ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES IN COMPOST BIOFILTERS ERIK SMET, HERMAN VAN LANGENHOVE∗ and KATRIEN MAES Department of Organic Chemistry, Faculty of Agricultural and Applied Biological Sciences, University of Ghent, Coupure Links 653, B-9000 Gent, Belgium (∗ author for correspondence, e-mail:
[email protected])
(Received 1 September 1998; accepted 6 May 1999)
Abstract. The performance of lab-scale compost biofilters for the purification of waste gases containing high (>70 mg m−3 ) ammonia concentrations was studied. When using fresh compost material, no effect of inoculating the compost material with a nitrifying culture was observed since high elimination capacities (up to 350 g NH3 m−3 d−1 ) were obtained in both the inoculated and the non-inoculated biofilter. Due to the physico-chemical interaction of NH3 with the compost material at the start of the experiment, no microbiological start-up period was observed and high removal efficiencies were obtained from the first day on. Next to this, no NH3 -toxicity was observed even at concentrations up to 550 mg NH3 m−3 . About 50% of the NH3 -removal was found to be nitrified, while the other 50% remained in the biofilter as NH+ 4 . As a result of this, no acidification of the carrier material was observed and NH4 NO3 accumulated in the biofilter. Due to osmotic effects, however, a complete inhibition in nitrification and NH3 -removal was obtained at a measured NH4 NO3 -concentration in the compost material of 6–7 g N kg−1 , corresponding to a cumulative NH3 -removal in the biofilter of ±6000 g m−3 . Finally, it was illustrated that the removal of the odorant dimethyl sulfide (Me2 S) in a Hyphomicrobium MS3-inoculated compost biofilter is completely inhibited due to NH3 -toxicity at a waste gas concentration of 100 mg NH3 m−3 . Next − to this, the NH+ 4 - and NO3 -concentrations in the compost material that were shown to inhibit the nitrification, also strongly affected the Me2 S-degrading activity of Hyphomicrobium MS3. Keywords: ammonia, biofilter, dimethyl sulfide, inhibition, inoculation, nitrification
1. Introduction The emission of ammonia-loaded waste gases is encountered in different bio-industrial processes, as e.g. rendering and composting plants (Lipski et al., 1994; Williams, 1995). In composting plants, ammonia is produced from either the aerobic or anaerobic decomposition of proteins and amino acids (Haug, 1993). During the aerobic composting of biowaste, concentrations in the waste gas up to 227 mg NH3 m−3 and a cumulative emission of 152 g NH3 ton−1 biowaste were reported (Smet et al., 1999), while NH3 -concenctrations up to 700 mg m−3 were found in exhaust gases from sludge composting (Haug, 1993). Although ammonia has a rather high odour threshold value (OTV) of 2.7 mg m−3 (Weckhuysen et al., 1994) and dilutes rapidly to below detection away from the composting facility (Haug, 1993), its emission deserves major attention because it contributes to soil acidificaWater, Air, and Soil Pollution 119: 177–190, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.
