Insights into the Complexity of Heterogeneous Liquid-Phase Catalysis: Case Study on the Cyclization of Citronellal

Article abstract of DOI:10.1021/acscatal.5b02493

Heterogeneously catalyzed liquid-phase reactions are highly complex chemical systems. As the local molecular composition close to the active site is often unknown, sophisticated spectroscopic tools are needed to gain insights on a molecular level. One solution to these challenges is the use of modulation excitation spectroscopy with attenuated total reflection infrared spectroscopy. We use this highly sensitive and selective technique to study the Lewis acid-catalyzed cyclization of citronellal over mesoporous Sn-SBA-15 and microporous Sn-Beta. We find that the reaction mechanism is generally similar for the two materials. However, the confined space at the active site within the zeolite stabilizes the coordination of citronellal to the Sn^IV site and prevents byproduct formation, as well as the reverse reaction due to size-exclusion of the product isopulegol. The use of the Lewis base acetonitrile as solvent reduces the catalytic performance with Sn-SBA-15 drastically, while Sn-Beta remains highly active. Infrared spectra reveal a simultaneous coordination of citronellal and acetonitrile to the tin site in Sn-Beta, whereas in Sn-SBA-15, the more-accessible Sn^IV site leads to much stronger and hence detrimental competitive adsorption. The results obtained in this study indicate that the substrate-catalyst-solvent combination needs to be optimized in order to maximize the performance of solid-liquid reactions.

Full text of DOI:10.1021/acscatal.5b02493

Subscriber access provided by University Libraries, University of Memphis
Article
Insights into the Complexity of Heterogeneous Liquid-
phase Catalysis: Case Study on the Cyclization of Citronellal
Philipp Mueller, Patrick Wolf, and Ive Hermans
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b02493 • Publication Date (Web): 16 Mar 2016
Downloaded from http://pubs.acs.org on March 22, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
ACS Catalysis
1
2
3
4
5
6
7
8
9
Insights into the Complexity of Heterogeneous Liquid-phase Cataly-
sis: Case Study on the Cyclization of Citronellal
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Philipp Müllera, Patrick Wolfa,b, Ive Hermansa,*
a Department of Chemistry & Department of Chemical and Biological Engineering, University of Wisconsin – Madison, 1101 Univer-
sity Avenue, Madison, WI 53706 (USA)
b Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir Prelog Weg 2, 8093 Zurich (Switzerland)
ABSTRACT: Heterogeneously catalyzed liquid-phase reactions are highly complex chemical systems. As the local molecular composition
close to the active site is often unknown, sophisticated spectroscopic tools are needed to gain insights on a molecular level. One solution to
these challenges is the use of Modulation Excitation Spectroscopy with Attenuated Total Reflection Infrared Spectroscopy. We use this highly
sensitive and selective technique to study the Lewis acid-catalyzed cyclization of citronellal over mesoporous Sn-SBA-15 and microporous Sn-
Beta. We find that the reaction mechanism is generally similar with the two materials. However, the confined space at the active site within the
zeolite stabilizes the coordination of citronellal to the SnIV site and prevents by-product formation, as well as the reverse reaction due to size-
exclusion of the product isopulegol. The use of the Lewis base acetonitrile as solvent reduces the catalytic performance with Sn-SBA-15 dras-
tically, while Sn-Beta remains highly active. Infrared spectra reveal a simultaneous coordination of citronellal and acetonitrile to the tin site in
Sn-Beta, whereas in Sn-SBA-15 the better accessible SnIV site leads to much stronger and hence detrimental competitive adsorption. The re-
sults obtained in this study indicate that the substrate-catalyst-solvent combination needs to be optimized in order to maximize the perfor-
mance of solid-liquid reactions.
KEYWORDS: modulation excitation, in situ infrared spectroscopy, confinement, solvent effect, stannosilicate, Sn-Beta
INTRODUCTION
croporous (viz., Sn-Beta and Sn-MFI) and mesoporous catalysts
(viz., Sn-MCM-41 and Sn-SBA-15), Osmundsen et al. observed a
distinct dependency of the activity for the conversion of biomass-
derived substrates on the framework structure.17
In heterogeneous catalysis it is widely accepted that the cata-
lytic performance does not only depend on the precise atomistic
coordination of the active sites, but also on the embedment of these
sites in the support material. For instance, catalysts with narrow
pores of molecular dimension can trigger size-exclusion reactions,1
or induce confinement effects that influence both the activity and
selectivity.2,3 Specifically for liquid-phase reactions, also the solvent
(composition) can affect the overall performance of the catalytic
system (vide infra). All these effects are often encountered in Lewis
acid-catalyzed liquid-phase catalysis. Indeed, solid Lewis acids are
widely used catalysts for various reactions, e.g. in biomass upgrad-
ing,4 selective oxidations/hydrogenations5,6 and the production of
fine chemicals.7 In particular, site-isolated Lewis acids incorporated
in siliceous frameworks such as zeolites or (mesoporous) silicas
have shown great potential, yet they are also notoriously complex.8-
As briefly mentioned above, a solvent can also affect the per-
formance of a catalyst, e.g. by disturbing the substrate adsorption,18
preferentially solubilizing certain molecules,19 or altering the active
site.20 For instance, Boronat et al. studied solvent effects in the cy-
clization of citronellal to (-)-isopulegol, an important step in the
artificial production of the flavoring agent (-)-menthol (see
Scheme 1) over Sn-Beta.21,22
12
Especially, SnIV incorporated in the BEA framework (Sn-Beta)
obtained significant attention due to its exceptional activity and
stability.13 Two different types of active sites have been proposed, a
four-fold framework-incorporated SnIV species (so-called closed
site) and a partially hydrolyzed (so-called open) site, with tin three-
fold bound to the zeolite framework.14,15
Scheme 1. Cyclization of (R)-(+)-citronellal to (-)-isopulegol,
followed by the hydrogenation to (-)-menthol.
