- Email: [email protected]
Improvement of butanol production from a hardwood hemicelluloses hydrolysate by combined sugar concentration and phenols removal
Accepted Manuscript Improvement of butanol production from a hardwood hemicelluloses hydrolysate by combined sugar concentration and phenols removal Fatma Mechmech, Hassan Chadjaa, Mohamed Rahni, Mariya Marinova, Najla Ben Akacha, Mohamed Gargouri PII: DOI: Reference:
S0960-8524(15)00676-8 http://dx.doi.org/10.1016/j.biortech.2015.05.012 BITE 14983
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
19 March 2015 6 May 2015 7 May 2015
Please cite this article as: Mechmech, F., Chadjaa, H., Rahni, M., Marinova, M., Akacha, N.B., Gargouri, M., Improvement of butanol production from a hardwood hemicelluloses hydrolysate by combined sugar concentration and phenols removal, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.05.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1
Improvement of butanol production from a hardwood
2
hemicelluloses hydrolysate by combined sugar
3
concentration and phenols removal
4
Fatma Mechmecha,c, , Hassan Chadjaab, Mohamed Rahnib, ,
5
Mariya Marinovac* ,Najla Ben Akachad and Mohamed Gargouria
6
a
7
Technology, National Institute of Applied Sciences and Technology (INSAT) Tunis,
8
University of Carthage, Tunisia
9
b
Biocatalysis and Industrial Enzymes Group, Laboratory of Microbial Ecology and
Centre National en Electrochimie et en Technologies Environnementales, Shawinigan,
10
QC, Canada
11
c
12
Biorefinery, Department of Chemical Engineering, Polytechnique Montréal, Montréal, QC,
13
Canada
14
d
15
Chemical Analysis (INRAP), Biotechno pole Sidi Thabet, Ariana, Tunisia
Research Unit on Energy Efficiency and Sustainable Development of the Forest
Laboratory of Natural Substances, National Institute of Research and Physical and
16 17 18 19 20 21 22
* Corresponding author: [email protected]; fax: +1 514 340 4159
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
23
Abstract
24
The feasibility of using hardwood hemicellulosic pre-hydrolysate recovered from a
25
dissolving pulping process for Acetone-Butanol-Ethanol (ABE) fermentation has been
26
investigated. Dilutions and detoxification methods based on flocculation and nanofiltration
27
were tested to determine the inhibitory concentration of phenolic compounds and to
28
evaluate the efficiency of inhibitors removal on fermentation. Flocculation carried out with
29
ferric sulfate was the most effective method for removal of phenolics (56%) and acetic acid
30
(80%). The impact on fermentation was significant, with an ABE production of 6.40 g/L
31
and 4.25 g/L when using flocculation or combined nanofiltration/flocculation respectively,
32
as compared to a non-significant production for the untreated hydrolysate. By decreasing
33
the toxicity effect of inhibitors, this study reports for the first time that the use of these
34
techniques is efficient to increase the inhibitory concentration threshold of phenols, from
35
0.3 g/L in untreated hydrolysate, to 1.1 g/L in flocculated and in nanofiltrated and
36
flocculated hydrolysates.
37
Highlights:
38
39
Hardwood pre-hydrolysates from dissolving pulp mill were used for ABE fermentation.
40
Flocculation was the most efficient method for inhibitors removal.
41
Detoxification increased the phenolics inhibitory level on ABE production up to 1.1
42
g/L.
43
Key words: Bio-butanol; flocculation; nanofiltration; phenolics; wood biomass
44
hydrolysate
45 2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
46 47
1. Introduction During the last decades, the growing environmental and economic concerns have
48
stimulated a worldwide adoption of new government policies and directives for the
49
production of renewable, biomass-derived fuels (Qureshi et al., 2013; Stoklosa & Hodge,
50
2014). Bio-butanol, a product of the ABE (Acetone-Butanol-Ethanol) fermentation, has
51
recently emerged as a superior biofuel and is considered as an attractive alternative to
52
conventional jet fuels, which may be an important factor in the success of the biofuel
53
industry. Its properties are similar to those of the conventional gasoline (Baral & Shah,
54
2014) and it has several advantages over ethanol: higher energy content, lower volatility,
55
less hydroscopic and less corrosive (Lee et al., 2008). Most important from an economic
56
perspective is the fact that its biological production by Clostridia species is largely
57
influenced by the price of the fermentation substrate (Green, 2011). For these reasons, the
58
use of lignocellulosic biomass, such as wood chips or wood wastes, as a feedstock to
59
produce bio-butanol has prompted the interest of the research community. Governments in
60
countries with a mature forestry sector, such as Canada and the Scandinavian countries,
61
have promoted the development and the implementation of the forest biorefinery and its
62
integration into existing pulp and paper (P&P) facilities, the largest global consumers of
63
woody biomass. The P&P industry in these countries has been facing for some time a
64
difficult economic situation (Marinova et al., 2009) and the concept of the integrated forest
65
biorefinery (IFBR) has been identified as a means to diversify its products portfolio,
66
increase its revenues, and regain profitability by manufacturing of a new value-added
67
bioproducts (Adriaan van et al., 2011; Huang et al., 2010). In a Kraft dissolving pulp
68
process, the hemicelluloses are extracted from wood chips prior to the pulping step by
69
steam or hot water partial hydrolysis. This hemicelluloses stream can be used as a
70
feedstock to an adjunct biorefinery plant, to be converted by hydrolysis into C5 and C6 3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
71
sugars, which will in turn be transformed into biofuels and bioproducts such as ethanol,
72
butanol or furfural (Marinova et al., 2014). However, the hemicelluloses pre-hydrolysate
73
contains other wood components besides the sugar oligomers, which are potential
74
fermentation inhibitors; those compounds are: acetic acid, furfural, hydroxymethylfurfural
75
and phenolic derivatives. Detoxification methods of the pre-hydrolysate solution prior to
76
ABE fermentation have been investigated by several authors; they include activated carbon
77
(Kudahettige-Nilsson et al., 2015) and alkaline peroxide treatment (Wang & Chen, 2011).
78
The removal of lignin degradation products by flocculation with primary focus on phenolic
79
compounds has also received much interest from the scientific community. It is known that
80
some phenolics found in lingocellulosic hydrolysates are colloidal particles with high
81
anionic charge that cannot be sedimented easily. During flocculation phenomena, a two-
82
step mechanism occurs. First, the flocculating agents or polyelectrolytes characterized by
83
an ionic charge form bridges between the colloidal particles. These linked particles in turn
84
form, through a mechanism of charge neutralization, agglomerates or patches which can be
85
removed from the solution more easily (Burke et al., 2011; Yasarla & Ramarao, 2012;
86
Yasarla & Ramarao, 2013). It has been reported that this method is effective for inhibitors
87
removal (Ajao et al., 2015b; Duarte et al., 2010) but to the best of our knowledge no
88
fermentation results on the flocculated material have yet been published. It is speculated
89
that the fermentation yield of butanol can be enhanced when combining the detoxification
90
step with the sugar concentration step by evaporation or membrane separation.
91
Nanofiltration may be a feasible membrane concentration technique for the recovery of
92
hemicelluloses from hydrolysates thanks to its selectivity and low energy requirement
93
(Huang et al., 2008; Wei et al., 2014). Ajao et al. (2015a) proposed recently a new process
94
configuration for the detoxification and concentration of a pre-hydrolysate from a
4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
95
dissolving pulp mill by combining nanofiltration and flocculation in a single step, prior to
96
the production of ethanol by fermentation.
97
The work presented herein extends the scope of this prior work and investigates the
98
applicability of flocculation or combined flocculation/nanofiltration to enhance the ABE
99
fermentation. Specifically, the objectives of this work were to:
100
Establish the feasibility of using an untreated or detoxified hemicellulosic wood
101
hydrolysate from a Kraft dissolving pulp mill as a carbon source for the ABE
102
fermentation by Clostridium acetobutylicum;
103
Demonstrate the impact of the proposed detoxification methods on the butanol
104
production (flocculation was first applied alone and later combined to nanofiltration for
105
inhibitors removal);
106 107
Determine the inhibitory concentration of phenolic compounds in each hydrolysate by comparing ABE production, fermentation yields and productivities.
108 109
2. Materials and methods
110
2.1. Microorganism, culture maintenance and inoculum preparation
111
Clostridium acetobutylicum ATCC 824 was obtained from American Type Culture
112
Collection (ATCC) and was cultured in sterilized RCM (Reinforced Clostridium Medium)
113
under anaerobic conditions at 37°C for 18-20 hours until an Optical Density (OD600) of
114
1.9-2. A stock culture was then prepared in 30% glycerol and stored at 80°C. For inoculum
115
preparation, RCM medium consisting of (in g/L): tryptose 10, beef extract 10, glucose 5,
116
sodium chloride 5, yeast extract 3, sodium acetate 1, soluble starch 0.5 and L-cysteine-HCl
117
0.5 was used. The medium was purged with a gas mixture (80% N2 and 20% CO2) to
118
dissolve oxygen and then autoclaved at 121°C for 15 minutes. The culture (inoculums) was
119
inoculated with stock culture and allowed to grow for 16-18 hours at 37°C under shaking 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
120
conditions at 110 rpm, until it was ready to be inoculated into the butanol production
121
medium at an OD600 between 0.6 and 0.9.
122 123 124
2.2. Preparation and treatment of hydrolysates Before testing the fermentability of the wood hydrolysate, different detoxification
125
approaches were evaluated (Figure 1).
126
2.2.1. Pre-hydrolysis: steam and hot water
127
A mixture of aspen (60%) and maple (40%) wood furnish obtained from an operating
128
Kraft pulping mill located in eastern Canada was used in this study. The pre-
129
hydrolysate rich in hemicelluloses was generated in a pilot digester at the FPInnovations
130
facilities (Pointe-Claire, Canada) using steam and hot water, whereas the cellulosic fraction
131
was directed to dissolving pulp production as described previously by Ajao et al. (2015b).
132
A 12L sample of pre-hydrolysate was withdrawn and shipped to the Centre National en
133
Electrochimie et en Technologies Environnementales (CNETE), where the experimental
134
program was conducted.
135
2.2.2. Sulfuric acid hydrolysis
136
To convert the polysaccharides into fermentable monomeric sugars, the pre-
137
hydrolysate was completely hydrolyzed with 1.5% wt/wt sulfuric acid at 121°C for 60 min.
138
The hydrolysate was filtered and sterilized by passing through a 1.5 µm filter. Untreated
139
and treated hydrolysates were neutralized with Ca(OH)2, from an initial pH of 1.34 to a
140
final pH of 6.5.
141
2.2.3. Hydrolysate detoxification
142
The procedure developed by Ajao et al. (2015a) in a previous work on the
143
production of ethanol from a similar hemicellulosic pre-hydrolysate extracted from wood
144
chips, was used for the hydrolysate detoxification in the current work. 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
145
Coagulation-Flocculation. Flocculation experiments were carried out in jar tests using
146
ferric sulfate Fe2(SO4)3 as flocculating agent. A volume of 1 L of a wood hydrolysate was
147
mixed with a 200 g/L of ferric sulfate solution to reach a ratio [Fe]/[phenols] of 1 and the
148
mixture was homogenized by agitation at 150 rpm. The pH was then adjusted to 6.8 with
149
Ca(OH)2 and the agitation was resumed for 30 min at 50 rpm. At the end of the agitation
150
period, the mixture was left to rest for 2 hours so that a sedimentation occurs. The
151
supernatant was then filtered with a 1.5 µm filter and used as a carbon source to carry out
152
the fermentation studies.
153
Concentration of hydrolysate by nanomembrane filtration. The pre-hydrolysate,
154
containing polysaccharides, was concentrated by naonomembrane filtration using the
155
organic membrane NF270 (MWCO 200-400 Da). A laboratory scale cross-flow flat-sheet
156
membrane test unit (SEPA CF II, GE Osmonics) was used in a batch mode and the
157
permeate stream collected in a small vessel. During the concentration experiments, 2.35 L
158
of permeate were withdrawn from an initial volume of 3.50 L of pre-hydrolysate to reach a
159
volumetric concentration factor of 3.43. Before and after concentration, samples from the
160
feed tank were taken in order to analyze the amounts of sugars, total phenols, acetic acid,
161
furfural and HMF.
