syringe needle or buret tip well into the
neck of the reaction vessel. The FeS04 or FeClQ and the CuS04 (or other
catalyst) was added next from a pressure buret in a similar manner* At
this time, a glass electrode was inserted into the helium stirred solu-
tion. An aliquot of the NaOH [or Ca(OH)2l stock solution, measured into
a syringe in a helium-filled glove bag, was added rapidly to the mixture.
After approximately 5 minutes, the pH of the mixture was recorded as the
initial value. An aliquot of the KNO^ stock solution was added from a
pressure buret and the mixture was brought to volume with additional
deaerated water. A Teflon-coated magnetic stirring bar was added, the
glass stopper was inserted, and the flask was placed in an 85° F c-ons-tant
temperature bath. The mixture was stirred magnetically for 24 hours
during which time the appearance of the suspended solids typically
changed from flocculant blue-green through gray-green to a readily
settleable black magnetic powder. After replacing the reaction vessel
under the helium-swept bell jar, the stopper was removed and the final
pH was determined. The mixture was transferred to a 200 ml screw cap
centrifuge bottle under helium and centrifuged for ca. 5 min« at 1725
rpm. The supernatant solution was analyzed immediately for nitrate,
nitrite, ammonia, and whatever other constituents of interest that
might be present as indicated in the section on analytical support.
In some experiments (as noted) the order of addition of the reagents was
varied to determine any effects on the reaction. In the typical
-------
experiment as described above the nitrate solution was added last because
under these conditions a more stable initial pH value could be obtained.
Certain experiments were carried out in the typical manner using larger
volumes of solution to permit multiple sampling for analyses at varying
reaction times. Under these circumstances a helium stream was directed
into the mixture during sampling to prevent entry of air. In other
experiments wherein continuous addition of reagents from a syringe pump
was used, a continuous sweep of water-saturated helium was used to main-
tain anaerobic conditions.
The experiments concerned with collection and mass spectrometric analysis
of the gaseous-N compounds required only a slight modification of the
closure for the reaction vessel. The conventional glass stopper was
replaced by another which extended into the neck of the flask to minimize
the ullage and which was equipped with a Teflon needle valve to permit
removal of the contents directly into a vacuum line. The ullage volume
2.53 cc was determined from the weight difference between the completely
water-filled apparatus and the apparatus filled to the 200 ml calibration
mark. An initial attempt was made to degas the solution by boiling off
the dissolved gases into the vacuum line and collecting the material
passing a -143 C trap for mass spectrometric analysis (Expts. 16, 18 and
25). This procedure was too lengthy and accordingly was modified by
inverting the reactor and discharging the entire reaction mixture into
an evacuated 1000 ml receiver where degassing was effected rapidly by
the freeze-thaw technique.
In the continuous flow experiments with solids retention (see Fig. l),
anaerobic conditions were maintained by the helium stream introduced
through G to stir the reaction mixture. The effluent helium from P was
lead over the gas intakes of the variable flowrate constant head siphons
used to supply reagent solutions which were introduced through the rubber
septum S by means of 3" #22 needles. In the series of annular chambers
rapid mixing was attained in A (upflow) and B (downflow) with moderate
agitation in C (downflow) with an upflow return to A from the bottom of
the reactor. Chamber D was an unstirred upflow clarifier which accounted
for 230 ml of the total 465 ml reactor volume. The effluent flowed from
0 to a sample collecting reservoir outside the constant temperature bath
L in which the reactor was immersed. The pH was measured with a glass
electrode introduced through E. The port R was used to introduce a small
(0.1 ml) cup in which the ferrous and cupric hydroxides were coprecipi-
tated before falling into the stirred chambers A and B.
The second design for a continuous flow reaction was simply an 8 mm I.D.
tube with a 15 cm section (7.5 nd) with inlets for introduction of
ferrous and cupric sulfates and sodium hydroxide near one end and a pH
electrode followed by an inlet for potassium nitrate at the other just
preceding a 121 cm thermostatted reaction section (60.5 ml). The reactor
was stirred by a stainless steel spiral (2 cm pitch) countercurrent to
-------
WATER
LINE
CROSS SECTION OP FLOW REACTOR
FIGURE 1
-------
the liquid flow at a speed just sufficient to suspend the precipitate.
The reagent inlets as well as the effluent in a sample collecting reser-
voir were maintained in a helium atmosphere. The FeS04 and NaOH solu-
tions were introduced with syringe pumps into the CuS04 solution (ca. 2Q%
of total flow) which was delivered from a variable flowrate constant head
siphon. After pH measurement of the preformed mixing hydroxides suspen-
sion, the KNO^ solution was introduced from a similar siphon. A periodi-
city in the pH indicated that the flowrates used (17 ml/hr. for CuS04)
were too low to obtain constant flow from the siphon as a result of the
periodic introduction of the helium bubbles into the siphon reservoir.
-------
DISCUSSION
In the discussion below of the effects of each of a number of variables
on the extent of reduction of nitrate ion and the resulting nitrogenous
product distribution, the pertinent experiments are grouped together.
The experimental detail for each experiment will be found in the appro-
priate appendix. As the discussion is arranged the experiments are not
necessarily in chronological order but this order may be deduced from
the numerical order of the experiments. In many cases, an experiment
will appear in more than one appendix for convenience in making compari-
sons if the data are pertinent to more than one of the variable parame-
ters.
In all of the experimental work a constant temperature of 85° F was used
first, because this appeared to be a realistic upper limit for waters
in which this denitrif ication process might find application, and, second,
because more rapid reaction at this upper temperature limit would expedite
the investigation of the numerous variables.
IDENTIFICATION OF THE GASEOUS-N SPECIES
In the earlier feasibility study (Gunderloy, 1968 ) an unsuccessful
attempt was made to identify the "missing-N" by mass spectrometric
analysis of the gaseous nitrogen products as a part of obtaining an
overall nitrogen material balance. In that experiment the wet analyses
for nitrate, nitrite and ammonia accounted for 96.4$ of the nitrogen
effectively precluding the mass spectrometric data. Additional experi-
ments reported herein have established that the conditions used were far
from optimum for reproducible formation of gaseous-N compounds. In the
successful experiments (tabulated in Appendix II) the initial pH was
lower (7.8 - 8.4 vs. ll) than was used in the early work. In the final
duplicate experiments 64 and 59$ of the NO^~-N was converted to a mix-
ture of N20 and N2 (N20/N2 = 1.46) with an overall material balance of
95 and 98$, respectively. In numerous subsequent experiments the yield
of the gaseous-N (determined by difference) was increased but these
experiments were not set up to collect and analyze the gaseous-N com-
pounds; it is not known how the ratio N20/N2 varies within the limits of
our experimental conditions or whether it is invariant.
The identification of ^0 as a product of the denitrif ication process
poses the question of air pollution if large quantities of this nitrogen
oxide were generated. The two primary photochemically induced decompo-
sition reactions of 0 are
N20 « N2 + 0 (1)
and
N20 -» NO + N (2)
-------
of which the first accounts for 88$ (McNesby, 1964). Secondary reactions
such as
N + 00 -» NO + 0
(3)
1)
and
NO + 02 -» N02 + 0 00
would form some pollutants but the concentration would probably be too
small to cause concern (Bates, 1967).
THE EFFECT OF INITIAL pH
The initial pH of the reaction mixture would be expected to influence
strongly the denitrification reaction since the reducing power of
ferrous iron increases as the ratio of Fe+++/Fe++ decreases. As a
result of the relative solubilities of Fe(OH)3 and Fe(OH)2 ferrous ion
becomes a more powerful reducing agent as pH increases. Complete reduc-
tion of nitrate to ammonia without catalysis has been demonstrated in
28$ sodium hydroxide solution (Carsley, 1930) but proceeds only to the
extent of 1-3$ in 3$ sodium hydroxide unless catalysts such as silver
or copper are used (Szabo, 1951).
In the feasibility study (Gunderloy, 1968) the copper catalyzed reduc-
tion of nitrate ion in dilute solution was shown not to proceed at
initial pH Ł 6. With increasing pH the extent of reaction increased to
near quantitative but products other than ammonia were observed. The
data were not without anomalies, however, prompting the reinvestigation
of the effect of pH reported herein.
At the time that most of the experimental data were being obtained, the
fact that OH /Fe mole ratio would be a better experimental parameter
than pH was not fully appreciated. The flatness of the potentiometric
titration curve for FeSO^ or FeCl2 causes pH to be a rather insensitive
parameter within the range of 0.25 - 1-75 for the OH /Fe mole ratio.
At low mole ratios the lack of buffering capacity resulting from the
limited amount of Fe(dH)2 permitted the pH to fall to a low value where
the reducing power of the ferrous iron decreased rapidly. As a result
both NO-j and Fe remained unused in many experiments. Conversely, at
high mole ratios the buffering capacity of the Fe(OH)2 maintains the pH
at a value favoring complete reduction to ammonia.
The pertinent experiments summarized in Appendix III all were run at
10 ppm NO-j -N with a Fe +/N03 m°le ratio of 8, a OH~/Fe++ mole ratio
of 1.45 - 1.46, and 5 ppm Cu^+ catalyst with both metal ions derived from
the respective sulfates. If the data from experiments 1-4 are not
included (initial pH values not reliable due to variable rates of intro-
ducing the NaOH solution to the nitrate-containing solution), a decreasing
10
-------
trend in the extent of nitrate ion reduction with decreasing initial pH
is clearly evident. After 24 hours at pH 11, no nitrate remained, only
about \% of the original amount at pH 8.7, 1% at pH 8,0, and 39$ at
pH 7,k, Particularly significant is experiment 273 which was analyzed
periodically during the 24-hour reaction period. Comparison of the
sample pH and the analytical data (See Appendix XXl) indicates that the
rate of conversion of nitrate to nitrite drops off sharply in the pH
range 7.0 - 7»3. The drop in pH noted during the first hour of reaction
was reversed during the second hour was presumably due to formation of
nitrite (basic in solution) which would overcome the acidifying effect
of the oxidizing iron (Equation 5)-
Fe(OH)2 + H20 -» Fe(OH)3 + H+ + e" (5)
Further reduction of the nitrite would result in the observed drop in pH
and formation of gaseous-N compounds. This rationale appears somewhat
less plausible when the reduction of nitrite is considered since the
same pH-time relationship was observed (see Beduction of Nitrite).
Although the experimental data offer no definitive proof, they are
consistent with the postulated formation of ferrous hyponitrite, an
unstable compound reported to form at pH 6.5 ~ 7«0 and to decompose
immediately with evolution of an unknown gas (Polydoropoulos, 196l).
With respect to the effect of pH on the yield of ammonia the experiments
summarized in Appendix III clearly show the trend to decreased yield
with decreasing pH but other factors not indicated definitively by these
data effect the yield as well. These factors are (l) the aforementioned
OH /Fe mole ratio which, if too high, maintains a pH favoring ammonia
formation, and (2) catalyst aging which will be discussed later also
favors formation of ammonia. In the uncatalyzed reduction of nitrite,
ammonia is the exclusive product at high initial OH /Fe mole ratio
(e.g., 1.82, see Reduction of Nitrite).
EFFECT OF THE BASE - NaOH VERSUS Ca(OH)2
In the majority of the experiments sodium hydroxide was used to adjust
the pH of the reaction mixture simply because of the ease of introducing
a solution compared to that of a slurry or a solid as is required with
calcium hydroxide. However, in one series of experiments a direct com-
parison of the effect of substituting a 1% Ca(OH)2 slurry for 0.98 M
NaOH was evaluated. The data tabulated in Appendix IV indicate that the
conversion of nitrate to reduced species was approximately 15$ less with
Ca(OH)2 than with NaOH but the yield of gaseous-N compounds was identical
within the experimental error. Thus, it appears that the presence of
calcium ion inhibits the nitrate to nitrite reduction step only.
EFFECT OF Fe++/N0 " MOLE RATIO
The reduction products observed in the denitrification process, NOg ,
NgO, Ng and NH_, theoretically require 2,4,5 and 8 moles of Fe++ per
11
-------
mole of NO-7 . However, the pH range in which the reduction reactions
occur to give these products falls within the range favoring formation
of ferrous ferrite (magnetite), FejO^, namely from pH 6.2 - 11.4 (Krause,
1955). Thus, the theoretical Fe+ /NO^" ratios are J, 6, 7-5 and 12,
respectively. It is immediately apparent that in order to minimize the
amount of soluble counterions (S0^= or Cl~) introduced into the receiving
waters the minimum amount of ferrous salt should be used in the denitri-
fication process.
Considering first the information derived from experiments using sulfate
only as the counterion (Appendix V), the conversion of NO^" to gaseous-N
appears not to change greatly from Fe++/N0^~ ratios from 12 to 5 (60 ± 5$)
and then decreases as the ratio drops to 2. The yields of gaseous-N
compounds imply only a slight downward trend over the Fe++/NO-7~ ratio
range 12 to 4 becoming more apparent as the ratio drops to 2. This con-
stancy in the yield of gaseous-N compounds over the wide range of
Fe++/NO-r~ ratios is interpreted to mean that the formation of gaseous-N
compounds is independent of the amount of ferrous iron used. From an
inspection of the data in Appendix V, it is evident that at high OE~/Ye++
mole ratios with sufficient ferrous iron ammonia is the major product
with concomitant decrease in gaseous-N products.
When ferrous chloride was used as the source of the reducing agent (see
Appendix Vl), ammonia appears as the major product even though the ratio
On~/Fe++ was no greater than 1.5^. As a_function of the Fe++/N0^" ratio
no trend is evident in the amount of NO^ converted to gaseous-N products.
Although the yield of gaseous-N products appears to increase with decreas-
ing Fe /NO-j ratio, the simultaneous decrease of the OH /Fe ratio might
better account for this observation. The seemingly anomalous results
among the experiments at a Fe /NO-j ratio of 8 is a result of the use
of CuSOzj catalyst in lieu of CuCl2 and, accordingly, will be discussed in
the section concerned with the effects of anions on the denitrification
process.
CATALYST EVALUATION
The reduction of nitrate with ferrous hydroxide has been studied quite
extensively as an analytical method and as such the experimental conditions
were selected to give quantitative reduction to ammonia. The catalysts
most commonly used are copper and silver salts both of which were used in
the feasibility study (Gunderloy, 1968). Silver salts were not further
considered in the present study for economic reasons.
The effect of the copper catalyst concentration was studied in more
detail than in the initial study (Gunderloy, 1968) wherein the concentra-
tions, 1, 5 and 10 ppm Cu were evaluated only at pH 11, a value far too
high for significant formation of gaseous-N compounds. Copper concentra-
tions from 0.05 ppm (resulting from a copper impurity in the reagent
ferrous sulfate) up to 5 ppm were evaluated at a ratio OH /Fe = 1.46
12
-------
(pH near 8.3). The data, listed in Appendix VII, indicate that nitrate
reduction occurred effectively at copper concentrations as low as 1 ppm.
The product distribution appeared not to change with copper concentration
in the range of 1 - 5 ppm where 65 - 15% of the initial NO^~-N was con-
verted to gaseous-N compounds and about 20% to ammonia.
The effect of the order of addition of the catalyst with respect to the
other reagents was studied briefly (Appendix VIIl), With all other con-
ditions held constant in 24-hour experiments, the effect on product
distribution of coprecipitation of the copper and iron hydroxides as
opposed to addition of the cupric salt to the precipitated ferrous hydrox-
ide was evaluated. A somewhat greater reduction of nitrate was observed
with the coprecipitated hydroxides--91$ vs. 81$. Individual differences
in the values of nitrite and gaseous-N products are not significant but
presumably reflect only the lowering of pH to a value where reduction
stopped. It may be noted that the summation of the nitrite and gaseous-N
is constant for each set of experimental conditions, i.e., ca. 6jft> of
the original nitrogen with coprecipitated hydroxides and ca. 52$ with
catalyst added last. The amount of ammonia produced was nearly the same
after 24 hours for either set of experimental conditions but with normal
addition of catalyst it was formed during the first hour.
Because of the obvious economic advantage the possibility of recycling
the catalyst was investigated. The absence of copper in the solution
after reaction indicated that all of the catalyst had coprecipitated with
the black magnetic Fe-jOjj. The first attempt at recycling the catalyst
was simply to add the black solids from a prior reaction to another
reaction mixture in lieu of the copper salt. The extent of nitrate
reduction was a little less but roughly comparable to that observed in
the reaction using fresh catalyst. However, the amount of ammonia
formed was approximately twice the usual quantity with a corresponding
decrease in the amount of gaseous-N compounds formed. Confirmation of
this observed increase in the yield of ammonia in the presence of the
copper-containing Fe-iO^ was obtained in a flow reaction (to be discussed
in more detail in another section). The data are tabulated in Appendix
IX. Included in Appendix IX is data from an experiment where the cupric
hydroxide had been coprecipitated with ferric hydroxide and heated at the
boiling point for 24 hours in an attempt to form cupric ferrite (Forestier,
1939? Longuet, 1941). Apparently, under the conditions of high dilution
used, the black ferromagnetic Cu(Fe02)2 did not form and the cupric
hydroxide dehydrated to the oxide (Sidgwick, 1950). This accelerated
aging of the catalyst resulted in lowered reduction of nitrate and a
high yield of ammonia.
