ww.sciencedirect.com

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6

Available online at w

ScienceDirect

http: / /www.elsevier .com/locate/biombioe

Thermophilic biohydrogen production usingpre-treated algal biomass as substrate

Shantonu Roy, Kanhaiya Kumar, Supratim Ghosh, Debabrata Das*

Department of Biotechnology, Indian Institute of Technology, Kharagpur 721302, India

a r t i c l e i n f o

Article history:

Received 19 August 2013

Received in revised form

5 December 2013

Accepted 8 December 2013

Available online 2 January 2014

Keywords:

Algal biomass

Pretreatment methods

Biohydrogen production

Thermophilic mixed culture

Gompertz equation

* Corresponding author. Tel.: þ91 3222 83758E-mail addresses: ddas.iitkgp@gmail.com

0961-9534/$ e see front matter ª 2013 Elsevhttp://dx.doi.org/10.1016/j.biombioe.2013.12.

a b s t r a c t

Algal biomass is rich in carbohydrates which can be utilized as a promising source of

substrate for dark fermentation. It becomes more significant when biomass is produced by

capturing atmospheric greenhouse gas, CO2. In the present study, clean energy was

generated in the form of biohydrogen utilizing algal biomass. Biohydrogen production was

carried out by thermophilic dark fermentation using mixed culture. The culture of Chlorella

sorokiniana was cultivated in helical airlift photobioreactor at 30 �C under continuous light

intensity of 120 mmol m�2 s�1 provided by white fluorescent lamps. Biomass reached to

stationary phase on 9th day giving maximum dry cell weight of 2.9 kg m�3. Maximum

carbohydrate and protein content observed was 145 g kg�1 and 140 g kg�1, respectively.

Maximum volumetric productivity of 334 g dm�3 d�1 was observed. Algal biomass was

subjected to various physical and chemical pre-treatments processes for the improvement

of hydrogen production. It was observed that the pretreatment with 200 dm3 m�3 HCl-heat

was most suitable pretreatment method producing cumulative hydrogen of 1.93 m3 m�3

and hydrogen yield of 958 dm3 kg�1 volatile suspended solid or 2.68 mol mol�1 of hexose.

Growth kinetics parameters such as mmax and Ks were estimated to be 0.44 h�1 and

120 g m�3, respectively. The relationship between biomass and hydrogen production was

simulated by the LuedekingePiret model showing that H2 production is growth associated.

The study thus showed the potential of algal biomass as substrate for biological hydrogen

production.

ª 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The recent exponential increase in worldwide energy demand

caused depletion of energy reserves at greater pace. The

combustion of fossil fuels has serious negative effects on

environment because of CO2 emission. Algal biomass culti-

vation is gaining importance in recent times as they can

capture atmospheric CO2 and can produce carbohydrates rich

biomass which can be used for production of biofuels [1].

Unlike other crops such as corn or soybeans, algae can use

; fax: þ91 3222 255303., ddas@hijli.iitkgp.ernet.iier Ltd. All rights reserved006

various water sources ranging from wastewater to brackish

water and can be grown in small, intensive plots on denuded

land. Hydrogen from algae is possible by two biological pro-

cesses. The first is the biophotolysis involving light-driven

splitting of the water [1,2]. Hydrogen production by bio-

photolysis had been extensively studied on Chlamydomonas

reinhardtii or Anabaena variabilis [3]. Secondly, dark fermenta-

tion of biomass utilizing carbohydrates present in algal cells

using thermophilic and mesophilic hydrogen producing bac-

teria. Thermophilic dark fermentation shows favorable ther-

modynamics of reaction and with reduced risk of

n (D. Das)..

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6158

methanogenic contamination, higher rate of hydrolysis and

higher hydrogen yields [4]. Algal biomass seems to be a po-

tential feedstock source for biofuel production because they

have higher growth rate, rich carbohydrate content and sim-

ple to harvest. Different processes are available for the pro-

duction of fuels from algae [5] which includes anaerobic

digestion of microalgae biomass to produce methane [6].

Moreover, direct extraction of fuel oils in the form of biodiesel

that remains accumulated in certain microalgae [7]. Micro-

algae can also be used to produce hydrogen photosyntheti-

cally by direct or indirect photolysis of water [8]. At this point,

there is one report on biohydrogen production from thermo-

philic dark fermentation using biomass of Chlorella vulgaris [9].

In this study, different pretreatments improved hydrogen

production. Different physico-chemical pretreatments were

employed to increase the accessibility of different complex

sugars entrapped in algal biomass into simpler form. Carbo-

hydrates in algal biomass are found as intracellular complex

polymeric form bounded with rigid algal cell walls [10,11].

Therefore, it is necessary to break the algal cell wall alongwith

complex carbohydrate to facilitate the release of simple sugar.

