June 1988
BIOLOGICAL DEGRADATION OF CYANIDE BY
NITROGEN-FIXING CYANOBACTERIA
by
Dr. C.J. Gantzer
Dr. W.J. Maier
University of Minnesota
Department of Civil and Mineral Engineering
Minneapolis, MN 55455
Project Officer
James S. Bridges
Office of Environmental Engineering and Technology Demonstration
Hazardous Waste Engineering Research Laboratory
Cincinnati, OH 45268
This study was conducted through
Minnesota Waste Management Board
St. Paul, MN 55108
and the
Minnesota Technical Assistance Program
University of Minnesota
Minneapolis, MN 55455
HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
-------
ABSTRACT
This study examined the ability of nitrogen-fixing Anabaena to biodegrade
cyanide in batch reactors. Mixed second-order rate constants were obtained
that described the biologically-mediated decrease in cyanide for reactors
containing initial cyanide concentrations of 3 ppo. For Anabaena cultures not
previously exposed to cyanide, the rate constants were a function of pH.
Faster rates of cyanide biodegradation were observed at higher pH values.
Anabaena cultures acclimated to the presence of cyanide had rate constants
that were at least 10 times faster than rate constants for unacclimated
cultures.
Mixed second-order rate constants were also obtained for the ability of
nitrogenase, the enzyme normally responsible for nitrogen-fixation, to reduce
hydrogen cyanide to methane and ammonia. In batch reactors with initial
cyanide concentrations of 30 ppb, the rate constants for methane production
were at least 10 times faster than expected based on literature values for
nitrogen fixation, suggesting that nitrogenase will preferentially use
hydrogen cyanide as a substrate as compared to molecular nitrogen. Also, the
rate constants for methane production were of the same order of magnitude as
the rate constants for total cyanide removal, indicating nitrogenase as an
important mechanism for the biodegradation of trace concentrations of cyanide.
The magnitude of the cyanide biodegradation rate constants suggests that
the utilization of nitrogen-fixing cyanobacteria in the treatment of cyanide
wastes can be a feasible process in some applications, i.e., secondary or
tertiary treatment at larger treatment facilities.
iv
-------
This project was partially supported with a United States
Environmental Protection Agency cooperative agreement through the
Minnesota Waste Management Board and the Minnesota Technical
Assistance Program.
Although the research described in this report has been funded in
part by the United States Environmental Protection Agency through
a cooperative agreement, it haa not been subjected to Agency
review, and therefore does not necessarily reflect the views of
the Agency and no official endorsement should be inferred.
-------
CONTENTS
Abstract iv
Tables vl
Figures vii
1. Introduction 1
2. Conclusions 4
3. Recommendations 7
4. Materials and Methods
Experimental Overview 8
Culture Media 8
Experimental Procedures
Growth Rate Experiments 10
Total Cyanide Removal Experiments
Objective and Apparatus 12
Cyanide Mass Balances 12
Kinetic Parameter Determination 13
Methane Formation Experiments
Objective and Apparatus 14
Kinetic Parameter Determination 15
Analytical Procedures 17
5. Results and Discussion
Growth Rate Determination 19
Total Cyanide Removal Rates
Limitations of Batch Experiments 23
Rate Constants for Total Cyanide Removal 24
Effect of Acclimation on Cyanide Removal Rates 32
Methane Production Rates
Rate Constants for Methane Production 33
Comparison of Methane Production with Nitrogen Fixation . . 39
Application of Kinetic Data to Cyanide Treatment 41
References 46
-------
TABLES
Number
1. Cyanobacteria culture media
2. Extent of biodegradation and volatilization in
selected batch experiments ................... 25
3. Overall volatilization rate constants (K^a) and
mixed second-order rate constants (Kv) for
total cyanide removal by unacclimatea Anabaena
cultures in selected batch experiments ............. 26
4. Comparison of mixed second-order rate constants
for total cyanide removal by unacclimated (batch
number 4) and acclimated (batch number 5)
Anabaena cultures in presence of thiosulfate .......... 33
vi
-------
FIGURES
Number Page
1. Logarithmic increase in Anabaena biomass (X) as
function of time, during a batch experiment 20
2. Changes in Anabaena biomass (X) versus time after
the start of feed flow through the reactor. The
dotted line represents the expected decrease in X
if dilution was the only removal mechanism 21
3. Decrease in total cyanide concentration (S) during
batch test number 3. The line represents the
numerical solution of equation (3) using the
Kf, and K^a values on Table 3 27
4. Relationship between log Kb [L/(ug chl hr)] versus
time-averaged pH 29
5. Relationship between log Kb [L/(ug chl hr)] versus
time-averaged ALPHA 30
6. Increase in headspace methane concentration with time.
For the illustrated batch test, the initial cyanide
concentration was 31.2 ug CN/L 34
7. Sensitivity of K_ obtained from equations (13) and (14)
to inputted Kh values. Both parameters have units
of L/(ug chl hr) 37
8. Sensitivity of the ratio between Kn and Kb to inputted
Kv values. Both parameters have units of
L/(ug chl hr) 38
vii
-------
SECTION 1
INTRODUCTION
The basic premise of this study was that the use of nitrogen-fixing
cyanobacteria (blue-green algae) in the biological treatment of small
concentrations of free cyanides (HCN and CN~) can be a cost-effective
alternative to existing treatment processes. A potential application of a
cyanobacteria-based process would be in the secondary treatment of the free
cyanides that escape alkaline-chlorination. Because the extent of cyanide
oxidation in alkaline-chlorination is an equilibrium-driven phenomena, use of
a microbial process to detoxify the last fraction of cyanide should result in
lower alkaline-chlorination operating costs. Another application of a
cyanobacteria-based process could be in the treatment of the cyanide
associated with metal-cyanide complexes via a two-step process. The first-
step would release cyanide from the metal-cyanide complexes by exposing the
complexes to ultraviolet irradiation. In the second-step, cyanobacteria would
detoxify the released cyanide.
The use of nitrogen-fixing cyanobacteria in the treatment of cyanide
wastes is a new concept. There are several potential advantages associated
with the use of nitrogen-fixing cyanobacteria in the treatment of small
concentrations of cyanide. First, the biological treatment of cyanide with
cyanobacteria should have much lower operating costs than alkaline-
chlorination. The operating costs for the biological treatment of cyanide
wastes with aerobic heterotrophs can be less than 10% the costs for alkaline-
-------
chlorination (Green and Smith, 1972). The costs associated with providing
aeration and with providing a supplemental energy source (organic substrate)
for the maintenance of large amounts of aerobic heterotroph biomass make up a
considerable portion of the total operating costs for the traditional
biological treatment of hazardous wastes. Because cyanobacteria are
photosynthetic, they do not require aeration for oxygen and do not require the
presence of organic substrates to maintain biomass (Kobayashi and Rittmann,
1982). Thus, in terms of operating costs, the use of nitrogen-fixing
cyanobacteria in the treatment of small amounts of cyanide should have an
economic advantage over the use of heterotrophic bacteria, and, consequently,
a significant economic advantage over alkaline-chlorination.
Second, nitrogen-fixing cyanobacteria have the ability to survive in low
to moderate concentrations of hydrogen cyanide. Hydrogen cyanide is toxic
because it inhibits the terminal cytochrome oxidase in respiration, which
normally reduces oxygen to water. Cyanobacteria have several terminal
oxidases—some of which are resistant to cyanide inhibition (Fogg, et al.,
1973; Degn, et al., 1978; Peschek, 1980; Henry, 1981). Cyanobacteria also
have several cyanide detoxification pathways, i.e., enzymatic pathways that
transform free cyanides into a less toxic form (Castric, 1981; Higgins, et
al., 1984). The most studied and perhaps the most important detoxification
pathway is mediated by the enzyme rhodanese, which transfers a sulfur from a
donating compound (e.g., thiosulfate) to cyanide to form thiocyanate (Westley,
1981). Other cyanide detoxification pathways result in the formation of amino
acids (Solomonson, 1981). Thus, due to the presence of cyanide detoxification
pathways and cyanide-resistant respiration, cyanobacteria can survive in
solutions containing free cyanides. For example, Howe (1963, 1965) observed a
-------
"luxuriant" growth of cyanobacteria on the filter stones of a biological
reactor that was treating wastes containing 300 ppra cyanide.
