Kinetic analysis of hexose conversion to methyl lactate by Sn-Beta: Effects of substrate masking and of water

Article abstract of DOI:10.1039/c8cy00335a

Simple sugars show promise as substrates for the formation of fuels and chemicals using heterogeneous catalysts in alcoholic solvents. Sn-Beta is a particularly well-suited catalyst for the cleavage, isomerization and dehydration of sugars into more valuable chemicals. In order to understand these processes and save resources and time by optimising them, kinetic and mechanistic analyses are helpful. Herein, we study substrate entry into the Sn-Beta-catalysed methyl lactate process using abundant hexose substrates. NMR spectroscopy is applied to show that the formation of methyl lactate occurs in two kinetic regimes for fructose, glucose and sucrose. The majority of methyl lactate is not formed from the substrate directly, but from methyl fructosides in a slow regime. At 160 °C, more than 40% of substrate carbon are masked (i.e. reversibly protected in situ) as methyl fructosides within a few minutes when using hydrothermally synthesised Sn-Beta, while more than 60% methyl fructosides can be produced within a few minutes using post-synthetically treated Sn-Beta. A significant fraction of the substrate is thus masked by rapid methyl fructoside formation prior to subsequent slow release of fructose. This release is the rate-limiting step in the Sn-Beta-catalysed methyl lactate process, but it can be accelerated by the addition of small amounts of water at the expense of the maximum methyl lactate yield.

Full text of DOI:10.1039/c8cy00335a

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: I. Tosi, A. Riisager,
E. Taarning, P. Jensen and S. Meier, Catal. Sci. Technol., 2018, DOI: 10.1039/C8CY00335A.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
ISSN 2044-4753
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Catalysis Science & Technology
Paper
Kinetic Analysis of Hexose Conversion to Methyl Lactate by Sn
Beta: Effects of Substrate Masking and of Water†
Irene Tosi,a Anders Riisager,a* Esben Taarning,b Pernille Rose Jensenc and Sebastian Meiera*
3Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
Simple sugars bear promise as substrates for the formation of fuels and chemicals using heterogeneous catalysts in
alcoholic solvents. Sn-Beta is a particularly well suited catalyst for the cleavage, isomerization and dehydration of sugars
into more valuable chemicals. In order to understand these processes and save resources and time by optimising them,
kinetic and mechanistic analyses are helpful. Herein, we study substrate entry into the Sn-Beta catalysed methyl lactate
process using abundant hexose substrates. NMR spectroscopy is applied to show that the formation of methyl lactate
occurs in two kinetic regimes for fructose, glucose and sucrose. The majority of methyl lactate is not formed from the
substrate directly, but from methyl fructosides in a slow regime. At 160 °C, more than 40% of substrate carbon are masked
(i.e. reversibly protected in situ) as methyl fructosides within few minutes when using hydrothermally synthesised Sn-Beta,
while more than 60% methyl fructosides can be produced within few minutes using post synthetically synthesised Sn-Beta.
A significant fraction of substrate thus is masked by rapid methyl fructoside formation prior to subsequent slow release of
fructose. This release is the rate limiting step in the Sn-Beta catalysed methyl lactate process, but can be accelerated by
the addition of small amounts of water at the expense of maximum methyl lactate yield.
17, 19
byproducts.12,
The transformation of simple sugars into
-hydroxy esters12-14,
11
16, 26
Introduction
methyl lactate10,
or other
α
is a
possible route for the sustainable production of polymers in
bio-based processes. PLA (polylactic acid) has attracted great
interest because of its mechanical and physical properties, the
possibility to combine lactic acid with other monomers and to
thus obtain a large variety of copolymer materials for diverse
applications.27 In addition, lactic acid is a platform for other
chemicals and alky lactates are promising green solvents.17
Knowledge-based approaches to improving biomass
conversion processes could benefit from detailed kinetic and
mechanistic understanding of the processes. Relatively little
attention has been devoted to the systematic study of reaction
kinetics in the Sn-Beta catalysed methyl lactate process.
Herein, we therefore employ quantitative NMR methodology
to derive a kinetic model of carbohydrate influx into the
pathway that ultimately converts fructose intermediate to
methyl lactate. Glucose and fructose had previously been
shown to yield similar product mixtures in the Sn-Beta
catalysed methyl lactate process, while the non-reducing
disaccharide sucrose resulted in higher methyl lactate yield,
possibly due to a slower release of reducing sugar. We
hypothesised that a slow release of reducing sugar may play a
generally neglected role in Sn-Beta catalysed conversions of
glucose, fructose and sucrose at temperatures near 160 °C,
masking all of these substrates as methyl fructoside. Previous
Simple sugars can be converted into fuels and chemicals,
including levulinic acid,1-6 5-hydroxymethyl furfural (HMF),5, 7-9
11
12, 13, 14 , 15-19
lactic acid10, and others9,
using zeolite-based
materials as heterogeneous catalysts. Sn-Beta in particular is
able to promote the cleavage, isomerization and dehydration
of carbohydrates into more valuable chemicals. Such reactions
include valorisations of abundant simple sugars by converting
them to rare sugars9, 11, 17-25 at moderate temperatures near
100 °C, or the formation of different hydroxyl esters as
prospective building block for the production of biomass-
derived polymers at temperatures near 160 °C.13, 14, 16, 26 Since
the stability of the catalysts for the production of Sn-Beta is
increased in alcoholic solvents as compared to water,6 the high
temperature process is normally carried out in short-chain
alcohol. The use of alcoholic solvents leads to the formation of
activated and pH-neutral methyl esters and glycoside
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Paper
Catalysis Science & Technology
results at 100 °C had shown that Sn-Beta and other catalysts as little as five minutes at 160˚C, especially when using Sn-Beta
with Lewis and weak Brønsted acidity sequester carbohydrate that contains defect sites. In the presence of added water at
as methyl-fructosides on the hours time scale.19 These levels that comply with catalyst stability, substrate masking
processes would be predicted to operate on the minutes time and reaction kinetics can be modulated to achieve faster
scale at 160 °C in competition to retroaldol cleavage of formation of methyl lactate.
