Ultrasonics Sonochemistry 26 (2015) 249–256
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultson
Mechanistic insight into ultrasound induced enhancement of simultaneous saccharification and fermentation of Parthenium hysterophorus for ethanol production Shuchi Singh a, Mayank Agarwal b,1, Shyamali Sarma a,1, Arun Goyal a,c,⇑, Vijayanand S. Moholkar a,b,⇑ a b c
Centre for Energy, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India Department of Biotechnology, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India
a r t i c l e
i n f o
Article history: Received 17 November 2014 Received in revised form 21 January 2015 Accepted 25 February 2015 Available online 17 March 2015 Keywords: Ultrasound Microturbulence Fermentation Bioethanol Parthenium hysterophorus Mathematical modeling
a b s t r a c t This paper presents investigations into mechanism of ultrasound assisted bioethanol synthesis using Parthenium hysterophorus biomass through simultaneous saccharification and fermentation (SSF) mode. Approach of coupling experimental results to mathematical model for SSF using Genetic Algorithm based optimization has been adopted. Comparison of model parameters for experiments with mechanical shaking and sonication (10% duty cycle) give an interesting mechanistic account of influence of ultrasound on SSF system. A 4-fold rise in ethanol and cell mass productivity is seen with ultrasound. The analysis reveals following facets of influence of ultrasound on SSF: increase in Monod constant for glucose for cell growth, maximal specific growth rate and inhibition constant of cell growth by glucose and reduction in specific cell death rate. Values of inhibition constant of cell growth by ethanol (K3E), and constants for growth associated (a) and non-growth associated (b) ethanol production remained unaltered with sonication. Beneficial effects of ultrasound are attributed to enhanced cellulose hydrolysis, enhanced trans-membrane transport of substrate and products as well as dilution of the toxic substances due to micro-convection induced by ultrasound. Intrinsic physiological functioning of cells remained unaffected by ultrasound as indicated by unaltered values of K3E, a and b. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Bioethanol production from lignocellulosic biomass has been a highly active research area for past several years, as ethanol has shown high promise as an alternate liquid transportation fuel as well as oxygenate blend for gasoline. The conventional process for bioethanol production has two steps, viz. pretreatment and
Abbreviations: CMCase, carboxymethylcellulase; GA, Genetic Algorithm; GC, gas chromatography; HPLC, high performance liquid chromatography; MTCC, microbial type culture collection; RI, refractive index; SHF, separate hydrolysis and fermentation; SSF, simultaneous saccharification and fermentation; TRS, total reducing sugar. ⇑ Corresponding authors at: Centre for Energy, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India. Tel.: +91 361 258 2208; fax: +91 361 258 2249 (A. Goyal). Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India. Tel.: +91 361 258 2251; fax: +91 361 258 2291 (V.S. Moholkar). E-mail addresses: [email protected] (A. Goyal), [email protected] (V.S. Moholkar). 1 Authors contributed equally. http://dx.doi.org/10.1016/j.ultsonch.2015.02.011 1350-4177/Ó 2015 Elsevier B.V. All rights reserved.
acid/enzymatic hydrolysis of lignocellulosic biomass followed by fermentation of the acid and/or enzymatic hydrolyzate. The cost of production of bioethanol is a major function of cost of fermentation substrate as well as the operating cost of the process. Lignocellulosic biomass available abundantly in the form of agroresidues, forest-residues, and waste biomass (weed/grass) forms a potential low-cost feedstock for bioethanol. Some typical examples of waste biomass whose carbohydrate moieties have been used for bioethanol production are Saccharum spontaneum [1], Lantana camara [2] and Prosopis juliflora [3]. In order to intensify the bioethanol productivity while reducing the cost of production, the process of simultaneous saccharification and fermentation (SSF) has also been extensively investigated. This process has distinct advantages of milder operating conditions, and requirement of a single fermentor vessel that combines the two steps of hydrolysis and fermentation mentioned above. In this process, hexose sugars released from enzymatic hydrolysis of cellulose in the biomass are simultaneously consumed by fermenting microorganisms. The enzymatic hydrolysis of cellulose itself is a two-step process, in which first cellulase hydrolyzes the cellulose into
250
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
Nomenclature Notation a, b constants for ethanol formation, growth associated and non-growth associated, respectively at available surface area for cellulose (B), (C), (E), (G), (X) concentrations of cellobiose, cellulose, ethanol, glucose and cell mass, respectively (E1)t total concentration of cellulase in the solution (E2) concentration of b-glucosidase in the solution kd specific rate of cell death k1, k2 specific rates of cellulose and cellobiose hydrolysis, respectively KI constant of cell growth inhibition by glucose Km Michaelis constant of b-glucosidase for cellobiose K3, K4 Monod constants of glucose for cell growth and ethanol synthesis, respectively
cellobiose (dimer hexose sugar units), which are later split into monomeric hexose sugar units by cellobiase (or b-glucosidase). SSF process reduces the inhibitory effect of substrate (sugar) concentration on enzymes and also the probability of contamination by undesired invasive microorganisms [4]. These features increase yield as well as kinetics of the saccharification as well as fermentation as compared to the conventional two-step process. More recently, another technique of ultrasound irradiation or sonication of the fermentation broth for intensification of bioethanol fermentation has been attempted. Ultrasound is a well known technique for intensification of diverse physical and chemical processes [5–10]. Ultrasound manifests its effect on the reaction system through phenomenon of cavitation, which is nucleation, growth, oscillation and implosive transient collapse of tiny gas or vapor bubbles, which is driven by pressure variation generated in the medium during passage of ultrasound wave. Both ultrasound and cavitation render several physical and chemical effects on the reactions system, which are beneficial in enhancing the kinetics of the system. The most peculiar feature of energy introduction into the medium via ultrasound and cavitation is that implosive collapse of cavitation bubbles creates intense energy concentration on an extremely small spatial and temporal scale. The main physical effect of ultrasound and cavitation is generation of intense micro-turbulence in the medium that gives very effective micromixing, which eliminates mass transfer limitations. The chemical effect of transient cavitation is generation of highly reactive radicals and other smaller species through dissociation of vapor entrapped in the bubble at the moment of transient collapse. Literature on application of ultrasound during bioethanol synthesis through SSF process is rather limited. Wood et al. [11] have reported bioethanol production using ultrasound (36 kHz, 150 W) assisted SSF process. The substrate used was waste paper and microbial strain of Klebsiella oxytoca was employed. Bioethanol yield was found to increase by 20% with sonication. OforiBoateng and Lee [12] have investigated bioethanol production using SSF process from oil palm fronds as substrate and Saccharomyces cerevisiae as the microbial strain. With ultrasound of 40 kHz frequency and 200 W intensity, 4-fold increment in bioethanol yield was observed within 5 h. In order to effectively utilize the potential of ultrasound on intensification of the SSF process for bioethanol production, it is essential to understand the basic underlying physical mechanism. This would essentially mean identifying the links between physics of ultrasound and cavitation and the biochemistry of fermentation. In this paper, we have addressed this important issue with the
K1B, K2B Inhibition constants of cellulase and b-glucosidase by cellobiose, respectively t time K1G, K2G inhibition constants of cellulase and b-glucosidase by glucose, respectively K1E, K2E, K3E Inhibition constants of cellulase, b-glucosidase and cell growth by ethanol, respectively m specific rate of substrate consumption for maintenance requirements lm maximal specific growth rate YX/G average yield coefficient of cell mass on substrate (glucose)
approach of coupling experimental results to the fermentation model of [13], which comprises of 5 ordinary differential equations, viz. one each for cellulose, cellobiose, glucose, microbial cell concentration and ethanol. This model takes into account the essential physiology of the SSF process. A major practical limitation of implementation of this model is difficulty in monitoring of the concentration of cellulose (which occurs in solid phase) and also the unstable intermediate of hexose-dimer cellobiose (which is rapidly decomposed into monomeric glucose) during fermentation [14]. Despite this limitation, fitting of the experimental data of microbial cell concentration and ethanol to their respective differential model equations reveals important mechanistic account of the influence of ultrasound on the SSF process. In our experiments, we have used waste biomass of Parthenium hysterophorus as substrate with S. cerevisiae as the microbial strain.
2. Materials, methods and mathematical model 2.1. Chemicals and reagents All components of fermentation medium were procured from HiMedia Pvt. Ltd., India. Glucose (99.5% purity, standard for HPLC and reducing sugar estimation) was procured from Sigma Aldrich, USA. Ethanol (99.5% purity) was procured from Tedia Chemicals, USA. All other chemicals were procured from Fischer Scientific, India.
2.2. P. hysterophorus biomass P. hysterophorus biomass was collected from the campus of our institute. Biomass was chopped (5 cm), washed with water, dried at 60 ± 3 °C for 24 h and ground to a particle size of 1 mm. Powdered biomass was pretreated with 1% (v/v) H2SO4 + 30 min autoclaving [15], and the solid residue was further delignified by ultrasound assisted alkaline treatment [16].
2.3. Source of enzymes Carboxymethylcellulase (CMCase) (1.0 U/mg, 1.7 mg/mL) was produced from Bacillus amyloliquefaciens SS35 [17–19] and b-glucosidase (250 U/mL) (Novozyme 188) was procured from Sigma Aldrich, USA.