162 163
2.3. Fermentation medium and conditions
164
The fermentations experiments were carried out in 250 mL screw capped Schott
165
bottles with 200 mL of fermentation medium (xylose 60 g/L and yeast extract 5g/L) (Choi
166
et al., 2012; Qureshi & Blaschek, 1999). Untreated or flocculated hydrolysates containing
167
inhibitory chemicals were used as carbon sources. The phenolic compounds concentrations
168
of the hydrolysates were first adjusted to stepwise decreasing levels by serial dilutions and
169
the final rate of sugars was raised to 60 g/L of xylose. A gas mixture consisting of 80% 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
170
N2and 20% CO2 was sparged for 5 min. through the samples to remove oxygen, before
171
they were sterilized at 121°C for 15 min. After cooling, loosely capped bottles were
172
transferred to anaerobic jars containing envelopes Gas Pak (BD Gas PakTMEZ Anaerobe
173
Container System) with indicators (BD BBLTM Dry Anaerobic Indicator Strips) for 48h to
174
create anaerobic conditions in the mixtures. Prior to inoculation, 2 mL of filtered sterilized
175
stock solution consisting of buffer (KH2PO4, 50 g/L; K2HPO4,50 g/L; ammonium acetate,
176
220 g/L), vitamins (para-aminobenzoic acid, 0.1 g/L; thiamin, 0.1 g/L; Biotin, 0.001 g/L)
177
and minerals (MgSO4.7H2O, 20 g/L; MnSO4.H2O, 1 g/L; FeSO4.7H2O, 1 g/L; NaCl, 1
178
g/L) were added to supply the equivalent nutriment concentration level as a synthetic
179
medium (Ezeji et al., 2007). Finally, 5% (v/v) of active growth inoculums was added and
180
the incubation was performed at 37°C under static conditions without pH control. Samples
181
were taken at regular intervals for optical density measurements. Sugars and ABE analysis
182
for all fermentation experiments were conducted at least in duplicates. During the course of
183
fermentation, 5 mL samples were periodically withdrawn to measure pH, OD600, residual
184
xylose, alcohols, acetic acid and butyric acid.
185 186
2.4. Analytical methods
187
Spectrophotometric analysis. The growth of C. acetobutylicum ATCC 824 was monitored
188
by measuring OD600 using a UV-visible spectrophotometer (Pharmacia Biotech
189
Novaspec®II).Total phenols concentration was measured by an UV-visible colorimetric
190
method using Folin-Ciocalteu reagent method from Singleton and Rossi (1965).
191
GC analysis. Butanol, acetone and ethanol were measured by gas chromatography (GC
192
7890A, Agilent Technologies), equipped with a flame ionization detector (FID, H2 flow
193
rate: 30 mL/min; air flow rate: 2.23 mL/min) and an OV 624 capillary column.
194 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
195
HPLC Analyses.
196
The concentration of simple sugars was determined by high performance liquid
197
chromatography (HPLC) using an Agilent 1260 system, a Refractive Index Detector
198
and an EC Nucleodur RP-NH2 column (250 mm x 4.6 mm, 5 μm ). Samples separation
199
was performed at 40°C with a mixture of acetonitrile and deionized water (ratio 75:25)
200
as a mobile phase and at a flow rate of 1.5 mL/min.
201
ABE productivity was calculated as total ABE produced in g/L, divided by the
202
fermentation time and is expressed as g/L.h. The ABE yield was calculated as the total
203
ABE produced, divided by the total sugars used and is expressed as g/g.
204
HMF and furfural were also analyzed by HPLC (Agilent Technologies, Germany)
205
using a 280 nm diode array detector (DAD) and a Nucleosil C18 (150 x4.6 mm)
206
Column. The inhibitors were eluted with a mixture of acetonitrile, deionized water and
207
acetic acid (ratio 84:15:1) at a flow rate of 0.8 mL/min.
208
Organic acids concentrations were determined using the same HPLC with a diode array
209
detector. Separation was carried out using Interstil ODS-3 column (150 x4.6 mm) at
210
40°C. The mobile phase was a mixture of 99% KH2PO4 at pH 2.8, 1% acetonitrile and
211
a flow rate of 1.25 mL/min.
212
Syringaldehyde, vanillin and gallic acid were analyzed by HPLC (Agilent
213
Technologies, Germany) using a Nucleosil C18 (150 x 4.6 mm) and a DAD at 313 nm
214
and 280 nm. The separation was achieved with a mixture of 15% acetonitrile – 85%
215
(0.1% phosphoric acid in water).
216 217 218 219 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
220
3. Results and discussion
221
3.1. Removal of phenolic compounds
222
During the lignocellulosic biomass pre-hydrolysis and hydrolysis, the conversion of
223
sugars from oligomeric to monomeric forms occurs together with the generation of
224
chemical compounds, including sugar and lignin degradation products having different
225
inhibitory activity levels (Baral & Shah, 2014; Ben Chaabane & Marchal, 2013). The
226
untreated hydrolysate used in this study contained 29.7 g/L of fermentable sugars, xylose
227
being the most abundant one (80% by weight of total sugars).The concentration of
228
potential fermentation inhibitors were: 10.7 g/L acetic acid, 3.8 g/L phenolic compounds,
229
0.1 g/L HMF and 0.18 g/L furfural. Additional parameters are given in Table 1. At these
230
levels, the inhibitory profile of the fermentation medium was critical and no growth of C.
231
acetobutylicum, nor butanol production was observed; results from this preliminary phase
232
of the work are not given herein. In order to decrease the inhibitory effect exerted by these
233
compounds, the hydrolysate was subjected to flocculation to remove the phenolics
234
produced by the degradation of lignin. Flocculation was achieved by addition of ferric
235
sulfate and the obtained results showed that this method was effective to remove 56.6% of
236
the phenolic compounds with a high sugar recovery (26.7 g/L). Those results also
237
demonstrate that flocculation does not remove significantly the fermentable sugars (Table
238
1). In fact, flocculation can preferentially precipitate out the colloidal fraction containing
239
lignin and lignin derived compounds (Duarte et al., 2010). The flocculation also caused the
240
removal of other inhibitors from the hydrolysate, 79.4% of acetic acid, 20% of HMF and
241
5.55% of furfural. Those results are in good agreement with the removal percentage found
242
by Ajao et al. (2015a). In addition, it can also be concluded that the removal of such
243
inhibitors, generally present in the solution, can be largely attributed to the use of lime for
244
pH adjustment as well as to the filtration (Table1). 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
245
The concentration of sugars obtained by pre-hydrolysis of wood chips is low and it is
246
not economically feasible to use this medium for microbial fermentation to produce high
247
value products (Qi et al., 2011). A second method combining concentration and
248
detoxification by nanofiltration was added to the procedure to increase the sugar
249
concentration. The acid hydrolysate thus produced was further detoxified by flocculation.
250
As shown in Table 2, when the NF270 membrane was used in the nanofiltration step,
251
phenols were the only inhibitors affected and their concentration ratio was increased by a
252
factor of 2.97, while it remained practically unchanged for acetic acid, furfural and HMF.
253
Qi et al. (2011) have also observed a negative retention of furfural when using NF270 to
254
concentrate a model solution of xylose, glucose and furfural. After flocculation, the
255
amount of acetic acid decreased by 77.7%, whereas the concentration of HMF and furfural
256
remained high. The detoxification strategy combining pre-hydrolysate concentration and
257
hydrolysate flocculation was effective in concentrating the sugars, but not in removing the
258
inhibitors.
259
As shown in Table 3, the dilution effect of various hydrolysates generated in the
260
course of the experimental program was investigated. It should be noted that these
261
dilutions were undertaken in order to determine the exact inhibitory concentration of the
262
total phenolic compounds in each hydrolysate and not as a detoxification method to be
263
considered at an industrial scale. This approach caused a reduction in sugar content that
264
may negatively affect the ABE production. Therefore, the hydrolysates were supplemented
265
with xylose prior to fermentation to raise the final sugar concentration to approximately 60
266
g/L, so that the inhibitors concentration remains the only parameter to be considered.
267 268 269 11
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
270 271
3.2. ABE fermentation of untreated hydrolysate In order to investigate the possible upgrade of the pre-hydrolysate obtained from a
272
Kraft dissolving pulp mill and to test its ability to be fermented by C. acetobutylicum for
273
ABE production, fermentation experiments were conducted using different hydrolysates,
274
as illustrated in Figure 1. A control experiment was first performed using 60 g/L of xylose
275
as model substrate and C. acetobutylicum ATCC 824 to benchmark the fermentation tests.
276
As shown in Figure 2A, during the acidogenic phase of the Clostridium growth which
277
lasted 72h, acetic and butyric acids were produced by fermentation. The final
278
concentrations of those two acids reached 2.66 g/L and 1.40 g/L, respectively. During the
279
acidogenic phase, the pH decreased to 4.8, which is considered as an optimum value,
280
ensuring the transition from the acidogenic to the solvatogenic phase. The solvatogenic
281
phase resulted in the production of 5.82 g/L ABE with 3.95 g/L of butanol (Figure 2B),
282
corresponding to an ABE productivity of 0.04 g/L.h. After 144h of fermentation, 25.2 g/L
283
of xylose were consumed; the corresponding ABE yield was 0.23 g/g and the acetic acid
284
production 2.4 g/L (Figure 2A). The obtained kinetic parameters are higher than those
285
previously reported by Jiang et al. (2014) from a study in which xylose was used as a
286
carbon source to produce butanol by C. acetobutylicum ATCC 824 under uncontrolled pH
287
conditions. The reported ABE productivity was 0.023 g/L.h and the yield 0.023 g/g.
288
Following the control experiment, fermentation was conducted using the untreated
289
wood hydrolysate as carbon source with different dilutions (Figure 3A). After 144h of
290
fermentation, an insignificant ABE production was observed at a phenolic concentration
291
higher than 0.3 g/L. This result may be linked to an accumulation of acetic acid of 10.1 g/L
292
which was not re-assimilated to produce solvents in the presence of 2.3 g/L of phenols. As
293
illustrated on Figure 3A, at phenolic concentrations of 0.17 g/L or 0.3 g/L, the total ABE
294
production was about 0.8 g/L with around 0.4 g/L of butanol. This production level is low 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
295
compared to the level achieved in the control experiment where xylose was used as carbon
296
source, suggesting that the hydrolysate remains toxic for the culture even after a 13-fold
297
dilution. The inhibition effect can be primarily attributed to the presence of totally soluble
298
phenolics that have been reported to be more toxic than furfural and HMF, even at low
299
concentrations (Baral & Shah, 2014). These compounds are known to undergo partitioning
300
in the biological membranes, thus reducing their permeability and their ability to play the
301
role of selective barriers and enzyme matrices (Ibraheem & Ndimba, 2013; Palmqvist &
302
Hahn-Hägerdal, 2000). It has also been shown in previous studies that low concentrations
303
of HMF and furfural produce a synergetic effect on the ABE fermentation (Ezeji et al.,
304
2007; Qureshi et al., 2012; Zhang et al., 2012). This would support the assumption that in
305
our work HMF and furfural cannot be considered as inhibition factors.
306
In order to investigate the monomers inhibitory concentration, major phenolic
307
compounds contained in the untreated hydrolysate were analyzed. As shown in Table 4,
308
the amount of vanillin was about 0.042 g/L, syringaldehyde 0.19 g/L, and gallic acid 0.047
309
g/L. The inhibitory concentrations in this work were much lower than those obtained by
310
Ezeji et al. (2007) (syringaldehyde 0.3-1 g/L) and Cho et al. (2009) (0.17 g/L for vanillin
311
and syringaldehyde) who studied the inhibitory effect of model phenolic compounds
312
solutions. This was assigned to the synergy that can occur due to the coexistence of
313
different inhibitory compounds. In fact, as explained by Zha et al. (2014) the threshold
314
toxicity of a specific compound present in a biomass hydrolysate can be much lower than
315
that evaluated in a synthetic medium.
316 317 318 319
3.3. Effect of hydrolysate flocculation on the ABE fermentation In the next phase of the experimental program, the hydrolysate was subjected to flocculation prior to being used for butanol fermentation at different dilutions, as described 13
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
320
previously. As illustrated on Figure 4A and 4B, for the same phenolic concentrations, the
321
flocculated hydrolysate shows a reduction of the lag phase for the growth of Clostridium,
322
relatively to the untreated hydrolysate, suggesting that the microorganism can easily adapt
323
to the fermentation medium. Avoiding extended lag phases would constitute a significant
324
advantage when scaling up the process to an industrial size.
325
As shown in Figure 3B, an increase of ABE production was observed, compared to the
326
control synthetic medium (5.82 g/L ABE) without inhibitors during a 144h fermentation; a
327
maximum solvent concentration of 6.00 and 6.40 g/L was reached in the presence of 0.3
328
and 0.17 g/L of phenols, respectively. These values are 88% higher than those obtained
329
with the untreated hydrolysate with the same phenolic compounds concentration. After 2.5
330
and 1.5 fold dilution, achieving 0.6 g/L and 1.1 g/L of phenols, the respective ABE
331
production was 2.6 g/L and 0.6 g/L. These results are important in view of the lack of
332
production observed with untreated hydrolysate at the same phenolics concentrations. An
333
accumulation of acetic acid, reaching concentrations of 6.7 g/L and 7.4 g/L for the same
334
concentrations of phenolics, was observed in the experiments with low ABE yield, as
335
illustrated on Figure3B. When using undiluted flocculated hydrolysate with a total
336
phenolics concentration of 1.65 g/L, the acetic acid was as high as 10 g/L with no ABE
337
production. In fact, for high concentrations of acetic acid, the culture does not enter in the
338
solvatogenic phase, because the accumulation of the acid is detrimental for the
339
solvatogenic clostridia nutrient uptake and the cell growth (Ezeji et al., 2010). The reason
340
behind this behavior may be related to the fact that phenolic compounds interfere with the
341
metabolic pathway by preventing the re-assimilation of the organic acids into acetyl-CoA
342
and butyryl-CoA and further conversion into ethanol and butanol (Baral & Shah, 2014).