Third, in addition to the above pathways, the nitrogen-fixing
cyanobacteria can destroy hydrogen cyanide with the enzyme nitrogenase. While
normally responsible for the reduction of molecular nitrogen (dinitrogen) to
ammonia, nitrogenase can also reduce hydrogen cyanide to methane and ammonia
(Hardy and Knight, 1967; Hardy and Burns, 1968; Biggins and Kelley, 1970;
Haystead, et al., 1970; Hwang and Burris, 1972; Hwang, et al., 1973; Zumft and
Mortenson, 1975; Stewart, 1980; Li, et al., 1982). In fact, nitrogenase will
preferentially reduce hydrogen cyanide instead of its normal substrate,
dinitrogen (Li, et al., 1982). Some researchers have proposed that the
original role of the nitrogenase system was to detoxify the cyanide and
cyanogen found in the primitive biosphere when the earth had a reducing
atmosphere (Silver and Postgate, 1973; Postgate, 1982).
Despite the considerable amount of information indicating that
cyanobacteria are able to survive in the presence of cyanide and are able to
detoxify cyanide, the kinetic data required to assess the feasibility of
utilizing cyanobacteria in the treatment of cyanide wastes does not exist.
This study provides an initial assessment of the rate at which nitrogen-fixing
cyanobacteria are able to degrade free cyanide. In particular, the mixed
second-order rate constants for the biologically-mediated removal of cyanide
by unacclimated cultures of Anabaena were determined in batch reactors. A
second set of batch experiments determined the mixed second-order rate
constants for the reduction of hydrogen cyanide by nitrogenase.
-------
SECTION 2
CONCLUSIONS
In batch reactors with initial cyanide concentrations of 3 mg/L, the
mixed second-order rate constants (Kb) for the removal of cyanide by nitrogen-
fixing Anabaena cultures was a function of pH. The biodegradation rate
constants increased with increasing pH for unacclimated Anabaena cultures,
i.e., cultures not previously exposed to cyanide. At a pH of 8.4 and a
temperature of 25°C, the observed rate constant in terms of chlorophyll (chl)
concentration was 5.0* 10~" L/(ug chl hr). When the pH was increased to 9.5,
Kj, increased to 2.2-10~* L/(ug chl hr). The observed relationship between Kj,
and pH suggested that the ratio of hydrogen cyanide to total free cyanide
concentrations, [HCN]/([HCN]+[CN~]), influenced biodegradation rates. Because
HCN is much more toxic and inhibitory than CN~, and because an increase in pH
would reduce HCN concentrations, the faster biodegradation rates observed at
higher pH values was probably due to a reduction in inhibition.
When previously exposed to cyanide, Anabaena biodegraded cyanide at a
faster rate. The Kj, value for an acclimated Anabaena culture was 10 times
greater than that for an unacclimated Anabaena culture. Thus, reactors
operating under steady-state conditions should have faster cyanide removal
rates than those observed in the batch experiments.
«
The mixed second-order rate constant for the reduction of cyanide to
methane by nitrogenase (Kn) was determined in small gas-tight batch reactors
with initial cyanide concentrations of 30 ug/L. A typical value of Kn was
-------
2.6-10"* L/(ug chl hr), which was at least an order of magnitude greater than
t
expected based on existing nitrogen-fixation kinetic data. The larger than
expected Kn supported the j.n_ vitro observation that nitrogenase will
preferential!/ reduce hydrogen cyanide rather than its normal substrate of
molecular nitrogen. Based on the amount of methane produced during the batch
tests, nitrogenase activity reduce* cyanide concentrations from 30 ug/L down
to 20 ug/L. Because few cyanide destruction processes are able to attack
cyanide at such low concentrations, the utilization of nitrogen-fixing
cyanobacteria in the treatment of trace-levels of cyanide is worth further
examination.
Based on the total cyanide biodegradation rate constants (Kjj) obtained
from batch experiments, use of nitrogen-fixing cyanobacteria in the secondary
or tertiary treatment of cyanide wastes could be a feasible process, provided
that the treatment process has adequate mean cell residence time. With a net
specific growth rate of 0.8 d~^f the unacclimated Anabaena culture can have
faster growth kinetics and trace-contaminant removal rates than existing
biological treatment processes. Because cyanobacteria are photosynthetic, the
amount of biomass in a reactor is not a function of organic substrate
concentration. Thus, the volumetric size of the reactors required to treat
low concentrations of cyanide would be small compared to traditional aerobic
processes for treating dilute wastes. For example, a chemostat model using
conservative rate and biomass parameters predicted that a hydraulic retention
time of only 5 days is required to reduce an influent cyanide concentration of
4 mg/L by 93 percent. In comparison, the 90 percent reduction of a 10 mg/L
influent BOD concentration in an aerobic chemostat would require a hydraulic
retention time of 21 days, based on typical BOD kinetic parameters (Metcalf
-------
and Eddy, 1979). However, the required size of the BOD chemostat would be
reduced if supplemental substrates were added to the reactor to increase
biomass concentrations.
Because of the potential to attack trace-concentrations of cyanide and of
the potential for low operating costs, the use of nitrogen-fixing
cyanobacteria In the biological treatment of cyanide wastes may be an
attractive alternative to exiting cyanide treatment technologies. However,
the capital costs and the requirement for trained personnel associated with
biological treatment processes, may make the cyanobacteria process unsuitable
for small-volume cyanide-waste generators. The cyanobacteria process is
probably better suited for larger cyanide treatment facilities.
-------
SECTION 3
RECOMMENDATIONS
This stud/ demonstrated the ability of nitrogen-fixing cyanobacteria to
biodegrade low concentrations of free cyanides in batch reactors. Future
studies need to examine the ability of the cyanobacteria to degrade cyanide
under steady-state conditions and to determine the stability of the steady-
state reactors to slight perturbations in cyanide concentrations. If such
laboratory-scale experiments continue to demonstrate the attractiveness of
utilizing nitrogen-fixing cyanobacteria in the treatment of low cyanide
concentrations, then a pilot-scale study should be performed to determine the
economic and technical feasibility of the process.
-------
SECTION 4
MATERIALS AND METHODS
EXPERIMENTAL OVERVIEW
The objectives of the study were 1) to determine the rates at which a
cyanobacteria culture removed free cyanides (HCN + CN~), and 2) to determine
the rate at which HCN was reduced to methane by the activity of the enzyme
nitrogenase. Rates of total cyanide removal were determined in a 1.3-liter
reactor under batch conditions. Attempts were also made to determine total
cyanide removal rates under continuous-feed conditions, i.e., in a chemostat.
The methane production experiments were conducted in small 0.037-liter, gas-
tight vials under batch conditions.
CULTURE MEDIA
The cyanobacteria used in the cyanide degradation experiments were
Anabaena sp. obtained from Carolina Biological Supply Co. (Burlington, NC,
catalog number 151710). None of the cultures used in the study were axenic,
i.e., none were free of bacteria.
The media used to maintain Anabaena cultures and used in the cyanide
degradation experiments was the Hughes-Gorham-Zehnder media described in Allen
(1973). The chemical composition of the media is shown on Table 1. To
increase the buffering capacity of the media at pH values slightly greater
than the pKa for hydrogen cyanide for media used in the batch experiments
0.186 grams of H603 was added to each liter of media (3 mN) . Adjustments in
pH were performed by the addition of either 1M NaOH or HC1. Prior to its use,
8
-------
TABLE 1. CYANOBACTERIA CULTURE MEDIA
Macronutrients
K2HP04
MgS04-7H20
CaCl2-2H20
Na2C03
NaSi03-H20
Citric Acid
EDTA
Ferric Citrate
Micronutrient Solution
Micronutrient Solution
H3B03
MnS04-H20
ZnS04-H20
(NH4)Mo7024-4H20
Co(N03)2'6H20
Na2W04-2H20
Or
KI
Cd(N03)2-4H20
NiS04-6H20
V205
A12(S04)-»'K7SOA-24H-,0
grams/liter
0.369
0.075
0.036
0.020
0.058
0.006
0.001
0.006
0.08 mL
grams/liter
3.10
1.69
0.287
0.088
0.0167
0.33
0.119
0.083
0.154
0.138
0.018
0.474
-------
the media was autoclaved. Ferric citrate was autoclaved separately from the
rest of the media to prevent the precipitation of ferric phosphates. Because
no fixed-nitrogen was added to the media, the Anabaena had heterocysts and
relied on the fixation of dinitrogen for their nitrogen needs.