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
2.4. Microorganism and fermentation conditions S. cerevisiae MTCC 170 was used for ethanol fermentation. Microbial culture was procured from Microbial Type Culture Collection (MTCC), Institute of Microbial Technology (IMTECH), Chandigarh, India and was maintained in YEPD medium containing yeast extract (10 g/L), peptone (20 g/L) and dextrose (20 g/L). 2.5. Simultaneous saccharification and fermentation (SSF): control experiments
standard. Dry cell mass for yeast cells was determined by using an indirect method by plotting a calibration curve using intracellular protein content (g/L) and dry cell mass (g/L) of the yeast cells [27]. Viability of S. cerevisiae MTCC 170 cells in the samples before and after sonication was determined using methylene blue staining method [28]. 0.1% (w/v) methylene blue solution was mixed with yeast broth in a ratio of 1:1 and incubated for 5 min. Cells were counted on hemocytometer at 40 magnification. The cell viability was calculated using the formula: Viability ð%Þ ¼
Simultaneous saccharification and fermentation (SSF) of pretreated and delignified P. hysterophorus was carried out by using the enzymes CMCase and b-glucosidase, and S. cerevisiae MTCC 170 culture. 3.88% w/v pretreated and delignified P. hysterophorus biomass was added to fermentation medium, which consisted of yeast extract (10 g/L), KH2PO4 (2 g/L), (NH4)2SO4 (1 g/L) and MgSO47H2O (1 g/L) (pH 5.0) [20]. This mixture was supplemented with CMCase and b-glucosidase with concentrations of 600 U/g and 50 FPU/g of delignified biomass, respectively. SSF was carried out in a 150 mL Erlenmeyer flask with working volume of 25 mL. The medium was inoculated with 10% (v/v) of previously grown culture of S. cerevisiae MTCC 170 and incubated at 30 °C and 150 rpm in an incubator shaker (Scigenics Biotech, Model: Orbitek). 2.6. Simultaneous saccharification and fermentation (SSF): test experiments Ultrasound-assisted fermentation (test experiment) was carried out in an ultrasound bath (Make: Elma, Germany; Model: Transonic T-460, 2L) operating at a frequency of 35 kHz and power rating of 35 W. The actual power input to the liquid medium in the bath (i.e. water) was determined calorimetrically as 18.58 W [21,22]. This power is transmitted through the transducer attached to bottom of the bath. The diameter of this transducer was 4 cm. Therefore, the power dissipation to sonication medium is 1.48 W/ cm2 and the corresponding acoustic power amplitude was 150 kPa. The ultrasound was applied in a duty cycle of 10% (i.e. 1 min of sonication and 9 min of mechanical shaking in every 10 min of fermentation). The total working volume of the bath was 2 L (dimensions: 25 cm 15 cm 10 cm). The bath was filled with water and the flask containing fermentation mixture was placed at the center of the bath. The flask was immersed to about 50% of its height in the water and the position of the Erlenmeyer flask was carefully maintained same in all experiments in view of spatial variation of the acoustic intensity [23,24]. The temperature of the water in the bath was maintained at 30o ± 2 °C by replacement of small portions of initial water at regular intervals. 200 lL aliquots of the fermentation broth were withdrawn intermittently in both control and test experiments to determine total reducing sugar, ethanol and cell mass concentration in fermentation broth. The total fermentation period for control and test experiments was set at 78 and 35 h, respectively, based on the concentration of residual sugar in the samples of fermentation broth. The fermentation was stopped when the sugar concentration became constant (< 2% difference) in the consecutive samples withdrawn from fermentation broth. Both test and control experiments were repeated thrice to assess the reproducibility of the results. 2.7. Determination of total reducing sugar, cell mass and cell viability Aliquots withdrawn from fermentation broth were treated using the method of Nelson [25] and Somogyi [26] for determination of the residual reducing sugar, using D-glucose as
251
Number of live ðunstainedÞ cells Number of live ðunstainedÞ cells þ Number of dead ðstainedÞ cells 100
2.8. Analytical methods Presence of glucose in the aliquots of fermentation broth was confirmed with HPLC analysis (Perkin-Elmer, Series 200, with a refractive index detector) using HiPlex-H column (300 mm 5 lm 4.6 mm, Varian). HPLC grade water (deionized water, Milli Q) was used as the mobile phase at a flow rate of 0.4 mL/min with glucose (99.5% purity) as standard in sugar analysis. Ethanol content of fermentation samples was determined by Gas Chromatography (Varian, CP 3800) using a CP Wax 52 CB capillary column (250 mm 0.25 mm 0.39 mm, Varian) with ethanol (99.5% purity) as the standard. The oven temperature was programmed from 45 °C to 100 °C with an increment of 3 °C/min and 100 °C to 200 °C with an increment of 5 °C/min. The injector and detector temperatures were kept at 230 °C and 250 °C, respectively. Nitrogen gas was used as a carrier at a flow rate of 2 mL/min. 2.9. Mathematical model We have used ethanol fermentation model for the SSF mode developed by Philippidis et al. [13] in this study. This model is essentially based on the HCH-1 model of enzymatic hydrolysis of cellulose reported by Holtzapple et al. [29]. This model makes several assumptions as follows: (1) No distinction between two components of cellulase, viz. endoglucanase and exoglucanase; (2) exogenous addition of cellulose hydrolysing enzymes, viz. cellulase and b-glucosidase to the SSF system; (3) cellulose is hydrolyzed to cellobiose by cellulase with negligible formation of glucose; (4) pH of the fermentation broth remains practically constant during SSF process; (5) major metabolic products of SSF process are ethanol and CO2, (6) the carbon source for metabolism (glucose) is derived from cellulose, while the growth medium provides sufficient excess of all other nutrients, and (7) the activities of cellulase and b-glucosidase enzymes are assumed to remain constant all through the process. The model of Philippidis et al. [13] proposes that cellulose to ethanol conversion process in SSF mode comprises of following steps: (1) cellulase diffusion towards cellulose, (2) adsorption of cellulase on surface of cellulose, (3) hydrolysis of cellulose to cellobiose, (4) diffusion of cellobiose into aqueous phase, (5) hydrolysis of cellobiose to glucose (catalyzed by b-glucosidase), (6) glucose diffusion towards cells, (7) glucose uptake by cells, (8) glucose to ethanol conversion and (9) ethanol secretion into aqueous phase. The model comprises of five ordinary differential equations for cellulose, cellobiose, glucose, cell mass and ethanol as follows: Cellulose:
dðCÞ k1 at ðE1 Þt K 1E i ¼ h ðBÞ ðGÞ K 1E þ ðEÞ dt K e þ ðE1 Þt 1 þ K 1B þ K 1G
252
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
Cellobiose:
fermentation with the model is depicted in the flowchart given in Supplementary material provided with the manuscript.
dðBÞ k1 at ðE1 Þt K 1E i ¼ 1:06 h ðBÞ ðGÞ K 1E þ ðEÞ dt K e þ ðE1 Þt 1 þ K 1B þ K 1G
k2 ðE2 ÞðBÞ K 2E h i h i ðBÞ K 2E þ ðEÞ K m 1 þ KðGÞ þ ðBÞ 1 þ K 2B 2G
Cell mass:
dðXÞ lm ðGÞ K 3E ¼ ðXÞ kd ðXÞ 2 dt K 3 þ ðGÞ þ ðGÞ K 3E þ ðEÞ KI
Glucose:
dðGÞ k2 ðE2 ÞðBÞ K 2E i h i ¼ 1:05 h ðGÞ ðBÞ K 2E þ ðEÞ dt K m 1 þ K 2G þ ðBÞ 1 þ K 2B 1 dðXÞ þ mðXÞ Y X=G dt Ethanol:
dðEÞ dðXÞ ðGÞ ¼ a þ bðXÞ dt dt K 4 þ ðGÞ The following changes have been made in original model equations of Philippidis et al. [13] in view of experimental conditions of present experiments. In view of acid + autoclaved biomass that results in high reduction of cellulose crystallinity and increase in porosity, the value of substrate reactivity coefficient (/) is set equal to 1. The values of two parameters, viz. adsorption constants for cellulase and b-glucosidase onto lignin are set to zero in view of use of delignified biomass as the substrate in SSF process. The concentration of lignin in the biomass is also assumed to zero. Zero value has been assigned to parameter K4 (Monod constant of glucose for ethanol synthesis) as per the results of Philippidis et al. [13] which demonstrated that ethanol formation was not directly dependent on the glucose concentration. As we have stated earlier, due to limitations of experimental facilities used in this study, we have not been able to monitor the concentration of cellulose (existing in solid phase) and the intermediate dimeric cellobiose during the fermentation process. However, we have monitored the concentration of glucose in the fermentation broth. Therefore, while fitting our experimental data, we have used only the equations for cell mass and ethanol. These equations have a total of 7 parameters, viz. K3, KI, K3E, kd, lm, a and b. The glucose concentration profile obtained during SSF experiments was fitted to a polynomial expression, and this expression was used to determine instantaneous glucose concentration in the broth at a time, during the numerical integration of the differential equations for cell mass and ethanol. Fitting of these equations to the experimental profiles of cell mass and ethanol would yield the numerical values of these model parameters, which represent kinetic and physiological facets of the SSF system. Variations in the values of these model parameters for the test and control experiments give a mechanistic account of the influence of ultrasound on the SSF system. For determination of the optimum values of the model parameters – so as to match the experimental and model predicted trends, the main model of two simultaneous ordinary differential equations was solved using Runge–Kutta 4th order method, and the solution was coupled to Genetic Algorithm (GA) optimization module of MATLAB. Optimization of the model parameters was done by calculating root mean square (RMS) error between experimental and model results. GA module minimizes this error, yielding optimized values of the model parameters. The overall algorithm of simulations of
2.10. Physical and chemical effects of ultrasound and transient cavitation The physical and chemical effects of ultrasound and cavitation are responsible for the acceleration of kinetics of processes. The physical effect is generation of strong micro-convection in the medium due to three phenomena: (1) oscillatory motion of fluid elements called ‘‘microstreaming’’ due to propagation of ultrasound in the form of compression/rarefaction cycles, (2) acoustic or shock waves emitted during transient collapse of cavitation bubbles, and (3) oscillatory motion of liquid called ‘‘micro-turbulence’’ induced by volumetric oscillation of cavitation bubbles. The chemical effect of cavitation is generation of highly reactive radicals due to thermal dissociation of solvent vapor molecules trapped in the cavitation bubble at the instance of transient collapse. The magnitudes of both physical and chemical effects can be quantitatively estimated using simulations of cavitation bubble dynamics. The physical and chemical effects of cavitation bubbles, as estimated by simulations of cavitation bubble dynamics using diffusion limited model of Toegel et al. [30] have been given in Table 1. Conditions used for simulations have been listed in the footnote of the table. Diffusion limited bubble dynamics model has been explained in greater details in several of our earlier papers [6,8,21]. It could be seen from the results presented in Table 1 that mild shock waves (with amplitude 30–70 kPa) emitted by the cavitation bubble can help enhance enzymatic hydrolysis due to greater access of enzymes to biomass and faster transport of sugars. The OH radicals generated by the cavitation bubbles can also boost enzymatic hydrolysis. The shock waves also cause rapid movement of the microbial cells. The kinetic energy gained from this movement can help enhance intracellular metabolism. As far as the convection generated by ultrasound is concerned, the magnitude of the micro-streaming velocity is given as, u ¼ P A =qC, where PA is pressure amplitude of ultrasound wave, q is density of medium and C is sonic velocity in medium. For typical values of PA = 150 kPa (or 1.5 bar), q = 1000 kg/m3 and C = 1500 m/s (for water as medium), u = 0.1 m/s. Relating this velocity to amplitude of oscillation as u ¼ 2pfa, for f (frequency) = 20 kHz, we get a (amplitude) = 0.8 l. In the context of present study, the spatial scale of micro-streaming is much smaller than the size of the S. cerevisiae cells (typically 5–10 l) through which diffusion of substrate (glucose) and product (ethanol) occurs. Hence, the convection generated by ultrasound is more effective than the convectional mechanical agitation in enhancing the transport on bulk as well as cellular level. 3. Results and discussion 3.1. Simultaneous saccharification and fermentation (SSF): with and without sonication Figs. 1 and 2 show the time profiles of glucose, ethanol and cell mass in control and test experiments, respectively. The summary of the results of the SSF process under control and test conditions is given in Table 2. The values of the physiological parameters in the differential equations for cell mass and ethanol, obtained with GA optimization are listed in Table 3. The salient features of the ethanol synthesis by SSF process in control (mechanical shaking) and test (sonication + mechanical shaking) conditions, which could be identified from results given in Figs. 1 and 2 and Tables 2 and 3 are as follows:
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
253
Table 1 Summary of cavitation bubble dynamics simulations indicating physical and chemical effects induced by transient cavitation. Species
Parameters for simulations Water Air bubble Air bubble Ro = 5 lm Ro = 10 lm Conditions at the first collapse of the bubble Tmax = 3258 K Tmax = 2304 K Pmax = 88.4 MPa Pmax = 384 MPa Vturb = 0.03 mm/s Vturb = 0.05 mm/s PAW = 72 kPa PAW = 31.6 kPa Equilibrium composition of bubble at transient collapse
N2 O2 O O3 H H2 NO NO2 N2O OH H2O HO2 H2O2 HNO HNO2
7.1952E01 1.6608E01 1.6723E03 6.4788E06 7.1808E05 1.8629E04 5.6173E02 1.3272E03 1.7158E04 7.3692E03 4.6767E02 4.3753E04 2.6760E05 2.2534E05 1.646E04
7.0137E01 1.8081E01 7.2220E05 – 1.3605E06 1.6934E05 1.4597E02 4.1249E04 2.0512E05 1.3447E03 1.0124E01 7.0556E05 5.7095E06 – 3.6992E–05
Fig. 2. Profiles of ethanol, glucose and cell mass in test experiments (with sonication of SSF system at 10% duty cycle).