Not only does the nature of the active site play a role, but also
its local molecular environment seems to influence the reactivity of
SnIV-containing materials.16 Testing SnIV-incorporated in mi-
Adding the Lewis base acetonitrile to citronellal improves both
activity and selectivity towards (-)-isopulegol. The authors hypoth-
esize that both acetonitrile and citronellal adsorb to the SnIV-site,
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 11
which results in enhanced stereoselectivity, due to a confined space
siliceous analogues Si-Beta35 and SBA-1536 were prepared. Charac-
terization of all synthesized materials is provided in the supporting
information (Figures S1, S2 and Table S1).
1
2
3
4
5
6
7
8
around the catalytic center. Moreover, product desorption from the
active site and diffusion inside the pores would be further facilitated
by the presence of acetonitrile. Other solvents, such as longer chain
nitriles and alcohols also affect the reaction. However, a general
understanding of these effects is lacking, as a correlation between
the catalytic performance and a single solvent characteristic, such as
the dielectric constant or the molecular diameter, is not sufficient
to capture the complexity of such catalytic systems.
The catalytic performance of these materials in the cyclization
of citronellal was determined under batch reaction conditions with
toluene and acetonitrile as solvent. Both solvents are widely used
for carbonyl-ene reactions and strongly differ in polarity and size.37
The diastereoselectivity was constant for all materials and solvents
throughout the course of the reaction (see Figures S3 and S4).
Overall higher activity was observed in toluene, where small
amounts of products are already formed without a catalyst. In the
presence of weakly Brønsted acidic SBA-15, both activity and selec-
tivity increased. Significantly better catalytic performance was ob-
served with the Lewis acidic stannosilicates, of which Sn-Beta
yielded highest activity and diastereoselectivity, followed by Sn-
SBA-15 and Sn-SiO2. The achieved diastereoselectivity for Sn-Beta
is slightly lower than reported values,37 which could be explained by
the post-synthetic preparation of the material. This leads to minor
amounts of remaining silanol groups, possessing Brønsted acidity
and slightly different active sites than in hydrothermal Sn-Beta.38
An analysis of the Arrhenius plots (see Figures S6 and S7) indicates
mass transfer to be influencing the reaction.
9
A sensitive approach towards the understanding of heterogene-
ous liquid phase reactions, in particular on a molecular level, is
vibrational spectroscopy.23,24 For the cyclization of citronellal, an
experimental FT-IR study has been performed, monitoring the
disappearance of the characteristic vibrations of pre-adsorbed cit-
ronellal on Sn-Beta.15 Two bands at 1729 cm-1 and 1715 cm-1 were
observed and assigned to weakly and strongly adsorbed carbonyl
groups, respectively. The presence of a C=C stretch vibration at
1645 cm-1 indicated the simultaneous formation of the isopulegol
product. While valuable information can be provided by such an
approach, many of the described effects are neglected, for instance
due to the absence of solvent. Conclusive experimental studies
under relevant conditions that address the mechanism on a mo-
lecular level are still missing.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Batch reactivity data for the cyclization of
citronellal over different catalysts.a
In situ spectroscopic studies would in principle be able to pro-
vide this information. However, a poor signal-to-noise ratio as a
consequence of low concentrations of active species is a frequently
encountered problem that limits the applicability and interpreta-
Catalyst
Solvent
Conv.
[%]b
TOF
[h-1]c
Sel.
Sel.dia
[%]e
[%]d
-
Toluene
4
-
50
67
92
90
40
89
-
67
66
64
65
70
70
-
tion of in situ spectroscopic studies. ꢀOne highly sensitive and
selective method to study reaction mechanisms at the solid-liquid
interface is the application of a modulation excitation spectroscopy
(MES) approach in combination with in situ attenuated total re-
flection infrared (MES-ATR-IR) and phase sensitive detection
(PSD).25 While the use of conditions that are close to actual reac-
tion conditions are desirable for a spectroscopic study, they also
impose significant complications. Indeed, in a solid-liquid reaction,
multiple phenomena such as adsorption, (by-)product formation
and desorption of various species occur simultaneously. It is hence
difficult to distinguish between species that are actually involved in
the surface reaction, i.e. active species, and those that just adsorb to
the solid-liquid interface and do not react further, i.e. spectator
species. PSD makes use of a mathematical treatment that is ex-
plained below, and transforms time-dependent spectra into phase-
resolved spectra.26 This transformation increases the signal-to-noise
ratio, allows for differentiation between active and spectator species
and simplifies the micro-kinetic analysis. This technique is hence
ideal to shed light on the mechanisms involved in complex multi-
phase reactions.27,28
SBA-15
Sn-SBA-15
Sn-SiO2
Si-Beta
Sn-Beta
-
Toluene
27
89
60
9
-
Toluene
115
42
-
Toluene
Toluene
Toluene
66
-
185
-
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
SBA-15
Sn-SBA-15
Sn-SiO2
Si-Beta
Sn-Beta
-
-
-
-
15
7
12
6
50
0
65
0
-
-
-
-
63
127
88
77
aReaction conditions: 50 mg of catalyst, 5 mL of 0.5 M citronellal so-
lution in solvent at 333 K, reaction time 4 hr.
bConversion = (moles of citronellal initial - moles of citronellal after
4 hr)/moles of citronellal initial
cTurnover frequency (TOF) = moles of citronellal converted/moles
of Sn after 1 hr
In the present work, we demonstrate the use of MES-ATR-IR
in the case of the cyclization of citronellal, a model reaction of con-
temporary interest. Correlating macroscopic differences in the
reactivity with a molecular understanding based on the information
obtained from the solid-liquid interface allows us to draw conclu-
sions on pore-size and solvent effects.
dSelectivity is defined as the moles of all produced isomers of isopu-
legol divided by the moles of converted citronellal
eDiastereoselectivity is defined as the moles of produced
(-)-isopulgeol divided by the moles of all isomers of isopulegol pro-
duced.
With acetonitrile as the solvent, the blank reaction and the reac-
tion with the pure silica materials, SBA-15 and Si-Beta, did not
show any activity. Sn-Beta performed similarly in terms of selectivi-
ty compared to the reaction in toluene but was slightly less active.
