Eb

Ma

b

a

ARRA

KMABNDI

1

ebis[e

trpSiw

d

1d

Chemical Engineering Journal 171 (2011) 1114– 1123

Contents lists available at ScienceDirect

Chemical Engineering Journal

jo u r n al hom epage: www.elsev ier .com/ locate /ce j

ffect of ammonium concentration on microbial population and performance of aiofilter treating air polluted with methane

arc Veillettea,1, Pascal Viensb,1, Antonio Avalos Ramireza, Ryszard Brzezinskib, Michèle Heitza,∗

Department of Chemical Engineering and Biotechnological Engineering, Faculty of Engineering, Université de Sherbrooke, 2500 Boulevard de l’Université, J1K 2R1 Québec, CanadaDepartment of Biology, Faculty of Sciences, Université de Sherbrooke, 2500 Boulevard de l’Université, J1K 2R1 Québec, Canada

r t i c l e i n f o

rticle history:eceived 13 January 2011eceived in revised form 22 April 2011ccepted 4 May 2011

eywords:ethane (CH4)

mmonium (NH4+)

iofiltrationitrificationenitrification

a b s t r a c t

The effect of ammonium concentration on the microbial population and performance of a biofilteroperated at methane concentrations of 0.3% (v/v) was studied. In a range of ammonium concentrationfrom 0.05 to 0.5 gN-NH4

+/L, the removal efficiency, carbon dioxide production rate, biomass produc-tion, ammonium conversion and nitrate production rate were determined. The middle biofilter sectionmicrobial population was analyzed using denaturing gradient gel electrophoresis (DGGE). DNA sequenceanalysis revealed that the bacteria colonizing the middle biofilter section belonged essentially to 4physiological groups: methanotrophs, methylotrophs, nitrifying and denitrifying bacteria. Ammoniumconcentration affected the presence of bacteria from each group and the interactions among them definedthe biofilter performance. In the range of ammonium concentrations tested, the methane removal effi-ciency decreased from 70 to 13%, the carbon dioxide production rate from 25 to 7.5 g/(m3 h), and the

nhibition mean of dry biomass content in the packed bed from 12.8 to 6.7 g biomass/kg filter bed. The nitrate pro-duction rate increased with ammonium concentration; however, it presented negative values at nitrateconcentrations around 0.5 gN/L. This suggests that denitrification occurred due to favourable conditionsfor the growth of denitrifiers. An analysis of methane biodegradation at different layers of the biofiltershows that the methanotrophic and nitrifying activities were more important in the upper section of thebiofilter.

. Introduction

According to the Kyoto protocol, several industrialised countriesngaged to reduce 1990 levels of greenhouse gas (GHG) emissionsy 6% between 2008 and 2012 [1]. In Canada, agriculture is an

mportant GHG producer and accounts for 9.6% of total GHG emis-ions. These emissions increased by 21% between 1990 and 20062]. In the province of Quebec, agriculture emitted 7.5% of all GHGmissions [3].

In Canada, pork industry is a very important sector of agricul-ure with annual incomes of more than 3.1 billion dollars, whichepresents 7% of all agricultural sectors revenues [4]. Quebec’s porkroduction provides 1/3 of all pork produced in Canada in 2010 [5].ince the middle of 1980s, the Quebec pork industry has increasedts production by 67% and as a consequence, its GHG emissions as

ell [6].Piggeries have two main sources of methane (CH4): animal

igestion and manure storage [7]. These two sources also produce

∗ Corresponding author. Tel.: +1 819 821 8000x62827; fax: +1 819 821 7955.E-mail address: Michele.Heitz@USherbrooke.ca (M. Heitz).

1 These two authors contributed equally to this work.

385-8947/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.cej.2011.05.008

© 2011 Elsevier B.V. All rights reserved.

carbon dioxide (CO2) – which is another GHG – hydrogen sul-phide and ammonia (NH3) [8,9]. The CH4 emitted from manurestorage is produced by anaerobic methanogenic bacteria [10]. In2006, GHG emissions of Canadian agriculture attributable to animaldigestion and manure management represented 58% of 69 MtonEq. CO2 emitted and 33% of those GHG were CH4 [11]. Methaneis a GHG which is 25 times more harmful than CO2 in regards toglobal warming [12]. Many factors influence CH4 emissions frommanure storage of piggeries: temperature [13], pH [14], oxygen[15], substrate availability (Biochemical Oxygen Demand) [16] andinhibitory substances such as ammonium (NH4

+) [17].In Quebec, manure storage is incontournable because laws

permit only land spreading on specific periods of the year [18].Methane concentrations emitted from manure storage are between0.1 and 20 g/m3 (0.015 to 3% v/v), depending on the storageconditions [19]. This range is too low for CH4 elimination by ther-mal oxidation, because this technology requires concentrationsbetween 5 and 15% (v/v) [20]. As a result, in the last years, the pro-cess of air biofiltration emerged as an alternative solution to treat

such low CH4 concentrations. This is technically feasible becausesome species of microorganisms can, in biofilters, oxidize CH4 andtransform it into heat, CO2, water, and biomass [21]. The CH4utilizing bacteria, known as methanotrophs, possess the enzyme

eering Journal 171 (2011) 1114– 1123 1115

mCo

mhmpoqtog

caoibiIbtbri[oimm[cchb[

ocNsdaoNfe(iNtbisefat

igocbt

M. Veillette et al. / Chemical Engin

ethane monooxygenase (MMO) that catalyzes the oxidation ofH4 to methanol (CH3OH). This pathway is a defining characteristicf methanotrophs [22].

Biofilters are complex systems whose efficiency depends onany factors, such as: temperature, pH, filter medium, filter bed

umidity, supply of nutrients and many others [23]. Macro andicronutrients supply is essential for microorganisms. Nitrogen,

hosphorus, sulphur and copper must be present in the filter mediar added by means of a nutrient solution. Among them, nitrogen isuantitatively the most important as it represents 10–14% (w/w) ofhe dry cell weight [24]. Several studies have addressed the effectf nitrogen on biofiltration efficiency with different waste streamases [25–28].

In most of these studies, increasing nitrogen (until an optimumoncentration) had a positive effect on biofilter performance. Asn example, Nikiema et al. [27] examined the effect of nitrogenn the efficiency of CH4 biofiltration. They used a “nitrate min-mal salt” (NMS) nutrient solution with nitrogen concentrationsetween 0.14 and 0.75 gN/L and the CH4 elimination capacity (EC)

ncreased from 5.4 to 29.2 g/(m3 h) for an inlet load of 71.2 g/(m3 h).n CH4 biofiltration, nitrate (NO3

−) is usually used instead of NH4+

ecause NH4+ is known to have adverse effects on CH4 oxida-

ion. Adding, for instance, NH4+ to a soil hosting methanotrophic

acteria can significantly reduce the oxidation of CH4 [29]. Thiseduction in performance could be directly related to the biochem-cal mechanism of competitive inhibition of the MMO by NH4

+

30–32]. It must, however, be underlined that the first productf oxidation of NH4

+ by methanotrophs is hydroxylamine, whichs further metabolized to nitrite (NO2

−). Thus, another proposedechanism suggests that hydroxylamine or/and NO2

− can inhibitethanotrophs because of their toxicity [32,33]. King and Schnell

32] studied the oxidation of CH4 inhibited by NH4+ in bacterial

ulture. Working at a CH4 concentration of 5% (v/v), they con-luded that CH4 oxidation is inhibited at NH4

+ concentrationsigher than 1–10 mM. Generally, most authors found that the inhi-ition of CH4 oxidation increased at high concentrations of NH4

+

34,35].On the other hand, some authors observed stimulation of CH4

xidation after addition of NH4+. One example comes from micro-

osm studies of rice rhizosphere, where it was observed thatH4

+ fertilisation enhanced CH4 oxidation [36,37]. Methane emis-ion from these microcosms was reduced by up to 26% followingiammonium hydrogen phosphate ((NH4)2HPO4) application. Theuthors suggest that the stimulating effect may be caused by a relieff possible N-limitation, or a direct stimulation of CH4 oxidation byH4

+ by an unknown mechanism. Bender and Conrad [38] per-ormed a study on 3 different soils and observed, in short-termxperiments (6 h), a stimulating effect of low NH4

+ concentrations5, 5 and 22 mM in soil water, respectively) on CH4-oxidizing activ-ty. They also suggested that this effect was due to the relief of an-limitation. In a study of landfill cover soils [39], an increase in

he CH4 oxidation rate after addition of NH4+ for short-term incu-

ation (less than 34 days) followed by a decrease for a long-termncubation (more than 34 days) was observed. All these studieshow that the influence of NH4

+ on the CH4 oxidation in complexnvironments such as soil is difficult to predict. Experiments per-ormed in a more controlled system such as a biofilter packed withn inorganic filter bed may be useful for a better understanding ofhe interplay between methanotrophy and NH4

+ supply.The main goal of the study presented here was to evaluate the

nfluence of NH4+ concentration on the performance of an inor-

anic packed bed biofilter treating CH4 contaminated air. The effect

f NH4

+ concentration on CH4 oxidation, CO2 production, nitrogenonversion and biomass content was determined. The evolution ofacterial community through the time was followed using dena-uring gradient gel electrophoresis (DGGE) and DNA sequencing in

Fig. 1. Biofilter set up for the biofiltration of CH4.

order to analyse the effect of NH4+ concentration on the biofilter

microbiota.