Note: conditions for simulations: ultrasound frequency = 35 kHz; ultrasound pressure amplitude = 150 kPa; equilibrium bubble radius = 5 and 10 lm; vapor pressure of water (in bar) is calculated using Antoine type correlation: 643:748 log10 P v ¼ 3:55959 T198:043 . Properties of water: density = 1000 kg/m3, kinematic viscosity = 106 Pa s, surface tension = 0.072 N/m and sonic velocity = 1481 m/s. Notation: Ro – initial radius of the cavitation bubble; Vturb – average velocity of the micro-turbulence in the medium generated by cavitation bubbles in the medium (estimated at 1 mm distance from bubble center); PAW – pressure amplitude of the acoustic wave generated by the cavitation bubble (estimated at 1 mm distance from bubble center); Tmax – temperature peak reached in the bubble at the time of first collapse; Pmax – pressure peak reached in the bubble at the time of first collapse.
(2)
(a)
(b) Fig. 1. Profiles of ethanol, glucose and cell mass in control experiments (with mechanical shaking of SSF system).
(1) SSF experiment with mechanical shaking resulted in a maximum ethanol concentration of 10.57 g/L after 54 h of fermentation, with an ethanol yield of 0.27 g/g of pretreated biomass or 0.15 g ethanol/g raw biomass). At the same time, the cell mass concentration of 4.37 g/L was achieved with a yield of 0.11 g cell mass/g of biomass. Ethanol and cell mass productivities in control experiments were 0.2 g/L/h and
(c)
0.08 g/L/h, respectively. The most notable effect of sonication on SSF process was in terms of reduction in time of fermentation. 3-Fold reduction in fermentation time was seen with sonication (18.3 h in test experiments against 54 h in control experiments). However, comparing with the SHF process with fermentation period of just 10 h (not inclusive of the time of cellulose hydrolysis prior to fermentation), the SSF process has still slower kinetics. In ultrasound assisted SSF process, ethanol titer was increased to 15.62 g/L with a yield of 0.4 g ethanol/g pretreated biomass or 0.21 g ethanol/g raw biomass. Due to reduction in fermentation time and enhancement in ethanol and cell mass concentrations, the ethanol and cell mass productivities were increased by 4-fold in the test experiments, viz. 0.85 g/L/h and 0.36 g/L/h, respectively, as compared to the control experiments. Comparative analysis of the values of the physiological parameters obtained from fitting of the fermentation model to the experimental data give a mechanistic account of the influence of ultrasound on the SSF process. The trends in the values of model parameters for the control and test experiments (as given in Table 3), and the explanations for these trends are as follows: The value of K3 (Monod constant for glucose for cell growth) has been reduced in ultrasound-assisted SSF (test experiment), which indicated higher utilization of substrate for cell growth. This is attributed to faster transport of glucose across cell membrane due to strong micro-convection generated by ultrasound. Due to enhanced mass transfer, lesser bulk concentration of glucose is required to achieve maximum specific growth rate. This is also evident from rise in the value of lm in test experiments. Under the influence of ultrasound, the maximal specific growth rate (lm) has been increased, and the specific cell death rate (kd) has been reduced. The main causes leading to cell death are depletion of nutrients or accumulation of toxic products in the medium. The microturbulence and intense mixing generated by ultrasound assists the transfer of nutrients across the cells and dilution of toxic substances in the vicinity of the cell, respectively. Both of these effects eventually result in enhanced growth phase of the cells, with reduced death rate in the test experiments. Rise in KI (inhibition constant of cell growth by glucose) indicated rise in tolerance of cells towards non-competitive inhibition by the substrate, or in other words, the inhibitory
254
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256 Table 2 Summary of experimental results under control and test conditions (results of SHF experiments are given in parentheses for comparison). Parameter
Control experiments (with mechanical shaking)*
Test experiments (with intermittent sonication at 10% duty cycle)#
Maximum ethanol concentration (g/L)
10.57 ± 0.15 (10.93 ± 0.2)** 0.27 (0.28) ** 0.15 (0.15) ** 0.2 (0.61) ** 4.37 ± 0.2 (5.26 ± 0.8) ** 0.11 (0.14) ** 0.08 (0.29) **
15.62 ± 0.2 (12.14 ± 0.5)## 0.40 (0.31)## 0.21 (0.17)## 0.85 (1.21)## 6.53 ± 0.3 (5.73 ± 0.5)## 0.17 (0.15)## 0.36 (0.57)##
Ethanol yield on pretreated/delignified biomass (g/g) Ethanol yield on raw biomass (g/g) Ethanol productivity (g/L/h) Maximum cell mass concentration (g/L) Cell mass yield on pretreated/delignified biomass (g/g) Cell mass productivity (g/L/h)
Concentration values are: mean ± standard error (n = 3). * Parameter values recorded at 54 h of fermentation in control experiments and at 18.3 h# in test experiments. ** Maximum ethanol and cell mass concentration at 18 h of fermentation in control experiments and 10 h## in test experiments.
Table 3 Kinetic and physiological parameters in fermentation model fitted to experimental data with GA optimization. Parameter
Control experiments
Test experiments
Monod constant for cell growth, K3 (g/L) Inhibition constant of cell growth by glucose, KI (g/L) Inhibition constant of cell growth by ethanol, K3E (g/L) Specific cell death rate, kd (1/h) Maximal specific growth rate, lm (1/h) Constant for growth associated ethanol formation, a (g/g) Non-growth associated specific ethanol production rate, b (g/g/h)
25.01 50.06
20.02 60.02
30.03
30.01
0.12 0.48 2.98
0.09 0.61 2.99
1.99
1.99
concentration of glucose shifted to a higher value. The reduction in both K3 and KI may be the result of the faster transport and utilization of glucose due to intense mixing introduced by ultrasound, and this essentially is the synergistic effect of sonication on fermentation. (d) Similar values of K3E (inhibition constant of cell growth by ethanol) in control and test experiments showed that the inhibition for cell growth by product (ethanol) remained practically unaltered. It indicated that this property solely depends on physiology of the cells, and not on its environment. Hence, it remained unaffected by the physical or
chemical effects induced by ultrasound and cavitation. This effect can be further explained on the basis of intrinsic properties of the cells related to ethanol inhibition. There are two major causes leading to cell growth inhibition by ethanol: (1) inhibition of enzymes involved in glycolytic pathway, and (2) effects on plasma membrane (fluidity, transport mechanisms or the enzymes associated with the membrane) [31]. These properties are mostly of intrinsic type, which remain unaffected by the physical or chemical effects of ultrasound. (e) Production of ethanol is directly related to the energy generation by microorganisms making this process growth-associated. However, a non-zero value (1.99 g/g/h) of b (non-growth associated specific ethanol production rate) indicates that ethanol formation takes place during stationary phase as well. The value of a (constant for growth associated ethanol formation) was greater than b in both control and test experiments, which suggested that ethanol production is predominantly a growth associated process. The results obtained in this study have been compared with the other reports in terms of ethanol titer, productivity and yield (with respect to raw biomass) in Table 4. It could be inferred from data presented in Table 4 that ethanol yield with respect to raw biomass in the present study is at par with the previous literature. This is a remarkable result despite use of cellulase enzyme from natural isolate for biomass hydrolysis. We attribute this result to the physical effect of ultrasound and cavitation, which enhance mass transfer in the system, and thus, kinetics and yield of the SSF process.
Table 4 Comparison of the results of present study with published literature. Biomass
Organism
Sonication
Ethanol titer (g/L)
Ethanol productivity (g/L/h)
Ethanol yield (g/g of raw biomass)
References
Parthenium hysterophorus Parthenium hysterophorus Waste paper Oil palm fronds Rice straw Saccharina japonica
Saccharomyces cerevisiae MTCC 170 Saccharomyces cerevisiae MTCC 170 Klebsiella oxytoca P2 Saccharomyces cerevisiae Saccharomyces cerevisiae AYH306 Pichia angophorae KCTC 17574, Pichia stipitis KCTC 7228, S. cerevisiae KCCM 1129, Pachysolen tannophilus KCTC 7937 Saccharomyces cerevisiae D4A Kluyveromyces marxianus Y01070 Saccharomyces cerevisiae
No Yes Yes Yes No No
10.57 15.62 36.6 18.2 58.7 7.7
0.20 0.85 0.38* 3.64 0.49 0.12*
0.27 0.40 N. D. N. D. 0.13 N. D.
This study This study [11] [12] [32] [33]
No No No
12.7 8.8 9.0
0.11* 0.12 0.13
N. D. 0.33 0.33
[34] [35] [35]
Rice straw Paper sludge Paper sludge *
As calculated from data reported in paper; N. D., not determined.
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
255
in ultrasound-assisted SSF (test experiment) could be a consequence of acceleration of fermentation as well as enzymatic hydrolysis under the influence of ultrasound. The ethanol productivity in SHF mode is higher than SSF. These values could be misnomer indicating SHF process to be more efficient than SSF. However, it should be noted that the productivity in SHF process has been determined on the basis of fermentation period only (and does not include the hydrolysis time). It is also noteworthy to see the trends in maximum cell mass concentration in SHF and SSF process under control and test conditions. Although in both SHF and SSF mode, higher cell mass concentration was seen in test experiments, comparison between SHF and SSF modes shows inverse variations. For control conditions (with mechanical shaking), SHF mode yields higher cell mass concentration, while for test conditions (sonication + mechanical shaking), SSF mode yields higher cell mass concentration. A plausible explanation for this could be given as follows: as stated earlier, ultrasound enhances not only fermentation but also enzymatic hydrolysis. In control experiments, due to slower hydrolysis due to relatively lower mixing induced by mechanical shaking, the instantaneous sugar concentration in the broth at any point of time is likely to be lower than in SHF, which results in lesser cell mass productivity. In test experiments, enhancement in cellulose hydrolysis due to strong micro-convection induced by sonication can significantly increase the instantaneous levels of sugar concentrations in the broth, even higher than that in the SHF mode, which could lead to higher cell mass production. 4. Conclusion
Fig. 3. Micrographs of methylene blue stained yeast cells after completion of fermentation in (A) control experiments and (B) test experiments.
