Metal Triflates for the Production of Aromatics from Lignin

Article abstract of DOI:10.1002/cssc.201600831

The depolymerization of lignin into valuable aromatic chemicals is one of the key goals towards establishing economically viable biorefineries. In this contribution we present a simple approach for converting lignin to aromatic monomers in high yields under mild reaction conditions. The methodology relies on the use of catalytic amounts of easy-to-handle metal triflates (M(OTf)_x). Initially, we evaluated the reactivity of a broad range of metal triflates using simple lignin model compounds. More advanced lignin model compounds were also used to study the reactivity of different lignin linkages. The product aromatic monomers were either phenolic C2-acetals obtained by stabilization of the aldehyde cleavage products by reaction with ethylene glycol or methyl aromatics obtained by catalytic decarbonylation. Notably, when the method was ultimately tested on lignin, especially Fe(OTf)₃ proved very effective and the phenolic C2-acetal products were obtained in an excellent, 19.3±3.2 wt % yield.

Full text of DOI:10.1002/cssc.201600831

DOI: 10.1002/cssc.201600831
Full Papers
Metal Triflates for the Production of Aromatics from Lignin
Peter J. Deuss,[a] Ciaran W. Lahive,[b] Christopher S. Lancefield,[b] Nicholas J. Westwood,[b]
Paul C. J. Kamer,[b] Katalin Barta,*[a] and Johannes G. de Vries*[a, c]
The depolymerization of lignin into valuable aromatic chemi-
cals is one of the key goals towards establishing economically
viable biorefineries. In this contribution we present a simple
approach for converting lignin to aromatic monomers in high
yields under mild reaction conditions. The methodology relies
on the use of catalytic amounts of easy-to-handle metal tri-
flates (M(OTf)x). Initially, we evaluated the reactivity of a broad
range of metal triflates using simple lignin model compounds.
More advanced lignin model compounds were also used to
study the reactivity of different lignin linkages. The product ar-
omatic monomers were either phenolic C2-acetals obtained by
stabilization of the aldehyde cleavage products by reaction
with ethylene glycol or methyl aromatics obtained by catalytic
decarbonylation. Notably, when the method was ultimately
tested on lignin, especially Fe(OTf)3 proved very effective and
the phenolic C2-acetal products were obtained in an excellent,
19.3Æ3.2 wt% yield.
Introduction
The development of fundamentally new catalytic methods is
of central importance for the production of bulk and fine
chemicals from renewable lignocellulose feedstocks.[1,2] In this
context, it is very important to valorize all main constituents of
lignocellulose, including lignin.[3] However, despite recent ef-
forts, the catalytic conversion of lignin has proven very chal-
lenging.[4] Acidolysis is an efficient method for the cleavage of
the most abundant b-O-4 linkage in lignin (Scheme 1).[5] In ad-
dition, acid-catalyzed depolymerization is a highly relevant
method especially in relation to the future biorefinery concept
as the most common organosolv lignin extraction methodolo-
gies also use acidic media.[6] One of the main challenges with
acid-mediated degradation of lignin is the reconstitution of re-
active fragments leading to more robust oligomeric structures
often referred to as biochar. In essence, acidolysis leads to phe-
nolic C2-acetaldehydes that are notoriously unstable under
acidic conditions. This is the major reason why this acidolysis
pathway has largely escaped attention for the production of
aromatic monomers from lignin until recently. In new studies
by our group and others, it was shown that by capturing these
reactive intermediates formed upon acidolysis, lignin can be
effectively depolymerized with suppression of recondensation
pathways leading to improved monomer yields (Scheme 1b).[7]
Of the different in situ stabilization methodologies applied,
in particular acetal formation with ethylene glycol led to a de-
fined set of major products in good yields from models and
lignin. The cleavage of the b-O-4 linkage was most efficiently
promoted by strong acids with non-coordinating anions such
as triflic acid (HOTf).[7a,d] However, HOTf is corrosive and incon-
venient to handle, which may lead to inconsistent results. On
the other hand, metal triflates (M(OTf)x) are weighable solids
that are less corrosive than HOTf and therefore much less haz-
ardous in handling.[8]
Additionally, in the case of other homogeneous catalysts
used, for example, in the iridium-catalyzed decarbonylation to
yield p-cresol and mono- and di-methoxylated cresols, the
strongly acidic conditions may affect the stability of the iridium
catalyst.[7a] For this reason, we were interested to assess the
usefulness of Lewis acids such as metal triflates and their com-
patibility with the homogeneous metal catalysts used.
In this contribution, we evaluate the reactivity of different
metal triflate salts for the cleavage of b-O-4, b-5, and b-b lignin
model compounds. We determine the quantities of ethylene
glycol acetals or methyl aromatics obtained after stabilization
of reactive intermediates by acetal formation or decarbonyla-
tion upon bond cleavage. Ultimately, the depolymerization of
walnut methanosolv lignin was successfully carried out and
the use of Fe(OTf)3 led to results surpassing those obtained
with HOTf.
