Chemical modifications of lignin for the preparation of macromers containing cyclic carbonates

Article abstract of DOI:10.1039/c6gc01028h

An epoxidized lignin derivative was prepared directly inserting epichlorohydrin on the phenolic functionalities. The epoxidized lignin was then converted to cyclic carbonates through the coupling reaction of CO₂ with the oxirane rings. Imidazolium based ionic liquids, acting as both solvents and catalysts, were successfully employed in the carbonation reaction. Moreover, the ionic liquid was reused up to three times without significant loss in activity. Finally, an exhaustive spectroscopic characterization was carried out on the epoxidized and carbonated lignins by quantitative ³¹P and ¹³C NMR analyses.

Full text of DOI:10.1039/c6gc01028h

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. Salanti, L. Zoia
and M. Orlandi, Green Chem., 2016, DOI: 10.1039/C6GC01028H.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/greenchem

Page 1 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
Chemical Modifications of Lignin for the Preparation of Macromers Containing
Cyclic Carbonates
Anika Salantia, Luca Zoiaa*, Marco Orlandia
Department of Earth and Environmental Sciences, University of Milano-Bicocca, Piazza della
Scienza, 1 Milan, I-20126 Italy.
* Corresponding author [telephone (0039) 02ꢀ6448ꢀ2709; fax (0039) 02ꢀ6448ꢀ2935; eꢀmail
luca.zoia@unimib.it]
Abstract. An epoxidized lignin derivative was prepared directly inserting epichlorohydrin on the
phenolic functionalities. The epoxidized lignin was then converted to cyclic carbonates through the
coupling reaction of CO2 with the oxirane rings. Imidazolium based ionic liquids, acting both as
solvents and catalysts, were successfully employed in the carbonation reaction. Moreover, the ionic
liquid was reused up to three times without significant loss in activity. Finally, an exhaustive
spectroscopic characterization was carried out on the epoxidized and carbonated lignin by
quantitative 31P and 13C NMR analysis.

Green Chemistry
Page 2 of 28
View Article Online
DOI: 10.1039/C6GC01028H
Introduction
Lignin is described as an irregular and tridimensional polymer whose structure depends on the
botanical source. The main functional groups encountered in lignin chemistry are hydroxyl (both
aliphatic and phenolic), methoxyl, carboxyl, and carbonyl moieties in different amount, depending
upon the type of extraction employed. As the principal sideꢀstream of the papermaking, bioethanol,
and agro industries, alternative uses of lignin have been sought by the whole scientific community
since the early ‘70s.1 Even if the specific process to which lignocellulose is subjected causes a
modification of the polymer characteristics, the abundance of lignin of different type as a waste
product has made such material an attractive proposition for the preparation of valueꢀadded
products.
The valorization of lignin through integrated approaches is of crucial importance to match the
requirements of the biorefinery concept, which is a comprehensive utilization of all those
lignocellulosic materials consisting of the residual nonꢀfood parts of current crops or other nonꢀfood
sources (known as secondꢀgeneration feedstock) by converting these biomass into fuels, energy,
chemicals and materials.2
Nowadays, the most explored fields for a convenient industrial application of lignin are the
incorporation into polymer compositions as a physical additive3,4 and, especially, the use of the
lignin carbon skeleton as a building block. Among the plethora of examples reported in literature,
lignin as a macromer has been introduced in phenolꢀformaldehyde and epoxy resins, as well as in
polyurethane and natural polyesters formulations5ꢀ9. At present, the methods to synthesize ligninꢀ
based resins consist of: modifying lignin derivatives directly and modifying lignin derivatives in
order to improve their reactivity.
Epoxidized lignin is a remarkably reactive derivative, useful both as a green substitute for epoxy
resins components and as intermediate in the synthesis of other ligninꢀbased macromer by means of
nucleophilic addition and insertion reactions.10,11 For example, the synthesis of cyclic carbonates
through the addition of CO2 to epoxy groups is an atomꢀeconomical reaction approach, and

Page 3 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
provides valueꢀadded monomers for the preparation of poly(hydroxy)urethanes, avoiding the use of
toxic isocyanates12. The introduction of cyclic carbonate groups in the lignin macromolecule is so
far not reported in literature.
In this work, two technical lignins, namely a soda pulping herbaceous lignin (HBL) and a hardwood
kraft lignin (HWKL), were modified through the direct insertion of epoxide groups by reaction
between lignin phenolic groups and epichlorohydrin. The epoxide groups were then converted to
cyclic carbonates through the fixation of CO2. Different solvents and experimental conditions were
tested to optimize the conversion to cyclic carbonates. In particular, in agreement with recent
reports, the use of ionic liquids was successfully investigated. In fact, the use of ionic liquids as
both solvent and catalyst, or their combination with metal salts, has been exploited for the synthesis
of cyclic carbonates13,14. Currently, imidazolium based ILs, are reported as the best choice for the
synthesis of cyclocarbonates from epoxide and CO2 due to the high solubility of CO2 observed in
this kind of ILs15,16. In this study, 1ꢀallylꢀ3ꢀmethylimidazolium based ionic liquids were selected as
solvents of choice as they are widely exploited for the dissolution of lignin and lignocellulose in
general17,18
.
An exhaustive spectroscopic characterization was finally performed on the epoxidized and
carbonated lignin samples. 31P NMR spectroscopy was used for a quantitative measurement of the
epoxide groups bonded to lignin after reaction with epichlorohydrin. The characteristic IR
absorption band of the cyclic carbonate moiety at 1790 cmꢀ1 was used to qualitatively evaluate the
extent of cyclic structures formed after reaction with CO2. Several qualitative 13C NMR analyses
were performed to confirm the formation of both the glycidyl and cyclic carbonate lignin derivative.
Once established the best set of experimental conditions for the carbonation reaction, quantitative
13C NMR spectra were acquired in the presence of an internal standard (dimethyl carbonate) to
retrieve the actual amount of cyclic structures formed. Moreover, glycidyl and cyclic carbonate
derivatives of lignin model compound (i.e., phenol, guaiacol, 2,6ꢀdimethoxyphenol, 4ꢀhydroxyꢀ3ꢀ
methoxybenzyl alcohol, and vanillic acid) were prepared and spectroscopically characterized to

