Upscaling of biohydrogen production process in semi-pilot scale biofilm reactor: Evaluation with food waste at variable organic loads

ww.sciencedirect.com

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 6

Available online at w

ScienceDirect

journal homepage: www.elsevier .com/locate/he

Upscaling of biohydrogen production process insemi-pilot scale biofilm reactor: Evaluation withfood waste at variable organic loads

Suresh Babu Pasupuleti 1, Omprakash Sarkar, S. Venkata Mohan*,1

Bioengineering and Environmental Sciences (BEES), CSIR-Indian Institute of Chemical Technology (CSIR-IICT),

Hyderabad 500 007, India

a r t i c l e i n f o

Article history:

Received 21 April 2013

Received in revised form

29 January 2014

Accepted 6 February 2014

Available online 2 April 2014

Keywords:

Acidogenic fermentation

pH

Volatile fatty acids

COD

Sequencing batch biofilm reactor

* Corresponding author. Bioengineering andHyderabad 500 007, India. Tel.: þ91 40 27191

E-mail addresses: [email protected] Academy of Scientific and Innovative Re

0360-3199/$ e see front matter Copyright ªhttp://dx.doi.org/10.1016/j.ijhydene.2014.02.0

a b s t r a c t

The present account focuses on upscaling of biohydrogen (H2) production at semi-pilot

scale bioreactor using composite food waste. Experiments were conducted at different

organic load (6, 12, 18, 30, 40, 50 and 66 g COD/l) conditions. H2 production increased with

an increasing organic load up to 50 g COD/l (9.67 l/h) followed by 40 g COD/l (6.48 l/h),

30 g COD/l (1.97 l/h), 18 g COD/l (0.90 l/h), 12 g COD/l (0.78 l/h) and 6 g COD/l (0.32 l/h). H2

production was affected by acidification (pH drop to 3.96) at 66 g COD/l operation due to the

excess accumulation of soluble metabolites (5696 mg VFA/l). Variation in organic load of

food waste influenced the overall hydrogen production efficiency.

Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

Fermentativehydrogen (H2) production is a promising approach

for economical and sustainable energy source generation in

which dark/acidogenic fermentation process is one of the

promisingmethod [1e3]. Dark fermentativeH2 production from

renewable resources (waste/wastewater) usingmixed consortia

appears to be the most attractive method compared to other

H2 production processes [4,5]. The main criteria for substrate

(waste/wastewater) selection are availability/nature, cost and

biodegradability [6e9]. Various types of wastewater, viz.,

designed synthetic, chemical, distillery and dairy wastewaters

Environmental Sciences664., [email protected] (S.search (AcSIR), India.2014, Hydrogen Energy P34

having different degrees of biodegradability and composition

especially for H2 production [4]. Apart from industrial waste-

waters, commercial wastes like food waste have attracted

attention in recent years from bioenergy recovery viewpoint,

due to its higher energy potential, biodegradability and inex-

haustibility [10]. At present, in India, a fresh estimate about Rs

580 billion worth of food items get wasted every year [11]. But

only 1% of these are utilized. Food waste is generally composite

in nature and composed of rich organics fraction along with

high moisture content and is highly variable depending on its

source. Exploitation of this highly biodegradable food waste as

source for biological H2 production with simultaneous treat-

ment canbe consideredasaviableandpromisingapproach [11].

(BEES), CSIR-Indian Institute of Chemical Technology (CSIR-IICT),

Venkata Mohan).

ublications, LLC. Published by Elsevier Ltd. All rights reserved.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 67588

H2 production potential depends not only on the nature of

wastewater, but also on operating conditions like reactor

configuration, mode of reactor operation, substrate load,

feeding redox condition and nature of substrate. Operation at

acidophilic conditions (pH 6.0), selective enrichment of parent

inoculum, biofilm configuration and batch mode operation

showed positive influence on the H2 production at bench scale

level reactors [4]. Beyond the optimization of H2 production

process at lab and bench scale level bioreactors, the next stage

of H2 production system is design and operation at the pilot

scale level. Upscaling of H2 production process is critically

important for understanding H2 production efficiency at

higher organic loading operation of wastewater. Optimization

of organic loading of wastewater at pilot scale level reactor for

efficient H2 production is a complex process, especially when

using mixed microflora and wastewater. Also it depends on

the bioreactor and conditions applied for the fermentation

process. Thus, the main objective of this report is to upscale

the H2 production process from bench scale level to semi-

pilot scale level. In this direction, the performance of the

bioreactor was assessed to find the optimum organic load of

food waste by monitoring H2 production rate (HPR), cumula-

tive hydrogen production (CHP), hydrogen conversion effi-

ciency (HCE), and change in the system pH, volatile fatty acids

(VFA) production and COD removal efficiency throughout the

cycle operation.

Materials and methods

Semi-pilot scale bioreactor

Semi-pilot scale biofilm configured anaerobic reactor was

designed and fabricated in the laboratory using ‘Perspex’

material. Fig. 1 depicts the schematic details of the semi-pilot

plant bioreactor. The bioreactor was designed to have an L/D

ratio of 6 (length � diameter; 120 � 20 cm) and was filled with

coir pith (void fractionw 0.18) as fixed bed packingmaterial to

support the growth of H2 producing microflora (Table 1). The

reactor has working/total volume of 20/34 l and gas holding

capacity of 4.5 l (head space). The bioreactor was operated in

batchmodewith an up-flow velocity of 0.50m/day (14 l/day) at

mesophilic temperature (30�2�C). The recirculation ratio

(recirculation: feed ratio) of 1:2 was maintained. Outlet from

the reactor was collected from the bottom and biogas gener-

ated during the reactor operation was collected by water

displacementmethod through an outlet provided at the top of

the bioreactor.

Acidogenic mixed consortia

The anaerobic mixed consortiumwas taken from the recycled

stream after the bio-treatment stage of a municipal waste-

water treatment plant in Hyderabad. The seed sludge was

kept under anaerobic condition for several days before being

used. It was used as seed inoculum after removing stone, sand

and other coarse material. Prior to use the culture was selec-

tively enriched three times in nutrient broth (32��C; 120 rpm;

48 h) under acidophilic conditions (pH 6) [12]. This permits

selective enrichment of acidogenic bacteria from anaerobic

culture. The resulting acidogenic consortia were used as

parent culture for inoculation. Prior to inoculation parent

culture was enriched in designed synthetic wastewater (DSW)

[glucosee 3 g COD/l; NH4Cle 0.5 g/l, KH2PO4 e 0.25 g/l, K2HPO4

e 0.25 g/l, MgCl2 e 0.3 g/l, CoCl2 e 25 mg/l, ZnCl2 e 11.5 mg/l,

CuCl2 e 10.5 mg/l, CaCl2 e 5 mg/l, MnCl2 e 15 mg/l, FeCl3 e

25 mg/l, chemical oxygen demand (COD e 3 g COD/l)] under

anaerobic-acidogenic microenvironment at pH 6.0.

Composite food waste

The food waste was collected from the canteen of CSIR-IICT,

Hyderabad and stored manually by removing any non-food

particles. CSIR-IICT canteen caters about 700e1000 people

per day and the generated waste is composite in nature

comprising uneaten food and food preparation leftovers

which mostly comprising of boiled rice (60�5%; wet weight

basis) followed by cooked vegetables (14�4%), un-cooked

vegetables (spoiled) (2�1%), cooking oil (6�2%), vegetable

peelings (3�2%), cooked meat (4�2%), cooked fish (2�1%) and

boiled spices (1.5�1%). The water content of waste varied

between 15 and 24%. The collected foodwaste wasmasticated

using electrical blender and filtered through stainless steel

sieve to remove coarse materials so as to avoid clogging

problems. Oil present in the waste was separated using an oil-

separating system, that works on the principle of gravity. The

oil free filtrate was used as a substrate after adjusting the

various OLs viz., 6, 12, 18, 30, 40, 50 and 66 g COD/l by tapwater

as required. The food waste was analyzed for pH, COD, ni-

trogen, protein TS and VS as per standard methods [12] and

the results are depicted in Table 2. Food waste showed a

high concentration of COD (5.15 kg COD/l) with a good

biodegradable fraction (BOD/CODe0.75). Prior to feeding the

bioreactor; the redox condition of the foodwastewas adjusted

to 6.0.

Operation and analysis

Experiments were conducted to evaluate the influence of

organic load (OL) on acidogenic consortia in semi-pilot scale

bioreactor utilizing food waste as substrate. Prior to start up,

the bioreactor was inoculated with anaerobic (acidogenic)

consortia [12] (20% of the total bioreactor volume) along with

DSW (3 g COD/l) to support the biofilm formation on the sup-

porting medium (coir pith) by adjusting substrate pH to 6. To

adopt the acidogenic microflora with food waste, initially the

bioreactor was operated at a lower OL of 6 g COD/l of food

waste. In the beginning of each cycle, immediately after

withdrawal (earlier sequence), a predefined volume (20 l) of

foodwastewaterwas fed to the reactorduringfill phaseand the

reactor volume was circulated in closed loop at recirculation

rate (recirculation volume to feed volume ratio) of 3 during the

reaction phase to achieve a homogeneous distribution of the

substrate aswell as uniformdistribution of requisite consortia

along the reactor depth. A peristalitic pump was used to

regulate the feed, recirculation, and decant operations. Bio-

hydrogen generated during the fermentation was estimated

using a microprocessor based pre-calibrated H2 sensor (ATMI

GmbH Inc., Germany). H2 monitoring was done under closed

conditions to avoid external environmental contamination.

Fig. 1 e Schematic representation of semi-pilot scale bioreactor.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 6 7589

The biogas generated during the reactor operation was also

collected by water displacement method through an outlet

provided at the top of the reactor. System pH, volatile fatty

acids (VFA), chemical oxygen demand (COD, closed dichro-

mate refluxing method) and buffering capacity were deter-

mined according to standard methods [13].Theoretically, 1 kg

of COD (wastewater) can produce 468.83 l of H2 (based on ac-

etate pathway of dark fermentation process) [14]. To calculate

the hydrogen conversion efficiency of the wastewater in a

specified reactor, requires cumulative hydrogen production

(CHP, l), organic loadingof the substrate (OL, gCOD/l), substrate

removal efficiency (CODR, %) and substrate feeding volume

(V, l) to the reactor.

Hydrogen conversion efficiencyðHCE; %Þ ¼ ðCHP� 10000Þ=ðOL� CODR � V � THYÞ (1)

where, CHP is cumulative hydrogen production (l), OL is the

organic loading (g COD/l), CODR is the substrate removal effi-

ciency (%), V is the substrate feeding volume to the reactor (l)

and THY is the theoretical hydrogen yield (0.468 l/g CODR).

Results and discussion

Organic loading is one of the most important factors which

influence H2 production. In order to optimize the bioreactor

for efficient H2 production, it is essential to define either a

range of the organic load at which the reactor can function

effectively, or an optimum organic load for a maximum H2

yield [4,15,16]. In some cases higher OL will decrease the H2

yield [15], whereas in some others cases higher OL will in-

crease the H2 yield [15,17]. Table 3 depicts the bench scale

experimental results obtained in our laboratory using

different kind of food based wastewater. The effect of OL of

Citrus limetta peeling on H2 production was evaluated at bench

scale level reactor operation. Among the three OLs of Citrus

limetta peelings, the optimum OL observed was 4.69 kg COD/

m3-day (6 g/l) where the maximum HPR of 10.07 mmol/day

and substrate removal efficiency of 45.45% was recorded [18].

Under similar conditions, the optimum OLs of food waste [19]

and vegetable based market waste [20] were reported as 10 g/l

and 30 g/l respectively. In this study, 6, 12, 18, 30, 40, 50 and

66 g COD/l of food waste was used to operate semi-pilot scale

reactor to assess its ability to convert food waste to H2 with

high yield efficiency.

Biohydrogen

Initially the reactor was started up with lower organic load of

food waste (6 g COD/l) and based on process stabilization

further OLs were applied in ascending order. Prior to feeding

the substrate pH was adjusted to 6.0. Fig. 2 illustrates the H2

Table 1 e Design criteria of semi-pilot scale bioreactor.

Sl. no Parameter

1 Length (L, cm) 120

2 Diameter (D, cm) 20

3 L/D ratio 6

4 Total/working volume (l) 34/20

5 Gas holding capacity (l) 4.5

6 Biofilm supporting material Coconut coir

7 Void ratio 0.18

8 Mode of operation Batch mode

9 Microenvironment Anaerobic

10 Biocatalyst Acidogenic

mixed microflora

11 Substrate feeding pH 6.0

12 Volumetric organic load

(kg COD/m3-day)

6 to 66

13 Operating temperature 30�2 �C14 Design flow (l/day) 14 (up-flow)

15 Recirculation rate

(Recirculation: Feed) (R/F)

1:2

16 Up-flow velocity (m/day)

at R/F of 2

0.5

Table 2 e Characteristics of food waste.

Sl. no Parameter Unit Number

1 COD mg/l 515000

2 Carbohydrates mg/l 24491

3 Protein mg/l 712

4 Alkalinity mg/l 810

5 Total solids mg/l 36290

6 Total suspended solids mg/l 23.68

7 Turbidity NTU 7710

8 pH e 4.2e4.5

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 67590

production rate (HPR) and cumulative H2 production (CHP)

with increasing OL of food waste as the function of operation

time. Majority of H2 production was observed within 24 h of

operation of the all OLs operated. During operation at 6 g COD/

l, the maximum HPR of 0.32 l/h was observed at 20th h fol-

lowed by 0.18 l/h (24th h), 0.16 l/h (16th h), 0.08 l/h (12th h),

0.04 l/h (8th h) and 0.01 l/h (4th h). At lower OL condition the

maximum HPR was observed at 20th h of operation. Thus the

Table 3 e Performance of lab and bench scale reactor with res

Reactorvolume

Wastewater Mode ofoperation

Organic loadingrate (OLR)

(kg COD/m3-day)(k

250 ml Citrus limetta

peelings

Batch/

Suspended

1.17, 2.35 and 4.69

250 ml Composite

food waste

Batch/

Suspended

5, 10, 15 (g/l)

250 ml Vegetable based

marked waste

Batch/

Suspended

4.8, 10.6, 20.0, 32.0,

40.0 and 57.6 (g/l)

2000 ml Chemical

wastewater

Batch/

Biofilm

6.3, 7.1, 7.3

2000 ml Dairy wastewater Batch/

Suspended

2.4, 3.5, 4.7

lag phase of the microflora was high, which might be due to

adaptation of bacteria to food waste and/or low organic con-

centration gradient in the substrate to the microflora. After

stabilization at 6 g COD/l, the reactor was operated at

12 g COD/l, where the maximum HPR of 0.88 l/h was obtained

at 16th h followed by 0.78 l/h (12th h), 0.58 l/h (20th h), 0.33 l/h

(8th h), 0.22 l/h (24th h), and 0.12 l/h (4th h). By increasing the

OL from 6 to 12 g COD/l, the H2 production rate increased from

0.32 l/h to 0.88 l/h. When the reactor was operated at

18 g COD/l showed maximum HPR of 0.90 l/h at 12th h fol-

lowed by 0.84 l/h (16th h), 0.67 l/h (20th h), 0.63 l/h (8th h),

0.48 l/h (4th h) and 0.43 l/h (24th h). By increasing the OL from

12 to 18 g COD/l, insignificant change in the H2 production rate

was noticed, but the production timewas reduced from 16th h

to 12th h. After 18 g COD/l, the reactor was operated at

30 g COD/l, where the maximum HPR of 1.96 l/h (12th h) fol-

lowed by 1.35 l/h (16th h), 1.24 l/h (8th h), 0.89 l/h (4th and

20th h) and 0.66 l/h (24th h) was observed. By increasing the OL

from 18 to 30 g COD/l, significant improvement in HPR (0.90 l/h

to 1.97 l/h) was observed. Subsequently, the reactor was

operated at 40 g COD/l, where the maximum HPR of 6.48 l/h

(8th h) followed by 5.69 l/h (12th h), 4.91 l/h (16th h), 2.95 l/h

(4th h), 2.46 l/h (20th h) and 1.96 l/h (24th h) was observed. By

increasing the OL from 30 to 40 g COD/l, significant improve-

ment was observed in HPR. This might be due to availability of

high organic gradient to the acidogenic microflora. The

reactor further operatedwith 50 g COD/l, themaximumHPR of

9.67 l/h (8th h) followed by 9.36 l/h (12th h), 6.98 l/h (16th h),

5.16 l/h (20th h), 3.96 l/h (4th h) and 2.20 l/h (24th h) was

observed. By further increasing the OL from 40 to 50 g COD/l,

significant improvement in HPR (9.67 l/h to 6.48 l/h) was

observed. By further increasing the OL from 50 to 66 g COD/l,

the HPR dropped significantly (1.967 l/h (4th h); 0.88 l/h

(20th h); 0.85 l/h (12th h); 0.61 l/h (16th h); 0.59 l/h (24th h);

0.56 l/h (8th h). The results showed that increasing OL had a

positive influence on H2 production up to certain extent

(Fig. 2). In particular, at 50 g COD/l of food waste showed

higher H2 production among all the OLs studied.

Comparing the overall performance of the semi-pilot

scale biofilm configured reactor, the variations in the

maximumH2 production rate (HPR) was 9.67 l/h at 50 g COD/l

followed by 6.48 l/h (40 g COD/l), 1.97 l/h (30 g COD/l), 1.96 l/h

(66 g COD/l), 0.90 l/h (18 g COD/l), 0.78 l/h (12 g COD/l)

pect to OLs of various wastewaters.

Optimum OLRg COD/m3-day)

Maximum HPR(mmol/day)

Substrateremovalefficiency

(%)

Reference

4.69 10.07 45.45 [18]

10 8.15 76.3 [19]

30 g/l 23.96 62 [20]

6.3 23.81 22.6 [21,24,27]

3.5 3.18 64.7 [22]

Fig. 2 e Biohydrogen production rate and cumulative hydrogen production with respect to organic load of food waste.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 6 7591

and 0.32 l/h (6 g COD/l). Reactor operation at 40 g COD/l

recorded maximum cumulative hydrogen production (CHP)

of 110 l followed by 50 g COD/l (106 l), 66 g COD/l (77 l),

30 g COD/l (54 l), 31 l (18 g COD/l), 13 l (12 g COD/l) and 5.49 l

(6 g COD/l).

Volatile fatty acids synthesis

Acidogenic process generates H2 along with CO2 and volatile

fatty acids (VFA, represented as a total of acids). The produc-

tion of VFA reflects the metabolic pathways, were affected by

both the carbon load and fermentation conditions. The

production of VFA also influences the pH and thus H2 pro-

duction. VFA and system pHweremonitored during the semi-

pilot scale reactor operation in order to assess the bioprocess

mechanism during H2 production. To enumerate the possible

relation between H2 production and VFA production, a plot

was constructed against fermentation time with different OLs

(Fig. 3). In the present study, VFA generation was found to

depend on the waste load applied to the bioreactor, which in

turn affects the system pH and H2 production rate. VFA gen-

eration was found to increase with increase in organic load

(Fig. 3). Initially, at 6 g COD/l, the maximumVFA generation of

2353 mg/l was recorded followed by increment to 2597 mg/l

Fig. 3 e VFA production profile and change in the system pH during H2 production from food waste at variable organic load.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 67592

(12 g COD/l), 3840 mg/l (18 g COD/l), 3143 mg/l (30 g COD/l),

4004 mg/l (40 g COD/l), 4059 mg/l (50 g COD/l) and 5639 mg/l

(66 g COD/l). During all organic loading conditions studied, a

gradual increment in VFA concentration was observed till

66 g COD/l. At 66 g COD/l, high VFA concentration was

observed (5639 mg/l) due to availability of higher organic

gradient of substrate and resulting acidogenic metabolism.

Increase in VFA concentration during H2 production illus-

trated the effective function of the acidogenic metabolic pro-

cess and was concordant with the corresponding H2

production. However, VFA concentration and system pH

conditions are integral expressions of anaerobic microenvi-

ronments. Production of acids gradually reduces the buffering

capacity of system, which, in turn, results in a decline in the

system pH condition due to accumulation of organic acids

leading to process inhibition [4,21,23]. If the system pH con-

dition is not maintained in the optimum range, termination of

H2 production would occur. H2 production from food waste in

semi-pilot scale level reactor with simultaneous production of

soluble acidic intermediates (VFA) influences the redox

microenvironment of the substrate. The variation in the sys-

tem pH is illustrated in Fig. 3. Accumulation of fatty acids

showed a gradual decline in the system pHwith respect to the

OL operation. The change in system pH condition and VFA

concentration of substrate provided information about

metabolic processes involved in H2 production. Upon opera-

tion of reactorwith OLs 6e50 g COD/l, the systempH condition

varied between 4.5 and 6.7 during cycle operation, which is

Fig. 4 e Substrate degradation rate and COD removal efficiency during H2 production in semi-pilot scale bioreactor with food

waste.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 6 7593

more or less optimum for the fermentative H2 production and

also supports growth of acidogenic bacteria. Whereas, at

66 g COD/l, maximum VFA was registered of 5696 mg/l with

system pH of 3.96. The resulting redox microenvironment

(pH<4.5) was considered unfavorable for the function of

acidogenic microflora which results in cessation of H2 pro-

duction. Operations with lower OLs (6e50 g COD/l), the pH of

the system can be maintained within the favorable redox

range till the end of the cycle operation without process

inhibition.

Substrate (COD) removal efficiency

Utilization of food waste for the H2 production in semi-pilot

scale biofilm reactor was evaluated in terms of substrate

degradation rate (SDR, kg COD/m3-day) and COD removal ef-

ficiency (Fig. 4). Gradual increase in the OL of food waste will

acclimatize the system microenvironment to higher OLs. SDR

was found to increase alongwith increase in loading condition

from 6 g COD/l to 50 g COD/l (Fig. 4). Maximum SDR was

observed at 50 g COD/l (12.30 kg CODR/m3-day) operation fol-

lowed by 40 g COD/l (8.02 kg CODR/m3-day), 30 g COD/l

(5.82 kg CODR/m3-day), 18 g COD/l (3.88 kg CODR/m

3-day),

12 g COD/l (2.4 kg CODR/m3-day) and 6 g COD/l (1.25 kg CODR/

m3-day). On the contrary to OLs 6e50 g COD/l, subsequent

increase in the organic loading condition (66 g COD/l) showed

decrement in SDR (9.39 kg CODR/m3-day). Thismight be due to

acidification of redox microenvironment which reduced the

function of microflora to metabolize the substrate. COD

removal efficiency varieswith increase in the carbon load. The

maximum COD removal efficiency was observed with

6 g COD/l (87.05%) followed by 40 g COD/l (86.61%), 30 g COD/l

(83.93%), 18 g COD/l (83.9%), 66 g COD/l (82.12%), 50 g COD/l

(81.20%) and 12 g COD/l (78.07%).

Fig. 5 e Biohydrogen conversion efficiency of food waste with respect to organic loading of food waste in semi-pilot scale

bioreactor.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 67594

Hydrogen conversion efficiency

Theoretically, 1 kg of COD (wastewater) can produce 468.83 l of

H2 (based on acetate pathway in dark fermentation process)

[14]. Fig. 5 illustrates the hydrogen conversion efficiency (HCE)

of semi-pilot plant bioreactor with respect to all OLs of food

waste. HCE gradually increased with increase in OL up to

40 g COD/l and declined at 50 and 66 g COD/l. MaximumHCE of

41.03% was observed at 40 g COD/l followed by 27.58%

(50 g COD/l), 25.73% (30 g COD/l), 24.78% (18 g COD/l), 16.48%

(12 g COD/l), 13.78% (66 g COD/l) and 12.47% (6 g COD/l). HCE

when calculated within 24 h of cycle operation showed

maximum values at 50 g COD/l. Thismight be due to higher H2

production as well as HPR observed at 50 g COD/l within 24 h

of operation only. The maximumHCE of 56.87% was observed

at 50 g COD/l followed by 48.42% (40 g COD/l), 21.57%

(12 g COD/l), 19.31% (30 g COD/l), 18.82% (18 g COD/l), 12.82%

(6 g COD/l) and 6.49% (66 g COD/l). It is apparent from the

experimental data that the OL of food waste as well as

retention time showed significant influence on the overall

performance of semi-pilot scale bioreactor for H2 production.

HCE value depends on the organic loading of waste, H2 pro-

duction values, COD removal and HRT.

H2 production is accompanied with acid production

coupled with solvent production. Production of acidic in-

termediates (VFA) reflects the changes in the metabolic

process involved and provides information for improving

the conditions favorable for H2 production. Visible differ-

ence in VFA production was observed between feeding OL

where, the maximum VFA concentration of 5696 mg/l was

documented at 66 g COD/l of food waste. A sudden shoot

up in VFA generation was observed at 66 g COD/l, which

remained more or less the same thereafter. So, the rapid

increase in VFA at OL 66 g COD/l the substrate redox

condition was dropped to 3.96 (Fig. 6). Higher drop in the

system pH visualized rapid ceased H2 production with

concomitant increase in VFA production. System pH con-

dition was reported to be one of the crucial factors

affecting dark fermentation due to its influence on hy-

drogenase activity. The optimal system pH condition range

of 4.5e6.5 is reported to be ideal to avoid both methano-

genesis and solventogenesis, which is important for effi-

cient H2 production. However, highly acidic pH condition

(<4.5) is detrimental to H2 production as it inactivates H2

producing bacteria. Effective H2 production was observed

by operating semi-pilot plant bioreactor till 50 g COD/l of

food waste. System pH condition is a fundamental oper-

ating parameter for the production of H2. Biofilm reactors

provide compact and high-rate processes [24]. The robust

performance with respect to H2 production in biofilm re-

actors under acidophilic conditions might be attributed to

its inherent potential to sustain high biomass concentra-

tion with homogeneous distribution throughout the

reactor [25e27]. Self-immobilization of microflora as bio-

film resulted in high biomass hold up, which enabled the

process to be operated significantly at higher liquid

throughput [24e28]. Biofilm systems are particularly useful

where slow growing organisms with special metabolic

capacities can be protected from washout [27,28]. Reten-

tion of high biomass increases the process stability and

resistance to shock loads.

Conclusions

Experimental results revealed that the increasing organic load

of food waste increases the fermentative H2 production till

50 g COD/l and at higher organic loading (66 g COD/l) the

system gets inhibited. The maximum H2 production rate of

9.67 l/h was observed at OL 50 g COD/l. At higher OL of 66 g/l,

Fig. 6 e Influence of VFA production on system redox condition and H2 production rate.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 6 7595

hydrogen production rate dropped to 1.97 l/h. The system pH

condition was very acidic (3.96) due to high VFA generation

(5696.46 mg/l) at 66 g COD/l, might inactivated the microflora.

Increase in organic loading showedmarginal variation in COD

removal efficiency and demonstrated relatively good correla-

tion with both pH and VFA concentration.

Acknowledgments

The authors wish to thank Director, CSIR-IICT for encour-

agement in carrying out this work. This work was supported

by Ministry of New and Renewable Energy (MNRE),

Government of India as Mission Mode Project on Biohydrogen

production (103/131/2008-NT). SBP duly acknowledges Council

of Scientific and Industrial Research (CSIR), New Delhi for

providing research fellowship.

r e f e r e n c e s

[1] Ren Y, Wang J, Liu Z, Ren Y, Li G. Hydrogen production fromthe monomeric sugars hydrolyzed from hemicellulose byEnterobacter aerogenes. Renew Energy 2009;34:2774e9.

[2] Mohanakrishna G, Venkata Subhash G, Venkata Mohan S.Adaptation of biohydrogen producing reactor to highersubstrate load: redox controlled process integration strategy

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 7 5 8 7e7 5 9 67596

to overcome limitations. Int J Hydrogen Energy2011;36:8943e52.

[3] Venkata Mohan S, Pandey A. Biohydrogen production: anintroduction. In: Pandey A, Chang JS, Hallenbeck PC,Larroche, editors. Elsevier: Burlington; 2013. pp. 1e24 (ISBN:978-0-444-59555-3).

[4] Venkata Mohan S. Waste to renewable energy: a sustainableand green approach towards production of biohydrogen byacidogenic fermentation. In: Om S, Steve H, editors.Sustainable biotechnology: renewable resources and newperspectives. New York: Springer; 2010. pp. 129e64.

[5] Rittmann BE. Opportunities for renewable bioenergy usingmicroorganisms. Biotechnol Bioeng 2010;100(2):203e12.

[6] Kapdan IK, Kargi F. Bio-hydrogen production from wastematerials. Enzyme Microb Technol 2006;38:569e82.

[7] Venkata Mohan S. Harnessing of biohydrogen fromwastewater treatment using mixed fermentative consortia:process evaluation towards optimization. Int J HydrogenEnergy 2009;34:7460e74.

[8] Venkata Mohan S, Mohanakrishna G, Raghavulu SV,Sarma PN. Enhancing biohydrogen production fromchemical wastewater treatment in anaerobic sequencingbatch biofilm reactor (AnSBBR) by bioaugmenting withselectively enriched kanamycin resistant anaerobic mixedconsortia. Int J Hydrogen Energy 2007;32(15):3284e92.

[9] Mohanakrishna G, VenkataMohan S, Sarma PN. Utilizingacid-rich effluents of fermentative hydrogen productionprocess as substrate for harnessing bioelectricity: anintegrative approach. Int J Hydrogen Energy2010;32(8):3440e9.

[10] Shin H, Youn J. Conversion of food waste into hydrogen bythermophilic acidogenesis. Biodegradation 2005;16(1):33e44.

[11] Goud RK, Suresh Babu P, Venkata Mohan S. Canteen basedcomposite food waste as potential anodic fuel forbioelectricity generation in single chambered microbial fuelcell (MFC): bio-electrochemical evaluation under increasingsubstrate loading condition. Int J Hydrogen Energy2011;36:6210e8.

[12] Venkata Mohan S, Lalit Babu V, Sarma PN. Effect of variouspretreatment methods on anaerobic mixed microflora toenhance biohydrogen production utilizing dairy wastewateras substrate. Bioresour Technol 2008;99:59e67.

[13] APHA. Standard methods for the examination of water andwastewater. 18thed.WashingtonDC,USA:APHA/AWWA;1998.

[14] Venkata Mohan S, Mohanakrishna G, Ramanaiah SV,Sarma PN. Simultaneous biohydrogen production andwastewater treatment in biofilm configured anaerobicperiodic discontinuous batch reactor using distillerywastewater. Int J Hydrogen Energy 2008;33(2):550e8.

[15] Hafez H, Nakhla G, Hesham M, Naggar El, Elbeshbishy E,Baghchehsaraee B. Effect of organic loading on a novelhydrogen bioreactor. Int J Hydrogen Energy 2010;35:81e92.

[16] Shen L, Bagley DM, Liss SN. Effect of organic loading rate onfermentative hydrogen production from continuous stirredtank and membrane bioreactors. Int J Hydrogen Energy2009;34(9):3689e96.

[17] Zhang JJ, Li XY, Oh SE, Logan BE. Physical and hydrodynamicproperties of flocs produced during biological hydrogenproduction. Biotechnol Bioeng 2004;88(7):854e60.

[18] Venkata Mohan S, Lenin Babu M, Venkateswar Reddy M,Mohanakrishna G, Sarma PN. Harnessing of biohydrogen byacidogenic fermentation of Citrus limetta peelings: effect ofextraction procedure and pretreatment of biocatalyst. Int JHydrogen Energy 2009;34:6149e56.

[19] Venkata Mohan S, Chiranjeevi P, Mohanakrishna G. A rapidand simple protocol for evaluating biohydrogen productionpotential (BHP) of wastewater with simultaneous processoptimization. Int J Hydrogen Energy 2012;37:3130e41.

[20] Venkata Mohan S, Mohanakrishna G, Kannaiah Goud R,Sarma PN. Acidogenic fermentation of vegetable basedmarket waste to harness biohydrogen with simultaneousstabilization. Bioresour Technol 2009;100:3061e8.

[21] Bhaskar YV, Venkata Mohan S, Sarma PN. Effect of substrateloading rate of chemical wastewater on fermentativebiohydrogen production in biofilm configured sequencingbatch reactor. Bioresour Technol 2008;99:6941e8.

[22] Venkata Mohan S, Lalit Babu V, Sarma PN. Anaerobicbiohydrogen production from dairy wastewater treatment insequencing batch reactor (AnSBR): effect of organic loadingrate. Enzyme Microb Technol 2007;41:506e15.

[23] Fang HHP, Li C, Zhang T. Acidophilic biohydrogen productionfrom rice slurry. Int J Hydrogen Energy 2006;31:683e92.

[24] Rao NC, Venkata Mohan S, Muralikrishna P, Sarma PN.Treatment of composite chemical wastewater by GAC-Biofilm configured sequencing batch reactor (SBGR) operatedin aerobic environment. J Hazard Mater 2005;124:59e67.

[25] Keskin T, Giusti L, Azbar N. Continuous biohydrogenproduction in immobilized biofilm system versus suspendedcell culture. Int J Hydrogen Energy 2012;37:1418e24.

[26] Zhang ZP, Show KY, Tay JH, Liang DT, Lee DJ. Biohydrogenproduction with anaerobic fluidized bed reactors e acomparison of biofilm-based and granule-based systems. IntJ Hydrogen Energy 2007;33(5):1559e64.

[27] Chiranjeevi P, Naresh Kumar A, Venkata Mohan S. Criticalassessment of biofilm and suspended growth reactorconfigurations for acidogenic biohydrogen production usingwastewater as a function of redox microenvironment.Int J Hydrogen Energy 2014;39(14):7561e71.

[28] Venkata Mohan S, Chandrasekhar Rao N, Sarma PN. Low-biodegradable composite chemical wastewater treatment bybiofilm configured sequencing batch reactor (SBBR). J HazardMater 2007;144:108e17.