[a] Dr. P. J. Deuss, Dr. K. Barta, Prof. J. G. de Vries
Stratingh Institute for Chemistry
University of Groningen
Nijenborgh 4, 9747 AG, Groningen (The Netherlands)
E-mail: k.barta@rug.nl
j.g.devries@rug.nl
[b] C. W. Lahive, Dr. C. S. Lancefield, Prof. N. J. Westwood, Prof. P. C. J. Kamer
School of Chemistry and Biomedical Science Research Complex
University of St Andrews and EaStCHEM
North Haugh, St Andrews, Fife, KY16 9ST (United Kingdom)
Results and Discussion
[c] Prof. J. G. de Vries
Metal triflate-catalyzed cleavage of b-O-4 model compound
1
Leibniz-Insitut fꢀr Katalyse e.V. an der Universitꢁt Rostock
Albert-Einstein-Straße 29a, 18059 Rostock (Germany)
E-mail: johannes.devries@catalysis.de
Acid-catalyzed cleavage of the b-O-4 model compound 1 re-
sults in the formation of 2-phenyl acetaldehyde (2) and guaia-
Supporting Information for this article can be found under http://
dx.doi.org/10.1002/cssc.201600831.
ChemSusChem 2016, 9, 1 – 9
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Scheme 1. a) Major linkage motifs in lignin. b) Acid-catalyzed cleavage of b-O-4 motif in model compounds and lignin with in situ conversion of reactive alde-
hyde intermediates. cod=1,5-cyclooctadiene
col (3) (Scheme 2),[5b,7a–b,9] and 2 is rapidly converted into its
aldol condensation products under these reaction condi-
tions.[7a,b] After evaluating a range of metal triflates, we found
that several metal triflates successfully catalyzed the cleavage
Scheme 2. Metal triflate-catalyzed cleavage of lignin b-O-4 model com-
pound 1.
of 1 (Table S2 in the Supporting Information). In toluene (Fig-
ure 1a), yields of 3 were similar or slightly higher for Al(OTf)3,
Figure 1. Cleavage of b-O-4 model compound 1 using metal triflates as catalysts in a) toluene and b) 1,4-dioxane (Reaction conditions shown in Scheme 2).
&
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Bi(OTf)3, Cu(OTf)2, Eu(OTf)3, Fe(OTf)3, Hf(OTf)4, Sc(OTf)3, and
Yb(OTf)3 when compared to HOTf. No significant reactivity was
detected for other metal triflate salts such as AgOTf, Fe(OTf)2,
Zn(OTf)2, and Ni(OTf)2. As expected, typically, only small
amounts of 2 were detected in these reactions in toluene.
Several metal triflates that showed significant cleavage activ-
ity in toluene were subsequently tested in 1,4-dioxane, which
is a suitable solvent for solubilizing lignin (Figure 1b). Here,
only Al(OTf)3, Bi(OTf)3, Fe(OTf)3, Hf(OTf)4, and Sc(OTf)3 showed
good reactivity compared to HOTf. Other metal triflates such
as Cu(OTf)3, Eu(OTf)3, and Yb(OTf)3 led to significantly lower
conversion of 1. Overall, higher yields of 2 and 3 were ob-
tained using 1,4-dioxane as solvent. Indeed, it is well-known
that 1,4-dioxane is a Lewis base and can form stable com-
plexes with Lewis acids;[10] therefore, it is expected that its use
further reduces the acidity of the reaction medium, allowing
the formation of substantial quantities of 2. To evaluate the
higher stability of 2 more quantitatively in 1,4-dioxane, we
monitored the reactions over time using Fe(OTf)3 and HOTf
(Figure 2).[7a] The reaction rates and product formation profiles
using a microsyringe either directly to the reaction mixture or
through a stock solution in the appropriate solvent.
All metal triflates that catalyzed the cleavage of 1 are known
to be strong Lewis acids and have already found many applica-
tions in organic synthesis.[11] Therefore, the question is whether
the cleavage of 1 is catalyzed by the metal triflates as Lewis
acids or alternatively by in situ formation of HOTf from the tri-
flate salts.[12] Other strong Lewis acids such as AlCl3 and FeCl3
did not show any significant activity in the cleavage of 1 in
1,4-dioxane or toluene (Table S2). The addition of non-nucleo-
philic bases such as 2,6-di-tert-butyl-4-methylpyridine[13] or
NaHCO3 completely quenched all reactivity for metal triflates
that were previously successful in the cleavage of 1 (Table S2).
These results, combined with the similar reaction progress, led
us to conclude that the triflate salts (Al(OTf)3, Bi(OTf)3, Cu(OTf)2,
Sc(OTf)3, Fe(OTf)3, Eu(OTf)3, Yb(OTf)3, and Hf(OTf)4) likely form
triflic acid in situ and are therefore capable of catalyzing the
acidolysis of 1. Nevertheless, the presence of the metal seems
to modulate the acidity somewhat, depending on the stability
of the triflate salt applied. Remarkably, a good correlation be-
tween the conversion of 1 and the hydrolysis constant of sev-
eral metal triflates were found in both toluene as well as diox-
ane (displayed in the Supporting Information, Figures S6 and
S7).