2. Materials and methods

2.1. Biofilter

The experimental set up of the biofilter used is presented inFig. 1. The upflow biofilter was built from of a 15 cm diameter Plex-iglas cylinder with a total height of 135 cm. The filtering based-bedwas split into 3 equal sections for a total bed height of 100 cm.The inorganic particles of the filter bed had an average diame-ter of 7 mm, a specific surface of 470 m2/m3 and a void space of45%. For confidential reasons, the nature of the filtering bed cannotbe revealed. Sampling ports were installed between the sections,allowing gas concentration measurements. At the bottom of thebiofilter, both CH4 (Praxair Canada, Inc.) and humidified air contain-ing oxygen (O2) were fed. Air was supplied by mean of a compressedair system. The air was saturated with water by passing air througha humidification column in order to avoid filter bed desiccation.The treated gas was evacuated at the top of the biofilter.

2.2. Operating conditions

In the 18 L biofilter, the air flow rate was set at 3 L/min foran empty bed residence time of 6 min. Nutrients were suppliedby means of a nitrate salts medium (NSM) nutrient solution [40]at a fixed nitrogen concentration of 0.5 gN/L. The CH4 inlet con-centration was set to 3000 ppmv (0.3% v/v) for a CH4 inlet loadof 20 g/(m3 h) using a mass flow meter (Brooks, USA). The con-centration of NH4

+ was gradually increased from 0 to 0.5 gN/L.Simultaneously, when NH4

+ concentration (as ammonium carbon-ate) was increased, NO3

− concentration (as sodium nitrate) wasdecreased to keep the total nitrogen concentration in the nutri-ent solution at 0.5 gN/L. The CH4 concentration in the gas phasewas measured at each sampling port using a hydrocarbon anal-yser equipped with a continuous flame ionisation detector (Horiba

model FIA-510, USA). The CO2 concentration was measured witha gas analyser detector (Ultramat 22P, Siemens, Germany) using anon-dispersive infra-red principle.

1 neerin

ti

I

R

E

wii((u

P

Y

woa2

s(pacda

Q

waN((o

paN(

mda(r

R

R

w(

116 M. Veillette et al. / Chemical Engi

Biofilter performance was determined with 2 main parameters:he CH4 removal efficiency, RE (%); and the CH4 elimination capac-ty, EC (g/(m3 h)); as follows:

L = Q × Cg0

V(1)

E = Cg0 − Cg

Cg0(2)

C = IL × RE = Q × Cg0 × REV

(3)

here IL is the inlet load (g/(m3 h)), Q is the air flow rate (m3/h), Vs the filter bed volume (m3, including void space), Cg0 is the CH4nlet concentration (g/m3) and Cg is the CH4 outlet concentrationg/m3). Other important parameters are CO2 production rate (PCO2 )g/(m3 h)) and CO2 yield coefficient (YCO2 ). They were calculatedsing the following formulas:

CO2 = Q × (Cd − Cd0)V

(4)

CO2 = PCO2

EC(5)

here Cd0 is the CO2 inlet concentration (g/m3) and Cd is the CO2utlet concentration (g/m3). Methane and CO2 concentrations weren average of values measured at least 3 times a week for at least

weeks.When the CH4-EC peudo steady state regime was reached, liquid

amples of the nutrient solution and of the leachate were collectedat the same time that the automatic watering). Biomass was alsoresent in the leachate throughout of the experiments. Nitrogenccumulated in the biofilm (biomass or denitrified) (Nbed) was cal-ulated from the mass balances of NH4

+, NO3− and NO2

− as theifference between their concentrations in the nutrient solutiondded to the biofilter and their outlet concentration in the leachate.

NS × [NH4i+ + NO3i

− + NO2i−] = QNS × [NH4o

+ + NO3o−

+ NO2o−] + Nbed (6)

here NH4i+, NO3i

−, NO2i− and NH4o

+, NO3o−, NO2o

− are the inletnd outlet concentrations (g/L) of nitrogen species respectively,bed is in gN/h and QNS is the nutrient solution flow of 0.0625 L/h

spread once a day). As the gaseous form of nitrogen (molecularN2), nitrous oxide) cannot be measured with the equipments atur disposal, Nbed includes the gaseous form of nitrogen.

Leachate samples were filtered with filter paper to remove sus-ended biomass. Filtered samples were acidified with sulphuriccid (pH < 2) and stored at around 4 ◦C [41]. Concentrations of NH4

+,O3

− and NO2− were measured by using ion chromatography

Dionex ICS-1000, Canada).The dry biomass concentration in the packed bed was deter-

ined (at pseudo steady state regime) by means of the weightifference of a packing material sample dried at 105 ◦C overnightnd calcined at 500 ◦C for 1 h. The specific nitrate production rateRNO3 ) and the nitrogen bed retention rate (Rbed) were calculatedespectively using:

NO3 =(NO3o

− − NO−3i

) × QNS

V × DB(7)

Nbed

bed =

V × DB(8)

here DB is the dry biomass content per volume unitkg biomass/m3), RNO3 and Rbed are in gN/(kg biomass h).

g Journal 171 (2011) 1114– 1123

2.3. Biofilm sampling and DNA isolation

For practical issues, microbial analysis was performed only inthe middle biofilter section. Samples of 15 g of packed bed from themiddle biofilter section were taken when the biofilter performancewas stable (pseudo steady state) for each NH4

+ concentrationtested. The 15 g samples were vortexed for 30 s, then shaken for30 min with 20 mL aliquots of an extraction buffer (0.1 wt.% sodiumpyrophosphate, pH 6.5; 2 wt.% sodium chloride (NaCl)). The result-ing suspension was centrifuged, initially at low speed (30 × g) for2 min to sediment the coarse debris and then at higher speed(4716 × g) for 15 min to pellet bacteria. Nucleic acids were extractedfrom pellets using the FastDNA® Spin kit for soil (MP Biomedicals,USA). DNA was quantified using the NanoDrop ND-1000 spec-trophotometer and stored at −20 ◦C.

2.4. Denaturing gradient gel electrophoresis (DGGE)

The DNA extracted from biofilm samples was used astemplate for touchdown PCR to amplify the V3, V4 and V5region of bacterial 16S rRNA gene with primers 341F-GC 5′-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGGCCTAC-GGGAGGCAGCAG-3′ and 907R 5′-CCGTCAATTCMTTTGAGTTT-3′

[42]. PCRs were performed with a thermocycler Biometra TPersonal (Biometra, Canada) using 50 �L (total volume) of amixture containing 0.25 �L of AmpliTaq Gold DNA polymerase(Applied Biosystems, USA), 5 �L of 10× PCR Gold Buffer, 6 �LMgCl2 (25 mM), 4 �L deoxyribonucleotides phosphates mixture(2.5 mM each), 2.5 �L primer 341F-GC (5 �M), 2.5 �L primer 907R(5 �M), and 1 �L DNA extract (20–50 ng/�L). Temperature cyclingwas done as described previously [42] with minor modifications.The denaturing steps were performed at 95 ◦C, the AmplitaqGoldwas added before starting the run and there was no step at 80 ◦C.

