Effect of the N-based ligands in copper complexes for depolymerisation of lignin

Article abstract of DOI:10.1039/c5nj03152d

Several organic soluble N-based ligands and their copper complexes were firstly investigated as catalysts to depolymerise organosolv lignin in the organic solvent, dimethylformamide (DMF) and an ionic liquid (1-ethyl-3-methylimidazolium xylenesulfonate, [

Full text of DOI:10.1039/c5nj03152d

NJC
View Article Online
View Journal
PAPER
Effect of the N-based ligands in copper
complexes for depolymerisation of lignin†
Cite this:DOI: 10.1039/c5nj03152d
a
a
b
b
Jinhuo Dai, Sepa Nanayakkara, Thomas C. Lamb, Andrew J. Clark,
a
a
a
a
Si-Xuan Guo, Jie Zhang, Antonio F. Patti and Kei Saito*
Several organic soluble N-based ligands and their copper complexes were firstly investigated as catalysts
to depolymerise organosolv lignin in the organic solvent, dimethylformamide (DMF) and an ionic liquid
(1-ethyl-3-methylimidazolium xylenesulfonate, [emim][ABS]). The results of screening depolymerisation reactions
in DMF and [emim][ABS] showed that all the copper–amine complexes catalysed lignin depolymerisation more
efficiently in ionic liquids than in DMF. Among the seven types of ligands, copper complexes with two types of
ligands (E)-N-(pyridin-2-ylmethylene)aniline and (E)-4-methoxy-N-(pyridin-2-ylmethylene)aniline depolymerised
the lignin more efficiently than the others. These two copper complexes with the N-based ligand were
further studied to determine the most efficient conditions for the depolymerisation of the lignin. The
most effective depolymerisation by conditions involved treatment at 180 1C for 12 h in [emim][ABS].
Cyclic voltammetric studies were carried out to investigate the reversible potential associated with the
copper centers of their complexes with these N-based ligands. The results suggest that two types of
ligands have more positive reversible potentials than those of other copper complexes.
Received (in Victoria, Australia)
9th November 2015,
Accepted 9th February 2016
DOI: 10.1039/c5nj03152d
www.rsc.org/njc
of softwood lignin to produce monomeric products (dihydro-
Introduction
coniferyl alcohol), however, this method required a pressure of
8
Lignin exists in the second cell wall of plants, and is well known 3.4 MPa with a temperature of 468 K. Koyama reported the
2 3
as one of the most abundant renewable materials from the hydrocracking of lignin model compounds by using Fe O –S,
1
–3
9
paper and pulp industry.
Lignin is also considered as a Fe
O
2 3
2 2 3 2 3
/Al O –S, and NiO–MoO –Al O between 613 and 723 K.
material with high commercial potential for producing fine
Some metal complexes with N-based ligands have been used
7
chemicals. This biopolymer contains three types of primary for lignin oxidation and depolymerisation under mild conditions.
phenylpropane units called monolignols and these creates Metalloporphyrin complexes, such as trisodium tetra-4-sulfo-
fundamental units that combine in different ways to give various natophthalocyanine iron(III), have been investigated to oxidize
4–6
10,11
forms of the complex natural polymer known as lignin. During lignin.
Zucca et al. also utilized immobilized Fe(III)–5,10,15,20-
the past few decades, lignin depolymerisation has become a tetrakis(pentafluorophenyl)porphyrin on a pyridyl-functionalized
research hotspot with many considerable methods reported, for poly(vinyl alcohol), to oxidize lignin model compounds at room
12
example, depolymerisation using oxidation, reduction, biological temperature. The drawbacks of these metalloporphyrins are that
7
methods and electrochemical methods. The main challenge of this these catalyst complexes normally have a complicated structure
7
research is to selectively cleave different bonds in lignin in order and lack recyclability and stability. A metallosalen catalyst has
7
to depolymerise lignin into fine chemicals with high value.
also been used as a novel lignin oxidation and depolymerisation
13
The depolymerisation of lignin with several metallic catalysts catalyst. Drago et al. reported that Co(salen) complexes oxidised
has also been reported. However, there are drawbacks in most lignin model compounds rapidly with the presence of molecular
1
4
existing depolymerisation methods using metallic catalysts such oxygen to produce vanillin.
as the need for high pressure and high temperature. Pepper
6
Copper complexes with N-based ligands are well known
et al. investigated the catalytic ability of a number of catalysts, oxidation catalysts for 2,6-dimethylphenol (DMP) to form an
s
such as Raney Ni, Pd/C, Rh/C and Ru/C, in the hydrogenation engineering thermoplastic poly(2,6-dimethyl-1,4-phenylene
oxide) (PPO) and also known as catalysts to depolymerise
PPO. For example, PPO was first developed by using a copper–
a
School of Chemistry, Monash University, Clayton, VIC 3800, Australia.
pyridine complex and its derivatives by the oxidative polymerisation
E-mail: Kei.Saito@monash.edu; Fax: +61-3-9905-8501; Tel: +61-2-9905-4600
Department of Chemistry, University of Warwick, Gibbbet Hill, Coventry,
CV4 7AL, UK
1
5,16
b
of DMP.