Metal triflate-catalyzed cleavage of b-O-4 model compounds
combined with decarbonylation
After establishing that metal triflates are attractive alternatives
to HOTf, we focused on the in situ catalytic decarbonylation of
aldehyde
2 towards highly desirable methyl aromatics
(Scheme 3).[7a,14] Thus, Fe(OTf)3 and Al(OTf)3 were tested under
decarbonylation conditions using [IrCl(cod)]2 and PPh3 in 1,4-di-
oxane (Figure 3, Tables S3 and S4).[7a,15] In these reactions,
Figure 2. Reaction profiles for the cleavage of 1, using 10 mol% Fe(OTf)3
(solid lines) and 10 mol% HOTf (dotted lines) in 1,4-dioxane at 1408C.
for these reactions showed a remarkably similar pattern, with
only slightly higher overall yields of 2. Additionally, the practi-
cal advantage of using the triflate salt compared to HOTf was
apparent during this experiment. The metal triflate catalyst
could easily be weighed and added using a stock solution in
1,4-dioxane or as a solid to the reaction. Because of this more
accurate data is provided compared to experiments that re-
quired the addition of microliter quantities of smoking HOTf
Scheme 3. Metal triflate-catalyzed cleavage of lignin b-O-4 model com-
pound 1 combined with in situ decarbonylation.
Figure 3. Cleavage of b-O-4 model compounds 1 and 1c catalyzed by metal triflates and in situ decarbonylation of 2 and 2c using 5 mol% [IrCl(cod)]2 and
10 mol% PPh3.
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
a lower metal triflate loading and milder temperature were
used to prevent the build-up of aldehyde 2, which was neces-
sary as the decarbonylation of 2 forming toluene 4 catalyzed
by Ir/PPh3 was found to be relatively slow. The results for
Fe(OTf)3 and HOTf at 50 mm substrate concentration and
10 mol% catalyst loading were similar. Higher yields of 3 could
be achieved with 25 mm substrate concentration and 5 mol%
catalyst loading but with prolonged reaction time. At this sub-
strate concentration, Fe(OTf)3 resulted in slightly better selec-
tivity towards 4 (76% vs. 70%). Using Al(OTf)3, the conversion
of 1 and selectivities towards 3 and 4 were significantly lower.
The same trends were observed for a similar b-O-4 model com-
pound 1c, yielding up to 83% 4c and showing a small in-
crease in decarbonylation efficiency in the presence of Fe(OTf)3
over HOTf (Figure 3, Table S5). The results demonstrate that
the balance between the rate of aldehyde formation and the
rate of decarbonylation is crucial to achieve high product
yields.
Scheme 4. Cleavage of lignin b-O-4 model compounds 1, 1b, and 1c in the
presence of ethylene glycol to form 5 and 5c catalyzed by M(OTf)x.
(Figure 4 and Table S6). Other triflate salts tested showed sig-
nificantly lower activity in the presence of ethylene glycol com-
pared to the previous runs in 1,4-dioxane alone (compare
Figure 4 and Figure 1b). In particular, Sc(OTf)3 showed much
lower conversion of 1 and only traces of 5 with some build-up
of 2 under these reaction conditions. In a separate set of ex-
periments, we showed that in particular Sc(OTf)3 is relatively in-
efficient in catalyzing the formation of acetal 5 from 2 and eth-
ylene glycol (Figure S5a).
Next, the reaction profiles for the cleavage of 1 and acetal
formation with Fe(OTf)3, Hf(OTf)3 and Bi(OTf)3 were compared
to HOTf (Figure 5). HOTf and Fe(OTf)3 again showed very simi-
lar conversion of 1 and yields of 3 and 5 (compare Figure 5a
and b). This similarity between HOTf and Fe(OTf)3 was also ob-
served for reactions in toluene as well as other b-O-4 model
compounds (1b and 1c, Figures S2–S4). With 10 mol%
Hf(OTf)4, full substrate conversion was seen, but in this case
within 1 h and a slight buildup of 2 was observed in the first
Metal triflate-catalyzed cleavage of b-O-4 model compounds
combined with ethylene glycol acetal formation
Next, we turned our attention to the cleavage of 1 with in situ
conversion of the formed aldehyde 2 to its more stable 1,3-di-
oxolane acetal 5 (Scheme 4) using ethylene glycol. Several tri-
flates such as Bi(OTf)3, Fe(OTf)3, and Hf(OTf)4 showed excellent
yields of 1,3-dioxolane acetal product 5 and guaiacol
3
Figure 4. Cleavage of b-O-4 model compound 1 using metal triflate catalysts. Reaction conditions shown in Scheme 4. (Slightly different results were obtained
with Fe(OTf)3 from different commercial sources, see Table S1).
Figure 5. Reaction progress of the cleavage of 1 and in situ acetal formation with ethylene glycol catalyzed by a) 10 mol% HOTf, b) 10 mol% Fe(OTf)3,
c) 10 mol% Hf(OTf)4, and d) 10 mol% Bi(OTf)3.
