Intraparticle Diffusional versus Site Effects on Reaction Pathways in Liquid-Phase Cross Aldol Reactions

Article abstract of DOI:10.1002/cphc.201701219

Chemo- and regioselectivity in a heterogeneously catalyzed cross aldol reaction were directed by tuning the nature of the sites, textural properties, and reaction conditions. Catalysts included sulfonic acid-functionalized resins or SBA-15 with varying particle size or pore diameter, H-BEA zeolites, and Sn-BEA zeotype; conditions were 25 °C to 170 °C in organic media. Benzaldehyde and 2-butanone yielded branched (reaction at -CH₂- of butanone) and linear (reaction at -CH₃) addition and condensation products; and fission of the branched aldol led to β-methyl styrene and acetic acid. Strong acids promoted the dehydration step, and regioselectivity originated from preferred formation of the branched aldol. Both, resins and functionalized SBA-15 materials yielded predominantly the branched condensation product, unless particle morphology or temperature moved the reaction into the diffusion-limited regime, in which case more fission products were formed, corresponding to Wheeler Type II selectivity. For H-form zeolites, fission of the branched aldol competed with dehydration of the linear aldol, possibly because weaker acidity or steric restrictions prevented dehydration of the branched aldol.

Full text of DOI:10.1002/cphc.201701219

Accepted Article
Title: Intraparticle Diffusional Effects vs. Site Effects on Reaction
Pathways in Liquid-Phase Cross Aldol Reactions
Authors: Koushik Ponnuru, Jinesh C Manayil, Hong Je Cho, Wei Fan,
Karen Wilson, and Friederike Jentoft
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPhysChem 10.1002/cphc.201701219
Link to VoR: http://dx.doi.org/10.1002/cphc.201701219
A Journal of
www.chemphyschem.org

10.1002/cphc.201701219
ChemPhysChem
The chemoselectivity towards condensation or fission products in the aldol reaction of benzaldehyde
and butanone is directed by using solid acid catalysts, including sulfonic acid-functionalized resins and
SBA-15 materials, and β-zeolites, and conditions favoring or preventing diffusion limitation.
1
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Intraparticle Diffusional Effects vs. Site Effects on Reaction Pathways in
Liquid-Phase Cross Aldol Reactions
Koushik Ponnuru,1 Jinesh C. Manayil,2 Hong Je Cho,1 Wei Fan,1 Karen Wilson,2,3 Friederike C. Jentoft1,*
1Department of Chemical Engineering, University of Massachusetts, Amherst, 159 Goessmann Laboratory, 686 N
Pleasant, Amherst, MA 01003, USA
2 European Bioenergy Research Institute, Aston University, Birmingham, B4 7ET, UK
3New address: School of Science, RMIT University, 124 La Trobe St, Melbourne VIC 3000, Australia.
*Corresponding author email: fcjentoft@umass.edu
Abstract
Chemo- and regioselectivity in a heterogeneously catalyzed cross aldol reaction were directed by tuning
the nature of the sites, textural properties and reaction conditions. Catalysts included sulfonic-acid
functionalized resins or SBA-15 with varying particle size or pore diameter, H-BEA zeolites, and Sn-BEA
zeotype; conditions were 25 °C to 170 °C in organic media. Benzaldehyde and 2-butanone yielded
branched (reaction at -CH2- of butanone) and linear (reaction at -CH3) addition and condensation
products; and fission of the branched aldol led to β-methyl styrene and acetic acid. Strong acids promoted
the dehydration step, and regioselectivity originated from preferred formation of the branched aldol.
Both, resins and functionalized SBA-15 materials yielded predominantly the branched condensation
product, unless particle morphology or temperature moved the reaction into the diffusion-limited regime,
in which case more fission products were formed, corresponding to Wheeler Type II selectivity. For H-
form zeolites, fission of the branched aldol competed with dehydration of the linear aldol, possibly
because weaker acidity or steric restrictions prevented dehydration of the branched aldol.
Keywords: acid-base catalysis, kinetics, C-C coupling, C-C bond cleavage, transport limitation
2
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
1. Introduction
Aldol reactions are versatile C-C bond formation reactions that are widely used in the synthesis of bulk
and fine chemicals. For example, higher alcohols and aldehydes are produced by aldol reactions (and
subsequent hydrogenation), mostly using liquid catalysts such as caustic soda or acetic acid.[1] Many
ingredients in the flavor and fragrance industry are produced via aldol reactions.[2-5] An emerging
application of aldol condensation is upgrading of biomass-derived pyrolysis oils for the production of fuels.