Green Chemistry
Page 4 of 28
View Article Online
DOI: 10.1039/C6GC01028H
evaluate the 13C NMR absorbance range of aliphatic alcohol, pꢀhydroxyphenyl, guaiacyl, syringyl,
and carboxylic acid moieties, respectively.
Experimental
Reagents and Materials. All reagents and solvents used were purchased from SigmaꢀAldrich and
used as received without further purification. Lignin specimens were supplied during the project
ProLignin, EraNets WoodWisdom/Bioenergy, 2012ꢀ14. The samples under investigation are a
herbaceous lignin (purity 98.2%, total phenolic content 3.55 mmolgꢀ1) obtained from a soda
pulping process sold under the trademark Protobind1000®, and a hardwood lignin recovered after a
kraft pulping process (purity 96.7%, total phenolic content 3.90 mmolgꢀ1).
General Procedure for the Epoxidation of Model Compound. 1g of a specific phenolic model
compound (phenol 10.6 mmol, guaiacol 8.0 mmol, 2,6ꢀdimethoxyphenol 6.5 mmol, 4ꢀhydroxyꢀ3ꢀ
methoxybenzyl alcohol 6.5 mmol, and vanillic acid 5.9 mmol) was dissolved in acetone (15 mL) in
the presence of epichlorohydrin (1:5 molar ratio of total hydroxyls moiety to epoxide; 4.2, 3.1, 2.6,
5.1, and 4.6 mL of epichlorohydrin for the reaction with phenol, guaiacol, 2,6ꢀdimethoxyphenol, 4ꢀ
hydroxyꢀ3ꢀmethoxybenzyl alcohol, and vanillic acid, respectively). The mixture was warmed to
70°C and then 50 mL of an aqueous solution of NaOH (1:5 molar ratio of hydroxyls moiety to
hydroxide; 2.12, 1.6, 1.3, 2.6, and 2.4 g NaOH for phenol, guaiacol, 2,6ꢀdimethoxyphenol, 4ꢀ
hydroxyꢀ3ꢀmethoxybenzyl alcohol, and vanillic acid, respectively) was added dropwise under
stirring. After 3 hours the mixture was allowed to cool to room temperature. Epoxidized phenol19,
epoxidized guaiacol20,and epoxidized 2,6ꢀdimethoxyphenol were extracted in petroleum ether (2
times, 30 mL each). Epoxidized 4ꢀhydroxyꢀ3ꢀmethoxybenzyl alcohol and epoxidized vanillic acid
were instead extracted in diethyl ether (2 times, 30 mL each). The collected organic phase was
washed with deionized water, anhydrified over Na2SO4, and filtered. After rotary evaporation of the
solvent a thick oil was left in the case of monoꢀhydroxy compounds and vanillic acid, whereas a
solid was obtained from 4ꢀhydroxyꢀ3ꢀmethoxybenzyl alcohol. Epoxidized vanillic acid was purified

Page 5 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
on silica column chromatography (Merck, pore size 60Å , 230ꢀ400 mesh) by elution with ethyl
acetateꢀpetroleum ether 1:1. Yields were included between 30ꢀ40% for epoxidized monoꢀhydroxy
model compounds and 4ꢀ6% for epoxidized 4ꢀhydroxyꢀ3ꢀmethoxybenzyl alcohol and vanillic acid.
1
Structure and purity of epoxidized model compounds were assessed by IR and H NMR analysis.
Characteristic absorptions of the different epoxide chains were recognized by 13C NMR analysis.
1
2-(2,6-dimethoxybenzyl)oxirane. Colourless oil, > 90% purity, yield 33%. H NMR (CDCl3, 100
MHz): δ = 6.98 (t, J = 8.42 Hz, 1H, Ar), 6.57 (d, J = 8.42 Hz, 2H, Ar), 4.10 (dd, J = 4.10 and 11.29
Hz, 1H, OCHHCH), 3.99 (dd, J = 5.85 and 11.29 Hz, 1H, OCHHCH), 3.83 (s, 3H, OCH3), 3.36
(m, 1H, CH2CHCH2), 2.79 (dd, J = 4.39 and 4.74 Hz, 1H, CHCHHO), 2.61 (dd, J = 2.67 and 5.03
Hz, 1H, CHCHHO) ppm . 13C NMR (DMSOꢀd6, 125 MHz): δ = 153.4 (Cq OMe), 136.8 (Cq), 123.9
(CAr), 105.2 (CAr), 74.0 (OCH2CH), 56.0 (OMe), 50.4 (CH2CH), 44.6 (CHCH2O) ppm.
Oxiran-2-ylmethyl-3-methoxy-4-(oxiran-2-ylmethoxy)benzoate.Colourless oil, > 90% purity, yield
1
4.3%. H NMR (CDCl3, 100 MHz): δ = 7.67 (d, J = 1.97 Hz, 1H, Ar), 7.56 (dd, J = 1.97 and 8.45
Hz, 1H, Ar), 6.93 (d, J = 8.50 Hz, 1H, Ar), 4.63 (dd, J = 3.09 and 12.29 Hz, 1H, OCHHCH PhOH),
4.33 (dd, J = 3.29 and 11.35 Hz, 1H, OCHHCH Acid), 4.13 (dd, J = 6.27 and 12.29 Hz, 1H,
OCHHCH PhOH), 4.07 (dd, J = 5.68 and 11.41 Hz, 1H, OCHHCH Acid), 3.92 (s, 3H, OCH3), 3.40
(m, 1H, CH2CHCH2 PhOH), 3.33 (m, 1H, CH2CHCH2 Acid), 2.90 (m, 2H, CHCHHO PhOH and
CHCHHO Acid), 2.77 (dd, J = 2.65 and 4.92 Hz, 1H, CHCHHO PhOH), 2.71 (dd, J = 2.65 and 4.93
Hz, 1H, CHCHHO Acid) ppm. 13C NMR (DMSOꢀd6, 125 MHz): δ = 166.1 (C=O), 152.3 (Cq
PhOH), 149.2 (CAr), 123.6 (Cq OMe), 112.6 (CAr), 69.8 (OCH2CH PhOH), 65.3 (OCH2CH Acid),
56.0 (OMe), 49.8 (CH2CH PhOH), 49.5 (CH2CH Acid), 44.7 (CHCH2O PhOH), 44.6 (CHCH2O
Acid) ppm.
General Procedure for the Carbonation reaction of Model Compounds. 200 mg of a selected
epoxidized model compound was solubilized with dimethyl formamide (10 mL) and gently loaded
into the pressure vessel. After purging the reactor with CO2 for 10 minutes the pressure was brought
up to 20 bar and the reactor heated to 80°C. After 20 hours reaction time, carbonated model

