Bioresource Technology 151 (2014) 244248

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Lignin removal and benzenealcohol extraction effects on lignin

measurements of the hydrothermal pretreated bamboo substrate
X.J. Ma, S.L. Cao, X.F. Yang, L. Lin, L.H. Chen, L.L. Huang
College of Material Engineering, Fujian Agriculture and Forestry University, Fuzhou 350002, PR China

h i g h l i g h t s

 The extractives of the pretreated substrate increased with the reaction severity.
 Benzenealcohol extraction gave rise to a low lignin measurement.
 Tappi standard should be modied.
 Lignin removal was the results of lignin degradation and migration.

a r t i c l e i n f o a b s t r a c t

Article history: Lignin content of hydrothermal pretreated bamboo chips was determined by the two methods: TAPPI
Received 5 September 2013 standard method (222om-06) and TAPPI standard method without benzenealcohol extraction (BAE).
Received in revised form 21 October 2013 The results showed that including BAE resulted in lower Klason lignin (KL) and acid soluble lignin
Accepted 23 October 2013
(ASL) measurements in the prehydrolyzed substrate, that is to say, BAE removed parts of KL and ASL.
Available online 1 November 2013
Therefore, the TAPPI standard method should be modied by omitting the BAE for lignin measurements
of pretreated substrate. The following lignin removal analysis suggested that lignin was removed from
Keywords:
the bamboo substrate during pretreatment by a combination of degradation reaction and deconstruction;
Benzenealcohol extraction (BAE)
Lignin
thereafter the pseudo lignin generated in the hydrothermal pretreatment and condensation reaction
Pseudo lignin between the lignin fragments accounted for the later KL increase.
Hydrothermal pretreatment 2013 Elsevier Ltd. All rights reserved.

1. Introduction production. Lignin is a key factor that hinders enzymatic hydrolysis

through physical inhibition and adsorption. Lignin is subjected to
To alleviate the energy crisis and climate change based on the chemical reaction and physical changes during pretreatment. One
use of fossil energy, biofuel production from renewable resources of the expected lignin chemical reaction pathways in dilute or
has received signicant attention (Ragauskas et al., 2006; Clark, hydrothermal pretreatment is the cleavage of b-O-4 linkage (Cao
2007). For sustainable development, lignocellulosics (including et al., 2012); the cleavage gives rise to a high content of lignin with
agricultural residues, forestry wood and energy crops) are regarded low molecular weight during hydrothermal pretreatment
as more appropriate feedstock for bioethanol production; lignocel- (Leschinsky et al., 2008a). In addition, polymerization reactions
lulosic ethanol is also called second-generation bioethanol (Blanca of lignin with itself or with carbohydrate degradation products also
and Juan, 2008). All lignocellulosic materials consist primarily of occur depending upon the reaction conditions (Leschinsky et al.,
cellulose, hemicelluloses, and lignin. In a heterogeneous enzymatic 2008; Bujanovic et al., 2012). Moreover, as lignin melts, it migrates
reaction, the presence of hemicelluloses and lignin and complex through the cell wall and forms droplets or coalesces on the
three-dimensional structure (called as recalcitrance) make the cel- biomass surface; this migration and redistribution of lignin has
lulose inaccessible to hydrolytic enzymes (Himmel et al., 2007). been viewed as an obstacle for enzyme hydrolysis (Bujanovic
Pretreatment can remove hemicelluloses/lignin and disrupt the et al., 2012). Additionally, lignin degradation products in the
structure of the lignocellulosic material to increase cellulose acces- hydrolysate impede the dissolution of hemicelluloses because of
sibility to enzymes, and thus improve the hydrolysis and fermenta- their tendency to form deposits (Leschinsky et al., 2008b). Further,
tion efciency. Employing pretreatment before enzymatic several studies have demonstrated that the Klason lignin (KL)
hydrolysis is essential in a multi-step process for cellulosic ethanol content of dilute acid pretreated biomass was higher than that of
the starting material (Mao et al., 2010; Chen et al., 2010; Jung
Corresponding author. Tel.: +86 15959041385; fax: +86 0591 83715175. et al., 2010; Sannigrahi et al., 2008). It has been suggested that
E-mail address: [email protected] (L.L. Huang).
aromatic materials formed by the combination of carbohydrate