Conversely, Sn-SBA-15 is one order of magnitude less active when
RESULTS
Materials & Batch reactivity
Three different SnIV-containing catalysts were synthesized fol-
lowing known procedures: an amorphous Sn-SiO2,29 a mesoporous
Sn-SBA-15,30,31 and a microporous Sn-Beta.32-34 In addition, the
2
ACS Paragon Plus Environment

Page 3 of 11
ACS Catalysis
employed in acetonitrile and only reaches a selectivity of 50%, In order to investigate the catalytic cycle, the adsorption and
1
2
3
4
5
6
7
8
compared to 92% in toluene. One possible explanation for this is
the competitive adsorption of acetonitrile and citronellal (both
Lewis bases) to the SnIV-site. However, it is not readily clear why
the activity only drops by 30% in the case of Sn-Beta, whereas it
drops significantly more (about 90%) for Sn-SBA-15. These obser-
vations prompted us to perform a detailed in situ spectroscopic
investigation. Due to the significantly lower activity of Sn-SiO2, this
material was not considered further.
reaction of citronellal was studied by switching between a solution
of 20 mM citronellal in toluene and pure toluene. Figure 1a shows a
projection of the TD of three averaged periods for the modulation
over SBA-15 at 333 K. In the TD one major signal appears at 1727
cm-1, which originates from the ν(C=O) vibration of liquid-phase
citronellal (see Figure S8 for IR spectra of pure substances in tolu-
ene). At 1605 cm-1 the aromatic C-C vibration of the toluene sol-
vent appears with a small negative absorbance band, indicating the
displacement of toluene from the solid-liquid interface by citron-
ellal. The transformation to PD allows for a more detailed analysis
of the micro-kinetics and a more sensitive and selective detection of
signals as only active species appear in the PD spectra. Further-
more, the clear assignment of an absolute phase delay to the corre-
sponding maximum absorbance of a vibration in the TD makes a
comparison between different experiments easier and allows for a
semi-quantitative analysis. Figure 1c shows the PD spectra corre-
sponding to the TD spectra from Figure 1a. The citronellal C=O
peak appears with an absolute phase delay (viz. the maximum in
PD) of 30°, which also corresponds to the minimum of the toluene
band at the same angle. Furthermore, it is slightly skewed to lower
wavenumbers, indicating only weak interactions with the solid
catalyst. Complete citronellal desorption occurs at a phase angle of
220°, which can be seen by a maximum in the toluene peak and a
minimum in the citronellal peak.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Citronellal adsorption monitored with ATR-IR-MES
A suitable method to study adsorption/desorption dynamics, as
well as reaction mechanisms at solid-liquid interfaces is ATR-IR
spectroscopy combined with MES.26,40,41 The latter follows an anal-
ogous principle as a lock-in amplifier by analyzing the response of a
system to a periodic external stimulation, such as a change in tem-
perature, radiation, pH, or as applied in the present work, reagent
concentration. This approach is not only able to decrease the noise
level to less than 10-5 absorbance units,41 but also allows for the
discrimination between active and spectator species. The applica-
tion of PSD [eq. (1)] to the time-domain (TD) spectra transforms
them into the phase-domain (PD), which further enhances the
sensitivity and enables the extraction of (semi)-quantitative micro-
kinetic data.
The incorporation of SnIV into the SBA-15 framework leads to
different observations (Figure 1d). Next to the liquid-phase citron-
ellal peak, a shoulder at 1712 cm-1 stemming from citronellal C=O
groups interacting with the SnIV site appears at 50°, i.e. 20° (or 6
seconds) delayed compared to the liquid-phase peak. At 70°, a peak
at 1643 cm-1, assigned to the ν(C=C) vibration of the primary dou-
ble bond in the product isopulegol, indicates that the catalytic reac-
tion is taking place. Simultaneously, a red-shifted shoulder next to
the liquid-phase peak appears, due to an interaction of isopulegol
with either a SnIV site and/or the silica framework.29,42
In equation (1), T is the length of one period, ω is the stimulation
frequency, ϕk is the demodulation phase angle, ϕkdelay is the phase
delay, k is the demodulation index (k=1 throughout this study),
A(t) is the active species response in the TD and Ak the response in
the PD. The transformation to the PD leads to a dependence of the
vibrational signals on the phase angle ϕ (adsorption from 0° to 180°
desorption from 180° to 360°) instead of the time (60s adsorption;
60s desorption). In the following PD spectra, 36 phase-resolved
spectra are shown with increments of 10° (corresponding to 3 se-
conds in the TD) which represent one whole period in the PD. In
an ideal experiment, i.e. without diffusion limitation and immediate
exchange of the cell volume (i.e. zero dead volume), the amplitude
of the active species response, which corresponds to absorbance in
the time-domain, would be 0 at 0° and 180°. A shift to larger phase
angles can be attributed to hindered transport to the active site or,
in the case of a reaction, to consecutive reaction steps. A detailed
analysis of the spectra at various phase angles therefore provides
valuable information on adsorption and reaction dynamics as will
be demonstrated below.
For Si-Beta almost identical PD spectra were obtained as for
SBA-15, but at larger phase angles (Figure 1e), indicating a longer
time for citronellal to diffuse in the pores. When SnIV is incorpo-
rated into the Beta framework, the peak at 1712 cm-1 becomes sig-
nificantly more intense than the liquid-phase citronellal peak at
1727 cm-1. The more intense peak at 1643 cm-1 indicates increased
productivity, in agreement with the batch experiments (see Table
1). Note that the C=O vibration of citronellal coordinated to SnIV
appears at the exact same spectral position (1712 cm-1) for both
SnIV-containing materials, i.e. both catalysts perturb the C=O
stretch vibration of citronellal to the same extent under reaction
conditions. Even though the metal content in Sn-Beta is ca. three
times lower than in Sn-SBA-15, the absolute coverage of citronellal
coordinated to SnIV is clearly much higher in case of the zeolite.
This observation suggests confinement effects as a likely reason for
the increased productivity, rather than the stronger Lewis acidity of
Sn-Beta (Ref. 16 and Figure. S10 for liquid-phase d3-acetonitrile
adsorption). The confined space inside the zeolite pore of molecu-
lar dimension (6.7 Å43) leads to more negative adsorption en-
thalpies due to van der Waals interactions44 which might additional-
ly stabilize the coordination of citronellal, resulting in increased
coverage of the tin site.