Green Chemistry
Page 6 of 28
View Article Online
DOI: 10.1039/C6GC01028H
compounds of phenol21, guaiacol22, 2,6ꢀdimethoxyphenol, and vanillic acid were recovered by
extraction in methylene dichloride (2 times 30 mL each). The organic phase was washed twice with
deionized water, anhydrified over Na2SO4, filtered, the solvent removed by rotary evaporation and
finally exsiccated under vacuum. All yields were included between 50ꢀ60%. Structure and purity of
carbonated model compounds was evaluated by IR and 1H NMR analysis. Characteristic absorption
of the different cyclocarbonate chains were established through 13C NMR analysis.
1
4-(2,6-dimethoxy benzyl)-1,3-dioxolan-2-one. Colourless oil, > 90% purity, yield 58%. H NMR
(CDCl3, 100 MHz): δ = 6.95 (t, J = 8.26 Hz, 1H, Ar), 6.52 (d, J = 8.15 Hz, 2H, Ar), 4.85 (m, 1H,
CH2CHCH2), 4.72 (dd, J = 6.41 and 8.38 Hz, 1H CHCHHO), 4.51 (t, J = 8.43 Hz, 1H, CHCHHO),
4.11 (dd, J = 4.44 and 9.26 Hz, 2H, OCH2CH), 3.77 (s, 3H, OCH3) ppm. 13C NMR (DMSOꢀd6, 125
MHz): δ = 155.1 (C=O), 153.2 (Cq OMe), 136.4 (Cq), 124.4 (CAr), 105.2 (CAr), 75.1 (OCH2CH),
71.3 (CH2CHCH2), 66.3 (OCH2CH), 56.0 (OMe) ppm.
(2-oxo-1,3-dioxolan-yl)methyl-3-methoxy-4-[(2-oxo-1,3-dioxolan-4-yl)methoxy]benzoate.Yellowish
oil, > 90% purity, yield 35%. 13C NMR (DMSOꢀd6, 125 MHz): δ = 166.2 (C=O Acid), 155.0 (C=O
cyclic carbonate PhOH), 154.6 (C=O cyclic carbonate Acid), 150.4 (Cq PhOH), 147.6 (Cq OMe),
128.7 (Cq Acid), 122.5 (CAr), 115.6 (CAr), 112.3 (CAr), 74.5 (OCH2CH PhOH), 74.0 (OCH2CH
Acid), 68.7 (CH2CHCH2 PhOH), 66.2 (OCH2CH Acid), 66.1 (OCH2CH PhOH), 63.7 (CH2CHCH2
Acid), 56.0 (OMe) ppm.
General procedure for the reaction of lignin with epichlorohydrin. Lignin (1 g, total phenolic
content 3.55 and 3.90 mmolgꢀ1 for herbaceous and hardwood kraft lignin, respectively) was
solubilized under vigorous stirring into 40 mL of NaOH aqueous solution (1:20 molar ratio of
phenolic moiety to hydroxide; 2.9 and 3.1 g NaOH for herbaceous and hardwood kraft lignin,
respectively). 10 mL of acetone was then added, along with epichlorohydrin (1:20 molar ratio of
phenol moiety to epoxide; 5.6 and 6.1 mL of epichlorohydrin for herbaceous and hardwood kraft
lignin, respectively). The mixture was heated to 70°C and reacted for 3 hours. After cooling to room
temperature, epoxidized lignin was precipitated out in cold acidic water (0.2 M HCl, 250 mL,

Page 7 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
additional concentrated HCl was added if needed to reach slightly acidic pH), filtered on a sintered
glass crucible (size 4), thoroughly washed with deionized water and lyophilized. Mass recovery was
about 70ꢀ75% for both the technical lignin.
General Procedure for the Carbonation reaction of Lignin. In a typical experimental procedure,
200 mg of epoxidized lignin was dissolved in a suitable amount of a specific solvent (20 mL of
organic solvents, 4 g of ionic liquid). When an ionic liquid (IL) was employed as solvent,
epoxidized lignin and the IL were placed into a screw top glass vial equipped with a magnetic
stirrer and the mixture heated with a heat gun (metabo H 16ꢀ500, 2 heating cycles of approximately
45 sec) and stirred overnight at 40°C to achieve complete solubility. The sample was then gently
loaded into the pressure vessel and the reactor purged with CO2 for 10 min. The pressure was
loaded up to 20 bar and the reactor heated to 80°C. After 20 hours reaction time, carbonated lignin
was precipitated out in excess cold water, filtered on a sintered glass crucible (size 4), thoroughly
washed with deionized water, and finally lyophilized. Carbonated lignin mass recovery was around
90% for both the herbaceous and hardwood kraft lignin.
Ionic Liquid Recycling. The IL 1ꢀallylꢀ3ꢀmethyl imidazolium chloride (amimCl) collected after the
carbonation reactions was regenerated according to the following procedure. The aqueous phase
containing amimCl was extracted with ethyl acetate (2 times, 40 mL each), and water first rotary
evaporated at 80°C and then exsiccated under vacuum for 2 h at 90°C, to give an extremely thick
gel. The required amount of regenerated ionic liquid (4 g) was further exsiccated under vacuum for
2 h at 90°C into the screw top vial used for the carbonation reaction, conveniently equipped with a
screw cap with pierced aperture and a glass stopcock.
FTIR Analysis. Fourier Transform Infrared Spectroscopy was used for a rapid screening of the
reactions outcome and to evaluate the relative yield of different lignin carbonation experiments by
the % absorbance of the characteristic cyclic carbonates band at 1790 cmꢀ1 after spectrum
normalization. The analyses were performed on a Nicolet iS10 spectrometer (Thermo Scientific)
equipped with a Smart iTR™ Total Attenuated Reflectance (ATR) sampling accessory (total scan

Green Chemistry
Page 8 of 28
View Article Online
DOI: 10.1039/C6GC01028H
16, range 4000ꢀ700 cmꢀ1, resolution 1 cmꢀ1). The data reported for the estimation of the
cyclocarbonation yield are the mean of 4 to 6 experiments; maximum standard deviation 0.010 a.u.,
maximum standard error 0.005 a.u.
31P-NMR Analysis. Accurately weighed lignin samples (30 mg) were dissolved in a pyridineꢀ
deuterated chloroform stock solution (1.6:1 v/v, 700 ꢁL) containing 1 mg/mL of chromium(III)
acetylacetonate, [Cr(acac)3] as relaxation agent. 100 ꢁL of an internal standard solution of endoꢀNꢀ
hydroxyꢀ5ꢀnorborneneꢀ2,3ꢀdicarboximide (eꢀHNDI, 121.5 mM, CDCl3/pyridine 4.5:0.5) was then
added. 2ꢀchloroꢀ1,3,2ꢀdioxaphospholane (RI, 100 ꢁL) was used as phosphorus derivatizing agent
for the detection and quantification of primary and secondary aliphatic alcohols23, whereas 2ꢀ
chloroꢀ4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaphospholane (RII, 100 ꢁL) was used for the detection and
quantification of phenolic and acidic moieties24. 31PꢀNMR spectra were recorded on 800 µl samples
with a Bruker Avance 500 MHz instrument at room temperature. The chemical shifts were referred
to the waterꢀphosphytilating agent adduct signal (121.5 ppm) when RI was used, and to the internal
standard signal (152 ppm) when RII was employed. Relaxation delay of 5 s was used between the
scans (90° pulse angle). Line broadening of 3 Hz was applied to FIDs before Fourier transform. For
each spectrum, typically 100 scans were accumulated.
13C NMR Analysis. Broadband decoupled 13CꢀNMR spectra were recorded on a Bruker Avance 500
MHz spectrometer at room temperature on 800 µl samples dissolved in deuterated
dimethylsulfoxide (DMSOꢀd6). The chemical shifts were referred to the methoxyl group signal (56
ppm). For qualitative evaluations relaxation delay of 4 sec was used between the scans (90° pulse
angle). Line broadening of 0.8 Hz was applied to FIDs before Fourier transform. For each spectrum,
typically about 6'000 scans were accumulated. Quantitative 13C NMR spectra were acquired for
carbonated herbaceous and hardwood kraft lignin in the presence of dimethyl carbonate (25 ul,
≥99%) as internal standard. Relaxation delay of 10 sec was applied and approximatively 10’000
scans were accumulated. The assignment of predominant signals was based on the chemical shift
data obtained from the synthesized lignin model compounds and from literature data25,19
.