178
E. SMET ET AL.
tion (Buijsman, 1987). With regard to the odour emission from composting plants, volatile organic sulfur compounds as e.g. dimethyl sulfide (OTV 1.6–103 µg m−3 ) and dimethyl disulfide (OTV 0.4–14 µg m−3 ) (De Zwart and Kuenen, 1992) deserve major attention (Pöhle and Kliche, 1996; Smet et al., 1998). In Flanders, all biowaste composting plants use biofilters to control the emission of ammonia and odorous compounds. There is, however, a lack of knowledge about the operational limits of this biotechnique when treating odorous waste gases containing high NH3 -concentrations. Due to its rather low Henry constant H20 ◦ C of 5.6 × 10−4 (Perry and Green, 1984) and its pH-dependent protonation, ammonia in biofilters is partly retained by physico-chemical processes. According to some authors (Shoda, 1991), ammonia in biofilters is mainly removed by adsorption onto the carrier material and by absorption into the water fraction of the carrier material. Next to this, others (Togashi et al., 1986; Cho et al., 1992) reported a chemical NH3 -removal in a biofilter loaded with NH3 and sulfur compounds due to the neutralisation reaction of NH3 with the metabolite sulfuric acid. Although recently the successful inoculation of a NH3 -loaded biofilter with the heterotrophic Arthrobacter oxydans CH8 strain was reported (Chung et al., 1997), and a heterotrophic nitrifier and aerobic denitrifier Alcaligenes faecalis strain was isolated from a full-scale NH3 -loaded biofilter (Lipski et al., 1994), most authors (Terasawa et al., 1986; Van Langenhove et al., 1988; Weckhuysen et al., 1994) suggest that microbiological ammonia removal in biofilters is achieved through nitrification by the autotrophic bacteria Nitrosomonas and Nitrobacter. Using a wood bark biofilter, maximum elimination capacities of 87 g NH3 m−3 d−1 and 7–27 g m−3 d−1 were obtained at NH3 -concentrations of 3–12 mg m−3 (Van Langenhove et al., 1988; Weckhuysen et al., 1994). In a pH-neutralised and inoculated peat biofilter, a 95% removal efficiency was reported when NH3 -concentrations were less than 14 mg m−3 and the loading rate was less than 43 g NH3 m−3 d−1 (Hartikainen et al., 1996). Sprinkling water on top of the biofilter was necessary in order to maintain humidity in the filter and to prevent toxic ammonia, nitrite and nitrate concentrations. While some authors (Heller and Schwager, 1996) reported that compost biofiltration of NH3 resulted in acidification of the carrier material, it was found by others (Don, 1985) that about 50% of the ammonia was nitrified and the other 50% remained in the filter as ammonium. As a result of this equilibrium between nitrification and NH3 -sorption, no acidification was observed. Due to toxification, the biofilter removal efficiency for NH3 was reported to drop drastically at a waste gas concentration level exceeding 45–50 mg NH3 m−3 (Don, 1985; Hartikainen et al., 1996). Next to this, inhibition of Nitrobacter species was found to occur when the free ammonia concentration in the biofilter percolate exceeded a value of 1 mg L−1 (Weckhuysen et al., 1994). The scope of this work was to evaluate the operational limits of a compost biofilter to purify waste gases containing high (>70 mg m−3 ) NH3 concentrations.
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
179
Next to this, the effect of these high NH3 concentrations on the odour removal potential of a compost biofilter was studied, using dimethyl sulfide as the model odorant compound.
2. Materials and Methods 2.1. N ITRIFYING
CULTURE
Fresh biowaste compost was used as inoculum for the enrichment of nitrifying micro-organisms. About 10 g of compost was added to a 5 L mineral medium solution. The mineral medium contained (in g L−1 ): K2 HPO4: 0.87; KH2 PO4 : 0.68; MgSO4 ·7H2 O: 0.25; CaCl2 ·2H2 O: 0.74 × 10−3 ; FeSO4 ·7H2 O: 2.5 × 10−3 ; CuSO4 : −3 0.08 × 10−3 . The NH+ 4 -N load was gradually increased from 0.02 to 0.2 kg m d−1 by adding NH4 Cl. Every day, the pH was readjusted to 7.5 and 1 L of mineral medium was renewed by a fill and draw procedure. 2.2. H YPHOMICROBIUM MS3 The enrichment procedure for Hyphomicrobium MS3 was described in a previous paper (Smet et al., 1996). The culture is converting dimethyl sulfide (Me2 S) stoichiometrically into H2 SO4 and is capable of converting several other sulfur compounds, as e.g. methanethiol and dimethyl disulfide. The mineral medium used for cultivation contained (in g L−1 ): K2 HPO4: 3.00; KH2 PO4 : 3.00; NH4 Cl: 3.00; MgSO4 ·7H2 O: 0.50; FeSO4 ·7H2 O: 0.01; final pH 7.0. 2.3. B IOFILTER
SET- UP
The biofilter consisted of a Plexiglass column with an internal diameter of 0.195 m. Gas sampling points were provided in the influent and effluent and at a height of 25 and 50 cm of the filled part of the biofilter. At a height of 25 and 50 cm, circular rings with internal diameter 0.135 m and outer diameter 0.195 m were installed in order to minimise preferential wall air flow. The air was humidified in a scrubber before adding the volatile compounds and before entering the biofilter. The compost used as a carrier material was not older than 2 months and was produced from source-separated municipal organic waste by the so-called double process, i.e. it was treated in an anaerobic thermophilic digestor and subsequently subjected to an aerobic treatment (Gellens et al., 1995). For the experiments with Me2 S, 36% (w/w) of dolomite particles (sieve fraction between 1.40 and 4.76 mm) were mixed with the compost biofilter material to control acidification due to accumulation of sulfuric acid. Ammonia was dosed using a cilinder containing the liquefied gas, a back pressure regulator and a flow controller. Me2 S was dosed by bubbling a calibrated N2 -flow through a thermostated gas bubbling bottle containing the pure liquid compound.