The PCR products were analyzed by DGGE with a 6% poly-acrylamide gel, containing a 35–65% gradient of denaturant, at aconstant voltage of 100 V for 18 h at 60 ◦C, and using a DCode appa-ratus (Bio-Rad Laboratories, USA). Following electrophoresis, thegels were stained for 40 min with SYBR Gold (Molecular Probes,USA) and visualized by UV transillumination. Selected bands wereexcised and the DNA was eluted as described previously [42]. ElutedDNAs were preamplified by PCR with the primer pair 341F (5′-CCTACGGGAGGCAGCAG-3′) and 907R. Sequencing of ampliconswas performed at Génome Québec Innovation Centre (McGill Uni-versity, Canada).

2.5. Metagenomic libraries construction and analysis

Three rRNA 16S gene libraries were constructed for thepseudo steady state biofiltration conditions reached at 0.05,0.3 and 0.5 gN-NH4

+/L. Gene segments were amplified bytouchdown PCR with AmpliTaqGold, using the primers BSF8/20 5′-AGAGTTTGATCATGGCTCAG-3′ and BSR 1541/20 5′-AAGGAGGTGATCCAGCCGCA-3′ [27] in the following conditions:an initial denaturing step at 95 ◦C for 5 min; 95 ◦C for 1 min;64–54 ◦C for 1 min at the ramp of 1 ◦C per 2 cycles (touchdown);72 ◦C for 3 min; then 15 cycles of 95 ◦C for 1 min, 54 ◦C for 1 min,72 ◦C for 3 min, and a final primer extension at 72 ◦C for 10 min.Amplicons were ligated into a linear form of pCR 2.1 vector usingthe TA-cloning procedure (Invitrogen Life Technologies, USA) andtransformed into Escherichia coli. After elimination of false posi-tive clones, 3 libraries were examined including, 96, 78 and 100

clones for, respectively, pseudo steady state conditions at 0.05, 0.3and 0.5 gN-NH4

+/L. The partial rRNA 16S gene sequences were thencompared with a database (GenBank database) to determine thetaxonomic position of the microorganisms from which they origi-

M. Veillette et al. / Chemical Engineering Journal 171 (2011) 1114– 1123 1117

-10

0

10

20

30

40

50

60

70

80

90

100

0.60.50.40.30.20.10.0

CH

4re

mov

al e

ffici

ency

(%)

NH 4+ inlet conce ntration in the nu trient solution (gN-NH 4

+/L)

Fig. 2. Removal efficiency of CH4 for the for the lower section (♦), middle section(i

nt

3

3

bNdn70

ucttuid

osttcR0b

sivf

3

osn

0

5

10

15

20

25

30

0.60.50.40.30.20.10

CO

2pr

oduc

tion

rate

(g/(m

3h)

)

NH 4+ inlet conce ntration in the nutrient solution (gN-NH 4

+/L)

y = 0.78x + 1.23

y = 1.69x + 0.97

y = 1.82x + 0.75

0

3

5

8

10

13

6543210

CO

2pr

oduc

tion

rate

(g/

(m3

h))

Elimination capacity (g/(m3 h))

b

a

Fig. 3. CO2 production rate (a) as a function of NH4+ concentration in nutrient solu-

tion (overall (©)), and (b) as a function of elimination capacity for the lower section(♦), middle section (�) and the upper section (�).

0

5

10

15

20

25

30

35

40

45

50

0.60.50.40.30.20.10

NH

4+co

nver

sion

(%)

NH4+ inle t conce ntration in the nu trient solution (gN-NH 4

+/L)

was observed in the range from 0.05 to 0.5 gN-NH /L with a

�), the upper section (�) and the overall (©) as a function of NH4+ concentration

n nutrient solution.

ated, using the BLAST-N algorithm at default settings, available athe National Center for Biotechnology Information server [43].

. Results

.1. Methane removal

Fig. 2 shows the CH4 removal efficiency (CH4-RE) for eachiofilter section and the overall CH4-RE as a function of theH4

+ concentration in the nutrient solution. The overall CH4-REecreased linearly with NH4

+ concentration presenting a determi-ation coefficient (R2) of 0.91. The overall CH4-RE dropped from0 to 13% in the range of NH4

+ concentration tested from 0.1 to.5 gN-NH4

+/L.The 3 sections followed a similar behaviour in terms of individ-

al RE. For example, for the upper section, in the range of NH4+

oncentration from 0.05 to 0.25 gN-NH4+/L, the CH4-RE presented

he highest value and it varied from 45 to 28%. For NH4+ concen-

rations ranging from 0.25 to 0.35 gN-NH4+/L, the CH4-RE in the

pper section dropped from 28 to 8%. For NH4+ concentration rang-

ng from 0.35 to 0.5 gN-NH4+/L, the CH4-RE in the upper section

ecreased from 8 to 0.6%.Fig. 3a shows the accumulated PCO2 at 100 cm (H) as a function

f the NH4+ concentration in the nutrient solution. The PCO2 at H

eems to have a linear tendency with the NH4+ concentration in

he nutrient solution, with a R2 of 0.83. Similar to CH4-RE, PCO2 forhe overall biofilter decreased from 23 to 13 g/(m3 h) when NH4

+

oncentration increased from 0.1 to 0.2 gN-NH4+/L. The results of

E and PCO2 suggest that a NH4+ concentration between 0.1 and

.2 gN-NH4+/L, microbial populations evolved and this is discussed

elow.Fig. 3b shows the PCO2 in the upper, the middle and the lower

ections of the biofilter as a function of the CH4-EC. The PCO2ncreased linearly with CH4-EC for all sections with R2 coefficientsarying respectively from 0.52 to 0.98 and YCO2 (slopes) varyingrom 0.78 to 1.82.

.2. Ammonium conversion

Fig. 4 shows the NH + conversion in the biofilter as a function

4f NH4

+ concentration in the nutrient solution. The NH4+ conver-

ion decreased from 37 to 15% when the NH4+ concentration in the

utrient solution increased from 0.05 to 0.5 gN-NH4+/L.

Fig. 4. Ammonium conversion as a function of NH4 concentration in nutrient solu-tion.

Fig. 5 shows the RNO3 in the biofilter as a function of theNH4

+ concentration in the nutrient solution. A linear tendency+

4R2 of 0.70. From 0.05 to 0.2 gN-NH4

+/L, the NO3− produced was

lower than that consumed by the bacteria; for example, theRNO3 varied from −0.03 to −0.01 gN/(kg biomass h). From 0.2 to

1118 M. Veillette et al. / Chemical Engineering Journal 171 (2011) 1114– 1123

-0.05

-0.03

0.00

0.03

0.05

0.08

0.10

0.60.50.40.30.20.10

NO

3- p

rodu

ctio

n ra

te (g

-N/(k

g bi

omas

s h)

)

NH4+ inlet conce ntration in the nu trient solution (gN-NH4

+/L)

F +

t

00tv

ct0f

3

tatfNsbt

Fs

4

6

8

10

12

14

0.60.50.40.30.20.10

Dry b

iom

ass (

g/kg

filte

r bed

)

NH4+ inlet concentration in the nutrient solution (gN-NH4

+/L)

pattern at 0.3 gN-NH4+/L presented several changes. Decreases of

PCO2, CH4-RE and dry biomass were also observed at this concen-tration (Figs. 2, 3a and 7).

ig. 5. Nitrate production rate as a function of NH4 concentration in nutrient solu-ion.

.5 gN-NH4+/L, the RNO3 became positive increasing from 0.05 to

.07 gN/(kg biomass h). For NH4+ concentrations ranging from 0.25

o 0.5 gN-NH4+/L, the RNO3 remained constant with an average

alue of 0.06 ± 0.01 gN/(kg biomass h).Fig. 6 shows the Rbed in the biofilter as a function of the NH4

+ con-entration in the nutrient solution. From 0.05 to 0.25 gN-NH4

+/L,he Rbed varied from 0.024 to 0.010 gN/(kg biomass h). From 0.3 to.5 gN-NH4

+/L, some values of the Rbed were negative and variedrom −0.013 to 0.010 gN/(kg biomass h).

.3. Dry biomass evolution

Fig. 7 shows the dry biomass content in the filter bed forhe lower, the middle and the upper sections of the biofilters a function of the NH4

+ concentration in the nutrient solu-ion. Fig. 7 shows that, the mean biomass content decreasedrom 12.8 to 6.7 g biomass/kg filter bed between 0.05 and 0.5 gN-H4

+/L. The dry biomass content in the upper and middle sectionseems to be less affected by the NH4

+ increase. For example, theiomass even increased between 0.25 and 0.5 gN-NH4

+/L from 8.0

o 9.1 g biomass/kg filter bed in the upper section.