Cost of pretreatment of biomass adds significantly to overall

hydrogen production process. Several methods of pre-

treatments such as physical (sonication, milling, grinding,

pyrolysis), chemical (acid, alkali, thermal, H2O2) and biological

methods (enzymatic, microbial) have been reported to break

algal cell wall, hydrolyze the complex carbohydrates and

release fermentable sugars. Each of the method has its own

merits and demerits [12]. Physical methods are based on

simpler technology but they are energy intensive processes

limiting their use for commercial application. Biological based

methods are costly and time consuming with low hydrolysis

rate. Preference of chemical method such as acid treatment

over others is because of higher conversion efficiency of

polymeric carbohydrates into simpler sugars in lesser time

[10,13]. Some reports are available on use of algal biomass as

substrate after pretreatment for hydrogen production using

mesophilic microorganisms [14]. In a study, algal biomasswas

used to produce H2 using Clostridium butyricum and subse-

quent use of produced organic acids for H2 production by

photo fermentation using Rhodobacter sphaeroides KD 131 [14].

Surface chemistry of biomass changes with different

pretreatments. X-ray diffraction (XRD) has been used to study

the changes in crystallinity with respect to different

pretreatment processes [15]. Effect of acid and enzymatic

pretreatment on surface chemistry of cornstalk for

improvement of biohydrogen productionwas studies by using

X-ray diffraction [16].

Similarly, fermentation of algal biomass by amixed culture

of lactic acid bacterium Lactobacillus amylovorus and photo-

synthetic bacterium Rhodobium marinum was used in the

single-step process to convert algal starch to H2 [11].

Monod model and Luedeking-Piret unstructured models

were used for the determination of the kinetic parameters.

These models allow us to relate microbial growth rate and

product formation in an aqueous environment to the con-

centration of a limiting nutrient. These kinetic parameters

helps us to understand the growth characteristics of the

microorganism, design the bioprocess based experiment and

scaling up the process.

Thus, the present study aimed to investigate the effect of

different pretreatment methods on saccharification of com-

plex carbohydrates present in algal biomass; Chlorella sor-

okiniana. Cell morphology was studied to investigate the

extent cell wall rupture using microscopic and X-ray diffrac-

tion (XRD) with respect to different pretreatment methods.

Optimized concentration of pretreated algal biomass was

further used as carbon source in H2 production medium for

thermophilic mixed culture. Attempts were also made to

determine the kinetics of hydrogen production using con-

ventional Monod, Luedeking-Piret and modified Gompertz

equations.

2. Materials and methods

2.1. Cultivation and harvesting of microalgae

The culture of C. sorokiniana was cultivated in helical airlift

photobioreactor at 30 �C under continuous light intensity of

120 mmol m�2 s�1 provided by white fluorescent lamps. The

helical photobioreactor was constructed of four parts: i) the

helical photostage; ii) the gas riser; iii) the degasser and iv) the

downstream tube. The construction material used was plex-

iglass (polymethacrylate). The helical photostage had tube

dimensions of 2.0 cm ID and 3.0 cm OD wound on a 15.0 cm

imaginary cylinder with a gap of 1.0 cm between the spirals. It

had 8 spirals in total with a height of 32.0 cm. The gas riser

consisted of a T-shaped joint which had a top diameter of

8.0 cm and a bottom diameter of 2.0 cm. The degasser was

cylindrical in shape with a height of 12.0 cm. The downflow

tube connected the bottom of the degasser with the end of the

helix via a T-piecewhich also served as a sampling port. Inside

the reactor, air sparging had a space velocity of 0.33 min�1 for

proper circulation andmixing of the culture. The C. sorokiniana

was grown in mTAP [-acetate] medium which was prepared

by substituting the nitrogen source NH4Cl by 1.5 kg m�3 of

NaNO3 [17]. Algal biomass was harvested at stationary phase

and centrifuged at 4053� g RCF for 5min. It was washed thrice

with distilled water and further air dried to get a powdered

algal biomass.

2.2. Microbial consortium

An enriched thermophilic mixed culture capable of producing

hydrogen was used in this study. Media used for this study

consist of Na2HPO4 (4.2 kg m�3), KH2PO4 (1.5 kg m�3), NH4Cl

(1.95 kg m�3), MgCl2 (0.18 kg m�3), yeast extract (2.0 kg m�3),

glucose (10 kg m�3), cysteine HCl (1 kg m�3), vitamins solution

(DSMZ medium No141, German Collection of Microorganisms

and Cell Cultures). Thismixed culture derived from a distillery

industry anaerobic digester [18] had predominantly Thermoa-

naerobacterium sp. genus and was mostly related to Thermoa-

naerobacterium thermosaccharolyticum which have hydrogen

production ability.

2.3. Algal biomass pretreatment

Dried algal biomass with initial concentration of 14 kg m�3 was

subjected to physical and chemical pre-treatments such as

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6 159

HCl�heat, H2O2, autoclaving, sonication etc prior to be used as

substrate for fermentation.Such typeofpretreatment facilitates

availability of simpler sugar frommorecomplexpolymeric form

of carbohydrates present in algal biomass [9]. In heat-HCl pre-

treatmentmethod, dried algal biomass wasmixedwith various

concentration of HCl from 50 dm3 m�3 to 250 dm3 m�3 and

autoclaved at 121 �C, 204.9 kPa for 20 min. Entire hydrolyzates

was neutralized with 0.4 g m�3 NaOH. In H2O2 treatment

method, dried algal biomass was mixed with 2% H2O2 for over-

night and then autoclaved at 121 �C, 204.9 kPa for 20 min. Son-

ication was performed at 50% of amplitude for 10 min (130 W,

SonicVibraCellUltrasonicProcessor). Theentire sonicatedalgal

biomass containing solubilizedcomplex carbohydratewasused

as substrate to prepare H2 production medium.