Some enzyme mimic complexes with N-based
ligands are also applied for PPO polymerisation. Higashimura
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/c5nj03152d et al. studied the oxidative polymerisation of DMP catalysed by
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
1,4,7-triisopropyl-1,4,7-triazacyclononane)copper(II) to produce
NJC
(
1
7
PPO. We have studied the depolymerisation of PPO involving
the redistribution mechanism to produce oligomeric PPO using
1
8
a copper–pyridine complex. This depolymerisation produced
2
4
M = 4.9 ꢀ 10 oligomeric PPO from PPO M = 1.0 ꢀ 10 .
n
n
We also reported the PPO depolymerisation mechanism as
1
8
follows. Under the oxidative conditions catalysed by copper
complexes, two kinds of radicals (monomeric phenoxyl radical
and polymeric phenoxyl radical) were generated. The redistribution
was induced via a quinone ketal intermediate after the phenoxyl
radical of PPO is attacked by the monomeric phenoxyl radical.
Further study also showed that PPO can be depolymerised by
copper–EDTA involving the redistribution mechanism in the
1
9
ionic liquid ([emim][ABS]).
Our group has applied this depolymerisation and demon-
strated that lignin can be successfully depolymerised using
copper–EDTA as a catalyst in both water and in ionic liquids
(1-ethyl-3-methylimidazolium xylenesulfonate [emim][ABS] and
1
-butyl-3-methylimidazolium methyl-sulfate [bmim][MeSO ]),
4
involving the redistribution mechanism under mild conditions
2
0
0 00 00
Scheme 1 N-Based ligands: L1: N,N,N ,N ,N -pentamethyldiethylene-
(
80 1C, atmospheric pressure).
Ionic liquids, because of their non-flammability, recyclability
and non-volatility, have been widely used as solvents for the
0
0
triamine (PMDETA); L2: N,N,N ,N -tetramethylenediamine (TMEDA); L3:
tris(2-pyridylmethyl)amine (TPA); L4: (E)-N-tert-butyl-1-(pyridin-2-yl)-
methanimine (TBPMA); L5: (E)-N-sec-butyl-1-(pyridin-2-yl)methanimine
21,22
organic reactions.
Binder et al. demonstrated that ionic liquids
(SBPMA); L6: (E)-N-(pyridin-2-ylmethylene)aniline (PMEA); L7: (E)-4-methoxy-
23
are excellent solvents for processing woody biomass and lignin.
N-(pyridin-2-ylmethylene)aniline (MPMEA).
St ¨a rk et al. presented a method for oxidative depolymerisation
of lignin in ionic liquid (1-ethyl-3-methylinidazolium trifluoro-
24
methanesulfonate [emim][CF SO ]) using Mn(NO ) as a catalyst.
3
3
3 2
N-based ligand as catalysts for lignin depolymerisation in
organic solvents and an ionic liquid.
In their study, this catalyst system was shown to be an efficient
reaction system as the conversion of lignin reached 66.3% after
reacting for 24 h at a temperature of 100 1C, however a pressure of
5
8
4 ꢀ 10 Pa air was required. Cox et al. reported using an acidic
Results and discussion
ionic liquid, 1-hexyl-3-methylimidazolium chloride (HMIMCl), as
25
both the solvent and the catalyst for lignin depolymerisation. In In this study, organosolv lignin was selected as a lignin for
this method, lignin, extracted from oak wood, was depolymerised depolymerisation research. This organosolv lignin was isolated by
in HMIMCl and proceeds by a hydrolysis reaction with alkyl–aryl removing the low molecular weight fraction by stirring in methanol
ether bond cleavage.
for 2 h to dissolve the low molecular weight fraction as we previously
2
0
In this study, we have investigated several types of copper reported. The methanol insoluble fraction, which was high
complexes with an N-based ligand for lignin depolymerisation molecular weight lignin, was filtered and dried. This part of
involving the redistribution mechanism in two solvents DMF lignin was defined as M-lignin used for all the depolymerisation
and an ionic liquid ([emim][ABS]). The N-based ligands shown reactions. The molecular weight and polydispersity of M-lignin
n w n
below (Scheme 1) have been used to form copper complexes were measured by GPC to give M = 12 500 and Ð (M /M ) = 1.86.
with copper halides and used as catalysts.
0
00 00
Screening depolymerisation of lignin in DMF and [emim][ABS]
Some of these ligands, such as N,N,N ,N ,N -pentamethyl-
0
0
diethylenetriamine (L1), N,N,N ,N -tetramethylenediamine (L2), Initial attempts to depolymerise lignin were conducted in separate
and tris(2-pyridylmethyl)amine (L3), are commercially available experiments, using [emim][ABS] and DMF with the seven N-based
and others are easy to synthesize and relatively stable. All these ligands (Scheme 2). DMF was chosen to enable solubilisation
ligands in this study have been proved to coordinate with of the N-based ligands, monomer (TBDMP) and M-lignin. It
2
6–33
copper to form copper complexes.
was thought that the solubility enhanced the depolymerisation
Initial screening studies with different copper complexes efficiency by making the solution of lignin, monomer and
with N-based ligands in DMF and IL were carried out to select catalysts homogeneous. Screening reactions were carried out
the most efficient N-based ligands for lignin depolymerisation. in DMF and [emim][ABS] at the temperatures of 120 1C and
On the basis of these results, ligands (E)-N-(pyridin-2-yl- 180 1C, respectively. All the screening reactions were carried out
methylene)aniline (L6) and (E)-4-methoxy-N-(pyridin-2-ylmethylene)- in these two solvents using 4-tert-butyl-2,6-dimethyl-phenol
aniline (L7) were selected for further study. The novelty of this (TBDMP) as an additive and Cu(I)/N-based ligands as catalysts
research lies in using several types of copper complexes with the under oxygen flow. These reactions were performed for 6 hours
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
Scheme 2 Schematic representation of the proposed lignin depolymerisation mechanism.
based on the result of previous research for lignin depolymerisation
2
0
in water with copper–EDTA. TBDMP was selected as a phenol
additive under the conditions as it could not polymerise due to the
tert-butyl group on the para-position in the redistribution process.
Table 1 summarizes the molecular weight and polydispersity
changes of M-lignin under the 6 h depolymerisation reaction
conditions using copper complexes with different N-based ligands.
Samples of every reaction were collected and the molecular
weight of each sample was measured by GPC (Fig. 1 and 2). As for
the initial samples, the high molecular weight peak, of which the
retention time is around 14 min, is M-lignin and the low molecular
weight peak with the retention time of 16.8 min is TBDMP.
After 6 h, the depolymerisation of M-lignin was observed as
the molecular weight of all samples decreased (Table 1). The
M
(M
n
n
value became smaller than the initial M
= 12 500), and the polydispersity became larger than Ð of
the initial M-lignin (M /M = 1.86). This indicated that all these
n
of the M-lignin
w
n
copper complexes with N-based ligands could catalyse lignin
depolymerisation with TBDMP involving the redistribution
mechanism. Moreover, the GPC data (Table 1) showed that the
n
M value of all the depolymerised lignin with different copper
complexes is smaller for reactions conducted in [emim][ABS],
compared with those conducted in DMF. This demonstrated that
lignin could be depolymerised more efficiently in [emim][ABS]
with all these different copper–amine complexes compared with
DMF. The reason could be that the ligands were more readily
soluble in [emim][ABS] other than in DMF.
n
Table 1 The M and Ð values of depolymerised samples from screening
reaction
In DMF
(g mol
In [emim][ABS]
ꢁ1
ꢁ1
Ligands
M
n
)
Ð (M /M
w n
)
M
n
(g mol
)
w n
Ð (M /M )
Fig. 1 GPC chromatograms of depolymerisation of M-lignin in DMF with
L1
L2
L3
L4
L5
L6
L7
11 500
11 300
7000
7100
8400
5600
5200
3.6
3.4
3.0
5.9
3.4
3.6
2.6
6000
5000
5000
5700
6200
2500
2500
2.5
2.7
3.2
2.4
2.5
3.2
2.4
copper complexes with different N-based ligands.
When comparing all the depolymerisation reactions, the M
and Ð (M /M ) of depolymerised lignin were different to each
other. This demonstrated that copper complexes with different
n
w
n
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
18 h, 24 h, 36 h, and 48 h. Samples from the different reaction
n w n
times were subjected to GPC to give M and Ð (M /M ).
The GPC chromatograms of depolymerised lignin are shown
below (Fig. 3).
Within 3 h of reaction time, the peaks of M-lignin and TBDMP
showed a molecular weight decrease from about 12 000 to about
1
0 000 and the Ð increased from 1.8 to 1.9. After 6 h, the average
molecular weight of M-lignin was reduced to M = 3900, with a
higher Ð (M /M = 2.09).
n
w
n
The depolymerisation reaction was continued for 12 h, and
the average molecular weight decreased to M = 2000 with a Ð
n
value of M /M = 2.63. The depolymerisation was continued for
w
n
4
1
8 h but the molecular weight remained at about 2000 after
2 h, which indicated that the best reaction time for depoly-
merisation with L6 was up to 12 h (Fig. 4).
This result showed that it was difficult to completely depoly-
merise lignin even using TBDMP as an additive, one reason
being that the redistribution mechanism involved is an equilibrium
3
4–36
reaction
and the low molecular weight part of depoly-
merised lignin can be repolymerised simultaneously with the
depolymerisation.
Further depolymerisation of lignin using the copper–L7 complex
with TBDMP was carried out. All the samples of depolymerised
lignin obtained from different reaction times were analysed by GPC
to determine the M
the average molecular weight of lignin decreased from M
2 500 to Mn = 4500 with Ð increasing from M /M = 1.86 to
n
and Ð (Fig. 5). Within 3 h of reaction time,
n
=
1
w
n
M /M = 3.08. This indicated that depolymerisation of the
w
n
mixture with the copper–L7 complex resulted in a significant
molecular weight decrease within 3 h of reaction time. After 6 h,
n
the average molecular weight was reduced to M = 2700 with Ð
(M
w
/M ) = 2.33. The depolymerisation reaction was carried out
n
for 48 h but there was no significant decrease in molecular
weight after 9 h, with the molecular weight remaining around
M = 2000 (Fig. 6). This reaction showed a similar result to the
n
Fig. 2 GPC chromatograms of depolymerisation of M-lignin in [emim][ABS]
with copper complexes with different N-based ligands.
N-based ligands showed different catalytic abilities for depoly-
merisation. Among all these ligands, the molecular weight of
depolymerised lignin from L6 and L7 was lower than that of
other ligands in these screening reactions and L6 and L7 were
considered to be more efficient than other ligands.
From the results from the above screening reactions, these
two N-based ligands were selected and several depolymerisation
reactions were carried out in [emim][ABS] with L6 and L7 with
different reaction times in order to find the optimum lignin
depolymerisation conditions using these two N-based ligands
and also to investigate the difference between these two ligands
as described below.
Depolymerisation of lignin in [emim][ABS] with L6 and L7
Two extended reactions were carried out with different reaction
times in [emim][ABS] with the copper complexes, L6 and L7.
These two reactions were undertaken over 48 h with samples
Fig. 3 GPC chromatograms of depolymerised M-lignin in [emim][ABS]
collected at the following reaction times: 3 h, 6 h, 9 h, 12 h, using copper–L6 with different reaction times.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
Fig. 6
M
n
of samples with different reaction times of lignin depolymer-
Fig. 4
n
M of samples with different reaction times of lignin depolymerisation
isation with copper–L7.
with copper–L6.
for this result is that the reversible potential of the copper–L7
complex could be more positive than that of the copper–L6
complex. The reversible potentials of different copper complexes are
known to change due to the structure of the coordinated ligand and
different N-based ligands in copper complexes can affect the radical
36–38
generation rate that can affect the lignin depolymerisation rate.
The yields of depolymerised products using copper–L6 and
1
copper–L7 were both ca. 90%. The H NMR spectrum (ESI,†
Fig. S3) of separated depolymerized lignin was analysed and
showed peaks in the range of 6.2 to 9.0 ppm (d values) resulting
from the aromatic protons in lignin units and peaks at 1.3 ppm
from the tert-butyl group resulting from TBDMP. However,
1
there were also several unidentified peaks in the H NMR spectrum
because of the nature of the lignin structure and we were not able to
identify or isolate individual depolymerized products.
Recovery and reusability of [emim][ABS]
Fig. 5 GPC chromatograms of depolymerisation of M-lignin in [emim][ABS]
using copper–L7 with different reaction times.
The recovery and reusability of [emim][ABS] have been studied.
First, the method reported by Tan et al. was followed to recover
3
9
1
the ionic liquid. Next, the ionic liquid was analysed by H NMR
to confirm its original structure remaining after the reaction
depolymerisation using the copper–L6 complex. The M-lignin
was not completely depolymerised by the copper–L7 complex
as the low molecular weight oligomers can repolymerise con-
currently with depolymerisation. Based on the results of these
depolymerisation reactions, the best reaction time for L7 was
found to be 9 h.
(ESI,† Fig. S1 and S2). As a result, ca. 75% of the ionic liquid was
recovered after each reaction and was able to reuse for the
depolymerisation reactions. In addition, catalysts were not able
to recover and reuse.
Electrochemical measurements
In summary, the GPC chromatogram of the lignin depoly-
merisation in Fig. 3 shows that lignin could be depolymerised In the depolymerisation reactions, Cu(I)Cl was added to coordinate
into oligomers with an average molecular weight of about M with different ligands to form copper complexes, in which the
000 in [emim][ABS] using copper–L6 and copper–L7 com- Cu center was then oxidized to Cu under air flow. Cyclic
n
=
+
2+
2
plexes within 12 h and 9 h, respectively. Comparing the results voltammetric measurements were carried out to determine the
of the depolymerisation catalysed by the copper–L6 complex reversible potentials of the copper centers of the copper com-
with those of the copper–L7 complex, the copper–L7 complex plexes with different N-based ligands. This part of work aimed
could catalyse lignin depolymerisation more efficiently. These to study if the differences in the reversible potentials of copper
copper complexes were used as catalysts to generate radicals on complexes could correlate with the different depolymerising
phenols to perform the depolymerisation process. The hypothesis ability of copper complexes with each ligand in [emim][ABS].
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
depolymerisation in DMF and the ionic liquid, [emim][ABS], as
solvents. Among all these seven ligands, both L6 and L7 were
the most efficient ligands to catalyse lignin depolymerisation in
[
emim][ABS]. Lignin can be depolymerised into oligomers with
TBDMP under oxidative conditions involving the redistribution
mechanism with the average molecular weight M = 2000 in
emim][ABS] using copper–L6 and copper–L7 complexes within
n
[
12 h and 9 h, respectively. Cyclic voltammetric results showed
that copper–L6 and copper–L7 have more positive reversible
potentials than other copper complexes and this could be the
reason for the high depolymerisation efficiency of these two
ligands compared to the other five ligands.
Experimental
Materials
Organosolv lignin, DMP, Cu(I)Cl, sodium dodecyl sulfate (SDS),
1
-ethyl-3-methylimidazolium chloride, PMDETA, TMEDA, TPA,
Fig. 7 Cyclic voltammograms of copper complexes with different ligands.
dichloromethane (DCM), 2-pyridinecarboxaldehyde, aniline,
and p-anisidine were purchased from Sigma-Aldrich. Hydrochloric
acid (32%) was obtained from Ajax Finechem. High purity oxygen
purchased from Air Liquid, Australia, was used for experiments
that required oxygen gas. Dimethylformamide (DMF), for gel
0/+
Table 2
m
E of copper complexes with different ligands vs. Fc
Ligands
L1
ꢁ695
L2
L3
L4
L5
L6
ꢁ193
L7
m
E /mV
ꢁ523
ꢁ760
ꢁ210
ꢁ320
ꢁ152 permeation chromatography (GPC) analysis, chloroform (CDCl
3
),
deuterium oxide (D O), and methanol were purchased from
2
Merck. All reagents were used as purchased from the supplier
without any further purification.