-0.015

-0.010

-0.005

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.60.50.40.30.20.10

Nitr

ogen

bed

rete

ntio

n ra

te (g

N/(k

g bi

omas

s h)

)

NH 4+ inle t con centration in the nutrie nt solution (gN-NH 4

+/L)

ig. 6. Nitrogen bed retention rate as a function of NH4+ concentration in nutrient

olution.

Fig. 7. Biomass concentration in the packed bed for the lower section (♦), middlesection (�), the upper section (�) and the mean of biomass ( ) as a function of NH4

+

concentration in nutrient solution.

3.4. Denaturing gradient gel electrophoresis

DGGE fingerprinting was performed to screen the bacterial com-munity of the middle section of the biofilter. Biomass sampleswere obtained after the biofilter reached pseudo steady state atdifferent NH4

+ concentrations between 0.05 and 0.5 gN-NH4+/L.

Representative DGGE profiles of all samples are shown in Fig. 8.Throughout the experiments, the fingerprint patterns were evolv-ing following the increase of NH4

+ concentrations, implying variouschanges in the microbial community. While very similar patternswere observed for samples taken at 0.05 and 0.1 gN-NH4

+/L, the

Fig. 8. DGGE gel of the PCR products representative of the bacterial commu-nity at different NH4

+ concentrations. (1) 0.05 gN-NH4+/L, (2) 0.1 gN-NH4

+/L,(3) 0.25 gN-NH4

+/L, (4) 0.3 gN-NH4+/L, (5) 0.35 gN-NH4

+/L, (6) 0.4 gN-NH4+/L, (7)

0.45 gN-NH4+/L, and (8) 0.5 gN-NH4

+/L.

M. Veillette et al. / Chemical Engineering Journal 171 (2011) 1114– 1123 1119

Table 1Similarity percentages of clones from three different NH4

+ concentrations as compared with the GenBank.

NH4+ concentration (gN/L) Number of clone (s) Clustersa Most probable identification (accession number) Physiological group Similarity (%)

0.05 26/96 CN005-1 Methylophilus methylotrophus (AB193724.1) Methylotroph 99.121/96 CN005-2 Methylocystis parvus (AF150805.1) Methanotroph 99.78/96 CN005-3 Paenibacillus sepulcri CCM 7311 (DQ291142.1) Denitrifying bacteria 93.9–98.27/96 CN005-4 Unidentified (FJ710755.1) 95.85/96 CN005-5 Sphingomonas asaccharolytica strain Y-345

(NR 029327.1)Denitrifying bacteria 98.2

Sphingomonas sp. C16y (GQ253122.1) 96Sphingomonas sp. LnR5-44 (EU332828.1) 99.6

1/96 CN005-11 Nitrobacter hamburgensis X14 (CP000319.1) Nitrifying bacteria 95.30.3 32/78 CN030-1 Methylocystis parvus (AF150805.1) Methanotroph 97.4

31/78 CN030-2 Methylocaldum sp. 05J-I-7 (EU275146.1) Methanotroph 93.82/78 CN030-3 Bacillus firmus (AY833571.2) Denitrifying bacteria 96.6–97.82/78 CN030-4 Pseudomonas peli (AM114534.1) Nitrifying bacteria 98.2

0.5 35/100 CN050-1 Uncultured Xanthomonadaceae bacterium cloneDMS04 (FJ536874.1)

Methylotroph 99.3

18/100 CN050-2 Methylocapsa acidiphila strain B2 (NR 028923.1) 97.8Methylocella silvestris BL2 (CP001280.1) 96.7Beijerinckia indica subsp. indica ATCC 9039(CP001016.1)

96.7

15/100 CN050-3 Rhodanobacter terrae strain SPg (FJ405366.1) 99.410/100 CN050-4 Methylocystis parvus (AF150805.1) Methanotroph 99.53/100 CN050-5 Nitrospira sp. clone g6 (AJ224039.1) Nitrifying bacteria 99.5

aceae

er con

st0tmacorrsbea[

3

otetcmcTmaCtAw

oaunb

3/100 CN050-6 Alcaligen

a Clusters group clones of bacteria with the same identification for a given biofilt

Furthermore, no PCR amplification was possible from biomassamples taken at 0.2 gN-NH4

+/L. Taken together, these observa-ions could indicate that the increase of NH4

+ concentration from.1 to 0.3 gN-NH4

+/L resulted in a serious rebuilding of the bac-erial community. These concentrations exceeded the tolerance of

any primary colonizers, resulting in massive cell lysis and deathnd release of PCR inhibitors in the biofilm. Thereafter, less drastichanges in the patterns were observed following further increasesf NH4

+ concentration. The identification of microorganisms rep-esented by the DGGE bands by DNA amplification and sequencingesulted in one readable sequence, as shown in Fig. 8. The solvedequences corresponded to members of the genus Methylocystis,est identified as Methylocystis parvus (96% similarity). We hypoth-size the failure to solve the other bands might be due to the fact

single DGGE band might have a blend of diverse DNA fragments44].

.5. 16S rRNA gene fragment libraries

Three gene fragment libraries of 96, 78 and 100 clones werebtained from biofilm samples from the middle section of the biofil-er for 0.05, 0.3 and 0.5 gN-NH4

+/L, respectively. The libraries werexamined by DNA sequencing and BlastN analysis. The best iden-ification of clusters of clones is given in Table 1. Clusters groupinglones with 90% or more of identity were assigned with an alphanu-eric designation (e.g. CN005-1). The 96 clones from the library

orresponding to 0.05 gN-NH4+/L were grouped into 12 clusters.

he largest cluster (CN005-1) is closely related to Methylophilusethylotrophus. It was followed by the cluster CN005-2, identified

s M. parvus and cluster CN005-3 assigned to Paenibacillus sepulcri.N005-4 could not be identified but it is closely related to an uncul-ured bacterium derived from an anaerobic NH4

+ oxidation reactor. NO2

− oxidizing population closed to Nitrobacter hamburgensisas also detected (cluster CN005-11).

The two most abundant clusters in the library of 78 clonesbtained at 0.3 gN-NH4

+/L were closed to M. parvus (CN030-1)

nd Methylocaldum sp. 05J-I-7 (CN030-2). Also, a denitrifying pop-lation related to Bacillus firmus (CN030-3) and a heterotrophicitrifying population related to Pseudomonas peli (CN030-4) haseen found.

bacterium BZ45 (GQ246952.1) 96.5

dition.

The cluster CN050-1 was largely the most abundant group in thelibrary of 100 clones obtained at 0.5 gN-NH4

+/L. The closest relativeto cluster CN050-1 is an uncultured Xanthomonadaceae bacteriumisolated from waste-activated sludge from municipal waste watertreatment plant. A clone cluster closed to M. parvus was still present(CN050-4) even if less abundant than in the other libraries. Addi-tionally, a NO2

− oxidizing population close to Nitrospira sp. wasdetected (CN050-5).

4. Discussion

4.1. Methane removal

For all the biofilter sections, the increase of NH4+ concentration

from 0.1 to 0.2 gN-NH4+/L caused a decrease of CH4 biooxidation

(Fig. 2). This kind of decrease was observed by Kravchenko [45] ina study on the effect of NH4

+ concentration on CH4 oxidation inboreal peat grass soil. Between 0.5 and 0.8 gN-NH4

+/kg soil, theyobserved a decrease of CH4-EC from 560 to 160 ng CH4/(g soil h).They also observed a less effect on the CH4-EC for NH4

+ concen-trations below 0.5 and over 0.8 gN-NH4

+/kg soil. The decrease ofCH4-RE may be imputable to the decline of some methanotrophicpopulations (Table 1). Studying CH4 oxidation in soil, King andSchnell [32] observed a similar decrease in relationship with anincrease of NH4

+ concentration. They explained this inhibition bythe presence of NO2

−, a known inhibitor of CH4 oxidation. However,in the present study, NO2

− production remained constant and wasrelatively low (<0.5 mgN/L) throughout the experiments (data notshown). In landfill covered soils, a linear relationship between CH4oxidation and NH4

+ concentration has also been reported [46,47],as observed in the present study.