Since the aged catalyst in the Fe-iO^ promoted formation of ammonia the
feasibility of extracting the copper from the solid was investigated
briefly. The acidic ferrous sulfate reagent solution was used at 100-
fold dilution to leach the precipitate under anaerobic conditions. The
concentration of copper in solution was determined periodically and
13
-------
indicated that approximately 40$ of the copper is recovered within 0.25
hour and over 80% in 24 hours (Appendix X). Although not definitively
established experimentally the variability of the copper concentration
may result from competition of reactions which decrease (equations 6 and
7) and increase (equation 8) the soluble cupric ion. The last reaction
Cu++ + Fe^- Cu + +Fe+++ (6)
2Cu+ - Cu++ + Cu
2Cu + 4H+ + 02 - 2Cu++ + H20 (8)
would be expected if air leaked into the system.
A series of experiments was conducted using elemental copper as the
catalyst source (presumably the surface oxide is the actual source). Two
forms of the metal, No. 1 shot and 150 mesh powder, were used to evaluate
the effect of surface area. The surface area of the powder was about 68
times that of the shot. In some experiments the copper was contacted
with the ferrous sulfate for an hour before addition of the base and
nitrate while in others all the reagents were added in rapid sequence.
The data, summarized in Appendix XI, show that the reduction of nitrate
was poor in the presence of shot but very good with the powder. Visual
observations of the color of the precipitated hydroxides as a measure
of catalyst concentration correlate well with extent of nitrate reduc-
tion; the color with shot was almost white and with powder a pale to dark
green. In addition, the color of the precipitate in all the experiments
darkened with time, but only in those using powder did the black Fe^O^
form. In the experiments using powder the high conversion of nitrate to
ammonia (47/0 and "to nitrite (50$) is not surprising at the high OH~/Fe +
ratio used (ca. 1,85, cf, data in Appendix III).
A atill more economical source of copper than scrap metal ia the copper-
cement recovered during regeneration of ferric chloride copper etchants.
The cement and spent etchant solution were analyzed for copper and iron
content but were not evaluated in denitrification experiments because the
samples were received just prior to the termination of the experimental
work. The spent etchant analyzed 3*7$ copper and 10.2$ iron while the
cement (after mechanical separation from 0.63 of its weight of tramp
iron) analyzed 27,4$ copper and 21.6$ iron.
Only a preliminary effort was made to evaluate other metals an catalysts.
The potential catalyst must be capable of entering into redox reactions
and should be less expensive than copper. These criteria limit the
potential candidates severely so that only lead was evaluated. Lead has
been reported to catalyze the oxidation of ferrous hydroxide by oxygen
(Krause, 1937) beyond the ferrous ferrite stage. Although the cost of
lead is only about one-fourth that of copper the greater atomic weight
14
-------
offsets most of the economic advantage of lead« Under the conditions
chosen for the preliminary experiments the reduction of nitrate was poor
and the principal product was ammonia (see Appendix VIl) so that evalua-
tion was not pursued further.
EFFECTS OF ANIONS
Since the denitrification process is the reduction of nitrate ion by
ferrous hydroxide, it may be assumed initially that only hydroxide ion
exerts an important effect on the course of the reaction. That this
assumption is naive was shown by a comparative study of the two economical
sources of ferrous ion, ferrous sulfate and ferrous chloride. The major
experimental effort was carried out using ferrous sulfate, which currently
in many locations is a surplus commodity requiring disposal. Dumping at
sea is being used where feasible as a more economic means of disposal
than recovery of the sulfuric acid and iron oxide values by existing
technology. The simpler recbvery of hydrochloric acid and iron oxide
values from ferrous chloride will doubtless mean that this salt will not
be as cheap a source of ferrous ion as the sulfate. However, the rapidly
growing importance of hydrochloric acid for pickling of steel (Anonymous,
1969) dictated that ferrous chloride should be evaluated as well. In
addition to the counter ions introduced with the iron and copper salts,
the effects of other anions commonly found in waters requiring denitri-
fication, such as phosphate, carbonate and silicate, were evaluated.
In all cases the system containing only sulfate in addition to hydroxide
and nitrate (and its reduction products) is used as the basis for com-
parison of the effects of other anions.
The mechanism by which the various anions influence the reaction rates
and/or the distribution of reduction products of the nitrate ion would be
of considerable importance to the understanding of the denitrification
process* To be considered is the ability of the anion (l) to buffer the
reaction medium, (2) to precipitate or complex either the catalyst or
the reducing agent or both, and (3) to alter the electrical surface
charge of the precipitated iron and copper compounds. Available experi-
mental data bearing on these considerations will be discussed within the
•tudy of the effects of the individual anibns.
Comparison of Chloride and Sulfate
An inspection of Appendices V and VI reveals that a direct comparison of
the effects of chloride vs. sulfate on the distribution of reduction
products is not possible because of differences in the OH'/Fe** ratio at
a given Pe^/NOj" ratio, However, where differences in the OH"/Fe
ratio at a given Fe^/NO*" ratio are not too great a valid comparison of
the effects of the individual anions can be made. For example, where
Fe^/NO*" - 8 and 1,46 < QH"/Fe++< 1,54 using NaQH for pH adjustment,
the yield of gaseous-N products was decreased from 77# (with §k% NO-j"
reduction) in the sulfate system to 1% (with 56# NOj" reduction) in the
15
-------
chloride system. The same effects were noted at Fe /NO-i = 5 where
QEr/Fe++= 1.72 in the sulfate system and QH~/Fe++= 1.51 in the chloride
system, i.e., gaseous-N yield 62$ (with 81% NO-j" reduction — sulfate)
and 31$ (with 36$ N0^~ reduction—chloride). With ferrous chloride the
apparent increase in yield of gaseous-N products as Fe++/NOT~ decreases
can be better explained as the result of the decrease in OH/Fe ratio.
In contrast in the ferrous sulfate experiments where OH~/Fe++ was held
constant as Fe+YN03~ decreased > tne 7ield of gaseous-N products falls
off significantly near Fe++/N03~= 3. Therefore, it appears that at any
given ratio of Fe/N(H~ and ffir/Fe"H' a greater yield of gaseous-N products
will be obtained using ferrous sulfate rather than ferrous chloride.
In a series of experiments at constant Fe++/NQ3~ = Q and OH /Fe++= 1.45
(Appendix IV ) where a variable mixture of ferrous sulfate and chloride
was used with a copper sulfate catalyst, the yield of gaseous-N products
drops only 10$ on replacing three-fourths of the ferrous sulfate with
ferrous chloride. The dramatic effect of sulfate on the yield of gaseous-N
products is evident from the experiments where only 7«6 ppm S04= (from
the CuSOjj catalyst) increased the amount of gaseous-N by an order of
magnitude over the sul'fate-free system (45$ vs. 4$). Thus, a relatively
small concentration of sulfate (ca. 100 - 150 ppm) is all that is
required to permit the denitrif ication process to proceed satisfactorily
in waters of high chloride concentration.
The amount of sulfate precipitated by the basic iron compounds decreased
from 5 - 10$ to less than 3$ as the OH'/Pe"1"1" ratio increased from 1.46
to 1.89 when all of the iron was supplied as sulfate and pH was adjusted
with NaOH (Appendix III). When only one-fourth of the iron was supplied
as sulfate and the remainder as chloride, 13$ of the initial sulfate
precipitated with the black iron compounds. In contrast, the amount of
chloride which precipitated with the iron compounds increased rather
than decreased with increasing anion concentration but the maximum was
only about 5$. These maxima were increased when calcium hydroxide was
substituted for sodium hydroxide for pH adjustment but only to 19$ for
the eulfate and 6$ for the chloride precipitation.
A possible means of limiting tbe amount of anions introduced into the
treated water has been reported (Gotz, 1963). The cyclic process which
utilizes concentrated (6 - 10$) solutions is based on the equations (9)
and (10). The separated gypsum may be disposed of as solid waste (or
+ CaClg - CaSO^ + FeClg (9)
FeCl2 + Ca(OH)2 - Fe(OH)2 + CaClg (10)
•old) while the separated Fe(OH)2 is used in the denitrif ication process.
Sufficient sulfate would accompany the Fe((H)2 to permit efficient deni-
trif ication.
16
-------
In keeping with the greater solubility of the basic iron chloride as com-
pared with the sulfate at a given pH, potentiometric titration of ferrous
sulfate shows that in the presence of chloride ion the precipitation of
hydroxide occurs at a higher pH (Levin, 1956). Potentiometric titrations
of dilute solutions of ferrous sulfate and ferrous chloride gave almost
superimposable curves (Figure 2) except that initial precipitation of the
hydroxide occurred at a lower pH in the sulfate system. The addition of
5 ppm Cu++ accentuated initial precipitation at lower pH in the sulfate
system more so than in the chloride system (Figures 2 and 3) as a result
of the oxidation of some of the iron to the ferric state (Arden, 1950)»
It is postulated that the lower pH of the basic sulfate as compared with
the basic chloride allows intermediate formation of ferrous hyponitrite
(Polydoropoulos , 196l) in this system while intermediate hydroxylamine
is formed at the higher pH in the chloride system. Presumably both of
these compounds arise from a common intermediate, nitroxyl (Brown, 1967),
and then further react to give the principal observed products, N2 and
from FeNgC^ decomposition and NH^ from NH20H reduction.
Effect of Phosphate
In the feasibility study (Gunderloy, 1968) the value of the experiments
concerned with the effect of phosphate on the denitrif ication process
was limited since a number of variables were changed simultaneously.
These experiments did serve to show (l) that phosphate had a strong
inhibitory effect on denitrif ication, and (2) that the inhibitory effect
could be overcome either by increasing the Fe^/NO*" ratio or substituting
silver for the copper catalyst. It is now known that the experimental
conditions used were such that little, if any, formation of nitrogen or
nitrous oxide would be expected.
A number of experiments were run at selected phosphate concentrations
and varying pH to delineate the extent of the inhibition and the influence
that phosphate might have on product distribution. If the series of
experiments at a constant 2 ppm POj^-P is considered first (Experiments
229 - 238, Appendix XIIl), it is apparent that over the pH range 7.3 -
9.5 the reduction of nitrate is significantly decreased. The extent of
reduction was only about lOji at the lower pH (OH'/Pe**- 0,63) and 28# at
the upper pH (OH"/Fe++- 1.80) with a maximum of 38# reduction at an inter-
mediate pH of 7.7 (QH'/Fe4*- 1.63), The principal reduction products at
CE'/Fe"1"1" * 1,29 were comparable amounts of gaseous-N compounds and ammonia
while at the QET/Fe++ ratio rose about 1,63 nitrite ion appeared in
increasingly greater proportion with a lesser amount of ammonia. In the
experiments where phosphate concentration was varied at a relatively
constant pH • 8,0 - 8.3, a decrease in the amount nitrate ion reduced
was noted as PO^-P concentration increased from 0,5 to 10 ppm,
The inhibition of the denitrification process might be caused either by
(l) precipitation or complexing of either the copper catalyst or the
ferrous reducing agent, or (2) superposition of the phosphate buffer on
17
-------
00
ml 0.988 M NaOH
FIGURE 2. Potentiowetric Titration Curves for FeSO, , FeSO.-CuSO, , FeCl_, and FeCl - CuCl
-------
FeSOj,-CuCl2 (5 ppm Cu)
ml 0.988 M NaOH
Figure 3. Potentiometric Titration Curves for FeSO -CuCl , Fed -CuSO. , and FeSO -CuSO -CaSO,
-------
the ferrous hydroxide buffer to give an unfavorable pH. Little reliable
quantitative data is to be found on either the solubility of phosphates of
each of the valence states of copper and iron or the complexing strength
of phosphate with these metal ions. If under the experimental conditions
of the denitrification process it is the copper catalyst alone which is
rendered inactive, then the reduction of nitrite ion which requires no
catalyst (Brown, 196?) should proceed unchanged in the presence of
phosphate. That it did not (vide infra, Reduction of Nitrite) implied
that either an iron-phosphate interaction or a buffering effect (or both)
were mainly responsible for the inhibition.
The decolorization of ferric solutions by complexing with phosphate ion
is well known in analytical chemistry. Data relating to phosphate
complexing with ferrous ion is more obscure but the known decrease in
the oxidation potential of the Fe++-Fe+++ (Van Vazer, 1958) is clear-
cut evidence that ferrous ion activity is diminished more than that of
ferric ion. Further data bearing on the relatively greater interaction
of phosphate with ferrous rather than ferric ion is found in the phos-
phate inhibition of magnetite, Fe^O^, formation in the air oxidation of
ferrous hydroxide (Krause, 1963). Magnetite forms in the pH range
6.2 - 11.4 between the isoelectric points of the negatively charged
ferric (5«4) and the positively charged ferrous (ll.5) hydroxides. Com-
plexing of the multivalent phosphate anions with the positively charged
ferrous hydroxide would be expected to change the isoelectric point and
prevent interaction of the two iron hydroxides. This same inhibition of
Fe^Oij formation was found in the denitrif ication process (with only a
few exceptions) when phosphate was present.
The possible superposition of the buffer actions of phosphate and ferrous
hydroxide is apparently overshadowed by a drastic lowering of the pH at
which the hydroxide precipitates from solution (Levin, 1956). Potentio-
metric titration curves at low concentration of ferrous sulfate in the
presence of copper catalyst and phosphate exhibit a greater lowering of
the pH than did sulfate (of. Figure 4 with Figure 2). The fact that
only partial reduction of nitrate occurred implies that the denitrifi-
cation reaction proceeded normally in a given pH range in the presence
or absence of phosphate but that in the presence of phosphate a lesser
amount of reaction was required to change the pH to a value where reaction
stopped. It should be noted on the one hand (Appendix XIl) that when the
CH/Fe^ ratio was low the final pH had dropped to about 5-6 whero
denitrification does not occur in the absence of phosphate (Gunderloy,
1968). On the other hand at high CH"/Fe++ ratio, the pH had risen
significantly higher than the initial value and nitrite ion had become
the major product*
Since the presence of phosphate results in precipitation of the basic
ferrous salt at an even lower pH than when sulfate is present in the
system, the addition of phosphate to this system should
20
-------
IO
FeSO^-CuSOj. (5 ppm Cu) - NaHPO/, (2 ppm P)
FeS04-Na2HPOA (2 ppm P)
L— NaHPOi, (2 ppm P ADDED FIRST) -
FeSOj^-CuSOj, (5 ppm Cu)
(5 ppm Cu) - Na2HPOj, (10 ppm P)
(5 ppm Cu) - Na2HPOi, (2 ppm P)
Q.
ml 0.988H NaOH
Figure 4. Potentiooetric Titration Curves for
, FeSO^CaSO. -Na HPO.
-------
result in an increased yield of gaseous-N products. This postulate was
verified experimentally in two series of experiments at different
initial OH~/Fe++ ratios and variable phosphate concentrations up to
5 ppm PO^=-P (Appendix XIIl)0 With increasing phosphate concentration
in both series of experiments the following observations were made:
(l) a decrease in the amount of nitrate reduced, (2) a simultaneous
increase in yield of gaseous-N compounds, and (3) a decrease in the
amount of ammonia (the major product) formed which was more pronounced
in the lower pH series. As a practical matter the beneficial effect
of phosphate (comparable to that of sulfate) on the formation of gaseous-N
products during the denitrif ication process is negated by the simultaneous
inhibition of nitrate reduction.
In order to overcome the deleterious effect of phosphate on the denitri-
f ication process, the water must be treated (l) to remove the phosphate
or (2) to render it inert with respect to the denitrif ication process.
The second approach was tried initially since it would be a simpler solu-
tion from engineering and cost considerations. The most common chemicals
for phosphate removal are iron salts, alum, and lime (Nesbitt, 1969) and,
accordingly, these materials were tried as additives. In separate experi-
ments, the addition of aluminum sulfate and ferric sulfate (Appendix XIV)
to phosphate-containing nitrate solutions failed to lessen the phosphate
inhibition of the denitrif ication process in that only 20 - 30$ reduction
of nitrate occurred. In further experiments using ferric iron (derived
from recycle of half the precipitated solids from a preceding experiment)
in conjunction with calcium hydroxide, less than 20$ reduction of
nitrate was observed with ammonia the principal product.