0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2013.10.078
 X.J. Ma et al. / Bioresource Technology 151 (2014) 244248 245

and lignin degradation products during acid pretreatment could be Controller

responsible for the increase of KL content (Hu et al., 2012). Katah-

ira, Sannnigrahi and Hu veried that dilute acid hydrolysis of holo- Thermometer for calibration

cellulose or cellulose could give rise to an increase of KL Oil bath digester sealing cover
measurements of the hydrolyzed substrate and the KL content in- Thermocouple
creased with increasing reaction severity (Sannigrahi et al., 2008;
Hu et al., 2012; Katahira et al., 2013). This gain of KL is termed Heat conducting oil
pseudo-lignin and could account for the surprising increase of KL
measurements and negatively affect enzymatic hydrolysis. As lig-
Cooking-pot jacket
nin is an obstacle to the enzymatic saccharication of biomass,
accurate determination of lignin content is important to the quan-
titative analysis of lignin effects. Cooking pot
Generally, the TAPPI standard method is used to determine lig-
nin content of biomass. Briey, benzenealcohol extraction is the
rst step in order to decrease the interference of extractives. There- Rotation axis

after, extractives-free biomass is depolymerized by hydrolyzing

holocellulose with 72% sulfuric acid and followed by hydrolysis
of the dissolved polysaccharide in 3% boiling sulfuric acid. The
residual solids are considered as acid insoluble or Klason lignin
(KL) while the acid soluble lignin (ASL) in the ltrate is determined Heat coil

by UVvisible spectroscopy at 205 nm.

Because of the complex chemical reactions and physical
Fig. 1. Schematic diagram of prehydrolysis apparatus.
changes that lignin undergoes during pretreatment, it is doubtful
that the TAPPI standard (222om-06) is applicable for lignin deter-
mination of pretreated substrate. Thus, a systematic hydrothermal
2.4. Analytical methods
pretreatment of bamboo chips was performed. The TAPPI standard
(222om-06) and TAPPI standard without benzenealcohol extrac-
The yields of pretreatment were determined by the weight and
tion (BAE) were used to determine the lignin content (KL and
moisture content of the starting bamboo chips and treated bamboo
ASL) and the differences between the two methods were explored
chips.
by quantitative analysis of the lignin measurements. In addition,
The extractives of benzene95% alcohol (2:1) extraction were
lignin removal was thoroughly analyzed by combination of sub-
determined by TAPPI standard method T264 cm-97.
strate, hydrolysate liquor and precipitate.
KL content was determined by TAPPI standard 222om-06 and
the TAPPI standard method without BAE. In order to compare the
2. Methods changes of KL removal, KL values were based on the original mate-
rial, i.e., the KL measurements were calculated by multiplying KL
2.1. Materials measurements of the pretreated substrate with the pretreatment
yield. ASL in the ltrate was determined by UVvisible spectros-
Bamboo chips (Dendrocalamopsis oldhami) were provided by a copy at 205 nm. Multiple measurements (34 runs) were con-
forest center (Fujian province, China, 2011). The bamboo chips ducted for all characterization techniques, and a mean value was
with size of (40 2)  (15 1)  5 mm were selected as the start- reported. The error margin of the KL and ASL measurements was
ing materials. 0.1% to 0.4% and 0.03% to 0.07%, respectively.
The pretreatment hydrolysate liquor was dripped onto a copper
mesh and vacuum dried for about 24 h. Images of the colloid par-
2.2. Pretreatment experiments ticles in hydrolysate liquor were taken with a Transmission Elec-
tron MicroscopeTEM (JEM1010, Japan).
18 experiments were carried out in an oil-bathed digester Bamboo and pretreated bamboo powder was characterized by
which equipped with 10 cooking pots. The reactions were carried solid state cross-polarization magic angle spinning (CPMAS) 13C
out at 160 to 180 C with the liquor/solid ratio 3:1 (180 g bamboo NMR (Advance III500, Bruker, Switzerland) operating at a fre-
chips) for times up to 180 min. The cooking pots were put into the quency of 125.75 MHz.
oil-bathed digester starting the reaction when the oil temperature Field Emission Scanning Electron Microscope (FE-SEM, FEI Nova
reached the target temperature; the cooking pots started to rotate NANOSEM 230, operated at an accelerating voltage of 15 kV) was
while the rotate button was switched on. The schematic diagram of used to characterize the morphology of the holocellulose and pre-
the digester was shown in Fig. 1. At the end of the reaction, the treated holocellulose samples.
cooking pots were taken out and cooled in cold water. The solids
and hydrolysate were separated by ltration with 80 mesh bags.
3. Results and discussion
The lignin content of the solid substrate was analyzed while the l-
trate was subjected to TEM analysis.
3.1. BAE effects on lignin measurements