fructose intermediate, and thus could impact on reaction
kinetics and substrate influx into the methyl lactate process.
Experimental Procedures
Scheme 1 depicts the equilibrium reactions of glucose and
fructose in methanol in the presence of solid Lewis and
Catalysts preparation
Brønsted acid catalysts. Glucose and fructose are
Sn-Beta zeolite was prepared via hydrothermal route as
previously described.28, 29 A mixture of 33.1 g of TEAOH (Sigma-
Aldrich, 35% in water) and 30.6 g of TEOS (Sigma-Aldrich, 98%)
was magnetically stirred for 1 h until the appearance of a
interconverted via isomerization through 1,2 hydride shift in
the presence of Lewis acidic sites (red equilibrium arrows in
Scheme 1), while Brønsted acidity promotes the formation of
methyl glycosides. Fructose is generally considered the more
reactive substrate than glucose due to its higher fraction of
five membered and acyclic forms.22 In addition, the ketose to
aldose isomerization of fructose or C1-C2 epimerization of
glucose can also produce mannose. C6 sugars are mostly
present as five- or six-membered rings termed furanose or
pyranose forms, respectively. In the cyclic form, carbohydrates
further exhibit stereoisomerism at the so-called anomeric
position deriving from the carbonyl group with two possible
.
homogeneous phase. Then, a solution of SnCl4 5 H2O (Sigma-
Aldrich, 98%) in water was added with the proportion Si/Sn =
150 and the mixture was kept stirring for 24 h. Afterwards, 3.1
g of HF (Fluka, 47-51%) in 1.6 g of H2O were added and
homogenised under manual stirring. The white granulose solid
was crystallised in a Teflon-lined stainless steel autoclave for
14 days at 140 °C, filtered, washed thoroughly with 4 L of
deionised water, dried at 80 °C overnight and finally calcined
at 550 °C for 6 h to obtain HT Sn-Beta (150).
anomeric forms called the
Brønsted acidity in alcoholic solvents, five- or six-membered
and -alkyl-glycosides are thus formed. Overall, a complex
α and β-forms. In the presence of
The Sn-Beta used for the synthesis of methyl fructosides
was prepared via post-treatment (PT).30 Initially, 1 g H-Beta
zeolite (Zeolyst, Si/Al = 12.5, verified experimentally) was
dealuminated by acidic treatment with HNO3 (VWR, 65%,
diluted to 13 M in water) at 100 °C overnight. The deAl-Beta
zeolite was then washed with water until neutral pH and dried
overnight at 120 °C. Subsequently, the Sn was incorporated via
incipient wetness impregnation using SnCl4·5H2O dissolved in
water added in a proportion for obtaining a Si/Sn ratio of 150.
The PT Sn-Beta (150) catalyst was finally dried overnight at 120
°C and calcined at 550 °C for 6 h before use.
α-
β
mixture of isomeric carbohydrates and glycosides is expected
to form even from pure carbohydrate substrates in reactions
using carbohydrate substrate, alcohol solvent and Sn-Beta
catalyst above 100˚C (Scheme 1).
We set out to study the formation of glycosides and how
they affect the influx of carbon into the retroaldol-reactions.
To this end, a quantitative high-resolution 1H-13C 2D NMR
approach was conducted at high magnetic field strength. A
kinetic model of the methyl lactate process is obtained.
Mechanistic and kinetic insights are applied to show that
exclusion of water can warrant a process allowing high
conversion of glucose and sucrose to methyl fructoside within
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 3 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Journal Name
Catalysis Science & Technology
Catalysts characterization
obtain response factors for the quantitative analysis of the
Catalyst characterization included XRD, nitrogen physisorption analytes of interest by the 2D HSQC spectra. 1H-13C-HSQC
and ICP-OES. Characterization by powder X-ray diffraction spectra were acquired including
(XRD) was performed on an X'Pert diffractometer (Philips). 1024(1H) 300(13C) complex data points centered around a 13C
XRD confirmed that the synthesis resulted in Beta carrier offset of 62 ppm and employing a spectral width of 20
a spectrum sampling
×
a
morphology without the formation of extra framework SnO2 ppm in the 13C dimension to sample 13C signal for 50 ms. This
(ESI, Fig. S1). Measurements of BET surface area and pore spectrum of 25 min duration provided sufficient resolution and
volume was performed by nitrogen adsorption/desorption sensitivity to identify and quantify sucrose as well as glucose
measurements at -196 °C using an Autosorb automatic surface and fructose anomeric and ring forms as well as the anomeric
area and pore size analyser (Quantachrome Instruments). The and ring forms of their epimers and glycosides. To this end,
samples were heated at 200 °C for 4 h under vacuum before signals of the C6 position were used as structural reporters.
