Determination of the carbonyl groups in native lignin utilizing Fourier transform Raman spectroscopy

Article abstract of DOI:10.1016/S1386-1425(01)00693-X

A near-infrared Fourier transform Raman (NIR-FTR) spectroscopic technique was utilized to determine the chemical structure of lignin in a woody matrix. In the NIR-FTR spectra of coniferaldehyde and coniferyl alcohol, the Raman bands for the carbonyl group and the α, β unsaturated bond were detected at 1620 and 1660 cm^-1, respectively. These peaks were also found in the NIR-FTR spectra of chemically synthesized lignins, isolated lignin from conifer wood, and conifer wood meal. Upon the reduction of carbonyl groups in the lignin samples and wood meal, the band at 1620 cm^-1 disappeared; on the other hand, the band at 1660 cm^-1 remained unchanged. However, upon the oxidation of reduced lignin at the benzyl hydroxyl group using dicyanodichrolobenzoquinone, the band at 1620 cm^-1 clearly appeared, strongly suggesting that the band at 1620 cm^-1 can be assigned as a carbonyl marker band. The hydrogenation reaction optimized for the reduction of the unsaturated bond in lignin caused the disappearance of the band at 1660 cm^-1, indicating that the band at 1660 cm^-1 is an α, β unsaturated bond marker band. The change in carbonyl content during the wood decay process was also shown to be monitored using the Raman intensity of the carbonyl marker band. It was indicated that the NIR-FTR spectroscopic techniques were suitable analytical method for a rapid and nondestructive analysis of wood samples.

Full text of DOI:10.1016/S1386-1425(01)00693-X

Spectrochimica Acta Part A 58 (2002) 2213–2221
www.elsevier.com/locate/saa
Determination of the carbonyl groups in native lignin
utilizing Fourier transform Raman spectroscopy
Momoko Kihara, Miyuki Takayama, Hiroyuki Wariishi *, Hiroo Tanaka
Department of Forest and Forest Products Sciences, Kyushu Uni6ersity, 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
Received 16 October 2001; accepted 26 November 2001
Abstract
A near-infrared Fourier transform Raman (NIR-FTR) spectroscopic technique was utilized to determine the
chemical structure of lignin in a woody matrix. In the NIR-FTR spectra of coniferaldehyde and coniferyl alcohol, the
−
1
Raman bands for the carbonyl group and the a, b unsaturated bond were detected at 1620 and 1660 cm
,
respectively. These peaks were also found in the NIR-FTR spectra of chemically synthesized lignins, isolated lignin
from conifer wood, and conifer wood meal. Upon the reduction of carbonyl groups in the lignin samples and wood
−
1
−1
meal, the band at 1620 cm
disappeared; on the other hand, the band at 1660 cm
remained unchanged. However,
upon the oxidation of reduced lignin at the benzyl hydroxyl group using dicyanodichrolobenzoquinone, the band at
−
1
−1
1620 cm
clearly appeared, strongly suggesting that the band at 1620 cm
can be assigned as a carbonyl marker
band. The hydrogenation reaction optimized for the reduction of the unsaturated bond in lignin caused the
−
1
−1
disappearance of the band at 1660 cm , indicating that the band at 1660 cm
is an a, b unsaturated bond marker
band. The change in carbonyl content during the wood decay process was also shown to be monitored using the
Raman intensity of the carbonyl marker band. It was indicated that the NIR-FTR spectroscopic techniques were
suitable analytical method for a rapid and nondestructive analysis of wood samples. © 2002 Elsevier Science B.V. All
rights reserved.
Keywords: Lignin; Carbonyl group; Nondestructive analysis; Fourier transform Raman spectroscopy; Near-infrared excitation laser
of novel biotechnologies. A lot of work has been
1. Introduction
directed towards the improvement of the planting
stock via breeding, the selection, clonal propaga-
tion of elite genotypes, and production of geneti-
cally engineered trees as well as post-harvest
improvement utilizing fungal or enzymatic treat-
ments [1,2]. Lignocellulosic materials are abun-
dant and renewable and their utilization is highly
expected to be used as alternatives to fossil re-
sources [3]. For chemical or biological treatment
of wood, such as pulping process and alcohol
A growing demand for wood and the preserva-
tion of natural resources strongly pertains not
only to tree plantations but also to a development
Abbre6iations: DHP, dehydrogenated polymerizate (syn-
thetic lignin); FTR, Fourier transform Raman; NIR, near-
infrared.