3.2. Effect of sonication on cell viability Methylene blue staining (Figs. 3A and B) and calculation of viability by cell count method revealed that no cell death occurred due to sonication. Yeast cell viability in control and test samples was 83% and 84%, respectively. We attribute this result to small duty cycle (of only 10%) and moderate ultrasound intensity of 1.48 W/cm2.
In this study, we have attempted to get a physical insight into ultrasound induced enhancement in bioethanol synthesis from waste biomass P. hysterophorus using the simultaneous saccharification and fermentation mode. Results of fermentation with and without sonication have been coupled to a model comprising of kinetic and physiological parameters that are characteristics of the mechanism of fermentation. The trends in these parameters give an interesting account of the links between physics of ultrasound and cavitation and biomechanics of ethanol fermentation. It is revealed that strong microturbulence and mild shock waves induced by ultrasound and cavitation augment trans-membrane transport of substrate and products as well as dilution of the toxic substances. These phenomena are manifested in terms of enhancement in cell growth with faster consumption of substrate and reduced cell death rate. Strong convection generated by ultrasound and cavitation also enhances the kinetics of cellulose hydrolysis by cellulase/cellobiase. As a consequence of these, the ethanol productivity as well as cell mass productivity shows a 4 enhancement with sonication, as compared to mechanically stirred system. Nonetheless, the parameters related to intrinsic mechanisms of the cells, such as the inhibition constant for ethanol or the constants for growth and non-growth associated ethanol production have remained unaffected by sonication. The work presented in this paper could give important inputs in further mechanistic research in ultrasound-assisted bioethanol fermentation. The methodology adopted in this work can form a framework for investigations in other fermentation processes employing suitable mathematical models.
3.3. Comparative analysis of fermentation in SHF and SSF modes Acknowledgements Table 2 presents the values of different yardsticks that evaluate performance of fermentation process in SSF mode. Comparison of these values with the SHF mode gives interesting accounts of links between physical effects of sonication and mode of fermentation operation. The higher value of maximum ethanol concentration
Authors are grateful to Prof. Mark T. Holtzapple, Department of Chemical Engineering, Texas A&M University, Texas, United States, for his clarification on HCH-1 model. Authors acknowledge the HPLC and GC (procured through FIST grant SR/FST/ETII-028/2010
256
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
from Department of Science and Technology, Government of India) facilities at Department of Chemical Engineering, Indian Institute of Technology Guwahati. Authors are also thankful to Mr. Akshay Verma for his help in development of the mathematical model for fermentation. Authors also thank anonymous referees of this paper for their meticulous evaluation and constructive criticism. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ultsonch.2015.02. 011. References [1] R. Kataria, S. Ghosh, Saccharification of Kans grass using enzyme mixture from Trichoderma reesei for bioethanol production, Bioresour. Technol. 102 (2011) 9970–9975. [2] R.C. Kuhad, R. Gupta, Y.P. Khasa, A. Singh, Bioethanol production from Lantana camara (red sage): pretreatment, saccharification and fermentation, Bioresour. Technol. 101 (2010) 8348–8354. [3] R. Gupta, S. Kumar, J. Gomes, R.C. Kuhad, Kinetic study of batch and fed-batch enzymatic saccharification of pretreated substrate and subsequent fermentation to ethanol, Biotechnol. Biofuels 5 (2012) 16. [4] M. Galbe, G. Zacchi, A review of the production of ethanol from softwood, Appl. Biotechnol. 59 (2002) 618–628. [5] H.A. Choudhury, S. Chakma, V.S. Moholkar, Mechanistic insight into sonochemical biodiesel synthesis using heterogeneous base catalyst, Ultrason. Sonochem. 21 (2014) 69–181. [6] B.R. Reddy, T. Sivasankar, M. Sivakumar, V.S. Moholkar, Physical facets of ultrasonic cavitational synthesis of zinc ferrite particles, Ultrason. Sonochem. 17 (2010) 416–426. [7] V.S. Nalajala, V.S. Moholkar, Investigations in the physical mechanism of sonocrystallization, Ultrason. Sonochem. 18 (2010) 345–355. [8] T. Sivasankar, V.S. Moholkar, Mechanistic approach to intensification of sonochemical degradation of phenol, Chem. Eng. J. 149 (2009) 57–69. [9] M.M.C.G. Warmoeskerken, P. van der Vlist, V.S. Moholkar, V.A. Nierstrasz, Laundry process intensification by ultrasound, Colloids Surf. A 210 (2002) 277–285. [10] V.S. Moholkar, V.A. Nierstrasz, M.M.C.G. Warmoeskerken, Intensification of mass transfer in wet textile processes by power ultrasound, AUTEX Res. J. 3 (2003) 129–138. [11] B.E. Wood, H.C. Aldrich, L.O. Ingram, Ultrasound stimulates ethanol production during the simultaneous saccharification and fermentation of mixed waste office paper, Biotechnol. Prog. 13 (1997) 232–237. [12] C. Ofori-Boateng, K.T. Lee, Ultrasonic-assisted simultaneous saccharification and fermentation of pretreated oil palm fronds for sustainable bioethanol production, Fuel 119 (2014) 285–291. [13] G.P. Philippidis, D.D. Spindler, C.E. Wyman, Mathematical modeling of cellulose conversion to ethanol by simultaneous saccharification and fermentation process, Appl. Biochem. Biotechnol. 34 (35) (1992) 543–556. [14] J. Shen, F.A. Agblevor, Modeling semi-simultaneous saccharification and fermentation of ethanol production from cellulose, Biomass Bioenergy 34 (2010) 1098–1107.
[15] S. Singh, S. Khanna, V.S. Moholkar, A. Goyal, Comparative assessment of pretreatment strategies for enzymatic saccharification of Parthenium hysterophorus, Appl. Energy 129 (2014) 195–206. [16] S. Singh, S.T.P. Bharadwaja, P.K. Yadav, V.S. Moholkar, A. Goyal, Mechanistic investigation in ultrasound-assisted (alkaline) delignification of Parthenium hysterophorus biomass, Ind. Eng. Chem. Res. 53 (2014) 14241–14252. [17] S. Singh, V.S. Moholkar, A. Goyal, Isolation, identification and characterization of a cellulolytic Bacillus amyloliquefaciens strain SS35 from rhinoceros dung, ISRN Microbiol. Article ID 728134 (2013) 7 pages. [18] S. Singh, V.S. Moholkar, A. Goyal, Optimization of carboxymethylcellulase production from Bacillus amyloliquefaciens SS35, 3 Biotech 4 (2014) 411–424. [19] S. Singh, P.K. Dikshit, V.S. Moholkar, A. Goyal, Purification and characterization of acidic cellulase from Bacillus amyloliquefaciens SS35 for hydrolyzing Parthenium hysterophorus biomass, Environ. Prog. Sustainable Energy (2014), http://dx.doi.org/10.1002/ep.12046. [20] M. Karuppaiya, E. Sasikumar, T. Viruthagiri, V. Vijayagopal, Optimization of process variables using response surface methodology (RSM) for ethanol production from cashew apple juice by Saccharomyces cerevisiae, Asian J. Food Agro-Ind. 3 (2010) 462–473. [21] T. Sivasankar, A.W. Paunikar, V.S. Moholkar, Mechanistic approach to enhancement of the yield of a sonochemical reaction, AIChE J. 53 (2007) 1132–1143. [22] S. Chakma, V.S. Moholkar, Mechanistic features of ultrasonic desorption of aromatic pollutants, Chem. Eng. J. 175 (2011) 356–367. [23] V.S. Moholkar, S.P. Sable, A.B. Pandit, Mapping the cavitation intensity in an ultrasonic bath using the acoustic emission, AIChE J. 46 (2000) 684–694. [24] V.S. Moholkar, M. Huitema, S. Rekveld, M.M.C.G. Warmoeskerken, Characterization of an ultrasonic system using wavelet transforms, Chem. Eng. Sci. 57 (2002) 617–629. [25] N. Nelson, A photometric adaptation of the Somogyi method for the determination of glucose, J. Biol. Chem. 153 (1944) 375–380. [26] M. Somogyi, A new reagent for the determination of sugar, J. Biol. Chem. 160 (1945) 61–68. [27] S. Abd-Aziz, G.S. Hung, M.A. Hassan, M.I.A. Karim, N. Samat, Indirect method for quantification of cell biomass during solid state fermentation of palm kernel cake based on protein content, Asian J. Sci. Res. 1 (2008) 385–393. [28] K. Painting, B. Kirsop, A quick method for estimating the percentage of viable cells in a yeast population, using methylene blue staining, World J. Microbiol. Biotechnol. 6 (1990) 346–347. [29] M.T. Holtzapple, H.S. Caram, A.E. Humphrey, The HCH-1 model of enzymatic cellulose hydrolysis, Biotechnol. Bioeng. 26 (1984) 775–780. [30] R. Toegel, B. Gompf, R. Pecha, D. Lohse, Does water vapor prevent upscaling sonoluminescence?, Phys Rev. Lett. 85 (2000) 3165–3168. [31] G. Moulin, B. Helen, P. Galzy, Inhibition of alcoholic fermentation, Biotechnol. Genet. Eng. Rev. 2 (1984) 365–382. [32] W.L. Sun, W.Y. Tao, Simultaneous saccharification and fermentation of rice straw pretreated by a sequence of dilute acid and dilute alkali at high dry matter content, Energy Source Part A 35 (2013) 741–752. [33] J.S. Jang, Y.K. Cho, G.T. Jeong, S.K. Kim, Optimization of saccharification and ethanol production by simultaneous saccharification and fermentation (SSF) from seaweed, Saccharina japonica, Bioprocess Biosyst. Eng. 35 (2012) 11–18. [34] J.K. Ko, J.S. Bak, M.W. Jung, H.J. Lee, I.G. Choi, T.H. Kim, K.H. Kim, Ethanol production from rice straw using optimized aqueous-ammonia soaking pretreatment and simultaneous saccharification and fermentation processes, Bioresour. Technol. 100 (2009) 4374–4380. [35] Zs. Kadar, Zs. Szengyel, K. Réczey, Simultaneous saccharification and fermentation (SSF) of industrial wastes for the production of ethanol, Ind. Crops Prod. 20 (2004) 103–110.