2.3. Holocellulose separation and hydrolysis From Fig. 2, the KL and ASL determined by the TAPPI standard
method showed that the KL decreased gradually at the initial stage
Holocellulose was obtained from bamboo powder following the and then showed an increase in the later stage (P170 C). In con-
procedure described by Cao et al. (2012). The extractives-free pow- trast, ASL decreased rst and then plateaued at extended pretreat-
der was subjected to oxidation reactions for removing lignin. Holo- ment times. Furthermore, the KL and ASL content determined by
cellulose pretreatment was conducted in the oil-bathed 25 mL TAPPI standard method was 1035% and 1575% lower than that
hydrothermal reactor at 170 C for up to 240 min. determined without BAE. Increasing reaction severity increased
246 X.J. Ma et al. / Bioresource Technology 151 (2014) 244248

o
2.4
Temperature, C
30 BAE M, 160 2.1
BAE M, 170
27 BAE M, 180 1.8

24 1.5
KL (%)

ASL (%)
1.2
21
0.9
18
0.6
15 0.3

12
(a) 0.0
(b)
0 30 60 90 120 150 180 0 30 60 90 120 150 180
Pretreating time (min) Pretreating time (min)

Fig. 2. The effects of BAE on measured lignin content (BAE: lignin quantication by TAPPI method T 222om-06 with BAE; M: lignin quantication by modied TAPPI method
(no BAE); (a) KL; (b) ASL).