Maintenance cultures were grown in autoclaved 100-ml Erlenraeyer flasks
that were stoppered with cheese-cloth-wrapped cotton plugs. The maintenance
cultures were illuminated by cool-white fluorescent light, and were not
continuously agitated nor was air bubbled through the media. Probably due to
C02 limitations, the Anabaena in the maintenance cultures had slow growth
rates.
The Anabaena used in the total cyanide removal and the methane production
batch tests were not directly transferred from maintenance culture to
respective batch reactors. Instead, an inoculum from a maintenance culture
flask was placed into a 500-mL gas wash bottle that contained fresh media.
The media was agitated by the introduction of filtered air passing through a
porous glass diffuser. In addition to providing agitation, the air bubbles
released by the diffuser provided a continuous supply of C02 to the system.
Anabaena in the gas wash bottle had rapid growth rates (doubling times on the
order of 1 day). Upon reaching a desired biomass (chlorophyll) concentration,
portions of the gas-wash-bottle cultures were transferred to the batch
experiment reactors.
EXPERIMENTAL PROCEDURES
Growth Rate Experiments
Anabaena growth rates were determined in a New Brunswick Scientific
(Edison, NJ) Bioflo Chemostat Model C32 with a Pyrex reaction vessel that held
1.3 liters of media. During the initial growth rate experiments, the reactor
10
-------
was operated in batch mode (no continuous feed) and no cyanide was present.
Temperatures were maintained at 25°C. Air was supplied at a rate of 15 L/hr.
Increases in biomass were determined by monitoring the increase in
chlorophyll-a (chl) concentrations with respect to time. The rate of increase
in biomass was assumed to be first-order with respect to biomass, i.e,
dX
— =uX (1)
dt
in which X is the chlorophyll-a concentration (ug chl/L), t is time (hr), and
u is the net specific growth rate (hr~*). The value of u was obtained by
determining the slope (least-squares fit) of the line produced when the
natural logarithm of X (In X) was plotted versus time. The amount of time
required for the value of X to double was determined from
0.693
td (2)
u
in which t
-------
Total Cyanide Removal Experiments
Objective and Apparatus—
The objective of the total cyanide removal experiments was to determine
the mixed-second order rate constants for the biologically mediated removal of
free cyanides from the 1.3-liter reactor operated in a batch mode. These
batch experiments were of short duration (4 to 5 hours), so that during the
course of the experiment any increases in Anabaena biomass due to growth would
be negligible. Initial total cyanide concentrations were approximately 3000
ug CN/L. Water temperatures were maintained at 25°C. Filtered air was added
to the reactor at a rate of 15 L/hr. and air exiting the reactor was drawn
through a gas-wash bottle containing a 1 N solution of NaOH to capture any HCN
that was volatilized from the reactor.
Cyanide Mass Balances—
The two mechanisms responsible for the decrease in total free cyanide
concentrations in the 1.3-L reactor were biodegradation of HCN and CN~ and
volatilization of HCN. Based on these two mechanisms, the mass balance
equation for the batch reactor is
dS
— - - Kb X0 S - Kra ALPHA S (3)
dt
in which S is the concentration of free cyanide (HCN + CN~) in the reactor
(ug CN/L), Kb is the mixed second-order rate constant for the biodegradation
of free cyanides by the microorganisms (L/(ug chl hr)), XQ is the
concentration of Anabaena in the reactor at the start of the batch experiment
(ug chl/L), K^a is the overall mass transfer coefficient for the movement of
HCN from the liquid phase to the gas phase (1/hr), and ALPHA is the ratio of
HCN to the total amount of free cyanide in solution,
12
-------
[HCN]
ALPHA = . (4)
[HCN] + [CN-]
The ALPHA parameter must be included in the volatilization term, because HCN
is the only form of free cyanide that is subject to volatilization. ALPHA is
a function of the pKa of HCN and of the pH of the solution. Because the pKa
of HCN at 25°C is 9.3, values of ALPHA can be obtained from (Snoeyink and
Jenkins, 1980)
10-PH
ALPHA = (5)
IO-PH + i
-------
being set at the same values. Thus, both Kb and l^a had to be determined from
the experimental data for each batch test—So, X0, S(t), pH(t), and Mf.
The values of Kb and l^a for each batch test were obtained via a trial
and error algorithm utilizing the Runge-Kutta method to solve equation (3).
First, an approximate value of KLa was determined by setting Kb to zero and
finding the K^a value that predicted the amount of cyanide collected in the
gas-wash bottle at the end of the batch experiment (Mf). This approximate
value of KLS accounted for changes in pH. Second, this approximate value of
K^a was increased and Kb values were increased until the Runge-Kutta model
approximated the observed decrease in total cyanide concentration (S) with
time. Third, values of KLa and Kb were then adjusted to yield the best
description of observed Mf and S(t) values.
Methane Formation Experiments
Objective and Apparatus—
The objective of the methane experiments was to determine the mixed-
second order rate constant for the conversion of HCN by nitrogenase to methane
and ammonia by monitoring the increase in headspace methane concentrations in
gas-tight reactors that were operated in a batch mode. The batch tests were
of short duration (1 to 2 hours), so that increases in Anabaena biomass or
changes in environmental conditions during the course of the experiment would
be insignificant. The Anabaena used in the methane production experiments had
not previously been exposed to cyanide. Initial total cyanide concentrations
were approximately 30 ug CN/L. To minimize the volatilization of HCN, Initial
pH values were adjusted to 10. Water, temperatures during the experiments were
2S°C. Methane concentrations were determined by injecting 100 uL samples of
headspace gas into a Hewlett-Packard 5340A gas chromatograph with a flame-
14
-------
ionization detector (FID).
The reactors used in methane experiments consisted of a 40-mL Wheaton
serum bottles fitted with a Supelco Mini-Valve stopper. The stopper was a
combination value and septum that provided a means of sampling the headspace
of the bottle, while being gas-tight when the headspace was not being sampled.
To maintain a uniform distribution of Anabaena in the media, a small Teflon-
coated magnetic stir-bar was placed inside the bottle and the bottle was
placed on top of a stirring plate. During the batch experiments, the bottle
contained 25 mL of media and had a headspace volume of 12 mL. The stem of the
stopper occupied the remaining 3 mL.
Kinetic Parameter Determination —
Each mole of methane appearing in the headspace of the serum bottle was
assumed to be produced from the reduction of one mole of hydrogen cyanide to
methane and ammonia by nitrogenase. Thus, the molar concentration of methane
in the headspace corresponds to a mass concentration of cyanide that was
reduced by nitrogenase,
[CH4] -- (7)
Vwater
in which P is the cyanide concentration in the water phase corresponding to
the mass of methane produced (ug CN/L), [CH^] is the molar concentration of
methane in the headspace gas (umoles Cfy/L of gas), MWCH4 is the molecular
weight of methane, MW^ is the molecular weight of the cyanide function group,
Vgas is the voiune of the headspace (0.012 L), and Vwater is the volume of the
water phase in the serum bottle (0.025 L). Because P represented another way
of describing methane concentration (the product of nitrogenase activity), the
value of P increased with time during the methane batch tests.
15
-------
The rate of methane production by nitrogenase was assumed to follow mixed
second-order kinetics,
dP
— = Kn X0 S (8)
dt
in which P is the water-phase concentration of cyanide corresponding to the
mass of methane produced (ug CN/L), t is time (hr), Rn is the mixed second-
order rate constant for the production of methane by nitrogenase expressed in
terms of chlorophyll-a concentration (L/(ug chl hr))t X0 is the concentration
of Anabaena at the start of the batch test (ug chl/L), and S is the total
cyanide concentration at time t (ug CN/L).
Because nitrogenase was not the only biological mechanism for reducing
cyanide concentrations (S) in the vials, the determination of Kn required that
the decrease in S in equation (8) reflect these additional mechanisms. The
total rate at which S decreases with time due to biological activity can be
described by
dS
*b xo s (9)
dt
in which Kjj is the mixed second-order rate constant for the biodegradation of
free cyanides by Anabaena cultures (L/(ug chl hr)). Integration of equation
(9) yields
S - S0 exp[-Kb X0 t] (10)
in which So is the initial cyanide concentration (ug CN/L). Substitution of
equation (10) for S in equation (8) yields the following equation:
dP
— =• Kn X0 S0 exp[-Kb X0 t] . (11)
dt
Integration of equation (11) from t=0 and P»PO««0 to t»t and P=P produces
16
-------
Kn So
P -- {l-exp[-Kb X0 t]J (12)
Kb
which describes methane concentration as a function of time.