*
Corresponding author. Tel./fax: +81-92-642-2993
E-mail address: hirowari@agr.kyushu-u.ac.jp (H. Wariishi).
1
386-1425/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1386-1425(01)00693-X

2
214
M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
production from cellulose, not only the quantity
but also the quality of lignin strongly affects the
properties of the products and the cost effi-
ciency. Rapid and nondestructive screening
methods are of great importance, since huge
numbers of progenies and chemically or biologi-
cally treated samples should be analyzed for the
quality control and the establishment of new
technology.
chemical methods, indicating that the Raman
method is a rapid, easy-handling, and nonde-
structive technique to characterize lignin in a
woody matrix.
2. Materials and methods
2.1. Wood samples
Vibrational studies such as infrared and UV–
vis spectroscopies have been providing the struc-
tural information of plant materials. Among
them, the Raman spectroscopic technique is ad-
vantageous since solid or powder samples can be
measured directly. However, a conventional Ra-
man spectroscopy using the excitation laser of
the visible region causes a fluorescent emission
interfering with the Raman measurement and
causes sample damage due to the high energy of
the visible laser. Using near-infrared (NIR) exci-
tation laser dramatically reduced these problems
Japanese cedar (Cryptomeria japonica D. Don)
was obtained from the experimental forest of
Kyushu University. The sapwood was ground
and fractionated using 100 and 42 mesh screens.
The wood meal was extracted with ethanol–tol-
uene (2:1 v/v) for 6 h in a Soxhlet apparatus [9].
Extractive-free wood meal (500 mg) was treated
with a solution of NaBH (100 mg) in 50% etha-
4
nol (50 ml) at 5 °C for 24 h with stirring [3].
After decantation, the wood meal was collected
and suspended in water (50 ml) containing acetic
acid (250 ml) and washed with water until the
reaction mixture was neutral.
[
4]. Since the NIR laser does not greatly excite
w(O–H) in water, the moisture in the samples
produces minimal effect on the spectra [5–7].
Therefore, the variation of samples from living
tissue was enhanced, and no complicated proce-
dures were required to obtain near-infrared
Fourier transform Raman (NIR-FTR) spectra of
lignocellulosic materials.
From these advantages, we utilized the NIR-
FTR spectroscopic technique to determine the
chemical structure of lignin. Previously, we have
assigned the Raman marker bands for syringyl
NaBH -reduced wood meal (300 mg) was
4
reoxidized with 2,3-dichloro-5,6-dicyano-1,4-ben-
zoquinone (DDQ) (150 mg) in anhydrous diox-
ane (200 ml) under reflux for 72 h [10]. The
color of wood meal after reoxidation was dark
brown. The wood meal was then washed with
ethanol and 1% NaOH.
2.2. Isolation of lignin
(
S-) and guaiacyl (G-) nuclei, and proposed a
Milled wood lignin (MWL) was isolated from
the cedar wood meal according to the method of
Bj o¨ rkman [11]. Wood meal was ground for 48 h
in toluene using the vibrational ball mill. MWL
was extracted with dioxane containing 10% wa-
ter, precipitated with ether, and then centrifuged.
The selective reduction of carbonyl groups in
rapid and nondestructive method to determine
the S/G ratios in wood [8]. In this study, we
examined the contributions of the carbonyl
groups and a, b unsaturated bonds found in the
side chain structure of lignin to the NIR-FTR
spectra. NIR-FTR measurement of the lignin
and a series of lignin model compounds, com-
bined with the chemical treatments of lignin
such as the reduction and reoxidation of sam-
ples, led us to assign the marker bands for the
carbonyl groups and unsaturated bonds in na-
tive lignin. The quantitative results of carbonyl
groups in lignin using the NIR-FTR were com-
pared with those obtained from the conventional
MWL with NaBH was conducted in 50% etha-
4
nol for 12 h as described [12]. The benzyl hy-
droxyl group in NaBH -reduced MWL was
4
reoxidized with DDQ in anhydrous dioxane for
20 h [13]. Hydrogenation was carried out in eth-
anol over palladium carbon (Pd; 5%) at room
temperature to reduce either carbonyl groups or
unsaturated bonds in alkyl side chain [14,15].

M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
2215
2.3. Model compounds
benzaldehyde) using horseradish peroxidase
Sigma) and H O , forming Alc-DHP [17,18], and
(
2
2
Coniferaldehyde was purchased from Aldrich.
Coniferyl alcohol was prepared by reducing
Ald-DHP [19,20], respectively. The yields were
60–70%. The colors of Alc-DHP and Ald-DHP
were pale brown and reddish-brown, respectively.
coniferaldehyde with NaBH [16]. Coniferyl alco-
4
hol was identified using GCMS as a ditrimethylsi-
lyl ether (m/z, M ; 324). They were purified using
+
2.4. Instrumentation
flash chromatography (hexane/ethyl acetate, 3/1)
before use. 4-Allyl-2-methoxyphenol (eugenol)
and 2-methoxy-4-(1-propenyl)phenol (isoeugenol)
were purchased from Tokyo Kasei and Wako
Pure Chem., respectively. The ring-conjugated un-
saturated bond in eugenol was hydrogenated in
ethanol over palladium carbon (Pd 5%) to pro-
duce 4-hydroxy-3-methoxyphenylpropane. The
hydrogenation reaction was carried out at room
temperature. The chemical structures of model
compounds were confirmed with MS spectrome-
try and shown in Scheme 1.
NIR-FTR spectra were collected using a Perkin
Elmer System 2000R spectrometer. The laser for
excitation was diode Nd:YAG operating at 1064
nm. All spectra were obtained in the 180° scatter-
ing geometry. A quartz beamsplitter and InGaAs
detector, averaging 200 scans with a resolution of
−
1
4 cm , and a laser power of 200 mW were used
to obtain FTR spectra in the range 3600–200
−
1
cm
or otherwise indicated. The curve-fitting
(
mixed Gaussian–Lorentzian) and peak area inte-
gration were performed using GRAMS/386TM
software (Galactic Co.). IR spectra of wood meals
and lignin model compounds were recorded using
a Perkin Elmer System 2000 Fourier transform
infrared (FTIR) spectrometer, with 64 or 100
Chemically synthesized lignin (Dehydrogenated
polymerizate; DHP) was prepared from either
coniferyl alcohol (4-hydroxy-3-methoxybenzyl al-
cohol) or coniferaldehyde (4-hydroxy-3-methoxy-
−
1
scans and a resolution of 4 cm
using the KBr
pellet method. FTIR spectra of wood meals were
also obtained by the diffuse reflectance method. A
GCMS analysis for lignin model compounds was
performed using a JEOL Atomass-15A II system
equipped with a fused silica column (NB-5; 30 m,
GL Science) at 70 eV and 120–300 °C (8 °C
−
1
min ). Samples were derivatized as previously
described [21].
3
. Results and discussion
Lignocellulosic materials found in wood and
other plant cell walls are abundant and renewable,
mainly consisting of cellulose, hemicellulose, and
lignin [10,22]. Cellulose is the most abundant bio-
material but its utilization is retarded by lignin,
which is one of the most recalcitrant biopolymers
on Earth [23]. Thus, the removal of lignin is a
very important issue not only for the pulping
industry but also for utilizing carbohydrates as a
starting material to produce sugars and alcohols
in an industrial scale to build up the renewable
and sustainable system (Green Industry) [2,24].
Scheme 1. Chemical structures of lignin model compounds
utilized in this study.

2
216
M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
For the chemical and biological treatment of
wood, structural information of lignin is essential.
It is known that structural features of lignin, such
as the ratio of syringyl nucleus to guaiacyl nucleus
and the amount of ring conjugated carbonyl
groups, exhibit a huge influence on the rate of
delignification reaction and the quality of the final
products in the pulping process [25]. It is also
known that the ring-conjugated carbonyl groups
in lignin cause the colorization of wood or the
final products, when they are exposed to heat or
UV light [26]. In the incipient stage of wood decay
by fungi, the content of the carbonyl groups
increase in proportion to the level of wood decay
[
27]. All these results suggest that ring-conjugated
carbonyl groups of lignin exhibit a strong effect
on the reactivity of lignin and the degree of
deterioration. Although, several spectral methods
to determine the total carbonyl content in lignin
have been utilized, such as the absorption spec-
troscopy using hydroxylamine hydrochloride or
sodium borohydride as a reducing reagent (Dm ;
r
reduction difference spectra), the accuracy of
these procedures is unpredictable because the
model compounds used as standards for UV de-
termination may not reflect the structure of native
lignin [10]. Furthermore, the Dm method cannot
r
Fig. 1. NIR-FTR spectra of coniferaldehyde (A), coniferyl
alcohol (B), isoeugenol (C), eugenol (D), 4-hydroxy-3-
methoxyphenylpropane (E), and 3-phenylpropionaldehyde (F).
be applied to the solid sample such as wood.
Isolation of lignin involves tedious and time-con-
suming treatments, which may cause autooxida-
tion to form a a-carbonyl group during the
extraction and purification steps. Development of
a rapid, nondestructive, and easy-handling proce-
dure to directly determine the ring-conjugated
carbonyl groups in wood would raise very unique
and useful techniques to estimate the chemical
features of lignin in a woody matrix.
−
1
cm . According to the structural features of two
compounds (Scheme 1), this band was thought to
be derived from the g-carbonyl group. However,
−
1
in a previous report, the band at 1661 cm
was
assigned to the g-carbonyl group using FTIR [29].
Then, those compounds were measured using
FTIR. Fig. 2A shows the FTIR spectrum of
coniferaldehyde, clearly showing the existence of
3.1. NIR-FTR spectra of lignin model compounds
−
1
the absorption bands at 1662 and 1618 cm . On
−
1
the other hand, the IR band at 1618 cm
was
Fig. 1A and B shows the NIR-FTR spectra of
disappeared as the g-carbonyl group was reduced
to form coniferyl alcohol (Fig. 2B). These obser-
vations clearly indicated that the band at 1620
coniferaldehyde and coniferyl alcohol, respec-
tively. Either coniferaldehyde or coniferyl alcohol
−
1
exhibited the bands at ꢀ1600 and 1660 cm
The bands at 1600 and 1585 cm
.
−
1
−1
were assigned
cm is most likely derived from the stretching of
to w(CꢀC) of an aromatic ring using the conven-
tional Raman [28]. Besides those bands, only
coniferaldehyde exhibited the extra band at 1620
the carbonyl group and that the band at ꢀ1660
−
1
cm
could be assigned to the ring-conjugated a,
b unsaturated bond.

M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
2217
To further characterize the contribution of the
ring-conjugated a, b unsaturated bond to the
NIR-FTR spectra, the NIR-FTR spectrum of
isoeugenol was measured. It showed a strong
−
1
band at 1656 cm
with the aromatic band at
−
1
1
611 cm , but no band was observed in 1620
−
1
cm
region (Fig. 1C). In addition, no Raman
band was detected in the region of 1655–1660
−
1
cm
in the NIR-FTR spectra of either eugenol
or 4-hydroxy-3-methoxyphenylpropane, contain-
ing no ring-conjugated a, b unsaturated bond
(
Fig. 1D and E). These results supported that the
−
1
Raman band found at 1655–1660 cm
could be
assigned to the ring-conjugated a, b unsaturated
bond. Phenylpropionaldehyde exhibited a band at
−
1
1
ꢀ
628 cm , supporting that the Raman band at
−
1
1620 cm
is the marker band for the carbonyl
stretching (Fig. 1F). It is also suggested from the
FTR spectra of model compounds (Fig. 1) that
the Raman shift of the carbonyl marker band is
dependent on whether the carbonyl is conjugated
to either an a, b unsaturated bond or to an
aromatic ring.
Fig. 3. NIR-FTR spectra of Ald-DHP (A) and Alc-DHP (C),
and their second derivative (B and D).
3.2. NIR-FTR spectra of polymeric lignin
Lignin in a woody matrix is an amorphous
aromatic polymer formed from phenylpropanoid
precursors, such as coniferyl and sinapyl alcohols,
via random radical coupling reactions [22]. Since
the chemical analysis of lignin requires tedious
and time-consuming steps, we initiated the utiliza-
tion of NIR-FTR spectroscopic techniques. Syn-
thetic polymer lignin samples (DHPs) were
prepared. Fig. 3A and C show the NIR-FTR
spectra of DHPs synthesized from coniferalde-
hyde (Ald-DHP) and coniferyl alcohol (Alc-
DHP), respectively. Only the NIR-FTR spectrum
−
1
of Ald-DHP exhibited the band at 1623 cm
,
supporting that this Raman band was derived
from the carbonyl group. These observations cor-
Fig. 2. FTIR spectra of coniferaldehyde (A) and coniferyl
alcohol (B).

2
218
M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
respond well with a previous report, which indi-
cated that the carbonyl groups are much more
abundant in Ald-DHP than those in Alc-DHP
Raman characteristics of the reoxidized DHP
were very similar to those of the original Ald-
DHP, there were some differences. The intensity
−
1
−1
[
30]. The band at 1625 cm
was also found in
ratio of the peak at 1622 cm
band at 1596 cm
to the aromatic
−
1
the NIR-FTR spectra of MWL and extractive-
free wood meal, albeit with weaker intensity (data
not shown).
seen in the spectrum of reoxi-
dized DHP was larger than that of the peak at
−1
1
623–1594 cm
seen in Ald-DHP. Since Ald-
DHP contains the g-carbonyl group and the reox-
idized DHP should contain the a-carbonyl group,
we tentatively concluded that the Raman shift of
either ring-conjugated a-carbonyl and g-carbonyl
groups appeared at the same region of 1620–1625
3
.3. Assignment of the carbonyl band in lignin
To further characterize the Raman bands
−
1
derived from lignin found at the 1500–1700 cm
region, a computer-fit analysis was applied to the
Raman data. The precise position and number of
the Raman bands were estimated from the second
derivative of the spectra. These values are re-
quired for the spectral curve-fit (peak component)
analysis to determine the detailed assignment of
the Raman bands [31]. The second derivatives of
the NIR-FTR spectra of Ald-DHP and Alc-DHP
clearly show the band position and number exist-
−
1
ing in 1500–1700 cm
region (Fig. 3B and D).
Then, the curve-fit analysis was performed, show-
ing a very good correlation between experimental
and computer-calculated curves (curves a and b in
Fig. 4). Using these analytical methods, the effect
of the reduction of carbonyl groups using NaBH4
on DHP was examined. NaBH has been reported
4
to reduce carbonyl groups in lignin structure to
corresponding hydroxyl groups but to leave the
ethylene group intact [12]. Fig. 4A and B show
the NIR-FTR spectra of Ald-DHP and its re-
duced sample. These spectra were subjected to a
peak component analysis. The reduction of all
carbonyl groups caused complete disappearance
−
1
of the bands at 1623 cm . The Raman charac-
teristics of NaBH -reduced DHP were almost
4
identical to Alc-DHP (Fig. 3C and Fig. 4B).
Then, reoxidation of NaBH -reduced Ald-DHP
4
using DDQ was carried out. DDQ has been
known to oxidize the benzyl hydroxyl group to
the corresponding carbonyl group [13,16]. In the
spectrum of DDQ-reoxidized DHP, a strong band
−
1
at 1622 cm
was again observed, which was
assigned to the carbonyl marker band as de-
Fig. 4. NIR-FTR spectra of Ald-DHP (A), A reduced with
scribed above. The aromatic CꢀC stretching band
NaBH (B), and B reoxidized with DDQ (C). Line a indicates
4
−
1
was seen at 1596 cm , which was observed at
the experimental spectrum and line b indicates the calculated
spectrum from curve-fit data.
−
1
1594 cm
in Ald-DHP (Fig. 4C). Although the

M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
2219
−
1
cm
and that the ring-conjugated a, b unsatu-
−1
rated bond is detected at 1660 cm , which was
almost unchanged during NaBH -reduction and
4
DDQ-reoxidation processes. It is also suggested
that the intensity of the carbonyl marker bands
depends on how strongly they conjugated to other
CꢀC bonds such as an aromatic ring and a, b
unsaturated bond. In addition, any bands derived
from DDQ did not disturb the measurement of
the NIR-FTR spectra of DHPs (data not shown).
The effect of NaBH -reduction and DDQ-reox-
4
idation treatments on native lignin was also exam-
ined. In the NIR-FTR spectra of MWL and wood
−
1
meal, the band at 1620–1625 cm
disappeared
upon NaBH reduction, and appeared again upon
4
reoxidation with DDQ (data not shown). From
these results, we concluded that the ring conju-
gated carbonyl groups are detected at ꢀ1625
−
1
cm
in the NIR-FTR spectra.
3.4. Assignment of the ring-conjugated h, i
unsaturated bond
Fig. 5. NIR-FTR spectra of MWL after reduction with
NaBH (A), and A further hydrogenated (B). Line a indicates
4
In the early stage of this study, we found that
the experimental spectrum and line b indicates the calculated
spectrum from curve-fit data.
−
1
the band at 1660 cm , once assigned to carbonyl
vibration from FTIR study [29], should be as-
signed to the a, b unsaturated bond in NIR-FTR
spectroscopy using model compounds. Therefore,
it ought to further characterize the a, b unsatu-
rated bond in lignin using the FTR technique.
The effect of the reduction of the a, b unsaturated
bond on the FTR spectra was examined using
Pd:C-catalyzed hydrogenation. Through this re-
duction reaction, both the unsaturated bonds and
the carbonyl groups in the lignin side chains are
completely reduced, but leave the aromatic ring
unchanged. In the NIR-FTR spectra of lignin
model compounds, the a, b unsaturated bond was
pearance of this band was also observed against
the hydrogenation of Ald-DHP, suggesting that
−
1
the peak at ꢀ1660 cm
is assigned to the
ring-conjugated a, b unsaturated bond in lignin.
3.5. Monitoring the change in ring conjugated
carbonyl content
To apply the NIR-FTR spectroscopic tech-
niques for monitoring the carbonyl group con-
tents in lignin, the Raman intensity ratio of
−
1
1
625–1600 cm
was compared with the results
−
1
detected at 1656–1660 cm (Fig. 1). The band at
obtained from the Dm method (Table 1). The
r
−
1
−1
ꢀ
1660 cm
was also observed in the NIR-FTR
aromatic marker band at 1600 cm
was utilized
spectra of Alc-DHP, Ald-DHP, MWL, and wood
meal. Fig. 5A shows the NIR-FTR spectrum of
as an internal standard to minimize the deviation
of the Raman intensities among the samples, since
both aromatic and carbonyl marker bands were
derived from lignin. The conventional method
revealed that the increase of the carbonyl groups
in MWL and Alc-DHP when treated with DDQ.
The ratio of the Raman bands for the carbonyl
NaBH -reduced MWL, where this band was still
4
−
1
observed even though the band at 1620 cm
completely disappeared. However, upon the hy-
drogenation of the sample, the band at 1660
−
1
cm
disappeared (Fig. 5B). The complete disap-

2
220
M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
Table 1
Contents of the ring conjugated carbonyl groups determined
by the conventional Dmr method and FTR spectroscopic
method
Lignin samples Dm method
FTR method
r
(
Number/100 C9
(1625/1600 cm−1)
units)
MWL
No treatment
DDQ-oxidation 17.4
13.2
1.50
1.85
Alc-DHP
No treatment
0
0
DDQ-oxidation 4.2
1.31
−
1
marker (1625 cm ) and the aromatic marker
(
Fig. 6. Time course of the change in Raman intensity ratio
−
1
−1
1600 cm ) also increased as the samples were
during wood decay process. 1660/1600 cm
decrease of a, b unsaturated bond (ꢀ) and 1620/1600 cm
indicates the
−
1
DDQ-oxidized. Since the Raman carbonyl marker
band counts both a- and g-carbonyl groups, fur-
ther consideration of lignin structure would be
still required for quantitative determination of the
ring-conjugated carbonyl group. However, it was
also indicated that the change in relative contents
of the ring-conjugated carbonyl groups in lignin
could be monitored by measuring the intensity
ratio of carbonyl marker band and the aromatic
marker band (Table 1).
indicates the increase of carbonyl contents (ꢁ) in wood sam-
ple (C. japonica) during fungal decay by C. 6ersicolor.
−1
1
660 cm
are assigned to the marker bands for
carbonyl groups and for the a, b unsaturated
bond, respectively. Using these marker bands, the
change in the relative contents of the carbonyl
groups and a, b unsaturated bonds can be moni-
tored. It was strongly suggested that NIR-FTR
spectroscopic technique would be useful for the
rapid and nondestructive determination of the
ring-conjugated carbonyl groups of lignin in a
woody matrix.
Upon fungal treatment of wood, it is known
that the content of the ring conjugated carbonyl
groups increases, and the content of the ring
conjugated a, b unsaturated bond decreases [23].
−
1
Therefore, for the bands at 1620 and 1660 cm
,
it would be useful to know the degree of wood
decay or deterioration. Fig. 6 shows the time
course of the change in Raman intensity ratio of
Acknowledgements
−
1
1620/1600 and 1660/1600 cm , indicating the
This research was supported by the Interna-
tional Joint Research Program from the New
Energy and Industrial Technology Development
Organization (NEDO) of Japan via Research In-
stitute of Innovative Technology for the Earth
(RITE) (to H. W.).
increase of carbonyl contents and the decrease of
unsaturated bonds in the early stage of fungal
decay of wood. During the early stage of wood
decay, the content of aromatic moiety could be
assumed to be unchanged.
4. Conclusion
References
From the Raman study of a series of lignin
model compounds, DHPs, MWL, and wood
meal, it was indicated that the bands at 1620 and
[
1] K.G. Eldridge, J. Davidson, C. Hardwood, Eucalypt
Domestication and Breeding, Clarendon Press, Oxford,
UK, 1993.

M. Kihara et al. / Spectrochimica Acta Part A 58 (2002) 2213–2221
2221
[
[
[
2] M. Kuwahara, Proceedings of International Symposium
on the Efficient Use of Biomass, Kyoto, 2001, p. 21.
3] M. Slesser, C. Lewis, Biological Energy Resources, Wiley,
New York, 1979.
4] R.P. Carey, Biochemical Applications of Raman and
Resonance Raman Spectroscopies, Academic Press, New
York, 1982.
[18] T. Yamasaki, K. Hata, T. Higuchi, Mokuzai Gakkaishi
19 (1973) 299.
[19] W.J. Connors, C.L. Chen, J.C. Pew, J. Org. Chem. 35
(
1970) 1920.
20] T. Ito, Proceedings of 1996 Tappi Pulping Conference,
996, p. 111.
21] H. Wariishi, K. Valli, M.H. Gold, Biochemistry 28 (1989)
017.
22] K.V. Sarkanen, C.H. Ludwig, Lignins: Occurrence, For-
mation, Structure and Reactions, Wiley-Intersciences,
New York, 1971.
[
[
[
1
6
[
[
[
5] R.H. Atalla, U.P. Agarwal, Science 227 (1985) 636.
6] U.P. Agarwal, R.H. Atalla, Planta 169 (1986) 325.
7] U.P. Agarwal, R.H. Atalla, Proceedings of the Eighth
International Symposium on Wood and Pulping Chem-
istry, Helsinki, vols. II and III, 1995, p. 67.
[
[
23] R.L. Crawford, Lignin Biodegradation and Transforma-
tion, Wiley-Intersciences, New York, 1981.
24] T. Haraguchi, Wood Chemistry, Academic Press, Japan,
[
[
8] M. Takayama, T. Johjima, T. Yamanaka, H. Wariishi, H.
Tanaka, Spectrochim. Acta 53 (1997) 1621.
9] J.D. Wilson, Forest Prod. J. 11 (1961) 260.
1
993.
[
[
[
25] J. Grier, I. Nor e´ n, Holzforschung 36 (1982) 123.
26] J. Gierer, S.Y. Lin, Svensk Papperstidn. 75 (1972) 233.
27] T. Higuchi, Adv. Enzymol. Related Areas Mol. Biol. 34
[
10] S.Y. Lin, C.W. Dence, Methods in Lignin Chemistry,
Academic Press, Berlin, Heidelberg, 1992.
11] A. Bj o¨ rkman, Svensk Papperstidn. 59 (1956) 477.
12] J. Marton, E. Adler, K.I. Persson, Acta Chem. Scand. 15
1961) 384.
13] H.D. Becker, E. Adler, Acta Chem. Scand. 15 (1961) 218.
14] J.L. Pepper, Y.W. Lee, Can. J Chem. 47 (1969) 723.
[
[
(
1971) 207.
28] U.P. Agarwal, R.H. Atalla, I. Forssk a˚ hl, Holzforschung
9 (1995) 300.
[
[
(
4
[
[
[
[
[
29] I. Forssk a˚ hl, J. Janson, Nord. Pulp Pap. Res. J. 7 (1992)
.
48.
15] E. Adler, J. Marton, Acta Chem. Scand. 15 (1961) 357.
16] P. Fenn, T.K. Kirk, J. Wood Chem. Tech. 4 (1984) 131.
17] M. Wayman, T.I. Obiaga, Can. J. Chem. 52 (1974) 2102.
[30] T. Higuchi, T. Ito, T. Umezawa, T. Hibino, D. Shibata, J.
Biotechnol. 37 (1994) 151.
[31] W.F. Maddams, Appl. Spectroscopy 34 (1980) 245.