the difference between the two methods; the difference in mea-

21
surements of KL and ASL were as much as 14% and 2%, respectively.
Lignin is subjected to acidic homolytic cleavage during hydrother-
18
mal pretreatment resulting in a substantial decrease in the molec-
ular weight (MW) of the lignin in the residual substrate
(Leschinsky et al., 2008a). Lignin with a low molecular weight 15
Extractives (%)
(LMW) could be extracted by BA from the pretreated substrate
thus causing the discrepancy between the two analytical methods. 12
The increase of KL may be tied to carbohydrate and lignin deg-
radation products, or pseudo-lignin. The results from holocellulose 9 o
hydrolysis are presented in Table 1. It was easily observed that KL Temperature, C
increased with pretreatment time and as with bamboo, BAE re- 160
6
sulted in a lower KL measurement. From these results, BAE could 170
180
extract 5060% pseudo lignin from pretreated holocellulose but
3
only 1035% lignin from pretreated bamboo substrate. The MW
of lignin or pseudo lignin might account for this difference. The 0 30 60 90 120 150 180
droplets (pseudo-lignin) appeared on the surface of the pretreated Pretreating time (min)
holocellulose substrate and the size of these droplets increased
with pretreatment time (Suppl. A). After the BAE procedure, the Fig. 3. Extractives content of pretreated bamboo as a function of pretreatment time
majority of these droplets disappeared but a few still remained and temperature.
on the most severely pretreated holocellulose (170 C, 240 min).
From Hus study, weight-average MW of pseudo-lignin extracted
from dilute acid pretreated poplar cellulose and holocellulose ran- content of the pretreated substrate ranged from 5% to 20%, depend-
ged from 3440 to 5970 g/mol, depending on reaction time and the ing on reaction temperature and time. At 160 C the extractives
original material (180 C, 4060 min, 0.10.2 M H2SO4); the MW of content increased linearly with reaction time, while at higher tem-
pseudo lignin increased with the reaction severity (Hu et al., 2012). peratures (P170 C) the content rst increased linearly before
Given Hus results, it may be that BAE can only remove low MW dropping gradually. When considered in tandem with the differ-
pseudo-lignin. ences in measure lignin content described above, it is reasonable
to conclude that a fraction of these extractives are KL and ASL. High-
er temperature and longer duration favor lignin migration and deg-
3.2. Extractives and its probably origin
radation (Ma et al., 2013), which could enhance the BAE effect.
However, during longer pretreatment condensation reactions of lig-
Generally, woody plant extractives are diverse substances, such
nin fragments could become more signicant (Leschinsky et al.,
as resin acid, fatty acid, turpenoid compounds and alcohols; most
2008a) leading to less soluble lignin that cannot be extracted by BA.
of these substances are soluble in water or neutral organic solvent
Additionally, the extractives were higher than the differences of
(Sheshmani, 2013). In bamboo (D. oldhami) material, there were
lignin measurements between the two methods (Suppl. B); inferring
only 3.6 0.2% extractives. Fig. 3 showed that the extractives
that in addition to lignin, extraction also removed other substances
from the pretreated substrate. In Yans study, monosaccharides ac-
Table 1
counted for 56  60% of the extractives recovered by extraction with
KL measurements of the pretreated holocellulose substrate.
95% ethanol (Yan et al., 2010). During hydrothermal pretreatment,
Time (min) Yield (%) TAPPI standard Without BAE the carbohydrates are degraded to produce monosaccharides such
Extractives KL KL as glucose and xylose. Although most of these monosaccharides
0 100 Trace 1.1 0.2 1.0 0.2 are dissolved in the liquor, a fraction may be retained in the pre-
30 79.5 0.2 1.0 0.2 0.6 0.3 1.5 0.2 treated substrate and subsequently removed by BAE.
90 73.3 0.2 1.3 0.3 0.8 0.3 2.0 0.3 In conclusion, due to the extractive effect of BA, the TAPPI stan-
150 71.7 0.2 2.2 0.3 1.2 0.3 3.1 0.2 dard method is not appropriate for lignin analysis of pretreated
240 72.3 0.2 3.6 0.4 2.8 0.3 6.2 0.4
bamboo therefore the BAE step should not be used.
 X.J. Ma et al. / Bioresource Technology 151 (2014) 244248 247