These oils contain a large fraction of short-chain carbonyl compounds, and aldol condensations have
drawn strong interest because they increase carbon chain length while removing oxygen without loss of
carbon. For instance, n-octane could be produced from the aldol condensation of furfural and acetone
followed by hydrodeoxygenation.[6]
For some of these reactions, the regioselective activation of carbonyl compounds with two alpha
hydrogens is of significant interest. Regioselectivity control in aldol condensation of unsymmetrical ket-2-
ones can be used to preferentially obtain branched or straight carbon chains that are crucial in the
synthesis of a variety of aroma chemicals.[2-5] The ability to control the formation of fuel precursors with
more or fewer branches would predestine the product for use as gasoline or diesel fuel.
A potentially useful side reaction is the fission of the aldol reaction products which is known to occur
during the conversion of acetone to diacetone alcohol and mesityloxide, yielding isobutene and acetic
acid. First observed long ago,[7,8] this pathway has drawn attention recently for the sustainable production
of isobutene,[9-11] which is a precursor of industrial products such as butyl rubber and ethyl tert-butyl
ether.[12] A variety of inorganic solid catalysts have been tested for C−C hydrolytic bond cleavage reactions
including mixed metal oxides[9,10,12,13,14] and acidic zeolites[15-19] with efforts focused on catalyst activity and
stability. The fission has thus far only been observed in vapor phase conversion at high temperatures
(>350 °C), and it is not clear whether these products result from the cleavage of aldol addition or the
3
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
cleavage of aldol condensation products. Moreover, little is known about optimizing chemoselectivity for
either condensation or fission products.
It is obvious from these examples that it is desirable to steer the regio- and chemoselectivity in aldol
reactions. Since homogeneous catalysis is dominant in this area, the known means are those of organic
chemistry. Pre-enolization is the most commonly applied method.[20] Variation of the catalyst can also be
successful; for example, acids activate hydrogen in the -CH2- group in butanone whereas bases activate
the –CH3 group.[21] Solvents have been reported to significantly affect selectivity in some cases, for
example, the enantioselectivity in the Mukayama aldol reaction[22] or the diastereoselectivity in the
reaction of tert-butyl acetate and 2-phenyl propanal.[23-27] While successful, these methods are not
necessarily suitable for large scale processing.
Solid catalysts, in addition to being more easily recoverable than soluble catalysts, offer additional
means to improve selectivity. Reactions may be steered by shape selectivity if the reactants and the pores
of the catalyst have the same dimensions, or by diffusion limitations in pores if the reactions to be
discriminated follow different rate laws and will respond differently to variations in local concentrations.
Shape-selective processes are well known in zeolite catalysis and include transition state shape selectivity
where formation of an undesired product is precluded, and reactant and product shape selectivity, which
can be seen as extreme cases of transport limitation. While the theory on the influence of diffusion
limitations on the selectivity in parallel and sequential reactions has been outlined by Wheeler long ago[28]
and further expanded by Roberts,[29] practical realizations of these concepts are scarce. Although it is
established that intraparticle concentration gradients have a considerable effect on the apparent activity
and selectivity of porous catalysts, the effect on activity has received significantly more attention than the
effect on selectivity,[30-32] even though selectivity is usually the most important variable in industrial
operations. The rate of reaction along the separate pathways in a reaction network can be influenced to
4
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
different extents by virtue of the relative abilities of the products and reactants to diffuse through the
porous medium.
This paper focuses on selectivities in cross aldol reactions. Butanone and benzaldehyde are chosen as
the test reactants to elucidate the reaction network. Butanone as an unsymmetrical ketone allows for
regioselective activation at the methyl or methylene group, while benzaldehyde can only act as
electrophile. Dehydration is not necessarily immediate,[33] and the aldols may be observed. Fission of the
aldol products may produce β-methylstyrene, a chemical intermediate in the manufacture of plasticizers,
resins, adhesives and various other valuable polymeric chemicals.[34-36] It will be shown that
regioselectivity is diagnostic of the rate-determining step in the aldol condensation[33] and that
chemoselectivity is diagnostic of the nature of sites and spatial constraints. Three types of strong Brønsted
acidic catalysts are investigated, resin catalysts characterized by mesopores, sulfonic-acid functionalized
SBA-15 materials with a combination of meso- and micropores, and zeolites with BEA framework and
either only micropores or both meso- and micropores. The influence of transport limitations on product
selectivity is also investigated by varying the reaction temperature and by comparing with homogeneous
catalysis by benzenesulfonic acid.