the nitrogen physisorption measurements. The elemental For the quantification of the sugars, the yield was calculated
composition of the synthesised Sn-Beta materials was after measuring their response factor from signal areas in the
determined by inductively coupled plasma-atomic emission 1H-13C HSQC spectral region of the primary alcohol (see Fig.
spectroscopy (ICP-OES) on a Perkin-Elmer Optima 3000 was 1B). Reaction mixtures were analysed in this way without
used to determine the elemental composition of the further purification or derivatization, owing to the high
hydrothermally and post-synthetically prepared Sn-Beta resolution of 1H-13C HSQC spectra on mixtures of small
materials. The results of catalyst characterization are molecule analytes when acquiring 13C signal for sufficiently
summarised in the ESI, Table S1.† Scanning Electron long acquisition times.31, 32 Equivalently, 1H-13C-HSQC spectra
Microscope images were recorded on a FEI Quanta 200 ESEM were acquired around a 13C carrier offset of 102 ppm in order
FEG instrument (Fig. S3).
to detect hemiacetal and acetal signals.
Spectral processing and kinetic fitting
Catalytic experiments
The catalytic experiments were carried out in a Biotage®
All spectra were processed with ample zero filling in all
Initiator+ microwave synthesizer. D-(-)-fructose (Sigma-Aldrich, dimensions and analysed in Bruker TopSpin 3.5pl7. Spectra
99.9%), D-(+)-glucose (Sigma-Aldrich, 99.5%) and sucrose were integrated and Methyl lactate integrals were fitted to
(Sigma-Aldrich, 99.5%) were used as substrates. First, the kinetic rate laws in profit 6.29 (Quantumsoft), while glucose,
sugar was solubilised in methanol (Sigma-Aldrich, 99.9%) to fructose, fructoside and methyl lactate signals were fitted to a
yield stock solutions of 0.132 M monomer concentration system of differential equations for the kinetic model of Fig. 5.
(0.066 M sucrose). Subsequently, 5 mL of stock solution and 50
Kinetic rate laws for mono- and biexponential fits of
were as follows: For the
mg of catalyst were reacted at 160 °C under magnetic stirring methyl lactate yields
Y
for different times using 80 µL of DMSO (Sigma-Aldrich, 99.5%) monoexponential fit Y=A(1-e-kt), where A is the maximum yield
as internal standard. In a typical time-resolved experiment, 10 and k is the apparent first order rate constant. For the
reactions using the same conditions were carried out with biexponential fit Y=B(1-e-lt)+C(1-e-mt), where B is the maximum
reaction times of 5 s, 10 s, 30 s, 90 s, 5 min, 10 min, 20 min, 40 yield of methyl lactate formed from fructose with apparent
min, 60 min and 240 min. After the reaction, the catalyst was first order rate constant l and C is the maximum yield of
removed by filtration using a 0.22
ꢀm Nylon syringe filter methyl lactate formed from methyl fructoside with apparent
(Frisenette) and the reaction mixture analysed using NMR first order rate constant m.
spectroscopy.
The system of differential equations for the kinetic model
of Fig. 5 comprised the equations
NMR spectroscopy
NMR spectra were recorded on a Bruker Avance III 800
MHz spectrometer equipped with a TCI cryoprobe at 25 °C.
Crude reaction mixtures in protonated anhydrous methanol
(Sigma-Aldrich, 99.8%) were analysed after addition of 100 µL
of deuterated methanol (Sigma-Aldrich, 99.8% D) to 500 µL of
(1) d[Glc]/dt=-k1[Glc]+k-1[Fru]
(2) d[Fru]/dt=k1[Glc]-k-1[Fru]- k2[Fru]+k-2[Me-Fru]-k3[Fru]
(3) d[Me-Fru]/dt= k2[Fru]-k-2[Me-Fru]
(4) d[Me-Lac]/dt=k3[Fru]
the reaction mixture. The spectra that were acquired included where [Glc], [Fru], [Me-Fru] and [Me-Lac] are normalised
a quantitative 1D 13C spectrum. This spectrum was acquired integrals of glucose, fructose, methyl fructoside and methyl
with a recycle delay of 30 s, sampling 65536 complex data lactate, respectively. Byproducts such as methyl glucosides or
points during an acquisition time of 1.36 s. Quantitative 13C reducing and glycoside forms of mannose and sorbose were
NMR spectroscopy was pursued by the comparison of 13C NMR omitted due to their low prevalence of few percent in the
signal areas in calibration samples relative to an internal reaction mixture at any given time. Integrals were normalised
standard (protonated DMSO).14
to account for the loss of carbon into other byproducts and all
In order to improve sensitivity and reduce time integrals were weighted equally. Parameters k1, k-1, k2, k-2 and
requirements, more sensitive 1H-13C HSQC spectra were k3 are rate constants that were determined by curve-fitting
subsequently acquired on the calibration samples in order to using Lmfit in the spyder (release 3.2.4) Python environment
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Paper
Catalysis Science & Technology
package. The kinetic parameters of equations (1)-(4) are Kinetic analysis of methyl lactate formation: Two regimes due to
defined in Fig. 5.