343
Interestingly, the flocculation reduced the level of inhibition exerted by phenols on
344
cell growth and enhanced the ABE production. Moreover, the flocculation removal 14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
345
efficiency on monomeric phenols was different from that observed on total phenols. As
346
shown in Table 4, where three monomeric phenols are represented, the flocculation
347
achieved a removal efficiency of 32% for vanillin, 55% for syringaldehyde and 100% for
348
gallic acid, as compared to 56.6 % for total phenols. These results demonstrate that
349
flocculation removes phenolic compounds in different proportions as function of their
350
ability to form aggregates with ferric sulfate. They also suggest that the proposed
351
detoxification method may contribute to the removal of other unidentified or non-analyzed
352
monomeric phenols that may have higher inhibitory effect. Thus, at the same phenols
353
concentration, an untreated and a flocculated hydrolysate do not contain the same
354
proportions of phenols monomers and as a consequence they will not present the same
355
inhibitory effect. Taken together, these reasons could explain the raise of the total phenols
356
inhibition levels from 0.3 g/L for an untreated hydrolysate to 1.1 g/L for a flocculated
357
hydrolysate (Figure 3A and 3B). This further supports the conclusion that flocculation is an
358
effective detoxification method for the removal of phenolic compounds.
359 360
3.4. Effect of combining pre-hydrolysate concentration and hydrolysate flocculation
361
on the ABE fermentation
362
In the final stage of the experimental program, fermentation experiments were
363
conducted using the hydrolysate produced by the third detoxification technique, illustrated
364
in Figure 1. In this case, the pre-hydrolysate was first concentrated using the organic
365
membrane NF270 to increase the sugar concentration by 2.17 folds (Table 2) and was
366
subsequently subjected to flocculation under the same conditions. Since it was not possible
367
to get reliable results of OD at 600 nm in the case of C. acetobutylicum growth, because of
368
the high turbidity of the culture medium, the sugar consumption and the solvent production
369
were used as criteria for comparison with the other experiments, as recommended by (Zha 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
370
et al., 2014). Figure 3C shows that by applying this detoxification strategy, the ABE
371
production was about 3.77 g/L and 4.25 g/L for phenolics respective concentrations of 0.3
372
and 0.17 g/L. These values are approximately 35% lower than those obtained with the
373
flocculated hydrolysate at the same phenolics concentrations, despite the fact that the
374
flocculation efficiency was approximately the same in both cases (58% VS 56.6%).
375
Thereby, the fact of overly diluting the hydrolysate in the seconf detoxification strategy to
376
achieve the same concentration of phenolic compounds may reduce the concentrations of
377
other compounds reported to be beneficial for the fermentation. Similar level of inhibition
378
was obtained at phenolics concentration of 1.1 g/L, an accumulation of 7.74 g/L of acetic
379
acid and ABE production of 0.2 g/L, as with a flocculated hydrolysate. By applying
380
nanofiltration, the main objective was to concentrate the pre-hydrolysate in order to
381
increase sugar concentration, to reduce the cost of the acid hydrolysis step (amount of acid
382
added, volume to be heated). During concentration, nanofiltraion has also allowed the
383
removal of acetic acid, HMF and furfural. However, its combination with flocculation was
384
not more efficient than flocculation alone. It is then recommended to apply another
385
technique before or after flocculation to improve the inhibitors removal efficiency. The
386
economic viability of the overall process should be taken into account.
387
A comparison of productivities and yields for different fermentation conditions are
388
shown in Figure 5. At phenolics concentrations of 0.17 g/L and 0.3 g/L, the fermentation
389
experiment using untreated hydrolysate did not reach more than 0.005 g/L.h of
390
productivity with sugar consumptions of 9.40 g/L and 11.9 g/L (Figure 5A) and yields of
391
0.08 and 0.07 g/g, respectively (Figure 5C). The corresponding productivities were
392
improved to 0.041 g/L.h and 0.044 g/L.h using flocculated hydrolysate, and to 0.023 g/L.h
393
and 0.03 g/L.h using combined nanofiltration and flocculation (Figure 5B). Improvements
394
of 65% and 74% in ABE yields could be achieved from flocculated hydrolysate with a 16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
395
xylose consumption of 24.2 g/L and 28.4 g/L. An increase of ABE yield of 3.12-folds and
396
3.57-folds with the third detoxified hydrolysate was also observed for the same phenolics
397
concentration.
398 399 400
3.5. Comparison to other studies Various lignocellulosic feedstocks for bio-butanol production via ABE fermentation
401
have been used in previous studies and reported in the scientific literature but few of these
402
studies discuss the use of wood hydrolysates obtained from pulp and paper mills in
403
operation, thus emulating the concept of the integrated forest biorefinery. A study of the
404
ABE fermentation by C. beijerinckii of wood hydrolysate with a sugar mixture of glucose,
405
xylose and arabinose has been reported by Lu et al. (2013). The authors have investigated
406
the use of various techniques for the removal of inhibitors: adsorption by activated
407
charcoal achieving concentration of 9.98 g/L of total ABE, resin adsorption (11.4 g/L of
408
total ABE), and combined resin adsorption and gas stripping (17.7 g/L of total ABE).
409
However, the work presented herein seems to be the first to consider the validation of
410
flocculation and combined nanofiltration and flocculation as hemicellulosic wood
411
hydrolysate detoxification methods, prior to the ABE fermentation. Despite the fact that
412
the work was done under uncontrolled pH conditions, the results are close to those
413
obtained recently by Kudahettige-Nilsson et al. (2015). They have reported that at a
414
phenolic concentration of 0.6 g/L, the batch fermentation of a detoxified hardwood
415
hydrolysate using active carbon with pH control produced 2.2 g/L of ABE. Their process
416
resulted in a yield of 0.13 g/g with a solvent production efficiency of 33%, as compared to
417
a yield of 0.17g/g with a production efficiency of 43% obtained in the present study when
418
fermenting a flocculated hydrolysate at the same phenolics concentration (Figure 5C). The
419
pH control of the medium is considered as a key factor for the ABE fermentation 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
420
enhancement (Jiang et al., 2014). This further demonstrates the effectiveness of
421
flocculation as a detoxification method and the feasibility of using a hemicellulosic wood
422
hydrolysate issued from dissolving Kraft pulp mill as an alternative renewable feedstock
423
for bio-butanol production via ABE fermentation. Further optimization work should be
424
undertaken to ensure detoxification of the hydrolysate to the feasible thresholds for the
425
ABE fermentation. The economic viability of the proposed process should also be
426
validated. Finally, means to reduce the cost of the fermentation medium should be
427
investigated. In this perspective, the use of low cost agricultural residues as a nitrogen
428
source in the stock solution is an opportunity worthy of further investigation.
429 430 431
4. Conclusions In this work, it has been demonstrated that the ABE fermentation of hemicellulosic
432
wood hydrolysates has been improved using dilution and flocculation or combined
433
nanofiltration and flocculation as downstream detoxification methods. Within the range of
434
conditions covered at a laboratory scale, the applied methods enhanced inhibitory levels
435
from 0.3 to 1.1 g/L of total phenols. Without dilutions, the residual total phenols
436
concentrations in the detoxified hydrolysates remain higher than the acceptable
437
fermentation thresholds. Therefore, it is recommended to investigate the use of different
438
hydrolysis methods generating less inhibitors, as well as combining flocculation with other
439
detoxification techniques such as enzymatic treatment.
440 441 442 443 444 18
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
445
Acknowledgments:This work was supported by a grant from the College-University I2I
446
Program of the Natural Sciences and Engineering Research Council of Canada (grant
447
number 437803-12).The authors wish to express their gratitude to the Centre National en
448
Electrochimie et Technologies Environnementales (Shawinigan, QC) which hosted the
449
experimental work presented in this manuscript and provided staff support and equipment
450
and, to FPInnovations (Pointe-Claire, QC) which supplied the hemicelluloses pre-
451
hydrolysate samples. The authors are also thankful to Professor Jean Paris from the
452
Department of Chemical Engineering at Polytechnique Montréal for his valuable
453
contributions to this work and for reviewing the manuscript.
19
References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1.
2.
3.
4. 5.
6.
7.
8.
9.
10.
11. 12.
13. 14.
15.
16.
17. 18.
Ajao, O., Le Hir, M., Rahni, M., Marinova, M., Chadjaa, H., Savadogo, O., 2015a. Concentration and Detoxification of Kraft Prehydrolysate by Combining Nanofiltration with Flocculation. Ind. Eng. Chem. Res. 54(3), 1113-1122. Ajao, O., Rahni, M., Marinova, M., Chadjaa, H., Savadogo, O., 2015b. Retention and flux characteristics of nanofiltration membranes during hemicellulose prehydrolysate concentration. Chem. Eng. J. 260, 605-615. Baral, N., Shah, A., 2014. Microbial inhibitors: formation and effects on acetone-butanolethanol fermentation of lignocellulosic biomass. Appl. Microbiol. Biotechnol. 98(22), 91519172. Ben Chaabane, F., Marchal, R., 2013. Upgrading the Hemicellulosic Fraction of Biomass into Biofuel. Oil Gas Sci. Technol. – Rev. IFP Energies nouvelles 68(4), 663-680. Burke, D.R., Anderson, J., Gilcrease, P.C., Menkhaus, T.J., 2011. Enhanced solid–liquid clarification of lignocellulosic slurries using polyelectrolyte flocculating agents. Biomass Bioenergy 35(1), 391-401. Cho, D., Lee, Y., Um, Y., Sang, B.-I., Kim, Y., 2009. Detoxification of model phenolic compounds in lignocellulosic hydrolysates with peroxidase for butanol production from Clostridium beijerinckii. Appl. Microbiol. Biotechnol. 83(6), 1035-1043. Choi, S., Lee, J., Jang, Y.-S., Park, J., Lee, S., Kim, I., 2012. Effects of nutritional enrichment on the production of acetone-butanol-ethanol (ABE) by Clostridium acetobutylicum. J. Microbiol. 50(6), 1063-1066. Duarte, G.V., Ramarao, B.V., Amidon, T.E., 2010. Polymer induced flocculation and separation of particulates from extracts of lignocellulosic materials. Bioresour. Technol. 101(22), 85268534. Ezeji, T., Milne, C., Price, N., Blaschek, H., 2010. Achievements and perspectives to overcome the poor solvent resistance in acetone and butanol-producing microorganisms. Appl. Microbiol. Biotechnol. 85(6), 1697-1712. Ezeji, T., Qureshi, N., Blaschek, H.P., 2007. Butanol production from agricultural residues: Impact of degradation products on Clostridium beijerinckii growth and butanol fermentation. Biotechnol. Bioeng. 97(6), 1460-1469. Green, E.M., 2011. Fermentative production of butanol—the industrial perspective. Curr. Opin. Biotechnol. 22(3), 337-343. Huang, H.-J., Ramaswamy, S., Al-Dajani, W.W., Tschirner, U., 2010. Process modeling and analysis of pulp mill-based integrated biorefinery with hemicellulose pre-extraction for ethanol production: A comparative study. Bioresour. Technol. 101(2), 624-631. Huang, H.-J., Ramaswamy, S., Tschirner, U.W., Ramarao, B.V., 2008. A review of separation technologies in current and future biorefineries. Sep. Purif. Technol. 62(1), 1-21. Ibraheem, O., Ndimba, B.K., 2013. Molecular Adaptation Mechanisms Employed by Ethanologenic Bacteria in Response to Lignocellulose-derived Inhibitory Compounds. Int. J. Biol. Sci. 9(6), 598-612. Jiang, W., Wen, Z., Wu, M., Li, H., Yang, J., Lin, J., Lin, Y., Yang, L., Cen, P., 2014. The Effect of pH Control on Acetone–Butanol–Ethanol Fermentation by Clostridium acetobutylicum ATCC 824 with Xylose and d-Glucose and d-Xylose Mixture. Chin. J. Chem. Eng. 22(8), 937-942. Kudahettige-Nilsson, R.L., Helmerius, J., Nilsson, R.T., Sjöblom, M., Hodge, D.B., Rova, U., 2015. Biobutanol production by Clostridium acetobutylicum using xylose recovered from birch Kraft black liquor. Bioresour. Technol. 176(0), 71-79. Lee, S.Y., Park, J.H., Jang, S.H., Nielsen, L.K., Kim, J., Jung, K.S., 2008. Fermentative butanol production by clostridia. Biotechnol. Bioeng. 101(2), 209-228. Lu, C., Dong, J., Yang, S.-T., 2013. Butanol production from wood pulping hydrolysate in an integrated fermentation–gas stripping process. Bioresour. Technol. 143(0), 467-475. 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