2. Experimental section
2.1. Benchmark catalysts
Cross-linked sulfonated polystyrene resins Amberlyst 36 (WET), Dowex 50WX8 (50-100 mesh), Dowex
50WX2 (200 mesh) were obtained from Sigma-Aldrich and served as solid benchmark catalysts, and
benzene sulfonic acid (Sigma-Aldrich 97%) served as soluble molecular benchmark catalyst.
5
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
2.2. Synthesis and functionalization of SBA-15
2.2.1. Synthesis of SBA-15
SBA-15 materials with different pore architectures were synthesized according to previously reported
procedures.[37] Twenty grams of Pluronic P123 (Aldrich) were dissolved in 614 ml of distilled water and
114 ml of 2 M HCl solution by stirring at a temperature of 35 °C. After addition of 46 ml of
tetraethylorthosilicate (TEOS), the contents were transferred to a jacketed reactor, combined with a
second identically prepared batch, and stirred continuously for 22 h at 35 °C. The resulting gel was divided
into four separate batches, which were aged under static conditions in closed polypropylene bottles.
Aging time and temperature were varied to obtain samples with different meso and micropore
dimensions. The following combinations of times and temperatures were applied: 1 day at 50 °C, 1 day at
80 °C, 2 days at 100 °C, or 7 days at 120 °C. Subsequently, the solid product was retrieved by filtration,
washed with DI water and calcined in static air at 550 °C for 5 h. The samples are denoted as SBA-15 (x °C)
where x corresponds to the aging temperature during the synthesis of the parent SBA-15.
2.2.2. Propyl sulfonic acid (PrSO3H) functionalization of SBA-15
The synthesis procedure was adapted from Ref. [38] Two grams of calcined SBA-15 (x °C) was stirred
with 2.5 ml of 3-mercapto n-propyl trimethoxysilane (MPTMS, Sigma) in 60 ml of toluene at a temperature
of 130 °C. The suspension was refluxed for 24 h; then the thiol-functionalized solid was retrieved by
vacuum filtration, washed with methanol and dried over night at 80 °C. The thiol-functionalized sample
was oxidized with H2O2 at room temperature for 24 h (30 ml of 33 wt% H2O2 per gram of material) to
obtain sulfonic acid-functionalized SBA-15. The samples are denoted as PrSO3H/SBA-15 (x °C) where x
corresponds to the aging temperature during the synthesis of the parent SBA-15.
6
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
2.3. Synthesis of zeolite and zeotype materials
2.3.1. Al-BEA (Fluoride medium)
The synthesis procedure was adapted from Ref. [39] Two separate solutions were first prepared that
contained 0.01 g of aluminum powder in 3.06 g of tetraethylammonium hydroxide (TEAOH) (35%, Alfa
Aesar) and 10 g of (TEOS) in 8 g of TEAOH, respectively. The two solutions were mixed in a PTFE cup and
stirred at room temperature until complete evaporation of water and ethanol. To this mixture, 0.944 ml
of hydrofluoric acid (48%, Sigma) was added. The resulting highly viscous gel was heated in a Teflon-lined
autoclave and held at a temperature of 140 °C for six days with 30.0-rpm rotation. The obtained solid was
washed, dried and calcined at 550 °C in flowing air for 12 h. This sample is denoted as Al-BEA/F-.
2.3.2. Preparation of Al-BEA (commercial)
The ammonium form of zeolite BEA (CP814E, Si/Al = 12.5) was obtained from Zeolyst. The catalyst
was calcined at 550 °C with a ramp rate of 1.0 °C min−1 in flowing air for 12 h. This sample is denoted as
Al-BEA.
2.3.3. Synthesis of tin-substituted zeotype
The synthesis of Sn-BEA was described in detail in previous reports.[40,41] Briefly, a starting solution
was created from tetraethylorthosilicate, TEAOH, tin (IV) chloride hydrate and water and stirred until
ethanol formed upon hydrolysis of TEOS had evaporated. A suspension of dealuminated zeolite BEA seeds
(corresponding to 4.0 wt.% seed on a silica basis) was added, followed by hydrofluoric acid. The final
composition of the synthesis gel was SiO2: 0.008 SnO2: 0.54 TEAOH: 0.54 HF: 7.5 H2O. The gel was heated
in an autoclave and held at a temperature of 140 °C for four days with 2.0 rpm rotation. The obtained
solid was washed, dried and calcined at 550 °C in flowing air for 12 h.