180
E. SMET ET AL.
2.4. BATCH
EXPERIMENTS
For the batch experiments, 10 mL of the Hyphomicrobium MS3 culture, pregrown on Me2 S, was brought in penicillin bottles (120 mL). Different concentrations (0– − − 10 g N L−1 ) of NH+ 4 , NO2 or NO3 were prepared by adding KNO3 , NaNO2 or NH4 Cl and the pH was readjusted to 6.6–6.9. The penicillin bottles were sealed with Teflon-lined Mininert valves (Alltech Ass.) and the culture was magnetically stirred. A 620 ppmv Me2 S headspace concentration (H25 ◦ C Me2 S = 0.07) (Przyjazny et al., 1983) was prepared in these penicillin bottles by injecting 60 µL of a Me2 S emulsion in water (6.8 g L−1 ). The initial Me2 S-degrading activity of Hyphomicrobium MS3 in the N-supplemented samples was monitored by headspace analysis and related to the control to obtain the relative activity. 2.5. A NALYSIS
OF VOLATILES
Ammonia was analysed using Gastec detector tubes (3L, 3LA and 3M). These tubes have an accuracy tolerance of 25%. Analysis of Me2 S was carried out with a Varian 3700 gas chromatograph, equipped with a flame ionisation detector (flow rates: H2 30 mL min−1 ; air 295 mL min−1 ). A 30 m DB-1 bounded phase column (100% dimethyl polysiloxane, internal diameter 0.53 mm, film thickness 1.5 µm) with He as a carrier gas (flow rate 4.9 mL min−1 ) was used. The oven temperature was 40 ◦ C. 2.6. OTHER
ANALYSIS
The pH of the carrier material was measured after mixing 10 g material with 100 mL distilled water during 15 minutes. The moisture content was calculated by the weight difference before and after drying at 105 ◦ C to constant weight. The wet bulk density was determined by weighing a column with known volume before and after filling with compost material. The BET specific surface was determined by nitrogen adsorption at liquid nitrogen temperature, using an AREA-meter 2 − (Ströhlein). Analysis of the NH+ 4 and NO3 contents were performed according to the standard methods (APHA, 1980), while volatile solids (VS) were determined by difference in weight between the dried (105 ◦ C during 24 h) and ashed (450 ◦ C during 3 h) culture samples.
3. Results 3.1. A MMONIA
ADSORPTION AND ABSORPTION CAPACITY OF A COMPOST
BIOFILTER
The bulk density of the compost material used was 578 kg m−3 while the specific surface area of the material was 1400 m2 kg−1 DW. To determine the adsorption
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
181
Figure 1. Influent (4) and effluent (N) NH3 concentration (mg m−3 ) as a function of time during the adsorption experiment. The polluent dosing was stopped at day 7.
capacity of a compost biofilter for NH3 , 7.3 kg of the compost carrier material was oven dried prior to use in order to exclude NH3 -absorption in the water fraction of the compost. Next to this, ammonia loaded nitrogen gas (N2 ) (22 L min−1 ) was sent through the biofilter instead of ammonia loaded air, in order to exclude aerobic microbiological NH3 -degradation. The surface loading rate was 44 m3 m−2 h−1 , corresponding to a superficial gas residence time t of 34 s. In Figure 1, it can be seen that it takes about 4 days before breakthrough of ammonia is observed at an influent concentration of ±113 mg NH3 m−3 . When integrating the area between influent and effluent concentration, an adsorption capacity of 1.41 g NH3 kg−1 dry compost, corresponding to 490 g NH3 m−3 biofilter (40% moisture content), is obtained at the 113 mg NH3 m−3 influent concentration. Six days after the NH3 dosing was stopped (day 7), the desorption was still not quantitative since a 7 mg NH3 m−3 concentration in the effluent of the biofilter was measured. Using Henry’s coefficient of NH3 at 20◦ C (H20 ◦ C = 5.610−4 ) (Perry and Green, 1984), the liquid phase NH3 concentration (NH3 )aq in equilibrium with the 113 mg NH3 m−3 gas phase concentration was calculated to be 202 g m−3 . For a compost biofilter with a 40% moisture content and a wet bulk density of 578 kg m−3 , this corresponds to an NH3 absorption capacity of 47 g m−3 . Depending on the pH (and