19. Marinova, M., Mateos-Espejel, E., Jemaa, N., Paris, J., 2009. Addressing the increased energy demand of a Kraft mill biorefinery: The hemicellulose extraction case. Chem. Eng. Res. Des. 87(9), 1269-1275. 20. Marinova, M., Perrier, M., Paris, J., 2014. Implementation of a Forest Biomass-based Biofuel Industry: A Canadian Experience. in: International Conference on Renewable Energies and Power Quality (ICREPQ’14), Vol. No.12, Renew. Energy Power. Qual. J. (RE&PQJ). Cordoba (Spain), 8th to 10th April, 2014. 21. Palmqvist, E., Hahn-Hägerdal, B., 2000. Fermentation of lignocellulosic hydrolysates. II: inhibitors and mechanisms of inhibition. Bioresour. Technol. 74(1), 25-33. 22. Qi, B., Luo, J., Chen, X., Hang, X., Wan, Y., 2011. Separation of furfural from monosaccharides by nanofiltration. Bioresour. Technol. 102(14), 7111-7118. 23. Qureshi, N., Blaschek, H.P., 1999. Butanol recovery from model solution/fermentation broth by pervaporation: evaluation of membrane performance. Biomass Bioenergy 17(2), 175-184. 24. Qureshi, N., Bowman, M.J., Saha, B.C., Hector, R., Berhow, M.A., Cotta, M.A., 2012. Effect of cellulosic sugar degradation products (furfural and hydroxymethyl furfural) on acetone– butanol–ethanol (ABE) fermentation using Clostridium beijerinckii P260. Food Bioprod. Process. 90(3), 533-540. 25. Qureshi, N., Liu, S., Ezeji, T.C., 2013. Cellulosic Butanol Production from Agricultural Biomass and Residues: Recent Advances in Technology. in: Advanced Biofuels and Bioproducts, (Ed.) J.W. Lee, Springer New York, pp. 247-265. 26. Singleton, V.L., Rossi, J.A., 1965. Colorimetry of Total Phenolics with PhosphomolybdicPhosphotungstic Acid Reagents. Am. J. Enol. Viticulture 16(3), 144-158. 27. Stoklosa, R.J., Hodge, D.B., 2014. Chapter 4 - Integration of (Hemi)-Cellulosic Biofuels Technologies with Chemical Pulp Production. in: Biorefineries, (Ed.) N.Q.B.H.A. Vertès, Elsevier. Amsterdam, pp. 73-100. 28. van Heiningen, A., Genco, J., Yoon, S., Tunk, S.M., Zou, H., Luo, J., Mao, H., Pendse, H., 2011. Integrated Forest Biorefineries ? Near-Neutral Process. in: Sustainable Production of Fuels, Chemicals, and Fibers from Forest Biomass, Vol. 1067, Am. Chem. Soc. , pp. 443-473. 29. Wang, L., Chen, H., 2011. Increased fermentability of enzymatically hydrolyzed steamexploded corn stover for butanol production by removal of fermentation inhibitors. Process Biochem. 46(2), 604-607. 30. Wei, P., Cheng, L.-H., Zhang, L., Xu, X.-H., Chen, H.-l., Gao, C.-J., 2014. A review of membrane technology for bioethanol production. Renew. Sustainable Energy Rev. 30(0), 388-400. 31. Yasarla, L.R., Ramarao, B.V., 2012. Dynamics of Flocculation of Lignocellulosic Hydrolyzates by Polymers. Ind. Eng. Chem. Res. 51(19), 6847-6861. 32. Yasarla, L.R., Ramarao, B.V., 2013. Lignin Removal from Lignocellulosic Hydrolyzates by Flocculation with Polyethylene Oxide. J.Biobased. Mater. Bio. 7(6), 684-689. 33. Zha, Y., Westerhuis, J., Muilwijk, B., Overkamp, K., Nijmeijer, B., Coulier, L., Smilde, A., Punt, P., 2014. Identifying inhibitory compounds in lignocellulosic biomass hydrolysates using an exometabolomics approach. BMC Biotechnol. 14(1), 22. 34. Zhang, Y., Han, B., Ezeji, T.C., 2012. Biotransformation of furfural and 5-hydroxymethyl furfural (HMF) by Clostridium acetobutylicum ATCC 824 during butanol fermentation. New Biotechnol. 29(3), 345-351.
21
Figure captions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Figure 1. Flow diagram – from pre-hydrolysate to fermentation: (1) Untreated hydrolysate; (2) Detoxification by flocculation after hydrolysis; (3) Pre-hydrolysate concentration by nanofiltration followed by flocculation of the obtained hydrolysate. Figure 2. Production of ABE in xylose synthetic medium at 37°C using C. acetobutylicum ATCC 824 under uncontrolled pH and static conditions: (A) acids production and pH profile; (B) biomass, xylose consumption and ABE production. Figure 3. Effect of phenols concentration on ABE and acids production; fermentation of wood hydrolysates for 144h by C. acetobutylicum ATCC 824 at 37°C and under controlled pH conditions: (A) untreated hydrolysate; (B) flocculated hydrolysate; (C) nanofiltered/flocculated hydrolysate. Figure 4. Growth behavior of C. acetobutylicum ATCC 824 (Biomass: OD 600nm) at 37°C, uncontrolled pH and static conditions, using as a carbon source: (A) untreated hydrolysate; (B) flocculated hydrolysate. Figure 5. Comparison of: (A) xylose consumption; (B) productivity; (C) ABE yield, obtained by C. acetobutylicum from untreated, flocculated or nonofiltrated/flocculated hydrolysate with different phenolics concentrations after 144h of fermentation at 37°C and uncontrolled pH.
22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Table1. Composition of untreated and treated hemicelluloses hydrolysates generated after detoxification. Compounds Concentration (g/L)
Untreated hydrolysatea
Total sugars Total phenols Acetic acid HMF Furfural
29.7±0.35 3.8±0.06 10.7±0.3 0.10±0.002 0.18±0.008
Neutralized hydrolysateb 29.2±0.36 2.3±0.05 2.6±0.08 0.08±0.001 0.17±0.007
Flocculated hydrolysatec 26.7±0.32 1.65±0.03 2.2±0.06 0.08±0.001 0.17±0.008
Pre-hydrolysate concentration / Hydrolysate flocculationd 63.2±1 3±0.02 2.2±0.06 0.2±0.004 1.02±0.04
The results are an average of three replicates with standard deviation. a : Untreated hydrolysate at pH 1.34. b : Hydrolysate neutralized with Ca(OH)2 at pH 6.8. c : Flocculated hydrolysate neutralized to pH 6.8 with Ca(OH)2. d : Hydrolysate generated after pre-hydrolysate concentration, hydrolysis, flocculation and neutralization to pH 6.8 with Ca(OH)2.
23
Table 2. Evolution of inhibitors and sugar concentration during treatment of hydrolysate by 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
combining flocculation and nanofiltration. Compounds concentration (g/L) Sugars Phenolic compounds Acetic acid HMF Furfural
Prehydrolysatea
Concentrated pre-hydrolysate b
Produced hydrolysatec
Flocculated hydrolysated
3.43±0.1
6.31±0.3
72.0±1.1
63.2±1
4.51±0.01
13.4±0.03
7.20±0.008
3.00±0.02
2.80±0.08 0.09±0.002 0.65±0.03
2.40±0.07 0.05±0.001 0.5±0.02
9.40±0.26 0.25±0.005 1.92±0.09
2.10±0.06 0.19±0.004 1.02±0.05
The results are an average of three replicates with standard deviation. a : Untreated pre-hydrolysate b : Concentrated pre-hydrolysate by nanofiltration using NF270 membrane. c : Corresponding concentrated hydrolysate generated after acid hydrolysis with sulfuric acid. d : Flocculated and neutralized concentrated hydrolysate at pH 6.8
24
Table 3. Reduction of phenolic compounds concentrations in generated hydrolysates with 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
several dilutions.
Hydrolysates
Dilution
Phenols
Xylose added
factor
(g/L)
(g/L)
0
2.30
32.0
3.5
0.66
52.0
7.5
0.31
56.3
13
0.17
57.9
0
1.65
33.3
1.5
1.10
35.7
2.5
0.66
49.3
3.5
0.30
52.4
9.5
0.17
57.2
Untreated hydrolysate
Flocculated hydrolysate
Pre-hydrolysate concentration by nanofiltraltion
0
3.00
0
2.5
1.20
6.94
+
5
0.60
9.47
Hydrolysate
10
0.30
10.7
17.5
0.17
11.2
Flocculation
25
Table 4. Concentration and percentage elimination of some phenolic compounds found in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
wood hydrolysate before and after flocculation with ferric sulfate. Untreated
Flocculated
hydrolysatea
hydrolysateb
(g/L)
(g/L)
Gallic acid
0.047±0.001
0
100
Vanilin
0.042±0.004
0.028±0.003
32
Syringaldehyde
0.19±0.003
0.049±0.001
55
Phenolic compounds
Removal (%)
The results are an average of three replicates with standard deviation. a : Untreated hydrolysate neutralized at pH 6.8. b : Flocculated and neutralized hydrolysate at pH 6.
26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Wood chip pre-hydrolysis
(1) Acid hydrolysis
(2)
(3)
Acid hydrolysis
Concentration by nanofiltration (NF270)
Coagulationflocculation
Acid hydrolysis
Coagulationflocculation
Dilution of hydrolysates
ABE fermentation by C. acetobutylicum ATCC 824 under static and uncontrolled pH conditions.
Figure 1.
27
A
6,2
3,0
6,0
2,5
5,8 5,6 5,4
1,5
1,0
pH
Acids (g/L)
2,0
5,2
pH Acetic acid Butyric acid
5,0
0,5
4,8 4,6
0,0 0
20
40
60
80
100
120
140
160
Fermentation time (h)
B
5 4 3 2 1 0
6 50 Acetone Ethanol Butanol Xylose Biomass Total ABE
4
40
2
Xylose (g/L)
6
60
Acids and solvants (g/L)
7
OD 600nm
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
30
0
20 0
20
40
60
80
100
120
140
160
Fermentation time (h)
Figure 2.
28
12 1,2
A 10 1,0
0,8
Acetone Ethanol Butnaol Total ABE Acetic Acid Butyric Acid
0,6
Acids (g/L)
Products (g/L)
8
6
4
0,4 2 0,2
0
0,0 0.17
0.3
2.3
Phenols (g/L)
7
12
6
10
5
8
4
6
Acetone Ethanol Butanol Total ABE Acetic Acid Butyric Acid
3
2
4
Acids (g/L)
Products (g/L)
B
2
1
0
0
-2 0.17
0.3
0.6
1.1
1.65
Phenols (g/L) 5
16
C
14
4
12 10
3
Acetone Ethanol Butanol Total ABE Acetic acid Butyric acid
2
8 6
Acids (g/L)
Products (g/L)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
4 2
1
0 0 0.17
0.3
0.6
1.1
3.0
Phenols (g/L)Fig.3.
Figure 3.
29
A 3,0 2.3 g/L of phenols 0.3 g/L of phenols 0.17 g/L of phenols
2,5
OD 600 nm
2,0
1,5
1,0
0,5
0,0
0
20
40
60
80
100
120
140
160
Fermentation time (h)
B
3,5
3,0 1.65 g/L of phenols 1.1 g/L of phenols 0.66 g/L of phenols 0.33 g/L of phenols 0.17 g/L of phenols
2,5
OD 600 nm
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
2,0
1,5
1,0
0,5
0,0 0
20
40
60
80
100
120
140
Fermentation Time (h)
Figure 4.
30
A Flocculated hydrolysate 30 Concentrated and flocculated hydrolysate
20 10
Untreated hydrolysate
0 1.1 0.6 0.3 0.17 Total phenolic compound (g/L)
Productivity (g/L.h)
B
Flocculated hydrolysate
0.06 0.04
Concentrated and flocculated hydrolysate
0.02
Untreated hydrolysate
0.00 1.1
0.6
0.3 0.17
Total phenolic compounds (g/L)
C
Concentrated and flocculated hydrolysate 0.3 Flocculated hydrolysate Yield (g/g)
Xylose consumption (g/L)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
0.2 0.1
Untreated hydrolysate
0 1.1
0.6
0.3
0.17
Total phenolic compounds (g/L)
Figure 5.