7
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
2.4. Characterization methods
2.4.1. SBA-15
The success of the synthesis was verified with a number of characterization techniques. Low angle
XRD patterns were recorded on a Bruker D8 Advance diffractometer fitted with an X’celerator detector
and Cu Kα (1.54 Å) source over a 2θ range from 0.3° to 10°. The textural properties were characterized by
measuring N2 adsorption and desorption isotherms at 77 K using a Quantachrome Nova 4000 porosimeter
and analyzing the adsorption branches with the BET and BJH methods. The samples were degassed
overnight under vacuum at 120 °C before analysis. The sample sulfur content was determined by CHNS
analysis on a Thermo Scientific Flash 2000 CHNS-O analyzer. The amount of physisorbed water on the
catalyst surface was estimated by performing thermogravimetric analysis (TGA) analysis. A Stanton
Redcroft STA780 thermal analyzer was used for this purpose, and 10 mg of sample were heated under N2
flow (30 ml min-1) at a ramp rate of 10 °C min-1 from 40 °C to 800 °C. Surface energies of the catalyst
samples were determined by inverse gas chromatography using a Surface Measurement System IGC with
IGC controller v1.8 software. Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was
performed with a Spectrum 100 spectrometer (PerkinElmer). A Praying Mantis™ diffuse reflectance
accessory (DRP-P11, Harrick Scientific) was used with a CHC-CHA-3 environmental chamber (Harrick
Scientific). Potassium bromide (Sigma, FT-IR grade) was dried in situ and used to record the background
spectrum. The catalysts were diluted in KBr (10 wt%) before the analysis. Scanning electron micrographs
of the samples were collected using a Magellan 400 XHR-SEM instrument (FEI) equipped with a field
emission gun operated at 3.0 kV. The samples were sputter-coated with platinum before imaging.
2.4.2. Zeolites and metal-substituted zeotypes
Detailed characterization procedures and additional results not repeated here can be found in Ref.
[42] The metal content was determined by inductively coupled plasma optical emission spectroscopy (ICP-
OES, iCap 6500 Dual view, Thermo Scientific). X-ray diffraction (XRD) patterns of the samples were
8
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
recorded on an XRD diffractometer (X’Pert Pro, PANalytical) with Cu Kα radiation operated at an
acceleration voltage of 45 kV and a current of 40 mA. N2 adsorption and desorption isotherms at 77 K
were obtained using an automated gas sorption analyzer (Autosorb iQ2, Quantachrome) after the samples
were degassed at 300 °C under vacuum for 12 h. Acid site titration by isopropylamine chemisorption with
subsequent temperature-programmed desorption (TPD) was performed for Al-BEA and Al-BEA/F- using a
thermogravimetric analyzer with mass spectrometer (TG-MS), specifically an SDT600 TGA system (TA
Instruments) equipped with a quadruple mass spectrometer (Extorr XT 300). The number of Brønsted acid
sites was determined from the mass loss associated with the evolution of propene, which was identified
by MS. DRIFTS and SEM analyses were performed using the same instruments as mentioned in Section
2.4.1.
2.5. Catalytic tests
Batch reactions were performed under autogenous pressure in an oven-dried pressure-resistant glass
tube of 15 ml volume equipped with a thermowell (Ace Pressure Tube, Ace Glass 8648-164). Typically, the
reaction mixture contained 1.0 M benzaldehyde (Sigma-Aldrich ≥99.5%) and 0.1 M hexadecane in excess
2-butanone (Fisher 99%). The benzaldehyde-to-site ratio was 517 unless otherwise indicated, with sites
defined as total moles for homogeneous catalysts, moles of sulfur for functionalized SBA-15, moles of
metal for the zeotypes, and moles of acid sites as titrated by isopropylamine chemisorption/TPD for the
zeolites, respectively. All solid catalysts were stored in air and used without pre-treatment. The reaction
mixture was magnetically stirred in the pressure tube for 20 h at a temperature of 140 °C (unless otherwise
indicated) in a temperature-controlled oil bath. The specified reaction temperature was measured by a
thermocouple placed in the thermowell. The reaction mixture was centrifuged to remove the solid
catalyst before GC analysis.
9
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
2.5.1. Reaction product analysis by GC
The reaction products were quantified by using gas chromatography with flame ionization detection
(FID); an Agilent 7890B system equipped with an HP-5 capillary column (length 30 m, i.d. 0.32 mm (wide
bore), film thickness 0.25 μm) was used. The oven temperature was held at 35 °C for 5 min, ramped at 10
°C min-1 to 250 °C, with a hold time of 10 min. The GC injection port was operated at 250 °C in 35:1 split
mode with N2 as the carrier gas at 3 ml min-1, and the injection volume was 0.5 μl. Response factors were
calculated using standards of reference compounds or structurally equivalent compounds.[43] Each
reaction sample was analyzed at least three times by GC, and the average responses were taken to
calculate the conversion, yield, and selectivity. Reported error bars represent standard deviations of
repeat catalytic experiments.