In the investigation of the other approach, namely separation of the
precipitated phosphate before attempting denitrif ication , the majority
of the experiments used recycled iron-containing solids in conjunction
with calcium hydroxide for pH adjustment because of the obvious economic
advantage if these reactants produced a practical pretreatment. From
the data summarized in Appendix XV it is evident that a substantial per-
centage of the phosphate can be precipitated by treatment with recycled
precipitate from the denitrif ication process. A problem seemingly
inherent in the process is that at low level of phosphate (ca. 0,5 ppm P)
the black Fe^O^ forms in the denitrif ication process, and this precipi-
tate is not as efficient at removing phosphate as is the green Fe^QHjg
which forms at higher phosphate levels (ca. 1 ppm P). Near the optimum
range (k - 6) for phosphate precipitation by ferric iron (llecht, 19?0)
the black Fe-jO^ removed about 80$ of the initial 10 ppm POj^-P while
the green Fe^(OH)g effected quantitative removal. Increased bascity
(up to pH 9) had little effect on the amount of phosphate removed by
recycled
After quantitative removal of 10 ppm PO^"~-P by recycled green
nitrate ion wag readily and almost completely reduced. However, in
these experiments the formation of gaseous-N products was significantly
22
-------
less than expected, i.e., 40$ rather than 10% conversion of nitrate.
The remainder of the nitrogen was almost completely accounted for as
ammonia. No explanation for this apparent anomaly has been found.
The use of lime for phosphate removal resulted in moderately success-
ful subsequent denitrification. At lime dosages of 280 - 300 ppm CaO
yields of gaseous-N products were near 10% (with 25 - 30$ conversions
of nitrate) with the remaining nitrogen found as ammonia. At higher
lime dosages more nearly equal amounts of gaseous-N products and
ammonia were observed without significant change in the amount of
nitrate reduced.
Among the other precipitants for phosphate which were evaluated cur-
sorily were ferrous iron and aluminum both of which are reported to be
efficient at near neutral pH (Eippert, 1958). The solutions resulting
after phosphate treatment with either ferrous iron alone or in combina-
tion with aluminum did not undergo denitrification readily. In each
case only about 10$ of the nitrate was reduced. Better nitrate reduction
(about 55$) was achieved on solutions which had been treated with lan-
thanum (Eecht, 1970) and in these experiments yields of gaseous-N
products were about 15% with the remainder ammonia.
Effect of Carbonate
In the pH range (6 - 8) of interest for the denitrification process
bicarbonate ion is the principal species with carbonate ion becoming
more prevalent at higher pH values. At a carbonate concentration of
100 ppm reduction of nitrate ion was found to be 70 - 95$ in the initial
pH range 8-10 but formation of gaseous-N products was significant (55$
yield and 52$ conversion) only at the lower pH value. As noted previously
(see section on Effect of Base) use of calcium hydroxide rather than
sodium hydroxide resulted in a decrease in the amount of nitrate reduc-
tion (95$ VB. 75$) at the same 100 ppm C0j= concentration. With
increasing carbonate concentration at an initial pH near 8 the extent
of nitrate reduction decreased sharply. At 200 ppm carbonate only 25$
of the nitrate was reduced (3 out of 4 experiments) and at 300 ppm
carbonate the amount reduced had fallen to 16$. The yield of gaseous-N
products was on the average just under half of the nitrate converted.
The single exceptional experiment at the 200 ppm COj" level resembled
those at the lower concentration in that the precipitate was the black
Fe-i04 rather than the green Fe^(OH)g observed in the other experiments
with 200 (and higher) ppm CO-j". In that experiment the reduction of
nitrate was over 50$ and the yield of gaseous-N products was 77$i These
different results from allegedly replicate experiments might have been
the result of inhomogeneity in the calcium hydroxide slurry. Regardless
of the cause, the results imply that at a concentration not far below
200 ppm COj" the inhibitory effect of this ion is minor.
23
-------
An inspection of the potentiometric titration curves for
solution containing NaHCC^ (Figure 5) shows that at the low pH end the
pH at which solids precipitate is not greatly different from that of
FeSOj1-CuS04 (Figure 2) alone. At the high pH end the pH rises more
gradually but sooner than when carbonate is absent. Thus, in the
presence of carbonate one might predict that in the initial stages of
the denitrification process larger amounts of ammonia would be formed
at a given OH"/Fe++ ratio but in the latter stages formation of gaseous-N
products would not be inhibited.
It was noted in the experiments at 100 ppm CQj~ that the pH-time curve
had the same shape as in the absence of carbonate (from the initial pH
through a minimum, then a maximum, and then a decline to final value)
but the time scale was extended about 4-fold (20 vs. 5 hrs.). Compar-
ison of a series of 24 and 48 hour experiments showed no significant dif-
ferences indicating reaction to be complete in less than 24 hours. The
data are summarized in Appendix XVI.
Effect of Silicate and Other Inhibitors
Early in the experimental work it was noted that the time required for
the change in color of the precipitate from green to black increased in
a series of replicate experiments. Concomitantly a decrease in the amount
of nitrate reduced was observed (see Appendix XVTl). This time dependent
phenomenon was rationalized as an inhibition caused by silicate or borate
leached from the Pyrex glass storage buret used for 1 M NaOH. Although
the conversion of nitrate to gaseous-N products decreased the yield
remained essentially constant. The shape of the pH-time curve was
normal but with the time scale extended even more than observed when
carbonate was present.
Since silica is one of the most common constituents (at 20 - 30 ppm) of
most natural as well as waste waters, the effectiveness of selected
silicate precipitants (Appendix XVIIl) and the effect on the denitrifi-
cation process (Appendix XIV ) were investigated. Of the sulfates of
Al(lll), Cu(ll) and Fe(lll) aluminum was most effective as a precipitant
with BOJt removal of an initial 25 ppm Si02« In the aeries of denitrifi-
cation experiments the reduction of nitrate decreased, conversion to
ammonia was high and constant, and both conversion and yield of gaaeoua-N
producta decreaaed as SiOg concentration increased. The time required
for the precipitate to change color from green to black waa generally
less than half the time in the absence of ailicate (l - 2 hra. vs.
4-5 braf) in keeping with the reported catalytic activity of ailioa
in the air oxidation of ferrous hydroxide (Krauae, 1958). Theae data
are not conaiatent with the premise that ailicate leached from the glaaa
by the etrong baae waa reaponeible for Blowing of the denitrifioation
prooeaa. Thus, borate remains as the euapected inhibitor. Insufficient
time waa available to determine the possible inhibitory effect of borate
under controlled conditions.
-------
to
VJl
NaHCOj (300 ppm 003 ADDED FIRST) - FeSOl»-CuSOi, (5 ppm Cu)
NaHCOo (100 ppm C07 ADDED FIRST) ^-
J
FeSOjj-CuSOi, (5 ppm Cu)
FeSOl,-CuSO/,-NaHC03 (100 ppm 003)
TAP WATER - FeSOi-CuSOij (5 ppm
NaHCO (100 ppm ADDED FIRST) - FeSO^-CuSO^ (5 ppm Cu)
ml 0.988M NaOH
Figure 5. Potentiometric Titration Curves for
-------
REDUCTION OF NITRITE
Nitrite is the first intermediate produced in the reduction of nitrate.
Its concentration was observed to be quite variable and particularly
dependent on the OH"/Fe++ mole ratio (see Appendix III). At high
OH /Fe++ ratios nitrite together with ammonia was observed as a major
product. A brief investigation of the uncatalyzed reduction of nitrite
(Brown, 196?) was undertaken to determine whether the denitrification
process might be broken into two steps--a catalyzed NO'r to NOg" step
and an uncatalyzed N02~ to final products step—with improved effi-
ciency in converting nitrate into NgO and N2« In addition, a study of
the effect of phosphate on nitrite reduction in a copper-free system was
conducted to aid in understanding the inhibitory mechanism.
The reported uncatalyzed reduction of both nitrite and hydroxylamine
(Brown, 196?) formed ammonia quantitatively but the experimental condi-
tions were much more drastic than had been used in the denitrification
process, i.e., strong base (pH 14.2 - 14.7)> concentrated oxidant
(1400 ppm N), and either higher temperature or longer reaction time
(llO° C for 30 min. or 25° C for 1-3 days). A series of experiments
at varying OH~/Fe++ ratio (see Appendix XX) indicated that ammonia was
the exclusive reduction product at OH~/Fe++ = 1.82 with the quantity
formed limited only by the amount of ferrous hydroxide. At OH~/Fe++ = 1.38
the conversion to gaseous-N products was 83$ and at OH"/Fe++ - 1.04, 80$
with yields of 83 - 91^. Conversion to ammonia decreased with decreasing
OH~/Fe++ ratio but no hydroxylamine was found. In some of the experiments
the changes in pH and concentrations of nitrogenous species was followed
aa a function time. The pH remained on the basic side until about
90 - 95$ of the nitrite had been reduced which required about 7-8 hours
with normal addition of reagents but only 4 hours when the NaOH was
added last. The conversion to ammonia was about half again as high
(1.4 vs. 0.9 ppm Nfl^-N) -when the NaOH was added last presumably as a
result of locally high pH.
The presence of 1 ppm P04 -P had an unusual and unexpected effect on
the course of the uncatalyzed reduction of nitrite. On the one hand, a
significant inhibition of reduction was noted (only half of the nitrite
was reduced) indicating that in the catalyzed reduction of nitrate the
poor results are not due to interaction of copper and phosphate alone.
The interaction of phosphate with the iron hydroxides thus appears to be
significant in the reduction of nitrite and most probably is operative
in the reduction of nitrate. On the other hand, a marked acceleration
in the rate of reduction was effected by the presence of phosphate. At
the end of 30 minutes when the reaction was first sampled, reduction
of nitrite had proceeded to 52% while in the absence of phosphate only
13^ of the nitrite had been reduced, Approximately 2,5 hours were
required to achieve 50# reduction. The yields of gaseous-N (by difference)
were comparable. Whether the slight increase in nitrite and corresponding
26
-------
decrease in gaseous-N during the 4 hour reaction time is real or due
to analytical error was not resolved.
DENITRIFICATION REACTION TIME AND ITS APPLICATION
TO A DYNAMIC SYSTEM
In order to optimize the denitrification process and use it in practical
water treatment applications the minimum reaction time must be deter-
mined. Aside from a single experiment of 19.5 hours duration (virtually
indistinguishable from 24-hour runs) only a few experiments of 1 -2 hour
duration were conducted (see Appendix VIIl) in the early part of the
program. Each of these experiments run at the high ratio of OH"/Fe++=1.83
indicated with normal addition of reagents that a fairly rapid (< 1 hour)
reduction of nitrate to nitrite occurred followed by a slow and incom-
plete conversion to gaseous-N products.
In order to obtain further data on the time dependent course of the
denitrification process a number of experiments were scaled up to per-
mit multiple sampling for analysis. In one of these experiments all of
the reagents were added initially in the usual manner while in two others
the coprecipitated Fe(OH)2-Cu(OH)2 mixture was added continuously during
the first 4 hours of the reaction. The data summarized in Appendix XXI
for the conventional experiment confirmed the previously observed forma-
tion of ammonia and nitrite in the first hour or so followed by conversion
of nitrite to gaseous-N products during the next 4-5 hours. At the low
OH~/Fe++ ratio (l.37) used the pH fell within 3 hours to 5.72 which
slowed reduction of nitrate to a virtual standstill near 60$ conversion.
An attempt to restart the reduction by addition of NaHCO^ to raise the pH
was unsuccessful.
The continuous addition of a preformed Fe(OH)2-Cu(OH)2 mixture to a
nitrate solution resulted in formation of higher than usual amounts of
ammonia. Presumably the reduction to ammonia was favored over formation
of gaseous-N products by the presence of the black solids formed early
in the experiment (vide supra - Catalyst Evaluation). Again the conver-
sion of nitrate was poor presumably as a result of too rapid a drop in
the pH but the yield of gaseous-N products was high (87$).
The efforts to effect the reduction of nitrate in a flow system were at
best preliminary and clearly indicated that flowrates greater than 50 -
100 ml/hr. will be required to obtain reliable data. Despite the mech-
anical problems associated with control of the low flowrates the contin-
uous reactor with recirculating solids was effective in achieving reduc-
tion of nitrate (see Appendix XXII, Experiment 277A - 277V).
The pH remained at 7«84 ± 0.04 during the time samples A-G were taken.
The high concentration of NOj'-N in the effluent (ca. 65$ of influent)
was interpreted as the result of too short a retention time (< 2 hrs.)
in contact with the iron-copper solids. That the major reduction product
27
-------
vas ammonia was attributed to the presence of the aged solids and perhaps
too high a pH. Accordingly, the flowrates were adjusted downward for
samples H-K. In this series the pH was indicated to be in the range 7»3 -
7-4 but the reliability of these valves is questionable because of a sub-
sequent electrode failure. The relative flowrates of the reagents would
indicate the pH to be of the correct order of magnitude. The formation
of gaseous-N products was 10-13$ indicating the doubled retention time
was beneficial. The higher proportion of unreacted N0^~ resulted from
the relatively higher N(H~ flowrate (a decreased Fe++/N(W~ ratio) and
the high proportion of the nitrogen reduced to ammonia.
An adjustment of the nitrate flowrates and the NaOH resulted in a low
influent nitrate concentration and a higher OH~/Fe++ ratio. These con-
ditions were found to give over 90^ reduction of nitrate but the pro-
portion of ammonia remained high. Gaseous-N products increased to 28%
accompanied by 27% nitrite (which should convert to gaseous-N products
with a longer retention time). The conditions for samples M-Q gave
results similar to those obtained initially.
Of particular significance was the consistent formation of ammonia in
relatively high yield similar to that observed when recycled solids were
used as the catalyst source. To confirm this effect in the flow system
the flowrates were left undisturbed while the black sludge was flushed
from the reactor. The residual solids were dissolved in concentrated
hydrochloric acid and the reactor was flushed with distilled water. It
will be noted that in samples R-V (obtained between 6.5 - 13 hrs. after
flushing the reactor with water and an approximate eight-hour retention
time) that the NH^-N concentration rose but not as high as before (using
the N03-N concentration as an internal standard). These data when con-
sidered with the pH time dependence suggest that a longer residence time
(estimated 5 ~ 6 hours) would be beneficial in a reactor in which solids
are not retained.
Accordingly, a stirred tubular flow reactor was employed in a brief
attempt to obtain some data on this design. Limitations such as flow-
rate control set the residence time at less than one hour so that an
effort was made to effect reduction of nitrate to nitrite only. Such
data as might be obtained would have a bearing on the feasibility of
separating the denitrification process into a two-step process where
only the first step is catalyzed. The data summarized in Appendix XXIII
indicated that 30 - 35# reduction of nitrate was achieved in 0,6 - 0,8
hour with NOg'/NHj ratio of about 3t It would appear from the single
experiment that separation of the denitrification process into two steps
would not be practical but that this reactor design would be useful for
a single step process.
28
-------
PRELIMINARY EVALUATION OF THE DENITRIFICATION PROCESS
USING BOTH TAP WATER AND SECONDARY EFFLUENT
Since the tap water contained both carbonate and silicate and the
secondary effluent contained phosphate in addition, a preliminary treat-
ment was given to the samples before denitrification was attempted. In
all nine anion removal pretreatment processes were used in an effort to
obtain data which would enable direct comparison of the efficiency of
the process in the two media.
Pretreatment with varying amounts of lime (process A - see Appendix XXIV")
alone for short times (0.25 - 0.5 hr.) at low temperatures (?0 - 85° F)
removed over 90$ of the phosphate and presumably some carbonate as well
although no analyses for the latter were made. The subsequent reduction
of nitrate in both tap water and effluent was low (5 - 17$) generally
with more gaseous-N products formed than ammonia. The yield of gaseous-N
products was generally near 60$.
The next two pretreatments involving ferric iron were evaluated because
recycled iron solids containing ferric iron resulted in higher than
normal conversions of nitrate to ammonia (see Appendix XV). Pretreat-
ment with ferric sulfate alone (process C) or preceded by lime treatment
(process B) were ineffective in that nitrate reduction in effluent was
poor (6 - 13$) and the yields of gaseous-N products were erratic.
Apparently, the previously observed increase in ammonia at the expense
of gaseous-N products results from the presence of the copper-containing
FejQii and not just ferric iron. Pretreatment of effluent with ferric
sulfate followed by lime with (process D) or without (process E) prior
acidification resulted in moderate improvements in both the amount of
nitrate reduced (23 - 28$) and the conversion to gaseous-N products
(26 - 34$), but the conversion to ammonia was increased as well.
An acidification pretreatment to remove carbonate was used on tap water
(process F) and on effluent (process H) which included subsequent lime
treatment to decrease phosphate concentration. With this treatment,
subsequent nitrate reduction in tap water and effluent, respectively,
was increased to 49$ and 45$ but ammonia was found as the major reduc-
tion product (33$ conversion) with 16$ and 27$ conversion to gaseous-N
products.