The fact that CH4-RE was higher in the upper section of the biofil-ter for NH4

+ concentrations lower than 0.25 gN-NH4+/L indicates

that NH4+ was quickly converted to NO3

− and a better availabil-ity of all nutrients (carbon, nitrogen, etc.) led to a higher removal.For this reason, the methanotrophic bacteria were less affected atlow NH4

+ concentrations. This performance could also be associ-

ated with the fact that there were more nitrifying bacteria in theupper section and less NH4

+ remaining. For NH4+ concentrations

above 0.25 gN-NH4+/L, these mechanisms of nitrogen conversion

were not sufficient anymore, resulting in a fall of CH4-RE.

1 neerin

0fpgtt

lttwucitFtt[

4

ebtinMb

lticmb

tgrC

la(TmitcirT

ftsisiht0i

120 M. Veillette et al. / Chemical Engi

Since the PCO2 had an important decrease between 0.1 and.2 gN-NH4

+/L (Fig. 3a), there was less CO2 available for the nitri-ying bacteria. Throughout all the experiments, the upper sectionroduced less CO2 than the other 2 sections, and this differencerew with the increase in NH4

+ concentration in the nutrient solu-ion. This was another sign of a more intense nitrifying activity inhe upper section.

Fig. 3b reveals that the upper section of the biofilter had theowest YCO2 (0.78) and the lowest R2 (0.52). The lowest R2 suggestshat the PCO2 in the upper section was less influenced by EC thanhe NH4

+ concentration. Again, this indicates that the nitrificationas more intense in the upper section as the nitrifying bacteriased CO2 to transform NH4

+ into NO3−. This is also related to the

onditions that the biofilter was operated: the NH4+ concentration

n the biofilm was higher in the upper section as the nutrient solu-ion flowed from top to bottom and NH4

+ was converted to NO3−.

or this reason, the higher YCO2 was observed in the bottom sec-ion. The maximum YCO2 of 1.82, observed in the lower section ofhe biofilter, is comparable to YCO2 of 1.9 obtained by Nikiema et al.27] using NO3

− as a source of nitrogen.

.2. Ammonium conversion

As seen in Fig. 4, the NH4+ conversion decreased all along the

xperiments. This demonstrates that the microbiota present in theiofilter had more difficulty to transform NH4

+ when its concentra-ion in the nutrient solution increased. Among the microbiota foundn middle biofilter section, the methanotrophic bacteria and theitrifying bacteria use similar enzymes for oxidizing CH4 and NH4

+:MO and ammonia monooxygenase (AMO) respectively [48]. So

oth groups can oxidize and be affected by both CH4 and NH4+.

The hydrogencarbonate (HCO3−) concentration could also be a

imiting factor because conventional autotrophic nitrification reac-ions require a HCO3

−/NH4+ molar ratio of 2 [49]. If this molar ratio

s smaller, some nitrifying bacteria could use CO2 as a source ofarbon. Since the HCO3

−/NH4+ molar ratio used for all the experi-

ents was 1, the autotrophic nitrifying bacteria found in the middleiofilter section (N. hamburgensis, Table 1) had to use some CO2.

As shown in Fig. 2, the CH4-RE decreased with NH4+ concen-

ration. This means that the CH4 concentration increase in theas phase and, consequently, also in the biofilm. This could haveeduced the NH4

+ conversion according to the inhibitory effect ofH4 on the nitrifying bacteria observed by Keener and Arp [50].

The interactions among the microbial consortium can be ana-yzed by the conversion of CH4 and NH4

+. For example, between 0.2nd 0.25 gN-NH4

+/L, the CH4-RE increased (Fig. 2), PCO2 decreasedFig. 3a) and the NH4

+ conversion increased from 31 to 48% (Fig. 4).his could be explained by the fact that the CO2 produced byethanotrophs by CH4 oxidation was consumed by the nitrify-

ng bacteria to transform NH4+ into NO3

−. At 0.25 gN-NH4+/L, in

he upper section was observed the highest CH4-RE for NH4+ con-

entrations higher than 0.2 gN-NH4+/L (Fig. 2). The CH4-RE could

ncrease because there was less NH4+ due to the high nitrification

ate in the middle section of the biofilter at this NH4+ concentration.

his shows that CH4-RE and nitrification are closely related.Fig. 5 shows that the RNO3 increased with NH4

+ concentrationrom 0.05 to 0.5 gN-NH4

+/L. The negative values of RNO3 (consump-ion of NO3

−) show that the microbial consortium metabolizedome NO3

− to form new biomass. Denitrifying bacteria were foundn the biofilter (Table 1) and this explains why RNO3 increasedlowly at NH4

+ concentrations above 0.05 gN-NH4+/L, as the dissim-

latory mode progressively took more importance. Other studies

ave shown that NH4+ can be nitrified and denitrified in a biofil-er as nitrogen gas [51]. For NH4

+ concentration between 0.2 and.25 gN-NH4

+/L, the value of RNO3 increases from negative to pos-tive value which indicates a decrease of denitrifying bacteria

g Journal 171 (2011) 1114– 1123

populations. The fact that RNO3 remained constant between 0.25and 0.5 gN-NH4

+/L also confirms that the increase of RNO3 observedat low NH4

+ concentration was attributable to the decrease of den-itrifying bacteria populations. To our best knowledge, the presentstudy shows for the first time the effect of the NH4

+ concentrationin the nutrient solution on a biofilter treating CH4 as well as thebehaviour of RNO3 .

The fact that the Rbed decreased with the amount of NH4+ con-

centration (Fig. 6) means that the bacteria produced less biomass asthe amount of NH4

+ increased. According to the results of CH4-REand PCO2 shown previously, the Rbed followed a similar tendencyto those parameters (Figs. 2 and 3a). It decreased with the NH4

+

concentration. Since the bioxidation of CH4 decreased due to aninhibition induced by NH4

+, the quantity of carbon and nitrogenassimilated into new biomass also decreased. The Rbed negativevalues observed at NH4

+ from 0.3 and 0.5 gN-NH4+/L, signify that

nitrogen was lost in the leachate. The loss of nitrogen could be dueto the necessity of acclimatization for each NH4

+ concentrationchange. During each NH4

+ increment, the microorganisms couldreduce the production of biomass and exopolysaccharides, affect-ing the adhesion of biofilm. As a result, fragments of biomass woulddetach and exit, suspended in the leachate. Since methanotrophs,nitrifiers and denitrifiers have slow growth rates, the recovery ofmetabolic activity in the biofilter and accumulation of biomasswould be slow.

Many authors have shown that CH4 or CH3OH produced bymethanotrophic bacteria could be used as an external source of car-bon for the step of denitrification [52–55]. The Rbed decrease couldbe associated with the decrease of denitrification in the biofilterbecause the carbon source for the denitrifiers, CH3OH produced bythe methanotrophs also decreased. The RNO3 could also be increasedby the activity of heterotrophic nitrifiers which can oxidize a rangeof reduced nitrogen compounds, but most of them can also deni-trify [56]. Moreover, at a value below the critical molar C/N ratiofrom organic carbon and NH4

+, the heterotrophic bacteria con-sume all mineral nitrogen for the formation of new cells [57]. In achemostat with chemolithotrophic NH4

+-oxidizing species such asNitrosomas europea, and heterotrophic species such as Arthrobacterglobiformis, Verhagen and Laanbroek [57] found critical molar C/Nratios between 9.6 and 11.6 using glucose as a source of carbon.In the present study, the inlet C/N molar ratio from N-NH4

+ andC-CH4 decreased from 513 to 103. Moreover, the organic carbonavailable different from CH4 which can be assimilated by most ofthe microorganisms decreased because less CH3OH would be pro-duced by the methanotrophic bacteria. The decrease in C/N molarratio could lead the heterotrophic bacteria to transform insolublebiomass into NH4

+ or NO3−. For example, the heterotrophic nitri-

fying bacteria (P. peli) identified by Vanparys et al. [58] was alsoidentified in the present study at a NH4

+ concentration of 0.3 gN-NH4

+/L. Using a biofilter packed with an inorganic material, Girardet al. [59] also obtained negatives values of Rbed ranging from−0.0004 to 0.212 gN/(m3 h) when NO3

− concentration decreasedfrom 0.25 to 0 gN-NO3

−/L.