In addition, gas chromatography coupled with mass spectrometry (GC-MS) was performed with an
Agilent 7890A system (7890 GC/7972 series mass selective detector, HP). Products were identified by
using mass spectral library matches and the AMDIS32 software or by measuring reference compounds.
The acetic acid in the reaction mixture was determined by derivatizing with trimethylsilyl imidazole (TMSI;
Sigma-Aldrich) and subsequent GC-MS analysis as shown in Figure S1.
2.5.2. In-situ ATR-FTIR spectroscopic analysis of the reaction mixture
In addition to GC analysis, the reaction mixture was spectroscopically analyzed in some experiments,
to obtain the liquid phase composition with high time resolution. A Mettler-Toledo ReactIR 15 equipped
with a DiComp (diamond-composite) attenuated total reflectance (ATR) insertion probe connected by an
AgX (silver halide) fiber conduit to a liquid nitrogen-cooled mercury cadmium telluride (MCT) detector
was used. The ATR probe was inserted into the Ace pressure tube such that the probe tip remained
submerged in the liquid reaction mixture at all times. IR spectra were collected in the wavenumber range
from 2000 to 650 cm-1 every 60 s with an average of 256 scans per spectrum at a spectral resolution of 4
cm-1. The acquired real-time continuous spectra were exported to the iC QuantTM module to build a
10
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
quantitative model for the prediction of absolute concentrations. Baseline correction and solvent
subtraction were applied before data processing. Calibration was performed using a univariate model by
profiling the characteristic band of benzaldehyde at 827 cm-1 from the spectra of reference reaction
standards collected at various reaction compositions. The quantitative analysis of the data obtained from
the reactIR is explained in detail in Section 1 of the supplementary information.
2.5.3. Calculations
The catalytic performance is reported in terms of benzaldehyde conversion, product selectivity and
turnover frequency defined as follows, where n is the number of moles and indices i, f, and p mean initial,
final and product:


nBenzaldehyde,i  nBenzaldehyde, f
Benzaldehyde conversion (%) 
100




nBenzaldehyde,i




np
Product yield (%) 
100




p
nBenzaldehyde,i






np


Product selectivityp (%) 
100
np





products
nbenzaldehyde,i  nbenzaldehyde, f
Turnover frequencyBenzaldehyde (h1) 
timenactive sites
3. Results
3.1. Catalyst characterization
3.1.1. Resin specifications
Properties of the resins as specified by the manufacturer are summarized in Table 1. Exchange
capacity was used to calculate the reactant-to-catalyst ratio. Amberlyst 36 is a macroreticular resin with
invariably accessible internal surface area, whereas the Dowex products are gel-type resins, which require
swelling in a solvent to make the pores accessible.
11
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Table 1 Properties of sulfonated cross-linked polystyrene resins as specified by supplier
Catalyst
Resin type
Exchange
capacity,
H+ (eq/l)
Mesh
size
Particle size
(cm)
Amberlyst 36
Dowex 50wx8
Dowex 50wx2
Dowex 50wx2
Macroreticular
1.95
1.70
0.60
0.60
-
0.030
0.010
0.004
0.010
Gel
Gel
Gel
50-100
200
50-100
3.1.2. Functionalized SBA-15 materials – structure, morphology, texture and composition
The small angle X-ray diffraction patterns and N2 adsorption isotherms (Figures S2 and S3) confirmed
the successful synthesis of the two-dimensional hexagonal lattice (p6mm) of SBA-15 with diffraction peaks
corresponding to (100), (110), and (200) reflections.[44] A gradual shift of the (100) reflection towards
lower angle was observed with aging temperature, implying an increase in d-spacing. The structure was
retained upon functionalization with propyl sulfonic acid groups. The mesopore sizes of the four SBA-15
samples and their functionalized variants ranged from 3.6 to 7.5 nm (Table 2 and Figure S4).
The rod-shaped morphology of the four SBA-15 samples synthesized at different temperatures is
evident from Figure 1a-d. The average particle size in these samples is very similar with only slight
variations in the length and width of the rods. Typical dimensions were 1 µm long and 0.5 µm wide.
12
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Figure 1. SEM images of (a) SBA-15 (50 °C), (b) SBA-15 (80 °C), (c) SBA-15 (100 °C), and (d) SBA-15 (120 °C).
