By: L.A. du Preez and J.P. Maree
Tomado de: Water Science and Technology
The discharge of effluents rich in sulphate and nitrate is of increasing concern, as
the former leads to mineralisation of surface water, and the latter is a plant nutrient. A
process is described whereby sulphate is reduced to sulphide under anaerobic conditions
with producer gas as substrate. Sulphide stripped off is then converted to elemental
sulphur by reacting it with biologically produced iron (III). The study consisted of
laboratory batch and continuous experiments, as well as pilot scale studies. Sulphate was
reduced to sulphide at a rate of 1.2 g SO4 1-1 F-1 with H2/CO as substrate, and
2.4 g SO4 1-1 d-1 with only CO as substrate.
Nitrate was converted to ammonia in the anaerobic reactor. Sulphide was removed from 2,000
mg 1-2 (as SO4) to less than 90 mg 1-1. The iron(I), produced after
redaction of iron (III) with H2S, was oxidised biologically to iron (III). The
volumetric reaction rate was 5.5 g Fe 1-1 d-1.
Sulphade; nitrate; producer gas; carbon monoxide; hydrogen; hydrogen sulphide; anaerobic; stripping; iron (II) oxidation.
The discharge of effluents rich in sulphate and nitrate is of increasing concern.
Discharging industrial effluents containing high sulphate concentrations into surface
waters contributes directly to mineralization and corrosion potential of the receiving
waters, while nitrate contributes to eutrophication and is toxic to infants and toddlers.
Sulphate originates from the use of sulphuric acid in manufacturing, chemical and
metallurgical processes, or from the oxidation of pyritic material in ore bodies under
natural conditions.
Nitrate originates from domestic wastewater, on from fertilizer and explosives
manufacturing. Unacceptably high concentrations of sulphate and nitrate may occur in
cooling water due to evaporation.
Generally, sulphate levels less than 200 to 500 mg 1-1 are acceptable for
discharge into public streams.
Nitrate is acceptable at levels of less than 10 mg 1-1. The Municipality of
Johannesburg and water authorities in this region (Transvaal) allow the discharge of water
containing high sulphate concentrations (e.g.2000 mg 1-1) into sewer systems or
rivers. The main reason for this is that the ratio of sulphate-rich water produced by
industrial activities from industries (e.g. mining, chemical, and power stations) to
surface water, is high in that region. It is expected that as soon as proven technologies
are available for the removal of sulphate at an acceptable cost, legislation will be
enforced to prevent the discharge of high sulphate concentrations into the receiving
waters.
Sulphate and nitrate can be removed from water by desalintaion processes such as
reverse osmosis and ion exchange, but these are costly. Hence, increasing attention has
been given to biological sulphate removal (e.g.Maree and Strydom, 1985; Maree et al.,
1986) and biological nitrate removal (Matais et al. 1983).
Sulphate can be removed biologically provided that a suitable energy source is available,
e.g. lactic acid (Middleton and Lawrence, 1977), wood dust and sewage sludge (Butlin et
al., 1949, 1960; Conradie and Grutz, 1973; Knivett, 1960; Sadana and Morey, 1962. Tuttle
et al., 1969). Good sulphate removal was obtained for all the carbon sources, but a long
retention time of 5-10 days was required for the latter two.
Upflow packed bed reactors were used by Maree and Strydom (1985) to establish well
developed microbial biofilms for sulphate remoal from mine water with either sugar, pulp
mill effluent or sewage as energy sources. They concluded that 1.6 g sugar, 16.7 ml of
spent sulphide liquor or 172 ml raw sewage sludge were necessary to remove 1 800 mg
sulphate.
Maree et al. (1986) showed that a three stage process (anaerobic - aerobic) employing upflow packed bed reactors) for anaerobic treatment, and an activated sludge system for aerobic treatment, could be used for producing reusable water from mining effluents. Sulphate was reduced from 2.5 g 1-1 to less than 0.5 g 1-1 with concomitant removal of H2S, carbonates, complexed cyanides, phenol and heavy metals. Molasses was used as the energy source. Olthof et al. (1985) described the "Biosulfix" process in which H2S gas produced during sulphate reduction is absorbed in sodium hydroxide and converted to sodium bisulphide. The latter product can be reused in pulp manufacturing . Cork (1982) proposed that H2S and CO2 removal from waste gases could be achieved through the use of Chlorobium thiosulfatophilum, a green photosynthetic sulphur bacterium. He susggested that such a biological process has much potential and that sulphur would be a by-product. Buisman (1989) showed that H2S produced during biological sulphate reduction can be oxidised only to elemental sulphur (and not to sulphate) provided that the oxygen level is kept low.