The goal of the methane batch experiments was to determine the value of
KU for the nitrogen-fixing Anabaena culture. For each batch test, all of the
parameters in equation (12) were known, except for Kn. The value of the
overall biodegradation rate constant (K^) was determined from the batch
experiments examining total cyanide removal. When an independent variable
(RHS) is defined for each sampling time by
RHS-— (1 -exp[-KbX0 t]J , (13)
Kb
equation (12) is reduced to a linear relationship,
P = Kn RHS . (14)
Thus, the value of Kn for each batch test was calculated from the slope of a
linear least-squares fit of the P(t) and RHS(t) data points collected during
the batch test. The y-intercept for equation (14) should equal zero, because
ANALYTICAL PROCEDURES
Cyanide and chlorophyll measurements followed the procedures presented in
Standard Methods (APHS, 1980).
Prior to analysis, cyanide samples were distilled in a commercially-
available cyanide distillation apparatus. In the boiling flask, the cyanide
sample was subjected to acidic conditions and to a magnesium chloride reagent.
This solution was re fluxed for 1 hour. The volatilized hydrogen cyanide was
collected in a gas-wash bottle containing 1.25 N NaOH solution.
For samples with cyanide concentrations greater than 1 mg CN/L, the
17
-------
cyanide concentrations in the alkaline distillate were determined by the
titrimetric method. The distillate was titrated with a standard silver
nitrate (AgNOj) solution to form a soluble cyanide complex, Ag(CN~>2. As soon
as all the CN had been completed and a small excess concentration of Ag+ had
been added, the excess Ag+ was detected by the silver-sensitive indicator,
paradimethylaminobenzalrhodanine, which turned from a yellow to salmon color
(APHS, 1980).
For samples with concentrations less than 1 rag CN/L, the cyanide content
of the alkaline distillate was determined by the colorimetric method. First,
the pH of a distillate sample was adjusted to approximately 8 with the
addition of a phosphate buffer. Second, chloramine-T was added, which
converted the CN~ in the distillate to CNC1. Third, with the addition of a
pyridine-barbituric reagent, the CNC1 formed a red-blue dye. The absorbance
of the aqueous dye at 578 nm was linearly proportional to CN concentration in
the distillate (APHS, 1980).
The concentration of chlorophyll-a in a sample was used as an indicator
of Anabaena biomass concentration. Chlorophyll-a concentrations were
determined by the cold acetone extraction method. A known volume of the
cyanobacteria suspension with a small amount of MgC03 was drawn through a
0.45-um membrane filter. The filter was placed was placed into a centrifuge
tube, and a small known volume of a 90% acetone (v/vwater) solution was added
to the centrifuge tube. The acetone solution was allowed to extract the
chlorophyll from the Anabaena for 24 hr at a temperature of A°C. After
centrifugation, the chlorophyll-a concentration in the supernatant was
calculated from the supernatant's absorbances at 663, 645, and 630 nm (APHS,
1980). Absorbances were measured with a Beckmann Spectrophotometer.
18
-------
SECTION 5
RESULTS AND DISCUSSION
GROWTH RATE DETERMINATION
For batch growth experiments conducted at pH values ranging from 8.0 to
9.5 in the absence of cyanide, the average net specific growth rate (u) of the
Anabaena was 0.832 d~^, which corresponded to a doubling time (tj) of 20 hr.
The fastest observed net growth rate was 1.07 d"1 (td=15.6 hr); the growth
curve for this batch experiment is shown on Figure 1. These growth rate
values were comparable to growth rates described by others (Stewart, 1977;
Fay, 1983; Bothe, et. al., 1984).
The Anabaena did not survive in maintenance cultures buffered at pH
values of 10 or greater.
While the batch growth experiments yielded actively growing Anabaena
populations, attempts to develop steady-state populations of Anabaena in a
chemostat failed. The chemostat studies were performed in the same reactor
and under the same environmental conditions as the batch growth experiments,
except that a continuous feed was added to the reactor. When the reactor was
subjected to a continuous feed, the concentration of chlorophyll in the
reactor decreased at a rate faster than would be predicted based on the
hydraulic dilution rate (D) for the reactor.
A typical response between chlorophyll concentration and time after the
start of feed flow is shown on Figure 2. After 22 hr of flow, the chlorophyll
concentration in the once-through reactor increased from 220 to 300 ug chl/L.
19
-------
1000
In X • 1.071
r 2. 0.981
2.12
t (day)
6
Figure 1. Logarithmic increase in Anabaena biomass (X) as function of time,
during a batch experiment.
20
-------
300-'
200"
0 30
Time (hr)
Figure 2. Changes in Anabaena biomasa (X) versus time after the start of feed
flow through the reactor. The dotted line represents the expected
decrease in X if dilution was the only removal mechanism.
21
-------
This biomass increase was equal to the increase predicted by the average net
specific growth rate (u) minus the dilution rate (D), i.e.,
Xt = XQ exp[(u-D) t] (15)
in which Xt is the chlorophyll concentration at time t (ug/L), X0 is the
chlorophyll concentration when flow was started (ug/L), u is the average net
specific growth rate for Anabaena obtained from the batch growth experiments
(0.0347 hr'1), D is the dilution rate for the reactor (0.0211 hr'1), and t is
the time since start of flow through the reactor (hr). Thus, for the first
22 hr of feed addition, the transient chlorophyll concentrations followed the
predicted growth curve for a once-through reactor.
After 22 hr, the transient chlorophyll concentrations no longer followed
the predicted trend. Instead of continuing to increase or of plateauing at a
steady-state value, the concentration of chlorophyll in the once-through
reactor decreased with time (Figure 2). The rate of chlorophyll decrease was
1.5 times greater than the dilution rate (D).
Possible explanations for the observed decrease in chlorophyll
concentrations included an Anabaena die-off or the presence of a selective
mechanism for removing the Anabaena trichomes from the reactor. An Anabaena
die-off could have occurred, due either to cell lysis caused by the lack of
nutrients or to a rapid increase in the concentration of grazers. Because
Anabaena die-offs did not occur during the batch growth experiments, and
because the batch growth experiments were of longer duration and obtained
higher chlorophyll concentrations than the situation displayed in Figure 2,
the die-off hypothesis was not considered a strong possibility.
The selective removal mechanism explanation was based on the fact that
the flow exited from the reactor through an overflow weir. Any mechanism that
22
-------
accumulated Anabaena in the neustron or in the upper regions of the reactor
would have allowed the chlorophyll concentrations to decrease at a rate faster
than predicted by the dilution rate. Possible mechanisms included the
entrainment of the filamentous Anfbaena on rising air bubbles and the buoyancy
conferred on Anabaena due to the presence of gas vesicles. An interesting
observation was that significant portions of the Anabaena collected in the
overflow-collection vessel were buoyant. However, the observed buoyancy could
have arisen from the different environmental conditions provided in the
overflow-collection vessel.
Regardless of the mechanism, it was not possible to establish steady-
state conditions in the once-through reactor either in the presence of or in
the absence of cyanide. Chlorophyll concentrations always decreased at a rate
faster than predicted by the dilution rate, despite variations in feed flow
rate, agitation speed, air flow rate, light intensity, and method of reactor
start-up. Because the available reactor could not function as a chemostat for
the Anabaena cultures, the kinetic coefficients for the degradation of cyanide
were determined from short-term batch experiments.
TOTAL CYANIDE REMOVAL RATES
Limitations of Batch Experiments
The mixed second-order rate constants (K^) describing the biologically-
mediated decrease in total cyanide concentrations by the Anabaena cultures
were determined via short-term batch experiments. The initial objective of
the batch experiments was to provide kinetic information for the setup of the
chemostat experiments. With inability of the available once-through reactor
to establish steady-state conditions, the results of the batch experiments, as
the only source of kinetic information, became more important than originally
23
-------
intended.
The short-term batch experiments were performed with Anabaena cultures
that had not previously been exposed to cyanide. Because the cultures were
not acclimated to the presence of cyanide (i.e., the optimal concentrations of
cyanide degrading enzymes were not developed), the reported rate constants for
total cyanide removal are probably slower than would be observed for
acclimated Anabaena cultures. Another reason why the reported rate constants
underestimate cyanide removal potential is that the maximum inhibitory effects
of cyanide on respiration and enzyme activity were expressed during a batch
test (i.e., protection mechanisms and pathways were not allowed to become
fully operational). Thus, a steady-state tertiary cyanide treatment process
utilizing Anabaena should have faster rate constants for the degradation of
cyanide than the reported values.