2.4. Hydrogen production in batch fermentation

Hydrogen production in batch with optimized pretreated algal

biomass was carried out in double jacketed bioreactor under

thermophilic condition. The volume of the bioreactor was

300 cm3 with working volume of 250 cm3. Temperature was

controlled at 60 �C by passing water in the outer jacket of the

bioreactor.

2.5. Analytical procedure

The amount of H2 was quantitatively measured by gas chro-

matography (GC, Agilent Technologies, USA) with a thermal

conductivity detector (TCD) with N2 as the carrier gas. The

column used for gas determination was a stainless column (1/

8 inch 15 ft) packed with 50/80 mesh Carboxen 1000 (Supelco)

[18].

Total sugar was estimated using Phenol sulfuric method as

described by Loewus [19]. The pH values were monitored

using a desktop pH meter (pH510, Cyberscan, Singapore). The

COD was measured according to APHA standard methods [20]

using a COD measurement instrument set (DRB200 & DR2800

Portable Spectrophotometer, HACH, USA).

The crystallinity index (CrI) of the algal biomass was

measured by XRD using an X-ray powder diffractometer

(Bruker Axs: D8 focus) with Co Ka radiation at 40 kV and

200mA. The CrI of the sampleswere calculated using thewide

angle X-ray diffraction counts at 2q angle close to 22� and 18�

according to the Segal empirical method [21,22].

CrI ¼ I22 � I18I22

� 100% (1)

where I22 is the peak intensity of the crystalline material

(2q ¼ 22�) and I18 is the peak intensity of the amorphous ma-

terial (2q ¼ 18�).

2.6. Determination of kinetic constants

The Monod model was used for determining the growth ki-

netics as follows [23]:

m ¼ mmaxSSþ Ks

(2)

where m (h�1) is the specific growth rate; mmax (h�1) is the

maximum specific growth rate respectively; Ks (g m�3) is the

substrate affinity coefficient; and S (g m�3) is the substrate,

concentration.

Linearization of Eq. (2) gives LineweavereBurk plot as

shown in Eq. (3):

1m¼ 1

mmax

þ Ks

mmax

1S

(3)

The relationship between biomass and the products for the

anaerobic hydrogen production by mixed anaerobic cultures

could be simulated by the LuedekingePiret model:

dPdt

¼ adxdt

þ bx (4)

The first term of Eq. (4), i.e., adX/dt, is referred as the

growth-associated product formation rate. This implies that

the growing cells produce the product in constant proportion

of their growth. Non-growth associated product formation

term (bX) shows that all the microorganisms produce the

product in a constant proportion of their concentration,

regardless of the growth phase. Eq. (4) could be changed into:

Qp ¼ dPxdt

¼ amþ b (5)

where dP/xdt is the specific product formation rate of product

(Qp) and m is the specific growth rate of the microorganisms. A

straight line could be obtained with an intercept of b and a

slope of a, if plotting Qp against m.

The modified Gompertz equation was used to fit cumula-

tive hydrogen production in the batch experiment performed

under optimized process parameters.

HðtÞ ¼ P exp

�� exp

�Rm

Pðl� tÞ þ 1

��(6)

where H (t) is the cumulative hydrogen production at time t

(h). The kinetic constants of modified Gompertz equation P

(mL H2), Rm (cm�3 h�1) and l (h) represent hydrogen produc-

tion potential, maximum hydrogen production rate and

hydrogen production lag time, respectively. Parameters (P, Rm

and l) were estimated using the solver function in Matlab

(Curve fitting tool box ver. 1.1.7.) [24].

3. Result and discussions

3.1. Production of C. sorokiniana biomass in helicalbioreactor

Growth curve, carbohydrate, protein and chlorophyll content

of C. sorokiniana was shown in Fig. 1.

The growth rate of C. sorokiniana was 0.28 d�1 which was

slower than the growth rate found in tubular airlift reactor

(0.48 d�1) [17]. Biomass reached to stationary phase on 9th day

giving maximum dry cell weight of 2.9 kg m�3. As the cells

reached stationary state, the carbohydrate and protein con-

tent also became constant. Maximum carbohydrate and pro-

tein content observed was 145 g kg�1 and 140 g kg�1

respectively. Volumetric productivity of 334 kg m�3 d�1 was

observedusing this helical reactorwhichwas comparablewith

the productivities that have been reported for different reactor

configurations. Volumetric productivity of 400e700 kgm�3 d�1

Fig. 1 e Profiles of growth, carbohydrate, protein and chlorophyll content for C. sorokiniana in helical airlift photobioreactor.