Maree (1989) described a process whereby sulphuric acid-rich water, rich in heavy metals, can be treated biologically for sulphate removal, without prior neutralisation with an alkali such as lime. This is possible because alkalinity is produced in the biological sulphate process. By contacting metal-containing water with H2S gas (stripped from the anaerobic reactor), heavy metals can be precipitated as metal sulphides. The heavy metals can in addition be precipitated selectively through pH control. The pH of the feed water can be reaised by recirculation of a side stream of the highly buffered water from the anaerobic reactor. The principle of selective recovery of heavy metals was also demonstrated by Hammack (1992) who showed that copper, sinc and iron can be separated selectively by bubbling H2S gas through three reactors, connected in series, at pH values of 1.6, 3.8 and 6.2 respectively. The pH in this case was controlled by dosing ammonia.
A major disadvantage when sulphate is removed biologically from an effluent with an organic carbon source as substrate, is the high residual organic carbon content, which requires downstream treatment after the anaerobic reactor. This disadvantage was overcome by replacing organic carbon sources with producer gas (mixture of H2, CO and CO2) (du Preez et al., 1991). Another benefit of using producer gas over organic carbon sources is that it can be produced in sufficient quantities for the treatment of large flows, as coal is the raw material.
The purpose of this investigation was to demonstrate the technical feasibility of an integrated process, consisting of the following stages: sulphate reduction, H2S-stripping, H2S-absorption in an Fe (III)-solution for sulphur production, and biological oxidation of Fe (II) (Fig.1). The specific aims of the study were to determine:
- the rate of sulphate reduction at pilot scale using hydrogen as energy
source;
- whether sulphate and nitrate can be removed simultaneously in an anaerobic
reactor using carbon monoxide as energy
source;
- the efficiency of H2S removal from solution through stripping;
- the rate of irong (II) oxidation to iron (III).
Fig.1. Flow diagram of the biological process for sulphate and
nitrate removal
and the recovery of sulphur
Sulphate and Nitrate Removal
Figure 2 shows a schematic diagram of the pilot plant. The anaerobic reactor had a diameter of 500 mm and a height of 4200 mm. It was filled with 200 1 of pelletized ash as support medium for bacterial growth. The void ratio of the pebble medium was 50%. A biologically active film was established on the pebble medium by inoculating the reactor with anaerobic sludge from a sulphate reducing laboratory plant.
Fig.2. Schematic diagram of pilot used for biological sulphate removal.
The temperature was kept constant at 35°C by means of a thermostat. Hydrogen gas was introduced into the anaerobic reactor by recirculation of a side stream downwards, over packing material, through a pressure vessel, filled with hydrogen gas at a pressure between 200 and 500 kPa. The water leaving the pressure vessel was saturated with hydrogen. The flow rate of the water recycled through the pressure vessel was controlled by a valve. The amount of hydrogen fed was a function of the flow rate through the pressure vessel and the pressure. Carbon monoxide was introduced in a similar way. Water was fed from the anaerobic reactor to the pressure. Carbon monoxide was introduced in a similar way. Water was fed from the anaerobic reactor to the pressure vessel with a positive displacement pump (mono pump). The pump was activated and stopped by lower and higher level controls in the pressurre vessel respectively. Feedstock was pumped from a 1500 1 holding tank to the anaerobic reactor at a rate between 40 and 120 1 d-1. The chemical composition ofthe feedstock was as follows: NaF, 3 mg 1-1; SrCl26H2O, 20mg 1-1; H3bo3, 30 mg 1-1 KBr, 100 mg 1-1; KC1, 700 mg 1-1; CaCl26H2O, 200 mg 1-1; Na2SO4, 2000; MgCl26H2O, 500 mg 1-1; NaSiO2,9H2O, 20 mg 1-1; Na4. EDTA, 1mg 1-1; NH4Cl. 1000 mg 1-1; K2HPO4, 500 mg 1-1; gypsum, 2500 mg 1-1.