&
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
15 min (Figure 5c). As aldehyde 2 was already shown to be un-
stable in the experiments above, this explains the overall lower
selectivity of 5 with Hf(OTf)4 as catalyst. The use of Bi(OTf)3 re-
sulted in slower reactions but similar product yields upon
180 min reaction time (Figure 5d). Overall, these reactions
showed a clear trend in the rate of cleavage of 1: Hf(OTf)4 >
HOTf=Fe(OTf)3 >Bi(OTf)3.
of the corresponding C2-aldehydes were detected as major
cleavage products, albeit at lower yields. This is due to a com-
peting cleavage pathway that leads to C3 aromatic products
that relate to the so-called Hibbert ketones detected in tradi-
tional acidolysis reaction mixtures (Scheme 6).[5c,16] It was also
shown that the b-5 linkage in 7–9 undergoes ring opening to
ultimately form the corresponding trans-stilbenes 12–14 as
major product (Scheme 5b and c). The product yields indicate
a preference for the cleavage of the b-O-4 moiety via a path-
way that releases the carbinol group as formaldehyde. The b-5
linkage is almost exclusively modified via a similar mecha-
nism.[17] This also implies that the formaldehyde released
during these reactions is probably trapped in the form of its
Reactivity of advanced lignin b-O-4, b-5, and b-b model
compounds
Next, we turned our attention to the reactivity of a more com-
plex b-O-4 motif containing all functionalities (carbinol-group
and additional electron-donating aromatic-ring substituents) as
well as other types of lignin linkages (b-5 and b-b) in the pres-
ence of metal triflates. For this purpose, we used a set of ad-
vanced lignin model compounds previously developed and
employed by our groups for mechanistic investigations.[7d]
These compounds contain b-O-4 (6, 8 and 9), b-5 (7) and b-
b (10) linkages comprising relevant functional groups that re-
flect the structure of these linkages in lignin (Scheme 5). Model
compounds of this level of complexity are rarely used due to
limited accessibility and analytical challenges associated with
product analysis. Herein, we have a unique opportunity to
gain more detailed insight into the effectiveness of the meth-
odology employed prior to testing lignin.[7d]
Model compounds 6–10 were exposed to the cleavage and
in situ acetal formation conditions in the presence of Fe(OTf)3
(Scheme 5). The b-O-4 linkages in compounds 6, 8, and 9 were
fully cleaved within 15 min and provided the corresponding
phenolic product 3 in high yields as determined by HPLC
(Scheme 5a and c). The ethylene glycol acetals 11, 13, and 14
Scheme 6. Competing “C2” and “C3” b-O-4 cleavage pathways after forma-
tion of the benzylic carbocation.
Scheme 5. Reaction of a) Complex b-O-4 model compound 6, b) b-5 model compound 7, c) Advanced b-O-4-b-5 model compounds 8 and 9, d) b-b model
compound 10. Reaction conditions: 10 mol% Fe(OTf)3, ethylene glycol, in 1,4-dioxane.
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
ethylene-glycol acetal 1,3-dioxolane as also confirmed in our
earlier studies.[7d]
The b-b model compound 10 underwent epimerization
upon reaction with Fe(OTf)3 in the presence of ethylene glycol
forming a mixture of epimers 10/10e1/10e2 at a ratio of
1:1:0.1 as determined by gas chromatography–flame ionization
detector (GC–FID; Scheme 5d). We have previously obtained
the same results using 10 mol% HOTf as catalyst in the same
solvent.[7d] Overall, it can be concluded that a very similar reac-
tivity was observed for the advanced model compounds 6–10
both in the presence of Fe(OTf)3 and HOTf. Regarding the dif-
ferent types of linkages, the b-O-4 linkage model compounds
were efficiently cleaved to two aromatic monomers (acetal and
phenol product); however, the b-5 and b-b linkage models re-
sulted in the formation of modified aromatic dimers. This
means that when applying these methods using lignin as sub-
strate, only the scission of the b-O-4 moiety will allow depoly-
merization. This highlights the importance of having a high b-
O-4 content in the lignin used to achieve high aromatic mono-
mer yields using the presented methodology.
Figure 6. Yields of P1-3 after depolymerization of 50 mg walnut methano-
solv lignin with 23 mmol M(OTf)x in 1 mL 1,4-dioxane with 60 wt% ethylene
glycol for 15 min at 1408C (Shown data are averages of 2–3 identical experi-
ments).