Page 9 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
Results and Discussion
Synthesis and Characterization of Epoxidized Lignin
A soda pulping herbaceous lignin (HBL) and a hardwood kraft lignin (HWKL) were subjected to
epoxide derivatization through nucleophilic substitution in the presence of excess epichlorohydrin.
The large excess of epichlorohydrin employed is dictated by need to achieve a thorough epoxide
functionalization of the lignin molecule minimizing undesired side reactions such as the formation
of high molecular species.26 Lower phenols:epichlorohydrin ratios were tested but they were found
to be not adequate to achieve the required epoxidation degree.
The formation of the glycidyl derivative was at first assessed by ATRꢀFTIR analysis (Fig.1,
middle). The presence of oxirane rings is confirmed by the appearance of weak to medium band at
760 and 835 cmꢀ1, by the higher intensity of the CꢀH stretching vibrations, and by the modification
of the ꢀOH stretching profile.
Figure 1 Top to bottom: ATRꢀFTIR spectra of reference, epichlorohydrin epoxidized, and
carbonated herbaceous lignin.

Green Chemistry
Page 10 of 28
View Article Online
DOI: 10.1039/C6GC01028H
31P NMR analyses were carried out on reference lignin and after treatment with epichlorohydrin.
Since a general underestimation of aliphatic hydroxyls groups was observed using 2ꢀchloroꢀ4,4,5,5ꢀ
tetramethylꢀ1,3,2ꢀdioxaphospholane (RII), labile hydroxyls quantification was accomplished in
duplicate on each sample: RII was dedicated to the quantification of phenolic and acidic moieties,
whereas 2ꢀchloroꢀ1,3,2ꢀdioxaphospholane (RI) was exploited for the quantification of aliphatic
alcohols. Moreover, RII affords a better resolution of the phenolic signals whereas RI provides a
better resolution of the peaks originated by reaction with primary and secondary alcohols.
For both the herbaceous and hardwood kraft lignin, a complete functionalization of the phenolic
moiety was achieved (Fig. 2, phosphorus reagent RII, below). More interestingly, the phosphorus
derivatizing agent is also able to react with the epoxide groups on lignin as it is evident by the
presence of two different strong and broad signals in the aliphatic region of the 31P NMR spectrum
(Fig. 2, phosphorus reagent RI, above). The signal recorded between 134.5 and 133 ppm lie in the
same region as the preponderant primary alcoholics groups in the reference lignin, but it is more
populated on account of the formation of hydroxypropylic structure after reaction with RI. On the
other hand, the signal recorded at lower fields between 137.0 and 134.5 ppm, in the secondary
alcohols region, should be exclusively attributed to the hydroxypropylic derivative27. It is worth
noticing that after phosphorus derivatization the formation of both a primary and a secondary
phosphitylated alcohol derivative is possible, indeed.

Page 11 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
Figure 2 Comparison between 31P NMR spectra of reference and epichlorohydrin epoxidized
hardwood kraft lignin after reaction with 2ꢀchloroꢀ1,3,2ꢀdioxaphospholane (RI, above) and 2ꢀ
chloroꢀ4,4,5,5ꢀtetramethylꢀ1,3,2ꢀdioxaphospholane (RII, below).

Green Chemistry
Page 12 of 28
View Article Online
DOI: 10.1039/C6GC01028H
The upper part of Table 1 reports the distribution of labile hydroxyls for reference and epoxidized
HBL and HWKL as found by 31P NMR analysis. After the epoxidation phenolic groups were no
longer detected, whereas the amount of primary and, especially, secondary aliphatic alcohols
dramatically increased. As mentioned above, two different kinds of aliphatic hydroxyl moieties can
be discerned: naturally occurring sideꢀchain aliphatic hydroxyls and hydroxypropylic alcohols
ensuing from epoxide rings opening after phosphitylation. The quantification of the
hydroxypropylic moiety is given in the penultimate line of Table 1. It was obtained by subtracting
to the total amount of aliphatics detected for the epoxidized lignins an opportunely corrected value
of the total amount of aliphatics found for the reference lignins. The aforementioned correction has
to be intended as a percentage mass correction accounting for changes in the molecular weight of
lignin C9 units after chemical functionalization. For the purpose of this work, a C9 unit containing a
methoxyl group, a phenolic hydroxyl, and a primary and a secondary aliphatic hydroxyl was
considered. The empirical formula of the illustrated structure is C10H15O4 and its molecular weight
is ca. 197 gmolꢀ1; the corresponding epoxidized unit is described by the formula C13H19O5 and its
molecular weight is ca. 253 gmolꢀ1. Corrected values of total aliphatics in reference lignins are
reported in the third to last line of Table 1. The resulting quantification of hydroxypropylic moieties
attained for the herbaceous and hardwood kraft lignin was 2.25 and 2.47 mmolgꢀ1, respectively.
As a rule of thumb, it could be stated that the amount of oxirane rings introduced is nearly the same
as the disappeared phenols, i.e., the total amount of phenolic groups quantified in reference lignin. It
is reminded that also this value has to be conveniently corrected according to lignin molecular
weight variation after the epoxidation reaction (Table 1, last line). The values associated to this
correction are 2.77 and 3.04 mmolgꢀ1 for HBL and HWKL, respectively. Moreover, it is highlighted
that all the hydroxypropylic units are associated to phenols functionalization. In fact, as confirmed
by the study of lignin model compounds, no reaction with epichlorohydrin was observed at the
aliphatic hydroxyls sites whereas a negligible yield in epoxidized carboxylic acids was achieved.
Discrepancies between theoretical (HBL 2.77 and HWKL 3.04 mmolgꢀ1) and experimental (HBL