31
Highlights:
Hardwood pre-hydrolysates from dissolving pulp mill were used for ABE fermentation.
Flocculation was the most efficient method for inhibitors removal.
Detoxification increased the phenolics inhibitory level on ABE production up to 1.1 g/L.
S0960-8524(15)00676-8 http://dx.doi.org/10.1016/j.biortech.2015.05.012 BITE 14983
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
19 March 2015 6 May 2015 7 May 2015
Please cite this article as: Mechmech, F., Chadjaa, H., Rahni, M., Marinova, M., Akacha, N.B., Gargouri, M., Improvement of butanol production from a hardwood hemicelluloses hydrolysate by combined sugar concentration and phenols removal, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.05.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1
Improvement of butanol production from a hardwood
2
hemicelluloses hydrolysate by combined sugar
3
concentration and phenols removal
4
Fatma Mechmecha,c, , Hassan Chadjaab, Mohamed Rahnib, ,
5
Mariya Marinovac* ,Najla Ben Akachad and Mohamed Gargouria
6
a
7
Technology, National Institute of Applied Sciences and Technology (INSAT) Tunis,
8
University of Carthage, Tunisia
9
b
Biocatalysis and Industrial Enzymes Group, Laboratory of Microbial Ecology and
Centre National en Electrochimie et en Technologies Environnementales, Shawinigan,
10
QC, Canada
11
c
12
Biorefinery, Department of Chemical Engineering, Polytechnique Montréal, Montréal, QC,
13
Canada
14
d
15
Chemical Analysis (INRAP), Biotechno pole Sidi Thabet, Ariana, Tunisia
Research Unit on Energy Efficiency and Sustainable Development of the Forest
Laboratory of Natural Substances, National Institute of Research and Physical and
16 17 18 19 20 21 22
* Corresponding author: [email protected]; fax: +1 514 340 4159
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
23
Abstract
24
The feasibility of using hardwood hemicellulosic pre-hydrolysate recovered from a
25
dissolving pulping process for Acetone-Butanol-Ethanol (ABE) fermentation has been
26
investigated. Dilutions and detoxification methods based on flocculation and nanofiltration
27
were tested to determine the inhibitory concentration of phenolic compounds and to
28
evaluate the efficiency of inhibitors removal on fermentation. Flocculation carried out with
29
ferric sulfate was the most effective method for removal of phenolics (56%) and acetic acid
30
(80%). The impact on fermentation was significant, with an ABE production of 6.40 g/L
31
and 4.25 g/L when using flocculation or combined nanofiltration/flocculation respectively,
32
as compared to a non-significant production for the untreated hydrolysate. By decreasing
33
the toxicity effect of inhibitors, this study reports for the first time that the use of these
34
techniques is efficient to increase the inhibitory concentration threshold of phenols, from
35
0.3 g/L in untreated hydrolysate, to 1.1 g/L in flocculated and in nanofiltrated and
36
flocculated hydrolysates.
37
Highlights:
38
39
Hardwood pre-hydrolysates from dissolving pulp mill were used for ABE fermentation.
40
Flocculation was the most efficient method for inhibitors removal.
41
Detoxification increased the phenolics inhibitory level on ABE production up to 1.1
42
g/L.
43
Key words: Bio-butanol; flocculation; nanofiltration; phenolics; wood biomass
44
hydrolysate
45 2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
46 47
1. Introduction During the last decades, the growing environmental and economic concerns have
48
stimulated a worldwide adoption of new government policies and directives for the
49
production of renewable, biomass-derived fuels (Qureshi et al., 2013; Stoklosa & Hodge,
50
2014). Bio-butanol, a product of the ABE (Acetone-Butanol-Ethanol) fermentation, has
51
recently emerged as a superior biofuel and is considered as an attractive alternative to
52
conventional jet fuels, which may be an important factor in the success of the biofuel
53
industry. Its properties are similar to those of the conventional gasoline (Baral & Shah,
54
2014) and it has several advantages over ethanol: higher energy content, lower volatility,
55
less hydroscopic and less corrosive (Lee et al., 2008). Most important from an economic
56
perspective is the fact that its biological production by Clostridia species is largely
57
influenced by the price of the fermentation substrate (Green, 2011). For these reasons, the
58
use of lignocellulosic biomass, such as wood chips or wood wastes, as a feedstock to
59
produce bio-butanol has prompted the interest of the research community. Governments in
60
countries with a mature forestry sector, such as Canada and the Scandinavian countries,
61
have promoted the development and the implementation of the forest biorefinery and its
62
integration into existing pulp and paper (P&P) facilities, the largest global consumers of
63
woody biomass. The P&P industry in these countries has been facing for some time a
64
difficult economic situation (Marinova et al., 2009) and the concept of the integrated forest
65
biorefinery (IFBR) has been identified as a means to diversify its products portfolio,
66
increase its revenues, and regain profitability by manufacturing of a new value-added
67
bioproducts (Adriaan van et al., 2011; Huang et al., 2010). In a Kraft dissolving pulp
68
process, the hemicelluloses are extracted from wood chips prior to the pulping step by
69
steam or hot water partial hydrolysis. This hemicelluloses stream can be used as a
70
feedstock to an adjunct biorefinery plant, to be converted by hydrolysis into C5 and C6 3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
71
sugars, which will in turn be transformed into biofuels and bioproducts such as ethanol,
72
butanol or furfural (Marinova et al., 2014). However, the hemicelluloses pre-hydrolysate
73
contains other wood components besides the sugar oligomers, which are potential
74
fermentation inhibitors; those compounds are: acetic acid, furfural, hydroxymethylfurfural
75
and phenolic derivatives. Detoxification methods of the pre-hydrolysate solution prior to
76
ABE fermentation have been investigated by several authors; they include activated carbon
77
(Kudahettige-Nilsson et al., 2015) and alkaline peroxide treatment (Wang & Chen, 2011).
78
The removal of lignin degradation products by flocculation with primary focus on phenolic
79
compounds has also received much interest from the scientific community. It is known that
80
some phenolics found in lingocellulosic hydrolysates are colloidal particles with high
81
anionic charge that cannot be sedimented easily. During flocculation phenomena, a two-
82
step mechanism occurs. First, the flocculating agents or polyelectrolytes characterized by
83
an ionic charge form bridges between the colloidal particles. These linked particles in turn
84
form, through a mechanism of charge neutralization, agglomerates or patches which can be
85
removed from the solution more easily (Burke et al., 2011; Yasarla & Ramarao, 2012;
86
Yasarla & Ramarao, 2013). It has been reported that this method is effective for inhibitors
87
removal (Ajao et al., 2015b; Duarte et al., 2010) but to the best of our knowledge no
88
fermentation results on the flocculated material have yet been published. It is speculated
89
that the fermentation yield of butanol can be enhanced when combining the detoxification
90
step with the sugar concentration step by evaporation or membrane separation.
91
Nanofiltration may be a feasible membrane concentration technique for the recovery of
92
hemicelluloses from hydrolysates thanks to its selectivity and low energy requirement
93
(Huang et al., 2008; Wei et al., 2014). Ajao et al. (2015a) proposed recently a new process
94
configuration for the detoxification and concentration of a pre-hydrolysate from a
4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
95
dissolving pulp mill by combining nanofiltration and flocculation in a single step, prior to
96
the production of ethanol by fermentation.
97
The work presented herein extends the scope of this prior work and investigates the
98
applicability of flocculation or combined flocculation/nanofiltration to enhance the ABE
99
fermentation. Specifically, the objectives of this work were to:
100
Establish the feasibility of using an untreated or detoxified hemicellulosic wood
101
hydrolysate from a Kraft dissolving pulp mill as a carbon source for the ABE
102
fermentation by Clostridium acetobutylicum;
103
Demonstrate the impact of the proposed detoxification methods on the butanol
104
production (flocculation was first applied alone and later combined to nanofiltration for
105
inhibitors removal);
106 107
Determine the inhibitory concentration of phenolic compounds in each hydrolysate by comparing ABE production, fermentation yields and productivities.
108 109
2. Materials and methods
110
2.1. Microorganism, culture maintenance and inoculum preparation
111
Clostridium acetobutylicum ATCC 824 was obtained from American Type Culture
112
Collection (ATCC) and was cultured in sterilized RCM (Reinforced Clostridium Medium)
113
under anaerobic conditions at 37°C for 18-20 hours until an Optical Density (OD600) of
114
1.9-2. A stock culture was then prepared in 30% glycerol and stored at 80°C. For inoculum
115
preparation, RCM medium consisting of (in g/L): tryptose 10, beef extract 10, glucose 5,
116
sodium chloride 5, yeast extract 3, sodium acetate 1, soluble starch 0.5 and L-cysteine-HCl
117
0.5 was used. The medium was purged with a gas mixture (80% N2 and 20% CO2) to
118
dissolve oxygen and then autoclaved at 121°C for 15 minutes. The culture (inoculums) was
119
inoculated with stock culture and allowed to grow for 16-18 hours at 37°C under shaking 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
120
conditions at 110 rpm, until it was ready to be inoculated into the butanol production
121
medium at an OD600 between 0.6 and 0.9.
122 123 124
2.2. Preparation and treatment of hydrolysates Before testing the fermentability of the wood hydrolysate, different detoxification
125
approaches were evaluated (Figure 1).
126
2.2.1. Pre-hydrolysis: steam and hot water
127
A mixture of aspen (60%) and maple (40%) wood furnish obtained from an operating
128
Kraft pulping mill located in eastern Canada was used in this study. The pre-
129
hydrolysate rich in hemicelluloses was generated in a pilot digester at the FPInnovations
130
facilities (Pointe-Claire, Canada) using steam and hot water, whereas the cellulosic fraction
131
was directed to dissolving pulp production as described previously by Ajao et al. (2015b).
132
A 12L sample of pre-hydrolysate was withdrawn and shipped to the Centre National en
133
Electrochimie et en Technologies Environnementales (CNETE), where the experimental
134
program was conducted.
135
2.2.2. Sulfuric acid hydrolysis
136
To convert the polysaccharides into fermentable monomeric sugars, the pre-
137
hydrolysate was completely hydrolyzed with 1.5% wt/wt sulfuric acid at 121°C for 60 min.
138
The hydrolysate was filtered and sterilized by passing through a 1.5 µm filter. Untreated
139
and treated hydrolysates were neutralized with Ca(OH)2, from an initial pH of 1.34 to a
140
final pH of 6.5.
141
2.2.3. Hydrolysate detoxification
142
The procedure developed by Ajao et al. (2015a) in a previous work on the
143
production of ethanol from a similar hemicellulosic pre-hydrolysate extracted from wood
144
chips, was used for the hydrolysate detoxification in the current work. 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
145
Coagulation-Flocculation. Flocculation experiments were carried out in jar tests using
146
ferric sulfate Fe2(SO4)3 as flocculating agent. A volume of 1 L of a wood hydrolysate was
147
mixed with a 200 g/L of ferric sulfate solution to reach a ratio [Fe]/[phenols] of 1 and the
148
mixture was homogenized by agitation at 150 rpm. The pH was then adjusted to 6.8 with
149
Ca(OH)2 and the agitation was resumed for 30 min at 50 rpm. At the end of the agitation
150
period, the mixture was left to rest for 2 hours so that a sedimentation occurs. The
151
supernatant was then filtered with a 1.5 µm filter and used as a carbon source to carry out
152
the fermentation studies.
153
Concentration of hydrolysate by nanomembrane filtration. The pre-hydrolysate,
154
containing polysaccharides, was concentrated by naonomembrane filtration using the
155
organic membrane NF270 (MWCO 200-400 Da). A laboratory scale cross-flow flat-sheet
156
membrane test unit (SEPA CF II, GE Osmonics) was used in a batch mode and the
157
permeate stream collected in a small vessel. During the concentration experiments, 2.35 L
158
of permeate were withdrawn from an initial volume of 3.50 L of pre-hydrolysate to reach a
159
volumetric concentration factor of 3.43. Before and after concentration, samples from the
160
feed tank were taken in order to analyze the amounts of sugars, total phenols, acetic acid,
161
furfural and HMF.