182
E. SMET ET AL.
Figure 2. Dissociation of NH3 in the water phase as a function of pH and temperature (Anthonisen et al., 1976).
to a smaller content on the temperature) of the liquid phase in the biofilter, however, part of the dissolved ammonia (NH3 )aq will be protonated into the ammonium ion (NH+ 4 ) (Figure 2), resulting in an increase in NH3 absorption capacity. As a result of this, it can be concluded that the minimum total physico-chemical sorption capacity of a compost biofilter at a 113 mg NH3 m−3 gas phase concentration is 537 g m−3 . 3.2. A MMONIA
REMOVAL IN A COMPOST BIOFILTER INOCULATED WITH A NITRIFYING CULTURE
The biofilter was filled over a height of 0.8 m with fresh compost material (V = 24 L). The carrier material was previously inoculated by mixing it with 1 L nitrifying culture (VS 0.82 g L−1 ). The surface loading rate was 22 m3 m−2 h−1 (t = 131 s). Due to the physico-chemical interaction of NH3 with the carrier material, no start-up period was observed since complete NH3 -removal is obtained from day 0 on (Figure 3). A 100% removal efficiency is obtained at a loading rate (LR) of 55 g NH3 m−3 d−1 (NH3 -concentration ±75 mg m−3 ) (day 0–11) while an average removal efficiency of 94% is obtained at a LR of 140–225 g NH3 m−3 d−1 (NH3 concentration 190–310 mg m−3 ) (day 11–30). Upon increasing the surface loading rate to 72 m3 m−2 h−1 (t = 40 s) on day 30, the NH3 -removal efficiency dropped to an average value of 64%. After re-installing the original surface loading rate on day 38, however, the biofilter recovered its original high removal efficiency for a
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
183
Figure 3. Loading rate (LR) (#) and elimination capacity (EC) (N) (g NH3 m−3 d−1 ) of the non-inoculated biofilter as a function of time (days), together with the pH of the carrier material (). From day 30 to 38 (vertical lines), superficial gas residence time was reduced from 131 to 40 s. TABLE I − − − + pH, NH+ 4 -N, (NO2 en NO3 )-N, molar ratio (NOx /NH4 ) and moisture content (%) of the carrier material during the biofilter experiment
Day
pH −1 NH+ 4 -N (g kg ) − − (NO2 + NO3 )-N (g kg−1 ) + NO− x /NH4 (mol/mol) Moisture content (%)
0
42
57
67
73
7.2 0 1.1 – 38
8.0 2.5 3.2 1.3 42
8.6 3.3 3.5 1.1 37
8.5 3.5 3.6 1.0 35
8.5 3.3 3.3 1.0 39
short time (4 days). The NH3 removal in the biofilter resulted in a strong increase of both the ammonia and nitrite/nitrate content of the compost, while the pH of the compost material slightly increased (Table I). From day 42 on, the EC for NH3 gradually decreased, while the pH of the carrier material increased to a value of 8.5. The temporary slight increase in EC on day 54–56 can be explained by the increased physico-chemical sorption due to the increased influent NH3 concentration. From day 58 on, the pH of the compost remained unaffected (pH 8.5) while the NH3 -removal completely stopped.
184
E. SMET ET AL.
Figure 4. Loading rate (LR) (#) and elimination capacity (EC) (N) (g NH3 m−3 d−1 ) of the non-inoculated biofilter as a function of time (days).