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6160

and 370 kgm�3 d�1was reported in shallowponds and outdoor

tubular reactors, respectively [25]. With Arthrospira platensis,

an indoor tubular coiled bioreactor reported productivity

values of 510 kg m�3 d�1 [26]. Where light could be one of the

limiting factors, comparing different bioreactor configurations

was difficult as each configuration have different surface areas

and illumination. The helical tubular design has certain ad-

vantages like it has a large surface-to-volume ratio, easy con-

trol on mass transfer property of gases and occupies a small

ground area thus considered as a convenient setup for indoor

pilot plant operation [27].

Fig. 2 e Hydrogen production profile in different

pretreatment process of algal biomass. a) Autoclaved

biomass raw algal biomass. b) Sonicated biomass. c)

200 dm3 mL3 HCl treatment. d) 50 dm3 mL3 HCl treatment.

e) H2O2 treatment f) Raw algal biomass. Process conditions

are pH 6.5, algal biomass concentration 14 kg mL3 and

temperature of 60 �C. Fermentation time was of 24 h.

3.2. Effect of different pre-treatment methods onhydrogen production

Different pretreatment methods used in this study resulted in

improvedhydrogenproductionas compared tountreatedalgal

biomass. Acid and heat treatment converts complex carbo-

hydrates to simpler sugar. Under the condition of HCl-heat

pretreatment, algal starch was hydrolyzed to different soluble

sugars mainly consisting of dextrin, cellobiose and glucose at

various contents depending on the used HCl concentration.

Dextrin was only presented in the acid hydrolytic solution of

algal biomass treated by 50 dm3 m�3 HCl-heat [9]. With using

the 200 dm3 m�3 HCl concentration for algal biomass treat-

ment, the algal starch was almost converted to glucose with

the 100% of glucose conversion efficiency. Similarly

200 dm3 m�3 H2O2 treatment generates nascent oxygen that

breaks the glycosidic bondspresent in complex carbohydrates.

Fig. 2 shows effect of different pretreatments for improvement

of hydrogen production and also degree of saccharification.

The 200 dm3 m�3 HCl-heat treatment gave highest

hydrogen production of 1.33 m3 m�3 with reducing sugar

concentration of 9.6 kg m�3 among the other pretreatment

methods. The probable reason could be that this pretreatment

could have resulted inmaximumconversion of complex sugar

to hexose. Pre-treatment with physical methods also

improved hydrogen production. Pretreatment of algal biomass

by sonication gave better hydrogen production of 0.42 m3 m�3

than autoclaved biomass 0.100 m3 m�3. This could be due to

smaller-sized starch molecules generated by sonication was

easier and more suitable to convert to H2 than pure starch.

Moreover, reducing sugar concentration in sonicated algal

biomasswas 4.8 kgm�3 was higher as compared to autoclaved

biomass extract. Moreover, autoclaved biomass gave

improved hydrogen production than raw biomass. Probable

reason of this could be the solubilization of some of the nu-

trients present in algal biomass under autoclaving condition.

3.3. Study of effect of pretreatment on algal biomass

3.3.1. Microscopic study of microalgal biomass disruptionPre-treatment processes impart distortion of cellular

morphology and integrity. Microalgae consist of different

types of carbohydrate which are mostly stored in the cell wall

[28]. Therefore disruption of the microalgae cell walls is

necessary to release the entrapped carbohydrates for use as a

Fig. 3 e Microscopic observation (403) of effect of different pretreatment on hydrogen production. a) Raw algal biomass. b)

Pretreatment of algae by 50 dm3 mL3 HCl-heat c) Pretreatment of algae by 200 dm3 mL3 HCl-heat d) Pretreatment of algae by

H2O2 e) Pretreatment of algae by autoclave. f) Pretreatment of algae by sonication.

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6 161

carbon source during the fermentation process. In this study,

acid pre-treatment was used for microalgae cell wall disrup-

tion. Different concentrations of hydrochloric acid were

interacted with the microalgae cells and microscopic analysis

was performed to examine any morphological change after

different pre-treatment processes was shown in Fig. 3.

The image before pre-treatment shows that the cells were

intact, mostly spherical in shape and clumped together

(Fig. 3a). This suggests that the carbohydrates are intracellular

entrapped. On the other hand, the microscopic images of the

pre-treated microalgae with different concentration of HCl

shows ruptured cell wall evidenced by the broken cells. With

200 dm3 m�3 HCl treatment showed greater cell debris for-

mation than 50 dm3 m�3 HCl as shown in Fig. 3b and c. This

shows the greater efficacy of higher concentration hydro-

chloric acid for cell wall disruption. Similar observations was

Fig. 4 e XRD patterns of different pretreated algal biomass.

also made by Razhi et al., where algal biomass gave clumped

ruptured morphology when treated with HCl for ethanol pro-

duction [12]. The change from intact cells to broken cells after

the acid pre-treatment process confirms the disruption of the

microalgae cell wall to release entrapped carbohydrates.