The textural properties and elemental analysis data of the parent SBA-15 (x °C) series and the
functionalized derivatives PrSO3H/SBA-15 (x °C) materials are summarized in Table 2. The mesopore size
of the SBA-15 sample increased from 4.1 to 7.5 nm with a concomitant decrease in micropore volume
from 0.17 to 0.05 cm3 g-1 upon increasing the aging temperature from 50 to 120 °C. This reduction in the
micropore volume is consistent with previously reported observations for SBA-15.[44] The walls between
the mesopores of SBA-15 precursors consist of an interpenetrating network of the P123 template’s
ethylene oxide chains and silica. Upon calcination, the template is removed, and the micropores are
formed in the mesopore walls. The wall thickness can be estimated from the pore diameter and the lattice
constant, and the SBA-15 samples with the largest mesopore sizes have thinner walls, resulting in reduced
micropore volumes.
13
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
No specific trend in BET surface area was observed with aging temperature. The samples aged at lower
temperatures (50 °C and 80 °C) were characterized by similar surface areas with an insignificant decrease
from 773 m2 g-1 at 50 °C to 759 m2 g-1 at 80 °C. The surface area then increased to 861 m2 g-1 at 100 °C and
decreased to 702 m2 g-1 with a further increase in aging temperature to 120 °C. Contrasting trends for BET
surface areas were observed in the literature for SBA-15 with increase in aging temperature.[44,45-47]
Interestingly, in all these reports, materials aged at a temperature of ≈100 °C always exhibited the highest
surface area as in the present investigation. One possibility to rationalize these trends is the dual effect of
thinner walls, which imply lower micropore volume but also decreasing density of the material. The BET
surface area decreased slightly upon derivatization with PrSO3H groups for all four SBA-15 samples,
whereas there were no clear trends in the behavior of mesopore diameter and mesopore volume.
Microporosity, as determined from t-plot analysis, consistently decreased upon introduction of PrSO3H
groups, indicating functionalization of the inner surface of the micropores. In more detail,
functionalization significantly reduced micropore volume for samples with small mesopores and high
initial micropore volume, and significantly reduced mesopore volume for samples with large mesopores
and small initial micropore volume. This result may signify a shift in location of the functional groups from
preferably in the micropores to partially in the mesopores.
The sulfur content varied from 0.54 wt% (0.17 mmol g-1) to 0.77 wt% (0.24 mmol g-1) without a
recognizable trend (Table 2). Acid sites concentrations were assumed to be equal to sulfur content.
14
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Table 2 Textural properties of functionalized SBA-15 catalysts
Materials
Surface BJH pore Total
Micropore Micropore Pore wall Sulfur
Surface TGA weight
(wt %) (mJ/m2)
(50-200 °C)
areaa
diameter BJH pore volume
Area
thicknessb, content energy loss
(m2 g-1)
(cm3g-1)
(m2g-1)
(nm)
volume
2Rwall
(cm3 g-1)
(nm)
(H2O,
mmol g-1)
SBA-15 (50 °C)
773
699
759
663
861
777
702
460
4.1
3.6
4..9
4.7
6.4
6.2
7.5
7.6
0.53
0.55
0.76
0.77
1.07
1.00
1.62
1.07
0.17
0.10
0.13
0.05
0.07
0.05
0.05
0.02
381
240
305
115
175
131
126
43
4.48
5.19
5.24
5.52
3.37
3.15
2.50
2.39
-
85
-
0.43
0.49
0.35
0.42
0.19
0.34
0.05
0.11
PrSO3H/SBA-15
(50 °C)
0.62
-
SBA-15 (80 °C)
76
-
PrSO3H/SBA-15
(80 °C)
0.77
-
SBA-15 (100 °C)
74
-
PrSO3H/SBA-15
(100 °C)
0.62
-
SBA-15 (120 °C)
55
-
PrSO3H/SBA-15
(120 °C)
0.54
aBET. bPore wall thickness = d-spacing − pore size; d-spacing calculated from the X-ray diffraction data for (100) plane
3.1.3. Functionalized SBA-15 materials – surface properties
The surface energy values obtained from IGC decreased with increasing aging temperature and
mesopore size, indicating that the surface becomes more hydrophobic. Thermogravimetric analysis of the
catalyst samples was performed to measure the physisorbed water which is weakly bound to the surface
and hence is lost at relatively low temperatures. Therefore, the first weight loss step observed between
50 °C and 150 °C was used to quantify the adsorbed water content of the unfunctionalized SBA-15 series.