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultson
Mechanistic insight into ultrasound induced enhancement of simultaneous saccharification and fermentation of Parthenium hysterophorus for ethanol production Shuchi Singh a, Mayank Agarwal b,1, Shyamali Sarma a,1, Arun Goyal a,c,⇑, Vijayanand S. Moholkar a,b,⇑ a b c
Centre for Energy, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India Department of Biotechnology, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India
a r t i c l e
i n f o
Article history: Received 17 November 2014 Received in revised form 21 January 2015 Accepted 25 February 2015 Available online 17 March 2015 Keywords: Ultrasound Microturbulence Fermentation Bioethanol Parthenium hysterophorus Mathematical modeling
a b s t r a c t This paper presents investigations into mechanism of ultrasound assisted bioethanol synthesis using Parthenium hysterophorus biomass through simultaneous saccharification and fermentation (SSF) mode. Approach of coupling experimental results to mathematical model for SSF using Genetic Algorithm based optimization has been adopted. Comparison of model parameters for experiments with mechanical shaking and sonication (10% duty cycle) give an interesting mechanistic account of influence of ultrasound on SSF system. A 4-fold rise in ethanol and cell mass productivity is seen with ultrasound. The analysis reveals following facets of influence of ultrasound on SSF: increase in Monod constant for glucose for cell growth, maximal specific growth rate and inhibition constant of cell growth by glucose and reduction in specific cell death rate. Values of inhibition constant of cell growth by ethanol (K3E), and constants for growth associated (a) and non-growth associated (b) ethanol production remained unaltered with sonication. Beneficial effects of ultrasound are attributed to enhanced cellulose hydrolysis, enhanced trans-membrane transport of substrate and products as well as dilution of the toxic substances due to micro-convection induced by ultrasound. Intrinsic physiological functioning of cells remained unaffected by ultrasound as indicated by unaltered values of K3E, a and b. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Bioethanol production from lignocellulosic biomass has been a highly active research area for past several years, as ethanol has shown high promise as an alternate liquid transportation fuel as well as oxygenate blend for gasoline. The conventional process for bioethanol production has two steps, viz. pretreatment and
Abbreviations: CMCase, carboxymethylcellulase; GA, Genetic Algorithm; GC, gas chromatography; HPLC, high performance liquid chromatography; MTCC, microbial type culture collection; RI, refractive index; SHF, separate hydrolysis and fermentation; SSF, simultaneous saccharification and fermentation; TRS, total reducing sugar. ⇑ Corresponding authors at: Centre for Energy, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India. Tel.: +91 361 258 2208; fax: +91 361 258 2249 (A. Goyal). Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781 039, Assam, India. Tel.: +91 361 258 2251; fax: +91 361 258 2291 (V.S. Moholkar). E-mail addresses: [email protected] (A. Goyal), [email protected] (V.S. Moholkar). 1 Authors contributed equally. http://dx.doi.org/10.1016/j.ultsonch.2015.02.011 1350-4177/Ó 2015 Elsevier B.V. All rights reserved.
acid/enzymatic hydrolysis of lignocellulosic biomass followed by fermentation of the acid and/or enzymatic hydrolyzate. The cost of production of bioethanol is a major function of cost of fermentation substrate as well as the operating cost of the process. Lignocellulosic biomass available abundantly in the form of agroresidues, forest-residues, and waste biomass (weed/grass) forms a potential low-cost feedstock for bioethanol. Some typical examples of waste biomass whose carbohydrate moieties have been used for bioethanol production are Saccharum spontaneum [1], Lantana camara [2] and Prosopis juliflora [3]. In order to intensify the bioethanol productivity while reducing the cost of production, the process of simultaneous saccharification and fermentation (SSF) has also been extensively investigated. This process has distinct advantages of milder operating conditions, and requirement of a single fermentor vessel that combines the two steps of hydrolysis and fermentation mentioned above. In this process, hexose sugars released from enzymatic hydrolysis of cellulose in the biomass are simultaneously consumed by fermenting microorganisms. The enzymatic hydrolysis of cellulose itself is a two-step process, in which first cellulase hydrolyzes the cellulose into
250
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
Nomenclature Notation a, b constants for ethanol formation, growth associated and non-growth associated, respectively at available surface area for cellulose (B), (C), (E), (G), (X) concentrations of cellobiose, cellulose, ethanol, glucose and cell mass, respectively (E1)t total concentration of cellulase in the solution (E2) concentration of b-glucosidase in the solution kd specific rate of cell death k1, k2 specific rates of cellulose and cellobiose hydrolysis, respectively KI constant of cell growth inhibition by glucose Km Michaelis constant of b-glucosidase for cellobiose K3, K4 Monod constants of glucose for cell growth and ethanol synthesis, respectively
cellobiose (dimer hexose sugar units), which are later split into monomeric hexose sugar units by cellobiase (or b-glucosidase). SSF process reduces the inhibitory effect of substrate (sugar) concentration on enzymes and also the probability of contamination by undesired invasive microorganisms [4]. These features increase yield as well as kinetics of the saccharification as well as fermentation as compared to the conventional two-step process. More recently, another technique of ultrasound irradiation or sonication of the fermentation broth for intensification of bioethanol fermentation has been attempted. Ultrasound is a well known technique for intensification of diverse physical and chemical processes [5–10]. Ultrasound manifests its effect on the reaction system through phenomenon of cavitation, which is nucleation, growth, oscillation and implosive transient collapse of tiny gas or vapor bubbles, which is driven by pressure variation generated in the medium during passage of ultrasound wave. Both ultrasound and cavitation render several physical and chemical effects on the reactions system, which are beneficial in enhancing the kinetics of the system. The most peculiar feature of energy introduction into the medium via ultrasound and cavitation is that implosive collapse of cavitation bubbles creates intense energy concentration on an extremely small spatial and temporal scale. The main physical effect of ultrasound and cavitation is generation of intense micro-turbulence in the medium that gives very effective micromixing, which eliminates mass transfer limitations. The chemical effect of transient cavitation is generation of highly reactive radicals and other smaller species through dissociation of vapor entrapped in the bubble at the moment of transient collapse. Literature on application of ultrasound during bioethanol synthesis through SSF process is rather limited. Wood et al. [11] have reported bioethanol production using ultrasound (36 kHz, 150 W) assisted SSF process. The substrate used was waste paper and microbial strain of Klebsiella oxytoca was employed. Bioethanol yield was found to increase by 20% with sonication. OforiBoateng and Lee [12] have investigated bioethanol production using SSF process from oil palm fronds as substrate and Saccharomyces cerevisiae as the microbial strain. With ultrasound of 40 kHz frequency and 200 W intensity, 4-fold increment in bioethanol yield was observed within 5 h. In order to effectively utilize the potential of ultrasound on intensification of the SSF process for bioethanol production, it is essential to understand the basic underlying physical mechanism. This would essentially mean identifying the links between physics of ultrasound and cavitation and the biochemistry of fermentation. In this paper, we have addressed this important issue with the
K1B, K2B Inhibition constants of cellulase and b-glucosidase by cellobiose, respectively t time K1G, K2G inhibition constants of cellulase and b-glucosidase by glucose, respectively K1E, K2E, K3E Inhibition constants of cellulase, b-glucosidase and cell growth by ethanol, respectively m specific rate of substrate consumption for maintenance requirements lm maximal specific growth rate YX/G average yield coefficient of cell mass on substrate (glucose)
approach of coupling experimental results to the fermentation model of [13], which comprises of 5 ordinary differential equations, viz. one each for cellulose, cellobiose, glucose, microbial cell concentration and ethanol. This model takes into account the essential physiology of the SSF process. A major practical limitation of implementation of this model is difficulty in monitoring of the concentration of cellulose (which occurs in solid phase) and also the unstable intermediate of hexose-dimer cellobiose (which is rapidly decomposed into monomeric glucose) during fermentation [14]. Despite this limitation, fitting of the experimental data of microbial cell concentration and ethanol to their respective differential model equations reveals important mechanistic account of the influence of ultrasound on the SSF process. In our experiments, we have used waste biomass of Parthenium hysterophorus as substrate with S. cerevisiae as the microbial strain.
2. Materials, methods and mathematical model 2.1. Chemicals and reagents All components of fermentation medium were procured from HiMedia Pvt. Ltd., India. Glucose (99.5% purity, standard for HPLC and reducing sugar estimation) was procured from Sigma Aldrich, USA. Ethanol (99.5% purity) was procured from Tedia Chemicals, USA. All other chemicals were procured from Fischer Scientific, India.
2.2. P. hysterophorus biomass P. hysterophorus biomass was collected from the campus of our institute. Biomass was chopped (5 cm), washed with water, dried at 60 ± 3 °C for 24 h and ground to a particle size of 1 mm. Powdered biomass was pretreated with 1% (v/v) H2SO4 + 30 min autoclaving [15], and the solid residue was further delignified by ultrasound assisted alkaline treatment [16].
2.3. Source of enzymes Carboxymethylcellulase (CMCase) (1.0 U/mg, 1.7 mg/mL) was produced from Bacillus amyloliquefaciens SS35 [17–19] and b-glucosidase (250 U/mL) (Novozyme 188) was procured from Sigma Aldrich, USA.