The pilot plant was run for a period of 50 days at a hydrogen:carbon monoxide ratio of 90:10 days at a ratio of 50:50, and for the remaining period at a ratio of 90:10 once more.
In parallel with the pilot plant, a laboratory plant was also operated in a similar way, but with only carbon monoxide as energy source. The reactor had a diameter of 150 mm and a height of 1500 mm.
Samples were taken daily and analyzed for sulphate, sulphide, alkalinity, calcium and pH. Complete analyses were carried out every fortnight.
The reactor was manufactured from stainless steel and had a volume of 500 ml. It was filled with 10 mm diameter raschig rings as support medium for bacterial growth. The rings had an active biofilm as they were taken from a laboratory sulphate reducing plant. The temperature was kept constant at 35° C by means of a thermostat. Carbon monoxide was fed from a gas cylinder. The pressure was kept constant for the course of an experiment at a specific pressure, except for short periods at specific intervals, when the gas was replaced with fresh carbon monoxide, in order to replace the produced H2S. The produced gas was bubbled through an iron (III) solution for H2S-trapping. After each experiment, the biomass was allowed to settle, for use in the next experiment. The following solutions were used as feedstock during the batch studies:
- The feedstock used for the pilot plant sutidies.
- An industrial effluent, rich in both sulphate and nitrate.
Samples were taken at regular intervals and analysed for sulphte, sulphide, nitrate, alkalinity, calcium and pH.
Batch studies were carried out to determine:
- Whether sulphate and nitrate can be removed simultaneously in an
anaerobic reactor using carbon monoxide as energy
source.
- The effect of pressure on the rate of sulphate reduction.
H2S Stripping
The reactor was filled with 25 mm diameter raschig rings. A fan (fan diameter = 120 mm; gas flow = 2.7 kl h-1; pressure = 240 mm water) was positioned in the top of the reactor to suck air through from the bottom, counter current to the flow of the H2S-rich water. The water was recycled between the H2S-stripping reactor (empty volume = 30 l) and a CO2-contacting (reactor volume = 10 1). CO2 was introduced into the water in the contacting reactor from a CO2 cylinder via a diffuser. The purpose of feeding CO2 was to reduce/maintain the pH below 7, where sulphide is mainly in the form of H2S. H2S-rich water was recycled at a rate of 1.8 1 min-1, while CO2 was fed at rates between 0.5 and 4.5 1 min-1. A synthetic solution, containing 2000 mg 1-1 H2S (as SO4) and 2000 mg 1-1 calcium (as SO4), was used. Sulphide removal was monitored by operating the system in a batch mode. Samples were analyzed at regular intervals for sulphide and pH.
Iron (II) Oxidation
Figure 3 is a schematic diagram of the reactor used for iron (II)-oxidation. The reactor had a diameter of 440 mm and a height of 620 mm. The water was aerated by blowing air through a diffuser using a compressor. A synthetic solution was used as feedstock, containing 33.4 g l FeSO4 7H2O, , 0.31 G 1-1 KH2PO4, O.39 G 1-1 MgSO4 and 0.39 mg 1-1 (NH4)2SO4. The solution was made up using tap water. The reactor was inoculated with acid mine water, rich in iron oxidising bacteria. The process was run in batch mode using the following steps.
- The reactor was filled with the solution, plus an inoculum of
iron-oxidising bacteria.
- Aeration was applied continuosly while filtered samples were taken at regular
intervals and analyzed for iron(II), acidity and
pH.
- When the iron (II) was completely oxidised to iron (III), aeration was stopped and
50% of the iron (III) solution was
replaced with a fresh iron (II) solution.
- Aeration was commenced and the procedure described above repeated.
Fig.3. Removal of sulphate using H2 and CO as energy (pilot scale results).
Analytical
Manual determinations of sulphate, sulphide, alkalinity, calcium and pH were carried out according to analytical procedures as described in Standard Methods (APHA, 1985). With the exception of sulphide, all the analyses were carried out on filtered samples. Acidity removal over the stripping reactor was measured by titrating 5 ml of the feed and treated solutions with 0.1 N NaOH to a specific pH.