Hydrophilic properties can be considered to scale with the water content (weight %) of the samples since
a H2O/Si-OH ratio close to 1 has been reported for SBA-15 in the literature.[48] TG traces (Figure S5a)
suggest that the sample aged at 120 °C during synthesis is the most hydrophobic in the series as it has the
least water loss (1.18%); 50 °C and 80 °C samples have the highest water losses (6.39 and 6.64%), and the
100 °C sample has an intermediate loss (3.89%). A similar hydrophobicity trend was observed for the
15
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
functionalized samples. These consistently attracted more water than their unfunctionalized counterparts
(Figure S5b and TGA weight loss values in Table 2). This increase in hydrophilicity can be attributed to the
polarity of the -SO3H groups.
DRIFTS analysis was performed to better understand the origins of hydrophobicity of the SBA-15 series
surfaces. The spectra were found to be amenable to quantitative interpretation, because the area of the
H2O (ν+δ) combination band at ≈5260 cm-1,[48, 49] which is characteristic of physisorbed water, was strictly
proportional to the amount of physisorbed water found by TG (Figure 2).
The silanol groups of the synthesized samples were characterized by absorption bands in the 3800-
3000 cm-1 and 4800-4200 cm-1 regions (Figures 3 and S6a,b respectively), which are in agreement with
Ref.[49] and can be ascribed to stretching and combinations modes ν(OH) and ν+δ(OH). The ν(OH) bands
appearing at ≈3520, 3670, 3710, and 3745 cm-1 can be assigned to H-bonded, internal, terminal, and
isolated silanol groups, respectively.[50] There was no obvious difference in the OH vibrations between
plain and propyl sulfonic acid-functionalized samples. The ν(OH) band of free -SO3H groups is around 2700-
3600 cm-1,[51] presumably, strong H-bonding prevented distinction of these vibrations.
16
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
Physisorbed H2O (mmol g-1)
Figure 2. Area of the (ν+δ) H2O combination band at 5260 cm-1 (original spectra in Figure S6a) of
sulfonic-acid functionalized SBA-15 (x °C) series versus the concentration of adsorbed water as
determined by thermogravimetry (Table 2). DRIFT spectra recorded at 50 °C. Water content
decreases with increasing mesopore diameter.
The concentration of unperturbed silanol groups (3745 cm-1) was about equal on all samples before
dehydration (Figure 3a). Comparison of spectra recorded before (Figure 3a) and after (Figure 3b) removal
of the physisorbed water (disappearance of band at 5260 cm-1, Figure S6b) shows a decrease in H-bonding
intensity between 3600 and 3000 cm-1 and an increase in isolated silanols at ≈3745 cm-1, indicating
physisorption of water on silanol groups.
There was a clear trend towards fewer H-bonded species with increasing aging temperature during
synthesis, as seen in Figure 3. This trend was also evident with unfunctionalized SBA-15 samples (not
shown). Interacting silanol groups are known to have different reactive properties than isolated silanols,
and hydrophilicity is, at least in part, originating from these groups. Presumably, the H-bonding silanol
groups are located in micropores; at least there is a rough correlation of the intensity of H-bonded species
in Figure 3b with the micropore volumes in Table 2. In summary, DRIFTS data are in agreement with IGC
and TG regarding hydrophilicity, as evidenced by H-bonded OH groups and by adsorbed water,
respectively.
17
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
(a)
(b)
Figure 3. DRIFT spectra of PrSO3H/SBA-15 samples of different aging temperatures in the 4000 –
3000 cm-1 region. a) Spectrum of calcined sample recorded at 50 °C. b) Spectrum recorded after
heating sample to 200 °C.
3.1.4. Zeolites and metal-substituted zeotypes – structure, composition, morphology and texture
XRD pattern and N2 adsorption isotherm for Sn-BEA are published in Ref.[40] Data for Al-BEA/F-, and
Al-BEA is shown in Figure S7 of the supplementary information. The XRD patterns of the final solids
showed the typical features of the BEA framework with no other crystalline phases being present. N2
adsorption isotherms measured at a temperature of 77 K indicated that micropore volumes were 0.15–
0.17 cm3 g1, consistent with the BEA topology (Table 3). Lewis acid site concentrations of zeotype
catalysts were assumed to be equal to the metal (tin) content as determined by elemental analysis (ICP-
OES), whereas the Brønsted acid site concentrations of the zeolites were determined by isopropylamine
TPD.
Table 3 Textural properties and composition of zeotype catalysts
Catalyst
Surface
areaa
Pore diameter Total pore
Micropore
volume
Si/metalb
Acid site
loading
(nm)
volume
(mol mol-1)
(m2 g-1)
(cm3 g-1)
(cm3 g-1)
(mmol g-1)
Sn-BEA
Al-BEA/F-
Al-BEA
488
528
563
0.75
0.75
0.75
0.32
0.26
0.59
0.17
0.16
0.15
126
42.5
12.5c
0.13b
0.28d
0.62d
aBET. bDetermined from ICP-OES analysis. c From the manufacturer (Zeolyst), dIsopropylamine TPD.