To perform an ATR-IR MES experiment, the catalyst needs to
be deposited as a thin film (ca. 15 µm) onto an ATR crystal. The
penetration depth of an ATR experiment depends on the refractive
indices of the sample and the crystal and is also wavelength de-
pendent.25 A rough estimation with the refractive index of SiO2
(1.45) yields a penetration depth of 1-2 µm at 1000 cm-1. Hence,
mainly the pores within the catalyst particles are probed. The con-
ditions for the spectroscopic experiments were chosen so that diffu-
sion effects within the ATR cell are minimized and differences in
phase delays can be attributed to diffusion within the catalyst as
well as consecutive reaction steps (see Ref. 46 and experimental
section for more details).
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. a) Time-domain spectra for the modulation of 20 mM citronellal in toluene at 333 K for SBA-15, b) Schematic representa-
tion of the most important molecular structures and their assignment, Phase-domain transformed spectra for c) SBA-15, d) Sn-SBA-
15, e) Si-Beta and f) Sn-Beta.
4
ACS Paragon Plus Environment

Page 5 of 11
ACS Catalysis
Interestingly, the red-shifted product shoulder (ca. 1625 cm-1),
lower wavenumbers than the Si-OH peaks. This was however not
observed. Hence, the present study does not confirm the direct
participation of a Sn-OH group in the reaction mechanism for both
Sn-Beta and Sn-SBA-15. Nevertheless, as many different species
(isopulegol, Si-OH, Sn-OH, water impurities) possibly show a
signal in this spectral region, and the concentration of Sn-OH spe-
cies is supposed to be rather small, the absence of evidence cannot
be interpreted as the evidence for absence. Testing another Beta
sample with a ten times higher tin loading resulted in similar spec-
tra and also no indication of a distortion of Sn-OH vibrations when
adsorbing citronellal (see Figure S26). Yet, the above findings ex-
emplify the consecutive steps needed for this reaction to occur: (1)
citronellal coordinates to the SnIV site, (2) citronellal is converted
to isopulegol inside the pores of Sn-Beta, (3) isopulegol interacts
with either the SnIV site and/or the pore walls until (4) isopulegol
diffuses out of the pore (vide infra). Additionally, the overall larger
phase delays found for Sn-Beta compared to Sn-SBA-15 suggest
diffusion limitations caused by the smaller pores in Sn-Beta.46 As
citronellal has a small kinetic diameter and is flexible (viz. it can
freely rotate around its C-C bonds), it can diffuse more easily inside
the zeolite pore channels and find its way to a tin site. The cycliza-
tion leads to a more rigid product with a larger kinetic diameter,
inducing diffusion limitations inside the micropores. However, as
the SnIV sites in Sn-Beta are embedded in a confined space stabiliz-
ing the coordination of citronellal, it still outperforms Sn-SBA-15.
1
i.e. adsorbed isopulegol, is also more intense compared to Sn-SBA-
15 and reaches a maximum at 80°, while the liquid-phase product
peak reaches a maximum at 100°. This implies that the shoulder
most likely originates from an isopulegol species inside the zeolite
pore that might relax to a different conformation once it leaves the
pore. Also the ν(OH) vibration of isopulegol appears as a broad
signal centered at 3250 cm-1, corroborating an interaction of isopu-
legol with the zeolite (Figure S25).
2
3
4
5
6
7
8
9
Under vacuum conditions, isolated Si-OH and Sn-OH bands
show a signal at 3740 and 3664 cm-1 respectively (see Figure S9 and
Ref. 3). Surprisingly, no significant change in either of these bands
could be observed for both of the SnIV-containing catalysts upon
the adsorption of citronellal from the liquid phase (Figures S14,
S15, S24, S25). However, it is likely that the presence of a solvent
shifts these signals to lower wavenumbers, broadens them and
weakens their intensity.45 The adsorption of citronellal on Sn-Beta
leads to two negative bands, one at 3600 cm-1 and a smaller one at
3700 cm-1, both bands are present in the low-defect Si-Beta as well
but with lower intensity (Figures S17 and S18). We therefore as-
sign the band at 3700 cm-1 to isolated Si-OH groups interacting
with citronellal in Beta zeolites. Dealuminated Beta shows a distinct
negative band at 3600 cm-1 at the same phase delay as the appear-
ance of citronellal, which we therefore assign to an interaction of
citronellal with silanol groups (Figures S19-S22). For an interac-
tion of citronellal with a Sn-OH group, we would expect a signal at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Phase-domain spectra of modulation experiments with 20 mM (-)-isopulegol in a solution of 20 mM of citronellal in tolu-
ene at 333 K for Sn-SBA-15 a) and Sn-Beta b).
isopulegol, indicating a production of citronellal from (-)-
isopulegol, i.e. the reverse reaction. However, compared to the
reaction from citronellal a considerably smaller red-shifted shoul-
Competitive adsorption of (-)-isopulegol and citronellal
In order to get insights into the competitive adsorption behav-
der next to the liquid-phase peak appears. In a further batch exper-
ior of substrate and product, the concentration of (-)-isopulegol
iment (-)-isopulegol was used as substrate over Sn-SBA-15 and
was modulated in presence of a constant citronellal concentration.
only little conversion was observed (see Figure S5), suggesting that
In the case of SBA-15 and Si-Beta, a relatively sharp ν(C=C) vibra-
this is only a minor effect on a macroscopic scale. When (-)-
tion is visible at 1643 cm-1, corresponding to liquid (-)-isopulegol
isopulegol is adsorbed on Sn-Beta (Figure 2b), its C=C stretch
(Figures S27-S30). For Sn-SBA-15 (Figure 2a), the spectra are
vibration appears at 1643 cm-1 and no red-shifted shoulder can be
significantly different. Other than expected, the liquid-phase cit-
observed. In addition, a small and broad feature at 1712 cm-1 is
ronellal peak (1727 cm-1) appears. The absolute phase delay for this
visible with an absolute phase delay of 200°. This strongly suggests
peak is 40°, which is 10° delayed from the C=C vibration of (-)-
the adsorption of isopulegol to the few accessible external or pore
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 11
mouth Lewis acid sites which leads to a distortion of the citronellal which leads to a significantly stronger decrease in activity for Sn-
1
2
3
4
5
6
7
8
signal. In the TD spectra (Figure S31), the change in intensity is
hardly visible, which further confirms that this is only a minor effect
and demonstrates the increased sensitivity when applying PSD.