In an attempt to improve the denitrification over that obtained with the
liming pretreatment, additional contacting with magnesium oxide at 85° F
and 194° F (process G) was evaluated assuming that silicate removal would
be beneficial. At the lower temperature no changes were apparent when
compared with liming alone (process A) for either tap water or effluent.
At the higher temperature, reduction was increased to 55$, meet of which
was converted to ammonia (42$) in tap water while in effluent the reduc-
tion was markedly inhibited (only 5$ reduction of nitrate).
29
-------
Inspection of the initial and final phosphate concentrations indicates
that the solids precipitated during the course of the denitrification
experiments are quite effective in removing this anion from solution.
It is not unreasonable to assume that the concentration of other anionic
species, such as carbonate, borate, silicate, etc., might be decreased
simultaneously. If inhibitory substances, either known or unknown, were
removed, then a second denitrification reaction should proceed more
efficiently in the same solution. Several experiments were conducted to
evaluate this anion removal process (process l). The following results
Mere obtained in tap water in the second denitrification for the indi-
cated first pretreatment: (a) no pretreatment, slight improvement in
nitrate reduction but marked decrease in conversion to ammonia with
corresponding increase in gaseous-N products, (b) lime pretreatment
(process A), nitrate reduction increased 2.5 times but conversions to
ammonia, nitrite and gaseous-N products all increased, and (c) lime plus
hot MgO pretreatment (process G), decreased nitrate reduction but more
gaseous-N products formed at the expense of ammonia.
Similar double denitrification experiments starting with secondary
effluent gave the following results: (a) lime pretreatment (process A)
nitrate reduction while still low (34$)had doubled as did formation of
gaseous-N products with a decrease in conversion to ammonia, and (b) lime
plus hot MgO pretreatment (process G), a large increase in nitrate reduc-
tion (from 5% to 5^%} with low conversion to ammonia (9%) and high con-
version to gaseous-N products (43$)•
The data imply that some inhibitory substances have been removed by the
iron-containing solids formed during the denitrification process but
further experiments will be required to identify these species.
ANALYTICAL SUPPORT
The analytical chemistry support activities during this program consisted
primarily of routine analyses for nitrate, nitrite, and ammonia. Addi-
tional routine analyses were carried out for copper, phosphate, and silica
content. From time to time some special analytical problems requiring
the application of non-routine analytical methods were performed primarily
to determine whether there were nitrogenous species formed other than
nitrate, nitrite, or ammonia during the denitrification process. Analyses
for sulfate and chloride were done conductometrically.
Routine Analyses
Ultraviolet Spectrophotometric Method for Nitrate and Nitrite lona. The
analysis procedure for nitrate and nitrite was virtually the same as used
in the feasibility study (Guuderloy, 1968). The absorbances at 200 mp
before and after treatment of sample with sulfamic acid were determined.
The total absorbance is the measure of NO-j" and N02~ , while tho decrease
aa a result of eulfamic acid addition is a measure of NOg". Interference
effects of iron, sulfate, chloride, and sulfamic acid were determined and
used to correct for these species when necessary,
30
-------
Tests with nitrate and nitrite samples at the 0-10 ppm N level indicate
an accuracy of ± 0.2 ppm for nitrate ion. Because of the lower extinction
coefficient of nitrite ion, the accuracy for it is only ±0.4 ppm.
Ammonia Distillation Method. Analysis for ammonia was performed by the
Kjeldahl distillation method used in the feasibility study (Gunderloy,
1968).
Copper. Copper at the part per million level was determined by atomic
adsorption spectrophotometry using a Perkin-Elmer Model 303 instrument.
The conditions of analysis are described in the manufacturer's litera-
ture (anonymous, 1966),
Silica. Soluble silicates were determined by the colorimetric molyb-
dosilicate method (Method B) described in the Standard Methods for the
Examination of Water (anonymous, 1965), using the Gary 14 recording
spectrophotometer. Method A, the gravimetric method was used to stan-
dardize the sodium silicate standard solutions. Phosphate ion up to
10 ppm P was found not to interfere with the colorimetric method.
Phosphate. The analysis for phosphate was performed by the molybdenum
blue spectrophotometric method using the Gary 14 recording spectro-
photometer (Furman, 1962).
Special Analytical Problems
Presence of Nitrogenous Species Other Than NO^ . NO? . and NH^. A com-
parison of the sum of the nitrogen found by separate analyses for NO^',
N02~, and NH^ with a total nitrogen method was made. Two methods for
the determination of total nitrogen were investigated. The purpose was
to convert all nitrogen-containing species, except dissolved N2 or ^0
to a single oxidation state before analysis.
The first method was an oxidation using hydrogen peroxide. Standard
solutions of nitrite ion and ammonium ion were run. Nitrite ion standards
were oxidized quantitatively to nitrate, but ammonium ion remained un-
changed. When experimental samples were analyzed by the peroxide oxida-
tion, however, it appeared that some nitrite remained unoxidized. The
oxidation procedure, therefore, was not considered satisfactory,
The second approach was a reduction method using ferrous sulfate in
highly alkaline media with silver sulfate catalyst to reduce all nitro-
genous species to NH3. Total nitrogen by the reduction method compared
favorable with the sum of nitrogen found by separate analyses for NOj",
N02", and NHj for experiments 25 and 26 (Table l). The following pro-
cedure was used for both standards and experimental samples,
A 20 ml aliquot of sample was added to the reaction vessel followed by
5 ml of 50# NaOH solution, 10 ml of saturated Ag^SO^ solution, and enough
31
-------
TABLE I
REDUCTION OF NITROGENOUS SPECIES TO AMMONIA
Sample
Nitrate Standard
Hydroxylamine
Hydrochloride
Standard
Nitrite Standard
Experiment 25
Experiment 26
Ug N Added
195.0
198.0
200.6
200.6
200. 6^*'
200.0
200.0
-
-
-
N Found by
Standard Methods
ppm, NH-r+NOo +NO-I
-
-
-
-
-
-
3.04
3.04
3.5^
N Found by
Reduction
(j_tg or ppm)
184 ug
198 ug
200.7 Ug
201. 3 ug
8.0 ug
200.6 ug
190. Oug
2.94 ppm
3.00 ppm
3.59 ppm
(a)
No FeSO^ added
-------
solution to provide a two-fold excess to convert NO^" (10 ppm N) to
NH}. The solution was heated to incipient boiling and held for 30 minutes
after which it was distilled into boric acid solution as is done in a
routine NEfrj determination. Nitrite and hydroxylamine were found to be
reduced without the JO minute heating time and without the addition of
catalyst.
A method for the determination of hydroxyamine was developed from reported
information (Nunes, 1970). Hydroxylamine was reported to be converted to
nitrous oxide by reacting with copper(ll) chloride in acidified solution.
Removal of nitrous oxide was accomplished by boiling the solution. The
analytical method developed was a two-step procedure. One aliquot of the
sample was reduced to ammonia using excess FeSO^ solution. Upon Kjeldahl
distillation, this gave the total ammonia and hydroxylamine present. A
second aliquot was treated with excess CuCl2 solution, acidified and
boiled one hour to destroy the hydroxylamine after which the ammonia was
Kjeldahl distilled from basic solution into boric acid and titrated with
standard acid. The difference between the first and second titrations
was a measure of the hydroxylamine. Hydroxylamine was found not to
distill with ammonia in the standard Kjeldahl distillation.
Experiments 278 and 279 were analyzed by this method. Each of these
solutions was found to contain nitrite-N by the uv method. Therefore,
the nitrite ion was destroyed by the addition of sulfamic acid prior to
analysis in Step 1. In Step 2, the nitrite ion was removed during the
one hour boiling. No hydroxylamine was found in either Sample 278 or
279. Accuracy on known mixtures containing all possible species, NO}" ,
N02~ , NH-j, and hydroxylamine was not as good as desired (Table II) , but
was considered adequate for the information desired in Experiments 278
and 279.
Sulfate and Chloride Analyses. A sample containing approximately 10
nancies of 804" was diluted to 5 ml with water and 3 ml of tetrahydro-
furan was added. The conductivity of the solution was recorded on a
stripchart recorder as it was titrated automatically with 0.010 M
Ba( OCOCH3)2. Extrapolation of the straight portions of the graph (i.e.,
the decreasing conductivity as BaSO^ formed and the increasing conducti-
vity as Ba(OCOCH3)n concentration increased) gave the time to precipitate
all the sulfate. Calibration was with standardized 0.010 M Na2S04
solution. In the samples contained calcium ion, it was replaced with
sodium by ion exchange before titration.
Chloride ion waa determined in a similar manner by titrating the samples
with Ot 010 M AgNOj, Tetrahydrofuran was not used nor waa it neceaaary
to remove calcium ion from the solutions.
33
-------
TABLE II
DETERMINATION OF HYDROXYLAMINE AND
AMMONIA IN KNOWN MIXTURES
MIXTURE A
0.25 ppm N0.j
0.50 ppm NO ~
1.4 ppm Nfl,
0.0 ppm NHgO
MIXTURE B
0.25 ppm NO ~
0.50 ppm N0~
1.4 ppm NH_
1.0 ppm NELOH
Mixture
A
B
NH,-N, ppm
Added Found
1.4 1.6
1.4 1.5
NH2OH-N, ppm
Added Found
0.0 -0.2
1.0 0.9
Total-N, ppm
Added Found
1.4 1.4
2.4 2.4
-------
ACKNOWLEDGEMENTS
The work described in this report was carried out over the periods
9 June 1969 to 9 February 1970 and 20 April 1970 to JO September 1970
by the Chemical Products unit of Rocketdyne Research. Dr. B. L. Tuffly
(Manager, Environmental Sciences and Technology) served as the Program
Manager. Dr. F. C. Gunderloy, Jr. (Manager, Chemical Products) was the
Responsible Scientist and Dr. R. I. Wagner was the Principal Investigator
with analytical assistance from Dr. V. H. Dayan and Members of the
Technical Staff, Dr. M. A. Rommel, Mr. R. P. Hollandsworth, Mr. S. R.
Gird, Mr. B. C. Neale, Jr., and Mr. G. G. DeChaine.
The Rocketdyne staff wishes to express their appreciation for the
interest and guidance provided by the Project Officer, Dr. R. B. Dean
(Ultimate Disposal) of the Advanced Waste Treatment Laboratory, Taft
Water Research Center, Cincinnati, Ohio.
The staff also wishes to acknowledge their gratitude to Mr. Lloyd Heden-
land, Manager, Sanitation Operations, Las Virgenes Municipal Water
District, Calabasas, California for his cooperation in providing from
the Tapia Water Reclamation Plant the secondary effluent used in the
reported work.
35
AWBERG LIBRARY U.S. EPA
-------
REFERENCES
Anonymous, Standard Methods for the Examination of Water and Wastewater
American Public Health Association, 1965, pp- 259~64.
Anonymous, Analytical Methods for Atomic Absorption Spectrophotometry,
Perkin-Elmer, November 1966, "Standard Conditions for Copper."
Anonymous, Chem. & Eng. News, 4?, 55 (l969)f "Pickle Liquor Recovery
Helps Sulfuric Market".
Arden, T. V., J. Chem. Soc., 1950. 882, "The Solubility Products of
Ferrous and Ferrosic Hydroxides".
Bates, D. R. and P. B. Hays, Planet. Space Sci. , _15_, 189 (196?),
"Atmospheric Nitrous Oxide".
Brown, L. L. and J. S. Drury, J. Chem. Phys., 46, 2833 (196?) "Nitrogen
Isotope Effects in the Reduction of Nitrate, Nitrite, and Hydroxylamine
to Ammonia. I. In Sodium Hydroxide Solution with Fe(ll)".
Forestier, H. and J. Longuet, Compt. Rend., 208. 1729 (1939), "Formation
of Copper Ferrite at Low Temperature".
Furman, N.H., Standard Methods of Chemical Analysis, Sixth Edition,
Vol. 1, D. Van Nostrand Co., Inc., 1962, p. 819.
Gotz, R., Hutnicke Liety, 18, 718 (1963), "The Possibility of Reprocessing
Waste Pickle Liquors in Metallurgical Operations".
Gunderloy, Jr., F. C., C. Y. Fujikawa, V. H. Dayan and S. R. Gird, Rpt.
No. TWRC-1, FWPCA, Cincinnati, Ohio, October 1968, "Dilute Solution
Reactions of the Nitrate Ion as Applide to Water Reclamation",
Krause, A., Z. Ernst and T. Grzeskowiak, Z.Anorg, Allgem. Chem. 254. 31
(1937), "Amorphous and Crystalline Oxide Hydrates and Oxides. XXXVII.
The Acceleration of the Air Oxidation of Iron(ll) Hydroxide by Lead(ll)
Hydroxide or Pb"1"1" Ion as Well as the Lattice Directing and Lattice
Stabilizing Influence Thereof",
Krauie, A, and A. Borkowska, Roczniki Chem., 2%, 999 (1955), "Oxidation
of Ferrous Hydroxide with Air and its Dependence on the H+ and OH"-Ion
Concentration!".
Krauie, A, and J. Lezuohowska, Roczniki Chem., J2, 29 (1998), "Tht
Catalytic Action of 8i02 on the Air Oxidation of Fe(OH)2".
36
-------
Krause, A. and A. Borkowska, Monatsh., $k, 460 (1963), "Influence of
Phosphate Ions on the Oxidation of Ferrous Hydroxide with Air, the
Inhibition of Magnetite Formation, and Structure of the Oxidation
Product".
Levin, A. I. and S. A. Pushkareva, J. Applied Chem. (U.S.S.R.-English
Translation), Ł9_, 1323 (1956), "The Influence of Anions on the pH of
Hydroxide Formation and on the Electrodeposition of Metal in Powder
Form from Ferrous Sulfate Solution."
Longuet, J., Compt. Rend., 213. 557 (l94l), "Study of the Role of Water
in the Reaction at Low Temperature Between Ferric Oxide and Metallic
Monoxides".
McNesby, J. R. and H. Okabe, "Vacuum Ultraviolet Photochemistry," in
V. A, Noyes, G. S. Hammond, and J. N. Pitts, Jr., Advances in Photo-
chemistry, Vol. 3, Interscience, New York, 1964, pp. 185-191.
Nesbitt, J. B., J. Water Pollut. Contr. Fed., kl, 701 (1969), "Phosphorus
Removal - The State of the Art".
Nunes, T. L. and R. E. Powell, Inorg. Chem., 9, 1912 (1970), "The Copper(l)-
Catalyzed Reduction of Nitric Oxide by Tin(ll) Chloride".
Polydoropoulos, C. N. and Th. Yannakopoulos, Chem. Chronika, 26. 70
(1961), "Heavy Metal Hyponitrites".
Recht, H. L. and M. Ghassemi, FWQA Report (in press), Cincinnati, Ohio,
1970, "Kinetics and Mechanism of Precipitation and Nature of the Pre-
cipitate Obtained in Phosphate Removal from Waste Water Using Aluminum(lll)
and Iron(lll) Salts".
Recht, H. L. and M. Ghassemi, FWQA Report (in press), Cincinnati, Ohio,
1970, "Phosphate Removal from Water Supplies Using Lanthanum Precipi-
tation".
Rippert, I., Gesundheits - Ing. 2%, 333 (1958), "Possibilities for
Elimination of Phosphate from Wastewater by Chemical Means."
Sidgwick, N. V., The Chemical Elements and Their Compounds, Oxford
University Press, 1950, Vol. I, p. 151.
Stone, H, W., Anal. Chem., 20, 747 (1948), "Storage and Titration with
Oxygen-Sensitive Solutions".
Szabo7, Z. G. and L, Bartha, Anal. Chim. Acta, 5, 33 (1951), "A New
Titrimetric Method for the Determination of Nitrate Ion".
Van Wazer, J. II., Phosphorus and Its Compounds, Vol. I, Intersoienoe,
New York, 1958, pp. 556-8.