+
/2+
The Cu
redox processes associated with the copper complexes
are shown in Fig. 7. At a more negative potential region, the
+
reduction of Cu to metallic Cu is evident since a characteristic Measurements
copper stripping peak was observed in the reverse anodic sweep
results not shown). In addition, the ligands without copper
1
H NMR spectra were recorded on a Bruker DRK-400 spectro-
meter operating at 400 MHz as solutions in CDCl and D O.
(
3
2
were also analysed by cyclic voltammetry (ESI,† Fig. S4). There
were no detected signals or visible reductive or oxidative
potential peaks from the cyclic voltammetry of all these ligands
from ꢁ800 mV to ꢁ100 mV.
The molecular weight of isolated lignin and depolymerised
samples was measured by GPC performed on a Tosoh EcosHLC-
8320 Gel Permeation Chromatograph equipped with both
refractive index (RI) and ultraviolet (UV) detectors (UV detec-
tion, l = 280 nm) using Tosoh alpha 4000 and 2000 columns.
DMF (with 10 mM LiBr) was used as the mobile phase with a
The E
m
(reversible potential) values (taken as the average
of the reduction and oxidation peak potentials) of copper
0/+
complexes with different ligands vs. Fc are shown in Table 2.
The copper centers of different copper complexes with different
N-based ligands show different Em values. The Em values (Table 2)
of copper–L6 and copper–L7 are more positive than those with
other ligands, indicating that they are stronger oxidizing agents
than other copper complexes coordinating with other ligands.
Based on the depolymerisation mechanism, this result could
explain why copper–L6 and copper–L7 more efficiently depoly-
merise lignin compared to the other ligands. The reversible
potential of copper–L7 is more positive than that of copper–L6,
which is consistent with the fact that copper–L7 is the best copper
complex for lignin depolymerisation involving the redistribution
mechanism among the seven ligands.
ꢁ1
flow rate of 1.0 mL min . A UV detector in GPC chromato-
grams set at 280 nm was used to analyze all the depolymerised
samples and isolated lignin. Calibration curves were obtained
using polystyrene standards.
Syntheses
4
-tert-Butyl-2,6-dimethylphenol (TBDMP). TBDMP was prepared
4
0
from DMP by following the reported literature procedure.
max (KBr)/cm
H NMR (400 MHz, CDCl ): d 7.03 (s, 2H, ArH), 2.28 (s,
H, CH
ꢁ1
n
3403 (O–H), 2917 (C–H), 1617 (CQC).
1
3
H
1
3
6
3 3 3
), 1.32 (s, 9H, CH ). C NMR (100 MHz, CDCl ):
d
C
16.3, 31.7, 34.0, 122.5, 125.7, 143.0, 150.0. MS (+ESI);
m/z; 178.14 [M+].
Ionic liquid 1-ethyl-3-methylimidazolium xylenesulfonate
(
[emim][ABS]). [emim][ABS] was prepared using the reported
Conclusions
3
9
ꢁ1
procedure.
n
max (KBr)/cm
3053, 2964, 2924, 2855, 1654,
O) d 8.55–8.45
form copper complexes and were firstly used to catalyse lignin (s, [emim] ArH), 8.21–8.19 (s, [ABS] ArH), 7.80–6.90 (m, [ABS]
1
Seven N-based ligands were found to coordinate with copper to 1465, 1327, 969, 771. H NMR (400 MHz, D
2
H
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
ArH), 4.15–4.00 (q, [emim] –CH
2
–CH
), 2.65–2.55 (q, [ABS] –CH
CH ), 2.53–2.40 (m, [ABS] CH –Ar), 2.28 (s, [ABS] –CH –Ar),
3
), 3.73 (s, [emim] –CH
3
), Isolation of the high molecular weight fraction of lignin using a
3.00–2.75 (m, [ABS] –CH
3
–CH–CH
3
2
–
solubility method
.98 g of dried organosolv lignin was added to methanol
600 mL) and the suspension was stirred (500 rpm) using a
3
3
3
5
(
2
1
.24–2.12 (m, [ABS] –CH –Ar), 1.45–1.25 (t, [C mim] –CH –CH ),
3 2 2 3
.20–1.11 (m, [ABS] –CH2 –CH , CH –CH–CH ). C NMR
1
3
3
3
3
magnetic stirrer for 2 h at room temperature (23 1C), and then
filtered through a pre-dried (105 1C overnight) weighed cellulose
filter paper. The solid lignin residue together with the filter
paper was then oven-dried overnight at 105 1C. After cooling in
a desiccator, the dried solid methanol-insoluble residue (M-lignin)
was weighed and collected. Evaporation of solvent from the filtrate
under vacuum afforded a soluble fraction. The percentage of the
insoluble material was calculated based on the dry weight. The
molecular weight distribution of the methanol-insoluble fraction
of M-lignin was determined by GPC.
(
1
100 MHz, D
2
O): d
23.04, 42.01, 27.22, 21.3 m/z (+ESI) 282.1.
E)-N-sec-Butyl-1-(pyridin-2-yl)methanimine (L5). The N-alkyl-(2-
pyridyl)methanimine ligand was synthesized using the reported
C
139.40, 137.21, 130.01, 128.24, 122.81,
(
4
1
ꢁ1
procedure. B.p. 74 1C at 5.0 Torr. nmax (KBr)/cm
1630,
1
1
1
1
1
3
1
571, 1454. H NMR (CDCl , 400 MHz): d 8.50 (d, 4.56 Hz,
3
H
H), 8.24 (s, 1H), 7.86 (d, 7.68 Hz, 1H), 7.57 (t, 7.36 Hz,
H), 7.14 (dd, 4.92, 6.32 Hz, 1H), 3.17 (sext, 6.32 Hz, 1H),
.50 (m, 7.72 Hz, 2H), 1.13 (d, 6.28 Hz, 3H), 0.72 (t, 7.36 Hz.
1
3
H).
C
NMR (CDCl
3
,
100 MHz):
d
C
161.60, 154.49,
49.16, 136.29, 124.38, 120.96, 69.36, 29.27, 22.50, 10.90. m/z
Depolymerisation
(ESI) 163.1.
(E)-N-tert-Butyl-1-(pyridin-2-yl)methanimine (L4). This com-
Screening depolymerisation of lignin in DMF. A typical
pound was obtained as a yellow liquid (322.2 mg, 85%) following lignin depolymerisation procedure is summarized as follows:
4
2
ꢁ1
the reported procedure. nmax (KBr)/cm 1637, 1569, 1451. for each reaction, lignin (0.31 g), TBDMP (0.45 g, 2.5 mmol),
1
H NMR (400 MHz, CDCl ) dH 8.63(d, 4.0 Hz, 1H), 8.36 (s, 1H), Cu(I)Cl (0.025 g, 0.25 mmol) and pyridine/ligands (2 mL,
3
8
1
1
.02 (d, 7.9 Hz, 1H), 7.73 (t, 7.7 Hz, 1H), 7.29 (ddd, 7.5, 4.9, 0.25 mmol) were added to DMF (20 mL). The reaction mixture
1
3
.2 Hz, 1H), 1.31 (s, 9H). C NMR (CDCl
57.04, 156.69, 150.01, 137.05, 125.26, 120.74, 57.94, 10.8. m/z mixture (5 mL) were withdrawn at the following time intervals:
1 h, 3 h and 6 h. At the end of 6 h, the reaction was stopped by
3
, 100 MHz): d
C
was stirred under oxygen at 120 1C and samples of the reaction
(ESI) 163.1.
4
1
General procedure for (E)-N-(pyridin-2-ylmethylene)aniline dropwise addition of 1 M HCl until pH 2 was reached, causing
L6) and (E)-4-methoxy-N-(pyridin-2-ylmethylene)aniline (L7). precipitation of the depolymerised product. The suspension
(
To a solution of pyridine carboxaldehyde (1.2 equiv.) in dichloro- was centrifuged several times with distilled water until the
methane DCM were added dry MgSO (5 g, 41.5 mmol) followed pH of the supernatant reached 7. The precipitate was then
4
by N-based ligands (1.0 equiv.). The reaction mixture was stirred dried under vacuum at room temperature. The initial acidified
at room temperature for 24 hours, and filtered and the solvent supernatant was then extracted with 3 volumes of dichloro-
was removed in vacuo. Ligands were then purified by distillation methane and dichloromethane was removed under vacuum to
at reduced pressure.
E)-N-(Pyridin-2-ylmethylene)aniline (L6). 2-Pyridinecarboxy- obtained in this way was analysed by GPC. A control reaction
aldehyde (0.32 g, 3.0 mmol, 1.2 equiv.), aniline (0.23 g, 2.5 mmol, was carried out without TBDMP (monomer) or catalyst (Cu(I)-
.0 equiv.) and dichloromethane (20 mL) were subjected to the Cl/ligands) under the same conditions in order to compare
give a residue. The molecular weight of all of the products
(
1
4
1
general procedure for the synthesis of ligands. Purification by the results.
vacuum distillation resulted in dark orange oil (0.45 g, 2.5 mmol, Screening depolymerisation of lignin in 1-ethyl-3-methylimid-
9%); nmax (neat, cm ) 3053, 1627, 1591, 1485; H NMR azolium xylenesulfonate ([emim][ABS]). M-lignin (0.005 g), TBDMP
ꢁ
1
1
9
ꢁ
2
ꢁ3
(
(
(
400 MHz, CDCl ): d
3 H
8.72 (1H, d, 4.8 Hz), 8.61 (1H, s), 8.21 (7 mg, 4 ꢀ 10 mmol), CuCl
2
ꢂ2H O (0.7 mg, 4 ꢀ 10 mmol) and
2
ꢁ
3
1H, d, 7.7 Hz), 7.82 (1H, td, 7.7, 1.6 Hz), 7.45–7.40 (2H, m), 7.37 ligands (4 ꢀ 10 mmol) were added to [emim][ABS] (0.5 g).
13
1H, ddd, 7.7, 4.8, 1.0 Hz), 7.32–7.24 (3H, m); C NMR The reaction mixture was stirred under oxygen at 180 1C for
(
100 MHz, CDCl ): d 160.7, 154.6, 151.0, 149.7, 136.7, 129.3, 6 h. At the end of 6 h, the pH of the mixture was adjusted to
3
C
1
26.8, 125.2, 121.9, 121.1; m/z (ESI) 183.1.
E)-4-Methoxy-N-(pyridin-2-ylmethylene)aniline (L7). 2-Pyridine- was collected and washed by centrifugation with distilled
carboxyaldehyde (0.32 g, 3.0 mmol, 1.2 equiv.) and p-anisidine water until the pH of the supernatant reached 7. The washed
0.31 g, 2.5 mmol, 1.0 equiv.) in dichloromethane (20 mL) precipitate was dried under vacuum at room temperature.
2 by dropwise addition of 1 M HCl. The resulting precipitate
(
(
4
1
were subjected to the general procedure for the synthesis of The molecular weight distributions of the depolymerised
ligands. Purification by vacuum distillation resulted in dark products were characterized by GPC. A control reaction was
orange oil (0.45 g, 2.12 mmol, 86%); nmax (neat, cm ) 3051, carried out in the absence of TBDMP and the catalyst (Cu(II)Cl/
ꢁ1
1
2834, 1624, 1579, 1503, 1242; H NMR (400 MHz, CDCl ): d
ligands) as described above in order to compare the results.
3
H
8.62 (1H, d, 4.7 Hz), 8.55 (1H, s), 8.10 (1H, d, 7.8 Hz), 7.71
Lignin depolymerisation in 1-ethyl-3-methylimidazolium
(
(
1
5
1H, td, 7.8, 1.6 Hz), 7.31–7.21 (3H, m), 6.94–6.81 (2H, m), 3.76 xylenesulfonate ([emim][ABS]). M-lignin (0.04 g), TBDMP
1
3
3H, s, OCH
3
);
57.6, 154.3, 149.0, 143.1, 136.0, 124.2, 122.1, 121.0, 113.8, ligands (0.66 mmol) were added to [emim][ABS] (4 g). The reaction
4.9; m/z (ESI) 213.1. mixture was stirred under oxygen at 180 1C for 48 h. And samples
C NMR (100 MHz, CDCl
3
): d
C
158.3, (0.056 g, 0.33 mmol), CuCl
2
ꢂ2H
2
O (0.0033 g, 0.033 mmol) and
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
were taken from the mixture at the following reaction time: 3 h,
6 P. Azadi, O. R. Inderwildi, R. Farnood and D. A. Kin, Renewable
Sustainable Energy Rev., 2012, 21, 506–523.
6
h, 9 h, 12 h, 18 h, 24 h, 36 h and 48 h. After every sample had
been taken from the mixture, we followed the previous screening
procedure for purification and characterisation.
7 J. Zakzeski, P. C. Bruijnincx, A. L. Jongerius and B. M.
Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
Recovery of [emim][ABS]. After each depolymerisation reaction,
the mixture was acidified with 1 M HCl and centrifuged with
8 J. M. Pepper and Y. W. Lee, Can. J. Chem., 1969, 47, 723–727.
9 M. Koyama, Bioresour. Technol., 1993, 44, 209–215.
distilled water to precipitate and remove depolymerised lignin. 10 N. A. Stephenson and A. T. Bell, J. Mol. Catal. A: Chem., 2007,
The acidic aqueous filtrate containing ionic liquid, copper
275, 54–62.
chloride, and hydrochloric acid was neutralised using 1 M 11 C.-M. Che and J.-S. Huang, Chem. Commun., 2009,
NaOH. Next, the mixture was left under high vacuum at the
3996–4015.
temperature of 70 1C overnight to remove water. Acetonitrile 12 P. Zucca, F. Sollai, A. Garau, A. Rescigno and E. Sanjust,
100 mL) was added to the mixture to dissolve only the ionic
J. Mol. Catal. A: Chem., 2009, 306, 89–96.
liquid and to remove all the insoluble residue (copper chloride 13 K. C. Gupta, A. K. Sutar and C.-C. Lin, Coord. Chem. Rev.,
and sodium chloride) by filtration. The ionic liquid was recovered
2009, 253, 1926–1946.
by removing acetonitrile at 70 1C under high vacuum. The 14 R. Drago, B. Corden and C. Barnes, J. Am. Chem. Soc., 1986,
(
1
recovered ionic liquid was then analysed by H NMR to confirm
its structure.
108, 2453–2454.
15 A. Hay, H. Blanchard, G. Endres and J. Eustance, J. Am.
Chem. Soc., 1959, 81, 6335–6336.
Electrochemical measurement. Electrochemical experiments
were carried out at the temperature of 23 ꢃ 2 1C in a standard 16 W. Brackman and E. Havinga, Recl. Trav. Chim. Pays-Bas,
three-electrode cell configuration using a Bioanalytical Systems
1955, 74, 1021–1039.
(
BAS West Lafayette, Indiana) Model 100B potentiostat at a scan 17 H. Higashimura, K. Fujisawa, M. Kubota and S. Kobayashi,
ꢁ1
rate of 100 mV s . A glassy carbon disc electrode (1 mm
J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 1955–1962.
diameter, eDAQ) was used as the working electrode. A platinum 18 K. Saito, T. Masuyama, K. Oyaizu and H. Nishide, Chem. –
wire was employed as the reference electrode and another
Eur. J., 2003, 9, 4240–4246.
platinum wire as the counter electrode. The reference potential 19 S. Nanayakkara, A. F. Patti and K. Saito, Aust. J. Chem., 2013,
+
was calibrated against that of the Fc/Fc (Fc = ferrocene) redox
66, 60–66.
couple as an internal reference from measurements made on 20 S. Nanayakkara, A. F. Patti and K. Saito, Green Chem., 2014,
4
3
the oxidation of 1 mM Fc present in the same solution.