Sulphate and Nitrate Removal
Sulphate removal at pilot scale. Figure 3 shows results obtained when sulphate was removed over a period of 80 days in the pilot plant with H2 and CO as substrate. Sulphate in the treated water decreased gradually from 2000 to 100 mg 1-1 (as SO4), except on two occasions. At day 58 the sulphate content increased to 2200 mg 1-1 because the ratio of H2/CO was changed from 90:10 to 50:50. Good sulphate removal was rapidly restored after the ratio was changed back to 90:10. This demonstrates that a long enough adaptation period must be provided after introduction of a new substrate composition. The decrease in sulphate removal on day 75 is due to a leak in the hydrogen feed line.
The volumetric sulphate reduction rate was 1.2 g SO4 1-1 d-1. As the volatile suspended solids (VSS) concentration was 1.9 g 1-1, the specific sulphate reduction rate was 0.63 g SO4 g VSS-1 d-1.
Figure 4 shows when pure CO is used as substrate for sulphate reduction in a laboratory plant. The maximum volumetric rate of sulphate reduction achieved was 2.4 g SO4 VSS-1 d-1, which is higher than when a mixture of H2/CO was used.
Fig.4. Removal of sulphate using only CO as energy source (laboratory scale results)
The reactions responsible for sulphate reduction when H2 and CO are used as energy sources are the following:
4H2 + SO42- --> S2- + 4H2O
4CO + SO42- --> S2- + 4CO2
Organisms responsable for reactions (1) and (2) could be. Desulfovibrio vulgaris (sub-specie Marburg) (Badziong and Thauer, 1978) and Desulfovibrio desulfuricans (Yagi and Tamiya. 1962) respectively.
From reactions (1) and (2), it is calculated that stoichiometrically 0.93 1 H2 or CO (equivalent to 0.083 g H2 and 1.17 g CO) is required for the reduction of 1 g SO4 to sulphide. The utilisation efficiencies of CO and a mixture of H2/CO were determined experimentally to be 1.27 1 CO g SO4-1 and 4.15 1 (CO + H2) g SO4 -1 CO was better utilised than H2 for sulphate reduction, possibly due to the fact that it has a higher solubility compared to H2 (23.2 ml CO 1-1 water), and because greater leakages of H2 (through the pipe lines) than CO, because of its small molecular size.
Effect of pressure. The aim of the study was to determine conditions required for maximum rate of sulphate reduction. From equations 3 and 4 is clear that this goal could be achieved by the biomass concentration in the reactor as well as the concentration of the substrate. The reaction rate (dx/dt) of biological processes is directly proportional to the biomass concentration (X) in the reactor (Pohland, 1992)
dx / dt = uX
where u = specific microorganism growth rate.
According to the Grau model (Grau, et al., 1975), the specific growth rate (u) is a function of the influent and effluent growth-limiting substrate concentrations (S0 and S).
Where: umax - maximum specific substrate utilization rate
b - specific
microorganism decay rate
The solubility of gases (H2 and CO) can be increased by increasing the pressure, as the solubility of a gas is directly proportional to pressure according to Henry/s law.
PA = HcA
Where Pa - partial pressure
H - Henry's
coefficient
cA - gas
concentration in solution
The results in Fig. 5 confirm that higher pressures, and accordingly a higher concentration of CO gas in solution, lead to higher reaction rates. It is expected that the reaction rate will increase further for a specific system when it is operated continuously under higher pressures, as the higher substrate concentration will support a higher biomass concentration in the reactor as illustrated by the following equation:
Where
X -
microorganism concentration
Õ - hydraulic retention time
(HRT)
Õc - microbial solids retention time
(SRT)
Y - growth yield coefficient
S0'S - influent and effluent concentration of
growth-limiting substrate
b - specific
microorganism decay rate
Simultaneous sulphate and nitrate removal. Figure 6 shows the results when
an industrial water, containing both sulphate (1650 mg 1-1) , was treated batchwise with
CO as substrate. This batch experiment was carried out in a system that was formerly used
for the treatment of only sulphate-rich water,.