substrate masking as methyl fructoside
The chemocatalytic conversion of abundant carbohydrates to
methyl lactate may benefit from a detailed understanding of
rate limiting steps in the process in order to become more
competitive to biocatalytic processes and provide optimised
kinetics or yields. Analysis of reaction mixtures by quantitative
ex situ NMR allowed the study of reaction progress. Fig. 2
displays the kinetics of methyl lactate formation from glucose
alongside mono- and biexponential fits of the data. The data
and kinetic fitting showed that methyl lactate was formed in a
reaction encompassing two kinetic regimes. The initial fast
regime accounted only for a minority of the methyl lactate
formed and was followed by an orders of magnitude slower
second regime. These data indicated that the majority of
methyl lactate derived from a form of the substrate that was
masked (protected in situ) in the chemocatalytic methyl
lactate process catalysed by Sn-Beta in methanol.
Results and Discussion
Chemical species detected in 2D NMR (1H-13C HSQC) spectra of
reaction mixtures of the Sn-Beta catalysed conversion of
hexoses to methyl lactate are exemplified in Fig. 1. Beyond
glucosides and fructosides, mannoside as well as acetal forms
of 3-deoxyglucosone signals were detected. In addition,
sorbose was formed by 1,5-hydride transfer from glucose33, 34
and was usually present in the reaction mixture for the first
few minutes in very small amounts (<1%). Mannose, which is
formed via a 1,2-carbon shift from glucose or by 1,2-hydride
shift from fructose (Fig. S4),33, was most abundant at the
34
beginning of the reaction with <4% of maximum yield after few
seconds of reaction from fructose, but was subsequently
converted further, analogously to other sugars. Time resolved
experiments using NMR spectroscopic detection akin to Fig. 1
were used to pursue a robust kinetic understanding of the
methyl lactate process.
The kinetic experiment of Fig. 2 was subsequently analysed
for the conversion of glucose and the formation of other
chemicals than methyl lactate over time. In addition, the
kinetics of glucose, fructose and sucrose conversion were
Fig. 2 Yield of methyl lactate from glucose in time resolved experiments. Fits of
mono- and biexponential kinetics to the data are displayed, showing that methyl
lactate forms in two kinetic regimes. Conditions: 120 mg glucose, 50 mg HT Sn-
Beta (150) catalyst, 5 mL methanol, MW reactor for the indicated time at 160 ⁰C.
Fig. 1 Regions of 1H-13C-HSQC spectra containing the anomeric C1-H1 groups of
aldoses (A) and the primary alcohol C6-H6 groups of aldoses and ketoses (B).
Detected signals are exemplified for the conversion of 120 mg fructose in 5 mL
methanol after 10 s at 160 °C using a MW reactor. Abbreviations: Furanose (fur),
pyranose (pyr), 3-deoxyglucosone (3DG), fructose (fru), glucose (glu), mannose
(man).
Fig. 3 Yields of glucose, fructose, methyl glycosides and methyl lactate in time
resolved experiments starting from glucose and fructose. Conditions: 120 mg
substrate, 50 mg HT Sn-Beta (150) catalyst, 5 mL methanol, MW reactor for the
indicated time at 160 ⁰C.
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 5 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Journal Name
Catalysis Science & Technology
compared, as shown in Fig. 3 and Fig. S5. Glucose and
fructose reached comparable methyl lactate levels and
comparable levels of methyl fructosides within 5 min. On the
order of 40-50% of all substrate was sequestered as methyl
fructosides in the Sn-Beta catalysed conversion of glucose,
fructose (Fig. 3A) and sucrose (Fig. S5), when using HT Sn-Beta
at 160 ˚C. After 5 min, a slow decline of methyl fructosides was
observed that correlated with the formation of methyl lactate
in the slower reaction regime. The carbohydrate composition
resulting from fructose and glucose substrate is compared in
Fig. 3. As Sn-Beta is highly active in the isomerization of
glucose to fructose, only a minor dependence on the starting
substrate was observed.
Molecular detail on the fructoside formation was further
probed by the distinction of the formed fructosides. Shape
selectivity of zeolite catalysts could favor the formation of
particular glycoside ring sizes. The kinetic experiment of Fig. 3B
showed, however, that five-membered rings (furanosides) are
the main forms as expected for Brønsted acid catalysed
glycosidation under kinetic reaction control. During the time
course of the reaction, the relative fraction of
fructopyranosides increased, as expected due to the
equilibration of the faster forming furanosides with pyranoside
forms35 and due to faster hydrolysis of fructofuranosides.36
Notwithstanding, fructofuranosides remained the main form
of masked substrate throughout the first 4 hours of the
reaction.