sis of the starting lignin.[7d] Overall, excellent yields of P1–P3
were obtained, reaching over 10 wt% for all metal triflate-cata-
lyzed reactions. Interestingly, Fe(OTf)3 performed better than
the other triflates and HOTf, reaching an excellent yield of P1-
P3 of 19.3Æ3.2 wt%. Thus far, only a few reports exist with
such high yields of aromatic monomers, especially of product
mixtures with limited complexity;[3a–d,18] thus, this method fur-
ther contributes to achieving the highly efficient valorization
of renewable resources and the production of distinct valuable
monomers in high yields.[19]
Metal triflate-catalyzed depolymerization of walnut metha-
nosolv lignin
Finally, after studying the reactivity of all main types of linkag-
es, the metal triflates Bi(OTf)3, Fe(OTf)3, and Hf(OTf)4 that
showed the most promising results in the model compound
studies were tested in lignin depolymerization in the presence
of ethylene glycol. The results were compared to those ob-
tained with HOTf (Scheme 7). For these reactions, methanosolv
walnut lignin was isolated from walnut shells. This lignin has
a higher b-O-4 content (26 b-O-4 linkage per 100 aromatic
units) compared to other organosolv lignins tested previously
by our group.[7a,d] The very simple catalytic methodology con-
sisted of runs using 50 mg lignin, 60 wt% ethylene glycol, and
catalytic amounts of M(OTf)x at 1408C for 15 min in 1,4-diox-
ane. The depolymerization mixtures were subjected to a fractio-
nation procedure in which the low-molecular weight material
was extracted using toluene (Table S7).
Conclusions
Overall, this study shows that several metal triflates (M(OTf)x)
can be excellent substitutes for HOTf in the depolymerization
of lignin. The general reactivities of lignin model compounds
compared well to those observed with triflic acid, and prelimi-
nary studies suggest that triflic acid formed in situ is responsi-
ble for the reactivity of these metal triflates. While model com-
pounds mirroring all three major lignin linkages (b-O-4, b-b, b-
5) showed similar behavior in the presence of triflates and trifl-
ic acid, there were variations in substrate conversion and prod-
uct yields depending on the type of metal triflate used. For ex-
ample, Hf(OTf)4 showed higher activity for the cleavage of b-O-
4 model compounds but lower product selectivity, whereas
other triflate salts such as Bi(OTf)3 showed lower activity but
To evaluate the effectiveness of the metal triflate catalysts,
we focused on the major monomeric products P1–P3, which
were quantified by GC–FID using an internal standard
(Figure 6, Table S8). The ratio of P1/P2/P3 was found to be
4:33:63 for all experiments, in almost perfect agreement with
the H/G/S ratio of 6:29:65 determined by 2D-HSQC NMR analy-
Scheme 7. Metal triflate-catalyzed depolymerization of walnut methanosolv lignin (representive structure shown) in the presence of ethylene glycol to yield
acetal phenols P1–P3.
&
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Metal triflate-catalyzed depolymerization of walnut methano-
solv lignin in combination with in situ acetal formation: Walnut
methanosolv lignin (50 mg) isolated by a reported procedure[7c]
was placed in a 20 mL microwave vial equipped with a magnetic
stirring bar. Solvent (1,4-dioxane, 1 mL), internal standard (n-octa-
decane, 10 mL from a 0.25m stock in 1,4-dioxane, 2.5 mmol) and
ethylene glycol (145 mL from a 7.1m stock in 1,4-dioxane) were
added. The catalyst M(OTf)x (23 mmol) was added as a solid (HOTf
was added from a 0.23m stock solution in 1,4-dioxane), and the
vial was sealed. The reaction was stirred at 1408C for 15 min
before being cooled rapidly in an ice bath. The mixture was filtered
over a plug of Celite and the flask and filter washed with about
0.5 mL 1,4-dioxane in three portions. The combined filtrate was
evaporated to dryness over 16 h at 408C in a Univapo 150 ECH ro-
tational vacuum concentrator. The residue was suspended in
150 mL DCM by extensive mixing (by vortex) after which 1.35 mL
toluene was added. The samples were vortexed and subsequently
centrifuged for 10 min at 13400 rpm using an Eppendorf minispin
tabletop centrifuge. The light organic liquid and solid or thick oily
residue were separated. This procedure for suspension/washing
with 10% DCM and 90% toluene was repeated three times after
which both the combined extracted fractions and the residue were
dried for 24 h at 408C in an Univapo 150 ECH rotational vacuum
concentrator (dried weights see Table S7). The oil containing the
low molecular weight components was dissolved in DCM and ana-
lyzed by GC–FID for quantification of P1–P3 (Table S8).
similar selectivities. In the depolymerization of organosolv
lignin, Bi(OTf)3, Fe(OTf)3 and Hf(OTf)4 all showed promising re-
sults and three main aromatic products were clearly identified
as major products. Interestingly, the best aromatic monomer
yields 19.3Æ3.2 wt% were obtained with Fe(OTf)3, reflecting
differences in reactivity in this case in favor of the metal triflate
compared to triflic acid. More specific reasons for this behavior
are currently investigated in our laboratories. Future studies
should also address the possibility of catalyst recycling[20a]
either by immobilization of the triflate salts[20b] or HOTf[20c]
.
Experimental Section
Metal triflate-catalyzed cleavage of b-O-4 model compounds:
Substrate (e.g., 1, 48.9 mg, 0.2 mmol) was weighed in a 20 mL mi-
crowave vial equipped with a magnetic stirring bar. Solvent (e.g.,
1,4-dioxane, 2 mL) and n-octadecane (25 mmol from a 0.25m stock
solution in the appropriate solvent) were added, and the vial was
sealed. The solution was stirred and heated to the appropriate
temperature and the catalyst (e.g., triflic acid, 10 mol%, 1 mL,
0.02 mmol or 200 mL of a freshly prepared 5 mgmLÀ1 Fe(OTf)3
stock in 1,4-dioxane, 10 mol%, 0.02 mmol) was added by a syringe
with a thin needle through the septum of the microwave vial. If
samples were taken, this was done by using a syringe equipped
with a long thin needle. The samples (100–150 mL) were filtered, di-
luted in dichloromethane (DCM) and analyzed by GC–FID and GC–
MS (Figure 2). Otherwise the reaction was stopped by cooling on
ice. The crude reaction mixture was filtered through Celite and an
aliquot was taken for GC–FID and GC–MS analysis (Table S2).
Acknowledgements
This work was funded by the European Union (Marie Curie ITN
‘SuBiCat’ PITN-GA-2013-607044, PJD, CWL, NJW, PCKL, KB, JGdeV)
as well as EP/J018139/1, EP/K00445X/1 grants (NJW and PCJK)
and an EPSRC Doctoral Prize Fellowship (CSL). Additionally, we
would like to acknowledge Dr. A. Narani for his contribution to
lignin extraction and T. Yan and M. Scott for their contributions
to the synthesis of model compounds 1, 1b, and 1c as well as
Dr. Fanny Tran for the kind gift of compound 10.
Metal triflate-catalyzed cleavage of b-O-4 model compounds in
combination with in situ decarbonylation: Inside a glovebox
a 20 mL microwave vial was charged with substrate (e.g., 1,
12.2 mg, 0.05 mmol) and n-octadecane (6.25 mmol) from a stock so-
lution in 1,4-dioxane. A premixed solution of PPh3 and [IrCl(cod)]2
in 1,4-dioxane (mixed for 15 min prior to addition) was added to
this mixture and the vial was sealed. The vial was stirred and
heated to the appropriate temperature and catalyst from a stock
solution in 1,4-dioxane was added by a syringe with a thin needle
through the septum of the microwave vial. Upon completion, the
reaction mixtures were cooled on ice and filtered through Celite.
Aliquots of the reaction mixtures were diluted in DCM and ana-
lyzed by GC–FID and GC–MS (Results in Tables S3–S5).
Keywords: acidolysis · aromatics · depolymerization · lignin ·
metal triflates
[1] a) C. O. Tuck, E. Pꢁrez, I. T. Horvꢂth, R. A. Sheldon, M. Poliakoff, Science
2012, 337, 695–699; b) P. N. R. Vennestrøm, C. M. Osmundsen, C. H.
Christensen, E. Taarning, Angew. Chem. Int. Ed. 2011, 50, 10502–10509;
Angew. Chem. 2011, 123, 10686–10694.
[2] a) M. Besson, P. Gallezot, C. Pinel, Chem. Rev. 2014, 114, 1827–1879;
b) P. J. Deuss, K. Barta, J. G. de Vries, Catal. Sci. Technol. 2014, 4, 1174–
1196; c) A. Ragauskas, G. T. Beckham, M. J. Biddy, R. Chandra, F. Chen,
M. F. Davis, B. H. Davidson, R. A. Dixon, P. Gilna, M. Keller, P. Langan,
A. K. Naskar, J. N. Saddler, T. J. Tschaplinski, G. A. Tuskan, C. E. Wyman,
Science 2014, 344, 1246843.
[3] For selected recent examples see: a) A. Rahimi, A. Ulbrich, J. J. Coon,
S. S. Stahl, Nature 2014, 515, 249–252; b) E. Feghali, G. Carrot, P. Thuꢁry,
C. Genre, T. Cantat, Energy Environ. Sci. 2015, 8, 2734–2743; c) I. Klein,
Ch. Marcum, H. Kenttamaa, M. M. Abu-Omar, Green Chem. 2016, 18,
2399–2405; d) S. Van den Bosch, W. Schutyser, R. Vanholme, T. Driessen,
S.-F. Koelewijn, T. Renders, B. De Meester, W. J. J. Huijgen, W. Dehaen,
C. M. Courtin, B. Lagrain, W. Boerjan, B. F. Sels, Energy Environ. Sci. 2015,
8, 1748–1763; e) T. vom Stein, T. den Hartog, J. Buendia, S. Stoychev, J.
Mottweiler, C. Bolm, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed.