It is noted that sulphate, which is expected, as nitrate is a stronger oxidising agent
than sulphate. This finding is in line with reported maximum specific growth rate (mmax),
namely 0.33 day -1 for sulphate (acetic acid as substrate) (Middleton and Lawrence, 1977)
and 1.8 day -1 for nitrate (methanol as substrate) (Stensel et. al., 1973). It was shown
by Mitchell et al. (1.8 day -1 for nitrate (methanol as substrate) (Stensel et. al.,
1973). It was shown by Mithell et al. (1986) that nitrate can be reduced by Desulfovibrio
natural isolate FBA20a. The rate of reduction when using this isolate is high compared to
other isolates of Desulfovibrio. Ammonia was the denitrification product, as was the case
in the studies described by Mitchell et al. (1986). This is confirmed by the increase in
the NH4+ concentrations with time. The reaction for the reduction of nitrate to
ammonia can be presented as follows:
NO3- + 4CO + 2H + H20 --> NH4 + 4CO2
Fig.6. Removal of both, sulphate and nitrate, using CO as energy source (batch results).
Fig.7. Removal of sulphide by stripping with air and CO2
H2S Stripping
Figure 7 shows that H2S can be removed from a level of 200 mg 1-1 to less than 100 mg 1-1 by stripping with air and CO2. Reaction times of 23 min. 28 min and 50 min and 50 min were required for CO2 flow rates of 0.5 1.5 and 4.5 min -1 respectively. The molar rations of CO2 fed after 21 min to H2S were initially 2.4, 7.2 and 21.6 respectively. The pH was 7.8 after H2S stripping and 6.9 after acidification with CO2.
Kinetics of Biological Iron (II) Oxidation
H2S which is produced in the anaerobic stage and removed from solution in the stripping stage, can be converted to elemental sulphur by oxidation with iron (III).
H2S + SFe3+ --> S + 2Fe2+ +2H+
The produced iron (II) is oxidised to iron (III) in the presence of acidophilic iron-oxidising bacteria, such as Ferrobacillus ferrooxidans, which acts as a catalyst.
2Fe2+ ½O2 + 2H + --> 2Fe3+ + H2+
The rate at which iron (II) is oxidised is influenced by factors such as iron (II) concentration (Fig.8), oxygen concentration, reactor depth (Fig.9) bacterial concentration, and temperature.
Fig.9. Effect of reactor depth on the rate of iron (II) oxidation
1. Sulphate can be reduced to sulphide at a rate of 1.2 g SO4 1-1 d-1 with H2/CO s substrate, and 2.4 g SO4 1-1 with only CO as energy source.
2. Nitrate is converted to ammonia in the anaerobic reactor
3. Sulphide can be removed from 2000 mg 1-1 (as SO4) to less than 90 mg 1-1 in a reaction time of 20 min.
4. Iron (II) can be oxidised to iron (III) biologically. The volumetric reaction rate was 5.5 g Fe 1-1 d-1.
The authors express their thanks to the following people.
- Dr. Matthias Graff, Institut für Mikrobiologie, der Technischen Universität
Braunschweig, Germany, for valuable interaction
during the course of the project during his visit to the CSIR (May-July
1993).
- Miss Julie Rackstraw, Farnborough College of Technology, Farnborough, England,
for conducting iron (II) oxidation studies
at the CSIR, s part of her training for a Higher National Diploma
(June-July 1992).
APHA (1985). Standard Methods for the Examination of Water and Wastewater Treatment,
12th Ed. American Public Health Association, New York.
Badziong, W. and Thauer, R.K. (1978). Growth yields and growth rates of Desulfovibrio
vulgaris (Marburg) growing on hydrogen plus sulfate and hydrogen plus thiosulfate as the
sole energy sources. Arch. Microbiol, 117,209-214.
Buisman, C.J.N. (1989), Biotechnological sulphide removal with oxygen, PhD Thesis,
Agricultural University of Waginingen.
Byrgess, S.C. and Wood, L.B (1961). Pilot plant studies in production of sulphur from
sulphate enriched sewage sludge, J. Sci Food Agric., 12 326-341.
Butlin, K.K. Adams, M.E.and Thomas, M. (1949). The isolation and cultivation of sulphate
reducing bacteria. J. Gen. Microbiol., 23, 158.