37
-------
APPENDIX I
REAGENT STOCK SOLUTIONS
Reagent
KN03
FeS04)
FeCl2
CuS04
CuCl2
Na2HP04
NaHC03
Na2Si03
NaOH
Ca(OH)2
(1% slurry)
NaN02Ca)
Concentration
Molarity
0.1428
0.5712
0.0022
1.1424
0.0157
0.0157
0.0646
0.3333
0.0162
0.98
0.1337
0.0144
ppm
2,000 N
55,080 804
31,900 Pe*+
81,000 Cl"
31,900 Fe++
1,000 Cu"1"1"
1,510 S04
1,000 Cu++
1,120 Cl"
2,000 P
20,000 CO*
972 Si02
--
--
202 N
(a)
Never stored over 4 days
38
-------
APPENDIX II
IDENTIFICATION OF GASEOUS-iN COMPOUNDS
E^ri-t
16w
18W
25
29tc)
30
31
1*1
Initial
8.27
8.10
7.87
~
8.40
8.42
Final
7.24
6.81
S.61
~
5.47
5.42
Nitrogen Material Balance, ppm
NHy-N
1.9
1.6
1.9
—
2.0
1.9
NO;-N
2.4
4.9
1.1
~
1.0
1.4
NO~-N
-1.3
0.3
0.1
—
0.1
0.5
Gas ecus -N Compounds
NO-N
1.0
0.9
(b)
—
3.8
3.5
N2-N
(a)
0.9
Cb)
—
2.6
2.4
NO-N
—
—
—
—
--
0.1
Unidentified
(by difference)
6.0
1.4
6.9
--
0.5
0.2
Mass Spectrometer Analysis
Gas
Volume,
cc
3.61
3.43
3.95
3.95
4.81
4.48
Mole, percent
He
89.0
87.8
--
97.7
78.8
75.3
N20
4.2
4.4
(b)
--
12.5
12.2
N2
4.7
7.0
(b)
2.3
8.7
11.0
NO
--
--
--
—
--
0.8
°2
1.6
0.7
--
~
--
0.6
co2
0.5
0.1
--
—
--
--
vO
Fixed Conditions:
Reactor ullage
Initial NO^-N concentration (1.000 ml 0.1428 M KN03/200 •!)
Fe**/H0^ mole ratio (2.000 ml 0.5712 M FeSO4 + 0.0022 M H
Catalyst concentration (1.000 ml 0.0157 M CaSO^/200 ml)
pH Adjustment (1.70 ml 0.98 ± 0.01 M NaOH/200 ml)
Reaction tine
Temperature
Atmosphere
ml)
2.53 cc
10 ppm
8
5 ppa Cu
24 hours
85 F
Heliui
(e)
stock solution used had been in contact with borosilicate glass for 28 days (Experiment 16) and 50 days (Experiment 18)
resulting in an inhibition of the denitrification process
Qomtitative data not nrailable due to malfunction of mass spectrometer and subsequent loss of sample
Blank {Ho DD} used)
to determine baseline value for heliim using improved degassing procedure (used in Experiments 30 and 31)
-------
APPENDIX III
EFFECT OF INITIAL pH ON DENITRIFICATION
*-
o
•D.
1
2
3
4
19
2O
21
S
6
13
14
IS
219
220
10
11
12
7
S
9
273
V61«e 0.9W
UOH Solatia*.
•1/2OOB1
1.130
1.293
1.9SO
1.95O
2.400
2.400
2.4OO
2.200
2.200
2.200
2.200
2.2OO
2.00
1.9S
1.9SO
1.9SO
1.9SO
1.700
1.700
1.700
1.20
OT/Fe**
Hole
Katio
0.97
1.11
1.67
1.67
2.O6
2.06
2.06
1.89
1.89
1.89
1.89
1.89
1.72
1.67
1.67
1.67
1.67
1.46
1.46
1.46
1.03
P«
Initial
8.80
8.00
9.50
8.45
10.96
10.94
10.97
10.46
10.60
8.92
8. 57
8.77
8.50
8.25
8.40
8.30
8.30
8.15
8.00
8.04
7.40
Final
5.60
5.41
5.66
5.6O
10.70
10.78
10.79
7.81
8.62
8.84
8.72
8.56
3.96
5.02
5.66
5.29
5.10
5.30
5.30
5.53
5.20
Nitrogen Material Balance, ppn
NHj-N
1.4
1.8
3.8
4.3
4.0
4.5
4.6
4.9
5.2
5.1
4.9
S.I
4.6
3.9
3.2
3.4
3.1
1.8
1.8
1.9
0.8
MOj-H
5.6
3.7
2.1
1.5
0.0
0.0
0.0
0.5
0.4
0.1
0.1
0.0
1.1
0.8
0.6
0.8
0.6
0.8
0.7
0.7
3.9
H02-N
-0.4
0.1
1.9
0.1
5.9
5.3
5.0
4.5
6.1
2.4
4.8
4.3
4.2
0.5
0.7
2.0
0.3
-0.3
0.3
O.S
0.0
Gaseous -N
0>y difference)
3.4
4.4
2.2
4.1
0.1
0.2
0.4
0.1
-1.7
2.4
0.2
0.6
0.1
4.8
5.5
3.8
6.1
7.7
6.2
6.9
5.3
Gaseous -N,
Percent Yield
77
70
28
48
1
?
4
1
--
24
2
6
1
52
59
41
65
84
67
74
87
Final 50^=
Concentration,
pp»
--
—
--
--
--
--
--
541
542
543
545
545
--
--
536
536
536
525
528
528
-
Fixed Conditions:
Initial nD~-M concentration (l.OOOnl 0.1428M KMO./200nl) 10 ppn
Fe*VBDj ile ratio (2.000nl O.S712M FeS04 * O.OO22M H2SO./200nl) 8
Catalyst concentration (l.OOOnl 0.0157N CuSCj,/200nl 5 pjm Cu*
•cnction Tine 24 hours
Tiamitmi 8S°F
Heliim
(a) l-»t-i«l SO = omcentration, 558 pfm
-------
APPENDIX IV
EFFECT OF THE BASE AND VARYING FeSO./FeCl_ MOLE RATIO
4 i
to.
145
14*
IS*
157
147
14*
1SS
159
149
ISO
160
161
151
152
162
163
153
154
164
165
fir1
Batia
.
-
-
-
3
3
3
3
1
1
1
1
O.33
0.33
0.33
0.33
O
0
0
e
tase
-------
APPENDIX V
EFFECT OF Fe"/N0" MOLE RATIO USING FeSO,,
o 4
Experiment
No.
198
199
38
39
32
33
271oo
7
8
9
145
146
156
157
155
34
37
41
60
61
40«"
I6«
I7<«
31
36
II
IB
14
•»,
70
71
66
17
Fe*7N03~
Mole
R.fio
16
16
12
12
10
10
8
8
8
8
8
1
8
8
8
6
6
6
6
6
6
6
6
6
6
1
1
4
4
5
1
2
2
Volume 0.98M
NaOH, ml
4.56^
4.72<«>
1.80
1.80
1.70
1.70
1.85
1.700
1.700
1.700
1.70
1.70
1.66ta>
1.66^
0.58
1.60
1.60
1,60
1.60
1.60
1.60
1.60
1.60
1.50
1.50
1.21
1.25
1.00
1.00
0.71
0.71
0.50
0,10
OH'/Fe**
Mole
Ratio
1.96
2.02
1.03
1.03
1.17
1.17
1.59
1.46
1.46
1.46
1.46
1.46
1.45
1.45
0.50
1.83
1.13
1.13
1.83
1.83
1.13
1.13
1.13
1.72
1.72
1,72
1.72
1.72
1.72
1.72
1.72
1.72
1.72
PH
Initial
8.21
8.39
8.04
8.10
8.34
8.17
7.72
8.15
8.00
8.04
8.42
8.34
8.23
8.31
8.17
8.82
8.84
1.60
8.80
1.71
1.76
1.71
8.79
1.40
8.52
1.41
1.32
1.41
1.45
1.30
1.29
1.51
1.26
Final
8.07
8.62
5.24
S.08
5.35
5.35
5.42
5.30
5.30
5.53
5.35
5.31
5.02
5.45
5.28
4.12
5.78
5.30
5.69
6.20
6.12
7.90
6.40
5.35
5.40
4.44
5.22
5.51
4.96
4,11
4.34
5.14
5.45
Nitrogen Material Balance, ppm
NHj-N
9.4
8.4
2.2
2.2
2.4
2.0
3.0
1.8
1.8
1.9
2.1
2.1
l.S
1.7
0.8
2.0
2.7
2.7
2,1
2.9
2.6
1.7
2.6
2.4
2.3
1.9
1.9
O.I
1.1
0.6
0.6
0.2
0.2
NO"-N
0.2
0.1
1,1
1.6
l.S
1.5
0.7
0.8
0.8
0.7
0.4
0.4
2.0
1.6
7.7
1.4
0.6
0.6
2.1
1.7
0.4
0.5
0,2
1.9
1.3
2.1
1.1
2.3
2.6
4.5
4.4
6.1
7.2
NO~-N
0.1
0.0
0.2
0.1
0.4
0.2
0.2
-0.3
0.3
0.5
0.3
0.3
0.3
0.1
0.7
5.0
3.3
5.2
2,0
3.2
6.3
7.3
6.9
0.5
0.4
1.6
0.9
1.6
1.4
1.2
2.6
1.6
2,5
Gaseous -N
(By Difference)
0.3
1.5
6.5
6.1
5.7
6.3
6.1
7.7
6.2
6.9
7.2
7.2
6.2
6.6
0.8
1.6
3.4
1.5
3.1
2.2
0.7
0.5
0.3
5.3
6.0
4.4
5.6
5.3
4,9
3.7
2.4
2.1
0.1
Ga«eou9-N,
Percent
Yield
3
15
73
73
67
74
66
84
71
74
75
75
78
79
35
19
36
16
39
27
7
5
3
61
69
56
67
69
61
17
41
14
4
Fixed ConditJam i
Initial NOj-N eenetntratlen (1,000 il 0,14JIM KNOj/200 •!)
ft** oencintration (0,210 ml 0,5712 M FrIO. + 0.0032M H.M./JOO •! for «toh unit vilui of
Fe*?HOJ Mil rttie) ' *
CiUlyit ooneintritien (1,000 hi 0.01I7M CuM./JOO •!)
Mittlon tiM
TMptrtturi
AtMiphiri
lOppn
I pirn Cu"
14 noun
II F
H.UUJI
(t) 1 pirotnt Ci(OH)} ilurry ui*d| eiloultttd •! 0.91 M N»OH derived by •ultiplieitien of il 1 ptr««nt CitOH),
by 0,111 '
bj 19,1 hour motion tiM
c) 1.1 hour ruction tiM
d) 1,0 hour ratatim tiM
42
-------
APPENDIX VI
EFFECT OF Fe /N03 MOLE RATIO USING FeCl2
Experiment
No.
90
91
92
iss""
154^
164
IttW
93
94
95
96
97
98
99
100
101 W
102 (a)
Fe++/N0:
Mole
Ratio
16
8
8
8
8
8
8
6
6
5
5
4
4
3
3
2
2
Volume
0.98M
NaOH.ml
2.50
1.80
1.80
1.70
1.70
1.66^
1.66
1.30
1.30
1.10
1.10
0.70
0.70
0.60
0.60
0.35
0.35
OH~/Fe*+
Mole
Ratio
1.07
1.54
1.54
1.46
1.46
1.45
1.45
1.49
1.49
1.S1
1.51
1.20
1.20
1.37
1.37
1.20
1.20
pH
Initial
8.06
8.29
8.20
8.31
8.36
8.33
8.30
8.26
8.12
8.17
8.12
8.11
8.02
8.09
8.07
8.16
7.96
Final
5.83
5.59
5.52
5.34
5.25
3.56
5.64
5.29
5.36
5.00
4.96
4.85
4.76
4.95
4.97
4.19
3.99
Nitrogen Material Balance, ppm
NH3-N
7.8
5.5
5.2
3.6
3.5
2.7
3.1
3.3
3.3
2.8
2.5
1.2
1.3
0.9
0.8
0.3
0.3
N03-N
4.5
4.3
4.5
3.5
3.6
4.6
4.3
5.6
5.7
6.3
6.5
7.5
7.5
7.5
7.5
8.7^
9.4^
NO~-N
0.8^
0.0
0.2
0.2
0.0
0.3
0.0
-0.2
0.1
-0.2
-0.1
-0.6
-0.7
-0.4
-0.4
-0.4
-0.1^
Gaseous-N
(by difference)
-3.1
0.2
0.1
2.7
2.9
2.4
2.6
1.3
0.9
1.1
1.1
1.9
1.9
2.0
2.1
1.4
0.4
Gaseous-N,
Percent
Yield
-
4
2
42
45
44
46
30
21
30
31
76
76
80
84
107
67
Fixed Conditions:
Initial NOj-N concentration (1-000 ml 0.1428M KN03/200 ml) ++ 10 ppm
Fe++ concentration (0.125 ml 1.1424M FeCl2/200 ml for each unit value of Fe++/N03 mole ratio)
Catalyst concentration (1.000 ml 0.0157M CuCl2/200 ml) S ppm Cu
Reaction Time 24 hours
Temperature 85°F
Atmosphere Helium
(a)Dark brown supernatant gave high background in spectrophotometric analysis for N03 and N02
due to inability to balance the instrument.
(b)l.OOO ml 0.0157M CuS04/200 ml used in lieu of equivalent amount of CuCl2.
(c)l percent'Ca(OH)2 slurry used; calculated ml 0.98M NaOH derived by multiplication of ml 1 percent Ca(OH)2 by 0.268.
-------
APPENDIX VII
CATALYST CONCENTRATION EFFECT
Experiment
No.
7
8
9
42
43
44
45
46
47
50
51
52
53
54
55
48
49
Cu+ +
Concentration,
ppm(a)
5
5
5
4
4
3
3
2
2
1
1
0.5
0.5
o.osflO
0.0500
(c)
(c)
pH
Initial
8.15
8.00
8.04
8.46
8.36
8.30
8.30
8.32
8.30
8.32
8.33
8.44
8.24
8.36
8.33
8.34
8.35
Final
5.30
5.30
5.53
5.41
5.35
3.84
5.52
5.38
5.44
5.66
5.45
5.73
7.50
7.64
7.74
7.28
7.91
Nitrogen Material Balance, ppm
NH3-N
1.8
1.8
1.9
2.0
2.1
1.8
2.1
1.1
2.1
2.2
2.2
1.9
2.4
0.5
0.5
1.1
1.2
NO"-N
o
0.8
0.7
0.7
1.0
0.6
1.2
1.0
0.6
0.4
1.0
1.2
3.3
3.7
9.5
9.5
8.8
8.1
NO"-N
-0.3
0.3
0.5
-0.2
0.2
0.5
0.0
0.6
0.2
0.3
0.1
0.0
1.2
0.0
0.5
-0.2
0.5
Gaseous-N
(by difference)
7.7
6.2
6.9
7.2
7.1
6.5
6.9
7.7
7.3
6.5
6.5
4.8
2.7
0.0
-0.5
0.3
0.2
Fixed Conditions:
Initial NOj-N concentration (1.000 ml 0.1428 M KN03/200 ml) 10 ppm
Fe+VNC>3 mole ratio (2.000 ml 0.5712 M FeS04 + 0.0022 M H2S04/200 ml) 8
OH'/Fe** mole ratio (1,70 ml 0,98 M NaOH/200 ml) 1.46
Reaction time 24 hours
Temperature 85 F
Atmosphere Helium
l>a>)For each ppm Cu** 0.200 ml 0.0157 M CuSO,/200 ml was used
fVO
1 ^Calculated concentration from the copper impurity in the FeSO.
'•C')16.3 ppm Pb** (molar equivalent of 5 ppm Cu**) added as 0.0478g solid PbSO>
44
-------
APPENDIX VIII
EFFECT OF ORDER OF ADDITION OF CATALYST
Experiment
No.
34
37
41
60
61
40
56
57
62
63
Order of
Catalyst
Addition^
Normal
Normal
Normal
Last
Last
Normal
Normal
Normal
Last
Last
pH
Initial
8.82
8.84
8.60
8.80
8.78
8.76
8.71
8.79
8.68
8.85
Final
4.12
5.78
5.30
5.69
6.20
6.12
7.90
6.40
8.98
8.60
Reaction
Time,
Hours
24
24
24
24
24
1.5
1
1
1
2
Nitrogen Material Balance, ppm
NH3-N
2.0
2.7
2.7
2.8
2.9
2.6
1.7
2.6
0.4
1.0
NO--N
1.4
0.6
0.6
2.1
1.7
0.4
0.5
0.2
4.7
3.4
NO~-N
5.0
3.3
5.2
2.0
3.2
6.3
7.3
7.9
4.9
5.3
Gaseous -N
(By Difference)
1.6
3.4
1.5
3.1
2.2
0.7
0.5
0.3
0.0
0.3
Gaseous -N,
Percent Yield
19
36
16
39
27
7
5
3
0
5
Fixed Conditions:
Initial NO^-N concentration (1.000 ml 0.1428 M KN03/200 ml)
Fe^NOs mole ratio (1.500 ml 0.5712 M FeS04 + 0.0022 M H2S04/200 ml)
OH~/Fe++mole ratio (1.60 ml 0.98 M NaOH/200 ml)
Catalyst concentration (1.000 ml 0.0157 M CuS04/200 ml)
Reaction Time
Temperature
Atmosphere
10 ppm
6
1.83
5 ppm Cu+
24 hours
85 F
Helium
(a) Normal addition of reagents: FeS04, CuS04, NaOH, KN03
-------
APPENDIX IX
EFFECT OF RECYCLED SOLID REACTION PRODUCTS
Experiment
No.
