Separable fluorous ionic liquids for the dissolution and saccharification of cellulose

Article abstract of DOI:10.1039/c1gc15776k

Ionic liquids are an attractive class of solvents for biomass conversion processes. The same properties that make them advantageous - high polarity, water solubility and negligible vapor pressure - hinder their recovery from carbohydrates. We report on th

Full text of DOI:10.1039/c1gc15776k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 2719
www.rsc.org/greenchem
COMMUNICATION
Separable fluorous ionic liquids for the dissolution and saccharification of
cellulose†
Benjamin R. Caes,a Joseph B. Binder,‡a Jacqueline J. Blankb and Ronald T. Raines*a,b
Received 30th June 2011, Accepted 9th August 2011
DOI: 10.1039/c1gc15776k
Ionic liquids are an attractive class of solvents for biomass
conversion processes. The same properties that make them
advantageous—high polarity, water solubility and negli-
gible vapor pressure—hinder their recovery from carbo-
hydrates. We report on the synthesis of seven fluorous
imidazolium chloride ionic liquids and on their ability to
dissolve cellulose. One of these ionic liquids, 3-methyl-1-
(2¢,2¢,3¢,3¢,3¢-pentafluoropropyl)-imidazolium chloride, dis-
solves cellulose. We found this fluorous ionic liquid to be
suitable for use in cellulose hydrolysis reactions, and we
achieved its recovery from glucose.
these intermolecular hydrogen bonds. Ionic liquids accom-
plish dissolution because their charged species disrupt the
hydrogen bonding network by forming electron donor–electron
acceptor complexes with the hydroxyl groups.3c,4 This results
in separation of the polymer chains, allowing dissolution.
Dialkylimidazolium chlorides, in particular, demonstrate the
capacity to dissolve high concentrations of cellulose,5 and have
been used in cellulose conversion processes.6 For example,
we accessed HMF in one step from lignocellulosic biomass
using a chromium catalyst in 1-ethyl-3-methylimidazolium
chloride ([EMIM]Cl)6c and fermentable sugars from ligno-
cellulosic biomass by hydrolysis in an [EMIM]Cl/HCl/H2O
system.6e Schu¨th and co-workers utilized solid acid catalysts
in 1-butyl-3-methylimidazolium chloride ([BMIM]Cl) for the
depolymerization of cellulose to cello-oligomers.6b Amarasekara
and Owereh accomplished cellulose hydrolysis to glucose and
other reducing sugars in the Brønsted acidic ionic liquids
1-(1-propylsulfonic)-3-methylimidazolium chloride and 1-(1-
butylsulfonic)-3-methylimidazolium chloride.6d We realized that
the physical properties that make ionic liquids ideal as a reaction
medium for these biomass conversion processes hinder their
separation and recovery from the desired carbohydrate products.
Yet, as ionic liquids are more expensive than traditional organic
solvents, it is imperative that they be recovered if they are to
serve as reaction media in economical processes.
Introduction
The implementation of a viable fuel and energy alternative
to sources such as coal, natural gas and petroleum is rapidly
becoming critical for a sustainable future. Lignocellulosic
biomass, both abundant and renewable, is one such alternative
resource. Its primary component, cellulose, can be hydrolyzed
to glucose, then transformed into the platform chemical 5-
(hydroxymethyl)furfural (HMF)1 or used as a feedstock for the
growth of engineered microbes. Unfortunately, the polymeric,
crystalline nature of cellulose makes it highly recalcitrant to
dissolution, and few organic solvents accomplish this feat.2 The
desire to utilize environmentally benign solvents for cellulose
dissolution aggravates the problem further.
Ionic liquids are an attractive class of solvents as they
are polar, water-soluble molten salts with negligible vapor
pressure.3 Importantly, ionic liquids can accomplish cellu-
lose dissolution. The crystallinity of cellulose results from
a complex network of intrastrand and interstrand hydrogen
bonds.2 To dissolve cellulose, a solvent must compete for
aDepartment of Chemistry, University of Wisconsin–Madison, 1101
University Avenue, Madison, Wisconsin, 53706-1322, USA
bDepartment of Biochemistry, University of Wisconsin–Madison, 433
Babcock Drive, Madison, Wisconsin, 53706-1544, USA.
E-mail: rtraines@wisc.edu; Fax: +1 608 890 2583; Tel: +1 608 262 8588
† Electronic supplementary information (ESI) available: Experimental
procedures and 1H NMR spectra of novel ionic liquids. See DOI:
10.1039/c1gc15776k
‡ Present address: Energy Biosciences Institute, University of California,
Berkeley, California, 94720, USA.
Fluorous tags are being increasingly employed to facilitate the
separation of products and catalysts in organic synthesis.7 They
have a high affinity for fluorous phases, which are distinct from
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2719–2722 | 2719
©

View Article Online
polar/non-polar and hydrophilic/hydrophobic phases.7c Heavy
fluorous tags containing >60% fluorine enable the extraction
of a tagged compound.7e Light fluorous tags (typically C3F7 to
C10F21) are suited for solid-phase extraction with a C8F17-bonded
phase (as in fluorous silica gel) or isolation by fluorous HPLC.7e
We envisioned that these same separation strategies could be
used with fluorous ionic liquids.
Fluorous ionic liquids are not available from commercial
vendors. A variety, however, have been synthesized, not only
to avail potential fluorine–fluorine interactions, but also to
modulate the physical properties of the ionic liquids.8 Mod-
ification of the fluorine-containing portion of the cation can
have a substantial impact on melting point, viscosity, density,
conductivity, solubility and thermostability.8d Additionally, the
identity of the anion can be modified to alter the properties
of an ionic liquid. The versatility is great, and a number of
imidazolium fluorous ionic liquids, in particular, have been
synthesized and tested as surfactants in gas separations and in
metal-ion extraction.8a–c,e–g,9 Nonetheless, fluorous ionic liquids
have not been employed in biomass- or bioproduct-conversion
processes, and no fluorous ionic liquids have been synthesized
with the chloride counterion necessary for cellulose dissolution.
Herein, we report on the synthesis of seven fluorous imidazolium
chloride ionic liquids (1–7) and on their capabilities as biomass
solvents. To our knowledge, ionic liquid 3 is novel. The other
six fluorous imidazolium groups have been described previously
(though not as chloride salts)8e,9,10 but have not been used in a
bioprocessing application.
Scheme 2
We again found the ionic liquid to be soluble in polar solvents
but unable to dissolve cellulose.
Then, we sought to consolidate the fluoro groups so as
to minimize alteration to the dialkylimidazolium core struc-
ture. Hence, we synthesized 3-methyl-1-(3¢,4¢,4¢,4¢-tetrafluoro-
3¢-trifluoromethyl-butyl)-imidazolium chloride (3) as a viscous
oil by following the procedure of Noble and co-workers.8e Once
again, while the ionic liquid was soluble in polar solvents, it was
unable to dissolve cellulose.
The behavior of ionic liquids 1–3 was unexpected, as ionic
liquids containing a chloride anion are typically able to dissolve
at least low concentrations of cellulose. We reasoned that the
size of the perfluorinated alkyl side chain on ionic liquids 1–3
could be hindering cellulose dissolution. A steric effect had been
observed previously, as imidazolium ionic liquids with N-alkyl
groups containing >8 carbons demonstrate a marked decrease
in cellulose solubility.5a,5b Perfluorinated alkyl groups are signif-
icantly more bulky than the corresponding alkyl groups, which
would explain why our fluorous ionic liquids were unable to
dissolve cellulose, unlike their corresponding alkyl ionic liquids.
For example, A-values indicate that a trifluoromethyl group is at
least as large as an isopropyl group; Taft-type steric parameters
indicate it to be as large as an isobutyl group.11 Therefore, we
sought to synthesize fluorous ionic liquids with smaller fluorous
side chains (Table 1).
Results and discussion
As its separation from non-fluorous molecules would be
enhanced by
a
high fluorine-content,7g we initially tar-
geted an ionic liquid with nine fluoro groups, 3-methyl-1-
(3¢,3¢,4¢,4¢,5¢,5¢,6¢,6¢,6¢-nonafluorohexyl)-imidazolium chloride
(1). Following the procedure of Noble and co-workers (Scheme
1),8e ionic liquid 1 was synthesized by the reaction of a fluo-
roalkyl iodide precursor with 1-methylimidazole. Ion-exchange
chromatography yielded ionic liquid 1 as a highly viscous oil.
This ionic liquid was soluble in polar solvents such as water,
methanol and acetonitrile, but demonstrated limited solubility
in toluene. We found cellulose to be insoluble in ionic liquid 1,
even with heating.
To access an ionic liquid with
a smaller fluorous
tag, we synthesized 3-methyl-1-(3¢,3¢,4¢,4¢,4¢-pentafluorobutyl)-
imidazolium chloride (4) as a viscous oil by following the proce-
dure of Noble and co-workers.8e This ionic liquid was unable to
dissolve cellulose, despite its solubility in polar solvents. Seeking
to truncate further the side chain of the cation, we synthe-
sized 3-methyl-1-(2¢,2¢,3¢,3¢,3¢-pentafluoropropyl)-imidazolium
chloride (5) by following the procedure of DesMarteau and
Table 1 Imidazolium volumes of fluorous ionic liquids 1–7
Fluorous ionic
liquid
Number of fluoro
groups
Imidazolium volume a
(Bohr3 mol-1)
Scheme 1
1
2
3
4
5
6
7
9
7
7
5
5
3
5
1835
1669
1352
1303
1213
1089
1652
Next, we synthesized 1-(2¢,2¢,3¢,3¢,4¢,4¢,4¢-heptafluorobutyl)-
3-methylimidazolium chloride (2), a fluorous ionic liquid con-
taining a smaller fluorous tag. The steric bulk of the proximal
fluoroalkyl group made the 1-position of heptafluorobutyl
iodide especially resistant to nucleophilic attack. Hence, a
hypervalent iodonium alkylating agent invented by DesMarteau
and co-workers was employed as an intermediate (Scheme 2).9,10b
a Calculated using Gaussian ’03, see the ESI.†
2720 | Green Chem., 2011, 13, 2719–2722
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
R
co-workers.9,10b This ionic liquid was obtained as a crystalline
solid at room temperature but melts at moderate temperatures
and is a viscous liquid at 100 ◦C. Importantly, upon heating to
140 ◦C, ionic liquid 5 demonstrated the dissolution of cellulose
to ≥2 wt%, forming a viscous solution similar to that observed
with typical dialkylimidazolium ionic liquids. Remarkably, this
difference in cellulose solubility was accomplished upon deletion
of a single methylene unit. This result demonstrates the profound
influence of imidazolium size on solubilizing cellulose. Further-
more, as dialkylimidazolium acetates are also known to dissolve
cellulose,5b,5c we transformed 5 into its acetate counterpart.
Again, a viscous solution formed upon heating and cellulose
dissolution was observed to ≥2 wt%.
fluorophilic phase on a FluoroFlashꢀ SPE cartridge. Although
the majority of ionic liquid 5 was retained on the cartridge after
water elution, we could detect the ionic liquid in the water eluent.
A methanol elution recovered the remaining ionic liquid. These
results suggested that the fluorous ionic liquid with only five
fluoro groups is too polar and insufficiently fluorous for standard
fluorous separation techniques. Fluorous SPE typically requires
molecules to contain at least seven fluoro groups,7e whereas our
ionic liquid contained only five.
The moderate retention of the fluorous ionic liquid on the
fluorous SPE cartridge led us to hypothesize that a chromato-
graphic separation might be possible. To this end, we prepared
a column using FluoroFlashꢀ silica gel, as in the SPE cartridge.
R
Encouraged by this success, we sought to decrease further
the size of the fluoroalkyl side chain. Hence, we again fol-
lowed the procedure of DesMarteau and co-workers9,10b to
synthesize 3-methyl-1-(2¢,2¢,2¢-trifluoroethyl)-imidazolium chlo-
ride (6) as a viscous oil. Cellulose was soluble in this ionic
liquid but somewhat less so than in 3-methyl-1-(2¢,2¢,3¢,3¢,3¢-
pentafluoropropyl)-imidazolium chloride (5). Apparently, the
imidazolium structure has subtle and (currently) inexplicable
influences on interactions with cellulose.
As a final possibility, we were interested in testing a perflu-
orinated aryl–alkyl imidazolium ionic liquid as a solvent for
cellulose. Aryl–alkyl ionic liquids allow for the potential to
modulate the physiochemical properties of an ionic liquid via s-
and p-based electronic effects.12 To this end, we synthesized 3-
methyl-1-pentafluorophenylmethylimidazolium chloride (7) as
a white solid by following the procedure of McGrandle and
Saunders (Scheme 3).10a Ionic liquid 7 demonstrated solubility
in polar solvents but was insoluble in toluene and had a melting
point of 150 ◦C. Due to its high melting point, we attempted
cellulose dissolution utilizing varying concentrations of the
ionic liquid in N,N-dimethylacetamide–LiCl.6c Yet, even when
heated to 140 ◦C, no cellulose dissolution was observed in these
solutions.
Using this column, we were able to separate ionic liquid 5 from
glucose using water as the eluent (Fig. 1). To conclude that
the fluorous tag was responsible for the separation of the ionic
liquid from the sugar, we performed a similar experiment using
pentafluorobutyl ionic liquid 4 and its non-fluorous counterpart
1-butyl-3-methylimidazolium chloride. The fluorous ionic liquid
again demonstrated a clean separation from glucose, while the
non-fluorous ionic liquid was not cleanly separated from the
glucose. This result is consistent with the fluorous tag enabling
the separation of the ionic liquid from glucose.
Scheme 3
Given the demonstrated ability of ionic liquid 5 to dissolve
cellulose, we next sought a method for its extraction. Ionic liquid
5 is highly soluble in water, and we did not recover an appreciable
quantity in an organic phase (e.g., ethyl acetate). Nonetheless,
we investigated whether a fluorous solvent (e.g., perfluorinated
hexane) could extract ionic liquid 5. As typical fluorous tags
used in this way contain >20 fluoro groups, we did not expect
our ionic liquids to separate into the fluorous phase. Indeed,
when tested for separation in water and perfluorinated hexane,
no ionic liquid 5 was detected in the fluorous solvent. Next,
we turned to fluorous solid-phase extraction (SPE) to remove
our lightly fluorinated ionic liquid from water. Following a
procedure from Fluorous Technologies (Pittsburgh, PA),13 we
used water as our fluorophobic phase and methanol as our
Fig. 1 Schematic of the isolation of ionic liquid 5 from sugars and
furanics by chromatography using FluoroFlashꢀR silica gel. The eluent
is water.
With the successful separation of the fluorous ionic liquids
from glucose, we sought to determine if ionic liquid 5 could be
used in a cellulose hydrolysis reaction. By following a modified
procedure that we have employed previously,6c we dissolved
cellulose at 2 wt% in ionic liquid 5 and assessed the reaction
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2719–2722 | 2721
©

View Article Online
mixture at known time points. After 2 h, we detected the presence
of both fructose and glucose in the reaction mixture at 27 and
11 mol%, respectively. After 4 h, we detected traces of HMF. Fi-
nally, we achieved separation of ionic liquid 5 from the reaction
(c) J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem., Int.
Ed., 2007, 46, 7164–7183.
2 A. Pinkert, K. N. Marsh and S. S. Pang, Ind. Eng. Chem. Res., 2010,
49, 11121–11130.
3 (a) N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37,
123–150; (b) M. Freemantle, An Introduction to Ionic Liquids, RSC
Publishing, 2009; (c) A. Pinkert, K. N. Marsh, S. S. Pang and M. P.
Staiger, Chem. Rev., 2009, 109, 6712–6728; (d) M. Petkovic, K. R.
Seddon, L. P. N. Rebelo and C. S. Pereira, Chem. Soc. Rev., 2010, 40,
1383–1403.
R
products by chromatography using FluoroFlashꢀ silica gel (Fig.
1), recovering >95% of 5. This result suggests that 3-methyl-
1-(2¢,2¢,3¢,3¢,3¢-pentafluoropropyl)-imidazolium chloride (5) can
indeed be used in cellulose conversion.
4 R. C. Remsing, R. P. Swatloski, R. D. Rogers and G. Moyna, Chem.
Commun., 2006, 1271–1273.
Conclusions
5 (a) R. P. Swatloski, S. K. Spear, J. D. Holbrey and R. D. Rogers, J.
Am. Chem. Soc., 2002, 124, 4974–4975; (b) J. Vitz, T. Erdmenger, C.
Haensch and U. S. Schubert, Green Chem., 2009, 11, 417; (c) M. E.
Zakrzewska, E. Bogel-Łukasik and R. Bogel-Łukasik, Energy Fuels,
2010, 24, 737–745.
We have synthesized seven fluorous imidazolium chloride ionic
liquids and assessed their ability to dissolve cellulose. We discov-
ered that the ability of a fluorous ionic liquid to dissolve cellulose
is highly sensitive to the structure of the imidazolium cation,
as only by decreasing the fluorous tag to a pentafluoropropyl
group was cellulose dissolution possible. An even smaller tag,
trifluoroethyl, also allowed some cellulose dissolution to occur.
Although the content of fluorine in these fluorous tags is too
small to allow for typical fluorous SPE methods, we accom-
plished the separation of cellulose hydrolysis products from our
optimal fluorous ionic liquid by utilizing chromatography with
ˇ
6 (a) J. Horvat, B. Klaic´, B. Metelko and V. Sunjic´, Croat. Chem. Acta,
1986, 59, 429–438; (b) R. Rinaldi, R. Palkovits and F. Schu¨th, Angew.
Chem., Int. Ed., 2008, 47, 8047–8050; (c) J. B. Binder and R. T. Raines,
J. Am. Chem. Soc., 2009, 131, 1979–1985; (d) A. S. Amarasekara and
O. S. Owereh, Ind. Eng. Chem. Res., 2009, 48, 10152–10155; (e) J. B.
Binder and R. T. Raines, Proc. Natl. Acad. Sci. U. S. A., 2010, 107,
4516–4521.
7 (a) D. P. Curran, Angew. Chem., Int. Ed., 1998, 37, 1174–1196; (b) N.
Spetseris, S. Hadida, D. P. Curran and T. Y. Meyer, Organometallics,
1998, 17, 1458–1459; (c) D. P. Curran and Z. Y. Lee, Green
Chem., 2001, 3, G3–G7; (d) J. van den Broeke, F. Winter, B.
J. Deelman and G. van Koten, Org. Lett., 2002, 4, 3851–3854;
(e) W. Zhang, Tetrahedron, 2003, 59, 4475–4489; (f) M. S. Yu,
D. P. Curran and T. Nagashima, Org. Lett., 2005, 7, 3677–3680;
(g) W. Zhang and D. P. Curran, Tetrahedron, 2006, 62, 11837–
11865; (h) N. L. Pohl, Angew. Chem., Int. Ed., 2008, 47, 3868–
3870.
R
FluoroFlashꢀ silica gel. Finally, we tested the pentafluoropropyl
ionic liquid in a cellulose hydrolysis reaction and discovered
its support of monomeric sugar production. These results
encourage the further development of fluorous imidazolium
chloride ionic liquids for biomass conversion processes.
8 (a) T. L. Merrigan, E. D. Bates, S. C. Dorman and J. H. Davis, Chem.
Commun., 2000, 2051–2052; (b) V. V. Rudyuk, D. V. Fedyuk and L.
M. Yagupolskii, J. Fluorine Chem., 2004, 125, 1465–1471; (c) Y. L.
Yagupolskii, T. M. Sokolenko, K. I. Petko and L. M. Yagupolskii, J.
Fluorine Chem., 2005, 126, 667–670; (d) H. Xue and J. M. Shreeve,
Eur. J. Inorg. Chem., 2005, 2573–2580; (e) J. E. Bara, C. J. Gabriel, T.
K. Carlisle, D. E. Camper, A. Finotello, D. L. Gin and R. D. Noble,
Chem. Eng. J., 2009, 147, 43–50; (f) O. Kysilka, M. Ryba´cˇkova´, M.
Skalicky´, M. Kv´ıcˇalova´, J. Cvacˇka and J. Kv´ıcˇala, J. Fluorine Chem.,
2009, 130, 629–639; (g) X. Li, D. W. Bruce and J. M. Shreeve, J.
Mater. Chem., 2009, 19, 8232–8238.
Acknowledgements
This work was supported by the Great Lakes Bioenergy Re-
search Center, a DOE Bioenergy Research Center. We are grate-
ful to A. Choudhary for his assistance with Gaussian calcula-
tions and R. L. Kubiak for her assistance with cellulose dissolu-
tion studies. This study made use of the National Magnetic Res-
onance Facility at Madison, which is supported by NIH grants
P41RR02301 (BRTP/NCRR) and P41GM66326 (NIGMS).
Additional equipment was purchased with funds from the
University of Wisconsin, the NIH (RR02781, RR08438), the
NSF (DMB-8415048, OIA-9977486, BIR-9214394) and the
USDA.
9 C. Lu, D. Vanderveer and D. D. DesMarteau, Org. Lett., 2008, 10,
5565–5568.
10 (a) S. McGrandle and G. C. Saunders, J. Fluorine Chem., 2005, 126,
451–455; (b) D. D. DesMarteau and C. Lu, J. Fluorine Chem., 2007,
128, 1326–1334.
11 T. Kitazume and T. Yamazaki, Introduction to Experimental
Methods in Organic Fluorine Chemistry, Kodansha Ltd., Tokyo,
1998.
12 S. Ahrens, A. Peritz and T. Strassner, Angew. Chem., Int. Ed., 2009,
Notes and references
48, 7908–7910.
13 http://fluorous.com/download/FTI_AppNote_F-SPE.pdf
(accessed August 4, 2010).
1 (a) J. Lewkowski, ARKIVOC, 2001, 17–54; (b) Y. Roma´n-Leshkov, C.
J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature, 2007, 447, 982–986;
2722 | Green Chem., 2011, 13, 2719–2722
This journal is
The Royal Society of Chemistry 2011
©