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
2.4. Microorganism and fermentation conditions S. cerevisiae MTCC 170 was used for ethanol fermentation. Microbial culture was procured from Microbial Type Culture Collection (MTCC), Institute of Microbial Technology (IMTECH), Chandigarh, India and was maintained in YEPD medium containing yeast extract (10 g/L), peptone (20 g/L) and dextrose (20 g/L). 2.5. Simultaneous saccharification and fermentation (SSF): control experiments
standard. Dry cell mass for yeast cells was determined by using an indirect method by plotting a calibration curve using intracellular protein content (g/L) and dry cell mass (g/L) of the yeast cells [27]. Viability of S. cerevisiae MTCC 170 cells in the samples before and after sonication was determined using methylene blue staining method [28]. 0.1% (w/v) methylene blue solution was mixed with yeast broth in a ratio of 1:1 and incubated for 5 min. Cells were counted on hemocytometer at 40 magnification. The cell viability was calculated using the formula: Viability ð%Þ ¼
Simultaneous saccharification and fermentation (SSF) of pretreated and delignified P. hysterophorus was carried out by using the enzymes CMCase and b-glucosidase, and S. cerevisiae MTCC 170 culture. 3.88% w/v pretreated and delignified P. hysterophorus biomass was added to fermentation medium, which consisted of yeast extract (10 g/L), KH2PO4 (2 g/L), (NH4)2SO4 (1 g/L) and MgSO47H2O (1 g/L) (pH 5.0) [20]. This mixture was supplemented with CMCase and b-glucosidase with concentrations of 600 U/g and 50 FPU/g of delignified biomass, respectively. SSF was carried out in a 150 mL Erlenmeyer flask with working volume of 25 mL. The medium was inoculated with 10% (v/v) of previously grown culture of S. cerevisiae MTCC 170 and incubated at 30 °C and 150 rpm in an incubator shaker (Scigenics Biotech, Model: Orbitek). 2.6. Simultaneous saccharification and fermentation (SSF): test experiments Ultrasound-assisted fermentation (test experiment) was carried out in an ultrasound bath (Make: Elma, Germany; Model: Transonic T-460, 2L) operating at a frequency of 35 kHz and power rating of 35 W. The actual power input to the liquid medium in the bath (i.e. water) was determined calorimetrically as 18.58 W [21,22]. This power is transmitted through the transducer attached to bottom of the bath. The diameter of this transducer was 4 cm. Therefore, the power dissipation to sonication medium is 1.48 W/ cm2 and the corresponding acoustic power amplitude was 150 kPa. The ultrasound was applied in a duty cycle of 10% (i.e. 1 min of sonication and 9 min of mechanical shaking in every 10 min of fermentation). The total working volume of the bath was 2 L (dimensions: 25 cm 15 cm 10 cm). The bath was filled with water and the flask containing fermentation mixture was placed at the center of the bath. The flask was immersed to about 50% of its height in the water and the position of the Erlenmeyer flask was carefully maintained same in all experiments in view of spatial variation of the acoustic intensity [23,24]. The temperature of the water in the bath was maintained at 30o ± 2 °C by replacement of small portions of initial water at regular intervals. 200 lL aliquots of the fermentation broth were withdrawn intermittently in both control and test experiments to determine total reducing sugar, ethanol and cell mass concentration in fermentation broth. The total fermentation period for control and test experiments was set at 78 and 35 h, respectively, based on the concentration of residual sugar in the samples of fermentation broth. The fermentation was stopped when the sugar concentration became constant (< 2% difference) in the consecutive samples withdrawn from fermentation broth. Both test and control experiments were repeated thrice to assess the reproducibility of the results. 2.7. Determination of total reducing sugar, cell mass and cell viability Aliquots withdrawn from fermentation broth were treated using the method of Nelson [25] and Somogyi [26] for determination of the residual reducing sugar, using D-glucose as
251
Number of live ðunstainedÞ cells Number of live ðunstainedÞ cells þ Number of dead ðstainedÞ cells 100
2.8. Analytical methods Presence of glucose in the aliquots of fermentation broth was confirmed with HPLC analysis (Perkin-Elmer, Series 200, with a refractive index detector) using HiPlex-H column (300 mm 5 lm 4.6 mm, Varian). HPLC grade water (deionized water, Milli Q) was used as the mobile phase at a flow rate of 0.4 mL/min with glucose (99.5% purity) as standard in sugar analysis. Ethanol content of fermentation samples was determined by Gas Chromatography (Varian, CP 3800) using a CP Wax 52 CB capillary column (250 mm 0.25 mm 0.39 mm, Varian) with ethanol (99.5% purity) as the standard. The oven temperature was programmed from 45 °C to 100 °C with an increment of 3 °C/min and 100 °C to 200 °C with an increment of 5 °C/min. The injector and detector temperatures were kept at 230 °C and 250 °C, respectively. Nitrogen gas was used as a carrier at a flow rate of 2 mL/min. 2.9. Mathematical model We have used ethanol fermentation model for the SSF mode developed by Philippidis et al. [13] in this study. This model is essentially based on the HCH-1 model of enzymatic hydrolysis of cellulose reported by Holtzapple et al. [29]. This model makes several assumptions as follows: (1) No distinction between two components of cellulase, viz. endoglucanase and exoglucanase; (2) exogenous addition of cellulose hydrolysing enzymes, viz. cellulase and b-glucosidase to the SSF system; (3) cellulose is hydrolyzed to cellobiose by cellulase with negligible formation of glucose; (4) pH of the fermentation broth remains practically constant during SSF process; (5) major metabolic products of SSF process are ethanol and CO2, (6) the carbon source for metabolism (glucose) is derived from cellulose, while the growth medium provides sufficient excess of all other nutrients, and (7) the activities of cellulase and b-glucosidase enzymes are assumed to remain constant all through the process. The model of Philippidis et al. [13] proposes that cellulose to ethanol conversion process in SSF mode comprises of following steps: (1) cellulase diffusion towards cellulose, (2) adsorption of cellulase on surface of cellulose, (3) hydrolysis of cellulose to cellobiose, (4) diffusion of cellobiose into aqueous phase, (5) hydrolysis of cellobiose to glucose (catalyzed by b-glucosidase), (6) glucose diffusion towards cells, (7) glucose uptake by cells, (8) glucose to ethanol conversion and (9) ethanol secretion into aqueous phase. The model comprises of five ordinary differential equations for cellulose, cellobiose, glucose, cell mass and ethanol as follows: Cellulose:
dðCÞ k1 at ðE1 Þt K 1E i ¼ h ðBÞ ðGÞ K 1E þ ðEÞ dt K e þ ðE1 Þt 1 þ K 1B þ K 1G
252
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
Cellobiose:
fermentation with the model is depicted in the flowchart given in Supplementary material provided with the manuscript.
dðBÞ k1 at ðE1 Þt K 1E i ¼ 1:06 h ðBÞ ðGÞ K 1E þ ðEÞ dt K e þ ðE1 Þt 1 þ K 1B þ K 1G
k2 ðE2 ÞðBÞ K 2E h i h i ðBÞ K 2E þ ðEÞ K m 1 þ KðGÞ þ ðBÞ 1 þ K 2B 2G
Cell mass:
dðXÞ lm ðGÞ K 3E ¼ ðXÞ kd ðXÞ 2 dt K 3 þ ðGÞ þ ðGÞ K 3E þ ðEÞ KI
Glucose:
dðGÞ k2 ðE2 ÞðBÞ K 2E i h i ¼ 1:05 h ðGÞ ðBÞ K 2E þ ðEÞ dt K m 1 þ K 2G þ ðBÞ 1 þ K 2B 1 dðXÞ þ mðXÞ Y X=G dt Ethanol:
dðEÞ dðXÞ ðGÞ ¼ a þ bðXÞ dt dt K 4 þ ðGÞ The following changes have been made in original model equations of Philippidis et al. [13] in view of experimental conditions of present experiments. In view of acid + autoclaved biomass that results in high reduction of cellulose crystallinity and increase in porosity, the value of substrate reactivity coefficient (/) is set equal to 1. The values of two parameters, viz. adsorption constants for cellulase and b-glucosidase onto lignin are set to zero in view of use of delignified biomass as the substrate in SSF process. The concentration of lignin in the biomass is also assumed to zero. Zero value has been assigned to parameter K4 (Monod constant of glucose for ethanol synthesis) as per the results of Philippidis et al. [13] which demonstrated that ethanol formation was not directly dependent on the glucose concentration. As we have stated earlier, due to limitations of experimental facilities used in this study, we have not been able to monitor the concentration of cellulose (existing in solid phase) and the intermediate dimeric cellobiose during the fermentation process. However, we have monitored the concentration of glucose in the fermentation broth. Therefore, while fitting our experimental data, we have used only the equations for cell mass and ethanol. These equations have a total of 7 parameters, viz. K3, KI, K3E, kd, lm, a and b. The glucose concentration profile obtained during SSF experiments was fitted to a polynomial expression, and this expression was used to determine instantaneous glucose concentration in the broth at a time, during the numerical integration of the differential equations for cell mass and ethanol. Fitting of these equations to the experimental profiles of cell mass and ethanol would yield the numerical values of these model parameters, which represent kinetic and physiological facets of the SSF system. Variations in the values of these model parameters for the test and control experiments give a mechanistic account of the influence of ultrasound on the SSF system. For determination of the optimum values of the model parameters – so as to match the experimental and model predicted trends, the main model of two simultaneous ordinary differential equations was solved using Runge–Kutta 4th order method, and the solution was coupled to Genetic Algorithm (GA) optimization module of MATLAB. Optimization of the model parameters was done by calculating root mean square (RMS) error between experimental and model results. GA module minimizes this error, yielding optimized values of the model parameters. The overall algorithm of simulations of
2.10. Physical and chemical effects of ultrasound and transient cavitation The physical and chemical effects of ultrasound and cavitation are responsible for the acceleration of kinetics of processes. The physical effect is generation of strong micro-convection in the medium due to three phenomena: (1) oscillatory motion of fluid elements called ‘‘microstreaming’’ due to propagation of ultrasound in the form of compression/rarefaction cycles, (2) acoustic or shock waves emitted during transient collapse of cavitation bubbles, and (3) oscillatory motion of liquid called ‘‘micro-turbulence’’ induced by volumetric oscillation of cavitation bubbles. The chemical effect of cavitation is generation of highly reactive radicals due to thermal dissociation of solvent vapor molecules trapped in the cavitation bubble at the instance of transient collapse. The magnitudes of both physical and chemical effects can be quantitatively estimated using simulations of cavitation bubble dynamics. The physical and chemical effects of cavitation bubbles, as estimated by simulations of cavitation bubble dynamics using diffusion limited model of Toegel et al. [30] have been given in Table 1. Conditions used for simulations have been listed in the footnote of the table. Diffusion limited bubble dynamics model has been explained in greater details in several of our earlier papers [6,8,21]. It could be seen from the results presented in Table 1 that mild shock waves (with amplitude 30–70 kPa) emitted by the cavitation bubble can help enhance enzymatic hydrolysis due to greater access of enzymes to biomass and faster transport of sugars. The OH radicals generated by the cavitation bubbles can also boost enzymatic hydrolysis. The shock waves also cause rapid movement of the microbial cells. The kinetic energy gained from this movement can help enhance intracellular metabolism. As far as the convection generated by ultrasound is concerned, the magnitude of the micro-streaming velocity is given as, u ¼ P A =qC, where PA is pressure amplitude of ultrasound wave, q is density of medium and C is sonic velocity in medium. For typical values of PA = 150 kPa (or 1.5 bar), q = 1000 kg/m3 and C = 1500 m/s (for water as medium), u = 0.1 m/s. Relating this velocity to amplitude of oscillation as u ¼ 2pfa, for f (frequency) = 20 kHz, we get a (amplitude) = 0.8 l. In the context of present study, the spatial scale of micro-streaming is much smaller than the size of the S. cerevisiae cells (typically 5–10 l) through which diffusion of substrate (glucose) and product (ethanol) occurs. Hence, the convection generated by ultrasound is more effective than the convectional mechanical agitation in enhancing the transport on bulk as well as cellular level. 3. Results and discussion 3.1. Simultaneous saccharification and fermentation (SSF): with and without sonication Figs. 1 and 2 show the time profiles of glucose, ethanol and cell mass in control and test experiments, respectively. The summary of the results of the SSF process under control and test conditions is given in Table 2. The values of the physiological parameters in the differential equations for cell mass and ethanol, obtained with GA optimization are listed in Table 3. The salient features of the ethanol synthesis by SSF process in control (mechanical shaking) and test (sonication + mechanical shaking) conditions, which could be identified from results given in Figs. 1 and 2 and Tables 2 and 3 are as follows:
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
253
Table 1 Summary of cavitation bubble dynamics simulations indicating physical and chemical effects induced by transient cavitation. Species
Parameters for simulations Water Air bubble Air bubble Ro = 5 lm Ro = 10 lm Conditions at the first collapse of the bubble Tmax = 3258 K Tmax = 2304 K Pmax = 88.4 MPa Pmax = 384 MPa Vturb = 0.03 mm/s Vturb = 0.05 mm/s PAW = 72 kPa PAW = 31.6 kPa Equilibrium composition of bubble at transient collapse
N2 O2 O O3 H H2 NO NO2 N2O OH H2O HO2 H2O2 HNO HNO2
7.1952E01 1.6608E01 1.6723E03 6.4788E06 7.1808E05 1.8629E04 5.6173E02 1.3272E03 1.7158E04 7.3692E03 4.6767E02 4.3753E04 2.6760E05 2.2534E05 1.646E04
7.0137E01 1.8081E01 7.2220E05 – 1.3605E06 1.6934E05 1.4597E02 4.1249E04 2.0512E05 1.3447E03 1.0124E01 7.0556E05 5.7095E06 – 3.6992E–05
Fig. 2. Profiles of ethanol, glucose and cell mass in test experiments (with sonication of SSF system at 10% duty cycle).