162 163
2.3. Fermentation medium and conditions
164
The fermentations experiments were carried out in 250 mL screw capped Schott
165
bottles with 200 mL of fermentation medium (xylose 60 g/L and yeast extract 5g/L) (Choi
166
et al., 2012; Qureshi & Blaschek, 1999). Untreated or flocculated hydrolysates containing
167
inhibitory chemicals were used as carbon sources. The phenolic compounds concentrations
168
of the hydrolysates were first adjusted to stepwise decreasing levels by serial dilutions and
169
the final rate of sugars was raised to 60 g/L of xylose. A gas mixture consisting of 80% 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
170
N2and 20% CO2 was sparged for 5 min. through the samples to remove oxygen, before
171
they were sterilized at 121°C for 15 min. After cooling, loosely capped bottles were
172
transferred to anaerobic jars containing envelopes Gas Pak (BD Gas PakTMEZ Anaerobe
173
Container System) with indicators (BD BBLTM Dry Anaerobic Indicator Strips) for 48h to
174
create anaerobic conditions in the mixtures. Prior to inoculation, 2 mL of filtered sterilized
175
stock solution consisting of buffer (KH2PO4, 50 g/L; K2HPO4,50 g/L; ammonium acetate,
176
220 g/L), vitamins (para-aminobenzoic acid, 0.1 g/L; thiamin, 0.1 g/L; Biotin, 0.001 g/L)
177
and minerals (MgSO4.7H2O, 20 g/L; MnSO4.H2O, 1 g/L; FeSO4.7H2O, 1 g/L; NaCl, 1
178
g/L) were added to supply the equivalent nutriment concentration level as a synthetic
179
medium (Ezeji et al., 2007). Finally, 5% (v/v) of active growth inoculums was added and
180
the incubation was performed at 37°C under static conditions without pH control. Samples
181
were taken at regular intervals for optical density measurements. Sugars and ABE analysis
182
for all fermentation experiments were conducted at least in duplicates. During the course of
183
fermentation, 5 mL samples were periodically withdrawn to measure pH, OD600, residual
184
xylose, alcohols, acetic acid and butyric acid.
185 186
2.4. Analytical methods
187
Spectrophotometric analysis. The growth of C. acetobutylicum ATCC 824 was monitored
188
by measuring OD600 using a UV-visible spectrophotometer (Pharmacia Biotech
189
Novaspec®II).Total phenols concentration was measured by an UV-visible colorimetric
190
method using Folin-Ciocalteu reagent method from Singleton and Rossi (1965).
191
GC analysis. Butanol, acetone and ethanol were measured by gas chromatography (GC
192
7890A, Agilent Technologies), equipped with a flame ionization detector (FID, H2 flow
193
rate: 30 mL/min; air flow rate: 2.23 mL/min) and an OV 624 capillary column.
194 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
195
HPLC Analyses.
196
The concentration of simple sugars was determined by high performance liquid
197
chromatography (HPLC) using an Agilent 1260 system, a Refractive Index Detector
198
and an EC Nucleodur RP-NH2 column (250 mm x 4.6 mm, 5 μm ). Samples separation
199
was performed at 40°C with a mixture of acetonitrile and deionized water (ratio 75:25)
200
as a mobile phase and at a flow rate of 1.5 mL/min.
201
ABE productivity was calculated as total ABE produced in g/L, divided by the
202
fermentation time and is expressed as g/L.h. The ABE yield was calculated as the total
203
ABE produced, divided by the total sugars used and is expressed as g/g.
204
HMF and furfural were also analyzed by HPLC (Agilent Technologies, Germany)
205
using a 280 nm diode array detector (DAD) and a Nucleosil C18 (150 x4.6 mm)
206
Column. The inhibitors were eluted with a mixture of acetonitrile, deionized water and
207
acetic acid (ratio 84:15:1) at a flow rate of 0.8 mL/min.
208
Organic acids concentrations were determined using the same HPLC with a diode array
209
detector. Separation was carried out using Interstil ODS-3 column (150 x4.6 mm) at
210
40°C. The mobile phase was a mixture of 99% KH2PO4 at pH 2.8, 1% acetonitrile and
211
a flow rate of 1.25 mL/min.
212
Syringaldehyde, vanillin and gallic acid were analyzed by HPLC (Agilent
213
Technologies, Germany) using a Nucleosil C18 (150 x 4.6 mm) and a DAD at 313 nm
214
and 280 nm. The separation was achieved with a mixture of 15% acetonitrile – 85%
215
(0.1% phosphoric acid in water).
216 217 218 219 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
220
3. Results and discussion
221
3.1. Removal of phenolic compounds
222
During the lignocellulosic biomass pre-hydrolysis and hydrolysis, the conversion of
223
sugars from oligomeric to monomeric forms occurs together with the generation of
224
chemical compounds, including sugar and lignin degradation products having different
225
inhibitory activity levels (Baral & Shah, 2014; Ben Chaabane & Marchal, 2013). The
226
untreated hydrolysate used in this study contained 29.7 g/L of fermentable sugars, xylose
227
being the most abundant one (80% by weight of total sugars).The concentration of
228
potential fermentation inhibitors were: 10.7 g/L acetic acid, 3.8 g/L phenolic compounds,
229
0.1 g/L HMF and 0.18 g/L furfural. Additional parameters are given in Table 1. At these
230
levels, the inhibitory profile of the fermentation medium was critical and no growth of C.
231
acetobutylicum, nor butanol production was observed; results from this preliminary phase
232
of the work are not given herein. In order to decrease the inhibitory effect exerted by these
233
compounds, the hydrolysate was subjected to flocculation to remove the phenolics
234
produced by the degradation of lignin. Flocculation was achieved by addition of ferric
235
sulfate and the obtained results showed that this method was effective to remove 56.6% of
236
the phenolic compounds with a high sugar recovery (26.7 g/L). Those results also
237
demonstrate that flocculation does not remove significantly the fermentable sugars (Table
238
1). In fact, flocculation can preferentially precipitate out the colloidal fraction containing
239
lignin and lignin derived compounds (Duarte et al., 2010). The flocculation also caused the
240
removal of other inhibitors from the hydrolysate, 79.4% of acetic acid, 20% of HMF and
241
5.55% of furfural. Those results are in good agreement with the removal percentage found
242
by Ajao et al. (2015a). In addition, it can also be concluded that the removal of such
243
inhibitors, generally present in the solution, can be largely attributed to the use of lime for
244
pH adjustment as well as to the filtration (Table1). 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
245
The concentration of sugars obtained by pre-hydrolysis of wood chips is low and it is
246
not economically feasible to use this medium for microbial fermentation to produce high
247
value products (Qi et al., 2011). A second method combining concentration and
248
detoxification by nanofiltration was added to the procedure to increase the sugar
249
concentration. The acid hydrolysate thus produced was further detoxified by flocculation.
250
As shown in Table 2, when the NF270 membrane was used in the nanofiltration step,
251
phenols were the only inhibitors affected and their concentration ratio was increased by a
252
factor of 2.97, while it remained practically unchanged for acetic acid, furfural and HMF.
253
Qi et al. (2011) have also observed a negative retention of furfural when using NF270 to
254
concentrate a model solution of xylose, glucose and furfural. After flocculation, the
255
amount of acetic acid decreased by 77.7%, whereas the concentration of HMF and furfural
256
remained high. The detoxification strategy combining pre-hydrolysate concentration and
257
hydrolysate flocculation was effective in concentrating the sugars, but not in removing the
258
inhibitors.
259
As shown in Table 3, the dilution effect of various hydrolysates generated in the
260
course of the experimental program was investigated. It should be noted that these
261
dilutions were undertaken in order to determine the exact inhibitory concentration of the
262
total phenolic compounds in each hydrolysate and not as a detoxification method to be
263
considered at an industrial scale. This approach caused a reduction in sugar content that
264
may negatively affect the ABE production. Therefore, the hydrolysates were supplemented
265
with xylose prior to fermentation to raise the final sugar concentration to approximately 60
266
g/L, so that the inhibitors concentration remains the only parameter to be considered.
267 268 269 11
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
270 271
3.2. ABE fermentation of untreated hydrolysate In order to investigate the possible upgrade of the pre-hydrolysate obtained from a
272
Kraft dissolving pulp mill and to test its ability to be fermented by C. acetobutylicum for
273
ABE production, fermentation experiments were conducted using different hydrolysates,
274
as illustrated in Figure 1. A control experiment was first performed using 60 g/L of xylose
275
as model substrate and C. acetobutylicum ATCC 824 to benchmark the fermentation tests.
276
As shown in Figure 2A, during the acidogenic phase of the Clostridium growth which
277
lasted 72h, acetic and butyric acids were produced by fermentation. The final
278
concentrations of those two acids reached 2.66 g/L and 1.40 g/L, respectively. During the
279
acidogenic phase, the pH decreased to 4.8, which is considered as an optimum value,
280
ensuring the transition from the acidogenic to the solvatogenic phase. The solvatogenic
281
phase resulted in the production of 5.82 g/L ABE with 3.95 g/L of butanol (Figure 2B),
282
corresponding to an ABE productivity of 0.04 g/L.h. After 144h of fermentation, 25.2 g/L
283
of xylose were consumed; the corresponding ABE yield was 0.23 g/g and the acetic acid
284
production 2.4 g/L (Figure 2A). The obtained kinetic parameters are higher than those
285
previously reported by Jiang et al. (2014) from a study in which xylose was used as a
286
carbon source to produce butanol by C. acetobutylicum ATCC 824 under uncontrolled pH
287
conditions. The reported ABE productivity was 0.023 g/L.h and the yield 0.023 g/g.
288
Following the control experiment, fermentation was conducted using the untreated
289
wood hydrolysate as carbon source with different dilutions (Figure 3A). After 144h of
290
fermentation, an insignificant ABE production was observed at a phenolic concentration
291
higher than 0.3 g/L. This result may be linked to an accumulation of acetic acid of 10.1 g/L
292
which was not re-assimilated to produce solvents in the presence of 2.3 g/L of phenols. As
293
illustrated on Figure 3A, at phenolic concentrations of 0.17 g/L or 0.3 g/L, the total ABE
294
production was about 0.8 g/L with around 0.4 g/L of butanol. This production level is low 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
295
compared to the level achieved in the control experiment where xylose was used as carbon
296
source, suggesting that the hydrolysate remains toxic for the culture even after a 13-fold
297
dilution. The inhibition effect can be primarily attributed to the presence of totally soluble
298
phenolics that have been reported to be more toxic than furfural and HMF, even at low
299
concentrations (Baral & Shah, 2014). These compounds are known to undergo partitioning
300
in the biological membranes, thus reducing their permeability and their ability to play the
301
role of selective barriers and enzyme matrices (Ibraheem & Ndimba, 2013; Palmqvist &
302
Hahn-Hägerdal, 2000). It has also been shown in previous studies that low concentrations
303
of HMF and furfural produce a synergetic effect on the ABE fermentation (Ezeji et al.,
304
2007; Qureshi et al., 2012; Zhang et al., 2012). This would support the assumption that in
305
our work HMF and furfural cannot be considered as inhibition factors.
306
In order to investigate the monomers inhibitory concentration, major phenolic
307
compounds contained in the untreated hydrolysate were analyzed. As shown in Table 4,
308
the amount of vanillin was about 0.042 g/L, syringaldehyde 0.19 g/L, and gallic acid 0.047
309
g/L. The inhibitory concentrations in this work were much lower than those obtained by
310
Ezeji et al. (2007) (syringaldehyde 0.3-1 g/L) and Cho et al. (2009) (0.17 g/L for vanillin
311
and syringaldehyde) who studied the inhibitory effect of model phenolic compounds
312
solutions. This was assigned to the synergy that can occur due to the coexistence of
313
different inhibitory compounds. In fact, as explained by Zha et al. (2014) the threshold
314
toxicity of a specific compound present in a biomass hydrolysate can be much lower than
315
that evaluated in a synthetic medium.
316 317 318 319
3.3. Effect of hydrolysate flocculation on the ABE fermentation In the next phase of the experimental program, the hydrolysate was subjected to flocculation prior to being used for butanol fermentation at different dilutions, as described 13
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
320
previously. As illustrated on Figure 4A and 4B, for the same phenolic concentrations, the
321
flocculated hydrolysate shows a reduction of the lag phase for the growth of Clostridium,
322
relatively to the untreated hydrolysate, suggesting that the microorganism can easily adapt
323
to the fermentation medium. Avoiding extended lag phases would constitute a significant
324
advantage when scaling up the process to an industrial size.
325
As shown in Figure 3B, an increase of ABE production was observed, compared to the
326
control synthetic medium (5.82 g/L ABE) without inhibitors during a 144h fermentation; a
327
maximum solvent concentration of 6.00 and 6.40 g/L was reached in the presence of 0.3
328
and 0.17 g/L of phenols, respectively. These values are 88% higher than those obtained
329
with the untreated hydrolysate with the same phenolic compounds concentration. After 2.5
330
and 1.5 fold dilution, achieving 0.6 g/L and 1.1 g/L of phenols, the respective ABE
331
production was 2.6 g/L and 0.6 g/L. These results are important in view of the lack of
332
production observed with untreated hydrolysate at the same phenolics concentrations. An
333
accumulation of acetic acid, reaching concentrations of 6.7 g/L and 7.4 g/L for the same
334
concentrations of phenolics, was observed in the experiments with low ABE yield, as
335
illustrated on Figure3B. When using undiluted flocculated hydrolysate with a total
336
phenolics concentration of 1.65 g/L, the acetic acid was as high as 10 g/L with no ABE
337
production. In fact, for high concentrations of acetic acid, the culture does not enter in the
338
solvatogenic phase, because the accumulation of the acid is detrimental for the
339
solvatogenic clostridia nutrient uptake and the cell growth (Ezeji et al., 2010). The reason
340
behind this behavior may be related to the fact that phenolic compounds interfere with the
341
metabolic pathway by preventing the re-assimilation of the organic acids into acetyl-CoA
342
and butyryl-CoA and further conversion into ethanol and butanol (Baral & Shah, 2014).