Page 13 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
2.25 and HWKL 2.47 mmolgꢀ1) values for the ensuing hydroxypropylic units and the reduced
amount of carboxylic acid detected after reaction with epichlorohydrin are caused by the of low
molecular weight lignin fractions during the purification process. In fact, intrinsic disadvantages
connected to lignin multifunctional nature and possible side reactions, such as oxirane ring opening,
confers an enhanced hydrophilic character to the substrate resulting in a limited mass recovery
(approximately 75% of the starting mass).
HBL
Epoxy
HWKL
Epoxy
31P NMR Assignment (mmol/g)
HBL
HWKL
1.31
0.92
0.39
3.55
1.96
1.05
0.54
1.04
1.02
ꢀ
3.27
1.50
1.77
n.d
1.73
1.20
0.53
3.90
2.87
1.03
n.d.
0.63
1.35
ꢀ
3.82
2.24
1.58
n.d
Total aliphatics (RI)
1° alcohol (RI)
2° alcohol (RI)
Total PhOH (RII)
S-OH + Cond. (RII)
G-OH (RII)
n.d.
n.d.
n.d.
0.18
ꢀ
n.d.
n.d.
n.d.
0.33
ꢀ
P-OH (RII)
COOH (RII)
Tot aliph reference (mass corrected)
Hydroxypropylic alcohols
Tot PhOH (mass corrected)
2.25
ꢀ
2.47
ꢀ
2.77
3.04
Table 1 31P NMR quantification of labile hydroxyl groups for reference and epoxidized herbaceous
and hardwood kraft lignin (phosphorus derivatizing agent used in brackets). Epoxide groups
quantification is labelled as “hydroxypropylic alcohols”. Mass correction accounts for the effect of
molecular weight variations after the epoxidation reaction on lignin hydroxyls distribution.

Green Chemistry
Page 14 of 28
View Article Online
DOI: 10.1039/C6GC01028H
31CꢀNMR analysis also confirmed the accomplishment of lignin epoxidation. In fact, strong
absorptions at about 44, 50, and 70ꢀ75 ppm were diagnostic for the presence of oxirane rings (vide
infra).
Carbonation Reaction of Epoxidized Lignin, Synthesis and Characterization
The feasibility of the insertion reaction of CO2 on the epoxidized lignin was then assessed.
At first, different organic solvents suitable for lignin dissolution, i.e., dimethylformamide (DMF),
dimethylsulfoxide (DMSO), methyl acetoacetate (Meacac), dioxane, and methylethyl ketone
(MEK) in association with a commonly employed salt catalyst (potassium iodide, KI) were tested.
Typical reaction conditions applied for carbonation reactions in organic solvent medium are, if not
otherwise stated: 20 bar pressure, 20 hours reaction time, 80°C as reaction temperature, 10% w/w
catalyst loading, and 1% w/w consistency28. For a rapid screening of the reaction outcome, the %
absorbance of the cyclic carbonate peak at 1790 cmꢀ1, as detected from ATRꢀFTIR analysis, was
considered (Fig. 1). 13C NMR analyses were performed on selected samples to evaluate the reaction
progress by the disappearance of characteristic epoxides carbon signal (spectra not reported). As a
general observation, IR absorbance of the characteristic cyclic carbonates peak (1790 cmꢀ1) above
0.30 a.u. means a complete conversion of epoxides groups according to 13C NMR analysis.
Results of the carbonation experiments are reported in Table 2. No significant conversion occurred
without the presence of any catalyst under these reaction conditions (entry 1). Dioxane and MEK
(entries 5 and 6) were not able to solubilize epoxidized lignin. Among DMF, DMSO, and Meacac,
in all cases in association with KI as catalyst (entries 2ꢀ4), the best performance in term of
carbonation yield was obtained with DMF, which was therefore selected as organic solvent of
choice for further investigation. Other catalysts were then tested: 1,8ꢀdiazabicyloundecꢀ7ꢀene
(DBU) and N,Nꢀdimethylaminopyridine (DMAP) were selected as metalꢀ and halideꢀfree catalysts29
while the onium salt tetrabutylammonium bromide (TBAB) was assayed both in its pure form and
in association with a Cobalt(II) salt30. With the exception of the trials involving DBU and DMAP as
catalysts (entries 7 and 8), the outcome of the CO2 fixation experiments using KI, pure TBAB, or

Page 15 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
the TBABꢀCoCl2 system were roughly the same (entries 2, 9, and 10). However, a slight increase in
the catalytic activity was observed in presence of CoCl2.
Given the prominent role of halideꢀ and oniumꢀbased catalysts in epoxidized lignin carbonation, the
catalytic capabilities of two imidazoliumꢀbased ionic liquids, i.e., 1ꢀallylꢀ3ꢀmethylimidazolium
chloride (amimCl) and 1ꢀallylꢀ3ꢀmethylimidazolium bromide (amimBr) in DMF medium was
investigated. The ILs were tested in their pure form and in the presence of CoCl2. Both the ILs were
found to be active catalysts but amimBr demonstrated superior performances over amimCl (entries
11 and 13). Furthermore, the addition of CoCl2 as coꢀcatalyst remarkably improved the yield in
cyclic carbonates in both cases (entries 12 and 14). The high catalytic activity of CoCl2 appears to
be linked to the high Lewis acidity of the metal ion. It was demonstrated that when CoCl2 and an
ammonium salt such as TBACl or an imidazoliumꢀbased IL were mixed together, a new ion pair,
[onium]+[CoCl3]−, was formed31,32. These ion pairs are believed to be the catalytically active species
in the synthesis of cyclic carbonates. In the particular case of ILs, the coordination of both cationic
and anionic moieties is believed to promote the reaction by dual activation of the epoxide and
carbon dioxide33. Another crucial factor is the good leaving ability of the IL counteranion that is
essential for ring closure and can also explain the lower activity of amimCl compared to amimBr.
The ability of imidazoliumꢀbased ILs to act as powerful catalyst in the fixation of CO2 to epoxides
is already well documented on simple molecules such as propylene oxide14,34 but it has never been
reported on epoxidized lignin.
Finally, the weight loading (w/v%) of epoxidized lignin in DMF in the presence of amimBr as
catalyst was forced to its limit. The solubility was assessed by eye inspection. The conversion to
cyclic carbonates was in line with previous experiments involving the same IL (entry 15).

Green Chemistry
Page 16 of 28
View Article Online
DOI: 10.1039/C6GC01028H
Weight loading
(w/v%)
IR Abs
(a.u.)
Entry
Solvent
Catalyst (w/w%)
1
2
DMF
DMF
None
KI (10%)
1
0.09
0.58
1
3
DMSO
Meacac
Dioxane
MEK
DMF
KI (10%)
1
0.48
4
KI (10%)
1
0.34
5
KI (10%)
1
Not soluble
Not soluble
0.44
6
KI (10%)
1
7
DBU (10%)
1
8
DMF
DMAP (10)
1
0.38
9
DMF
TBAB (10%)
TBABꢀCoCl2* (10%)
amimCl (10%)
amimClꢀCoCl2* (10%)
amimBr (10%)
amimBrꢀCoCl2* (10%)
amimBr (10%)
1
0.53
10
11
12
13
14
15
DMF
1
0.62
DMF
1
0.49
DMF
1
1
0.63
DMF
0.59
DMF
1
0.66
DMF
6.7%
0.62
*TBABꢀCoCl2, amimClꢀCoCl2, and amimBrꢀCoCl2 1:1 mol/mol
Table 2 List of carbonation experiments performed on HBL lignin at 20 bar for 20 hours at a fixed
temperature of 80°C, reporting: adopted solvent, % weight loading, catalyst type and loading,
pressure, reaction period, and absorbance of the cyclic carbonate peak at 1790 cmꢀ1.
The natural continuation of the previously described results is the exploitation of the sole IL as both
reaction medium and catalyst. Table 3 shows the effect of different ILs and experimental
conditions. Typical conditions applied for carbonation reactions in IL medium are, if not otherwise

Page 17 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
stated: 20 bar pressure, 20 hours reaction time, 80°C as reaction temperature, and 5% w/w
consistency.
amimCl was the first IL chosen for this investigation due to its widely spread use in the dissolution
of lignocellulosic materials. The obtained outcome (entry 16) was in line with trials performed in
DMF medium in the presence of amimCl, proving its catalytic activity even under different
experimental conditions. As previously discussed, the IL counteranion nature plays a crucial role in
promoting CO2 fixation with epoxides. Accordingly, amimBr was selected as an IL bearing a better
leaving group than amimCl for further experiments. According to IR analysis, the attained result
was the best among all the trials presented in term of cyclic carbonates yield (entry 20). It is
highlighted that, when ILs are used as proper solvent, the addition of CoCl2 to the reaction mixture
did not result in an increased catalytic activity, as was instead observed in DMF medium (entries 19
and 21).
Additional experiments involving the use of recycled amimCl were also performed to evaluate the
catalytic and solvating efficiency of the IL after purification. At the third run, neither a minor
conversion nor a decrease in selectivity of cyclic carbonates formation was observed according to
IR and 13C NMR analysis (entries 16ꢀ18).
Furthermore, others commercially available imidazoliumꢀbased ILs bearing chloride counteranions
were tested for the cycloaddition reaction: 1ꢀbutylꢀ3ꢀmethylimidazolium chloride (bmimCl) and 1ꢀ
benzylꢀ3ꢀmethylꢀimidazolium chloride (BzmimCl). In agreement with the scarce influence of the
cation nature reported in literature25, their efficiency in promoting the formation of cyclic
carbonates was the same to that observed for amimCl (entries 16,22,and 23).
Again, the weight loading of epoxidized lignin was increased to evaluate the catalytic activity of the
IL in extreme reaction conditions. The solubility was assessed by eye inspection. Going from 5 to
20% lignin loading, epoxides conversion to cyclic carbonates was only slightly affected (entries 20,
24, and 25). It is worth noticing that the corresponding limit in DMF medium was 6.7%. Even if the
recovery process is similarly smooth (water precipitation followed by filtration), the use of ILs

Green Chemistry
Page 18 of 28
View Article Online
DOI: 10.1039/C6GC01028H
allows for the use of minimal amount of solvent and, as direct consequence, antisolvent if compared
to DMF. Moreover, as organic salts with negligible vapor pressure, imidazoliumꢀbased ILs are far
more easily regenerated by highꢀvacuum water distillation.
Solvent &
Catalyst
Additional
Catalyst (w/w%)
Consistency
(w/v%)
IR Abs
(a.u.)
Entry
16
17
18
19
20
21
22
23
24
25
amimCl
amimCl cycle 2
amimCl cycle 3
amimCl
ꢀ
5
5
0.57
0.51
0.55
0.60
0.68
0.58
0.57
0.56
0.62
0.61
ꢀ
ꢀ
5
CoCl2 (10%)
5
amimBr
ꢀ
5
amimBr
CoCl2 (10%)
5
bmimCl
ꢀ
ꢀ
ꢀ
ꢀ
5
BzmimCl
amimBr
5
10
20
amimBr
Table 3 List of carbonation experiments performed on HBL lignin at 20 bar for 20 hours at a fixed
temperature of 80°C, reporting: adopted solvent, % weight loading, type and % of catalyst,
pressure, reaction period, and absorbance of the cyclic carbonate peak at 1790 cmꢀ1.
Besides, 13C NMR spectra acquired on epoxidized and carbonated specimens of herbaceous and
hardwood kraft lignin were thoroughly interpreted. More interestingly, quantitative 13C NMR
spectra were acquired on carbonated HBL and HWKL in the presence of dimethyl carbonate as
internal standard to provide a reliable quantification of the cyclic carbonate moiety. Carbonated
specimens were prepared according to the best experimental conditions found, i.e., pure amimBr as

Page 19 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
reaction solvent. Epoxidized and carbonated lignin was not subjected to acetylation (traditionally
performed in a pyridine/acetic anhydride solution) in order to avoid degradation of glycidyl and
cyclic carbonate moieties.
Accurate assignment of epoxides and cyclic carbonates Carbon nuclei, as a function of the type of
labile hydroxyl (phenolic or from carboxylic acid) and degree of orthoꢀsubstitution of the phenolic
moiety (not substituted, monoꢀ, and diꢀsubstituted), was based on spectroscopic data obtained by
13C NMR analyses of epoxidized and carbonated lignin model compounds (Table 4). As already
reported for allylated lignin35, the C1 position of the epoxide chain is the most affected by the
surrounding chemical environment (ꢂC1 ≈ 8.7 ppm, ꢂC2 ≈ 0.9 ppm, ꢂC3 ≈ 0.4 ppm). Surprisingly,
this is no longer true for carbonated derivatives, in which the most perturbed site is the C3 position
and the least influenced the C1 (ꢂC1 ≈ 0.2 ppm, ꢂC2 ≈ 1.1 ppm, ꢂC3 ≈ 7.6, ꢂC4 ≈ 0.6 ppm).
Assignment
(ppm)
2,6-dimethoxy Vanillic acid Vanillic acid
Phenol
Guaiacol
phenol
(phenol)
(acid)
O
RO
C2
68.3
50.2
44.7
66.3
74.3
67.0
154.8
70.4
50.3
45.0
66.3
74.9
69.4
155.2
74.0
69.8
65.3
C1
C3
O
RO
C2
50.4
44.6
66.3
75.1
71.3
155.1
49.8
44.7
66.1
74.5
68.7
155.0
49.5
44.6
66.2
74.0
63.7
154.6
C1
C3
O
RO
C2
C1
C3
O
C3
C2
C4
RO
RO
RO
RO
O
O
O
O
C1
C1
C1
C1
O
O
C3
C2
C4
C4
C4
O
O
O
C3
C2
O
O
C3
C2
Table 4 13C NMR assignment for the epoxide and cyclic carbonate side chain in representative
model compounds of lignin after labile hydroxyls functionalization.

Green Chemistry
Page 20 of 28
View Article Online
DOI: 10.1039/C6GC01028H
Figure 3 reports the 13C NMR spectra acquired for epoxidized and carbonated herbaceous (top) and
hardwood kraft (bottom) lignin. In the NMR spectrum of epoxidized HBL, partially overlapped
signals located at 43.5 and 50.4 ppm were assigned to the C3 and C2 positions of condensed and
syringyl derivatives, while peaks at 43.9 and 49.9 ppm were attributed to the C3 and C2 positions of
guaiacyl and pꢀhydroxyphenyl derivatives, respectively. In the case of epoxidized HWKL, partially
overlapped signals at 43.4 and 50.4 ppm were assigned to the C3 and C2 positions of condensed and
syringyl derivatives, while peaks located at 43.9 and 49.9 ppm were attributed to the C3 and C2
positions of guaiacyl derivatives, respectively. In agreement with labile hydroxyls distributions in
the two lignin under investigation (Table 1), the relative intensity of signals pertaining to condensed
and syringyls (C&S) compared to guaiacyls and pꢀhydroxyphenyls (G&H) is similar for the
herbaceous lignin (C&S ≈ G&H) whereas favorable to disubstituted phenols for the hardwood kraft
lignin (C&S › G&H).
The accurate assignment of the C1 Carbon of the epoxide chain was more challenging. A number of
different and weak signals were indeed detected, as a consequence of the strong influence of the
surrounding chemical environment to which this position is subjected. Anyway, it is for certain that
downfielded peaks have to be attributed to condensed and syringyl derivatives, whereas upfielded
peaks have to be related to guaiacyl and pꢀhydrophenyl derivatives (in agreement with the study
performed on model compounds). The broad signal in the aromatic region (not entirely reported) is
characterized by a single wellꢀdefined peak related to quaternary aromatic Carbon linked to
methoxyl groups.
After CO2 fixation, no signal pertaining to the C2 and C3 position of the epoxide chain (around 50
and 44 ppm, respectively) was detected for both the lignins examined, proving a total conversion of
the glycidyl group. On the other hand, the appearance of wellꢀdefined peaks at 65.9 and 155.1 ppm
were unambiguously attributed to the C1 and C4 position of the cyclocarbonate structure,
respectively. C2 and C3 Carbon signals were numerous and scattered along a broad ppm range.

Page 21 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
Approximately, their position lay between 73ꢀ76 and 68ꢀ72 ppm, respectively, being in both cases
the more downfielded peaks attributed to condensed and syringyl derivatives, and the more
upfielded peaks related to guaiacyl and pꢀhydrophenyl derivatives (in agreement with the study
performed on model compounds).
13C NMR spectra of carbonated lignins were also recorded in the presence of dimethyl carbonate
(25 µl) as internal standard to provide a reliable quantification of the cyclic carbonate moiety. The
carbonyl group Carbons of the internal standard (155.9 ppm) and carbonyls groups of cyclic
carbonates on lignin (155.1 ppm) were used as reference signals for the quantification. The strong
signal detected at 54.8 ppm is related to methyl groups in dimethyl carbonate.
Table 5 reports the experimental (first line) and theoretical (third line) quantification of the cyclic
carbonates groups for HBL and HWKL, expressed as mmolgꢀ1. The theoretical data were calculated
starting from the quantification of epoxides groups, as obtained by 31P NMR analysis (Table 1). The
amount of hydroxypropylic alcohols ensuing from phosphorus derivatization (penultimate line,
Table 5) ought to be conveniently corrected according to lignin molecular weight variation after
CO2 fixation. In this case, the C9 lignin unit, already described in the section concerning the
epoxidation reaction, is illustrated by the empirical formula C14H17O7 and its molecular weight is
ca. 297 gmolꢀ1. Corrected values for hydroxypropylic alcohols are reported in the last line of Table
5 (1.91 and 2.10 mmolgꢀ1 for HBL and HWKL, respectively).
According to 13C NMR analysis, the cyclic carbonates moiety quantification is 2.01 and 1.78
mmolgꢀ1 for the herbaceous and hardwood kraft lignin, respectively, in very good agreement with
calculated theoretical values.
The occurrence of degradation phenomena leading to cyclic carbonates ring opening was excluded
by both IR and NMR analyses. IR spectra of carbonated lignin specimens did not evidenced the
presence of peaks around 1750 cmꢀ1, which is the typical absorption region of linear carbonates. On
the other hand, 13C NMR spectra acquired for qualitative purposes, i.e., without dimethyl carbonate,

Green Chemistry
Page 22 of 28
View Article Online
DOI: 10.1039/C6GC01028H
never did displayed absorption around 156 ppm, which is the characteristic signal of linear
carbonates.

Page 23 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H

Green Chemistry
Page 24 of 28
View Article Online
DOI: 10.1039/C6GC01028H
Figure 3 13C NMR spectra depicting epichlorohydrin epoxidized and carbonated herbaceous and
hardwood kraft lignins. Spectra normalized on methoxyls signal (56 ppm). Epoxidized and
carbonated lignins were not subjected to acetylation to avoid degradation of glycidyl and cyclic
carbonate moieties.
Quantification (mmol/g)
HBL
HWKL
Experimental
Cyclic carbonates (13C NMR)
2.01
1.78
Theoretical
Hydroxypropyl –OH (31P NMR)
Hydroxypropyl –OH (mass corrected)
2.25
1.91
2.47
2.10
Table 5 Theoretical and experimental quantification of cyclic carbonates groups in herbaceous and
hardwood kraft lignin, calculated according to 31P and 13C NMR analysis, respectively. Mass
correction considers the effect of molecular weight variations after the carbonation reaction on
lignin hydroxyls distribution.
Conclusion
Lignins from different sources were directly reacted with epichlorohydrin, providing a pure
epoxidized derivative. A quantitative evaluation of the epoxide groups inserted was attained
through 31P NMR analysis. The addition reaction of CO2 to the epoxide groups was for the first
time reported on lignin. Ionic liquids were successfully used as both solvent and catalyst for the
reaction. With the aid of conveniently functionalized model compounds, epoxidized and carbonated
lignin was fully characterized by 13C NMR analysis. A precise determination of the cyclic carbonate
structures formed was also given by quantitative 13C NMR.

Page 25 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
Acknowledgement
The authors would like to thank the Italian Minister of Agriculture for financial support (ProLignin,
EraNets WoodWisdom/Bioenergy, 2012ꢀ14 project)
References
1.
Proceedings of the eighth cellulose conference I. Wood chemicals ꢀ a future challenge.
Journal of Applied Polymer Science. Applied Polymer Symposia n. 28, Timell T.E. (Wiley &
Sons), USA, 1975.
2.
3.
S. Fernando, S. Adhikari, C. Chandrapal, N. Murali, Energy Fuels, 2006, 20(4), 1727ꢀ1737.
K.Chaochanchaikul, K. Jayaraman, V. Rosarpitak, N. Sombatsompop, Bioresources, 2012,
7(1), 38ꢀ55.
4.
5.
P. Frigerio, L. Zoia, M. Orlandi, T. Hanel, L. Castellani, Bioresources, 2014, 9(1), 1387ꢀ
1400.
A. Gandini, M.N. Belgacem, in Monomers, polymers and composites from renewable
materials, A. Gandini, M.N. Belgacem (Elsevier), Oxford, UK, 2008, p. 243.
T. Koike, Polym. Eng. Sci., 2012, 52(4), 701ꢀ717.
6.
7.
C.A. Cateto, M.F. Barreiro, A.E. Rodrigues, M.N. Belgacem, Ind. Eng. Chem. Res., 2009,
48(5), 2583ꢀ2589.
8.
9.
L. Hu, H. Pan, Y. Zhou, M. Zhang, Bioresources, 2011, 6(3), 3515ꢀ3525.
Y.L. Chung, J.V. Olsson, R.J. Li, C.W. Frank, R.M. Waymouth, S.L. Billington, E.S. Sattely,
ACS Sustainable Chem. Eng., 2013, 1, 1231ꢀ1238.
10. C. Sasaki, M. Wanaka, H. Takagi, S. Tamura, C. Asada, Y. Nakamura, Industrial Crops and
Products, 2013, 43, 757ꢀ761.
11. P. G. Parzuchowski, M. JurczykꢀKowalska, J. Ryszkowska, G. Rokicki, J. Appl. Polym. Sci.,
2006, 102, 2904–2914.

Green Chemistry
Page 26 of 28
View Article Online
DOI: 10.1039/C6GC01028H
12. B. Nohra, L. Candy, J.F. Blanco, C. Guerin, Y. Raoul, Z. Mouloungui, Macromolecules,
2013, 46(10), 3771ꢀ3792.
13. S. Zhang, Y. Chen, F. Li, X. Lu, W. Dai, R. Mori, Catal. Today, 2006, 115, 61ꢀ69 .
14. J. Sun, S.I. Fujita M. Arai, J. Organomet. Chem., 2005, 690, 3490ꢀ3497.
15. Z. Lei, C. Dai, B. Chen, Chem. Rev., 2014, 114, 1289ꢀ1326.
16. J.L. Anthony, J.L. Anderson, J.M. Edward, J.F. Brennecke, J. Phys. Chem. B, 2005, 109,
6366ꢀ6374.
17. Y. Pu,N. Jiang, A.J. Ragauskas, J. Wood Chem. Technol., 2007, 27(1), 23ꢀ33.
18. I. Kilpeläinen, H. Xie, A. King, M. Granstrom, S. Heikkinen, D.S. Argyropoulos, J. Agric.
Food Chem., 2007, 55(22), 9142ꢀ9148.
19. Spectral Database of Organic Compounds, National Institute of Advanced Industrial Science
and Technology, Japan. Available over Internet at http://sdbs.db.aist.go.jp
20. M. Maciejewski, K. Poltorak, J. Kaminska, Polish Journal of Chemistry, 2009, 83(4), 595ꢀ
604.
21. H. Khoshro, H. R. Zare, M. Namazian, A. A. Jafari, A. Gorji, Electrochimica Acta, 2013,
113, 263ꢀ268.
22. A. A. Bredikhin, Z. A. Bredikhina, D. V. Zakharychev, A. V. Pashagin, Tetrahedron:
Asymmetry, 2007, 18(10), 1239ꢀ1244.
23. D. S. Argyropoulos, J. Wood Chem. Technol., 1994, 14(1), 45ꢀ63.
24. A. Granata, D.S. Argyropoulos, J. Agric. Food Chem., 1995, 43, 1538ꢀ1544.
25. S.A. Ralph, J. Ralph, L.L. Landucci, NMR database of lignin and cell wall model compounds,
2004. Available over Internet at http://www.dfrc.ars.usda.gov/software.html
26. NIIR Board of Consultants and Engineers, Synthetic Resins Technology Handbook, ASIA
PACIFIC BUSINESS PRESS Inc., 2005.
27. B. Ahvazi, O. Wojciechowicz, T.M. TonꢀThat, J. Hawari, J. Agric. Food Chem.,
2011, 59(19), 10505ꢀ10516.

Page 27 of 28
Green Chemistry
View Article Online
DOI: 10.1039/C6GC01028H
28. M. North, R. Pasquale, C. Young, Green Chem., 2010, 12, 1514ꢀ1539.
29. J. Sun, W. Cheng, Z. Yang, J. Wang, T. Xu, J. Xin, S. Zhang
3078.
, Green Chem., 2014, 16, 3071ꢀ
30. A. Sibaouih, P. Ryan, M. Leskela, B. Rieger, T. Repo, Appl. Catal. A, 2009, 365, 194ꢀ198.
31. F. Li, L. Xiao, C. Xia, B. Hu, Tetrahedron Lett., 2004, 45, 8307ꢀ8310.
32. L.F. Xiao, F.W. Li, J.J. Peng, C.G. Xia, J. Mol. Catal. A: Chem., 2006, 253, 265ꢀ269.
33. A.L. Girard, N. Simon, M. Zanatta, S. Marmitt, P. Gonçalves, J. Dupont, Green Chem., 2014,
16, 2815ꢀ2825
34. J. Peng, Y. Deng, New J. Chem., 2001, 25, 639ꢀ641.
35. L. Zoia, A. Salanti, P. Frigerio, M. Orlandi, Bioresources, 2014, 9(4), 6540ꢀ6561.

Green Chemistry
Page 28 of 28
View Article Online
DOI: 10.1039/C6GC01028H
198x78mm (96 x 96 DPI)