7
8
9
58
59
277 Q
277 V
240
Catalyst
Source
(a)
(a)
(a)
(b)
(b)
(c)
(d)
(e)
Fe++/N0
Ratio
8
8
8
8
8
6.5
6.5
8
OH /Fe+*
Ratio
1.46
1.46
1.46
1.46
1.46
1.23
1.23
1.50
pH
Initial
8.15
8.00
8.04
8.36
8.30
--
--
8.11
Final
5.30
5.30
5.53
5.04
5.48
--
--
6.38
Nitrogen Material Balance, ppm
NH3-N
1.8
1.8
1.9
3.7
3.2
3.1
1.8
3.4
NO"-N
o
0.8
0.8
0.7
1.2
1.2
5.7
6.0
4.4
NO~-N
-0.3
0.3
0.5
0.2
-0.1
0.9
0.7
-0.2
Gaseous-N
(by difference)
7.7
6.2
6.9
4.9
5.7
-0.1
1.1
2.4
Fixed Conditions:
Initial NO^-N concentration (1.000 ml 0.1428 M KN03/200 ml)
Reaction time
Temperature
Atmosphere
10 ppm
24 hours
85 F
Helium
^Fresh catalyst control 1.000 ml 0.0157 M CuS04/200 ml equivalent to 5 ppm Cu+ +
Black solids equivalent to 5 ppm Cu from Experiments 56 (used in 58) and 57 (used in 59)
which may be found in Appendix IV
( cl
Black solids accumulated after flow reactor had run for approximately 100 hours
(d)
(e)
Black solids accumulated 13 hours after cleaning without altering flowrates used
for Experiment 277 Q
Presumably CuO equivalent to 5 ppm Cu with FeOOH from an attempt to prepare Cu(FeO )
-------
APPENDIX X
DISSOLUTION OF COPPER OR COPPER COMPOUNDS IN
ACIDIC FERROUS SULFATE
Copper Source
fa!
Recovered^ J
Fe3°4
Recovered '
150 Mesh
Copper Powder
As-Received
150 Mesh
Copper Powder
Leaching
Time,
Hours
0.25
0.50
0.75
1.0
1.5
5.0
24.0
0.25
0.50
0.75
1.0
1.25
24.0
68.0
0.25
0.50
0.75
1.0
1.25
24.0
68.0
Cu Concentration,
ppm
2.35
2.05
1.90
1.85
2.15
1.90
4.22
0.30;
0.38;
0.41;
0.48;
0.52;
4.92;
0.65;
1.35;
1.77;
2.03;
2.22;
S6.4&
0.09(C)
0.13(C)
0.22CC)
1.40(C)
-
0.60(C)
0.75
1.02(C)
1.23(C:>
» ~
Fixed Conditions;
Leaching medium (2.000 ml 0.5712 M FeS04
+ 0.0022M H2S04/200 ml)
Temperature
Atmosphere
5.712 x 10"3M FeS04
2.2 x 10'5M H2S04
85 F
Helium
(a) From experiment 271, Appendix V
(b) From experiments 262-264, Appendix XI
(c) Second leaching with fresh solution
(d) Air had leaked into reactor blackening copper and
precipitating brown FeOOH
47
-------
APPENDIX XI
ELEMENTAL COPPER AS CATALYST SOURCE
*-
OD
Experiment
No.
zsa00
n>i
259l°;
260
261
262
263
264
Catalyst
Form (a)
0.2610g Shot
0.2554g Shot
0.2S81g Shot
0.2694g Shot
0.2207g Powder
0.2285g Powder
0.2302g Powder
Volume 0.98M
NaOH, ml
2. IS
2. IS
2.20
2.18
2.12
2.12
2.13
Oil /Fe"
Hole
Ratio
1.84
1.84
1.89
1.87
1.82
1.82
1.83
PH
Initial
8.12
8.26
8.34
8.16
8.10
8.10
8.20
Final
8.12
8.11
8.08
8.01
8.15
6.44
8.33
Nitrogen Material Balance, ppra
NHj-N
0.6
0.9
0.5
0.2
4.7
4.7
4.7
NO -N
9.4
9.5
9.7
9.9
0.5
0.9
0.4
N02-N
0.7
-0.1
0.2
0.2
4.5
4.9
5.5
Gaseous-N
(by difference)
-0.7
-0.3
-0.4
-0.3
0.3
-0.5
-0.6
Fixed Conditions:
Initial NDj-N concentration (1.000 ml 0.1428 M KNOj/200 •!)
Fe*?N03 ratio (2.000 ml 0.5712 M FeS04 + 0.0022 M H2S04/200 •!)
Reaction ti»e
Temperature
Atmosphere
10 ppm
8
24 hours
85 F
Helium
(a) Calculated approximate surface area (assuming spherical particles) of No. 1 shot
(0.1S7 cm2) and ISO mesh powder (10.6 cm2)
(b) Metal heated for 1 hour at 85 F with FeSO before addition of NaOH and KNO
-------
APPENDIX XII
EFFECT OF PHOSPHATE ON DENITRIFICATION USING FeSO.-CuSO
4 4
Experiment
Ho.
72
73
74
75
76
77
237
238
229
230
231
232
235
236
233
234
78
79
132""
II.™
PO=-P
Concentration,
PP»
Initial
10
10
5
5
2
2
2
2
2
2
2
2
2
2
2
2
1
1
1
0.5
0.5
Final
-
-
-
-
-
-
0.0
0.0
0.0
0.0
0.5
0.5
0.9
0.7
0.8
0.7
-
0.1
0.1
0.0
Volume 0.98M
NaCH, ml
0.85
0.85
l.SO
1.50
1.70
1.70
0.74
0.74
1.50
1.50
1'.90
1.90
2.02
2.06
2.10
2.10
1.90
1.90
9.70
10.00
14.00(C)
OT/Fe**
Hole
Ratio(a)
0.73(0.84)
0.73(0.84)
1.29(1.34)
1.29(1.34)
1.46(1.48)
1.46(1.48)
0.63(0.66)
0.63(0.66)
1.29(1.31)
1.29(1.31)
1.63(1.65)
1.63(1.65)
1.73(1.76)
1.77(1.79)
1.80(1.82)
1.80(1.82)
1.63(1.64)
1.63(1.64)
1.66(1.66)
1.72(1.72)
1.74(1.74)
PH
Initial
8.16
7.98
8.26
8.22
8.03
7.99
7.25
7.28
7.31
7.34
7.78
7.65
8.08
8.68
9.42
9.64
8.16
8.19
8.24
8.25
8.30
Final
6.14
6.36
6.40
5.41
6.55
6.70
6.06
6.16
5.63
5.71
9.14
9.16
9.72
9.82
9.91
10.01
8.40
9.68
7.35
6.30
4.94
Nitrogen Material Balance, ppm
«43-N
0.8
1.2
1.6
1.4
1.3
1.3
0.5
0.4
1.3
1.2
1.8
1.5
0.7
0.6
0.5
0.3
1.2
1.6
1.3
4.6
4.3
NO~-N
9.4
9.4
7.0
7.2
6.2
6.6
8.7
9.2
6.8
6.6
6.3
6.2
6.9
7.5
8.1
7.6
5.7
5.4
6.6
1.7
0.1
NO--N
0.0
0.9
0.4
0.4
0.6
0.2
0.0
0.0
0.0
0.8
2.1
1.3
2.6
1.3
1.1
0.9
0.0
0.2
1.8
2.4
4.1
Gaseous -N
(by difference)
-0.2
-l.S
1.0
1.0
1.9
1.9
0.8
0.4
1.9
1.4
-0.2
1.0
-0.2
0.6
0.3
1.2
3.1
2.8
0.3
1.3
l.S
Gaseous-N,
Percent Yield
-
-
33
36
50
56
62
50
59
41
-
26
-
24
16
50
72
61
9
16
15
Fixed Conditions:
Initial HOj-N concentration (1.000 ml 0.1428M KNDj/200 •!) 10 ppn
Fe*7M>3 mole ratio (2.000 ml 0.5712M FeSO4 + 0.0022H H2SO4/200 nl) 8
Catalyst concentration (1.000 ml 0.0157N CuS04/200 ml) 5 ppm Cu+
Initial POf-P concentration (0.100 ml 0.0646N Ka2HK>4/200 nl for each ppm)
•action time 24 hours
Tiajii nl in I 85 F
Atmosphere Heliun
(a) Parenthetical nine assumes 2 NaOH available fron each NaJffO.
(b) 1000 •! reaction volune
(c) 0.30 nl 12M HC1 added to lower initial pH
-------
APPENDIX XIII
EFFECT OF PHOSPHATE ON DENITRIFICATION USING FeCl2-CuCl2
vx
o
Experiment.
No.
241
242
243
244
245
246
247
248
91
92
249
250
251
252
253
254
255
256
POJ-P
Concentration,
PP»
0
0
1
1
2
2
5
5
0
0
0
0
1
1
2
2
5
5
Volune 0.98M
NaOH, ml
2.20
2.20
2.15
2.15
2.10
2.10
2.05
2.05
1.80
1.80
1.50
1.50
1.50
1.50
1.50
1.50
1.50
1.50
OT/Fe**
Mole f -
Ratio la)
1.89
1.89
1.84(1.86)
1.84(1.86)
1.80(1.82)
1.80(1.82)
1.76(1.82)
1.76(1.82)
1.54
1.54
1.29
1.29
1.29(1.30)
1.29(1.30)
1.29(1.31)
1.29(1.31)
1.29(1.34)
1.29(1.34)
pH
Initial
8.39
8.28
8.20
8.37
8.06
8.22
7.94
7.96
8.29
8.20
7.75
7.73
7.73
7.86
7.83
7.70
7.67
7.75
Final
8.96
9.18
8.30
8.66
5.96
7.45
6.66
8.87
S.S9
S.52
5.65
5.55
5.77
5.89
6.97
7.12
7.14
7.04
Nitrogen Material Balance, ppm
NH3-N
5.4
5.3
5.1
5.2
5.3
5.1
S.I
1.7
5.5
S.2
4.3
4.2
3.4
3.5
0.9
0.9
1.4
1.4
NO--N
1.6
0.7
1.0
0.8
2.7
1.6
3.2
5.8
4.3
4.5
5.4
5.6
5.2
5.0
6.5
6.5
7.4
7.5
N02"N
3.1
4.2
3.6
4.4
1.4
2.8
0.3
0.8
0.0
0.2
0.0
0.2
0.0
0.0
l.S
0.9
0.0
0.0
Gaseous -N
(by difference)
-0.1
-0.2
0.3
-0.4
0.6
0.5
1.4
1.7
0.2
0.1
0.3
0.0
1.4
1.5
1.1
1.7
1.2
1.1
Gaseous -N,
Percent Yield
-
-
3
-
8
6
21
40
4
2
7
0
29
30
31
49
48
44
Filed Conditions:
Initial NDj-N concentration (1.000 nl 0.1428M KNOj/200 »1) 10 ppm
Fe*V»3 ">le ratio (1-000 ml 1.1424M FeCl2/200 ml) 8
Catalyst concentration (1.000 nl 0.0157M CuCl2/200 •!) 5 ppn Cu*
Initial POj-P concentration (0.100 ml 0.0646M tU^OPO^/200 »1 for each pp«)
Reaction tine 24 hours
Tenperatnre 85 F
Atnospbere Heliun
(a) Parenthetical valne ass
2 NaOH available fron each Na
2HP04
-------
APPENDIX XIV
EFFECT OF ADDITIVES ON DENITRIFICATION IN THE PRESENCE OF PHOSPHATE
VJI
Experiment
Ho.
82
83
(6
87
80
81
84
85
111(C)
115 *d'
119td)
112Cc}
116(d)
120 ^
113(c,e)
117(d,e)
121 Cd)
114(c'e)
ns'd'e)
122Cd,e)
Additive
Concentration,
7.S Al***
7.5 Al***
7.5 Al***
7.5 Al***
14.4 Fe***
14.4 Fe***
14.4 Fe***
14.4 Fe***
RS 109
RS 111
RS 115
RS 109
RS 112
RS 116
RS 110
RS 113
RS 117
RS 110
RS 114
RS 118
POJ-P
Concentration,
PP»
Initial
10
10
1
1
10
10
1
1
10
10
10
10
10
10
10
10
10
10
10
10
Final
_
-
-
-
-
-
-
-
0.4
0.4
0.4
0.4
0.4
0.3
0.4
0.5
0.4
0.3
0.3
0.3
Volume 0.98H
NaOH, ml or
wt CaO, mgO>)
1.60
1.60
1.80
1.80
1.40
1.40
1.90
1.70
(59.9)
(60.2)
(60.2)
(59.9)
(60.5)
(60.0)
(59.8)
(59.7)
(59.7)
(60.1)
(60.3)
(60.2)
OH'/Fe**
Hole
Ratio
1.37
1.37
1.54
1.54
1.20
1.20
1.63
1.46
1.87
1.88
1.88
1.87
1.89
1.87
1.87
1.86
1.86
1.88
1.88
1.88
PH
Initial
8.16
8.04
8.08
7.90
8.20
7.60
8.10
8.08
7.39
7.34
7.36
6.40
7.52
7. 55
7.22
7.59
7.62
7.34
7.40
7.22
Final
6.80
6.66
6.62
7.02
6.31
5.93
7.50
6.42
8.00
7.28
8.20
7.52
8.17
8.26
8.43
8.30
8.35
8.50
8.40
8.50
Nitrogen Material Balance, ppm
NH3-N
1.2
1.1
1.0
1.0
0.7
0.6
1.1
1.0
1.3
1.4
1.4
1.3
1.4
1.5
1.8
1.5
1.6
1.7
1.5
1.8
NOj-N
7.9
8.2
6.8
7.0
7.9
7.8
6.6
7.0
8.S
7.9
8.2
8.1
8.0
8.3
7.8
8.0
8.1
7.9
7.8
7.7
NO~-N
0.4
-0.1
0.0
0.0
0.3
0.8
0.2
0.3
0.0
0.0
0.0
0.0
0.4
0.3
0.0
0.0
0.5
0.0
0.0
0.8
Gaseous-N
(by
difference)
0.5
0.8
2.2
2.0
1.1
0.8
2.1
1.7
0.2
0.7
0.4
0.6
0.2
0.2
0.4
0.5
0.3
0.4
0.7
-0.3
Gase-
ous-N,
Percent
Yield
24
44
69
67
52
36
62
57
13
33
22
32
10
12
18
25
16
19
32
-
Fixed Conditions:
Initial NOj-N concentration (1.000 •! 0.1428H KNOj/200 •!) 10 ppm
Fe*yHOj Bale ratio (2.000 ml 0.5712H FeSO4 + 0.0022M H2S04/200 •!) 8
Catalyst concentration (1.000 ml 0.0157H CaSO^/200 •!) 5 ppm Cu*
Reaction ti»e 24 hours
Temperature 85 F
Atmosphere Helium
(a) 7.S ppm Al***derived from 2.50 •! A12(S04)3 solution analyzing 600 ppm Al; 14.4 ppm Fe**derived from l.SO ml Fe2(S04)3
solution analyzing 1925 ppm Fe; RS 109 designates that one-half of the solids precipitated in experiment 109 were
recycled by suspension in helium deaerated distilled water which was used as the matrix for a subsequent experiment.
(b) Parenthetical data indicates weight of solid CaO (N.F. Grade lime) in milligrams.
(c) Order of addition of reagents: Recycled solids suspension, Na2HPO4, 1003, CaO, FeSO4, CuS04.
(d) Order of addition of reagents: CaO, recycled solids suspension, NajHPO^, KNOj, FeSO4, CuSO4.
(e) Catalyst concentration (0.500 ml 0.0157M CuS04/200 ml); 2.5 ppa Cu**.
-------
APPENDIX XV
EFFECT OF VARIOUS PHOSPHATE REMOVAL METHODS ON DENITRIFICATION
to
Experiment
No.
89
88
A
103
104
D
109
110
B
105
106
C
107
108
Ł«=>
167
168
F(c)
166 ^
169 °°
170 (h)
171 Cc)
172
173
174
175
PO|-P
Concentration,
ppm
Initial
10
10
10
-
-
10
2.0
2.0
10
-
-
10
-
-
10
0.2
0.2
10
0.5
10
10
0.9
6.8
6.9
0.2
0.2
Final
-
-
-
-
-
2.0
0.0
0.0
-
-
-
-
-
-
0.2
0.0
0.0
0.9
0.0
6.8
6.9
0.2
0.1
0.1
0.1
0.1
Phosphate
Precipitation
Source
80 pp» Fe**(S04)
80 ppm Fe**(S04) +
7.5 ppm A1***(S04)
280 ppm CaO
-
-
301 ppm CaO
-
-
560 ppm CaO
-
-
1120 ppm CaO
-
-
46 ppm La""'"f(Cl)
-
-
900 ppm CaO
is blk prcpt 166
-------
APPENDIX XV
(Concluded)
VJl
to.