Fig. 4 (A) Comparison of fructose, glucose, and sucrose conversion. (B) Fructose
and glucose conversion and yields of early intermediates in the time-resolved
experiments shown in Fig. 3. Conditions: 120 mg substrate, 50 mg HT Sn-Beta
(150) catalyst, 5 mL methanol, MW reactor for the indicated time at 160 ⁰C.
fructoside formed faster from fructose than from glucose in
the initial rapid regime of methyl lactate formation (Fig. 4B).
Methyl glucosides were formed in the reaction mixtures both
from glucose and from fructose in yields lower than 3% by
residual Brønsted acidity. The high Lewis acidity of Sn-Beta
catalysed the isomerization between glucose and fructose, but
led to an initial accumulation of the isomeric tautomer only to
a yield of 7-10% due to the rapid masking of fructose as
fructosides (Fig. 4B). Glucose-to-fructose isomerization thus
avoids the accumulation of the thermodynamically stable
glucosides20 due to the kinetically preferred sequestration of
fructosides both from glucose and fructose substrate.19
Different reactivity of glucose, fructose and sucrose
The kinetics of conversion for glucose, fructose and sucrose
substrate was subsequently compared (Fig. 4A). Albeit the Sn-
Beta-catalysed conversion of carbohydrates is often conducted
for several hours, the substrates are consumed on the seconds
to minutes time scale. Fructose was converted more rapidly
than glucose, while sucrose got converted slowest. Thus,
sucrose solvolysis and glucose to fructose isomerization
preceded subsequent reactions of fructose. Fructose is the
most reactive species both in Brønsted acid catalysed
formation of glycoside and in Lewis acid catalysed retroaldol
cleavage. It is worth noting that higher yields of methyl lactate
from sucrose than from glucose and fructose had previously
been observed and had been ascribed to sucrose releasing
both sugars more slowly from the disaccharide.37 Beyond the
slower conversion of sucrose, we detected higher methyl
fructoside levels from sucrose than from glucose and fructose
substrate (Fig. S5). Such increased formation of methyl
fructoside from sucrose indeed implies a particularly high
degree of substrate masking using sucrose substrate and a low
availability of free fructose.
Kinetic model
Time-resolved experimental data permit proposing a plausible
kinetic model for the methyl lactate process: Glucose and
fructose isomerize rapidly. Fructose can undergo Lewis acid-
catalysed retro-aldol cleavage and form methyl lactate or it
can undergo Brønsted acid-catalysed masking with methanol
to form methyl fructoside and water. This latter reaction is
reversible under the reaction conditions, while the conversion
to methyl lactate is irreversible.
A corresponding model with five kinetic rate constants is
shown in Fig. 5A. The time-resolved experimental data for
glucose conversion by HT Sn-Beta at 160 ˚C were fit to the
kinetic model of glucose conversion to yield fits displayed in
Fig. 5B and Fig. S6. The kinetic model fitted the data
reasonably well. Glucose and fructose isomerised with rate
constants k1 and k-1 of approximately 1.8 min-1. Fructose can
Initial processes in the conversion of the monosaccharides
glucose and fructose were deduced from kinetic profiles for
the first 10 min of the reaction (Fig. 4B). Due to the higher
availability of free fructose, methyl lactate and methyl
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Paper
Catalysis Science & Technology
Fig. 6 Yields of methyl glycosides and methyl lactate from glucose in time
resolved experiments using post treated Sn-Beta catalyst. Conditions: 120 mg
substrate, 50 mg PT Sn-Beta (150) catalyst, 5 mL methanol, MW reactor for the
indicated time at 160 ⁰C. Experiments using 50 mg HT Sn-Beta (150) catalyst are
overlayed for comparison.
Fig. 5 (A) Reaction model for the conversion of glucose to methyl fructoside and
methyl lactate. (B) Fit of experimental glucose, fructose, methyl fructoside and
methyl lactate concentrations in time resolved experiments to the model of (A).
Conditions: 120 mg substrate, 50 mg HT Sn-Beta (150) catalyst, MW reactor for
the indicated time at 160 ⁰C.
treated Sn-Beta catalyst. Higher Brønsted acidity in this defect-
containing PT Sn-Beta catalyst led to a higher accumulation of
methyl glucosides than for the HT Sn-Beta. Notably, sufficient
Lewis acidity to warrant rapid isomerization and higher
Brønsted acidity resulted in higher levels of methyl fructosides
when using the PT Sn-Beta catalyst, than when using the HT
Sn-Beta catalyst (Fig. 6). Yields of 60% methyl fructoside could
thus be achieved within 5-10 min from glucose at 160 ˚C using
PT Sn-Beta, while the competing formation of methyl lactate
was reduced relative to Brønsted acid catalysed Fischer
glycosidation. These data showed that balanced Lewis and
Brønsted acidity permit glucose to fructose isomerization at
competitive yields within few minutes at 160 °C, especially
when using PT Sn-Beta.
also react to form methyl fructoside and methyl lactate.