The presented data demonstrate that (-)-isopulegol does not enter
the zeolite pore and thus can only compete for the active site if it is
formed inside the pores. Consequently the reverse reaction is hin-
dered and a higher productivity is obtained. This finding is further
confirmed by a batch experiment, in which 0.5 M (-)-isopulegol
was added to the reaction mixture (Table S2 entries 10, 11, 22, 23),
SBA-15 than Sn-Beta, due to this size-exclusion effect.
The main by-products of the studied reaction are isopulegol
ethers formed through condensation reactions.47 As such an acid-
catalyzed bimolecular reaction proceeds through a bulky transition
state, it is highly unlikely to occur in the small pores of the zeolite,
where most of the Lewis and Brønsted acid sites are located. Sn-
SBA-15 on the other hand possesses readily accessible SnIV and
silanol sites and can catalyze the reverse reaction, as well as by-
product formation.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Phase-domain spectra for the modulation of 20 mM citronellal in a solution of 20 mM acetonitrile in toluene (top) and 60
mM acetonitrile in toluene (bottom) for Sn-SBA-15 a) and c) and Sn-Beta b) and d).
Competitive adsorption of acetonitrile and citronellal
the concentration of citronellal was modulated in presence of a
constant acetonitrile concentration. As expected from batch exper-
iments, SBA-15 is hardly active in the presence of acetonitrile. Only
a liquid-phase citronellal signal appears at 1727 cm-1 and a small
peak around 1640 cm-1, assigned to minor production of isopulegol,
is visible for an acetonitrile/citronellal molar ratio of 1:1. At a ratio
of 3:1, the minor isopulegol signal completely disappears (Figures
S33-S36). No distortion of the ν(CN) region (ca. 2263 cm-1) was
To rationalize the vast difference in catalytic performance of
Sn-SBA-15 and Sn-Beta in presence of acetonitrile (Table 1), a
competitive adsorption study was performed. As the physicochem-
ical properties of acetonitrile (higher polarity, lower refractive in-
dex than toluene) impede sensitive ATR measurements,23,25 we
decided to assess the influence of acetonitrile by adding it to tolu-
ene in different acetonitrile/citronellal ratios. In these experiments,
6
ACS Paragon Plus Environment

Page 7 of 11
ACS Catalysis
observed for either of the two ratios. The adsorption on Si-Beta when pure acetonitrile is used as solvent, which results in poor
1
2
3
4
5
6
7
8
results in similar spectra, only showing a small product peak at 1643
cm-1 and the liquid-phase citronellal peak. However, when the ace-
tonitrile content was increased to a 3:1 ratio with respect to citron-
ellal, additionally a ν(CN) vibration with negative absorbance was
observed, as well as an even smaller product peak.
activity (Table 1).
For Sn-Beta, an activated C=O peak is visible at 1715 cm-1 for
an acetonitrile/citronellal ratio of 1:1 and 3:1 (Figures 3b and 3d).
This peak is shifted by 3 cm-1 compared to pure toluene as solvent,
indicating a minor distortion of the citronellal coordination by
acetonitrile. Nevertheless, inside the zeolite pores, citronellal is able
to displace acetonitrile as can be seen by the negative ν(CN) band
at 2263 cm-1. Furthermore, the absolute phase delay of the (liquid
phase) product peak at 1646 cm-1 is 70°, which is 30° less than in
pure toluene. A 3:1 acetonitrile/citronellal ratio also leads to a fast-
er transport and/or adsorption of citronellal to the SnIV site, which
is reflected in the 40° absolute phase delay of the peak at 1715 cm-1,
i.e. 10° less than in pure toluene.
In the case of Sn-SBA-15 (Figure 3a), the addition of acetoni-
trile in a 1:1 ratio leads to the complete disappearance of the shoul-
der at 1712 cm-1 observed in pure toluene. No negative peak in the
ν(CN) region can be observed which means that almost no ace-
tonitrile is displaced from the solid-liquid interface by citronellal. A
small product band at 1646 cm-1 reveals still minor activity. By fur-
ther increasing the acetonitrile/citronellal ratio to 3:1 (Figure 3c)
this band gets even smaller and still no perturbed ν(CN) peak is
visible. Additionally, the small product peak shows a maximum at a
phase angle of 100° showing an increased induction time until a
steady product concentration is reached. Even though only a
shoulder is observed in the liquid-phase CD3CN adsorption exper-
iment (Fig S10), indicating weak Lewis acidity, this competitive
study implies that acetonitrile binds more strongly to the SnIV sites
in Sn-SBA-15 than citronellal. This effect will be further amplified
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These findings demonstrate that the addition of acetonitrile leads
to facilitated transport of citronellal and isopulegol inside the zeo-
lite pores, possibly due to the smaller molecular size of the additive
and competitive adsorption/desorption. The coordination of cit-
ronellal is also stabilized in the presence of the Lewis base acetoni-
trile. For the mesoporous Sn-SBA-15, however, the SnIV sites are
more accessible, which leads to detrimental competitive adsorp-
tion.
Scheme 2. Representation of effects occurring during the cyclization of citronellal over Sn-SBA-15 and Sn-Beta in the presence of
acetonitrile and toluene.
7
ACS Paragon Plus Environment

ACS Catalysis
CONCLUSIONS
Page 8 of 11
DISCUSSION
1
On the timescale of the ATR-IR MES experiments, no signifi-
cant deactivation was observed. Hence, the obtained spectroscopic
data can be correlated with the reactivity of the initial phase of the
batch reactions. Based on our experimental observations, the dif-
ferent effects influencing the cyclization of citronellal over Sn-SBA-
15 and Sn-Beta are thus summarized in Scheme 2. In the case of Sn-
SBA-15 in toluene, citronellal can easily diffuse to the SnIV site. The
citronellal coverage on SnIV sites is relatively low, as solvent mole-
cules can effectively interfere with the coordinated citronellal. Once
the product is formed, it can easily diffuse away, but it can also re-
coordinate to the site and undergo the reverse reaction, or form
undesired by-products. The size-restricted micropore structure of
the Beta zeolite increases the coverage of citronellal on SnIV sites.