4.3. Dry biomass evolution

As seen in Fig. 7, the mean of the biomass decreased with theincrease of the NH4

+ concentration in the nutrient solution. Thiscould be related to the decrease of methanotrophic activity and,consequently, to the decrease of CH4-RE, as previously discussed(Fig. 2). Consequently, as said above the decrease in assimilablenitrogen (NO3

−) and carbon, led to the decrease of biomass pro-

duction. For example, De Visscher and Van Cleemput [35] foundan increase of CH4 oxidation when NH4

+ concentration decreasesand stabilizes at less than 5 mgN/kg soil. The present study explainswhy there was less carbon available for biomass formation as the

eerin

NcFCfnrbsa

4

iiniqcttwcfbNis0cacm

omrpphcpa

p0f[tiusatI

dpCwtrtat

M. Veillette et al. / Chemical Engin

H4+ concentration increased. The decrease in the biomass content

ould also be associated with the heterotrophic nitrifying bacteria.or example, in the middle section, from 0.25 to 0.3 gN-NH4

+/L, theH4-RE increased from 10 to 13% (Fig. 2), but the biomass decreased

rom 8.8 to 6.3 g biomass/kg filter bed (Fig. 7). The heterotrophicitrifying bacteria could use the carbon of exopolysaccharides,esponsible of cell agglomeration and caused a decrease in theiomass content. On the other hand, as the microbiological analy-is was restricted to the middle section of the biofilter, the previousssumption is limited to this section of the biofilter.

.4. Microbiological analysis

In the present study, the increase of NH4+ concentration in an

norganic packed bed biofiter dedicated to CH4 oxidation resultedn a decrease of CH4-RE and PCO2 . The structure of bacterial commu-ities, analyzed by two culture-independent methods, gave some

nsight in the reasons behind this loss of efficiency. While onlyualitative, the DGGE analysis revealed a continuously changingommunity as a result of adaptation to increasing NH4

+ concen-ration. This adaptation was, however, insufficient to maintainhe initial performance of CH4 oxidation in the biofilter, whichas obtained at 0.05 gN-NH4

+/L. Interestingly, the most drastichange in DGGE pattern correlated with a drop in the CH4-RErom 28 to 12%, observed between 0.1 and 0.2 gN-NH4

+/L in middleiofilter section. This indicates that the concentration of 0.2 gN-H4

+/L constitutes a threshold for the inhibition of CH4 oxidizersn the biofilter. As shown in Fig. 8, most DGGE bands were con-tantly observed in the concentrations range between 0.25 and.5 gN-NH4

+/L but were different from those present at low NH4+

oncentrations (0.05 and 0.1 gN-NH4+/L). Furthermore, biofilm

bundance systematically decreased in the higher range of NH4+

oncentrations, reflecting an increasing difficulty to use CH4 asain carbon source by the bacterial microbiota.As revealed by the DNA sequence analysis, the biofilter was col-

nized by bacteria belonging essentially to 4 physiological groups:ethanotrophs, methylotrophs, nitrifying and denitrifying bacte-

ia. Clusters CN005-2, CN030-1 and CN050-4 were related to M.arvus, an aerobic type-II methanotroph [60] which was indeedresent throughout the entire period of experimentation. M. parvusas been already identified in a biofilter treating CH4 at an inlet con-entration of 0.7–0.75% (v/v) [27]. It seems that species related to M.arvus are well-adapted to the biofilter ecological conditions, event NH4

+ concentration of 0.5 gN-NH4+/L in the nutrient solution.

A cluster related to Methylocaldum sp. (CN030-2) is anotherotential methanotroph found in abundance, but transiently at.3 gN-NH4

+/L. This microorganism is a possible member of theamily Methylococcaceae grouping aerobic type I methanotrophs61]. The appearance of a type-I methanotroph population at par-icular periods of the experiment can be explained by their relativensensitivity to NH4

+ in this range of concentrations. In fact, a stim-latory effect of NH4

+ on type I methanotrophs was described byeveral authors [62–64], while inhibition was particularly severegainst type II methanotrophs [65]. However, the disappearance ofhis group at 0.5 gN-NH4

+/L reflected the higher sensitivity of type methanotrophs to environmental changes [66].

The second physiological group, methylotrophs, has no yetirect important reported role in CH4 oxidation. M. methylotro-hus (CN005-1) is an obligate methylotroph that grows readily onH3OH as sole carbon and energy source [67,68]. This methylotrophas the most abundant specie at 0.05 gN-NH4

+/L, exceeding evenhe methanotrophic population. Similarly, the cluster CN050-1,

epresented by an unidentified member of the family of Xan-homonadaceae known as formate utilizers [69,70] was the mostbundant microbial population at 0.5 gN-NH4

+/L, also exceedinghe methanotrophic population. Even considering that the abun-

g Journal 171 (2011) 1114– 1123 1121

dance of methanotrophs may be underestimated because of a biasin the cell extraction method [71], this finding still emphasizesthe importance of methylotrophs in CH4 biofiltration. Their abun-dance indicates that the CH4 biooxidation by the methanotrophsin the biofilter is often incomplete and, as a result, these bacte-ria excrete CH3OH or other one-carbon compounds (C1) such asformaldehyde or formate. These C1 compounds are metabolized bymethylotrophs, which contribute to the overall conversion of CH4into biomass and CO2. The relationship between these two groupscan be interpreted as mutualistic: the methanotrophs supply themethylotrophs in carbon sources while methylotrophs “detoxify”the environment by keeping CH3OH, formaldehyde and formateat low concentrations. The situation in the middle biofilter sec-tion seems to be different from natural environments: a study on amicrobial food web in rice field soil driven by CH4 failed to obtainDNA sequences affiliated to methylotrophs out of a library of 50clones and concluded that methanotrophs did not excrete productsof incomplete CH4 oxidation [72].

A complete switch in the dominant methylotrophic species wasobserved between 0.05 gN-NH4

+/L and 0.5 gN-NH4+/L. This indi-

cates that the methylotrophic consortia had also been stronglyaffected by the increases of NH4

+ concentration. Suppositions of apotential impact of NH4

+ on methylotrophic communities (such asCH3OH-utilizing communities) have already been formulated [73]but, to our best knowledge, the present study is the first to observesuch an effect experimentally.

The two last physiological groups, nitrifying and denitrifyingbacteria, while less abundant, were nevertheless present at all stepsof this biofiltration experiment. Denitrifiers were more abundantthan nitrifiers at 0.05 gN-NH4

+/L, probably due to the high con-centration of NO3

− at this stage. The occurrence of denitrifiers inassociation with methanotrophs was observed by Knowles [54] ingel-stabilized agarose columns that were inoculated with swampsoil. In such conditions, denitrifiers may use C1 compounds likeCH3OH released by the methanotrophs as substrates for growth[54]. Their presence indirectly revealed that anoxic layers maybe created inside the biofilm, which might locally inhibit oxida-tion of CH4. Nitrite oxidizing populations have been observed at0.05 gN-NH4

+/L and 0.5 gN-NH4+/L but no obligate NH4

+ oxidizingbacteria were found at those concentrations. Members of Methy-locystis genus, such as M. parvus may play this role, as suggestedby the kinetic data collected by Nyerges and Stein [74] for NO2

−

production from NH3 at CH4 concentrations higher than 0.05 (v/v).Besides, NH4

+-oxidizing bacteria may be underestimated becauseof a bias in the cells recovery for this bacterial group [75].

5. Conclusion

The results of the study suggest that the inhibition of CH4 oxi-dation by NH4

+ in an inorganic packed bed biofilter occurred onlywhen the NH4

+ concentration reached the tolerance limit of themicrobiota (between 0.1 and 0.2 gN-NH4

+/L). Furthermore, datademonstrate that the CH4 oxidation is the result of a mutualis-tic relation between methanotrophs and methylotrophs and thatboth physiological groups were influenced by NH4

+ concentration.Thus, henceforth for CH4 oxidation, biofilter conditions should beadapted to support a mixed bacterial community.

Acknowledgements

This work was supported by a Strategic grant from the Natu-ral Sciences and Engineering Council of Canada to Michèle Heitz inpartnership with le Centre de Recherche Industrielle du Québec andViaporc inc. The authors want to express their gratitude to the labo-

1 neerin

ri

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

122 M. Veillette et al. / Chemical Engi

atories staff of Chemical and Biochemical Engineering departmentn Université de Sherbrooke.

eferences

[1] Government of Canada, Climate change plan for Canada. http://dsp-psd.pwgsc.gc.ca, 2002 (11.11.2009).