These products form in parallel as competing products of
Lewis acid- and Brønsted acid-catalysed reactions. Formation
of methyl fructoside was faster (k2=3.6 min-1) than of methyl
lactate (k2=0.72 min-1), consistent with a higher conversion of
fructose to fructoside than to methyl lactate in the initial fast
regime of the methyl lactate process. Hydrolysis of methyl
fructoside was limiting, with a rate k-2 of approximately 0.036
min-1, a factor of 20 slower than fructose conversion to methyl
lactate. Hence, the methyl lactate process proceeds in two
different kinetic regimes, as the fructose intermediate can
react in a Lewis acid catalysed pathway to methyl lactate, but
gets predominantly masked as methyl fructoside due to faster
Fischer glycosidation by weak Brønsted acidity. Masked
substrate subsequently reacted more than one order of
magnitude slower due to rate limiting unmasking by methyl
fructoside hydrolysis. The reaction scheme and rates thus
indicated that addition of water could be used to influence
unmasking and hence affect pathway kinetics in the methyl
lactate process. In addition, the high yield of methyl fructoside
within few minutes and its slower conversion indicated that a
glucose to fructoside conversion process is feasible at 160 °C
using catalysts with suitably balanced Brønsted and Lewis
acidity.
While fructosides are reactive forms of masked substrate,
glucosides accumulate on the low hour time scale and
glucoside formation results in a deactivation of the substrate
in the methyl lactate process (Fig. 6). These findings were
validated by the use of commercial methyl glucoside as
substrate for the methyl lactate process, showing that water
contents above 75% were required to unmask significant
fractions of methyl glucosides on the hours timescale. Such a
high water content compromises catalyst stability at 160 ˚C
and is thus not desirable.
Effect of water on rate and yield of methyl lactate formation
Kinetic data of the methyl lactate process showed that
substrate masking occurred, and that unmasking of methyl
fructoside by hydrolysis was rate limiting for the formation of
the majority of methyl lactate. Hence, we probed the effect of
water addition on kinetics and yields of the methyl lactate
process, using glucose as the substrate and water levels that
complied with catalyst stability at 160 °C. We hypothesised
that small amounts of water would aid the unmasking of
methyl fructoside and thus accelerate the formation of methyl
lactate. Fig. 7A shows the kinetic profile of methyl fructoside in
the presence of varying amounts of water. The masking of
substrate in the form of fructosides decreased in the presence
Glucose isomerization to 60% fructoside within few minutes
The faster formation of methyl fructoside than of methyl
lactate opens a possibility for production of methyl fructosides
from glucose on the low minute timescale (rate constant of 3.6
min-1 at 160 °C using HT Sn-Beta). Improved conversion of
glucose to fructoside should be especially feasible when
tailoring Brønsted and Lewis acidic sites to sequester fructose
more efficiently as fructosides. To this end, we used a post
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 7 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Journal Name
Catalysis Science & Technology
Fig. 7 (A) Methyl fructoside formation and consumption using glucose as the
substrate in the presence of different amounts of water in methanol (5 mL total
volume) at 160⁰C and 50 mg HT Sn-Beta catalyst. Accumulation of methyl
fructosides decreases as the water content increases. (B) Faster formation of
methyl lactate in the presence of water, and higher final yield in the absence of
water. The sum yield of lactate and methyl lactate is displayed.
of water as anticipated. This decrease in the accumulation of
methyl fructoside correlated with faster hydrolysis of methyl
fructoside. Thus, the addition of 2%, 5% and 10% water
accelerated the hydrolysis of methyl fructosides by factors of
5.6, 9.9 and 15.5, respectively. In this manner, the addition of
water at moderate concentrations that are not obstructive to
catalyst stability could accelerate the mobilization of the
masked methyl fructoside in the chemocatalytic conversion of
carbohydrates in methanol.
Fig. 8. (A) Fructose formation from glucose in the presence of different amounts
of water. (B) Methyl fructoside and methyl lactate formation from glucose within
20 minutes of reaction time in the presence of varying amounts of water.
Conditions: 120 mg glucose, 50 mg HT Sn-Beta (150) catalyst, MW reactor for the
indicated time at 160 ⁰C in the presence of indicated amounts of water (% v/v).
quantitatively mobilized within 40 minutes at water
concentrations as low as 5% (Fig. 7A). These results are
consistent with recent findings indicating that the methyl
lactate process under flow conditions is best performed in the
presence of water in the 1-10% range, with an optimum near
5%.39
Rates of methyl lactate formation were increased in the
presence of increasing amounts of water, even though water
may be expected to compete with the binding of substrate to
the catalyst Sn sites.38 Notably, the reduced masking of
substrate in the presence of added water was accompanied by
lower final yields of methyl lactate (Fig. 7B). In addition,
reduced methyl fructoside pools in the presence of added
water largely reduced the slow kinetic regime and increased
the fraction of methyl lactate formed in the initial fast regime.