Confinement effects, as indicated by the same spectral position of
the ν(C=O) vibration but a vastly different coverage for the two
materials, most likely lead to a more stabilized coordination which
increases the productivity. The micropore structure hinders the
product from diffusing out of the pores, but also reduces by-
product formation and the reverse reaction becomes unlikely as the
product cannot re-enter the pore. When the Lewis base acetonitrile
is used as solvent, the higher solvent concentration of the smaller
molecule leads to a significant loss in activity. Our MES-ATR-IR
study with 1:1 and 3:1 acetonitrile/citronellal ratios suggests a
stronger interference of the solvent with citronellal coordinated to
the SnIV site as a possible explanation. On the other hand, a simul-
taneous coordination of citronellal and acetonitrile to the SnIV site
inside the micropore of Sn-Beta induces better diastereoselectivity.
Similar to a bulky ligand in a homogeneous catalyst,48,49 acetonitrile
reduces the space around the active site and hence the formation of
(-)-isopulegol is favored. The latter hypothesis was suggested be-
fore, but could never be observed experimentally.21 Finally, acetoni-
trile leads to easier diffusion of both reactant and product in the
micropores, as indicated by smaller absolute phase delays, which
likely enhances the overall reactivity. The amorphous Sn-SiO2 pre-
sents the other extreme case as the active SnIV sites are present at
the external surface where the solvent can easily interfere with the
coordinated citronellal. According to the liquid-phase acetonitrile
adsorption (Figure S10), Sn-SiO2 was found to be more acidic than
Sn-SBA-15, but it is still less active. Hence, a plausible explanation
for the difference in activity could be the much smaller total mi-
cropore volume of Sn-SiO2 compared to Sn-SBA-15 (Table S1), i.e.
some of the SnIV sites in Sn-SBA-15 are present in a microporous
environment which might explain the higher activity. A similar
effect has also been found for the conversion of tetrose sugars over
SnIV-containing materials, where Sn-SBA-15 showed a different
temperature-dependency than the purely mesoporous Sn-MCM-
41.2 However, this is just a hypothesis and needs further experi-
mental validation.
In conclusion, we studied the cyclization of citronellal as a
case study to get insights into solid Lewis acid catalyzed liquid-
phase reactions. A combination of batch experiments and ATR-IR-
MES spectroscopic investigations revealed important details on
pore size and solvent effects. The reaction mechanism over Sn-
SBA-15 and Sn-Beta could be followed spectroscopically and in-
volves the coordination of citronellal with its carbonyl group to the
SnIV site, followed by the cyclization step and product desorption.
The coordination is amplified in Sn-Beta because of the stronger
Lewis acidity and confinement effects inside the micropore. As the
product isopulegol is too big to enter the pores of the Sn-Beta zeo-
lite, by-product formation and the reverse reaction are hindered
due to size exclusion resulting in a better catalytic performance. In
Sn-SBA-15 on the other hand, the SnIV sites are readily accessible
for substrate, product and solvent, resulting in a lower overall activi-
ty and selectivity. The addition of the Lewis base acetonitrile leads
to a stronger loss in catalyst performance for Sn-SBA-15 due to
competition for the Lewis acid site. This effect is not as pro-
nounced in the microporous Sn-Beta because the local acetonitrile
concentration is lower as citronellal is able to displace acetonitrile
from the active site. Additionally, the simultaneous coordination of
citronellal and acetonitrile enhances stereoselectivity as it benefi-
cially reduces the space around the active site and accelerates sub-
strate and product diffusion inside the micropores. As shown with
these examples, confinement and solvent effects play an important
role in understanding why certain catalytic systems are highly ac-
tive. As the local molecular composition close to the active site is
often not known and the reaction mechanisms are highly complex,
a sensitive spectroscopic technique is crucial to obtain insights on a
molecular level. ATR-IR spectroscopy in combination with MES is
an ideal tool in that regard and can be applied to many other rele-
vant heterogeneously catalyzed reactions as well.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL DETAILS
Materials synthesis
The mesoporous SBA-15 material was synthesized according to
a literature procedure:36 3.5 g of Pluronic P123 was fully dissolved
in 62.8 g of water and 2 g of HCl (37%) at 308 K. After the addition
of 7.44 g tetraethylorthosilicate (TEOS), the solution was stirred
for a day. Then, the mixture was hydrothermally treated in a Tef-
lon-lined autoclave for 42 hours at 373 K. The obtained SBA-15
was washed with 5 L of water, dried in an oven overnight at 373 K
and calcined in air at 823 K for 6 h.
The mesoporous Sn-SBA-15 material was synthesized using a
method reported by Selvaraj et al.30,31, 4 g of Pluronic P123 was
stirred with 25 mL of water to yield a clear solution. Afterwards, a
dilute HCl solution was prepared by mixing 12.8 g HCl (37%) with
144.8 g of water, added to the solution, and stirred for 1 h. Then, 9
g of TEOS and 3.06 g of SnCl4 were added and stirred for 24 h at
313 K. The solid products were recovered by filtration, washed
several times with water and dried overnight at 373 K. Finally, the
sample was calcined in air at 823 K for 6 h.
As shown with the illustrative examples above, the Lewis acidi-
ty, even when measured under in situ conditions, is not a sufficient
descriptor for the reactivity of stannosilicates. It is therefore highly
important to gain insights into the molecular composition near the
active site to understand complex heterogeneous liquid-phase ca-
talysis. The use of MES-ATR-IR provides a unique tool to probe
dynamics at the solid-liquid interface with high sensitivity.