2015, 54, 5859–5863; Angew. Chem. 2015, 127, 5957–5961; f) S. Dabral,
J. Mottweiler, T. Rinesch, C. Bolm, Green Chem. 2015, 17, 4908–4912.
Metal triflate catalyzed cleavage of b-O-4 model compounds in
combination with in situ acetal formation: Substrate (e.g., 1,
48.9 mg, 0.2 mmol) was weighed in a 20 mL microwave vial
equipped with a stirring bar. Solvent (e.g., 1,4-dioxane, 2 mL) and
n-octadecane (25 mmol from a 0.25m stock solution in the appro-
priate solvent), diol (e.g., ethylene glycol 16 mL 0.3 mmol) were
added, and the vial was sealed. The solution was stirred and
heated to the appropriate temperature and catalyst (e.g., triflic
acid, 10 mol%, 1 mL, 0.02 mmol or 200 mL of a freshly prepared
5 mgmLÀ1 Fe(OTf)3 stock in 1,4-dioxane, 10 mol%, 0.02 mmol) was
added by a syringe with a thin needle through the septum of the
microwave vial. If samples were taken, this was done by using sy-
ringe equipped with a long thin needle. The samples (100–150 mL)
were filtered, diluted in DCM, and analyzed by GC–FID and GC–MS
(see Figures 3 and S2–S4). Otherwise the reaction was stopped by
cooling on ice. The crude reaction mixture was filtered over Celite
and an aliquot was taken for GC–FID and GC–MS analysis (Results
in Table S6).
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[4] See recent reviews and references therein: a) C. Xu, R. A. D. Arancon, J.
Labidi, R. Luque, Chem. Soc. Rev. 2014, 43, 7485–7500; b) P. J. Deuss, K.
Barta, Coord. Chem. Rev. 2016, 306, 510–532; c) S. K. Hanson, R. T. Baker,
Acc. Chem. Res. 2015, 48, 2037–2048; d) R. Rinaldi, R. Jastrzebski, M. T.
Clough, J. Ralph, M. Kennema, P. C. A. Bruijnincx, B. Weckhuysen, Angew.
Chem. Int. Ed. 2016, 55, 8164–8215; Angew. Chem. 2016, 128, 8296–
8354.
[5] a) T. Yokoyama, J. Wood Chem. Technol. 2014, 35, 27–42; b) M. R. Stur-
geon, S. Kim, K. Lawrence, R. S. Paton, S. C. Chmely, M. Nimlos, T. D.
Foust, G. T. Beckham, ACS Sustainable Chem. Eng. 2014, 2, 472–485;
c) K. Lundquist, Appl. Polym. Symp. 1976, 28, 1393–1407.
[6] For selected references on early work on acidolysis related to organo-
solv extraction see: a) E. West, A. S. MacInnes, H. Hibbert, J. Am. Chem.
Soc. 1943, 65, 1187–1192; b) J. M. Pepper, P. E. T. Baylis, E. Adler, Can. J.
Chem. 1959, 37, 1241–1248; c) K. Lundquist, Acta Chem. Scand. 1973,
27, 2597–2606; recent selected references on organosolv lignin isola-
tion: d) A. Guerra, I. Filpponen, L. A. Lucia, D. S. Argyropoulos, J. Agric.
Food Chem. 2006, 54, 9696–9705; e) S. Bauer, H. Sorek, V. D. Mitchell,
A. B. IbꢂÇez, D. E. Wemmer, J. Agric. Food Chem. 2012, 60, 8203–8212;
f) R. El Hage, N. Brosse, P. Sannigrahi, A. Ragauskas, Polym. Degrad. Stab.
2010, 95, 997–1003; g) N. Brosse, P. Sannigrahi, A. Ragauskas, Ind. Eng.
Chem. Res. 2009, 48, 8328–8334.
[7] a) P. J. Deuss, M. Scott, F. Tran, N. J. Westwood, J. G. de Vries, K. Barta, J.
Am. Chem. Soc. 2015, 137, 7456–7467; b) M. Scott, P. J. Deuss, J. G. de
Vries, M. H. G. Prechtl, K. Barta, Catal. Sci. Technol. 2016, 6, 1882–1891;
c) A. Kaiho, M. Kogo, R. Sakai, K. Saito, T. Watanabe, Green Chem. 2015,
17, 2780–2783; d) C. W. Lahive, P. J. Deuss, C. S. Lancefield, Z. Sun, D. B.
Cordes, C. M. Young, F. Tran, A. M. Z. Slawin, J. G. de Vries, P. C. J. Kamer,
N. J. Westwood, K. Barta, J. Am. Chem. Soc. 2016, 138, 8900–8911.
[8] For the metal triflate salts used in this study the skin corrosive category
is set to 1b for Cu(OTf)2 and Al(OTf)3 and lower for others compared to
1a for HOTf (ECHA categories). Category 1a indicates acute corrosion.
Category 1b indicates corrosion after >3 min of skin contact.
[9] a) O. Karlsson, K. Lundquist, S. Meuller, K. Westlid, Acta Chem. Scand. B
1988, 42, 48–51; b) T. Yokoyama, Y. Matsumoto, J. Wood Chem. Technol.
2010, 30, 269–282.
Kobayashi, Eur. J. Org. Chem. 2015, 5485–5499; f) K. Mori, K. Kurihara, T.
Akiyama, Chem. Commun. 2014, 50, 3729–3731; g) B. Wang, Y. Chen, J.
Wang, C.-H. Tung, Z. Xu, J. Org. Chem. 2015, 80, 12718–12724; h) B.
Cacciuttolo, S. Poulain-Martini, E. DuÇach, Eur. J. Org. Chem. 2011,
3710–3714.