As a result, it took approximately 70, 35, 12 and 7 minutes to
reach a yield of 20% methyl lactate upon addition of 0%, 2%,
5% and 10% water, respectively (Fig. 7B). Higher water
contents were not studied, as they are detrimental to catalyst
stability39 and to final methyl lactate yields (Fig. 7B), but favor
formation of carbonaceous material and of Brønsted acid
catalyzed products.10 Furthermore, methyl fructoside was
Increased unmasking of methyl fructoside and faster
formation of methyl lactate correlated to increased free
fructose intermediate in the presence of added water (Fig. 8A).
The increased availability of free fructose during the first 20
min of reaction allowed methyl lactate yields to be reached in
the presence of 5-10% (v/v) water that would take hours to
achieve in the absence of added water (Fig. 7B). These findings
indicate that the yield of methyl lactate per unit time could be
optimised by water addition, with possible implications for the
production capacity both by batch and flow setups. Yields of
methyl lactate, lactate, methyl fructoside and fructose after 20
min are summarised in Fig. 8B. Notably, only minor amounts of
lactate were formed relative to the predominant methyl
lactate product, in the range of 0-10% added water. Similar
results were obtained when using fructose instead of glucose
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Paper
Catalysis Science & Technology
4
5
B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Chem. Eng.
Res. Des., 2006, 84, 339-349.
V. Choudhary, S. H. Mushrif, C. Ho, A. Anderko, V. Nikolakis,
N. S. Marinkovic, A. I. Frenkel, S. I. Sandler and D. G. Vlachos,
J. Am. Chem. Soc., 2013, 135, 3997-4006.
S. Saravanamurugan and A. Riisager, ChemCatChem, 2013, 5,
1754-1757.
T. Wang, M. W. Nolte and B. H. Shanks, Green Chem., 2014,
16, 548-572.
as the substrate. Also in this case, fructoside formation was
suppressed by the addition of small amounts of water, while
the conversion to methyl lactate was accelerated in the
presence of water (Fig. S7). These reactions of fructose
underline possible benefits of water addition in improving the
production of methyl lactate per unit time by modulating
kinetic profiles of the methyl lactate process with water.
6
7
8
R.-J. van Putten, J. C. van der Waal, E. de Jong, C. B.
Rasrendra, H. J. Heeres and J. G. de Vries, Chemical Rev.,
2013, 113, 1499-1597.
9
M. Moliner, Y. Román-Leshkov and M. E. Davis, Proc. Nat.
Acad. Sci., 2010, 107, 6164-6168.
Conclusions
10 M. S. Holm, S. Saravanamurugan and E. Taarning, Science,
2010, 328, 602-605.
11 E. Taarning, S. Saravanamurugan, M. S. Holm, J. Xiong, R. M.
West and C. H. Christensen, ChemSusChem, 2009, 2, 625-
627.
12 S. Tolborg, S. Meier, I. Sadaba, S. G. Elliot, S. K. Kristensen, S.
Saravanamurugan, A. Riisager, P. Fristrup, T. Skrydstrup and
E. Taarning, Green Chem., 2016, 18, 3360-3369.
13 S. G. Elliot, C. Andersen, S. Tolborg, S. Meier, I. Sádaba, A. E.
Daugaard and E. Taarning, RSC Adv., 2017, 7, 985-996.
14 S. G. Elliot, S. Tolborg, I. Sádaba, E. Taarning and S. Meier,
ChemSusChem, 2017, 10, 2990-2996.
15 H. S. Chen, A. Wang, H. Sorek, J. D. Lewis, Y. Román-Leshkov
This work describes kinetic and mechanistic insight into the
by Sn-Beta catalysed methyl lactate process in methanol with
an emphasis of carbohydrate influx into the pathway. Plausible
kinetic differences in the conversion of glucose, fructose and
sucrose can be detected, while methyl lactate formation from
all substrates shows two different kinetic regimes. The
majority of methyl lactate is formed in the slow reaction
regime. This slow regime results from the accumulation of
methyl fructoside (especially in its furanoside form) as a
masked form of substrate carbon for glucose, fructose and
sucrose. Sucrose, known to yield higher methyl lactate levels
than glucose and fructose, also yields higher methyl fructoside
levels. Kinetic insight is practically applied by showing that the
fast methyl fructoside formation and its slow hydrolysis can be
used to produce more than 60% fructoside within few minutes
using a defect-containing Sn-Beta catalyst. In addition, masking
of substrate as methyl fructoside permits modulating process
kinetics by the presence of water. The conversion to methyl
lactate can thus be accelerated by the addition of small
amounts of water that are not prohibitive to catalyst stability.
Such small amounts of water may in some instances be
endemic to the process, if carbohydrate substrate is supplied
as syrup. The findings described herein have implications for
strategies to improve and tailor the methyl lactate process,
specifically in increasing productivity in the presence of water
and for the rapid isomerization of glucose to methyl
fructosides at high temperature in the absence of water.
and A. Bell, ChemistrySelect, 2016, 1, 4167-4172.
16 R. De Clercq, M. Dusselier and B. F. Sels, Green Chem., 2017,
19, 5012-5040.
17 M. Paniagua, S. Saravanamurugan, M. Melian-Rodriguez, J.
A. Melero and A. Riisager, ChemSusChem, 2015,
1094.