Pure Si-Beta zeolite was prepared following a literature proce-
dure.35 To 25.5g tetraethylammonium hydroxide solution (35 wt%,
SACHEM) 23g tetraethyl orthosilicate (TEOS, 99% Sigma Al-
drich) was added. The reaction mixture was stirred under ambient
conditions allowing ethanol and water to evaporate until the de-
sired final composition was obtained. After adding 2.4g of hydro-
8
ACS Paragon Plus Environment

Page 9 of 11
ACS Catalysis
fluoric acid (HF, 48 wt%, Sigma Aldrich) dropwise and homoge-
tor. Typically 32 scans with a resolution of 4 cm-1 were co-added to
1
2
3
4
5
6
7
8
nizing the synthesis gel with a PTFE spatula, the thick paste was
transferred into a 45 mL stainless steel autoclave equipped with a
PTFE liner. Hydrothermal crystallization was carried out at 140 °C
and 60 rpm for 7 days. The final product was filtered, washed with
water and acetone and dried at 120 °C over night. To remove the
organics the sample was calcined at 853 K for 6h.
give one spectrum. The switching between the two liquid reservoirs
was accomplished through air-actuated valves (Parker) and modu-
lation frequencies of 10-50 mHz. The ATR accessory was home-
made and made use of a 45° ZnSe crystal. Measurements with a
bare crystal showed a total exchange of the cell volume within less
than 2 seconds under typical conditions. The liquids were drawn by
a peristaltic pump (Ismatec) at flow rates of 3-25 ml/min. See Fig-
ure S45 for a schematic of the setup. Absorbance is given in arbi-
trary units for all spectra in the time- and phase-domain.
The Sn-Beta zeolite was prepared through a post-synthetic
method reported elsewhere.32-34 Dealumination of the parent com-
mercial Al-Beta zeolite (Zeolyst, SiO2/Al2O3=300) was done by
acid leaching (13 M HNO3, 20 mL g-1, 373 K, 20 h). Tin was in-
corporated via solid-solid ion-exchange by grinding dealuminated
Beta with the appropriate amount of the tin(II) acetate precursor
followed by subsequent heat treatments under N2 and air for 3 h
each at 823 K.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information. Additional batch reactivity data, IR
spectra, estimation of rate of diffusion, schematic of setup and ex-
planation of MES experiment.
The amorphous Sn-SiO2 material was synthesized via a xerogel
method.29 7 g of TEOS was diluted with 8.7 g of water. 1.9 g of a 0.1
M HCl solution was added and the resulting solution was stirred
for 2 hours. Then, a solution of 0.2 g of SnCl4 ·5 H2O in 2 g of water
was prepared and added dropwise to the first solution. After anoth-
er hour of stirring, a 40 wt% tetrapropylammonium hydroxide solu-
tion was added dropwise until the gel hardened. The resulting gel
was dried at 373 K overnight and calcined in air at 853 K for 6 h.
AUTHOR INFORMATION
Corresponding Author
*E-mail: hermans@chem.wisc.edu
ACKNOWLEDGEMENTS
The authors thank Sam Burt for help with XRD and Connor Firth
for assistance with ICP-OES measurements. Financial support
from the University of Wisconsin – Madison, the Wisconsin Alum-
ni Research Foundation (WARF), the US Army Research Office
under grant W911NF-15-1-0134 and the Swiss National Science
Foundation (grant 200021_146661) is kindly acknowledged.
Ex Situ Characterization
The different materials were analyzed with various techniques:
Si, Al and Sn contents were determined by ICP-OES after digesting
the solids in HF. The small-angle powder XRD-patterns for cal-
cined samples were recorded under ambient conditions on a
Bruker D8 advance diffractometer with Cu Kα radiation (λ=1.5406
Å). N2-physisorption measurements were performed on a Mi-
cromeritics 3flex apparatus. The samples were degassed prior to
measurement for 4 hours at 373 K. Adsorption isotherms were
collected at 77 K and analyzed using BET, t-plot and DFT meth-
ods. DRUV-Vis analysis was performed with a Maya2000 Pro spec-
trometer (Ocean Optics) equipped with a deuterium/halogen light
source (DH-200-BAL from Mikropack) using BaSO4 as matrix.
ABBREVIATIONS
ATR-IR, Attenuated Total Reflection Infrared; FT-IR, Fourier
Transform Infrared; MES, Modulation Excitation Spectroscopy;
PD, Phase-Domain; PSD, Phase Sensitive Detection; TD, Time-
Domain.
Batch reactivity
Batch reactions were performed in 10 mL thick-walled glass re-
actors sealed with PTFE caps under vigorous stirring at 600 rpm
and were heated in a temperature controlled oil bath. Typically, the
reactions were performed at 333 K with 50 mg of catalyst, and 5 ml
of a 0.5 M citronellal solution for 4 hours. Biphenyl was used as an
internal standard. Samples were taken periodically and analyzed
using an HP6890 gas chromatograph (Agilent) equipped with an
HP-5 column. By-products were mainly condensation products.
REFERENCES
(1)
Weisz, P. B.; Frilette, V. J.; Maatman, R. W.; Mower, E. B. J.
Catal. 1962, 1, 307-312.
(2)
De Clercq, R.; Dusselier, M.; Christiaens, C.; Dijkmans, J.;
Iacobescu, R. I.; Pontikes, Y.; Sels, B. F. ACS Catal. 2015, 5, 5803-5811.
(3)
Hermans, I. ChemCatChem 2015, 7, 3270-3278.
(4) Zhou, C. H.; Xia, X.; Lin, C. X.; Tong, D. S.; Beltramini, J.
Chem. Soc. Rev. 2011, 40, 5588-5617.
Conrad, S.; Verel, R.; Hammond, C.; Wolf, P.; Göltl, F.;
(5)
Corma, A.; Domine, M. E.; Valencia, S. J. Catal. 2003, 215,
ATR-IR & MES
294-304.
(6)
The preparation of a catalyst layer was done by adding 10 mL of
a 1:25 times diluted Collodion solution (5%, in 75:25 diethyl
ether:ethanol, Fisher Scientific) to 100 mg of catalyst. The resulting
suspension was stirred for 1 hour and then sonicated for 5 minutes.
1 mL of this solution was deposited onto the ZnSe crystal and put
in an oven to dry at 353 K for 30 minutes to yield ca. 9 mg of depos-
ited catalyst. The influence of the added Collodion on reactivity
was tested in a batch reaction and was only minor (see Table S2).
All ATR-IR measurements were performed on a Bruker Vertex 70
spectrometer equipped with a liquid nitrogen cooled MCT detec-
Qi, G. S.; Yang, R. T.; Chang, R. Appl. Catal. B-Environ. 2004,
51, 93-106.