[2] Environment Canada, Canada’s greenhouse gas emissions: understand-ing the trends, 1990–2006. http://www.ec.gc.ca/pdb/ghg/inventoryreport/2008 trends/trends fra.cfm#toc 1, 2008 (17.11.2009).

[3] Gouvernement du Québec, Inventaire québécois des émissionsde gaz à effet de serre en 2006 et leur évolution depuis 1990,Développement durable, Environnement et Parcs Québec, 15 p.http://www.mddep.gouv.qc.ca/changements/ges/2006/index.htm1, 2008.

[4] Statistics Canada, Statistiques financières des exploitations agricoles: recettesmonétaires agricoles. http://www40.statcan.gc.ca/l02/cst01/agri03a-fra.htm,2010 (April 2010).

[5] Fédération des producteurs de porcs du Québec, Portrait de la production.http://www.leporcduquebec.qc.ca/fr/fppq/prod-2.html, 2010 (12.04.2010).

[6] Canadian Pork Council, Statistic: description of Canadian hog farms.http://www.cpc-ccp.com/statistics-farms-f.php, 2009 (02.04.2010).

[7] C. Laguë, Management practice to reduce greenhouse gas emissions from swineproduction systems, Adv. Pork Prod. 14 (2003) 287–300.

[8] J.A. Zahn, J.L. Hatfield, D.A. Laird, T.T. Hart, Y.S. Do, A.A. DiSpirito, Functionalclassification of swine manure management systems based on effluent and gasemission characteristics, J. Environ. Qual. 30 (2001) 635–647.

[9] C. Laguë, E. Gaudet, J. Agnew, T.A. Fonstad, Greenhouse gas emissions fromliquid swine manure storage facilities in Saskatchewan, Trans. ASAE 48 (2005)2289–2296.

10] J.C. Zeikus, The biology of methanogenic bacteria, Bacteriol. Rev. 41 (1977)514–541.

11] Agriculture and Agri-Food Canada, Climate change: questions and answers.http://www4.agr.gc.ca/AAFC-AAC/, 2007 (05.09.2009).

12] P. Forster, Changes in atmospheric constituents and in radiative forcing, in:Climate Change 2007: The Physical Science Basis. Contribution of WorkingGroup I to the Fourth Assessment Report of the Intergovernmental Panel onClimate Change, Cambridge University Press, United Kingdom, New York, NY,USA, 2007.

13] S.G. Sommer, S.O. Petersen, P. Sørensen, H.D. Poulsen, H.B. Møller, Methane andcarbon dioxide emissions and nitrogen turnover during liquid manure storage,Nutr. Cycl. Agroecosyst. 78 (2007) 27–36.

14] E. Kebreab, K. Clark, C. Wagner-Riddle, J. France, Methane and nitrous oxideemissions from Canadian animal agriculture: a review, Can. J. Anim. Sci. 86(2006) 135.

15] M.B. Vanotti, A.A. Szogi, C.A. Vives, Greenhouse gas emission reduction andenvironmental quality improvement from implementation of aerobic wastetreatment systems in swine farms, Waste Manage. 28 (2008) 759–766.

16] J. Martinez, F. Guiziou, P. Peu, V. Gueutier, Influence of treatment techniquesfor pig slurry on methane emissions during subsequent storage, Biosyst. Eng.85 (2003) 347–354.

17] K.H. Hansen, I. Angelidaki, B.K. Ahring, Anaerobic digestion of swine manure:inhibition by ammonia, Water Res. 32 (1998) 5–12.

18] Gouvernement du Québec, Règlement sur les exploitations agricoles, Article31. http://www2.publicationsduquebec.gouv.qc.ca, 2005 (12.11.2008).

19] R.W. Melse, A.W. Van Der Werf, Biofiltration for mitigation of methane emissionfrom animal husbandry, Environ. Sci. Technol. 39 (2005) 5460–5468.

20] R.H. Perry, D.W. Green, J.O. Maloney, Perry’s Chemical Engineers’ Handbook,seventh ed., McGraw-Hill, New York, 1997.

21] J. Nikiema, R. Brzezinski, M. Heitz, Elimination of methane generated from land-fills by biofiltration: a review, Rev. Environ. Sci. Biotechnol. 6 (2007) 261–284.

22] R. Hanson, T. Hanson, Methanotrophic bacteria, Microbiol. Mol. Biol. Rev. 60(1996) 439.

23] M. Delhoménie, M. Heitz, Biofiltration of air: a review, Crit. Rev. Biotechnol. 25(2005) 53–72.

24] M.L. Shuler, F. Kargi, Bioprocess Engineering: Basic Concept, second ed., Pren-tice Hall International Series in the Physical and Chemical Engineering Sciences,Upper Saddle River, NJ, 2002.

25] F.L. Smith, G.A. Sorial, M.T. Suidan, A.W. Breen, P. Biswas, R.C. Brenner, Develop-ment of two biomass control strategies for extended, stable operation of highlyefficient biofilters with high toluene loadings, Environ. Sci. Technol. 30 (1996)1744–1751.

26] M. Delhoménie, L. Bibeau, J. Gendron, R. Brzezinski, M. Heitz, Toluene removalby biofiltration: influence of the nitrogen concentration on operational param-eters, Ind. Eng. Chem. Res. 40 (2001) 5405–5414.

27] J. Nikiema, L. Bibeau, J. Lavoie, R. Brzezinski, J. Vigneux, M. Heitz, Biofiltrationof methane: an experimental study, Chem. Eng. J. 113 (2005) 111–117.

28] A. Avalos Ramirez, S. Bénard, A. Giroir-Fendler, J.P. Jones, M. Heitz, Treatmentof methanol vapours in biofilters packed with inert materials, J. Chem. Technol.

Biotechnol. 83 (2008) 1288–1297.

29] P.L.E. Bodelier, H.J. Laanbroek, Nitrogen as a regulatory factor of methane oxi-dation in soils and sediments, FEMS Microbiol. Ecol. 47 (2004) 265–277.

30] P.A. Steudler, R.D. Bowden, J.M. Melillo, J.D. Aber, Influence of nitrogen fertiliza-tion on methane uptake in temperate forest soils, Nature 341 (1989) 316–341.

[

g Journal 171 (2011) 1114– 1123

31] M. Bender, R. Conrad, Microbial oxidation of methane, ammonium and carbonmonoxide, and turnover of nitrous oxide and nitric oxide in soils, Biogeochem-istry (1994) 97–112.

32] G. King, S. Schnell, Effect of increasing atmospheric methane concentrationon ammonium inhibition of soil methane consumption, Nature 370 (1994)282.

33] S.C. Whalen, Influence of N and non-N salts on atmospheric methane oxidationby upland boreal forest and tundra soils, Biol. Fertil. Soils 31 (2000) 279–287.

34] P. Dunfield, R. Knowles, Kinetics of inhibition of methane oxidation by nitrate,nitrite, and ammonium in a humisol, Appl. Environ. Microbiol. 61 (1995) 3129.

35] A. De Visscher, O. Van Cleemput, Induction of enhanced CH4 oxidation in soils:NH4

+ inhibition patterns, Soil Biol. Biochem. 35 (2003) 907–913.36] P.L.E. Bodelier, P. Roslev, T. Henckel, P. Frenzel, Stimulation by ammonium-

based fertilisers of methane oxidation in soil around rice roots, Nature 403(2000) 421–425.

37] P.L.E. Bodelier, A.P. Hahn, I.R. Arth, P. Frenzel, Effects of ammonium-based fertil-isation on microbial processes involved in methane emission from soils plantedwith rice, Biogeochemistry 51 (2000) 225–257.

38] M. Bender, R. Conrad, Effect of CH4 concentrations and soil conditions on theinduction of CH4 oxidation activity, Soil Biol. Biochem. 27 (1995) 1517–1527.

39] A. De Visscher, M. Schippers, O. VanCleemput, Short term response of enhancedmethane oxidation in landfill cover soils to environmental factors, Biol. Fertil.Soils 33 (2001) 231–237.

40] A. Cornish, K.M. Nicholls, D. Scott, B.K. Hunter, W.J. Aston, I.J. Higgins, J.K.M.Sanders, In vivo 13C NMR investigations of methanol oxidation by the obligatemethanotroph Methylosinus trichosporium OB3b, J. Gen. Microbiol. 130 (1984)2565–2575.