Treatment with 200 dm3 m�3 H2O2 gave lesser cell debris for-

mation (Fig. 3d). Nascent oxygen released from H2O2 oxidizes

the chlorophyll content of the cell along with other proteins

whichwere required formaintaining the integrity of cell. Thus

transparent cells were observed with H2O2 treatment. Sur-

prisingly, lesser cell debris formedwith physical pretreatment

viz. autoclaving and sonication (Fig. 3e and f). A large number

of small and transparent particles can be observed.

3.3.2. XRD (X-ray diffraction) analysisFermentation has been greatly influenced by the crystallinity

of the biomass as it governs the ease of hydrolysis. The

biomass composition significantly influences the CrI value

[29]. For algal biomass, the CrI measures the relative amount

of crystalline starch in the total solid.

The X-ray diffraction pattern of the algal biomass treated

under above mentioned pretreatments was shown in Fig. 4,

from which the CrI was summarized in Table 1.

Table 1 e Crystallinity index of pretreatedalgal biomass.

Sample CrI (%)

Autoclaved biomass 38.69158

Sonication 33.11965

200 dm3 m�3 HCl 3.80818

200 dm3 m�3 H2O2 24.26778

Table 2 e Kinetic parameters with various pretreatmentmethods.

Differentsubstrates

l

(h)Rm

(m3 m�3 h�1)P

(m3)H2 yield

on VS (dm3 kg�1)

Starch 1.17 0.1242 1.583 833

200 dm3 m�3

HCl-heat treated

algal biomass

1.05 0.3353 2.155 958

50 dm3 m�3

HCl treated algal

biomass

1.26 0.121 1.520 760

H2O2 treated algal

biomass

3.2 0.1120 0.180 63

Autoclaved algal

biomass

2.2 0.540 0.650 338

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6162

As shown in Table 1, the CrI value of 38.69 for autoclaved

algal biomass was the highest suggesting that the higher

crystallinity of starch as compared to other samples. The CrI

value of 10.38 was lowest for 200 dm3 m�3 HCl treated

biomass. The probable reason for such observation could be

due to complete hydrolysis of starchy component of algal

Fig. 5 e a) Hydrogen production profile in different initial biom

Process conditions are pH 6.5 and temperature of 60 �C. Fermen

fermented pretreated algal biomass.

biomass under acidic conditions. Another pretreatment that

showed decrease in CrI values is H2O2 treatment. Sonication

didn’t influenced CrI value much. The CrI value of 33.11 was

observed for sonicated biomasswas comparable to autoclaved

biomass. This could suggest that sonication didn’t change the

crystallinity of the starchy components of the algae. The re-

sults from the above seemed to suggest that there was a

profound effect of different pretreatment methods on the

crystalline structure of the biomass. Similar observation was

also made by Xing et al., in which corn stock was treated

enzymatically to improve hydrogen production [30]. They

showed similar decrement of crystallinity index of corn stock

under enzymatic pretreatment.

3.4. Improvement of hydrogen production with differentpretreatment methods

The three kinetic parameters were studied (Table 2) along

with H2 yield and maximum specific H2 production rate.

Result shows that lag time was least in the case of

200 dm3 m�3 HCl treated biomass than other pretreatments.

ass concentration pretreated with 200 dm3 mL3 HCl (v/v).

tation time was of 24 h b) Volatile fatty acid profile of

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6 163

Moreover highest hydrogen production potential of

2.15m3m�3 and rateofhydrogenproductionof 0.34m3m�3 h�1

was also observed in 200 dm3 m�3 HCl treated biomass. Prob-

able reason for higher hydrogenproduction could be due to the

greater availability of fermentable sugar in this pretreatment.

When compared to pure starch, the lag time (l) for fermenta-

tion of 200 dm3 m�3 HCl treated biomass wasmore. This could

be due to presence of some inhibitory components like furfu-

rals that were generated during pretreatment processes.

Pretreatment with H2O2 gave very less amount of hydrogen.

This could be due to excessive oxidation of the fermentable

sugars present algal biomass. Autoclaved biomass gave better

Fig. 6 e a) Batch fermentation using 50 dm3 mL3 HCl-heat treat

LineweavereBurk plot. c) Product kinetics study using Leudekin

hydrogenproduction thanH2O2 pretreatment. On autoclaving,

sugarsget solubilized thus improveshydrogenproduction.The

kinetic parameters were calculated from Gompertz model by

curve fitting with R2 value greater than 0.99.

3.5. Effect of pretreated initial biomass concentration onhydrogen production

After optimizing the pretreatment processes, 200 dm3 m�3

HCl-heat treatment was found to be most suitable amongst

the other pretreatment methods. The effect of microalgae

loading on the acid pre-treatment process was also studied.

ed algal biomass. b) Growth kinetics study by using

g Piret model.

Table 3 e Comparative study on hydrogen yield by the thermophilic mixed culture.