In accordance with this, no increase in ammonia and nitrite/nitrate content of the compost was observed from day 57 on (Table I). The cumulative NH3 -removal over this 73-day experiment was 6550 g NH3 m−3 or 9.3 g NH3 -N kg−1 compost. − − The total amount of NH+ 4 /NO2 /NO3 -N that accumulated in the compost material −1 over this period was 5.5 g kg compost (Table I), corresponding to 59% of the cumulative NH3 -removal. 3.3. A MMONIA
REMOVAL IN A NON - INOCULATED COMPOST BIOFILTER
The biofilter was filled over a height of 0.8 m with fresh compost material (V = 24 L) and the surface loading rate was 22 m3 m−2 h−1 (t = 131 s). No inoculation of the fresh compost material was applied, while the influent NH3 -concentration was installed at a value of ±550 mg m−3 . Up to day 18, high removal efficiencies were obtained (average removal efficiency day 0–18: 87%) with EC up to 350 g NH3 m−3 d−1 (Figure 4). From day 18 on, however, a strong reduction in EC was obtained, while a complete NH3 breakthrough was observed on day 27. The cumulative NH3 -removal over this 27-day experiment was 5560 g NH3 m−3 or 7.9 g NH3 -N kg−1 compost.
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
185
Figure 5. Loading rate (LR) (#) and elimination capacity (EC) (N) (g NH3 m−3 d−1 ) of the non-inoculated biofilter as a function of time (days). On day 4 (vertical line), the biofilter carrier material was mixed with 0.5 L of a 456 g L−1 NH4 NO3 -solution, corresponding to 3 g NH+ 4 -N −1 compost. kg−1 compost and 3 g NO− -N kg 3
3.4. S UPPLEMENTATION
OF AMMONIUM NITRATE TO A COMPOST BIOFILTER
The biofilter was filled over a height of 0.7 m with fresh compost material (V = 21 L) and the surface loading rate was 28 m3 m−2 h−1 (t = 90 s). No inoculation of the fresh compost material was applied. After a 4-day period with 100% NH3 removal efficiency, the compost material was mixed with 0.5 L of a 456 g L−1 NH4 NO3 -solution, corresponding to a nitrogen supplementation of 3 g NH+ 4 -N −1 kg−1 compost and 3 g NO− -N kg compost (Figure 5). The NH -removal in the 3 3 biofilter strongly dropped from day 4 on and the NH3 -removal completely stopped on day 10. The cumulative NH3 -removal over this 10-day experiment was 1175 g + NH3 m−3 or 1.7 g NH3 -N kg−1 compost. The calculated (NO− x + NH4 )-N content −1 of the compost material at the end of this experiment is 8 g N kg (sum of NH3 removal and NH4 NO3 -supplementation). 3.5. E FFECT
OF AMMONIA AND ITS METABOLITES ON THE BY HYPHOMICROBIUM MS3
Me2 S- REMOVAL
The biofilter was filled over a height of 0.65 m with fresh compost material (V = 19 L) and the surface loading rate was 84 m3 m−2 h−1 (t = 27 s). The carrier material was previously inoculated by mixing it with 0.5 L Hyphomicrobium MS3
186
E. SMET ET AL.
Figure 6. Removal of Me2 S in the Hyphomicrobium MS3-inoculated compost biofilter before, during (day 73–79) and after the additional loading with NH3 . Symbols: Me2 S loading rate (LR) (#) and elimination capacity (EC) (N) (g Me2 S m−3 d−1 ); NH3 influent concentration () and NH3 effluent concentration () (mg m−3 ).
culture (VS 2.3 g L−1 ) to promote the Me2 S-removal. At a Me2 S loading rate of 600 g m−3 d−1 , the biofilter obtained an EC of 200–300 g m−3 d−1 after a startup period of 5 days. When NH3 was dosed to the biofilter as a second substrate at a concentration of 14 mg m−3 (LR 45 g m−3 d−1 ) during a 24-hour period, only a minor effect (EC for Me2 S increased from 270 to 295 g m−3 d−1 ) was observed (data not shown). However, when NH3 was supplemented to this biofilter as a second substrate at a concentration of ±100 mg m−3 (LR 320 g m−3 d−1 ) during a 6-day period, complete inhibition in Me2 S-removal (EC dropped from 200 to less than 10 g m−3 d−1 ) was obtained (Figure 6), while the NH3 -absorption resulted in a pH-increase of the carrier material from 4.3 to 7.7. Due to physicochemical desorption from the carrier material, NH3 was detected in the effluent of the biofilter up to 5 days after interruption of the NH3-dosing (Figure 6). Taking into account this desorption, the total NH3 removal efficiency during this biofilter experiment was 74%. From day 84 on, the biofilter Me2 S-degradation restarted and reached an EC of 285 g Me2 S m−3 d−1 . − − Finally, the effect of NH+ 4 , NO2 and NO3 on the activity of Hyphomicrobium MS3 was studied in liquid batch experiments. Complete inhibition in Me2 S-removal by Hyphomicrobium MS3 was obtained at a nitrite concentration of 1.3 g NO− 2 -N − −1 L , while the 50% inhibition concentration was found to be 0.60 g NO2 -N L−1 . For ammonia and nitrate, the 50% inhibition concentration was found to be 7.3 g −1 −1 NH+ and 4.9 g NO− 4 -N L 3 -N L , respectively (Figure 7).