41] USEPA (U.S. Environmental Protection Agency), Sample preservation, in: USEPA(Ed.), Methods for Chemical Analysis of Water and Wastes, USEPA, Cincinnati,OH, USA, 1983, pp. 15–20.

42] G. Muyzer, Denaturing gradient gel electrophoresis (DGGE) in microbialecology[J], in: G.A. Kowalchuk (Ed.), Molecular Microbial Ecology Manual, Dor-drecht, Kluwer, 2004, pp. 743–769.

43] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignmentsearch tool, J. Mol. Biol. 215 (1990) 403–410.

44] M. Nikolausz, R. Sipos, S. Revesz, A. Szekely, K. Marialigeti, Observation of biasassociated with re-amplification of DNA isolated from denaturing gradient gels,FEMS Microbiol. Lett. 244 (2005) 385–390.

45] I.K. Kravchenko, Methane oxidation in boreal peat soils treated with variousnitrogen compounds, Plant Soil 242 (2002) 157–162.

46] P. Boeckx, O. Van Cleemput, I. Villaralvo, Methane emission from a landfill andthe methane oxidising capacity of its covering soil, Soil Biol. Biochem. 28 (1996)1397–1404.

47] S.Y. Park, K.W. Brown, J.C. Thomas, The use of biofilter to reduce atmosphericmethane emissions from landfills: part 1. Biofilter design, Water Air Soil Pollut.155 (2004) 63–85.

48] C. Bédard, R. Knowles, Physiology, biochemistry, and specific inhibitors of CH4,NH4

+, and CO oxidation by methanotrophs and nitrifiers, Microbiol. Rev. 53(1989) 68–84.

49] Metcalf and Eddy Inc., G. Tchobanoglous, F.L. Burton, H.D. Stensel, WastewaterEngineering: Treatment, Disposal and Reuse, fourth ed., McGraw-Hill, NewYork, USA, 2003.

50] W.K. Keener, D.J. Arp, Kinetic studies of ammonia monooxygenase inhibition inNitrosomonas europaea by hydrocarbons and halogenated hydrocarbons in anoptimized whole-Cell Assay, Appl. Environ. Microbiol. 59 (1993) 2501–2510.

51] T. Sakuma, S. Jinsiriwanit, T. Hattori, M.A. Deshusses, Removal of ammonia fromcontaminated air in a biotricklingfilter – denitrifying bioreactor combinationsystem, Water Res. 42 (2008) 4507–4513.

52] F. Thalasso, A. Vallecillo, P. Garcia-Encina, F. Fdz-Polanco, The use of methaneas a sole carbon source for wastewater denitrification, Water Res. 31 (1997)55–60.

53] O. Modin, K. Fukushi, K. Yamamoto, Denitrification with methane as externalcarbon source, Water Res. 41 (2007) 2726–2738.

54] R. Knowles, Denitrifiers associated with methanotrophs and their potentialimpact on the nitrogen cycle, Ecol. Eng. 24 (2005) 441–446.

55] A. Eisentraeger, P. Klang, B. Vansbotter, E. Heymann, W. Dott, Denitrificationof groundwater with methane as sole hydrogen donnor, Water Res. 35 (2001)2261–2267.

56] G.J. Kuenen, L.A. Robertson, Combined nitrification–denitrification process,FEMS Microbiol. Rev. 15 (1994) 109–117.

57] J.M. Verhagen, H.J. Laanbroek, Competition for ammonium between nitrifyingand heterotrophic bacteria in dual energy-limited chemostats, Appl. Environ.Microbiol. 57 (1991) 3255–3263.

58] B. Vanparys, K. Heylen, L. Lebbe, P. De Vos, Pseudomonas peli sp. nov. and Pseu-domonas borbori sp. nov., isolated from a nitrifying inoculum, Int. J. Syst. Evol.Microbiol. 56 (2006) 1875–1881.

59] M. Girard, A. Avalos Ramirez, G. Buelna, M. Heitz, Biofiltration of methane atlow concentrations representative of the piggery industry – influence of themethane and nitrogen concentrations, Chem. Eng. J. 168 (2011) 151–158.

60] J.P. Bowman, L.I. Sly, P.D. Nichols, A.C. Hayward, Revised taxonomy ofthe methanotrophs: description of Methylobacter gen. nov., emendation of

methylococcus, validation of Methylosinus and Methylocystis species, and aproposal that the family Methylococcaceae includes only the group I methan-otrophs, Int. J. Syst. Bacteriol. 43 (1993) 735–753.

61] L. Bodrossy, E.M. Holmes, A.J. Holmes, K.L. Kovacs, J.C. Murrell, Analysis of 16SrRNA and methane monooxygenase gene sequences reveals a novel group of

eerin

[

[

[

[

[

[

[

[

[

[

[

[

M. Veillette et al. / Chemical Engin

thermotolerant and thermophilic methanotrophs, Methylocaldum gen. nov.,Arch. Microbiol. 168 (1997) 493–503.

62] M. Noll, P. Frenzel, R. Conrad, Selective stimulation of type I methanotrophsin a rice paddy soil by urea fertilization revealed by RNA-based stable isotopeprobing, FEMS Microbiol. Ecol. 65 (2008) 125–132.

63] Y. Zheng, L. Zhang, Y. Zheng, H. Di, J. He, Abundance and community compo-sition of methanotrophs in a Chinese paddy soil under long-term fertilizationpractices, J. Soils Sediments 8 (2008) 406–414.

64] Q.F. Qiu, M. Noll, W.R. Abraham, Y.H. Lu, R. Conrad, Applying stable isotopeprobing of phospholipid fatty acids and rRNA in a Chinese rice field to studyactivity and composition of the methanotrophic bacterial communities in situ,ISME J. 2 (2008) 602–614.

65] S.R. Mohanty, P.L.E. Bodelier, V. Floris, R. Conrad, Differential effects of nitroge-nous fertilizers on methane-consuming microbes in rice field and forest soil,Appl. Environ. Microbiol. 72 (2006) 1346–1354.

66] L.Q. Wu, K. Ma, Y.H. Lu, Rice roots select for type I methanotrophs in rice fieldsoil, Syst. Appl. Microbiol. 32 (2009) 421–428.

67] J.A. Greenwood, C.W. Jones, Environmental regulation of the methanol oxi-dase system of Methylophilus methylotrophus, J. Gen. Microbiol. 132 (1986)1247–1256.

68] M. Madhaiyan, S. Poonguzhali, S.-W. Kwon, T.-M. Sa, Methylophilus rhi-zosphaerae sp. nov., a restricted facultative methylotroph isolated from

[

[

g Journal 171 (2011) 1114– 1123 1123

rice rhizosphere soil, Int. J. Syst. Evol. Microbiol. 59 (2009) 2904–2908.

69] O.G. Nercessian, E. Noyes, M.G. Kaluyzhnaya, M.E. Lidstrom, L. Chistoserdova,Bacterial populations active in metabolism of C1 compounds in the sedimentof Lake Washington, a freshwater lake, Appl. Environ. Microbiol. 71 (2005)6885–6899.

70] A. Cébron, L. Bodrossy, N. Stralis-Pavese, A.C. Singer, I.P. Thompson, J.I. Prosser,J.C. Murrell, Nutrient amendments in soil DNA stable isotope probing experi-ments reduce the observed methanotroph diversity, Appl. Environ. Microbiol.73 (2007) 798–807.

71] A. Priemé, J.I.B. Sitaula, A.K. Klemedtsson, L.R. Bakken, Extraction of methane-oxidising bacteria from soil particles, FEMS Microbiol. Ecol. 21 (1996)59–68.

72] J. Murase, P. Frenzel, A methane-driven microbial food web in a wetland ricesoil, Environ. Microbiol. 9 (2007) 3025–3034.

73] S. Kolb, Aerobic methanol-oxidizing bacteria in soil, FEMS Microbiol. Lett. 300(2009) 1–10.

74] G. Nyerges, L.Y. Stein, Ammonia co-metabolism and product inhibition varyconsiderably among species of methanotrophic bacteria, FEMS Microbiol. Lett.297 (2009) 131–136.

75] A. Aakra, M. Hesselsoe, L.R. Bakken, Surface attachment of ammonia-oxidizingbacteria in soil, Microbiol. Ecol. 39 (2000) 222–235.