(7)
(8)
Corma, A.; Garcia, H. Chem. Rev. 2003, 103, 4307-4365.
Holm, M. S.; Saravanamurugan, S.; Taarning, E. Science 2010,
328, 602-605.
(9)
Dapsens, P. Y.; Mondelli, C.; Perez-Ramirez, J. Chem. Soc. Rev.
2015, 44, 7025-7043.
(10)
Román-Leshkov, Y.; Moliner, M.; Labinger, J. A.; Davis, M. E.
Angew. Chem., Int. Ed. 2010, 49, 8954-8957.
(11)
Gunther, W. R.; Michaelis, V. K.; Caporini, M. A.; Griffin, R. G.;
Román-Leshkov, Y. J. Am. Chem. Soc. 2014, 136, 6219-6222.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 11
(12)
Wolf, P.; Valla, M.; Rossini, A. J.; Comas-Vives, A.; Núñez-
(31)
(32)
Selvaraj, M.; Choe, Y. Appl. Catal., A 2010, 373, 186-191.
Hammond, C.; Conrad, S.; Hermans, I. Angew. Chem., Int. Ed.
Zarur, F.; Malaman, B.; Lesage, A.; Emsley, L.; Copéret, C.; Hermans, I.
1
2
3
4
5
6
7
Angew. Chem., Int. Ed. 2014, 126, 10343-10347.
(13)
Van de Vyver, S.; Román-Leshkov, Y. Angew. Chem., Int. Ed.
2015, 54, 12554-12561.
(14)
Boronat, M.; Concepción, P.; Corma, A.; Renz, M.; Valencia, S.
J. Catal. 2005, 234, 111-118.
(15)
Boronat, M.; Concepcion, P.; Corma, A.; Navarro, M. T.; Renz,
M.; Valencia, S. Phys. Chem. Chem. Phys. 2009, 11, 2876-2884.
(16)
Lew, C. M.; Rajabbeigi, N.; Tsapatsis, M. Microporous
8
9
Mesoporous Mater. 2012, 153, 55-58.
(17)
Soc. A 2012, 468, 2000-2016.
Osmundsen, C. M.; Holm, M. S.; Dahl, S.; Taarning, E. Proc. R.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(18)
389-396.
(19)
Chung, Y.-M.; Rhee, H.-K. J. Mol. Catal. A: Chem. 2000, 159,
(38)
Dijkmans, J.; Dusselier, M.; Janssens, W.; Trekels, M.;
Vantomme, A.; Breynaert, E.; Kirschhock, C.; Sels, B. F. ACS Catal. 2016,
6, 31-46
Wettstein, S. G.; Alonso, D. M.; Chong, Y.; Dumesic, J. A.
Energy Environ. Sci. 2012, 5, 8199-8203.
(20)
514-518.
(21)
State, Materials and Surface Science: A Book in Celebration of the Life and
Work of Sir John Meurig Thomas; The Royal Society of Chemistry: 2008,
p 639-650.
(22)
(23)
39, 4643-4655.
(24)
(25)
4584.
(26)
4902-4909.
(27)
Chem., Int. Ed. 2012, 51, 8212-8216.
(28)
– Europ. J. 2014, 20, 1298-1309.
(29)
(39)
(40)
658.
(41)
3792.
(42)
Clarke, M. L.; France, M. B. Tetrahedron 2008, 64, 9003-9031.
Urakawa, A.; Bürgi, T.; Baiker, A. Chem. Phys. 2006, 324, 653-
Farnesi Camellone, M.; Marx, D. J. Phys. Chem. Lett. 2013, 4,
Boronat, M.; Corma, A.; Renz, M. In Turning Points in Solid-
Baurecht, D.; Fringeli, U. P. Rev. Sci. Instrum. 2001, 72, 3782-
Clayden, J.; Greeves, N.; Warren, S. Organic Chemistry; OUP
Oxford, 2012.
(43)
Eccles, R. J. Pharm. Pharmacol. 1994, 46, 618-630.
Mojet, B. L.; Ebbesen, S. D.; Lefferts, L. Chem. Soc. Rev. 2010,
Baerlocher, C., McCusker L.B. Database of Zeolite Structures.
http://www.iza-structure.org/databases/, (accessed Oct 10, 2015).
(44)
(45)
85, 371-380.
(46)
2003, 107, 13061-13068.
(47)
R.; Pariente, J. P.; Corma, A. Appl. Catal. 1989, 47, 367-374.
(48)
Gounder, R.; Iglesia, E. Chem. Commun. 2013, 49, 3491-3509.
Allerhand, A.; von R. Schleyer, P. J. Amer. Chem. Soc. 1963,
Zaera, F. Chem. Soc. Rev. 2014, 43, 7624-7663.
Andanson, J.-M.; Baiker, A. Chem. Soc. Rev. 2010, 39, 4571-
Urakawa, A.; Bürgi, T.; Baiker, A. Chem. Eng. Sci. 2008, 63,
Meemken, F.; Maeda, N.; Hungerbühler, K.; Baiker, A. Angew.
Urakawa, A.; Wirz, R.; Bürgi, T.; Baiker, A. J. Phys. Chem. B
Fuentes, M.; Magraner, J.; De Las Pozas, C.; Roque-Malherbe,
Hori, Y.; Iwata, T.; Okeda, Y.; Process for producing isopulegol,
Meemken, F.; Baiker, A.; Schenker, S.; Hungerbühler, K. Chem.
US20020133046 A1: 2002.
(49)
List, B. J. Amer. Chem. Soc. 2015, 137, 13268-13271.
Liu, L.; Leutzsch, M.; Zheng, Y.; Alachraf, M. W.; Thiel, W.;
Pacheco, J. J.; Davis, M. E. Proc. Natl. Acad. Sci. U. S. A. 2014,
111, 8363-8367.
(30)
Selvaraj, M.; Kawi, S. J. Mater. Chem. 2007, 17, 3610-3621.
10
ACS Paragon Plus Environment

Page 11 of 11
ACS Catalysis
Graphic for Table Of Contents
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment