Conversion of Cellulose into Amphiphilic Alkyl Glycosides Catalyzed by Aquivion, a Perfluorosulfonic Acid Polymer

Article abstract of DOI:10.1002/cssc.201700903

The perfluorosulfonic acid (PFSA) Aquivion PW98 is an amphiphilic solid superacid which is shown to catalyze the conversion of cellulose into amphiphilic alkyl glycosides (AAGs) in 85 % yield (with 97 % selectivity). The process involves a mechanocatalytic depolymerization of cellulose followed by a direct glycosylation with n-dodecanol. In comparison to H₂SO₄ and solid acid catalysts commonly employed in cellulose processing, Aquivion PFSA PW98 is not only recyclable but also exhibits superior catalytic performances in terms of yield, selectivity, and reactor productivity.

Full text of DOI:10.1002/cssc.201700903

Accepted Article
Title: Conversion of cellulose to amphiphilic alkyl glycosides catalyzed
by Aquivion, a perfluorosulfonic acid polymer
Authors: François Jérôme, Karine De Oliveira Vigier, Ayman Karam,
Boris Estrine, Sinisa Marincovic, and Claudio Oldani
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201700903
Link to VoR: http://dx.doi.org/10.1002/cssc.201700903
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
Conversion of cellulose to amphiphilic alkyl glycosides catalyzed
by Aquivion, a perfluorosulfonic acid polymer
[
a]
[a]
[b]
[b]
[c]
Ayman Karam, Karine De Oliveira Vigier, Sinisa Marinkovic, Boris Estrine, Claudio Oldani and
François Jérôme*[a]
Dedication ((optional))
®
(poly)glycerol, sucrose, glucose or sorbitan for instance.[4] AAGs
are industrially produced by glycosylation of glucose with an
excess of fatty alcohols in the presence of an acid catalyst,
Abstract: Aquivion PFSA PW98 is an amphiphilic solid superacid
capable of converting cellulose to Amphiphilic Alkyl Glycosides
(
AAG) in 85% yield (i.e. 97% selectivity). The present process
involves (1) mechanocatalytic depolymerization of cellulose
followed by (2)
comparison to H
2 4
typically H SO
or dodecylbenzene sulfonic acid.[5] Generally, a
a
a
direct glycosylation with n-dodecanol. In
direct glycosylation with fatty alcohols is achieved from refined
glucose while from glucose syrup the process involves (1) a
glycosylation with a short chain alcohol (typically n-butanol)
followed by (2) a transglycosylation reaction with fatty alcohols.
2
SO
4
and solid acid catalysts commonly employed
®
in cellulose processing, Aquivion PFSA PW98 is not only recyclable
but also exhibits superior catalytic performances in terms of yield,
selectivity and reactor productivity.
Despite attractive physicochemical properties and
a low
environmental footprint, the price of AAGs is unfortunately still
too high (~ 2 €/kg) to compete with surfactants from a fossil
origin which limits their large deployment in our Society.
Beside process optimization, the production of AAGs from a
cheaper source of glucose such as cellulosic waste (0.1 €/kg vs
Introduction
With a worldwide production of 12 million tons per year and an
annual growth of 4.35%, the market of surfactants has
enormous potential in terms of innovation, product development
and employment in many industrial sectors.[1] This growth is
boosted by the strong demand of developing countries.
Surfactants are nowadays mainly produced from petrochemicals.
Among them, alkylbenzene sulfonates, ethoxylated alcohols,
alkylether sulfates and ethoxylated alkylphenols represent the
most popular families.[2] The necessity to diversify our resources
and the general trend of reducing our environmental impact has
opened opportunities to produce surfactants derived from
renewable feedstocks.[3] The annual market growth of
renewably-sourced surfactants is estimated around 3.5% with a
predicted turnover of 1.8 billion euros in 2018. The industry of
the personal care and the detergence are among the strongest
demander for the supply of “biosurfactants”.
0
.45 €/kg for refined glucose) is clearly an interesting option.
Conversely to hemicellulose from which amphiphilic alkyl
xylosides are catalytically produced in large scale,[6] the
production of AAGs from cellulose remains a challenging task.