Organism used Substrate Operationtemperature/pH

Volumetric hydrogenproduction on VS (dm3 kg�1)

References

Mixed culture Starchy wastewater 55 �C/7.0 365 [35]

Mixed culture Cassava starch 55 �C/7.0 249.3a [36]

Mixed culture Cellulosic wastewater 55 �C/5.5 102a [37]

Mixed culture Glycerol 60 �C/5.5 34.0 [38]

Mixed culture Wheat straw (untreated) 55 �C/8.0 24.5a [39]

Thermotoga neoplanita Algal biomass 60 �C/6.5 2.4b [7]

Mixed culture Algal biomass 60 �C/6.5 172 [40]

Mixed culture Algal biomass 60 �C/6.5 958

2.51bThis study

a dm3 kg�1 substrate.b mol mol�1 of glucose.

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6164

Thus with 200 dm3 m�3 HCl-heat treatment, different

initial biomass concentrations were used for thermophilic

biohydrogen production (Fig. 5a). Initial algal biomass was

varied from 8 kg m�3 to 16 kg m�3. At biomass concentration

of 14 kg m�3 biomass concentration gave highest cumulative

hydrogen production of 1.50 m3 m�3. Fermentation was car-

ried out for 24 h at 60 �C. A decrease in hydrogen production

beyond 14 kgm�3 was observed. Highermicroalgae loading up

to 16 kg m�3 resulted in lower hydrogen production. Although

higher concentrations of biomass was expected to release

higher amounts of fermentable sugars during the pre-

treatment process, it has been reported that higher biomass

concentrations reduces hydrogen production levels due to

substrate inhibition, metabolic burden and accumulation of

inhibitory substances like furfurals [31]. The volatile fatty

acids (VFAs) production is always associated with H2 pro-

duction [32]. In this experiment, the acetate and butyrate

accounted for 70e80% of total VFA concentration and rest was

ethanol (Fig. 5b). More or less same amounts of all three

components of VFAs were produced during H2 production

with acid pretreated cornstalk waste [33]. With increase in

initial biomass concentration butyrate and acetate concen-

tration increased. It has been reported that butyrate to acetate

(B to A) ratios are directly proportional to H2 yields [34].

Highest butyrate to acetate ratio was observed during

fermentation of 14 kg m�3 pretreated initial biomass.

3.6. Batch fermentation under optimized processparameters

Fermentation was performed with 200 dm3 m�3 HCl�heat

pretreated algal biomass having initial biomass concentration

of 14 kg m�3 for 24 h.

Cumulative hydrogen production, substrate consumption,

pH and dry cell weight profile are shown in Fig. 6a. Hydrogen

production ceased after 20th hour of fermentation. Total cu-

mulative hydrogen production and hydrogen yield were

1.93 m3 m�3 and 2.68 � 0.2 mol mol�1 of hexose, respectively.

Substrate conversion efficiency of 62% was observed.

Maximum dry cell weight and final pH at the end of the

fermentationwere found to be 1.9 kgm�3 and 4.2, respectively.

The pH of the system may be decreased due to production of

volatile fattyacidsas fermentationof reducible sugaroccurred.

Table 3 shows a comparative study on hydrogen produc-

tion with algal biomass.

Hydrogen yield obtained in this study was 958 dm3 kg�1 of

volatile suspended solid which was higher than the report of

365 dm3 kg�1 of volatile suspended solid with starchy waste-

water at thermophilic condition [41]. In another study,

hydrogen yield of 9 mol kg�1 COD reduced was observed with

pretreated algal biomass as substrate using Enterobacter cloacae

IIT-BT 08 under mesophilic conditions [10]. Thus algal digest

could be a promising substrate for hydrogen production.

3.7. Kinetic analysis

Plotting m�1 against S�1, a straight line was obtained with an

intercept of m�1max and a slope of Ks m

�1max. This plot was shown in

Fig. 6b, from which mmax and Ks were estimated was 0.44 h�1

and 120 g m�3, respectively. The regression line had a corre-

lation coefficient of 0.96, suggesting the applicability of Eq. (3).

The Ks value represents the substrate concentration required

to achieve 50% of the maximum specific growth rate.

The hydrogen production kinetics was shown in Fig. 6c for

hydrogen. The high correlation coefficients (R2 ¼ 0.98) indicate

that the LuedekingePiret model could properly describe the

relationship between biomass and product in the anaerobic

hydrogen production process. The estimated values of a and b

were 17 (mmol g�1 dry cell weight) and 0.726, respectively

indicating the dependence of product formation and biomass

concentration. These values could be used to predict the

productivity in the bioreactor design. Since b values were near

to zero the formation of hydrogen in the anaerobic hydrogen

production was growth associated [42].

4. Conclusion

Algal biomass of C. sorokiniana was used in dark fermentation

for H2 production using thermophilic mixed consortia.

Different physical and chemical pretreatments methods were

applied to rupture the algal cell walls, hydrolyze intracellular

complex carbohydrates and to facilitate the release of

fermentable sugars. HCl (200 dm3 m�3)-heat was found the

most suitable pretreatment method. Hydrogen production

was studied in batch in double jacketed bioreactor with opti-

mized concentration of pretreated algal biomass. Microscopic

and XRD analysis confirmed the rupture of cell walls neces-

sary for the release of fermentable sugars. Monod, Luedeking-

Piret equations were used to determine the hydrogen

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6 165

production kinetics. Hydrogen production was found to be

growth associated. Modified Gompertz equation was found to

reasonably fit with the experimental hydrogen production

data.