[12] More on this subject: a) S. Kobayashi, S. Nagayama, T. Busujima, J. Am.
Chem. Soc. 1998, 120, 8287–8288; b) N. Legrave, A. Couhert, S. Olivero,
J.-R. Desmurs, E. DuÇach, Eur. J. Org. Chem. 2012, 901–904; c) T. T.
Dang, F. Boeck, L. Hintermann, J. Org. Chem. 2011, 76, 9353–9361.
[13] For an example where the addition of a non-nucleophilic base shuts
down metal-triflate activity proving in situ acid formation, see: a) T. C.
Wabnitz, J.-Q. Yu, J. B. Spencer, Chem. Eur. J. 2004, 10, 484–493; for an
example where a non-nucleophilic base does not shut down metal-tri-
flate activity hence providing evidence for Lewis acid-catalyzed reac-
tions, see: b) S. Kundal, S. Jalal, K. Paul, U. Jana, Eur. J. Org. Chem. 2015,
5513–5517.
[14] Jastrzebski, PhD Thesis, University of Utrecht (NL), 2016. This thesis was
published online while we were preparing this manuscript. Here, a simi-
lar catalytic cleavage was reported using Sc(OTf)3 and several other
water-tolerant Lewis acid metal triflate salts. Higher reaction tempera-
tures (175–2008C) were required for cleavage and [Rh(cod]Cl]2/dppp[21]
(dppp=1,3-bis(diphenylphosphino)propane) was used as decarbonyla-
tion catalyst.
[15] T. Iwai, T. Fujihara, Y. Tsuji, Chem. Commun. 2008, 6215–6217.
[16] L. Mitchell, H. Hibbert, J. Am. Chem. Soc. 1944, 66, 602–694.
[17] S. Li, K. Lundquist, Holzforschung 1999, 53, 39–43.
[18] a) K. Barta, G. R. Warner, E. S. Beach, P. T. Anastas, Green Chem. 2014, 16,
191–196; b) C. S. Lancefield, O. S. Ojo, F. Tran, N. J. Westwood, Angew.
Chem. Int. Ed. 2015, 54, 258–262; Angew. Chem. 2015, 127, 260–264.
[19] a) C. Li, X. Zhao, A. Wang, G. W. Huber, T. Zhang, Chem. Rev. 2015, 115,
11559–11624; b) I. Delidovich, P. J. C. Hausoul, R. Pfꢃtzenreuter, M. Rose,
R. Palkovits, Chem. Rev. 2016, 116, 1540–1599.
[20] a) E. Wiebus, B. Cornils: Supported Aqueous Phase Catalysis and P. Was-
serscheid and M. Haumann: Supported Ionic Liquid Phase Catalysis in Cat-
alyst Separation, Recovery and Recycling (Eds.: D. Cole-Hamilton, R.
Tooze), Springer, Netherlands, Dordrecht, 2006, vol. 30, chap. 5 and
chap. 7; b) S. Kobayashi, S. Nagayama, J. Am. Chem. Soc. 1998, 120,
2985–2986; c) P. N. Liu, F. Xia, Q. W. Wang, Y. J. Ren, J. Q. Chen, Green
Chem. 2010, 12, 1049–1055.
[10] a) K. S. Surprenant, Dioxane, Ullmann’s Encyclopedia of Industrial Chemis-
try, Electronic Release, Wiley-VCH Weinheim, 2000; b) J. M. McIntosh,
Sulfur Trioxide-1,4-Dioxane. e-EROS Encyclopedia of Reagents for Organic
Synthesis, Wiley, 2001.
[11] For examples see: a) S. Kobayashi, M. Sugiura, H. Kitagawa, W. W.-L.
Lam, Chem. Rev. 2002, 102, 2227–2302; b) R. Akiyama, S. Kobayashi,
Chem. Rev. 2009, 109, 594–642; c) T. Ollevier, Org. Biomol. Chem. 2013,
11, 2740–2755; d) R. Ghosh, S. Maiti, J. Mol. Catal. A 2007, 264, 1–8; se-
lected recent examples: e) H. Ishitani, H. Suzuki, Y. Saito, Y. Yamashita, S.
[21] M. Kreis, A. Palmelund, L. Bunch, R. Madsen, Adv. Synth. Catal. 2006,
348, 2148–2154.
Received: June 21, 2016
Published online on && &&, 0000
&
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
P. J. Deuss, C. W. Lahive, C. S. Lancefield,
N. J. Westwood, P. C. J. Kamer, K. Barta,*
J. G. de Vries*
&& – &&
Metal Triflates for the Production of
Aromatics from Lignin
Nutty triflates: Metal triflates are intro-
duced as an easy-to-handle alternative
to triflic acid for the cleavage of lignin
b-O-4 linkages in conjunction with stabi-
lization of reactive intermediates. The
reactivity of several model compounds
and lignin is studied. In particular,
iron(III) triflate proved effective, provid-
ing as much as 19.3Æ3.2 wt% yield of
a set of three aromatic C2-acetals from
lignin.
ChemSusChem 2016, 9, 1 – 9
www.chemsuschem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