123(f)
124**'
121
12SW
129
12.™
130
U7W
131
132(f>
133<">
137
1J4(b)
131
13S""
139
I**'
140
141
142
143
144
Concentration,
PP"
laitial
O.S
10
2.1
10
1.4
10
2.7
10
2.0
1
10
0.0
10
0.0
10
0.0
10
0.0
10
10
10
10
Fiaal
0.1
2.1
1.6
1.4
0.0
2.7
0.0
2.0
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
1.9
7.1
4.S
Phosphate
Precipitation
Source
-
V blk prcpt 123
-
k blk prcpt 123
-
k blk prcpt 123
-
k blk prcpt 123
_
.
k grn prcpt 132
_
% grn prcpt 132
_
3 mole ratio (2.000 nl 0.5712N FeSO4 + 0.0022M H2SO4/200 a]
1.1424M F0C12/2OO nl)
Catalyst coacentntloa (1.000 nl 0.0157M CuSO4/2OO nl)
" > concentration (0.100 nl 0.0646M MajHPO^OT al for
1 or 1.833
each ppa)
al 0.57121 FeS04 * 0.0022M H2SO4 and 0.083 al
(a) a.miaOB. 1 percent Ca(OH)2 slurry (eouualeat to 0.134M). or O.SM H2S04
(a) pB mljmftft nth innu il raluae of «N B^SO^
(c) SOT ml teaniun nlnae
W nl varict from S.49-1.B nth helioa flowate
(•1 Salatiaa rnntiinrJ O.lO247g •a200I-ll2O (100 ppa OOj)
(f) UN nl reaction nlnne
(f) ».» ml UN BC1 aided to loner initial pH
m
CO <
10 ppa
S ppa Cu
24 hours
85 F
Heliua
-------
APPENDIX XVI
EFFECT OF CARBONATE ON DENITRIFICATION
Experiment
No.
(b)
(b)
28
178
179
184
185
180
181
186
187
182
183
188
189
co-f , ,
( 3.1
Concentration,
ppm
100
100
100 (c)
100
100
100
100
200
200
200
200
300
300
300
300
OH~/Fe++
Mole
Ratio
-
1.46(1. 75) (d)
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
1.40
PH
Initial
10.3
9.0
8.20
8.06
8.11
8.08
8.07
8.15
8.04
8.11
8.23
8.18
8.17
8.14
8.13
Final
10.2
8.4
6.84
6.64
6.78
6.22
6.33
7.14
7.30
7.10
7.15
7.25
7.33
7.27
7.25
Reaction
Time,
Hours
24
24
24
24
24
48
48
24
24
48
48
24
24
48
48
Nitrogen Material Balance, ppm
NH3-N
3.5
4.4
3.2
2.4
2.6
3.0
2.9
0.8
1.2
1.2
1.2
0.7
0.5
0.7
0.7
NO--N
1.8
3.2
0.6
2.8
2.8
2.5
2.2
7.4
4.7
7.6
7.3
8.4
8.4
8.0
8.4
NO"-N
4.3
1.8
1.0
0.4
0.1
0.5
0.5
0.5
0.0
0.3
0.9
-0.1
0.4
0.3
0.4
Gaseous-N
(By Difference)
0.4
0.6
5.2
4.4
4.5
4.0
4.4
1.3
4.1
0.9
0.6
1.0
0.7
1.0
0.5
Gaseous-N,
Percent Yield
5
9
55
61
63
53
56
50
77
38
22
63
44
50
31
VJ1
*•
Fixed Conditions:
Initial N03-N concentration (1.000 ml 0.1428M KN03/200 ml)
Fe*VN03 mole ratio (2.000 ml 0.5712M FeS04 + 0.0022M H2S04/200 ml)
Catalyst concentration (1.000 ml 0.0157M CuS04/200 ml)
C03~ concentration (1.00 ml 0.3333M NaHC03/200 ml for each 100 ppm)
pH adjustment (6.40 ml 1 percent Ca(OH)2 slurry)
Temperature
Atmosphere
10 ppm
8
5 ppm Cu++
85 F
Helium
(a) Carbonate added after iron and copper salts
(b) Data from feasibility study (Gunderloy, 1968) where 0.3333M K2C03 and 6M NaOH solutions were used.
(c) 1.00 ml 0.3334M Na2C03/200 ml initially of which 66 percent remained in solution after 24 hours.
(d) pH adjusted with 1.70 ml 0.98M NaOH; parenthetical value assumes 1 NaOH available for each Na2C03.
-------
APPENDIX XVII
EFFECT OF BASE-BOROSILICATE GLASS CONTACT TIME ON DENITRIFICATION
Experiment
No.
22
23
24
25
26
28
7
8
9
16
17
18
27
NaOH
Age,
Days
0
0
0
0
0
0(a)
7
7
7
28
50
55
71
PH
Initial
7.90
7.80
8.08
7.87
7.59
8.20
8.15
8.00
8.04
8.27
8.28
8.10
7.82
Final
4.35
5.35
4.25
5.61
5.11
6.84
5.30
5.30
5.53
7.24
7.30
6.81
7.44
Nitrogen Material Balance, ppm
NH3-N
1.9
1.9
2.0
1.9
2.1
3.2
1.8
1.8
1.9
1.9
1.1
1.6
0.9
NO--N
0.8
0.6
0.6
1.1
1.2
0.6
0.8
0.7
0.7
2.9
5.0
4.9
4.4
NQ--N
0.7
0.4
0.5
0.1
0.2
1.0
-0.3
0.3
0.5
0.7
0.0
0.3
0.3
Gaseous-N
(By Difference)
6.6
7.1
6.9
6.9
6.5
5.2
7.7
7.2
6.9
4.5
3.9
3.2
4.4
Gaseous-N,
Percent Yield
71
76
73
78
74
55
84
77
74
63
78
63
79
Fixed Conditions:
Initial NOy-N concentration (1.000 ml 0.1428M KN03/200 ml) 10 ppm
Fe*7l»3 »ole ratio (2.000 ml 0.5712M FeS04 + 0.0022M H2S04/200 ml) 8
Catalyst concentration (1.000 ml 0.0157M CuS04/200 ml) 5 ppm Cu++
OH~/Fe**mole ratio (1.70 ml 0.98M NaOH) 1.46
Reaction time 24 hours
Temperature 85 F
Atmosphere _ Helium
(a) Stored 6 days in polyethylene
-------
APPENDIX XVIII
PRECIPITATION OF SILICA WITH ALUMINUM, CUPRIC, OR FERRIC SULFATES
Precipitant
Al (SO )
L, *T +J
CuSO.
*+
Fe (SO )
I* *T w
M/Si
Mole Ratio
2
1
2
Initial
pH
4.03
5.05
5.93
6.93
7.60
7.80
8.00
8.20
9.01
10.01
11.00
3.94
4.93
6.19
6.99
8.03
9.00
10.02
10.98
3.96
5.05
5.96
6.93
8.03
8.99
9.97
Si02 Remaining
In Solution,
ppm
22.4
18.5
14.9
9.4
5.8
6.2
4.9
4.7
19.0
23.8
23.9
23.8
23.1
23.4
18.0
18.0
17.9
18.2
19.6
21.2
19.4
16.2
12.7
12.1
16.0
16.3
Si02
Precipitated,
percent
7.8
23.9
38.7
61.3
76.1
74.5
79.8
80.7
21.8
2.1
1.6
2.1
4.9
3.7
25.9
25.9
26.3
25.1
19.3
12.8
20.2
33.3
47.7
50.2
34.2
32.9
Fixed Conditions:
Initial Si02 concentration (50.0 ml
solution obtained by 20-fold dilution
of 0.0162 M Na2Si03/100 ml)
Reaction time
Temperature
Atmosphere
24.3 ppm
1 hour
85 F
Helium
56
-------
APPENDIX XIX
EFFECT OF SILICA ON DENITRIFICATION
Experiment
No.
219
220
221
222
223
224
225
226
227
228
Si02
Concentration,
PP»
Initial
0
0
1.0
1.0
4.9
4.9
9.7
9.7
19.4
19.4
Final
U)
0.9
1.4
1.4
3.5
3.1
5.0
3.8
5.6
7.3
Volume 0.98 M
NaOH,
ml
2.00
1.95
1.90
1.95
1.90
1.95
1.95
2.00
1.85
1.92
OH-/Fe*+
Mole
Ratio
1.72
1.67
1.63
1.67
1.63
1.67
1.67
1.72
1.59
1.65
P*1
Initial
8.50
8.25
8.07
7.94
8.13
8.05
8.23
8.25
8.13
8.31
Final
3.96
5.02
5.27
5.59
5.37
5.40
5.82
6.39
6.59
6.66
Nitrogen Material Balance, ppm
NH3-N
4.6
3.9
3.6
4.1
4.6
4.7
4.8
4.8
4.9
4.4
_
N03-fl
1.1
0.8
0.5
0.5
1.3
0.8
1.8
0.6
3.6
2.8
_
NO -N
4.2
0.5
1.1
0.5
0.8
1.0
1.5
3.4
0.4
1.9
Gaseous-N
(by difference)
0.1
4.8
4.8
4.9
3.3
3.5
1.9
1.2
1.1
0.9
Gaseous-N,
Percent Yield
1
52
51
52
38
38
23
13
17
13
VJI
Fixed Conditions:
Initial NO^-N concentration (1.000 »1 0.1428 M KN03/200 •!)
Fe**/l»3 mole ratio (2.000 ml 0.5712 M FeS04 * 0.0022 M H2S04/200 •!)
Catalyst concentration (1.000 »1 0.0157 M CuS04/200 ml)
Initial Silica concentration (1.00 ml 0.0162 M Na^iO^OO ml for each 4.9 ppm Si02)
Reaction time
Temperature
Atmosphere
10 ppm
8
5 ppm Cu
24 hours
85 F
Helium
-------
APPENDIX XX
UNCATALYZED REDUCTION OF NITRITE
Experiment
No.
265
266
267
268
278
279
269
270
272(c.d)
274(C)
275(c.e)
Volume
0.988H
NaOH.il
1.05
1.05
0.80
0.80
0.80
0.80
0.60
0.60
4.00
4.00
4.00
OH'/Fe"
tele
Ratio
1.82
1.82
1.38
1.38
1.38
1.38
1.04
1.04
1.38
1.31
1.31
PH"
Initial
8.12
8.40
7.90
7.85
8.57
8.42
7.60
7.62
7.60
7.70
7.52
Final
8.60
8.35
5.85
5.96
5.31
4.98
5.95
4.11
7.24
7.39
7.40
7.51
6.24
6.02
5.98
7.80
7.46
6.94
7.08
7.04
6.01
4.90
7.20
7.14
7.20
7.30
7.34
Reaction
Time,
Hours
24
24
24
24
24
24
24
24
0.5
1
2
3
4
5
24
0.5
1
2
3
4
7.5
23
0.5
1
2
3
4
Nitrogen Material Balance, ppn
NH3-N
4.3
4.2
1.3
1.5
1.4
1.2
1.0
1.0
1.1
1.2
1.2
1.2
1.3
1.3
1.4
0.2
0.3
0.6
0.6
0.7
0.8
0.9
0.4
0.5
0.5
0.5
0.5
NO--N
0.0
0.1
0.0
0.1
0.1
0.0
0.1
0.1
0.0
0.0
0.0
0.0
0.1
0.1
0.1
0.0
0.0
0.1
0.1
0.1
0.1
0.0
0.1
0.0
0.0
0.0
0.0
NO~-N
5.9
5.9
0.3
0.0
0.6
0.5
0.7
1.3
7.0
6.5
4.8
1.1
0.5
0.0
0.0
8.7
8.2
6.0
4.5
3.4
0.8
0.0
4.8
4.8
5.0
5.1
5.3
Gaseous-N
(by difference)
-0.1
-0.2
8.5
8.5
S7.^"'
8.3
7.7
2.0
2.4
4.1
7.8
8.2
8.7
8.6
1.2
1.6
3.4
4.9
5.9
8.4
9.2
4.7
4.7
4.5
4.4
4.2
Gaseous-N,
Percent
Yield
8.7
8.4
8.3
8.6
8.8
8.8
65
67
77
87
85
86
85
86
84
83
88
88
90
91
90
90
90
90
89
Fixed Condition!:
Initial NOJ-N
Fe /NOj acle
Temperature
Atnoapher*
concentration (10.0 •! 0.0144M NaN02/200 •!)
ratio (1.000 ml 0.5712M FeS04 » 0.0022M
ml)
10.1 ppll
4
85°P
Helium
(i)With the exception of experiMmts 278 and 279 the pH values an suspected to be loo due to a failing electrode.
(b)Analysis for hyaroxylaadne in experlMnti 278 and 279 gave -0.01 and 0.01 pp» NH2OH-N, respectively.
(c)lOOO Hi reaction volua*.
(d)NaOH added la«t.
(e)Initial NO:>-N concentration, 10.0 pp»; initial PO*-P concentration, 1 ppm (0.500 ail 0.0646M NajHPO^/1000 ml)
-------
VJI
APPENDIX XXI
EFFECT OF REACTION TIME ON DENITRIFICATION
Experiment
No.
275A
I
C
D
E
F
G
H
239A™
I
C
D
E
F
G
H
2S7A0"
B
C
0
E
F
G
Mole J
Ratio
6
.w
.w
Volume
0.98H
NaOH,
ml/1000 ml
6.00
8.50
10.50
OtT/Fe**
Mole
Ratio
1.37
1.46
1.80CC)
P"
Initial
7.40
9.4
5.68
Final
7.05
7.06
7.34
5.72
5.45
5.48
5.20
7.04
7.55
7.6
7.6
7.6
7.6
7.5
7.0
4.05
7.64
7.80
7.84
7.78
7.69
6.91
S.74
Reaction
Time.
Hours
O.S
1
2
3
4
5
24(i)
29
1
2
3
4
5
6
7
24
1
2
3
4
5
6
24
Nitrogen Material Balance, ppn
NH -N
0.4
O.S
O.S
0.8
0.8
0.8
0.8
0.8
0.1
0.4
1.1
1.9
2.4
2.8
2.8
3.0
0.1
0.6
1.4
2.6
3.1
3.3
3.4
NO~-N
5.8
S.8
S.S
4.4
4.0
3.9
3.9
3.8
8.6
7.8
6.7
5.6
5.3
4.4
4.4
4.5
8.7
7.0
5.0
4.5
3.8
3.8
3.4
NO~-N
3.5
S.S
0.8
0.8
0.4
0.3
0.0
0.0
1.2
1.0
1.3
1.3
0.6
2.3
0.7
1.9
0.7
2.4
3.1
2.0
2.7
1.6
-0.1
Gaseous-N
(by difference)
0.3
0.2
3.2
4.0
4.8
S.O
5.3
5.4
0.1
0.8
0.9
1.2
1.7
0.5
2.1
0.6
O.S
0.0
0.5
0.9
0.4
1.3
3.3
Gaseous-N,
Percent
Yield
7
5
71
71
80
82
87
87
7
36
27
27
36
9
37
11
38
0
10
20
6
21
SO
Fixed Conditions:
Initial HJj-N concentration (5.000 •! 0.1428M KNOj/1000 •!) 10 ppm
Final Fe**/N0j mole ratio (1.25 »1 O.S712M FeS04 » 0.0022M H2S04/1000 »1 for each unit value of ratio)
Final catalyst concentration (5.000 »1 0.01S7M CuSO4/1000 •!) 5 ppm Cu*
Temperature 85° F
Atmosphere Helium
(a)Immediately after withdrawal of analytical sample 2.00 •! 0.3333M NaHC03 was added to residual
300 ml of reaction mixture to raise pH to 6.52 (133 ppm COj).
(b)All nitrogen analyses normalized to 10 ppm N.
(c)In these experiments the indicated volumes of stock solutions were diluted to 100 ml and added at 25 ml/hr
through a mixing chamber to an initial 950 ml of dilute KNOj solution: 8.50 ml (Experiment 239) or 10.50 ml
(Experiment 257) of 0.98M NaOH in 100 ml; 10.000 ml 0.5712M FeS04 + 0.0022H H2S04 and 5.000 nl 0.01S7M CuS04 in 100 ml.
(d)Final mole ratio assuming reagents had been added to reaction mixtures to give final volume of 1000 ml.