3.3. Lignin removal counts for the signicant lignin removal observed by Liu and
Wyman.
Typically, hydrothermal pretreatment can cause lignin frag- At the later stage of hydrolysis, the KL content exceeded the ori-
mentation and subsequent dissolution. Therefore, following hydro- ginal KL, indicating that, some acid insoluble substance formed
thermal pretreatment, lignin is partitioned among three locations: during pretreatment. The gain of KL probably originated from
(1) because hydrothermal pretreatment only causes slight deligni- pseudo-lignin, which is the product of reactions between carbohy-
cation, most of lignin is retained in the pretreated substrate, (2) a drate and lignin degradation products. From Table 1, it is observed
small amount of lignin is suspended or dissolved in the hydrolysate that pseudo-lignin formation increases with pretreatment time.
liquor, and (3) a small amount of lignin precipitates onto the wall
of the reactor. It is apparent that the latter two parts are responsi- 4. Conclusions
ble for lignin removal.
Fig. 2b shows that the quantity of ASL of the pretreated sub- BAE has a signicant effect on the lignin measurements of pre-
strate was lower than that of the starting material and decreased treated substrate. Benzenealcohol extraction removed a fraction
with pretreatment time until nally plateauing. ASL is basically of lignin from the pretreated substrate; therefore lignin measure-
composed of low molecular weight degradation products and ments determined by TAPPI standard method (222om-06) are low-
hydrophilic derivatives of lignin (Yasuda et al., 2001). The lignin er than measurements made without extraction. Therefore, TAPPI
degradation product with low MW dissolved in the hydrolysate li- standard without BAE should be used for lignin determination of
quor would account for the decrease of ASL in the substrate. pretreated substrate. KL decreased rst due to the depolymeriza-
The KL decreased rst and then increased in the course of pre- tion reaction and migration of lignin; KL in the pretreated substrate
treatment (Fig. 2a). The phenol groups of lignin (assigned to then increased due to pseudo lignin produced from degradation
147 ppm) increased and non-phenolic groups decreased products and condensation reaction between dissolved lignin
(152 ppm) clearly showing cleavage of lignininter-unit linkages fragments.
(Suppl. C) (Wikberg and Maunu, 2004). Theoretically, the cleavage
of arylether bonds within the lignin structure resulted in an in- Acknowledgements
crease of phenol groups and depolymerization of lignin macromol-
ecules. The cleavage of b-O-4 linkage together with other The authors gratefully acknowledge the support from Natural
degradation reaction probably resulted in lignin with low MW as Science Foundation of Fujian Province, China, 2012J01065 and For-
soluble species in the liquor. estry Department of Fujian Province, China, [2012]2 and Scientic
Additionally, Selig and Donohoe suggested that thermalchem- Research Foundation of Graduate School of Fujian Agriculture and
ical pretreatment at temperatures exceeding the melting tempera- Forestry University, 324-1122yb002.
ture of lignin allow the lignin within the cell wall matrix to migrate
and to form spheres on contacting with water (Selig et al., 2007; Appendix A. Supplementary data
Donohoe et al., 2008). During the pretreatment, a portion of these
droplets could separate from the cell wall matrix by the washing Supplementary data associated with this article can be found, in
effect of the liquor as the reactor rotated and ultimately fall into the online version, at http://dx.doi.org/10.1016/j.biortech.2013.10.
the hydrolysate liquor. The hydrolysate liquor could then be con- 078.
sidered as a colloid containing lignin droplets as colloidal particles.
It was suggested that droplets were only present in the liquor dur- References
ing the initial stages of pretreatment (Suppl. D). The occurrence of
the droplets in the liquor likely depended on the characteristics of Blanca, A.L., Juan, L.T., 2008. Second-generation biofuels and local bioenergy
the liquor (pH, zeta potential) (Yasarla and Ramarao, 2012); the de- systems. Biofuels Bioprod. Bioren. 2, 455469.
Bujanovic, B.M., Goundalkar, M.J., Amidon, T.E., 2012. Increasing the value of a
gree of hemicelluloses degradation and condensation reactions of biorenery based on hot-water extraction: lignin products. Tappi J. 11, 1926.
lignin fragments may also contribute to this phenomenon. From Cao, S.L., Pu, Y.Q., Studer, M., Wyman, C., Ragauskas, A.J., 2012. Chemical
the Leschinsky study, treatment of prehydrolysate liquor under transformations of populus trichocarpa during dilute acid pretreatment. RSC
Adv. 2, 1092510936.
hydrothermal pretreatment conditions in the absence of wood res-
Chen, W.H., Tu, Y.J., Sheen, H.K., 2010. Impact of dilute acid pretreatment on the
idue led to formation of deposits on the wall of the reactor; further structure of bagasse for bioethanol production. Int. J. Energy Res. 34, 265274.
analysis conrmed that the deposit consisted mainly of lignin deg- Clark, J.H., 2007. Green chemistry for the second generation biorenery-sustainable
radation products with a higher average MW than that of the de- chemical manufacturing based on biomass. J. Chem. Technol. Biotechnol. 82,
603609.
posit separated directly from the hydrolysate by centrifugation Donohoe, B.S., Decker, S.R., Tucker, M.P., Himmel, M.E., Vinzant, T.B., 2008.
(Leschinsky et al., 2008b). Condensation reactions between the lig- Visualizing lignin coalescence and migration through maize cell walls
nin droplets and subsequent precipitation might be the reason for following thermochemical pretreatment. Biotechnol. Bioeng. 101, 913925.
Himmel, M.E., Ding, S.Y., Johnson, D.K., 2007. Biomass recalcitrance: engineering
the disappearance of the droplets at the later stages of pretreat- plants and enzymes for biofuels production. Science 315, 804807.
ment. Also, deposits formed from droplets or dissolved lignin frag- Hu, F., Jung, S., Ragauskas, A., 2012. Pseudo-lignin formation and its impact on
ments would precipitate back onto the substrate surface and thus enzymatic hydrolysis. Bioresour. Technol. 117, 712.
Jung, S., Foston, M., Sullards, M.C., Ragauskas, A.J., 2010. Surface characterization of
partially cause the increase of KL content. It could be concluded dilute acid pretreated Populus deltoides by ToF-SIMS. Energy Fuels 24, 1347
that lignin removal not only depends on lignin dissolution as solu- 1357.
ble species, but also on the lignin migration as colloidal particles in Katahira, R., Sluiter, J.B., Schell, D.J., Davis, M.F., 2013. Degradation of carbohydrates
during dilute sulfuric acid pretreatment can interfere with lignin measurements
the hydrolysate liquor. An earlier study by Liu and Wyman found in solid residues. J. Agric. Food Chem. 61, 32863292.
that lignin removal was limited to less than 35% for batch runs Leschinsky, M., Zuckersttter, G., Weber, H.K., Patt, R., Sixta, H., 2008a. Effect of
but ow-through operation increased lignin removal to as high autohydrolysis of Eucalyptus globulus wood on lignin structure. Part 1:
comparison of different lignin fractions formed during water prehydrolysis.
as 75% when a high ow rate was used (Liu and Wyman, 2003).
Holzforschung 62, 645652.
Although the signicant effects of lignin removal were observed, Leschinsky, M., Zuckersttter, G., Weber, H.K., Patt, R., Sixta, H., 2008b. Effect of
the relationship between lignin removal and ow rate were not autohydrolysis of Eucalyptus globulus wood on lignin structure. Part 2: inuence
fully addressed. Given recent studies, lignin migration combined of autohydrolysis intensity. Holzforschung 63, 653658.
Liu, C., Wyman, C.E., 2003. The effect of ow rate of compressed hot water on xylan,
with the washing effect of hot water at high ow rate probably ac- lignin, and total mass removal from corn stover. Ind. Eng. Chem. Res. 42, 5409
5416.
248 X.J. Ma et al. / Bioresource Technology 151 (2014) 244248