Butlin, K.K., Selwyn, S.C. and Wakerby, D.S. (1960). Microbial sulphide production from
sulphate-enriched sewage sludge. J. Appl. Bact., 23, 158.
Conradie, P.J.A. and Grütz, P.W.E. (1973). The treatment of acid mine waste in admixture
with raw sewage sludge in an anaerobic digester. Report to the Chamber of Mines (File No.
W6/534/3). National Institute for Water Research, Pretoria (Restricted).
Cork, D. J. (1982). Acid waste bioconversion - an altemative to Clauss desulpbhrization
process. Dev. Ind. MicrobiaL, 23, 379-387.
du Preez, L.A., Maree, J.P. and Jackson-Moss, C.A. (1991). Biological sulphate removal
from mining effluents by utilising producer gas as energy source. In: Proc. of the Int.
Mine Water Conf., Ljubliana, Yugoslavia. Sept.
Grau. P., Dohanyas, M. and Chudoba, J. (1975). Kinetics of multicomponent substrate
removal by activated sludge, Water Res., 9, 637-642.
Hammack, R. W. (1992). U S Bureau of Mines, Pittsburg, Pennsylvania, Personal
communication.
Knivett, V. A. (1960). The microbiological production of Vitamin B12, and
sulphate from sewage. Prog. Indusi. Microbiol., 2, 27.
Marais, G. v. R., Loewenthal, R. E. and Siebritz, I. P. (1983). Observations supporting
phosphate reinoval by biological excess uptake - review. Wat. Sci. Tech., 15(8/9),
15-41.
Maree, J. P. and Strydom, W. F. (1985). Biological sulphate removal from a packed bed
reactor, Water Res., 19(9), 1101-1106.
Maree, J. P., Gerber, A. and Strydom, W. F. (1986). A biological process for sulphate
removal from industrial effluent. Water SA, 12(3), 139-144.
Maree, J. P. (1989). Biological treatment of water, Unifed States Patent No.
4,839,052, 13 June (Filed: 23 June 1988).
Maree, J. P. and Hill, E. (1989). Biological removal of sulphate from industrial effluents
and concomitant production of sulphur.
Wat. Sci. Tech., 21(4/5), 265-276.
Middleton A. C. and Lawrence, A. W. (1977). Kinetics of microbial sulphate reduction. J.
Wat. PolL Contr. Fed., 1659-1670. Mitchell, G. J., Jones, J. G. and Jeffrey, A. C.
(1986). Distribution and regulation of nitrate and nitrite reduction by Desulfovibrio
and Desulfotomaculum species. Arch. Microbial., 144, 35-40.
Olthof, M., Kelly, W. R., Oleszkiewicz, J. and Weinreb, H. (1985). Development of
anaerobic treatment process for wastewaters containing high sulfates. Presented at Purdue
Industrial Waste Conference, May. 16.
Pohland, F. G. (1992). Anaerobic treatment: Fundamental concepts, applications, and new
horizons, In: Design of anaerobic processes for the treatment of industrial and
municipal wastes. Edited by Malina, J. F. and Pohland, F. G., Pennsylvania, pp 19-20.
Sadana, J. C. md Morey, A. V. (1962). Microbiological production of sulphide from gypsum. 1.
Scient. Indust. Res. 21C, 124.
Stensel, H. D., Loehr, R. C. and Lawrence, A. W. (1973). Biological kinetics of suspended
growth denitrification. Jour. Water PolL Contr. Fed., 45, 249-261.
Tuttle, J. H., Dugan, P. R. and Ranles, C. I. (1969). Microbial sulphate reduction and its
potencial utility as an acid mine water
pollution abatement procedure. Appl MicrobioL, 17, 297-302.
Umita T., Nenov, V., Omura, T., Aizawa, J. and Onuma, M. (1988). Biological ferrous-iron
oxidation with fluidized bed reactor. Water Pollution Control in Asia (Adv. Wat.
Pollut. Control no. 6),, pp. 479-485.
Yagi, T. and Tamiya, N. (1962). Biochim. Biophys. Acta, 65, 508-509.
Date upgrated
Mar/30/99
Coments to Webmaster
[ Homepage CEPIS
]