343
Interestingly, the flocculation reduced the level of inhibition exerted by phenols on
344
cell growth and enhanced the ABE production. Moreover, the flocculation removal 14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
345
efficiency on monomeric phenols was different from that observed on total phenols. As
346
shown in Table 4, where three monomeric phenols are represented, the flocculation
347
achieved a removal efficiency of 32% for vanillin, 55% for syringaldehyde and 100% for
348
gallic acid, as compared to 56.6 % for total phenols. These results demonstrate that
349
flocculation removes phenolic compounds in different proportions as function of their
350
ability to form aggregates with ferric sulfate. They also suggest that the proposed
351
detoxification method may contribute to the removal of other unidentified or non-analyzed
352
monomeric phenols that may have higher inhibitory effect. Thus, at the same phenols
353
concentration, an untreated and a flocculated hydrolysate do not contain the same
354
proportions of phenols monomers and as a consequence they will not present the same
355
inhibitory effect. Taken together, these reasons could explain the raise of the total phenols
356
inhibition levels from 0.3 g/L for an untreated hydrolysate to 1.1 g/L for a flocculated
357
hydrolysate (Figure 3A and 3B). This further supports the conclusion that flocculation is an
358
effective detoxification method for the removal of phenolic compounds.
359 360
3.4. Effect of combining pre-hydrolysate concentration and hydrolysate flocculation
361
on the ABE fermentation
362
In the final stage of the experimental program, fermentation experiments were
363
conducted using the hydrolysate produced by the third detoxification technique, illustrated
364
in Figure 1. In this case, the pre-hydrolysate was first concentrated using the organic
365
membrane NF270 to increase the sugar concentration by 2.17 folds (Table 2) and was
366
subsequently subjected to flocculation under the same conditions. Since it was not possible
367
to get reliable results of OD at 600 nm in the case of C. acetobutylicum growth, because of
368
the high turbidity of the culture medium, the sugar consumption and the solvent production
369
were used as criteria for comparison with the other experiments, as recommended by (Zha 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
370
et al., 2014). Figure 3C shows that by applying this detoxification strategy, the ABE
371
production was about 3.77 g/L and 4.25 g/L for phenolics respective concentrations of 0.3
372
and 0.17 g/L. These values are approximately 35% lower than those obtained with the
373
flocculated hydrolysate at the same phenolics concentrations, despite the fact that the
374
flocculation efficiency was approximately the same in both cases (58% VS 56.6%).
375
Thereby, the fact of overly diluting the hydrolysate in the seconf detoxification strategy to
376
achieve the same concentration of phenolic compounds may reduce the concentrations of
377
other compounds reported to be beneficial for the fermentation. Similar level of inhibition
378
was obtained at phenolics concentration of 1.1 g/L, an accumulation of 7.74 g/L of acetic
379
acid and ABE production of 0.2 g/L, as with a flocculated hydrolysate. By applying
380
nanofiltration, the main objective was to concentrate the pre-hydrolysate in order to
381
increase sugar concentration, to reduce the cost of the acid hydrolysis step (amount of acid
382
added, volume to be heated). During concentration, nanofiltraion has also allowed the
383
removal of acetic acid, HMF and furfural. However, its combination with flocculation was
384
not more efficient than flocculation alone. It is then recommended to apply another
385
technique before or after flocculation to improve the inhibitors removal efficiency. The
386
economic viability of the overall process should be taken into account.
387
A comparison of productivities and yields for different fermentation conditions are
388
shown in Figure 5. At phenolics concentrations of 0.17 g/L and 0.3 g/L, the fermentation
389
experiment using untreated hydrolysate did not reach more than 0.005 g/L.h of
390
productivity with sugar consumptions of 9.40 g/L and 11.9 g/L (Figure 5A) and yields of
391
0.08 and 0.07 g/g, respectively (Figure 5C). The corresponding productivities were
392
improved to 0.041 g/L.h and 0.044 g/L.h using flocculated hydrolysate, and to 0.023 g/L.h
393
and 0.03 g/L.h using combined nanofiltration and flocculation (Figure 5B). Improvements
394
of 65% and 74% in ABE yields could be achieved from flocculated hydrolysate with a 16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
395
xylose consumption of 24.2 g/L and 28.4 g/L. An increase of ABE yield of 3.12-folds and
396
3.57-folds with the third detoxified hydrolysate was also observed for the same phenolics
397
concentration.
398 399 400
3.5. Comparison to other studies Various lignocellulosic feedstocks for bio-butanol production via ABE fermentation
401
have been used in previous studies and reported in the scientific literature but few of these
402
studies discuss the use of wood hydrolysates obtained from pulp and paper mills in
403
operation, thus emulating the concept of the integrated forest biorefinery. A study of the
404
ABE fermentation by C. beijerinckii of wood hydrolysate with a sugar mixture of glucose,
405
xylose and arabinose has been reported by Lu et al. (2013). The authors have investigated
406
the use of various techniques for the removal of inhibitors: adsorption by activated
407
charcoal achieving concentration of 9.98 g/L of total ABE, resin adsorption (11.4 g/L of
408
total ABE), and combined resin adsorption and gas stripping (17.7 g/L of total ABE).
409
However, the work presented herein seems to be the first to consider the validation of
410
flocculation and combined nanofiltration and flocculation as hemicellulosic wood
411
hydrolysate detoxification methods, prior to the ABE fermentation. Despite the fact that
412
the work was done under uncontrolled pH conditions, the results are close to those
413
obtained recently by Kudahettige-Nilsson et al. (2015). They have reported that at a
414
phenolic concentration of 0.6 g/L, the batch fermentation of a detoxified hardwood
415
hydrolysate using active carbon with pH control produced 2.2 g/L of ABE. Their process
416
resulted in a yield of 0.13 g/g with a solvent production efficiency of 33%, as compared to
417
a yield of 0.17g/g with a production efficiency of 43% obtained in the present study when
418
fermenting a flocculated hydrolysate at the same phenolics concentration (Figure 5C). The
419
pH control of the medium is considered as a key factor for the ABE fermentation 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
420
enhancement (Jiang et al., 2014). This further demonstrates the effectiveness of
421
flocculation as a detoxification method and the feasibility of using a hemicellulosic wood
422
hydrolysate issued from dissolving Kraft pulp mill as an alternative renewable feedstock
423
for bio-butanol production via ABE fermentation. Further optimization work should be
424
undertaken to ensure detoxification of the hydrolysate to the feasible thresholds for the
425
ABE fermentation. The economic viability of the proposed process should also be
426
validated. Finally, means to reduce the cost of the fermentation medium should be
427
investigated. In this perspective, the use of low cost agricultural residues as a nitrogen
428
source in the stock solution is an opportunity worthy of further investigation.
429 430 431
4. Conclusions In this work, it has been demonstrated that the ABE fermentation of hemicellulosic
432
wood hydrolysates has been improved using dilution and flocculation or combined
433
nanofiltration and flocculation as downstream detoxification methods. Within the range of
434
conditions covered at a laboratory scale, the applied methods enhanced inhibitory levels
435
from 0.3 to 1.1 g/L of total phenols. Without dilutions, the residual total phenols
436
concentrations in the detoxified hydrolysates remain higher than the acceptable
437
fermentation thresholds. Therefore, it is recommended to investigate the use of different
438
hydrolysis methods generating less inhibitors, as well as combining flocculation with other
439
detoxification techniques such as enzymatic treatment.
440 441 442 443 444 18
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
445
Acknowledgments:This work was supported by a grant from the College-University I2I
446
Program of the Natural Sciences and Engineering Research Council of Canada (grant
447
number 437803-12).The authors wish to express their gratitude to the Centre National en
448
Electrochimie et Technologies Environnementales (Shawinigan, QC) which hosted the
449
experimental work presented in this manuscript and provided staff support and equipment
450
and, to FPInnovations (Pointe-Claire, QC) which supplied the hemicelluloses pre-
451
hydrolysate samples. The authors are also thankful to Professor Jean Paris from the
452
Department of Chemical Engineering at Polytechnique Montréal for his valuable
453
contributions to this work and for reviewing the manuscript.
19
References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1.
2.
3.
4. 5.
6.
7.
8.
9.
10.
11. 12.
13. 14.
15.
16.
17. 18.
Ajao, O., Le Hir, M., Rahni, M., Marinova, M., Chadjaa, H., Savadogo, O., 2015a. Concentration and Detoxification of Kraft Prehydrolysate by Combining Nanofiltration with Flocculation. Ind. Eng. Chem. Res. 54(3), 1113-1122. Ajao, O., Rahni, M., Marinova, M., Chadjaa, H., Savadogo, O., 2015b. Retention and flux characteristics of nanofiltration membranes during hemicellulose prehydrolysate concentration. Chem. Eng. J. 260, 605-615. Baral, N., Shah, A., 2014. Microbial inhibitors: formation and effects on acetone-butanolethanol fermentation of lignocellulosic biomass. Appl. Microbiol. Biotechnol. 98(22), 91519172. Ben Chaabane, F., Marchal, R., 2013. Upgrading the Hemicellulosic Fraction of Biomass into Biofuel. Oil Gas Sci. Technol. – Rev. IFP Energies nouvelles 68(4), 663-680. Burke, D.R., Anderson, J., Gilcrease, P.C., Menkhaus, T.J., 2011. Enhanced solid–liquid clarification of lignocellulosic slurries using polyelectrolyte flocculating agents. Biomass Bioenergy 35(1), 391-401. Cho, D., Lee, Y., Um, Y., Sang, B.-I., Kim, Y., 2009. Detoxification of model phenolic compounds in lignocellulosic hydrolysates with peroxidase for butanol production from Clostridium beijerinckii. Appl. Microbiol. Biotechnol. 83(6), 1035-1043. Choi, S., Lee, J., Jang, Y.-S., Park, J., Lee, S., Kim, I., 2012. Effects of nutritional enrichment on the production of acetone-butanol-ethanol (ABE) by Clostridium acetobutylicum. J. Microbiol. 50(6), 1063-1066. Duarte, G.V., Ramarao, B.V., Amidon, T.E., 2010. Polymer induced flocculation and separation of particulates from extracts of lignocellulosic materials. Bioresour. Technol. 101(22), 85268534. Ezeji, T., Milne, C., Price, N., Blaschek, H., 2010. Achievements and perspectives to overcome the poor solvent resistance in acetone and butanol-producing microorganisms. Appl. Microbiol. Biotechnol. 85(6), 1697-1712. Ezeji, T., Qureshi, N., Blaschek, H.P., 2007. Butanol production from agricultural residues: Impact of degradation products on Clostridium beijerinckii growth and butanol fermentation. Biotechnol. Bioeng. 97(6), 1460-1469. Green, E.M., 2011. Fermentative production of butanol—the industrial perspective. Curr. Opin. Biotechnol. 22(3), 337-343. Huang, H.-J., Ramaswamy, S., Al-Dajani, W.W., Tschirner, U., 2010. Process modeling and analysis of pulp mill-based integrated biorefinery with hemicellulose pre-extraction for ethanol production: A comparative study. Bioresour. Technol. 101(2), 624-631. Huang, H.-J., Ramaswamy, S., Tschirner, U.W., Ramarao, B.V., 2008. A review of separation technologies in current and future biorefineries. Sep. Purif. Technol. 62(1), 1-21. Ibraheem, O., Ndimba, B.K., 2013. Molecular Adaptation Mechanisms Employed by Ethanologenic Bacteria in Response to Lignocellulose-derived Inhibitory Compounds. Int. J. Biol. Sci. 9(6), 598-612. Jiang, W., Wen, Z., Wu, M., Li, H., Yang, J., Lin, J., Lin, Y., Yang, L., Cen, P., 2014. The Effect of pH Control on Acetone–Butanol–Ethanol Fermentation by Clostridium acetobutylicum ATCC 824 with Xylose and d-Glucose and d-Xylose Mixture. Chin. J. Chem. Eng. 22(8), 937-942. Kudahettige-Nilsson, R.L., Helmerius, J., Nilsson, R.T., Sjöblom, M., Hodge, D.B., Rova, U., 2015. Biobutanol production by Clostridium acetobutylicum using xylose recovered from birch Kraft black liquor. Bioresour. Technol. 176(0), 71-79. Lee, S.Y., Park, J.H., Jang, S.H., Nielsen, L.K., Kim, J., Jung, K.S., 2008. Fermentative butanol production by clostridia. Biotechnol. Bioeng. 101(2), 209-228. Lu, C., Dong, J., Yang, S.-T., 2013. Butanol production from wood pulping hydrolysate in an integrated fermentation–gas stripping process. Bioresour. Technol. 143(0), 467-475. 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