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
187
Figure 7. Relative Me2 S-degrading activity of Hyphomicrobium MS3 as a function of the NH+ 4 -N (g −1 ) ( ) and NO− -N (g L−1 ) (#). L−1 ) (N), NO− -N (g L 2 3
4. Discussion
Due to significant physico-chemical retention phenomena of ammonia on the compost carrier material used (absorption, adsorption, ion exchange reactions), no startup period could be observed during all biofiltration experiments performed with NH3 (Figures 3–5). No effect on the NH3 -removal in the compost biofilter was observed by inoculating it with a nitrifying culture since high elimination capacities were obtained in both the inoculated and the non-inoculated biofilter (Figures 3 and 4). Apparently, the fresh compost material used in this study is already a good inoculum for nitrifying micro-organisms. The EC obtained in the compost biofilters (up to 350 g NH3 m−3 d−1 ) strongly exceeded maximum removal capacities obtained by others (7–87 g NH3 m−3 d−1 ) using wood bark and peat as a carrier material (Terasawa et al., 1986; Van Langenhove et al., 1988; Weckhuysen et al., 1994). The high metabolic activity in compost biofilters can be explained by the superior nutritional level of this material in comparison with the other materials mentioned (Smet et al., 1996). Using different types (source, composting process) of compost as a biofilter carrier, however, strong differences in biofilter
188
E. SMET ET AL.
performance can be expected due to the variability in nutritional level of these materials. Contrary to others (Don, 1985; Hartikainen et al., 1996), no toxicity effect of ammonia was observed on the nitrifying activity of the biofilter, even at NH3 concentrations up to 550 mg m−3 . This is an important observation with regard to the biofiltration of composting waste gases where these high NH3 -concentrations can be found (Haug, 1993). In accordance with Don (1985), analysis of the compost material revealed that only 50% of the NH3 -input in the biofilter was nitrified, while the other 50% remained in the filter as NH+ 4 (Table I). As a result of + this, a molar NO− /NH ratio of ±1 was observed in the compost material. In x 4 our experiments, only 59% of the cumulative NH3 -removal was analysed in the − − carrier material as NH+ 4 , NO2 or NO3 . As a possible reason for this 41% gap, N-immobilisation in the biomass and N2 - and N2 O-emission can be suggested (Morgenroth et al., 1996; Czepiel et al., 1996). Although nitrification was taking place in the compost biofilters, no acidification of the carrier material was observed (day 0–42: pH 7–8) (Figure 3) due to the equilibrium between the microbiological and physico-chemical NH3 removal. Both in the non-inoculated and the inoculated biofilter, however, a sharp reduction in NH3 -removal was observed after a cumulative NH3 -removal of ±6000 g NH3 m−3 . According to Hunik et al. (1992), no NH+ 4 -inhibition is to be expected for Nitrosomonas europaea and Nitrobacter agilis in concentrated NH3 -loaded − waste streams, while also NO− 2 - and NO3 -inhibition is not to be expected at a pHvalue of 8.5. However, osmotic pressure due to high salt concentrations was found to inhibit mainly the activity of Nitrosomonas europaea according to (Hunik et al., 1992): V = 0.994 − 0.00187(Csalt ) Vmax where V Vmax Csalt
= = =
activity of Nitrosomonas europaea maximum activity of Nitrosomonas europaea salt concentration (mmol salt L−1 )