Rate Constants for Total Cyanide Removal
The two mechanisms responsible for reducing total cyanide concentrations
during a batch test were biodegradation and volatilization. The length of the
batch tests listed on Table 2 was 4 hours, with the exception of batch test 1
which was 5 hours long. The extent of biodegradation and volatilization
varied considerably between the batch tests. With the exception of batch test
1, the extent of cyanide loss due to volatilization increased as the time
averaged pH value decreased. For example, for batch test 2 with a time-
averaged pH value of 8.44, 1112 ug CN was volatilized, while the amount of
cyanide volatilized during batch test 3 with a time-averaged pH of 9.5 was 138
ug CN. This was the expected result, because as pH dropped the fraction of
cyanide existing as volatile hydrogen cyanide increased. In contrast, the
extent of biodegradation increased as the time averaged pH value increased
24
-------
TABLE 2. EXTENT OF BIODEGRADATION AND VOLATILIZATION
IN SELECTED BATCH EXPERIMENTS
Batch
Number
1
2
3
4*
5**
Time
Ave.
PH
9.59
8.44
9.50
9.00
9.00
Initial
Biomass
*o
(ug chl/L)
412
4926
203
560
367
Total
Mass CN
Removed
(ug CN)
752
1188
443
360
583
Volatilized
Mass of
Cyanide
(ug CN)
534
1112
138
324
410
Biodegraded
Mass of
Cyanide
(ug CN)
218
76 .
305
36
173
* media for this batch test included 0.5 ppm thiosulfate
** re-exposure of biomass in batch test #4 to cyanide and
thiosulfate
(for this observation ignore batch test 5, which is a batch test examining the
potential for acclimation). For example, at a time-averaged pH of 8.44 (batch
test 2), biodegradation accounted for the removal of 76 ug CN. At a time-
averaged pH of 9.5 (batch test 3), biodegradation accounted for the removal of
305 ug CN. Batch test 3 had a greater extent of biodegradation than batch
test 2, despite it having only 4% of the biomass. These observations suggest
that pH played an important role in determining rate of total cyanide removal,
whether the removal mechanisms was predominantly by volatilization or by
biodegradation.
The biodegradation and volatilization rate constants in equation (3) were
determined from a computer algorithm. The resulting rate constants provided
25
-------
the simultaneous best fit of two sets of data: the decrease in cyanide
concentration with time in the reactor and the mass of cyanide collected in
the gas-wash bottle. The calculated Kb and Yi^a values for batch tests 1
through 4 are listed on Table 3. As shown on Figure 3 for batch test 3, the
calculated values of Kj, and K^a described the decrease in total cyanide
concentration. Similar fits of the data were obtained for the other batch
tests.
TABLE 3. OVERALL VOLATILIZATION RATE CONSTANTS (Kia) AND MIXED SECOND-ORDER
RATE CONSTANTS (KO FOR TOTAL CYANIDE REMOVAL BY UNACCLIMATED ANABAENA
CULTURES IN SELECTED BATCH EXPERIMENTS
Batch
Number
1
2
3
4*
Time
Averaged
pH
9.59
8.44
9.50
9.00
Time
Averaged
ALPHA
0.339
0.879
0.387
0.666
KLa
(hr-1)
0.17
0.22
0.052
0.075
Kb
[L/(ug chl hr)]
8.5-10-5
5.0-KT6
2. 2 -10-*
8.0- 10'6
* media for this batch test included 0.5 ppm thiosulfate
Examination of Table 3 indicates that the fitted values of Kb and
varied considerably between the 4 batch tests. Because the effect of temporal
changes in pH on volatilization rates were accounted for by the ALPHA term in
equation (3), the 4-fold variation in the calculated K^a values was due to
differences in aeration rates and agitation intensities. These differences
existed, despite the mechanical settings on the experimental reactor being the
same for each batch test.
26
-------
3300
u
53
2900
2500
O sample
— equation (3)
2 3
Tim® (fw)
Figure 3. Decrease in total cyanide concentration (S) during batch test
number 3. The line represents the numerical solution of equation
(3) using the Kb and KLa values on Table 3.
27
-------
The mixed second-order biodegradation rate constants (Kb) for the
unacclimated Anabaena cultures ranged from 5.0-10~6 to 2.2-10'4 L/(ug chl hr).
This observed 44-fold variation in the Kb values appeared to be a function of
pH. As shown on Figure 4, the biodegradation rate constant increased as the
time-averaged pH for each batch test increased. The observed relationship
between Kb and the time-averaged pH value was approximated by the following
linear regression:
log Kb «, 1.44 ^H - 17.7 (16)
r2 = 0.836
in which Kb has the units L/(ug chl hr), pH is the time-averaged pH value for
the batch test, log is the base 10 logarithm, and r2 is the square of the
linear regression coefficient.
Because Kb values increased as the time-averaged pH increased, the
observed Kb values were probably responding to ALPHA,
[HCN]
ALPHA = (4)
[HCN] + [CM~]
Since the initial total cyanide concentration for the batch tests were similar
(aprox. 3000 ug CN/L), comparing Kb to the time-averaged ALPHA value would
indicate the effect of HCN on biodegradation rates. The observed relationship
between Kb and the time-averaged ALPHA value suggested that for smaller
concentrations of HCN the rate of total cyanide biodegradation by the Anabaena
cultures Increased (Figure 5). The relationship between Kf, and ALPHA was
approximated by the following linear regression:
log Kb = -2.91 ALPHA - 2.88 (17)
r2 - 0.862
in which ALPHA was calculated from the time-averaged pH for each batch test.
28
-------
•4-
log Kb=l.44pH-l7,7
r2 » 0.836
Figure A. Relationship between log Kb [L/(ug chl hr)] versus
time-averaged pH.
29
-------
log Kh--2.91 ALPHA
-2.88
r2a 0.862
Figure 5. Relationship between log Kb [L/(ug chl hr)] versus
time-averaged ALPHA.
30
-------
The rate at which unacclimated Anabaena cultures reduced total cyanide
concentrations was optimized by reducing the fraction of the cyanide that
existed as HCN (reduced ALPHA value). This observation is consistent with the
concept that HCN is more toxic and inhibitory than CN~ (Doudoroff, et al.,
1966; Doudoroff, 1976). The ALPHA value is smallest at pH values greater than
the pKa for HCN. Operating a reactor at pH values above the pKa for HCN also
reduces the rate of volatilization. These factors suggest that the
biodegradation efficiency of a cyanide-treatment process based on nitrogen-
fixing cyanobacteria would increase as the pH of the system is increased.
This improvement in reactor efficiency with pH would occur only up to pH
values of 10, because Anabaena culture did not survive at pH values greater
than 10.
The maximum observed Kb for the degradation of cyanide by unacclimated
Anabaena cultures is larger than the mixed second-order rate constants for
existing biological treatment processes. Converting the observed K^ for batch
test 3 into units of L/(mg VSS d) where VSS is volatile suspended solids, the
maximum observed rate constant was 0.0195 L/(mg VSS d). In comparison, the
biodegradation of acetate by methanogens (i.e., anaerobic heterotrophs that
produce methane) can have a mixed second-order rate constant of 0.0054
L/(mg VSS d) (Rittmann and McCarty, 1980). Thus, the biodegradation of
cyanide by unacclimated Anabaena cultures can have faster rate constants than
the biodegradation of acetate by methanogens. The observed net specific
growth rate for Anabaena under optimum conditions was 0.832 d , while the
maximum specific growth rate for the acetate-utilizing methanogens was
reported as 0.25 d~* (Rittmann and McCarty, 1980). Because Anabaena can have
faster growth rates and contaminant removal rates than methanogens, and
31
-------
because successful wastewater treatment facilities have been developed using
methanogens-, there appears to be no kinetic limitation for the development of
biological treatment processes utilizing Anabaena. However, other factors may
limit process development, such as construction costs. As with the anaerobic
treatment of wastes with suspended microorganisms, large reactors may be
required to provide adequate cyanide removal and to ensure process stability.
Effect of Acclimation on Cyanide Removal Rates
By re-exposing an Anabaena culture to cyanide, the impact of acclimation
on biodegradation rates was assessed. Batch test 5 was a repeat of batch test
4 (Table 2). Thiosulfate (0.5 ppm) was added to the culture media used in
batch test 4 to induce the production of rhodanese, the enzyme that catalyzes
the formation of thiocyanate from free cyanide and thiosulfate. The addition
of thiosulfate was assumed to have no effect on the K^ value for batch test 4.
However, the induction of rhodanese and other detoxification pathways during
batch test 4, were expected to increase the rate of total cyanide removal
during batch test 5. Comparison of the Kj, values for batch tests 4 and 5
would indicate the difference in cyanide biodegradation kinetics between
unacclimated and acclimated Anabaena cultures.
Several hours after completion of batch test 4, aeration and agitation
were shut off in the experimental reactor. The Anabaena was allowed to settle
overnight. The next morning the supernatant was siphoned off and replaced
with new culture media containing thiosulfate. Aeration and agitation were
restarted. Two hours later, the acclimated Anabaena culture was subjected to
a second cyanide batch experiment.
The acclimated Anabaena culture reduced total cyanide concentrations 12
times faster than the unacclimated Anabaena culture (Table 4). The acclimated
32
-------
TABLE 4. COMPARISON OF MIXED SECOND-ORDER RATE CONSTANTS (Kb) FOR TOTAL
CYANIDE REMOVAL BY UNACCLIMATED (BATCH NUMBER 4) AND ACCLIMATED (BATCH
NUMBER 5) ANABAENA CULTURES IN PRESENCE OF THIOSULFATE
Batch
Number
4
5
Time
Averaged
pH
9.00
9.00
Time
Averaged
ALPHA
0.666
0.666
KLa
(hr-1)
0.075
0.085
Kb
[L/(ug chl hr)]
8.0-10-6
1.0- 10~4
Anabaena culture (batch test 5) had a mixed second-order rate constant (K^) °
1.0-10~4 L/(ug chl hr), while KJJ for the unacclimated Anabaena culture (batch
test 4) was 8.0-10~6 L/(ug chl hr).
Thus, when an Anabaena culture has time for the induction of rhodanese
and other enzymatic pathways, the rate constants for cyanide biodegradation
increase. Because the Anabaena in an operating cyanide-treatment process
would be acclimated to a steady-state cyanide concentration, the rate
constants for cyanide biodegradation would likely be faster than those listed
on Table 3.
METHANE PRODUCTION RATES
Rate Constants for Methane Production
For the methane production data shown on Figure 6, an unacclimated
Anabaena culture (Xo»1146 ug chl/L) produced 14 ug CH^/L (gas phase) in 1.75
hours when the initial total cyanide concentration in the vial was 31.2
ug CN/L (water phase). This amount of methane production corresponded to the
reduction of 11.0 ug CN/L (water phase) by nitrogenase in 1.75 hours. The
ability of nitrogen-fixing cyanobacteria to reduce cyanide concentrations
33
-------
16
3Q-T
u
I
Tim« (hr)
Figure 6. Increase in headspace methane concentration with time. For the
illustrated batch test, the initial cyanide concentration was
31.2 ug CN/L.
34
-------
below 20 ug/L may be significant, because no cyanide destruction process
has demonstrated the ability to reduce total cyanide concentrations to
levels less than 25 ug/L (Brunker, 1980). Thus, nitrogen-fixing
cyanobacteria may be well suited for use in the secondary or tertiary
treatment of cyanide wastes.
The suitability of utilizing the nitrogen-fixing capacity of
cyanobacteria to treat small concentrations of cyanide can be related to the
mechanism by which nitrogenase synthesis is regulated. Nitrogenase synthesis
is induced by the lack of ammonia (Stewart, 1977; 1980). Thus, the
intracellular concentration of nitrogenase is not a function of substrate
concentration (N2 or HCN), but is a function of product concentration
(ammonia). By promoting the active growth of cyanobacteria, ammonia is kept
in short supply, which results in the maintenance of maximal levels of
nitrogenase (Stewart, 1977), regardless of cyanide concentration. In
constrast to nitrogenase, the synthesis and maintenane of high intracellular
concentrations of rhodanese is a function of cyanide concentration (Atkinson,
1975; Atkinson, et al., 1975; Brunker, 1980). The ability to maintain maximal
activity levels of nitrogenase at low cyanide concentrations implies that
nitrogenase may be more effective in the removal of low levels of cyanide than
cyanide-induced enzymatic pathways.
The determination of the rate constant for the reduction of cyanide by
nitrogenase required the calculation of the mixed second-order rate constant
for the overall removal of cyanide from the vial. As the pH in the vial was
10, the value of mixed second-order rate constant for the total cyanide
removal rate (Kb) calculated from equation (17) was 4.33'10~* L/(ug chl hr).
This Kb value was used in equations (13) and (14) to calculate the mixed
35
-------
second-order rate constant (Kn) for the reduction of cyanide by nitrogenase.
For the methane production shown in Figure 6, the calculated Kn value was
2.58-KT4 L/(ug chl hr).
A quick comparison of the calculated Kn and Kb values would suggest that
for an initial cyanide concentration (S0) of 31.2 ug CN/L, the rate of HCN
reduction by nitrogenase accounted for 60% of the total cyanide removal rate.
One problem with this comparison is that the calculation of Kb ^ro° equation
(17) assumes that SQ 13 approximately 3000 ug CN/L. Because the value of Kb
was found to increase with decreasing HCN concentration, the use of equation
(17) in the analysis of above methane production experiment may greatly
underestimate Kb. Fortunately, the Kn value obtained from equations (13) and
(14) was relatively insensitive to changes in Kb. For inputted total cyanide
biodegradation rate constants (Kb) ranging from 2.58-10~4 (all cyanide removal
due to nitrogenase) to I.O'IO"-* L/(ug chl hr) (only 5% of original cyanide
remained in reactor after 1.75 hr), the Kn values obtained from equations (13)
and (14) only varied by a factor of 2 (Figure 7). For the inputted range of
Kb values, the ratio of nitrogenase activity to total cyanide biodegradation
rates (Kn/Kb) ranged from 0.39 to 1.00 (Figure 8). These calculations suggest
that the reduction of HCN to methane by nitrogenase is a significant mechanism
of cyanide biodegradation for unacclimated Anabaena cultures exposed to low
cyanide concentrations (SQ-31.2 ug CN/L).
As the initial concentration of cyanide increased, the mixed second-order
rate constant for nitrogenase activity (Kn) appeared to decrease. For a
methane production experiment with an initial cyanide concentration (SQ) of
400 ug CN/L and a pH of 9.9, the calculated Kn value was 2.62-10~6
L/(ug chl hr). This Kn value was 2 orders of magnitude slower than the Kn
36
-------
.0004- •
.0002- -
-3.75
—I—
-3.35
log Kb
-235
Figure 7. Sensitivity of Kn obtained from equations (13) and (14) to inputted
Kb values. Both parameters have units of L/(ug chl hr).
37
-------
i.o-
0.3
oH—
-3.73
-3.33
log
-2.95
Figure 8. Sensitivity of the ratio between Kn and Kb to inputted
Both parameters have units of L/(ug chl hr).
values.
38
-------
value calculated when SQ was 31.2 ug CN/L. The decrease in Kn with increasing
S0 was probably due to HCN inhibition of the ATP generating pathways in the
heterocysts, because cyanide at the experimental concentrations has no effect
on the rate of electron flow through nitrogenase (Li, et al., 1982). Thus,
the effect of cyanide concentrations on Kn may diminish with acclimation.
Comparison of Methane Production with Nitrogen Fixation
The observed rate of HCN reduction (methane production) by nitrogenase
was compared to literature values of dinitrogen (^) reduction rates. The
dinitrogen reduction rates were converted to HCN reduction rates based on
three assumptions. First, the rate of electron flow through nitrogenase is
independent of HCN concentration, so that the rate of electron flow for
nitrogen fixation is assumed equal to electron flow for HCN reduction (Li, et
al., 1982). This is a reasonable assumption for the experimental conditions
described above. Second, all of the cyanide inside the heterocyst is in the
form of HCN. This is a reasonable assumption, because the pH inside of a
heterocyst is 7.3 (Stewart, 1977), two pH units below the pKa for HCN. Third,
for each mole dinitrogen reduced there is the simultaneous reduction of 2
protons to make 1 mole of hydrogen gas, i.e., 8 electrons must flow through
nitrogenase for each mole of N2 fixed. This is a conservative estimate of the
moles of H2 produced per mole of ^2 re^uced by nitrogenase (Burris and
Peterson, 1978; Lean, et al., 1973; Postgate, 1982; Fay, 1983; Bothe, et al.,
1984).
Ramos, et al., (1987) reported that an Anabaena culture had a maximum
specific dinitrogen reduction rate of 30 umole N2/(mg chl hr). Based on the
above assumptions, this corresponded to a calculated maximum specific HCN
reduction rate (k) of 1.08 ug HCN/(ug chl hr). This maximum rate was adjusted
39
-------
to account for the effect of ambient HCN concentrations by the following
modified Monod equation:
k S
rHCN (18)
Km + S
in which FHCN is the specific HCN reduction rate [ug HCN/(ug chl hr)], k is
the maximum specific HCN reduction rate [1.08 ug HCN/(ug chl hr)], S is the
ambient HCN concentration (ug HCN/L), and Km is the Michaelis-Menten constant
for the production of methane from HCN by nitrogenase (121,500 ug HCN/L) as
reported by Li, et al.f (1982). This Km value included both the effect of
enzyme affinity for HCN and of H2 production on r^of. Because Km is much
larger than the ambient cyanide concentrations used in the methane production
experiments, equation (18) can be reduced to
k
rHCN = — S (W)
Km
in which k/Km has the value 8.9-10~6 L/(ug chl hr). Equation (19) is
equivalent to the following rearrangement of equation (8):
1 dP
*„ S (20)
X0dt
in which K,, is the experimentally-derived mixed second-order rate constant for
the production of methane by nitrogenase [L/(ug chl hr)]. The calculated k/Km
value was compared to the experimentally-derived Kn values.
The observed values of Kn ranged from 2.62-10~6 L/(ug chl hr) when S0 was
400 ug CN/L to 2.58-10-4 L/(ug chl hr) when SQ was 31.2 ug CN/L. The
calculated k/Km value of 8.9-10~6 L/(ug chl hr) lies between the two observed
&n values. Thus, the observed rates of methane production were reasonable
compared to known in. vivo nitrogen-fixation rates and in, vitro HCN reduction
-------
rates.
The observed Kn value for So=31.2 ug CN/L was almost 30 times larger than
the calculated k/Km value. This suggests that Anabaena may reduce low
concentrations of HCN at a more rapid rate than dinitrogen. Such a conclusion
agrees with the observation by Li, et al., (1982) that the activation
requirements for nitrogenase to reduce HCN are lower than the activation
requirements to reduce dinitrogen. Because lower activation energies usually
translate into faster kinetics, nitrogenase will preferentially reduce HCN.
Thus, if nitrogenase's requirements for ATP and electrons can be continuously
satisfied, then the enzymatic apparatus in Anabaena normally responsible for
nitrogen-fixation may play an important role in determining the rate at which
low concentrations of cyanide are biodegraded in a treatment process. In the
methane production experiments with So»31.2 ug CN/L, the rate of nitrogenase
activity was responsible for at least 39% of the total cyanide biodegradation
rate (Figure 8).
APPLICATION OF KINETIC DATA TO CYANIDE TREATMENT
The purpose of this section is to predict the strength of a cyanide waste
that can treated by a process utilizing suspended cultures of nitrogen-fixing
Anabaena. This prediction will assume steady-state conditions in a once-
through competely-mixed reactor. Because the kinetic data collected in this
study were not obtained under steady-state conditions, application of the
following discussion to the design of an operating cyanide treatment process
should be done with great caution.
Assuming mixed second-order biodegradation kinetics and ignoring
volatilization, the mass balance equation for cyanide in a completely-mixed
once-through reactor is as follows:
41
-------
dS
V_=QS1-QS-KbXS (21)
dt
in which V is the volume of the reactor (L), S is the concentration of cyanide
in the effluent and in the reactor (ug CN/L), SA is the cyanide concentration
in the influent (ug CN/L), t is time (hr), Q is the volumetric flow rate
through the reactor (L/hr), X is the concentration of Anabaena in the reactor
(ug chl/L), and Kb is the mixed second-order rate constant for the
biodegradation of cyanide by Anabaena [L/(ug chl hr)]. Under steady-state
conditions (i.e., dS/dt=0 and dX/dt=0), the influent concentration can be
described by
S4 = S [1 + Kb X 9] (22)
in which 9 is the hydraulic retention time (hr) and is equal to V/Q. Thus, by
assigning values of S, Kb, X and 9, the strength of the cyanide waste can be
estimated from equation (22).
In predicting the influent cyanide concentrations that can be treated by
an Anabaena process, the following parameters were assumed:
pH - 9.5 temperature - 25°C
X = 500 ug chl/L S - 300 ug CN/L
9 - 120 hr Kb = 0.00022 L/(ug chl hr)
The values of the above parameters are considered to be conservative. The
value of S is an order of magnitude smaller than the cyanide concentration at
which Kb was determined. The above Kb value assumes an unacclimated Anabaena
culture, under steady-state conditions Kb should be larger. The biomass
concentatlon (X) is an order of magnitude smaller than the highest
concentration observed in the experimental reactor. A 9 of 120 hr (5 days)
provides a safety factor of 10 above the average net specific growth rate for
42
-------
Anabaena. Substitution of the above parameters into equation (22) yields a SA
of 4260 ug CN/L (4.26 mg CN/L). Thus, if a steady-state Anabaena
concentration of 500 ug chl/L can be maintained in a chemostat with a
hydraulic retention time of 5 days, then the reactor should be able to reduce
an influent cyanide concentration of 4.26 mg/L to an effluent concentration of
0.3 mg CN/L, a 93 percent reduction. As the rate of cyanide biodegradation is
first-order with respect to cyanide concentration (equation 22), similar
removal efficiencies should be observed for lower influent cyanide
concentrations.
The above calculations predict that a chemostat with a hydraulic
retention time (0) of 5 days is required to achieve a 93% reduction in
cyanide. For some cyanide-waste generators, the costs associated with the
construction of a reactor large enough to hold the volume of wastes generated
in 5 days may be prohibitive. However, the volumetric size of a cyanobacteria
reactor could be reduced by using methods to make the mean cell residence time
(Oc) larger than the hydraulic retention time (0).
One possible method for increasing Oc involves the recycling of
cyanobacteria biomass in the effluent line to the reactor. The activated-
algae reactor was originally proposed by McGriff and McKinney (1971). The
success of the activated-algae reactor is dependent on the ability to separate
algal cells from the reactor effluent by settling. Despite the presence of
gas vesicles, cyanobacteria will sink when their growth is limited by fixed
nitrogen availability (Kleiner, et al., 1982). Because nitrogen-fixing
cyanobacteria only produce one mole of ammonia per cyanide reduced instead of
the two obtained from dinitrogen, the availability of nitrogen may limit
cyanobacteria growth rates in reactors used to treat cyanide wastes. While
43
-------
cyanobacteria may sink, the potentially-slow settling velocities of the
filamentous nitrogen-fixers might require clarifiers with large surface areas.
Thus, the reduction in construction costs for the activated-algae reactor
compared to a chemostat could be partially negated by the construction costs
for a clarifier.
Another method of increasing 9C would be the use of attached-growth
reactors instead of suspended-growth chemostats. By growing cyanobacteria on
attached media steady-state concentration of biomass becomes less dependent on
hydraulic flow rates. One problem with a cyanobacteria biofilm reactor is
assuring that sufficient light intensity reaches all of the cyanobacteria
biofilms. If the light availability problem can be solved, then the
combination of the large QC associated with biofilm reactors and the ability
of nitrogenase to detoxify trace-concentrations of cyanide suggests that a
nitrogen-fixing cyanobacteria biofilm reactor would be well suited for the
secondary or tertiary treatment of cyanide wastes.
In addition to providing adequate mean cell retention times, another
concern is protecting the cyanobacteria from fluctuations in cyanide
concentration. Despite the acclimation of a microbial process to a given
cyanide concentration, small short-term increases in cyanide concentrations
can be inhibitory to the microorganisms and, thus, disrupt the microbial
process (Gaudy, et al., 1982). One means of protection is to dampen the
magnitude of the influent cyanide concentration fluctuations by primary
treatment of the cyanide wastes before biological treatment. Primary
«
treatment of the cyanide wastes would also reduce the cyanide load to the
cyanobacteria process. Thus, the steady and small concentrations of cyanide
in the effluent of an alkaline-chlorination or other primary treatment process
44
-------
would be conducive to the maintenance of cyanobacteria in a secondary
treatment process. Utilization of nitrogen-fixing cyanobacteria to detoxify
the last fraction of cyanide, instead of attempting to treat all of the
cyanide by the primary process, should result in lower operating costs for
larger cyanide treatment facilities.
45
-------
REFERENCES
Allen, M.M. 1973. Methods for Cyanophyceae. In: J.R. Stein (ed.), Handbook
of Phycological Methods - Culture Methods and Growth Measurements,
Cambridge University Press, Cambridge, pp. 127-138.
APHS. 1980. Standard methods for the examination of water and wastewater,
16th edition. American Public Health Association, Washington, D.C.
Atkinson, A. 1975. Bacterial cyanide detoxification. Biotech. Bioeng.
17(3): 457-460.
Atkinson, A., C.G.T. Evans, and R.G. Yeo. 1975. Behavior of Bacillus
stearothermophilis grown in different media. J. Appl. Bact. 38: 301-
304.
Biggins, D.R., and M. Kelley. 1970. Interaction of nitrogenase form
Klebsiella pneumoniae with ATP or cyanide. Biochimica et Biophysica Acta
205: 288-299.
Bothe, H., H. Nelles, K.P. Eager, H. Papen, and G. Neuer. 1984. Physiology
and biochemistry of nitrogen-fixation by cyanobacteria. In: C. Veeger
and W.E. Newton (eds.), Advances in Nitrogen Fixation Research, Dr. W.
Junk Publishers, Boston, pp. 199-210.
Brunker, R.L. 1980. The biological degradation of cyanides by autotrophic
organisms. Proceedings of the 12th Mid-Atlantic Industrial Waste
Conference, pp. 146-151.
Burris, R.H., and R.B. Peterson. 1978. Nitrogen-fixing blue-green algae:
their Ho metabolism and their activity in freshwater lakes. In: U.
Granhall (ed.), Environmental Role of Nitrogen-Fixing Blue-Green Algae
and Asymbiotic Bacteria. Ecol. Bull. (Stockholm) 26: 28-40.
Gastric, P.A. 1981. The metabolism of hydrogen cyanide by bacteria. In: B.
Vennesland, E.E. Conn, C.J. Knowles, J. Westley, and F. Wissing (eds.),
Cyanide In Biology, Academic Press, London, pp. 233-261.
Degn, H., D. Lloyd, and G.C. Hill (eds.). 1978. Functions of alternative
terminal oxldases. Pergamon Press, Oxford, 196 pp.
Doudoroff, P., G. Leduc, and C.R. Schneider. 1966. Acute toxicity to fish of
solutions containing complex metal cyanides, in relation to
concentrations of molecular hydrocyanic acid. Trans. Am. Fish. Soc.
95: 6-22.
46
-------
Doudoroffi P. 1976. Toxicity of fish to cyanide and related compounds: a
review. USEPA Office of Research and Development, Duluth, MN, Ecological
Research Series EPA-600/3-76-038, 154 pp.
Fay, P. 1983. The blue-greens. The Institute of Biology's Studies in
Biology No. 160, Edward Arnold, Baltimore, Maryland, 88 pp.
Gaudy, A.F., E.T. Gaudy, Y.J. Feng, and G. Brueggemann. 1982. Treatment of
cyanide waste by the extended aeration process. J. Wat. Pollut. Contr.
Fed. 54: 153-164.
Fogg, G.E., V.D.P. Stewart, P. Fay, and A.E. Walsby. 1973. The blue algae.
Academic Press, London, 459 pp.
Green, J., and D.H. Smith. 1972. Processes for the detoxification of waste
cyanides. Metal Finish. Journal, August issue, pp. 229-232.
Hardy, R.W.F., and R.C. Burns. 1968. Biological nitrogen fixation. Ann.
Rev. Biochem. 37: 331-358.
Hardy, R.W.F., and E. Knight, Jr. 1967. ATP-dependent reduction of azide and
HCN by nitrogen-fixing enzymes of Azotobacter vinelandii and Clostridium
pasteurianum. Biochimica et Biophysica Acta 139: 69-90.
Haystead, H., R. Robinson, and W.D.P. Stewart. 1970. Nitrogenase activity in
extracts of heterocystous and non-heterocystous blue-green algae. Arch.
Mikrobiol. 74: 235-243.
Henry. M.F. 1981. Bacterial cyanide-resistant respiration: a review. In:
B. Vennesland, E.E. Conn, C.J. Knowles, J. Vestley, and F. Visaing
(eds.), Cyanide in Biology, Academic Press, London, pp. 415-436.
Higgins, I.J., D. Scott, and R.C. Hammond. 1984. Transformation of Cl
compounds by microorganisms. In: D.T. Gibson (ed.), Microbial
Degradation of Organic Compounds, Marcel Dekker, Inc., New York, pp. 43-
87.
Howe, R.H.L. 1963. Recent advance in cyanide waste reduction practice.
Proceedings of the 18th Industrial Waste Conference at Purdue University,
pp. 690-705.
Howe, R.H.L 1965. Bio-destruction of cyanide wastes—advantages and
disadvantages. Int. J. Air Water Pollut. 9: 463-478.
Hwang, J.C., and R.H. Burris. 1972. Nitrogenase-catalyzed reactions.
Biochimica et Biophysica Acta 283: 339-350.
Hwang, J.C., C.H. Chen, and R.H. Burris. 1973. Inhibition of nitrogenase-
catalyzed reductions. Biochimica et Biophysica Acta 292: 256-270.
47
-------
Kleiner, A.R., J. Feuillade, and M. Feuillade. 1982. Cyanobacteria blooms:
carbon and nitrogen limitation have opposite effects on the buoyance of
Oscillator!. Science 215(4540): 1629-1631.
Kobayashi, H., and B.E. Rittmann. 1982. Microbial removal of hazardous
organic compounds. Environ. Sci. Tech. 16(3): 170a-183a.
Lean, D.R.S., C.F.H. Liao, T.P. Murphy, and D.S. Painter. 1978. The
importance of nitrogen fixation in lakes. In: U. Granhall (ed.),
Environmental Role of Nitrogen-Fixing Blue-Green Algae and Asymbiotic
Bacteria. Ecol. Bull. (Stockholm) 26: 41-51.
Li, J.G., B.K. Burgess, and J.L. Corbin. 1982. Nitrogenase reactivity:
cyanide as substrate and inhibitor. Biochem. 21: 4393-4402.
McGriff, E.G., and R.E. McKinney. 1971. Activated algae: a nutrient removal
process. Water and Sewage Works 118(11): 377-379.
Metcalf and Eddy, Inc. 1979. Wastewater engineering: treatment, disposl,
reuse. McGraw-Hill, New York, 920 pp.
Peschek, G.A. 1980. Electron transport reactions in respiratory particles of
hydrogenase-induced Anacvstis nidulans. Arch. Microbiol. 125: 123-131.
Postgate, J.R. 1982. Fundamentals of nitrogen fixation. Cambridge
University Press, Cambridge, 252 pp.
Ramos, J.L., M.G. Guerrero, and M. Losada. 1987. Factors affecting the
photoproduction of ammonia from dinitrogen and water by the cynobacterium
Anabaena sp. strain ATCC 33047. Biotech. Bioeng. 29: 566-571.
Rittmann, B.E., and P.L. McCarty. 1980. Design of fixed-film processes with
steady-state biofilm model. Prog. Wat. Tech. 12: 271-281.
Silver, W.S., and J.R. Postgate. 1973. Evolution of asymbiotic nitrogen
fixation. J. Theor. Biol. 40: 1-10.
Snoeyink, V.L., and D. Jenkins. 1980. Water Chemistry. John Wiley & Sons,
New York, 463 pp.
Solomonson, L.P. 1981. Cyanide as a metabolic inhibitor. In: B.
Vennesland, E.E. Conn, C.J. Knowles, J. Westley, and F. Visaing (eds.),
Cyanide in Biology, Academic Press, London, pp. 11-28.
Stewart, W.D.P. 1977. Blue-green algae. In: R.W.F. Hardy and W.S. Silver
(eds.), A Treatise on Dinitrogen Fixation—Section III. Biology, Wiley-
Interscience, New York, pp. 63-123.
Stewart, W.D.P. 1980. Some aspects of structure and function in nitrogen-
fixing cyanobacterla. Ann. Rev. Microbiol. 34: 497-536.
48
-------
Westley, J. 1981. Cyanide and sulfane sulfur. In: B. Vennesland, E.E.
Conn, C.J. Knowles, J. Westley, and F. Vissing (eds.), Cyanide in
Biology, Academic Press, London, pp. 29-49.
Zuraft, W.G., and L.E. Mortenson. 1975. The nitrogen-fixing complex of
bacteria. Biochimica et Biophysica Acta 416: 1-52.
49
------- |