Hydrogen bonds, Van der Waals interactions and electronic
effects assuring the cohesion of the cellulose crystal
unfortunately render the glycosidic bonds highly recalcitrant to
chemical reactions.[7]
To date, the glycosylation of cellulose has been reported with
short chain alkyl alcohols (methanol or ethanol) in the presence
of homogeneous (H
heterogeneous (sulphated ZrO
sulfonated carbon, zeolites, etc.) acid catalysts and often at
elevated temperatures.[8] Examples of catalytic glycosylation of
cellulose with long chain alkyl alcohols are however very scarce
due to their low reactivity and immiscibility with sugars rendering
the process even more difficult. In 2010, de Vos and Corma
independently reported the conversion of cellulose to AAGs in
ionic liquids (ILs) containing Amberlyst-15 as a solid acid
catalyst.[9] In 1-butyl-3-imidazolium chloride, cellulose was
2
SO
4
, HCl, H
3
PW12
O40, sulfamic acid, etc.) or
2
,
Nafion, Amberlyst-15,
Amphiphilic alkyl glycosides (AAG) are of particular interest in
the field of surfactants not only because of their intrinsic
physicochemical properties but also because of their low toxicity,
biodegradability and higher stability in water than esters of
soluble which greatly enhanced its reactivity. AAGs were
4 8
obtained with more than 80 % mass yield starting from C to C
[
a]
Dr A. Karam, Dr K. De Oliveira Vigier, Dr F. Jérôme
Institut de Chimie des Milieux et Matériaux de Poitiers,
CNRS/Université de Poitiers, 1 rue Marcel Doré, ENSIP, TSA 41105
alkyl alcohols. Although 99% of the ILs was recycled, few
barriers remain for deployment at a larger scale such as the cost
of 1-butyl-3-imidazolium chloride and the instability of Amberlyst-
15 under the process conditions. In another example, Estrine
reported the possible glycosylation of cellulose with fatty
86073 Poitiers cedex 9 (France).
E-mail: francois.jerome@univ-poitiers.fr
Dr S. Marincovic, Dr B. Estrine
ARD-Agro-industrie Recherches et Développements, Green
Chemistry Department, Route de Bazancourt, F-51110 Pomacle,
France
Dr C. Oldani
[
[
b]
c]
(>80%).[6b] This
alcohols in excess of concentrated H
2
SO
4
process has however shortcomings such as safety,
corrosiveness and costly downstream purification processes.
Very recently, Corma showed it was possible to perform the
glycosylation of cellulose with n-dodecanol in the presence of a
Solvay Speciality Polymers, Viale Lombardia 20, 20021 Bollate MI,
Italy
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
catalytic amount of
H
2
SO
4
at 160-200°C. Yields in alkyl
[10]
glycosides of about 20-35% were claimed. As an alternative
route to the direct glycosylation reaction, Corma demonstrated
that amphiphilic glycosides can be also obtained from cellulose
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
through a tandem methanolysis /transglycosylation reaction with
fatty alcohols in the presence of sulfonated carbons.[11] Inspired
by previous works of Schüth and Rinaldi,[12] we have recently
reported that microcrystalline cellulose could be converted to
butyl glycosides (70% yield) using a mechanocatalytic process
PW98 failed. Inspired by our recent study,[13] MCC was ball-
milled in the presence of Aquivion® PFSA PW98 before the
glycosylation reaction. Typically, MCC (1 g) was mixed with 0.5
®
g of Aquivion PFSA PW98 and the powder was ball-milled for
24 h at 400 rpm in a bowl of 125 mL. Optimization of the MCC/
Aquivion® PFSA PW98 mass ratio, stirring rate and stirring time
are discussed later.
(
grinding of cellulose with H
2
SO
4
).[13] Two drawbacks were
SO which cannot be
however associated to this process (1) H
2
4
recycled and (2) this process was not applicable to alkyl alcohols
with a chain length higher than 4 carbons.
Very recently, we have reported that Aquivion PFSA PW98, a
perfuorinated solid superacid (H
0
= -12)[14] available as coarsely
grained powder, was capable of catalyzing the glycosylation of
glucose and even glucose syrup with fatty alcohols with a higher
.[15] Furthermore,
2 4
selectivity and reactor productivity than H SO
Aquivion® PFSA PW98 chemically and mechanically stands
quite high operating temperature (up to 150-160°C) allowing its
long term recycling.[16] Last but not least, Aquivion® PFSA PW98
is suspected to have amphiphilic properties and is thus well
adapted for catalysis in biphasic conditions,[15, 16d, 17] an important
aspect in the synthesis of AAGs.
Whereas MCC was not soluble in water, after the mix-milling
step with Aquivion® PFSA PW98, 90% of MCC was solubilized
in water. Size Exclusion Chromatography revealed a deep
depolymerization of MCC with the formation of low molecular
weight oligosaccharides with an average degree of
polymerization (DP) of 6 (Figure S1). Despite discrimination
against high mass, an observable DP of up to 20 was observed
by MALDI-TOF suggesting an important polydispersity.
Monocationized pseudomolecular peaks with sequential m/z
increments of 162 uma were observed indicating that glycosyl
units were not damaged during the mechanocatalytic process
(Figure S2). In addition, analysis of the water soluble
disaccharide fraction by HPLC revealed the presence of different
®
Here, we report
a
two-step process involving (1)
a
glycosidic
linkages
suggesting
that
®
mechanocatalytic depolymerization of cellulose with Aquivion
PFSA PW98 followed by (2) a direct glycosylation with n-
dodecanol (Scheme 1). This process affords AAGs in a yield up
to 85% and does not involve the presence of any hazardous
chemicals such as ionic liquids or mineral acids for instance.
Aquivion® PFSA PW98 was found highly stable and was
successfully recycled without loss of its performances.
systematic comparison with H SO was also included to shed
light on the remarkable performances of Aquivion PFSA PW98
for cellulose hydrolysis/glycosylation.
depolymerization/repolymerization of sugars occurred during the
mechanocatalytic process. This result agrees well with previous
works of Rinaldi, Schüth[ and Beltramini
12]
[18]
who investigated
the mechanocatalytic depolymerization of MCC in the presence
of H SO
2
4
.
[19]
In contrast to clays previously reported by Blair, no formation
of glucose, levoglucosan or furanic derivatives was observed
and Aquivion® PFSA PW98 similarly behaved as H
the mechanocatalytic process. The plot of the solubilization of
MCC in water as a function of the milling time is provided for
information in Figure S3.
A
2
4
2 4
SO during
®
The recovered oligosaccharides (1.5 g) containing Aquivion®
PFSA PW98 were next suspended in 10 g of n-dodecanol and
heated at 150°C. The reaction (yield, conversion) was monitored
by gas chromatography. To assess the catalytic performances of
Aquivion® PFSA PW98, three parameters were considered (1)
the initial production rate of AAGs, also denoted rAAG which
mechanocatalytic
depolymerization
Cellulose
glycosylation
n-dodecanol
150 C
AAG
corresponds to the number of mol of AAGs produced per mol of
+
AQUIVION PFSA
+
H
and per hour (determined at a conversion rate lower than
m = 1,2
n = 8,8
20%), (2) the reactor productivity corresponding to the mass of
AAGs produced (in kg) per m3 of solution and per hour
recycling
Aquivion®
PFSA PW98
AQUIVION PFSA
(
determined at the maximum yield of AAGs) and (3) the yield of
AAGs. For benchmarking purposes, the catalytic glycosylation
was also investigated from MCC ball-milled with H SO . During
2
4
Scheme 1. Strategy for the synthesis of AAG from cellulose
the glycosylation, AAGs may be produced with different DPs.
For instance, starting from glucose, AAGs are produced with a
DP typically ranging from 1.1 to 1.5 (i.e. 1.1 to 1.5 glycosyl unit
per fatty chain).[14] For the sake of clarity, only AAGs with a
DP=1 and denoted as AAG-1 were considered in this section.
Formation of AAGs with a higher DP is discussed later. During
the glycosylation reaction, the yield in AAG-1 passed through a
maximum and then decreased due to their partial degradation in
the reaction medium. Otherwise noted, the yield of AAG-1
provided in this work corresponds to the maximum yield. A
typical kinetic profile is provided in the supporting information
(Figure S4). Results are summarized in Table 1.
Results and Discussion
In this work, microcrystalline cellulose AVICEL PH 200 (MCC)
was selected because of its high purity to glucose which
facilitates analyses. The degree of polymerization (DP) of
AVICEL PH 200 is 250, the particle size is 250 m and the water
content is 5 wt%. As expected, MCC was highly recalcitrant to
the glycosylation reaction with n-dodecanol and all our attempts
to convert MCC to AAGs in the presence of Aquivion® PFSA
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
cellobiose or MCC-derived oligosaccharides, AAG-1 was
produced with a similar rAAG-1 suggesting the important role
played by the amphiphilic nature of Aquivion PFSA PW98 in
Table 1. Results on the catalytic glycosylation of MCC in the presence of an
acid catalyst
[
a]
®
this case (entries 2, 6).
To shed light on the great performances of Aquivion® PFSA
PW98, other solid acid catalysts such as kaolinite,
Montmorillonite and Amberlyst-15 were tested (entries 7-9). In
all cases, 30 wt% of solid catalyst was employed. Although
these solid acid catalysts were capable of partly depolymerizing
MCC during the mechanocatalytic step, no formation of AAG-1
was however observed during the subsequent glycosylation
reaction presumably due to the lack of amphiphilic properties.
®
Furthermore, in contrast to Aquivion PFSA PW98, we observed
Entry
Catalyst
Carbohydrates
AAG-1
yield
r
AAG-1
(h )
-
1
that Kaolinite, Montmorillonite and Amberlyst-15 were
deactivated in a large extent after the mechanocatalytic step
which may also explain their inactivity in the glycosylation
reaction. Indeed, all our attempts to recycle Amberlyst-15,
Kaolinite and Monrillonite in the mechanocatalytical step failed.
(%)
1
2
3
4
5
6
7
8
9
H
2
SO
4
MCC
MCC
0
0
40
430
0
Aquivion PFSA PW98
70
94
0
®
Aquivion PFSA PW98 is stable up to 150°C which prompted us
H
H
2
SO
4
4
glucose
Cellobiose[a]
glucose
Cellobiose[a]
MCC
to investigate its recycling, a pre-requisite step considering the
higher price of heterogeneous catalysts than mineral acids.
Typically, when the yield of AAG-1 reached a maximum value of
70%, the glycosylation reaction was stopped by cooling down
the reaction medium to 25°C. Ethanol was then added to fluidize
the crude and to precipitate unreacted carbohydrates. Then,
Aquivion® PFSA PW98 and the precipitated carbohydrates were
2
SO
Aquivion PFSA PW98
Aquivion PFSA PW98
Amberlyst-15
92
75
0
101
40
0
filtered off and washed with ethanol to remove physically
Kaolinite
MCC
0
0
®
adsorbed
AAG-1.
The
recovered
Aquivion
PFSA
PW98/carbohydrates mixture was then re-charged into the
planetary ball-mill and fresh MCC was added. The amount of
Montmorillonite
MCC
0
0
+
added
carbohydrates/Aquivion
Remarkably, we found that Aquivion PFSA PW98 was highly
robust under our working conditions and could be recycled at
least 5 times without obvious decrease in the AAG-1 yield. The
fresh
MCC
was
adjusted
to
keep
a
[
(
a] for solid acid catalysts (Aquivion PFSA PW98 (1mmol H /g), Amberlyst-15
+
®
1.7 mmol H /g), Kaolinite, Montmorillonite), 30 wt% (relative to cellulose) was
PFSA PW98 mass ratio of 2.
®
employed; [b] as in the case of MCC, cellobiose was first ball-milled with either
SO
or Aquivion PW98 before the glycosylation reaction (3h of milling at
00 rpm).
H
4
2
4
Whereas MCC ball-milled with H
2
SO
4
can be converted to
3
amount of -SO H groups (measured by acid-base titration) and
[
13]
butylglycosides by reflux in n-butanol, no reaction occurred
with n-dodecanol at 150°C (entry 1). This result might be
explained by the huge difference of polarity between hydrophilic
oligosaccharides and hydrophobic n-dodecanol making their
interaction difficult. Remarkably, starting from MCC mix-milled
with Aquivion® PFSA PW98, AAG-1 were produced with 70%
yield after only 30 min of reaction which corresponded to reactor
sulfur (measured by elemental analysis) on Aquivion PFSA
remained unchanged before and after use which demonstrates
the robustness of Aquivion PFSA under our experimental
conditions. The absence of fluorine (determined by 19F NMR)
and sulfur (determined by elemental analysis) in the liquid phase
is also consistent with this claim. To the best of our knowledge,
Aquivion® PFSA PW98 is the first example of solid acid catalyst
recyclable in the mechanocatalytic depolymerization of MCC.
productivity as high as 250 kg.m h and a rAAG-1 of 40 h-1 (entry
-3 -1
2
). This results indicates that (1) Aquivion® PFSA PW98 was still
1
00
active after the mechanocatalytic step and (2) the amphiphilic
properties of Aquivion® PFSA PW98 favored a better interaction
between hydrophilic oligosaccharides and hydrophobic n-
dodecanol. To support this claim, the glycosylation of glucose
9
0
80
70
60
5
0
and cellobiose were investigated either in the presence of H
or Aquivion® PFSA PW98. With H
with 94% yield from glucose (entry 3). However, from cellobiose,
no reaction occurred indicating that an increase in the DP of
2 4
SO
40
2
SO , AAG-1 was produced
4
30
20
10
0
carbohydrates had a detrimental effect on the H
(
2 4
SO efficiency
entry 4). Conversely, in the presence of Aquivion® PFSA PW98,
1
2
3
4
5
AAG-1 was produced with 92% and 75% yield from glucose and
cellobiose, respectively (entries 5, 6). Interestingly, from
Catalytic run
Figure 1. Recycling of Aquivion® PFSA PW98
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
From a mechanistic point of view, two reactions pathways may
was weighted. A mass of 0.77 g was obtained indicating that
0.27 g of carbohydrates did not react. It corresponds to a
conversion of 72% indicating that the selectivity of the reaction
to AAG-1 is 97%. The high selectivity towards AAG-1 was also
evidenced by H NMR and GC analyses (Figure S5).To confirm
the high selectivity to AAG-1, the crude product was also passed
theoretically occur (Fig. 2). The path
1
involves
a
transglycosylation reaction between n-dodecanol
and
oligosaccharides. The path 2 involves a glycosylation reaction
on the reductive ends of oligosaccharides which is accompanied
by the release of water. Water next hydrolyzes oligosaccharides
which generate additional reductive ends that can be next
glycosylated and so on.
To elucidate the dominant path, the reaction was first carefully
examined by GC, HPLC and mass spectrometry. No trace of
free glucose or disaccharides was observed which suggest the
dominance of path 1. To further support this claim, the reaction
was conducted at 150°C either under vacuum or in a closed
reactor. If water is produced as a co-product (case of path 2),
the vacuum is expected to quickly distill it out of the reaction
vessel. Hence, vacuum should dramatically impact the rate of
the path 2. Interestingly, AAG-1 was produced in 70% yield after
1
2
through a plug of silica and purified using a mixture MeOH/H O
as eluent. Only the excess of n-dodecanol, AAG-1 and
didodecylether (5 % yield) were recovered confirming the
selective conversion of oligosaccharides to AAG-1. In all cases,
the pyranoside and furanoside forms ( and  isomers) of AAG-
1
13
1 were clearly observable on the H and C NMR spectra (in d6-
DMSO) and on the GC chromatogram as well. Quantification of
each form was carried out by GC. The pyranoside form was
dominant and the pyranoside/furanoside ratio was 9. As
1
13
suggested by H and C NMR, both forms co-existed as a
mixture of  and  anomers in a ratio / of 2 and 1 for the
pyranoside and furanoside form, respectively. It is worth
mentioning that when the n-dodecanol/oligosaccharide ratio was
decreased from 10 to 3, formation of AAGs with a DP of 1, 2 and
3 (average DP=1.2) were clearly observed by GC indicating that
the DP of AAGs can be tuned by adjusting the experimental
conditions.
30 min of reaction either under vacuum or in a closed reactor
which indicate that path 2 was less likely to occur. Altogether,
these results suggest that, from oligosaccharides, the
transglycosylation reaction was the dominant path. The exact
mechanism for transglycosylation of cellulose is still under
debate in the literature (protonation of the anomeric oxygen vs
protonation of the oxygen inside the ring in the closed form).[20]
This aspect is the topic of current investigations in our groups
and it will be published in due course.
Having all these results in our hands, the process was next
optimized. The time of the mechanocatalytic step was
successfully reduced from 24 h to 3 h without any effect either
on the rAAG-1, AAG-1 yield and reactor productivity which is an
important aspect with respect to the energy consumption of the
mechanocatalytic technology (Table 3, entries 1, 2). A further
PATH 1
drop of the milling time from 3 h to 1 h however decreased the
r
+
H
+
, successive
transglycosylation
AAG-1 and the reactor productivity to 18 h-1 and 156 kg/m /h,
3
respectively (entry 3). Furthermore, the maximum yield in AAG-1
obtained in the glycosylation step was also decreased from 70%
to 45% suggesting that, after 1h of milling, MCC partly remained
recalcitrant to the glycosylation reaction.
H , transglycosylation
+
®
Next, the amount of Aquivion PFSA PW98 was varied. In these
experiments, the mechanocatalytic time was fixed to 3 h.
Increasing the MCC/ Aquivion® PFSA PW98 mass ratio from 2
to 5 during the mechanocatalytic step led to an increase of the
2
H O
H , glycosylation
+
r
AAG-1 from 37 to 47 h-1 and also improved the maximum yield in
AAG-1 from 70% to 85% suggesting a better efficiency of the
mechanocatalytic step under these conditions (entries 1, 4). Due
to the lower amount of Aquivion® PFSA PW98, the reactor
productivity of the glycosylation reaction was logically decreased
+
H
2
O
H , hydrolysis
3
but remained as high as 203 kg/m /h (entry 4). A further
increase in the MCC/ Aquivion® PFSA PW98 mass ratio from 5
to 10 however markedly decreased the rAAG-1 to 15 h-1 (entry 5).
Note that a maximum yield of AAG-1 of 60% was still obtained in
this case but after 350 min of reaction which dramatically
+
PATH 2
Figure 2. Two possible pathways for the conversion of oligosaccharides to
AAG-1
decreased the reactor productivity of the glycosylation step from
3
203 to 18 kg/m /h (entry 5).
The crude product was next carefully analyzed by means of
different analytical methods to determine the selectivity of the
reaction. First of all, after the first catalytic run, the liquid fraction
was separated from the solid and the recovered powder
containing Aquivion® PFSA PW98 and unreacted carbohydrates
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
Table 3. Optimization of the mechanocatalytic step and impact on the glycosylation reaction
Entry
MCC/ Aquivion® PFSA
PW98 mass ratio
Time (h)
Stirring speed (rpm)
Initial AAG
formation rate (h )
Maximum
AAG yield
Reactor productivity
-
1
3
(kg/m /h)
(%)
1
2
3
4
5
6
7
2
2
12
3
400
400
400
400
400
200
600
37
31
18
47
15
1
70
70
45
85
60
85
50
250
250
156
203
18
2
1
5
3
10
5
3
3
8
5
3
32
111
The milling speed was also investigated. In this section, the
®
mechanocatalytic time was fixed to 3 h and a MCC/ Aquivion
PFSA PW98 mass ratio of 5 was employed. A decrease of the
mechanocatalytic stirring rate from 400 to 200 rpm led to the
formation of AAG-1 with unacceptable yield, rAAG-1 and reactor
productivity (entry 6). Under these conditions, MCC remained
highly recalcitrant. Similarly, an increase of the mechanocatalytic
stirring rate from 400 to 600 also decreased the rAAG-1 from 47 to
Experimental Section
Chemical and reagents: Microcrystalline cellulose of the type
AVICEL PH 200 was used and purchased to FMC Biopolymers.
MCC has glucose content higher than 99%, a water content of 5
wt% and a particle size mean of 150-250 µm. n-Dodecanol and
-1
3
32 h , the reactor productivity from 203 to 111 kg/m /h and the
H
2
SO
4
were purchased to Sigma-Aldrich and were used without
maximum yield of AAG-1 from 85 to 50% (entry 7). At 600 rpm, it
is worth mentioning that MCC turned brown suggesting a partial
degradation presumably caused by the formation of hot spot
®
further purification. Aquivion PFSA PW98, was kindly provided
by Solvay Specialty Polymers. Amberlyst-15 was purchased to
ACROS.
temperatures.
Hence,
optimal
conditions
for
the
General procedure for the synthesis of AAGs from
mechanocatalytic step was 400 rpm, 3 h and a MCC/ Aquivion®
PFSA PW98 mass ratio of 5. These conditions led to the
formation of AAG-1 in 85% yield, a rAAG of 47 h-1 and a reactor
cellulose: Typically, 1 g of cellulose was mixed with the desired
amount of Aquivion® PFSA PW98. The solid mixture was then
introduced into a planetary ball mill (Retsch PM 100) and the
mixture was milled at a spinning rate of 400 rpm for the desired
time. Next, the mix-milled mixture was suspended in 10 g of n-
dodecanol in a 50 mL round bottom flask equipped with a
magnetic stirring bar. The solution was heated in an oil bath at
3
productivity of 203 kg/m /h.
Conclusions
1
50°C under vacuum (15 mmHg) for the desired reaction time.
Aquivion® PFSA PW98 is
a solid superacid capable of
Note that vacuum is used to prevent oxidation reactions. The
reaction was monitored by GC.
converting cellulose to AAG-1 in 85% yield (i.e. 97% selectivity).
The remarkable catalytic performances of Aquivion® PFSA
PW98 was ascribed to its high acid strength caused by the
presence of perfluorinated chains but also to its suspected
amphiphilic nature favoring a better contact between hydrophilic
Gas Chromatography analyses: AAGs were analyzed on a
Varian 3350 equipped with an on-column injector and a flame
ionization detector. Before analysis, AAGs were silylated.
Typically, a sample (0.1 g) was taken from the reactor and
diluted in 2 mL of a solution of pyridine containing 1 mg.mL-1 of
sorbitol (internal standard). After solubilization, 1.0 mL of HDMS
and 1.0 mL of TMSCl were successively added. The solution
was then centrifuged for 10 minutes and 0.5 µL of the solution
was injected. A HT 5 column was used (L = 25 m, Ø = 0.32 mm,
e = 0.25 µm). The oven was heated from 50°C to 220°C with a
heating ramp of 10°C/min and then to 320°C with a heating
ramp of 10°C/min.
g
sugars and hydrophobic n-dodecanol. Thanks to a T of 150°C,
Aquivion® PFSA PW98 was stable in our conditions and was
recycled without obvious loss of activity or selectivity either in
the mechanocatalytic or glycosylation step.
We are conscious that the mechanocatalytic step is an energy-
consuming technology and may represent, at this stage, a
limitation for a deployment at a larger scale. As a perspective to
this work, we are now exploring different reactors more adapted
for this purpose.
Recycling experiments: After the first catalytic run, the
Aquivion® PFSA was filtered off at 100°C. Then, Aquivion®
PFSA was washed with water and with ethanol to remove most
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
of adsorbed glucose and AAGs. The recovered catalyst was
then dried for one hour at ambient temperature under vacuum
Dorris, Angew. Chem. Int. Ed. 2011, 50, 5438-5466; c) I. Siro, D.
Plackett, Cellulose 2010, 17 (3), 459-494; e) R. J. Moon, A. Martini, J.
Nairn, J. Simonsen, J. Youngblood, Chem. Soc. Rev. 2011, 40 (7),
(
15 mmHg) and directly re-engaged in the next catalytic run
without any further purification. Experimental parameters
reaction time, temperature, n-dodecanol/glucose ratio, catalyst
3941-3994.
[
[
8]
9]
a) S. Dora, T. Bhaskar, R. Singh, D. V. Naik, D. K. Adhikari, Biores.
Technol., 2012, 120, 318–321; b) W. Deng, M. Liu, Q. Zhang, X. Tan, Y.
Wang, Chem. Commun. 2010, 46, 2668-2670; c) W. Deng, M. Liu, O.
Zhang, Y. Wang, Catal. Today, 2011, 164, 461–466; d); G. Guchhait, A.
K. Misra, Catal. Commun., 2011, 14, 52–57; e) Y. Ishikawa, S. Saka,
Cellulose, 2001, 8, 189-195.
(
loading,etc.) remained unchanged between each catalytic run.
Acknowledgements
a) N. Villandier, A. Corma, Chem. Commun., 2010, 46, 4408-4410 ; b)
N. Villandier, A. Corma, ChemSusChem, 2011, 4 (4), 508-513 ; c) I. A.
Ignatyev, P. G. N. Mertens, C. Van Doorslaer, K. Binnemans, D. E. de
Vos, Green Chem., 2010, 12, 1790-1795.
Authors are grateful to the CNRS, the French Ministry of
Research, ARD, SOLVAY and ADEME (project APG-2G) for
their financial supports.
[10] A. V. Puga, A. Corma, ChemistrySelect, 2017, 2, 2495-2498.
[11] M. J. Climent, A. Corma, S. Iborra, S. Martnez-Silvestre
ChemSusChem 2016, 9,3492-3502.
Keywords: biomass • cellulose • mechanocatalysis •
glycosylation • glycosides
[
12] a) N. Meine, R. Rinaldi, F. Schüth, ChemSusChem, 2012, 5, 1449-
1454; b) F. Schuth, R. Rinaldi, N. Meine, M. Käldström, J. Hilgert, M. D.
Kaufman Rechulski, Catal. Today, 2014, 234, 24-30.
[
1]
a)
published
by
Transparency
Market
Research
[13] F. Boissou, N. Sayoud, K. De Oliveira Vigier, A. Barakat, S. Marinkovic,
B. Estrine, F. Jérôme, ChemSusChem, 2015, 8, 3263-3269.
[14] a) H. Hideshi, Y. Ono, Y. in "Solid Acid Catalysis: From Fundamentals
to Applications" CRC Press, 2015, 299; b) G. A. Olah, G.S.K. Prakash,
A; Molnar, J. Sommer in "Superacid Chemistry" 2009, 2nd Edition John
Wiley and Sons ; New York
(
-
www.transparencymarketresearch.com ) "Specialty Surfactants Market
Global Scenario, Raw Material and Consumption Trends, Industry
Analysis, Size, Share & Forecast 2011-2017; b) Market Report Global
Surfactant Market 4th Edition, Market Report, 2016.
[
2]
R. J. Farn in Chemistry and Technology of Surfactants, 2006, Blackwell
Publishing Ltd, Oxford; b) P.G. Jessop, F. Ahmadpour, M.A. Buczynski,
T.J. Burns, N.B. Green II, R. Korwin, D. Long, S.K. Massad, J.B.
Manley, N. Omidbakhsh, R. Pearl, S. Pereira, R.A. Predale, P.G. Sliva,
H. VanderBilt, S. Weller, M.H. Wolf, Green Chem. 2015, 17, 2664-2678.
a) M. Kjellin, I. Johansson in Surfactants form Renewable Resources,
[15] A. Karam, K. De Oliveira Vigier, S. Marinkovic, B. Estrine, C. Oldani, F.
Jérôme ACS Catal., 2017, 7 (4), 2990–299.
[16] a) W. Fang, S. Wang, A. Liebens, F. De Campo, H. Xu, W. Shen, M.
Pera-Titus, J.-M. Clacens, Catal. Sci. Technol. 2015, 5, 3980-3990; b)
R. Tassini, V. D. Rathod, S. Paganelli, E. Balliana, O. Piccolo, J. Mol.
Catal. A : Chem. 2016, 41, 257-263; c) L. Bianchi, E. Ballerini, M. Curini,
D. Lanari, A. Marrocchi, C. Oldani, L. Vaccaro, ACS Sustainable Chem.
Eng. 2015, 3, 1873-1880; d) A. Karam, N. Sayoud, K. De Oliveira Vigier,
J. Lai, A. Liebens, C. Oldani, F. Jérôme, J. Mol. Catal. A : Chem. 2016,
422, 84-88.
[
3]
4]
2010, John Wiley & Sons, Ltd; b) P. Foley, A. Kermanshahi, E.S. Beach,
J.B. Zimmerman, Chem. Soc. Rev., 2012, 44, 4, 1499-1518.
[
a) D. Balzer, H. Lüders, In Nonionic surfactants: alkyl polyglucosides,
Marcel Dekker:ꢀ New York and Basel, Switzerland, 2000, vol. 91, p19;
b) W. von Rybinski, K. Hill, Angew. Chem. Int. Ed. 1998, 37, 1328-
1
345; c) F. Martel, B. Estrine, R. Plantier-Royon, N. Hoffmann, C.
[17] a) W.-J. Zhou, L. Fang, Z. Fan, B. Albela, L. Bonneviot, F. De Campo,
M. Pera-Titus, J.-M. Clacens, J. Am Chem. Soc. 2014, 136, 4869-4872;
b) W. Fang, S. Wang, A. Liebens, F. De Campo, H. Xu, W. Shen, M.
Pera-Titus, J.-M. Clacens, Catal. Sci. Technol. 2015, 5, 3980-3990; c)
W. Fang, Z. Fan, H. Shi, S. Wang, W. Shen, H. Xu, J.-M. Clacens, F.
De Campo, A. Liebens, M. Pera-Titus, J. Mater. Chem. A 2016, 4,
4380-4385.
Portella, Top. Curr. Chem., 2010, 294, 79-115; d) A. Ismail, S. Soultani,
M. Ghoul, J. Biotechnol., 1999, 69, 135–143; e) A. Ismail, S. Soultani,
M. Ghoul, J. Biotechnol., 1999, 69, 145–149; f) J. Y. Li, Y. J. Liu, G.
Zheng, Y. Sun, Y. N. Hao, T. Fu, Adv. Mater. Res., 2012, 550-553, 75-
79; g) K. Böge, L. F. Tietze, Fett/Lipid, 1998, 100, S. 36–41; h) A.
Corma, S. Iborra, S. Miquel, J. Primo, J. Catal. 1998, 180, 218– 224.
a) K. Hill, W. von Rybinski, G. Stoll, In Alkyl Polyglycosides: Technology,
Properties and Applications, Eds VCH Verlagsgesellschaft mbH
Weinheim, 1997, 9-22; b) J. Nowicki, J. Woch, M. Mościpan, E.
Nowakowska-Bogdan, Appl. Catal. A, Gen., 2017, 539, 13-18.
[
5]
[18] A. Shrotri, L. K. Lambert, A. Tanksale, J. Beltramini, Green Chem.,
2013, 15, 2761-2768.
[19] S. M. Hick, C. Griebel, D. T. Restrepo, J. H. Truitt, E. J. Buker, C. Bylda,
R. G. Blair, Green Chem. 2010, 12, 468-474.
[
[
6]
7]
a) F. Bouxin, S. Marinkovic, J. Le Bras, B. Estrine, Carbohydrate Res.
[20] Y. Pierson, F. Bobbink, N. Yan, Chem Eng Process Tech., 2013, 1(2),
1014.
2010, 345, 2469-2473; b) N. Hausser, S. Marincovic, B. Estrine,
Cellulose, 2013, 20, 2179-2184 and references cited therein.
as selected reviews on cellulose, please see a) D. Klemm, B. Heublein,
h.-P. Fink, A. Bohn, Angew. Chem. Int. Ed. 2005, 44, 3358-3393; b) D.
Klemm, F. Kramer, S. Moritz, T. Lindström, M. Ankerfors, D. Gray, A.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201700903
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 2:
FULL PAPER
Aquivion PFSA
Ayman Karam, Karine De Oliveira
Vigier, Sinisa Marinkovic, Boris Estrine,
Claudio Oldani and François Jérôme
AAG ( and  forms)
Page No. – Page No.
Conversion of cellulose to
amphiphilic alkyl glycosides
catalyzed by Aquivion, a
Oligosaccharides
+
Aquivion PFSA
Aquivion PFSA is an amphiphilic superacid capable of converting microcrystalline
cellulose to amphiphilic alkyl glycoside in yields, selectivity and reactor
productivities unreported to date
perfluorosulfonic acid polymer
This article is protected by copyright. All rights reserved.