Note: conditions for simulations: ultrasound frequency = 35 kHz; ultrasound pressure amplitude = 150 kPa; equilibrium bubble radius = 5 and 10 lm; vapor pressure of water (in bar) is calculated using Antoine type correlation: 643:748 log10 P v ¼ 3:55959 T198:043 . Properties of water: density = 1000 kg/m3, kinematic viscosity = 106 Pa s, surface tension = 0.072 N/m and sonic velocity = 1481 m/s. Notation: Ro – initial radius of the cavitation bubble; Vturb – average velocity of the micro-turbulence in the medium generated by cavitation bubbles in the medium (estimated at 1 mm distance from bubble center); PAW – pressure amplitude of the acoustic wave generated by the cavitation bubble (estimated at 1 mm distance from bubble center); Tmax – temperature peak reached in the bubble at the time of first collapse; Pmax – pressure peak reached in the bubble at the time of first collapse.
(2)
(a)
(b) Fig. 1. Profiles of ethanol, glucose and cell mass in control experiments (with mechanical shaking of SSF system).
(1) SSF experiment with mechanical shaking resulted in a maximum ethanol concentration of 10.57 g/L after 54 h of fermentation, with an ethanol yield of 0.27 g/g of pretreated biomass or 0.15 g ethanol/g raw biomass). At the same time, the cell mass concentration of 4.37 g/L was achieved with a yield of 0.11 g cell mass/g of biomass. Ethanol and cell mass productivities in control experiments were 0.2 g/L/h and
(c)
0.08 g/L/h, respectively. The most notable effect of sonication on SSF process was in terms of reduction in time of fermentation. 3-Fold reduction in fermentation time was seen with sonication (18.3 h in test experiments against 54 h in control experiments). However, comparing with the SHF process with fermentation period of just 10 h (not inclusive of the time of cellulose hydrolysis prior to fermentation), the SSF process has still slower kinetics. In ultrasound assisted SSF process, ethanol titer was increased to 15.62 g/L with a yield of 0.4 g ethanol/g pretreated biomass or 0.21 g ethanol/g raw biomass. Due to reduction in fermentation time and enhancement in ethanol and cell mass concentrations, the ethanol and cell mass productivities were increased by 4-fold in the test experiments, viz. 0.85 g/L/h and 0.36 g/L/h, respectively, as compared to the control experiments. Comparative analysis of the values of the physiological parameters obtained from fitting of the fermentation model to the experimental data give a mechanistic account of the influence of ultrasound on the SSF process. The trends in the values of model parameters for the control and test experiments (as given in Table 3), and the explanations for these trends are as follows: The value of K3 (Monod constant for glucose for cell growth) has been reduced in ultrasound-assisted SSF (test experiment), which indicated higher utilization of substrate for cell growth. This is attributed to faster transport of glucose across cell membrane due to strong micro-convection generated by ultrasound. Due to enhanced mass transfer, lesser bulk concentration of glucose is required to achieve maximum specific growth rate. This is also evident from rise in the value of lm in test experiments. Under the influence of ultrasound, the maximal specific growth rate (lm) has been increased, and the specific cell death rate (kd) has been reduced. The main causes leading to cell death are depletion of nutrients or accumulation of toxic products in the medium. The microturbulence and intense mixing generated by ultrasound assists the transfer of nutrients across the cells and dilution of toxic substances in the vicinity of the cell, respectively. Both of these effects eventually result in enhanced growth phase of the cells, with reduced death rate in the test experiments. Rise in KI (inhibition constant of cell growth by glucose) indicated rise in tolerance of cells towards non-competitive inhibition by the substrate, or in other words, the inhibitory
254
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256 Table 2 Summary of experimental results under control and test conditions (results of SHF experiments are given in parentheses for comparison). Parameter
Control experiments (with mechanical shaking)*
Test experiments (with intermittent sonication at 10% duty cycle)#
Maximum ethanol concentration (g/L)
10.57 ± 0.15 (10.93 ± 0.2)** 0.27 (0.28) ** 0.15 (0.15) ** 0.2 (0.61) ** 4.37 ± 0.2 (5.26 ± 0.8) ** 0.11 (0.14) ** 0.08 (0.29) **
15.62 ± 0.2 (12.14 ± 0.5)## 0.40 (0.31)## 0.21 (0.17)## 0.85 (1.21)## 6.53 ± 0.3 (5.73 ± 0.5)## 0.17 (0.15)## 0.36 (0.57)##
Ethanol yield on pretreated/delignified biomass (g/g) Ethanol yield on raw biomass (g/g) Ethanol productivity (g/L/h) Maximum cell mass concentration (g/L) Cell mass yield on pretreated/delignified biomass (g/g) Cell mass productivity (g/L/h)
Concentration values are: mean ± standard error (n = 3). * Parameter values recorded at 54 h of fermentation in control experiments and at 18.3 h# in test experiments. ** Maximum ethanol and cell mass concentration at 18 h of fermentation in control experiments and 10 h## in test experiments.
Table 3 Kinetic and physiological parameters in fermentation model fitted to experimental data with GA optimization. Parameter
Control experiments
Test experiments
Monod constant for cell growth, K3 (g/L) Inhibition constant of cell growth by glucose, KI (g/L) Inhibition constant of cell growth by ethanol, K3E (g/L) Specific cell death rate, kd (1/h) Maximal specific growth rate, lm (1/h) Constant for growth associated ethanol formation, a (g/g) Non-growth associated specific ethanol production rate, b (g/g/h)
25.01 50.06
20.02 60.02
30.03
30.01
0.12 0.48 2.98
0.09 0.61 2.99
1.99
1.99
concentration of glucose shifted to a higher value. The reduction in both K3 and KI may be the result of the faster transport and utilization of glucose due to intense mixing introduced by ultrasound, and this essentially is the synergistic effect of sonication on fermentation. (d) Similar values of K3E (inhibition constant of cell growth by ethanol) in control and test experiments showed that the inhibition for cell growth by product (ethanol) remained practically unaltered. It indicated that this property solely depends on physiology of the cells, and not on its environment. Hence, it remained unaffected by the physical or
chemical effects induced by ultrasound and cavitation. This effect can be further explained on the basis of intrinsic properties of the cells related to ethanol inhibition. There are two major causes leading to cell growth inhibition by ethanol: (1) inhibition of enzymes involved in glycolytic pathway, and (2) effects on plasma membrane (fluidity, transport mechanisms or the enzymes associated with the membrane) [31]. These properties are mostly of intrinsic type, which remain unaffected by the physical or chemical effects of ultrasound. (e) Production of ethanol is directly related to the energy generation by microorganisms making this process growth-associated. However, a non-zero value (1.99 g/g/h) of b (non-growth associated specific ethanol production rate) indicates that ethanol formation takes place during stationary phase as well. The value of a (constant for growth associated ethanol formation) was greater than b in both control and test experiments, which suggested that ethanol production is predominantly a growth associated process. The results obtained in this study have been compared with the other reports in terms of ethanol titer, productivity and yield (with respect to raw biomass) in Table 4. It could be inferred from data presented in Table 4 that ethanol yield with respect to raw biomass in the present study is at par with the previous literature. This is a remarkable result despite use of cellulase enzyme from natural isolate for biomass hydrolysis. We attribute this result to the physical effect of ultrasound and cavitation, which enhance mass transfer in the system, and thus, kinetics and yield of the SSF process.
Table 4 Comparison of the results of present study with published literature. Biomass
Organism
Sonication
Ethanol titer (g/L)
Ethanol productivity (g/L/h)
Ethanol yield (g/g of raw biomass)
References
Parthenium hysterophorus Parthenium hysterophorus Waste paper Oil palm fronds Rice straw Saccharina japonica
Saccharomyces cerevisiae MTCC 170 Saccharomyces cerevisiae MTCC 170 Klebsiella oxytoca P2 Saccharomyces cerevisiae Saccharomyces cerevisiae AYH306 Pichia angophorae KCTC 17574, Pichia stipitis KCTC 7228, S. cerevisiae KCCM 1129, Pachysolen tannophilus KCTC 7937 Saccharomyces cerevisiae D4A Kluyveromyces marxianus Y01070 Saccharomyces cerevisiae
No Yes Yes Yes No No
10.57 15.62 36.6 18.2 58.7 7.7
0.20 0.85 0.38* 3.64 0.49 0.12*
0.27 0.40 N. D. N. D. 0.13 N. D.
This study This study [11] [12] [32] [33]
No No No
12.7 8.8 9.0
0.11* 0.12 0.13
N. D. 0.33 0.33
[34] [35] [35]
Rice straw Paper sludge Paper sludge *
As calculated from data reported in paper; N. D., not determined.
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
255
in ultrasound-assisted SSF (test experiment) could be a consequence of acceleration of fermentation as well as enzymatic hydrolysis under the influence of ultrasound. The ethanol productivity in SHF mode is higher than SSF. These values could be misnomer indicating SHF process to be more efficient than SSF. However, it should be noted that the productivity in SHF process has been determined on the basis of fermentation period only (and does not include the hydrolysis time). It is also noteworthy to see the trends in maximum cell mass concentration in SHF and SSF process under control and test conditions. Although in both SHF and SSF mode, higher cell mass concentration was seen in test experiments, comparison between SHF and SSF modes shows inverse variations. For control conditions (with mechanical shaking), SHF mode yields higher cell mass concentration, while for test conditions (sonication + mechanical shaking), SSF mode yields higher cell mass concentration. A plausible explanation for this could be given as follows: as stated earlier, ultrasound enhances not only fermentation but also enzymatic hydrolysis. In control experiments, due to slower hydrolysis due to relatively lower mixing induced by mechanical shaking, the instantaneous sugar concentration in the broth at any point of time is likely to be lower than in SHF, which results in lesser cell mass productivity. In test experiments, enhancement in cellulose hydrolysis due to strong micro-convection induced by sonication can significantly increase the instantaneous levels of sugar concentrations in the broth, even higher than that in the SHF mode, which could lead to higher cell mass production. 4. Conclusion
Fig. 3. Micrographs of methylene blue stained yeast cells after completion of fermentation in (A) control experiments and (B) test experiments.