8, 1088-
18 S. Saravanamurugan, M. Paniagua, J. A. Melero and A.
Riisager, J. Am. Chem. Soc., 2013, 135, 5246-5249.
19 S. Saravanamurugan, I. Tosi, K. H. Rasmussen, R. E. Jensen, E.
Taarning, S. Meier and A. Riisager, Catal. Sci. Techn., 2017, 7,
2782-2788.
20 S. Saravanamurugan, A. Riisager, E. Taarning and S. Meier,
ChemCatChem, 2016, , 3107-3111.
8
21 Y. Román-Leshkov, M. Moliner, J. A. Labinger and M. E.
Davis, Angew. Chem. Int. Ed., 2010, 49, 8954-8957.
22 R. Bermejo-Deval, R. S. Assary, E. Nikolla, M. Moliner, Y.
Román-Leshkov, S.-J. Hwang, A. Palsdottir, D. Silverman, R. F.
Lobo, L. A. Curtiss and M. E. Davis, Proc. Nat. Acad. Sci.,
2012, 109, 9727-9732.
23 J. Dijkmans, D. Gabriels, M. Dusselier, F. de Clippel, P.
Vanelderen, K. Houthoofd, A. Malfliet, Y. Pontikes and B. F.
Sels, Green Chem., 2013, 15, 2777-2785.
24 N. Rajabbeigi, A. I. Torres, C. M. Lew, B. Elyassi, L. Ren, Z.
Wang, H. Je Cho, W. Fan, P. Daoutidis and M. Tsapatsis,
Chem. Eng. Sci., 2014, 116, 235-242.
Conflicts of interest
There are no conflicts to declare.
25 S. Li, T. Josephson, D. G. Vlachos and S. Caratzoulas, J. Catal.,
2017, 355, 11-16.
26 M. Dusselier, P. Van Wouwe, S. De Smet, R. De Clercq, L.
Verbelen, P. Van Puyvelde, F. E. Du Prez and B. F. Sels, ACS
Catal., 2013, 3, 1786-1800.
27 S. Farah, D. G. Anderson and R. Langer, Adv. Drug Delivery
Rev., 2016, 107, 367-392.
28 S. Tolborg, A. Katerinopoulou, D. D. Falcone, I. Sadaba, C. M.
Osmundsen, R. J. Davis, E. Taarning, P. Fristrup and M. S.
Holm, J. Mat. Chem. A, 2014, 2, 20252-20262.
29 A. Corma, L. T. Nemeth, M. Renz and S. Valencia, Nature,
2001, 412, 423.
Acknowledgements
We gratefully acknowledge funding by the Innovation Fund
Denmark (case number 5150-00023B). 800 MHz NMR spectra
were recorded on the spectrometer of the DTU NMR center.
1
2
3
J. Jow, G. L. Rorrer, M. C. Hawley and D. T. A. Lamport,
Biomass, 1987, 14, 185-194.
H. Li, Z. Fang, J. Luo and S. Yang, Appl. Catal. B: Environ.,
2017, 200, 182-191.
L. Peng, H. Li, L. Lin and K. Chen, Catal. Commun., 2015, 59
10-13.
30 C. Hammond, S. Conrad and I. Hermans, Angew. Chem. Int.
Ed., 2012, 51, 11736-11739.
31 M. Bøjstrup, B. O. Petersen, S. R. Beeren, O. Hindsgaul and S.
Meier, Anal. Chem., 2013, 85, 8802-8808.
,
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 9 of 10
View Article Online
DOI: 10.1039/C8CY00335A
Journal Name
Catalysis Science & Technology
32 B. O. Petersen, O. Hindsgaul and S. Meier, Analyst, 2014, 37 M. Dusselier and B. F. Sels, in Selective Catalysis for
139, 401-406.
33 H. Nguyen, V. Nikolakis and D. G. Vlachos, ACS Catal., 2016,
Renewable Feedstocks and Chemicals, ed. K. M. Nicholas,
Springer International Publishing, Cham, 2014, DOI:
10.1007/128_2014_540, pp. 85-125.
6, 1497-1504.
34 R. Bermejo-Deval, R. Gounder and M. E. Davis, ACS Catal., 38 W. N. P. van der Graaff, C. H. L. Tempelman, G. Li, B. Mezari,
2012, , 2705-2713. N. Kosinov, E. A. Pidko and E. J. M. Hensen, ChemSusChem,
2016, , 3145-3149.
2
35 G. S. Bethell and R. J. Ferrier, Carbohydrate Res., 1973, 31
,
9
69-80.
36 L. J. Heidt and C. B. Purves, J. Am. Chem. Soc., 1938, 60
1206-1210.
39 D. Padovan, S. Tolborg, L. Botti, E. Taarning, I. Sadaba and C.
Hammond, Reaction Chemistry & Engineering, 2018, DOI:
10.1039/C7RE00180K.
,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Catalysis Science & Technology
Page 10 of 10
View Article Online
DOI: 10.1039/C8CY00335A
TOC entry
Strategies to tailor the Sn-Beta catalysed methyl lactate process are identified by kinetic and mechanistic
insight.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 10
Please do not adjust margins