The voltammetric investigation was carried out in 0.5 mL of 21 Z. C. Zhang, Adv. Catal., 2006, 49, 153–237.
emim][ABS] in the presence of 0.0165 mmol Cu(I)Cl, and 22 H. Olivier-Bourbigou, L. Magna and D. Morvan, Appl. Catal.,
.33 mmol ligands. Prior to voltammetric experiments, the
A, 2010, 373, 1–56.
glassy carbon electrode was polished with 0.3 mm alumina 23 J. B. Binder, M. J. Gray, J. F. White, Z. C. Zhang and J. E. Holladay,
slurry on a clean polishing cloth (Buehler, USA), rinsed with
Biomass Bioenergy, 2009, 33, 1122–1130.
deionized water, washed with acetone and finally dried with 24 K. St ¨a rk, N. Taccardi, A. B o¨ smann and P. Wasserscheid,
16, 1897–1903.
[
0
nitrogen gas.
ChemSusChem, 2010, 3, 719–723.
2
2
5 B. J. Cox and J. G. Ekerdt, Bioresour. Technol., 2013, 134, 59–65.
6 A. J. Clark, A. E. Collis, D. J. Fox, L. L. Halliwell, N. James,
R. K. O’Reilly, H. Parekh, A. Ross, A. B. Sellars and H. Willcock,
J. Org. Chem., 2012, 77, 6778–6788.
Acknowledgements
The financial support of the ARC Industrial Transformation
Research Hub – Bioprocessing Advanced Manufacturing Initiative
2
7 D. M. Haddleton, D. J. Duncalf, D. Kukulj, M. C. Crossman,
S. G. Jackson, S. A. Bon, A. J. Clark and A. J. Shooter, Eur.
J. Inorg. Chem., 1998, 1799–1806.
8 A. J. Clark, D. J. Duncalf, R. P. Filik, D. M. Haddleton, G. H.
Thomas and H. Wongtap, Tetrahedron Lett., 1999, 40, 3807–3810.
9 A. J. Clark, G. M. Battle, A. M. Heming, D. M. Haddleton and
A. Bridge, Tetrahedron Lett., 2001, 42, 2003–2005.
(BAMI), Monash University and the China Scholarship Council
(CSC) is gratefully acknowledged.
2
2
Notes and references
1
F. S. Chakar and A. J. Ragauskas, Ind. Crops Prod., 2004, 20, 30 D. K. Seth and S. Bhattacharya, Polyhedron, 2011, 30, 2438–2443.
1
31–141.
31 S. Dehghanpour, N. Bouslimani, R. Welter and F. Mojahed,
Polyhedron, 2007, 26, 154–162.
32 A. J. Clark, Chem. Soc. Rev., 2002, 31, 1–11.
2
3
4
M. P. Pandey and C. S. Kim, Chem. Eng. Technol., 2011, 34, 29–41.
J. H. Grabber, Crop Sci., 2005, 45, 820–831.
J. Ralph, K. Lundquist, G. Brunow, F. Lu, H. Kim, P. F. 33 T.-H. Huang and M.-H. Zhang, Inorg. Chim. Acta, 2014, 410,
Schatz, J. M. Marita, R. D. Hatfield, S. A. Ralph and J. H.
Christensen, Phytochem. Rev., 2004, 3, 29–60.
150–155.
34 K. Saito, N. Kuwashiro and H. Nishide, Polymer, 2006, 47,
5
D. S. Argyropoulos, in Biotechnology in the Pulp and Paper
6581–6584.
Industry, ed. Karl-Erik L. Eriksson, Springer, Heidelberg, 35 K. Saito, T. Masuyama and H. Nishide, Green Chem., 2003, 5,
1st edn, 1997, ch. 3, pp. 127–158.
535–538.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
36 R. Bailey, S. Kozak, T. Michelsen and W. Mills, Coord. Chem. 40 G. A. Olah, M. W. Meyer and N. A. Overchuk, J. Org. Chem.,
Rev., 1971, 6, 407–445.
1964, 29, 2313–2315.
3
3
7 E. T. Adman, Adv. Protein Chem., 1991, 42, 145–197.
8 E. A. Ambundo, M.-V. Deydier, A. J. Grall, N. Aguera-Vega,
L. T. Dressel, T. H. Cooper, M. J. Heeg, L. Ochrymowycz and
D. Rorabacher, Inorg. Chem., 1999, 38, 4233–4242.
9 S. S. Tan, D. R. MacFarlane, J. Upfal, L. A. Edye, W. O. Doherty,
41 D. M. Haddleton, M. C. Crossman, B. H. Dana, D. J. Duncalf,
A. M. Heming, D. Kukulj and A. J. Shooter, Macromolecules,
1999, 32, 2110–2119.
42 K. Kloc, E. Kubicz, J. Młochowski and L. Syper, Synthesis,
1987, 1084–1087.
3
A. F. Patti, J. M. Pringle and J. L. Scott, Green Chem., 2009, 11, 43 J. Zhang and A. M. Bond, Anal. Chem., 2003, 75,
39–345. 2694–2702.
3
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.