Upon substitution of the NH4 NO3 -concentration analysed in the compost material on day 73 (6.6 g NH4 NO3 -N kg−1 = 16.9 g NH4 NO3 -N L−1 ) (Table I) in this equation, a complete (100%) inhibition in nitrification activity due to osmotic effects is found. This was confirmed by the supplementation of 6 g NH4 NO3 -N kg−1 to the biofilter (Figure 5), resulting in an immediate reduction in NH3 -removal of the biofilter. While no toxicity effect of high NH3 concentrations on the biofiltration of NH3 was observed, a dual effect of NH3 on the Me2 S-removal in a Hyphomicrobium MS3-inoculated biofilter was obtained. While the supplementation of 14 mg m−3
ABATEMENT OF HIGH CONCENTRATED AMMONIA LOADED WASTE GASES
189
NH3 to the gas phase yielded only a minor effect (9% increase in Me2 S-removal), a complete inhibition in Me2 S removal was observed upon the supplementation of NH3 at a 100 mg m−3 gas phase concentration. Apparently, NH3 can be used as a nitrogen source by Hyphomicrobium MS3 at lower concentrations. At higher concentrations, however, the toxic effect of NH3 is dominating. After strippingoff and/or nitrifying part of the NH3 added to the biofilter, however, the biofilter recovered and even surpassed its original EC for Me2 S due to the nutritional effect of the N-supplementation (Morgenroth et al., 1996). Finally, it was illustrated that the Me2 S degradation by Hyphomicrobium MS3 at pH 6.6–6.9 is strongly inhibited by small concentrations of NO− 2 -N. Next to this, + − the NH4 -N and NOx -N concentrations which were found to inhibit the nitrifying biomass due to osmotic effects (8.5 g N L−1 ) were also shown to strongly (>60%) inhibit the Me2 S-degradation by Hyphomicrobium MS3 at pH 6.6–6.9. As a conclusion, it can be stated that high NH3 -removal efficiencies can be obtained in compost biofilters, even at input NH3 -concentrations up to 550 mg m−3 . Due to the accumulation of toxic NH4 NO3 -concentrations, however, regular renewal of the compost material in the biofilter has to be applied. With regard to the removal of Me2 S by Hyphomicrobium MS3, a dual effect of NH3 supplementation was observed with a slight stimulation of the Me2 S-removal due to nutritional effects at low (14 mg NH3 m−3 ) concentrations, but complete inhibition due to toxicity at higher (100 mg NH3 m−3 ) concentrations. Moreover, in long-term biofilter experiments with NH3 and Me2 S, both the Me2 S-removal by Hyphomicrobium MS3 and the nitrification will ultimately be affected by the accumulation of NH4 NO3 , irrespective of the NH3 -concentration in the waste gas.
Acknowledgements This work was financed by a scholarship of the Flemish Institute for support of Scientific/Technological Research in the Industry (IWT) (OZM/960008).
References Anthonisen, A. C., Loehr, R. C., Prakasam, T. B. S. and Srinath, E. G.: 1976, ‘Inhibition of nitrification by ammonia and nitrous acid’, Journal WPCF 48, 835–852. APHA: 1980, Standard methods for the examination of water and waste water, 15th edition American public health association, Washington, 1134 pp. Buijsman, E., Maas, H. F. M. and Asman, W. A. H.: 1987, ‘Anthropogenic NH3 emissions in Europe’, Atmospheric Environment 21, 1009–1022. Cho, K.-S., Hirai, M. and Shoda, M.: 1992, ‘Enhanced removal efficiency of malodorous gases in a pilot-scale biofilter inoculated with Thiobacillus thioparus DW44’, J. Ferment. Bioeng. 73, 46–50. Chung, Y. C., Huang, C. P. and Tseng, C. P.: 1997, ‘Biotreatment of ammonia from air by an immobilized Arthrobacter oxydans CH8 biofilter’, Biotechnology Progress 13, 794–798.