Mao, J.D., Holtman, K.M., Franqui, V.D., 2010. Chemical structures of corn stover and stems retards enzymatic hydrolysis of cellulose. Biotechnol. Prog. 23, 1333
its residue after dilute acid prehydrolysis and enzymatic hydrolysis: insight into 1339.
factors limiting enzymatic hydrolysis. J. Agric. Food Chem. 58, 1168011687. Sheshmani, S., 2013. Effects of extractives on some properties of bagasse/high
Ma, X.J., Cao, S.L., Lin, L., Luo, X.L., Chen, L.H., Huang, L.L., 2013. Surface density polypropylene composite. Carbohydr. Polym. 2013 (94), 416419.
characterizations of bamboo substrates treated by hot water extraction. Wikberg, H., Maunu, S.L., 2004. Characterisation of thermally modied hard- and
Bioresour. Technol. 136, 757760. softwoodsby 13C CPMAS NMR. Carbohydr. Polym. 58, 461466.
Ragauskas, A.J., Williams, C.K., Davison, B.H., et al., 2006. The path forward for Yan, J., Hu, Z., Pu, Y., Brummer, E.C., Ragauskas, A.J., 2010. Chemical compositions of
biofuels and biomaterials. Science 311, 484489. four switch grass populations. Biomass Bioenergy 34, 4853.
Sannigrahi, P., Ragauskas, A.J., Miller, S.J., 2008. Effects of two-stage dilute acid Yasarla, L.R., Ramarao, B.V., 2012. Dynamics of occulation of lignocellulosic
pretreatment on the structure and composition of lignin and cellulose in hydrolyzates by polymers. Ind. Eng. Chem. Res. 51, 68476861.
loblolly pine. Bioenerg. Res. 1, 205214. Yasuda, S., Fukushima, K., Kakehi, A., 2001. Formation and chemical structures of
Selig, M.J., Viamajala, S., Decker, S.R., Tucker, M.P., Himmel, M.E., Vinzant, T.B., 2007. acid-soluble lignin in sulfuric acid treatment time and acid-soluble lignin
Deposition of lignin droplets produced during dilute acid pretreatment of maize content of hardwood. J. Wood Sci. 47, 6972.