Acknowledgment

The authors gratefully acknowledge Council of Scientific and

Industrial Research (CSIR), Govt. of India for fellowship; IIT

Kharagpur and Ministry of New and Renewable Energy

(MNRE), Govt. of India for the financial support.

r e f e r e n c e s

[1] Nayak BK, Pandit S, Das D. Biohydrogen chapter 15. In:Kennes C, Veiga MC, editors. Air pollution prevention andcontrol: bioreactors and bioenergy. Chichester: John Wiley;2013. pp. 345e82.

[2] Melis A, Happe T. Hydrogen production. Green algae as asource of energy. Plant Physiol 2001;127(3):740e8.

[3] Levin DB, Pitt L, Love M. Biohydrogen production: prospectsand limitations to practical application. Int J Hydrogen Energ2004;29(2):173e85.

[4] Roy S, Vishnuvardhan M, Das D. Improvement of hydrogenproduction by newly isolated Thermoanaerobacteriumthermosaccharolyticum IIT BT-ST1. Int J Hydrogen Energ.http://dx.doi.org/10.1016/j.ijhydene.2013.06.128; 2013.

[5] Kosaric N, Velikonja J. Liquid and gaseous fuels frombiotechnology: challenge and opportunities. FEMS MicrobiolRev 1995;16(2):111e42.

[6] Markou G, Angelidaki I, Georgakakis D. Carbohydrate-enriched cyanobacterial biomass as feedstock for bio-methane production through anaerobic digestion. Fuel2013;111:872e9.

[7] Sawayama S, Inoue S, Dote Y, Yokoyama S-Y. CO2 fixationand oil production through microalga. Energy ConversManag 1995;36(6):729e31.

[8] Melis A, Zhang L, Forestier M, Ghirardi ML, Siebert M.Sustained photobiological hydrogen gas production uponreversible inactivation of oxygen evolution in the green algaChlamydomonas reinhardtii. Plant Physiol 2000;122(1):127e35.

[9] Nguyen TAD, Kim KR, Nguyen MT, Kim MS, Kim D, Sim SJ.Enhancement of fermentative hydrogen production fromgreen algal biomass of Thermotoga neapolitana by variouspretreatment methods. Int J Hydrogen Energy2010;35(23):13035e40.

[10] Kumar K, Roy S, Das D. Continuous mode of carbon dioxidesequestration by C. sorokiniana and subsequent use of itsbiomass for hydrogen production by E. cloacae IIT-BT 08.Bioresour Technol 2013;145:116e22.

[11] Kawaguchi H, Hashimoto K, Hirata K, Miyamoto K. H2production from algal biomass by a mixed culture ofRhodobium marinum A-501 and Lactobacillus amylovorus. JBiosci Bioeng 2001;91(3):277e82.

[12] Harun R, Danquah MK. Influence of acid pre-treatment onmicroalgal biomass for bioethanol production. ProcessBiochem 2011;46(1):304e9.

[13] Nguyen MT, Choi SP, Lee J, Lee JH, Sim SJ. Hydrothermal acidpretreatment of Chlamydomonas reinhardtii biomass forethanol production. J Microbiol Biotechnol 2009;19(2):161e6.

[14] Kim MS, Baek JS, Yun YS, Jun Sim S, Park S, Kim SC.Hydrogen production from Chlamydomonas reinhardtiibiomass using a two-step conversion process: anaerobic

conversion and photosynthetic fermentation. Int J HydrogenEnergy 2006;31(6):812e6.

[15] Zhao X, Chen J, Chen F, Wang X, Zhu Q, Ao Q. Surfacecharacterization of corn stalk superfine powder studied byFTIR and XRD. Colloids Surf B Biointerfaces2013;104:207e12.

[16] Pan CM, Ma HC, Fan YT, Hou HW. Bioaugmented cellulosichydrogen production from cornstalk by integrating diluteacid-enzyme hydrolysis and dark fermentation. Int JHydrogen Energy 2011;36(8):4852e62.

[17] Kumar K, Das D. Growth characteristics of Chlorellasorokiniana in airlift and bubble column photobioreactors.Bioresour Technol 2012;116(1):307e13.

[18] Roy S, Ghosh S, Das D. Improvement of hydrogen productionwith thermophilic mixed culture from rice spent wash ofdistillery industry. Int J Hydrogen Energ2012;37(21):15867e74.

[19] Loewus FA. Improvement in anthrone method fordetermination of carbohydrates. Anal Chem1952;24(1):219e22.

[20] APHA/AWWA/WEF. In: Eaton AD, Clesceri LS, Rice EW,Greenberg AE, editors. Standard methods for theexamination of water and wastewater. 21st ed. Washington,DC: American Public Health Association/American WaterWorks Association/Water Environment Federation; 2005.

[21] Segal L, Creely JJ, Martin Jr AE, Conard CM. An empiricalmethod for estimating the degree of crystallinity of nativecellulose using the X-ray diffraction. Textile Res J1959;29:786e94.

[22] Tserki V, Matzinos P, Kokkou S, Panayiotou C. Novelbiodegradable composites based on treated lignocellulosicwaste flour as filler. Part I. surface chemical modification andcharacterization of waste flour. Compos Part A Appl Sci2005;36(7):965e74.

[23] Shuler ML, Kargi F. Bioprocess engineering: basic concepts.2nd ed. Prentice Hall; 2001.

[24] Khanna N, Kotay SM, Gilbert JJ, Das D. Improvement ofbiohydrogen production by Enterobacter cloacae IIT-BT 08under regulated pH. J Biotechnol 2011;152(1):9e15.

[25] Richmond A. Large scale microalgal culture and applications.Prog Phycol Res 1990;7:1e62.

[26] Watanabe Y, de la Noue J, Hall DO. Photosyntheticperformance of a helical tubular photobioreactorincorporating the cyanobacterium Spirulina platensis.Biotechnol Bioeng 1995;47(2):261e9.

[27] Scragg AH, Illman AM, Carden A, Shales SW. Growth ofmicroalgae with increased calorific values in a tubularbioreactor. Biomass Bioenergy 2002;23(1):67e73.

[28] Hu Q, Kurano N, Kawachi M, Iwasaki I, Miyachi S. Ultrahigh-cell-density culture of a marine green alga Chlorococcumlittorale in a flat-plate photobioreactor. Appl MicrobiolBiotechnol 1998;49(6):655e62.

[29] Chu Y, Wei Y, Yuan X, Shi X. Bioconversion of wheat stalk tohydrogen by dark fermentation: effect of different mixedmicroflora on hydrogen yield and cellulose solubilisation.Bioresour Technol 2011;102(4):3805e9.

[30] Xing Y, Fan SQ, Zhang JN, Fan YT, Hou HW. Enhanced bio-hydrogen production from corn stalk by anaerobicfermentation using response surface methodology. Int JHydrogen Energy 2011;36(20):12770e9.

[31] Cao GL, Ren NQ, Wang AJ, Guo WQ, Xu JF, Liu BF. Effect oflignocellulose-derived inhibitors on growth and hydrogenproduction by Thermoanaerobacterium thermosaccharolyticumW16. Int J Hydrogen Energy 2010;35(24):13475e80.

[32] Prakasham RS, Sathish T, Brahmaiah P, Subba Rao C,Sreenivas Rao R, Hobbs PJ. Biohydrogen production fromrenewable agri-waste blend: optimization using mixerdesign. Int J Hydrogen Energy 2009;34(15):6143e8.

b i om a s s a n d b i o e n e r g y 6 1 ( 2 0 1 4 ) 1 5 7e1 6 6166

[33] Zhang ML, Fan YT, Xing Y, Pan CM, Zhang GS, Lay JJ.Enhanced biohydrogen production from cornstalk wasteswith acidification pretreatment by mixed anaerobic cultures.Biomass Bioenergy 2007;31(4):250e4.

[34] Kim MS, Oh YK, Yun YS, Lee DY. Fermentative hydrogenproduction from anaerobic bacteria using a membranebioreactor. In: Proc. 16th world hydrog. energy conf. Lyon Fr,vol. 50; 2006.

[35] Lay JJ. Modeling and optimization of anaerobic digestedsludge converting starch to hydrogen. Biotechnol Bioeng2000;68(3):269e78.

[36] O Thong S, Hniman A, Prasertsan P, Imai T. Biohydrogenproduction from cassava starch processing wastewater bythermophilic mixed cultures. Int J Hydrogen Energy2011;36(5):3409e16.

[37] Liu H, Zhang T, Fang HHP. Thermophilic H2 production froma cellulose-containing wastewater. Biotechnol Lett2003;25(4):365e9.

[38] Ito T, Nakashimada Y, Senba K, Matsui T, Nishio N. Hydrogenand ethanol production from glycerol-containing wastes

discharged after biodiesel manufacturing process. J BiosciBioeng 2005;100(3):260e5.

[39] Kongjan P, Angelidaki I. Extreme thermophilic biohydrogenproduction from wheat straw hydrolysate using mixedculture fermentation: effect of reactor configuration.Bioresour Technol 2010;101(20):7789e96.

[40] Carver SM, Hulatt CJ, Thomas DN, Tuovinen OH.Thermophilic, anaerobic co-digestion of microalgal biomassand cellulose for H2 production. Biodegradation2011;22(4):805e14.

[41] Yang Y, Tsukahara K, Sawayama S. Biodegradation andmethane production from glycerol-containing syntheticwastes with fixed-bed bioreactor under mesophilic andthermophilic anaerobic conditions. Process Biochem2008;43(4):362e7.

[42] Luedeking R, Piret EL. A kinetic study of the lactic acidfermentation. Batch process at controlled pH. Reprintedfrom Journal of Biochemical and Microbiological TechnologyEngineering Vol. I, No. 4. Pages 393-412 (1959). BiotechnolBioeng 2000;67(6):636e44.