-------
APPENDIX XXII
DENITRIFICATION IN A FLOW REACTOR WITH RECIRCULATING SOLIDS
Ho.
nt*M
•
c
D
E
277A«>
1
C
D
E
F
G
H
I
J
I
L
X
N
0
f
q
R
S
T
U
V
Total
Flovnt*.
sd/hr
113
122
58
57
S*
69
M
55
Filed Conditions:
Teape rat lire
Ataosphere
Retention
Tia».
hours
4.1
3.8
8.0
8.2
8.3
6.7
7.8
8.S
Relative
Flowrates,
D0}
110
23.5
16.5
5.5
8.5
21.0
FeS04-CuS04
(b)
29.5
13.5
8.0
11.0
23.5
NaOH
0>)
29.0
13. S
11.0
8.S
21.0
Elapsed Tie*
at Steady Flow.
hours
5.25
17
22.33
24.5
41
49.5
11
12
13
14
IS
16
17
15.5
18.5
20. S
22. S
11
10.5
12. S
21
23
25
6.5
8
9. 75
11.5
13
PP"
9.7
8.6
11.4
6.7
9.1
9.6
Fe /NO,"
Mole
Ratio
(b)
6.0
4.9
8.8
7.8
6.7
OH-/F.**
Hole
Ratio
0>)
1.70
1.38
1.90
1.07
1.23
Nitrogen Material Balance,
PP"
NH -N
2.0
0.7
0.7
0.7
0.7
1.7
3.2
2.6
2.4
2.3
2.3
2.5
2.3
3.2
2.9
2.8
3.0
2.4
3.4
3.3
3.4
3.2
3.1
0.2
0.3
0.6
1.2
1.8
NO~-H
5.3
8.8
9.2
9.3
8.6
8.2
4.9
5.2
5.6
5.4
5.5
5.5
54.
5.8
6.2
6.4
6.4
0.6
3.9
4.5
5.3
S.S
5.7
2.0
3.S
4.6
5.6
6.0
N0"-»l
2.0
0.0
0.9
0.2
0.0
0.0
2.0
1.4
0.2
1.2
0.9
0.9
1.1
1.1
1.1
0.7
1.1
1.8
1.3
0.3
0.3
0.4
0.9
0.6
0.6
0.7
0.2
0.7
Gaseous-N
(by difference)
0.4
0.2
-1.1
-0.5
0.4
-0.2
-1.5
-0.6
0.4
-0.3
-0.1
-0.3
-0.2
1.3
1.2
l.S
0.9
1.9
0.5
1.0
0.6
0.5
-0.1
6.8
5.2
3.7
2.6
1.1
Gaseous-N.
percent yield
0
22
-
36
:
-
13
-
-
23
23
30
18
31
10
22
14
12
89
85
74
65
31
8S F
Heliias
'^Reagents: 0.5781, Of), in 1000 al . 0.0007147M CNO, (10 ppa MOj-N)
75 aJ 0.5712M FeS04 • 0.0022N H2S04 in 125 al • 0.3427M FeS04 » 0.0013M H,
60 al 0.988M NaOH in 125 ail • 0.474N NaOH
(c)
Coafcined FeS04 and NaOH flow was 3 to 4 percent of total flow but due to numerous Mechanical problesu flowrates were not sufficiently
steairjr to provide Meaningful calculation of relative flowrates or Bole ratios of reagents.
Reajenti: 1.5159g DUj in 7000 »1 • 0.002142M COj (30 pp« "Oj-NJ
25.0107g Fe^-TH^ * 0.4124j CuS04 • 5H20 • 0.735 ml 0.474M H2S04 in 7000
126 ml 0.989M NaOH In 7000 ml • 0.0178H NaOH
• 0.01290M FeSO4 . 0.0002360M CuSO. . 0 000049714 H,SO.
-------
APPENDIX XXIII
DENITRIFICATION IN A TUBULAR FLOW REACTOR
Experiment
No.
2SOA<>)
B
C
D
E
F(d)
G
H
I
J
K
Total
Flowrate,
•1/hr
70
69
75
77
98
76
Itet6nti.on
Time.
hours
0.86
0.88
0.81
0.79
0.62
0.80
Relative
FlOWI*4t6$
™°3
80
76
80
83
70
CuS04
20
19. S
20
21
13
Flowrates,
•1/hr
FeS04
3.30
3.86
3.61
3.53
NaOH
3,72
3.73
3.61
3.85
Initial pH
8.3 - 9.4
8.3 - 9.4
7.S - 8.l(<0
7.7 - 7.8
7.5 - 7.6
8.2 - 8.7
8.6 - 10.2
8.6 - 9.0
8.8 - 10.2
8.7 - 10.2
9.1 - 10.4(e)
HOj-N,
PP"
9.8
9.7
9.8
10.0
9.8
9.8
10.0
10.0
10.3
10.3
10.3
Fe+*/NO *
ro f i^j-
Mole *
Ratio
2.57
2.62
2.76
2.73
2.73
S.ll
3.91
3.83
4.63
4.63
4.63
OH~/Fe**
Mole
Ratio
1.95
1.95
1.67
.67
.67
.73
.73
.89
.89
1.89
1.89
Nitrogen Material Balance,
PP"
NHj-N
0.3
O.S
0.1
0.3
0.2
0.2
0.3
0.3
0.6
0.8
1.0
NO~-N
9.3
8.4
9.1
9.3
9.3
7.2
6.8
7.0
6.7
6.4
6.6
NO--NCa)
_„
1.7
1.5(0.4)
0.0(0.1)
1.3(0.1)
0.5
1.1
1.4
1.9
2.3
2.3
Gaseous -N
(by difference)
__
-0.9
-0.9(0.2)
0.4(0.3)
-1.0(0.2)
1.9
1.8
1.3
0.8
0.8
0.4
Gaseous-N,
percent yield
__
—
-(29)
57(43)
-(40)
73
56
43
22
21
11
Fixed Conditions:
Temperature 85 F
Atmosphere Helium
(a) Parenthetical values obtained by reduction to NH, rather than spectrophotometrically.
(b) . . - .
arna aue ,
Reagents (280 A-E) : O.S898g KN03 in 6000 •! = 0.000972M KN03 (13.6 ppm NOj-N)
(c)
(d)
Effluent pH = 6.40
Reagents (280 F-K) :
(e) Effluent pH = 7.55
0.3944g CuS04-5H20 in 4000 nl = 0.000395M CuS04 (25.1 ppm Cu*+)
3.000 ml O.S712M FeSO4 + 0.0022M H2S04 in 45 ml = 0.03808 FeS04
3.00 ml 0.989M NaOH in 45 ml = 0.0659M NaOH
KN03 and CuS04 unchanged
FeS04 and NaOH concentrations doubled
0.00015M H2S04
-------
APPENDIX XXIV
DENITRIFICATION IN TAP WATER AND SECONDARY EFFLUENT SUBSEQUENT TO REMOVAL OF SILICATE
AND/OR CARBONATE AND PHOSPHATE
Experiment
No.
176
177
190
191
192
193
194
195
196
197
200
201
202
207
208
205
206
210
209
203
204
211
212(8)
21500
214(h)
216
21300
21700
218(>0
Type of
Matrix W
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
T«P
Tap
Tap
Tap
Tap
Effluent
Effluent
Effluent
Effluent
Distilled
Fro» 207
FroB 205
FroB 210
FFOB 203
Froa 211
FroB 212
An ion
Removal
Process (b)
A
A
A
A
B
B
C
C
D
D
E
E
E
None
F
A
G
G
H
A
G
G
None
I
1
I
I
I
I
Volume 1%
Ca(OH)2.
ml
5.45
5.45
3.80
5.20
4.50
7.20
7.60
8.20
7.30
8.30
8.00
8.00
8.00
8.20
10.00
7.40
7.30
8.00
9.40
6.00
6.20
8.20
10.00
1.60
3.20
3.60
2.60
3.00
3.40
pH
Initial
8.43
8.41
8.27
8.25
8.25
8.31
8.30
8.23
8.34
8.29
8.20
8.18
8.09
8.35
8.07
8.24
8.14
8.39
8.34
8.11
8.15
8.38
8.36
8.40
8.32
8.40
8.38
8.33
8.36
Final
7.65
7.69
7.18
6.18
6.89
6.91
6.97
7.06
7.10
7.25
6.89
6.78
6.82
7.35
6.66
7.57
7.37
7.36
9.00
7.59
7.18
7.90
6.54
7.20
6.38
7.68
6.87
7.01
(i)
Initial NOs-N
Concentration,
ppmCc)
15.9
15.9
12.1
12.1
12.1
12.1
12.2
12.2
12.0
12.0
12.1
12.1
12.1
10.2
10.2
10.2
10.2
10.2
10.8
10.8
10.8
10.8
10
10.0
10.0
10.0
10.0
10.3
10.0
Fe**/N0j
Mole
Ratio
5.03
5.03
7.92
7.92
7.92
7.92
7.88
7.88
8.02
8.02
8.02
8.02
8.02
7.82
7.82
7.82
7.82
7.82
8.16
8.16
8.16
8.16
8.00
8.00
8.00
8.00
8.00
7.76
8.00
Nitrogen Material Balance, ppm
NH3-N(d)
6.2(0.7)
6.3(0.8)
0.6(0.4)
0.4(0.2)
0.3(0.1)
0.9(0.7)
1.0(0.9)
0.6(0.5)
1.7(1.6)
1.4(1.3)
1.5(1.4)
1.5(1.4)
1.5(1.4)
1.5
2.4
1.6
1.5
4.3
3.6(2.7)
2.1(1.2)
1.8(0.9)
1.7(0.8)
1.5
2.3(0.8)
4.3(2.7)
5.5(1.2)
3.0(0.7)
2.6(0.9)
--
NOj-N
14.1
13.9
11.0
11.5
11.4
10.0
11.0
10.6
8.6
9.3
8.8
8.8
8.7
7.6
5.0
7.4
7.5
4.4
5.9
9.0
8.7
10.3
6.2
7.3
3.3
5.9
6.6
4.6
--
N0--Ntd)
0.0
0.0
0.0
0.0
1.8
-0.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-0.3
0.5
0.0
0.5(0.0)
1.1(0.6)
1.1(0.6)
0.5(0.0)
0.9
0.0
1.8
0.0
1.1(0.5)
0.5
--
Gaseous -N
(by
ifference;
1.1
1.2
0.7
0.4
-1.2
1.8
0.3
1.1
1.8
1.4
1.9
1.9
2.0
1.1
1.7
1.5
0.7
1.5
2.7
0.5
1.1
0.2
1.4
1.9
2.5
2.9
2.6
4.3
--
Gaseous -N,
Percent Yield
61
60
63
67
--
86
25
69
53
52
58
58
59
42
33
54
26
26
50 (f)
23(«
44(«
20 («
37
70
37
71
63 C«
75
-
PO|-F
Concentration , ppm
Initial^
0.7
0.7
0.8
0.8
0.5
0.5
6.9
6.9
0.9
0.9
0.2
0.2
0.1
0.04
--
0.00
0.00
0.00
0.02
0.06
0.18
0.05
1
0.00
0.00
0.00
0.00
0.00
0.00
Final
0.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.00
--
0.00
--
0.00
0.00
0.00
--
0.00
0.00
--
"
--
--
--
--
-------
APPENDIX XXIV
(Concluded)
Fixed Conditions:
Catalyst Concentration (1.000 ml 0.01S7M CuSO /200 ml) 5 ppm Cu++
Reaction Time 24 hours
Temperature 85 F
Atmosphere Helium
^Effluent analysis (ppra N) : N03 5.8 (experiments 176 and 177), 12.1 (experiments 190 through 197, 200 through 202),
10.8 (experiments 203, 204, 209, and 211); NC>2 0.5 (experiments 203, 204, 209, and 211).
^ ^A--10.0 ml 1-percent Ca(OH)2 slurry/1000 ml for 15 minutes at 85 F (final pH = 9.53) followed by centrifugation (for
experiments 176 and 177); 6.0 ml 1-percent Ca(OH)2 slurry/1000 ml for 15 minutes at 85 F (final pH = 9.75) followed
by centrifugation (for'experiments 190 and 191); 8.3 ml 1-percent Ca(OH)2 slurry/1000 ml for 30 minutes at ambient
(final pH = 9.96) followed by centrifugation (experiment 205); 10.8 ml 1-percent Ca(OH)2 slurry/1000 ml for 30 minutes
at ambient (pH = 9.97) followed by centrifugation (experiment 203)
B--6.0 ml 1-percent Ca(OH)2 slurry/1000 ml for 15 minutes at 85 F, centrifugally clarified, 0.88 ml Fe2(S04)3 (analyzing
1925 ppm Fe )/1000 ml during 1 hour at 85 F (final pH = 9.48) followed by centrifugation
C--0.86 ml Fe2(SC>4)3 (analyzing 1925 ppm Fe't"f)/1000 ml during 1 hour at 85 F (final pH = 8.17) followed by centrifugation
D--0.86 ml Fe2(S04), (analyzing 1925 ppm Fe++)/1000 ml during 1 hour at 85 F, centrifugally clarified, 6.0 ml 1-percent Ca(Oll),
slurry/1000 ml for IS minutes at 85 F (final pH = 9.51) followed by centrifugation.
^ E--1 ml 0.5M H2S04/1000 ml (pH = 7.06), 0.98 ml Fe2(S04)3 (analyzing 1925 ppm Fe*+)/1000 ml during 1 hour at 85 F (pH = 7.03),
centrifugally clarified, 6.0 ml (experiment 200) or 9.2 ml (experiment 201) or 12.4 ml 1-percent Ca(OH)2 slurry/1000 ml for
15 minutes at 85 F (final pH = 9.60, 10.05, and 10.55, respectively) followed by centrifugation
F--3.2 ml O.SM HjSC^/lOOO ml, bubbled helium for 1 hour at ambient, and added 1.2 ml more acid
G--8.3 ml (for tap) or 10.8 ml (for effluent) 1-percent Ca(OH)2 slurry/1000 ml for 30 minutes at ambient (pH = 9.96), centri-
fugally clarified, added 0.0380 ±0.0005g MgO (~150 ppm) for 15 minutes at ambient (experiments 204 and 206) or 90 C (experi-
ments 210 and 211) followed by centrifugation
H--4.0 ml O.SM H2S04/1000 ml, bubbled helium for 1 hour at ambient, 12.0 ml Ca(OH)2 slurry/1000 ml for 15 minutes at ambient
(pH = 10.95), centrifugally clarified, and added 0.4 ml more acid
I--Solutions from indicated experiment reused (after adiusting NOj concentration)
Includes that already present plus added KNOj
Parenthetical values refer to ammonia or N02 ion formed in the denitrification process only while the other numerical values
include that present in the solution initially
Subsequent to the preliminary anion removal process. The samples of secondary effluent analyzed 8.9 ppm POj-P (used for experi-
ments 176 and 177, 190 through 197, and 200 through 202), and 10.9 ppm POJ-P (used for experiments 203 and 204, 209, 211 213
and 217)
Yield based on conversion of sum of initial NOj + N02
^250 ml reaction volume with 0.125 ml 0.0646M Na2HP04 (1 ppm PO|-P), 1.25 ml 0.3333M NaHCOj (100 ppm COj), and 0.15 ml O.SM H-,S04 added
^ -*100 ml reaction volume
Reaction vessel broke during run
-------
1
Access/on Number
w
5
2
Organization
Rocketdyne, a Division
Canoga Park, California
Subject Field & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
of North American Rockwell Corporation t
Title
DEVELOPMENT OF A CHEMICAL DENTCRTJTCATION PROCESS
10
Authors)
Gunderloy, Frank
Wagner , Ross !«, ,
Dayan, Victor Ho
C* , Jr0 ,
and
16
21
Project Designation
Project No. 1T010EEX
Note
22
Citation
23
Descriptors (Starred First)
«Denitrification, *Nitrates, ""Reduction^ Potable water, Wastewater
25
Identifiers (Starred First)
Ferrous ion, catalysis
Abstract
Chemical denitrification of dilute nitrate solutions has been achieved with high con-
version of the nitrate primarily to a mixture of nitrous oxide and nitrogen with lesser
amounts of ammonia and nitrite. The investigated variables of the process, based on the
anaerobic copper catalyzed ferrous iron reduction of nitrate in basic media, were initial
pH (or OH~/Fe++ ratio), Fe^/NO^" ratio, lime vs» NaOH for pH adjustment, catalyst con-
centration, catalyst recycle, anions introduced with the iron and copper, and anions
present in the treated water* The optimum conditions as presently defined include:
initial pH near 8 (ratio OB~/Fe++ = 1*5), ratio Fe++/N0j~ = 8, Fe^ delved from FeS04,
Cu = 1-5 ppfflj NaOH for pH adjustment, and pretreatment to remove phosphate and decrease
carbonate content* Preliminary investigation of reduction in a continuous flow system was
encouraging* The by-product magnetite, Fe-^, was easily separated from the treated water
in the absence of appreciable phosphate which inhibits its formation* Cursoiry studies in
potable water and secondary effluent indicate the presence of as yet unidentified
inhibitory substances* Data from 280 experiments are tabulated. (Vagner, Rocketdyne)
Abstractor _
Ross
I.
Wagner
Institution
Rocketdyne,
a
Division
of
North
American
Rockwell
Corp0
WR:102 (REV. JULY 1969)
WRSI C
SEND. WITH COPY OF DOCUMENT, TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
* GPO: 1 970-389-930
------- |