19. Marinova, M., Mateos-Espejel, E., Jemaa, N., Paris, J., 2009. Addressing the increased energy demand of a Kraft mill biorefinery: The hemicellulose extraction case. Chem. Eng. Res. Des. 87(9), 1269-1275. 20. Marinova, M., Perrier, M., Paris, J., 2014. Implementation of a Forest Biomass-based Biofuel Industry: A Canadian Experience. in: International Conference on Renewable Energies and Power Quality (ICREPQ’14), Vol. No.12, Renew. Energy Power. Qual. J. (RE&PQJ). Cordoba (Spain), 8th to 10th April, 2014. 21. Palmqvist, E., Hahn-Hägerdal, B., 2000. Fermentation of lignocellulosic hydrolysates. II: inhibitors and mechanisms of inhibition. Bioresour. Technol. 74(1), 25-33. 22. Qi, B., Luo, J., Chen, X., Hang, X., Wan, Y., 2011. Separation of furfural from monosaccharides by nanofiltration. Bioresour. Technol. 102(14), 7111-7118. 23. Qureshi, N., Blaschek, H.P., 1999. Butanol recovery from model solution/fermentation broth by pervaporation: evaluation of membrane performance. Biomass Bioenergy 17(2), 175-184. 24. Qureshi, N., Bowman, M.J., Saha, B.C., Hector, R., Berhow, M.A., Cotta, M.A., 2012. Effect of cellulosic sugar degradation products (furfural and hydroxymethyl furfural) on acetone– butanol–ethanol (ABE) fermentation using Clostridium beijerinckii P260. Food Bioprod. Process. 90(3), 533-540. 25. Qureshi, N., Liu, S., Ezeji, T.C., 2013. Cellulosic Butanol Production from Agricultural Biomass and Residues: Recent Advances in Technology. in: Advanced Biofuels and Bioproducts, (Ed.) J.W. Lee, Springer New York, pp. 247-265. 26. Singleton, V.L., Rossi, J.A., 1965. Colorimetry of Total Phenolics with PhosphomolybdicPhosphotungstic Acid Reagents. Am. J. Enol. Viticulture 16(3), 144-158. 27. Stoklosa, R.J., Hodge, D.B., 2014. Chapter 4 - Integration of (Hemi)-Cellulosic Biofuels Technologies with Chemical Pulp Production. in: Biorefineries, (Ed.) N.Q.B.H.A. Vertès, Elsevier. Amsterdam, pp. 73-100. 28. van Heiningen, A., Genco, J., Yoon, S., Tunk, S.M., Zou, H., Luo, J., Mao, H., Pendse, H., 2011. Integrated Forest Biorefineries ? Near-Neutral Process. in: Sustainable Production of Fuels, Chemicals, and Fibers from Forest Biomass, Vol. 1067, Am. Chem. Soc. , pp. 443-473. 29. Wang, L., Chen, H., 2011. Increased fermentability of enzymatically hydrolyzed steamexploded corn stover for butanol production by removal of fermentation inhibitors. Process Biochem. 46(2), 604-607. 30. Wei, P., Cheng, L.-H., Zhang, L., Xu, X.-H., Chen, H.-l., Gao, C.-J., 2014. A review of membrane technology for bioethanol production. Renew. Sustainable Energy Rev. 30(0), 388-400. 31. Yasarla, L.R., Ramarao, B.V., 2012. Dynamics of Flocculation of Lignocellulosic Hydrolyzates by Polymers. Ind. Eng. Chem. Res. 51(19), 6847-6861. 32. Yasarla, L.R., Ramarao, B.V., 2013. Lignin Removal from Lignocellulosic Hydrolyzates by Flocculation with Polyethylene Oxide. J.Biobased. Mater. Bio. 7(6), 684-689. 33. Zha, Y., Westerhuis, J., Muilwijk, B., Overkamp, K., Nijmeijer, B., Coulier, L., Smilde, A., Punt, P., 2014. Identifying inhibitory compounds in lignocellulosic biomass hydrolysates using an exometabolomics approach. BMC Biotechnol. 14(1), 22. 34. Zhang, Y., Han, B., Ezeji, T.C., 2012. Biotransformation of furfural and 5-hydroxymethyl furfural (HMF) by Clostridium acetobutylicum ATCC 824 during butanol fermentation. New Biotechnol. 29(3), 345-351.
21
Figure captions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Figure 1. Flow diagram – from pre-hydrolysate to fermentation: (1) Untreated hydrolysate; (2) Detoxification by flocculation after hydrolysis; (3) Pre-hydrolysate concentration by nanofiltration followed by flocculation of the obtained hydrolysate. Figure 2. Production of ABE in xylose synthetic medium at 37°C using C. acetobutylicum ATCC 824 under uncontrolled pH and static conditions: (A) acids production and pH profile; (B) biomass, xylose consumption and ABE production. Figure 3. Effect of phenols concentration on ABE and acids production; fermentation of wood hydrolysates for 144h by C. acetobutylicum ATCC 824 at 37°C and under controlled pH conditions: (A) untreated hydrolysate; (B) flocculated hydrolysate; (C) nanofiltered/flocculated hydrolysate. Figure 4. Growth behavior of C. acetobutylicum ATCC 824 (Biomass: OD 600nm) at 37°C, uncontrolled pH and static conditions, using as a carbon source: (A) untreated hydrolysate; (B) flocculated hydrolysate. Figure 5. Comparison of: (A) xylose consumption; (B) productivity; (C) ABE yield, obtained by C. acetobutylicum from untreated, flocculated or nonofiltrated/flocculated hydrolysate with different phenolics concentrations after 144h of fermentation at 37°C and uncontrolled pH.
22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Table1. Composition of untreated and treated hemicelluloses hydrolysates generated after detoxification. Compounds Concentration (g/L)
Untreated hydrolysatea
Total sugars Total phenols Acetic acid HMF Furfural
29.7±0.35 3.8±0.06 10.7±0.3 0.10±0.002 0.18±0.008
Neutralized hydrolysateb 29.2±0.36 2.3±0.05 2.6±0.08 0.08±0.001 0.17±0.007
Flocculated hydrolysatec 26.7±0.32 1.65±0.03 2.2±0.06 0.08±0.001 0.17±0.008
Pre-hydrolysate concentration / Hydrolysate flocculationd 63.2±1 3±0.02 2.2±0.06 0.2±0.004 1.02±0.04
The results are an average of three replicates with standard deviation. a : Untreated hydrolysate at pH 1.34. b : Hydrolysate neutralized with Ca(OH)2 at pH 6.8. c : Flocculated hydrolysate neutralized to pH 6.8 with Ca(OH)2. d : Hydrolysate generated after pre-hydrolysate concentration, hydrolysis, flocculation and neutralization to pH 6.8 with Ca(OH)2.
23
Table 2. Evolution of inhibitors and sugar concentration during treatment of hydrolysate by 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
combining flocculation and nanofiltration. Compounds concentration (g/L) Sugars Phenolic compounds Acetic acid HMF Furfural
Prehydrolysatea
Concentrated pre-hydrolysate b
Produced hydrolysatec
Flocculated hydrolysated
3.43±0.1
6.31±0.3
72.0±1.1
63.2±1
4.51±0.01
13.4±0.03
7.20±0.008
3.00±0.02
2.80±0.08 0.09±0.002 0.65±0.03
2.40±0.07 0.05±0.001 0.5±0.02
9.40±0.26 0.25±0.005 1.92±0.09
2.10±0.06 0.19±0.004 1.02±0.05
The results are an average of three replicates with standard deviation. a : Untreated pre-hydrolysate b : Concentrated pre-hydrolysate by nanofiltration using NF270 membrane. c : Corresponding concentrated hydrolysate generated after acid hydrolysis with sulfuric acid. d : Flocculated and neutralized concentrated hydrolysate at pH 6.8
24
Table 3. Reduction of phenolic compounds concentrations in generated hydrolysates with 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
several dilutions.
Hydrolysates
Dilution
Phenols
Xylose added
factor
(g/L)
(g/L)
0
2.30
32.0
3.5
0.66
52.0
7.5
0.31
56.3
13
0.17
57.9
0
1.65
33.3
1.5
1.10
35.7
2.5
0.66
49.3
3.5
0.30
52.4
9.5
0.17
57.2
Untreated hydrolysate
Flocculated hydrolysate
Pre-hydrolysate concentration by nanofiltraltion
0
3.00
0
2.5
1.20
6.94
+
5
0.60
9.47
Hydrolysate
10
0.30
10.7
17.5
0.17
11.2
Flocculation
25
Table 4. Concentration and percentage elimination of some phenolic compounds found in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
wood hydrolysate before and after flocculation with ferric sulfate. Untreated
Flocculated
hydrolysatea
hydrolysateb
(g/L)
(g/L)
Gallic acid
0.047±0.001
0
100
Vanilin
0.042±0.004
0.028±0.003
32
Syringaldehyde
0.19±0.003
0.049±0.001
55
Phenolic compounds
Removal (%)
The results are an average of three replicates with standard deviation. a : Untreated hydrolysate neutralized at pH 6.8. b : Flocculated and neutralized hydrolysate at pH 6.
26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Wood chip pre-hydrolysis
(1) Acid hydrolysis
(2)
(3)
Acid hydrolysis
Concentration by nanofiltration (NF270)
Coagulationflocculation
Acid hydrolysis
Coagulationflocculation
Dilution of hydrolysates
ABE fermentation by C. acetobutylicum ATCC 824 under static and uncontrolled pH conditions.
Figure 1.
27
A
6,2
3,0
6,0
2,5
5,8 5,6 5,4
1,5
1,0
pH
Acids (g/L)
2,0
5,2
pH Acetic acid Butyric acid
5,0
0,5
4,8 4,6
0,0 0
20
40
60
80
100
120
140
160
Fermentation time (h)
B
5 4 3 2 1 0
6 50 Acetone Ethanol Butanol Xylose Biomass Total ABE
4
40
2
Xylose (g/L)
6
60
Acids and solvants (g/L)
7
OD 600nm
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
30
0
20 0
20
40
60
80
100
120
140
160
Fermentation time (h)
Figure 2.
28
12 1,2
A 10 1,0
0,8
Acetone Ethanol Butnaol Total ABE Acetic Acid Butyric Acid
0,6
Acids (g/L)
Products (g/L)
8
6
4
0,4 2 0,2
0
0,0 0.17
0.3
2.3
Phenols (g/L)
7
12
6
10
5
8
4
6
Acetone Ethanol Butanol Total ABE Acetic Acid Butyric Acid
3
2
4
Acids (g/L)
Products (g/L)
B
2
1
0
0
-2 0.17
0.3
0.6
1.1
1.65
Phenols (g/L) 5
16
C
14
4
12 10
3
Acetone Ethanol Butanol Total ABE Acetic acid Butyric acid
2
8 6
Acids (g/L)
Products (g/L)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
4 2
1
0 0 0.17
0.3
0.6
1.1
3.0
Phenols (g/L)Fig.3.
Figure 3.
29
A 3,0 2.3 g/L of phenols 0.3 g/L of phenols 0.17 g/L of phenols
2,5
OD 600 nm
2,0
1,5
1,0
0,5
0,0
0
20
40
60
80
100
120
140
160
Fermentation time (h)
B
3,5
3,0 1.65 g/L of phenols 1.1 g/L of phenols 0.66 g/L of phenols 0.33 g/L of phenols 0.17 g/L of phenols
2,5
OD 600 nm
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
2,0
1,5
1,0
0,5
0,0 0
20
40
60
80
100
120
140
Fermentation Time (h)
Figure 4.
30
A Flocculated hydrolysate 30 Concentrated and flocculated hydrolysate
20 10
Untreated hydrolysate
0 1.1 0.6 0.3 0.17 Total phenolic compound (g/L)
Productivity (g/L.h)
B
Flocculated hydrolysate
0.06 0.04
Concentrated and flocculated hydrolysate
0.02
Untreated hydrolysate
0.00 1.1
0.6
0.3 0.17
Total phenolic compounds (g/L)
C
Concentrated and flocculated hydrolysate 0.3 Flocculated hydrolysate Yield (g/g)
Xylose consumption (g/L)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
0.2 0.1
Untreated hydrolysate
0 1.1
0.6
0.3
0.17
Total phenolic compounds (g/L)
Figure 5.
31
Highlights:
Hardwood pre-hydrolysates from dissolving pulp mill were used for ABE fermentation.
Flocculation was the most efficient method for inhibitors removal.
Detoxification increased the phenolics inhibitory level on ABE production up to 1.1 g/L.