3.2. Effect of sonication on cell viability Methylene blue staining (Figs. 3A and B) and calculation of viability by cell count method revealed that no cell death occurred due to sonication. Yeast cell viability in control and test samples was 83% and 84%, respectively. We attribute this result to small duty cycle (of only 10%) and moderate ultrasound intensity of 1.48 W/cm2.
In this study, we have attempted to get a physical insight into ultrasound induced enhancement in bioethanol synthesis from waste biomass P. hysterophorus using the simultaneous saccharification and fermentation mode. Results of fermentation with and without sonication have been coupled to a model comprising of kinetic and physiological parameters that are characteristics of the mechanism of fermentation. The trends in these parameters give an interesting account of the links between physics of ultrasound and cavitation and biomechanics of ethanol fermentation. It is revealed that strong microturbulence and mild shock waves induced by ultrasound and cavitation augment trans-membrane transport of substrate and products as well as dilution of the toxic substances. These phenomena are manifested in terms of enhancement in cell growth with faster consumption of substrate and reduced cell death rate. Strong convection generated by ultrasound and cavitation also enhances the kinetics of cellulose hydrolysis by cellulase/cellobiase. As a consequence of these, the ethanol productivity as well as cell mass productivity shows a 4 enhancement with sonication, as compared to mechanically stirred system. Nonetheless, the parameters related to intrinsic mechanisms of the cells, such as the inhibition constant for ethanol or the constants for growth and non-growth associated ethanol production have remained unaffected by sonication. The work presented in this paper could give important inputs in further mechanistic research in ultrasound-assisted bioethanol fermentation. The methodology adopted in this work can form a framework for investigations in other fermentation processes employing suitable mathematical models.
3.3. Comparative analysis of fermentation in SHF and SSF modes Acknowledgements Table 2 presents the values of different yardsticks that evaluate performance of fermentation process in SSF mode. Comparison of these values with the SHF mode gives interesting accounts of links between physical effects of sonication and mode of fermentation operation. The higher value of maximum ethanol concentration
Authors are grateful to Prof. Mark T. Holtzapple, Department of Chemical Engineering, Texas A&M University, Texas, United States, for his clarification on HCH-1 model. Authors acknowledge the HPLC and GC (procured through FIST grant SR/FST/ETII-028/2010
256
S. Singh et al. / Ultrasonics Sonochemistry 26 (2015) 249–256
from Department of Science and Technology, Government of India) facilities at Department of Chemical Engineering, Indian Institute of Technology Guwahati. Authors are also thankful to Mr. Akshay Verma for his help in development of the mathematical model for fermentation. Authors also thank anonymous referees of this paper for their meticulous evaluation and constructive criticism. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ultsonch.2015.02. 011. References [1] R. Kataria, S. Ghosh, Saccharification of Kans grass using enzyme mixture from Trichoderma reesei for bioethanol production, Bioresour. Technol. 102 (2011) 9970–9975. [2] R.C. Kuhad, R. Gupta, Y.P. Khasa, A. Singh, Bioethanol production from Lantana camara (red sage): pretreatment, saccharification and fermentation, Bioresour. Technol. 101 (2010) 8348–8354. [3] R. Gupta, S. Kumar, J. Gomes, R.C. Kuhad, Kinetic study of batch and fed-batch enzymatic saccharification of pretreated substrate and subsequent fermentation to ethanol, Biotechnol. Biofuels 5 (2012) 16. [4] M. Galbe, G. Zacchi, A review of the production of ethanol from softwood, Appl. Biotechnol. 59 (2002) 618–628. [5] H.A. Choudhury, S. Chakma, V.S. Moholkar, Mechanistic insight into sonochemical biodiesel synthesis using heterogeneous base catalyst, Ultrason. Sonochem. 21 (2014) 69–181. [6] B.R. Reddy, T. Sivasankar, M. Sivakumar, V.S. Moholkar, Physical facets of ultrasonic cavitational synthesis of zinc ferrite particles, Ultrason. Sonochem. 17 (2010) 416–426. [7] V.S. Nalajala, V.S. Moholkar, Investigations in the physical mechanism of sonocrystallization, Ultrason. Sonochem. 18 (2010) 345–355. [8] T. Sivasankar, V.S. Moholkar, Mechanistic approach to intensification of sonochemical degradation of phenol, Chem. Eng. J. 149 (2009) 57–69. [9] M.M.C.G. Warmoeskerken, P. van der Vlist, V.S. Moholkar, V.A. Nierstrasz, Laundry process intensification by ultrasound, Colloids Surf. A 210 (2002) 277–285. [10] V.S. Moholkar, V.A. Nierstrasz, M.M.C.G. Warmoeskerken, Intensification of mass transfer in wet textile processes by power ultrasound, AUTEX Res. J. 3 (2003) 129–138. [11] B.E. Wood, H.C. Aldrich, L.O. Ingram, Ultrasound stimulates ethanol production during the simultaneous saccharification and fermentation of mixed waste office paper, Biotechnol. Prog. 13 (1997) 232–237. [12] C. Ofori-Boateng, K.T. Lee, Ultrasonic-assisted simultaneous saccharification and fermentation of pretreated oil palm fronds for sustainable bioethanol production, Fuel 119 (2014) 285–291. [13] G.P. Philippidis, D.D. Spindler, C.E. Wyman, Mathematical modeling of cellulose conversion to ethanol by simultaneous saccharification and fermentation process, Appl. Biochem. Biotechnol. 34 (35) (1992) 543–556. [14] J. Shen, F.A. Agblevor, Modeling semi-simultaneous saccharification and fermentation of ethanol production from cellulose, Biomass Bioenergy 34 (2010) 1098–1107.
[15] S. Singh, S. Khanna, V.S. Moholkar, A. Goyal, Comparative assessment of pretreatment strategies for enzymatic saccharification of Parthenium hysterophorus, Appl. Energy 129 (2014) 195–206. [16] S. Singh, S.T.P. Bharadwaja, P.K. Yadav, V.S. Moholkar, A. Goyal, Mechanistic investigation in ultrasound-assisted (alkaline) delignification of Parthenium hysterophorus biomass, Ind. Eng. Chem. Res. 53 (2014) 14241–14252. [17] S. Singh, V.S. Moholkar, A. Goyal, Isolation, identification and characterization of a cellulolytic Bacillus amyloliquefaciens strain SS35 from rhinoceros dung, ISRN Microbiol. Article ID 728134 (2013) 7 pages. [18] S. Singh, V.S. Moholkar, A. Goyal, Optimization of carboxymethylcellulase production from Bacillus amyloliquefaciens SS35, 3 Biotech 4 (2014) 411–424. [19] S. Singh, P.K. Dikshit, V.S. Moholkar, A. Goyal, Purification and characterization of acidic cellulase from Bacillus amyloliquefaciens SS35 for hydrolyzing Parthenium hysterophorus biomass, Environ. Prog. Sustainable Energy (2014), http://dx.doi.org/10.1002/ep.12046. [20] M. Karuppaiya, E. Sasikumar, T. Viruthagiri, V. Vijayagopal, Optimization of process variables using response surface methodology (RSM) for ethanol production from cashew apple juice by Saccharomyces cerevisiae, Asian J. Food Agro-Ind. 3 (2010) 462–473. [21] T. Sivasankar, A.W. Paunikar, V.S. Moholkar, Mechanistic approach to enhancement of the yield of a sonochemical reaction, AIChE J. 53 (2007) 1132–1143. [22] S. Chakma, V.S. Moholkar, Mechanistic features of ultrasonic desorption of aromatic pollutants, Chem. Eng. J. 175 (2011) 356–367. [23] V.S. Moholkar, S.P. Sable, A.B. Pandit, Mapping the cavitation intensity in an ultrasonic bath using the acoustic emission, AIChE J. 46 (2000) 684–694. [24] V.S. Moholkar, M. Huitema, S. Rekveld, M.M.C.G. Warmoeskerken, Characterization of an ultrasonic system using wavelet transforms, Chem. Eng. Sci. 57 (2002) 617–629. [25] N. Nelson, A photometric adaptation of the Somogyi method for the determination of glucose, J. Biol. Chem. 153 (1944) 375–380. [26] M. Somogyi, A new reagent for the determination of sugar, J. Biol. Chem. 160 (1945) 61–68. [27] S. Abd-Aziz, G.S. Hung, M.A. Hassan, M.I.A. Karim, N. Samat, Indirect method for quantification of cell biomass during solid state fermentation of palm kernel cake based on protein content, Asian J. Sci. Res. 1 (2008) 385–393. [28] K. Painting, B. Kirsop, A quick method for estimating the percentage of viable cells in a yeast population, using methylene blue staining, World J. Microbiol. Biotechnol. 6 (1990) 346–347. [29] M.T. Holtzapple, H.S. Caram, A.E. Humphrey, The HCH-1 model of enzymatic cellulose hydrolysis, Biotechnol. Bioeng. 26 (1984) 775–780. [30] R. Toegel, B. Gompf, R. Pecha, D. Lohse, Does water vapor prevent upscaling sonoluminescence?, Phys Rev. Lett. 85 (2000) 3165–3168. [31] G. Moulin, B. Helen, P. Galzy, Inhibition of alcoholic fermentation, Biotechnol. Genet. Eng. Rev. 2 (1984) 365–382. [32] W.L. Sun, W.Y. Tao, Simultaneous saccharification and fermentation of rice straw pretreated by a sequence of dilute acid and dilute alkali at high dry matter content, Energy Source Part A 35 (2013) 741–752. [33] J.S. Jang, Y.K. Cho, G.T. Jeong, S.K. Kim, Optimization of saccharification and ethanol production by simultaneous saccharification and fermentation (SSF) from seaweed, Saccharina japonica, Bioprocess Biosyst. Eng. 35 (2012) 11–18. [34] J.K. Ko, J.S. Bak, M.W. Jung, H.J. Lee, I.G. Choi, T.H. Kim, K.H. Kim, Ethanol production from rice straw using optimized aqueous-ammonia soaking pretreatment and simultaneous saccharification and fermentation processes, Bioresour. Technol. 100 (2009) 4374–4380. [35] Zs. Kadar, Zs. Szengyel, K. Réczey, Simultaneous saccharification and fermentation (SSF) of industrial wastes for the production of ethanol, Ind. Crops Prod. 20 (2004) 103–110.