190
E. SMET ET AL.
Czepiel, P., Douglas, E., Hariss, R. and Crill, P.: 1996, ‘Measurements of N2 O from composted organic wastes’, Environ. Sci. Technol. 30, 2519–2525. De Zwart, J. M. M. and Kuenen, J. G.: 1992, ‘C1 -cycle of sulfur compounds’, Biodegradation 3, 37–59. Don, J. A.: 1985, ‘The rapid development of biofiltration for the purification of diversified waste gas streams’, VDI Berichte 561, 63–73. Gellens, V., Boelens, J. and Verstraete, W.: 1995, ‘Source separation, selective collection and in reactor digestion of biowaste’, Antonie van Leeuwenhoek 67, 79–89. Hartikainen, T., Ruuskanen, J., Vanhatalo, M. and Martikainen, P. J.: 1996, ‘Removal of ammonia from air by a peat biofilter’, Environmental Technology 17, 45–53. Haug, R. T.: 1993, The practical handbook of compost engineering, Lewis Publishers, Boca Raton, 717 pp. Heller, W. and Schwager, H.: 1996, ‘Luftreinigung durch Kompost-Biofiltration’, Agrarforschung 3, 520–522. Hunik, J. H., Meijer, H. J. G. and Tramper, J.: 1992, ‘Kinetics of Nitrosomonas europaea at extreme substrate, product and salt concentrations’, Appl. Microbiol. Biotechnol. 37, 802–807. Lipski, A., Ahrens, A., Reichert, K. and Altendorf, K.: 1994, ‘Identifizierung von Alcaligenes faecalis als heterotrophem Nitrifizierer und aerobem Denitrifizierer in Biofiltern’, VDI Berichte 1104, 251–259. Morgenroth, E., Schroeder, E., Chang, D. and Scow, K.: 1996, ‘Nutrient limitation in a compost biofilter degrading hexane’, J. Air and Waste Manage. Assoc. 46, 300–308. Perry, R. H. and Green, D.: 1984, Perry’s Chemical Engineers’ Handbook, Sixth edition. McGrawHill international editions, Chemical Engineering Series, New York. Pöhle, H. and Kliche, R.: 1996, ‘Geruchsstoffemissionen bei der Kompostierung von Bioabfall’, Zbl. Hyg. 199, 38–50. Przyjazny, A., Janicki, W., Chrzanowski, W. and Staszewski, R.: 1983, ‘Headspace gas chromatographic determination of distribution coefficients of selected organosulphur compounds and their dependence on some parameters’, J. Chromatogr. 280, 249–260. Rutherford, D. W. and Chiou, C. T.: 1992, ‘Effect of water saturation in soil organic matter on the partition of organic compounds’, Environ. Sci. Technol. 26, 965–970. Shoda, M.: 1991, ‘Methods for the Biological Treatment of Exhaust Gases’, in A. M. Martin (ed.), Biological Degradation of Wastes, Elsevier Science Publishers, New York, pp. 31–46. Smet, E., Van Langenhove, H. and De Bo, I.: 1999, ‘The emission of volatile compounds during the aerobic and the combined anaerobic/aerobic composting of biowaste’, Atmospheric Environment 33, 1295–1303. Smet, E., Van Langenhove, H. and Verstraete, W.: 1996, ‘The effect of inoculation and the type of carrier material used on the biofiltration of methyl sulphides’, Appl. Microbiol. Biotechnol. 45, 293–298. Terasawa, M., Hirai, M. and Kubota, H.: 1986, ‘Soil deodorization systems’, Biocycle 27, 28–32. Togashi, I., Suzuki, M., Hirai, M., Shoda, M. and Kubota, H.: 1986, ‘Removal of NH3 by a peat biofilter without and with nitrifier’, J. Ferment. Technol. 64, 425–432. Turk, A., Sakalis, E., Lessuck, J., Karamitsos, H. and Rago, O.: 1989, ‘Ammonia injection enhances capacity of activated carbon for hydrogen sulfide and methyl mercaptan’, Environ. Sci. Technol. 23, 1242–1245. Van Langenhove, H., Lootens, A. and Schamp, N.: 1988, ‘Elimination of ammonia from pigsty ventilation air by wood bark biofiltration’, Med. Fac. Landbouww. Rijksuniv. Gent 53, 1963– 1969. Weckhuysen, B., Vriens, L. and Verachtert, H.: 1994, ‘Biotreatment of ammonia- and butanalcontaining waste gases’, Appl. Microbiol. Biotechnol. 42, 147–152. Williams, T. O.: 1995, ‘Odors and VOC emissions control methods’, Biocycle 36, 49–56.