18
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
The SEM image of the Al-BEA in Figure 4a shows agglomeration of the nano-crystallites to larger
particles of various sizes with rough appearance and no distinct boundaries, making it difficult to estimate
the crystal size from the image. This material was characterized by a Type II isotherm (Figure 4c) and
considerably higher total pore volume than micropore volume, consistent with ample void space inside
the agglomerates. Diffractograms showed broad reflections consistent with small crystallites. Synthesis
of Al-BEA in a fluoride medium resulted in smaller particles, with a diameter of ≈1 µm as seen in Figure
4b. These particles appeared to consist of densely packed, perhaps even intergrown crystallites,
consistent with the observed Type I isotherm and the absence of significant amounts of mesopores.
Diffractograms were characterized by narrow reflections, indicating large crystallites as observed by
others.[39]
3.1.5. Zeolites – surface properties
The strong band at 3735 cm-1 in the IR spectrum of Al-BEA (Figure 5) indicates the presence of a large
number of isolated silanol groups, which has been correlated to crystal defects and surface hydrophilicity
by Camber et al.[52] The decrease in framework Al content by synthesis in fluoride medium produced a
sample with notably fewer defects and silanol groups implying a higher surface hydrophobicity in
comparison to Al-BEA.
19
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Figure 4. SEM images of (a) Al-BEA, (b) Al-BEA/F-, and c) N2 adsorption isotherms of Al-BEA and Al-
BEA/F-.
3735
3590
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Al BEA/F-
Al-BEA
4000
3800
3600
3400
3200
3000
Wavenumber (cm-1)
Figure 5. IR spectra of dehydrated zeolite samples recorded at 50 °C.
20
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
3.2. Catalytic data
3.2.1. Principal products
Scheme 1 shows the reaction pathways (referred to as Pathways A and B) in the cross-aldol
condensation of 2-butanone and benzaldehyde. Reaction at either of the alpha carbons of 2-butanone
leads to different addition products (aldols I and II), which can dehydrate to give the condensation
products III and IV, respectively. All intermediates and products of the aldol condensation were observed,
in varying amounts depending on the catalyst used and reaction conditions.
Scheme 1: Reaction network for aldol reaction of butanone with benzaldehyde
Three addition products were chromatographically distinguished. A diastereomeric mixture of
erythro- and threo-4-phenyl-3-methyl-butan-4-ol-2-one (referred to collectively as I) indicated
condensation at the methylene carbon of the ketone, and 1-phenyl-pentan-1-ol-3-one (II), isomeric with
(I), indicated condensation at the methyl carbon of the ketone. The dehydration of the isomeric β-aldols
21
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
(I) and (II) resulted in α,β-unsaturated ketones, 3-methyl-4-phenyl-3-buten-2-one (III) and 1-phenyl-1-
penten-3-one (IV), respectively.
Previous investigations[33] showed that condensation product yields in excess of 90% are possible,
indicating the absence of equilibrium limitations under typical reaction conditions. Further products
observed in significant amounts were β-methyl-styrene (V) and acetic acid (VII), formally the result of a C-
C bond cleavage in Pathway A. These products will be referred to as fission products, and their formation
may be envisioned from the aldol (I) or from the condensation product (III).[9, 10, 13, 15] The corresponding
products of Pathway B, styrene (VI) and propionic acid (VIII), were only detected in traces in a few cases
and will not be further discussed. The rate constants are represented as k and indicated on the arrows of
the reaction pathways in Scheme 1. The superscript “b” refers to the branched products formed in
Pathway A and “l” refers to the linear products in Pathway B. The subscript represents the product number
formed or disappeared from its corresponding reaction as shown in Scheme 1.
3.2.2. Homogenous catalysis by benzenesulfonic acid
Benzenesulfonic acid was used as a soluble catalyst to provide a benchmark case without diffusion
limitation. Independent of the conditions, this catalyst promoted Pathway A, consistent with the
regioselectivity reported in the literature. Acid catalysts are known to lead to reaction at the methylene
group and formation of the branched addition product.[21, 53] As can be seen in Table 4, Entries 1 to 4, the
selectivity to the addition product decreased with conversion whereas the selectivity to the condensation
product increased, as is expected for an intermediate product in a series reaction. The regioselectivity of
the reaction steadily decreased with increasing temperature (Entries 5-9).
The selectivity toward fission products was significant only at low conversions when aldols can be
detected (Entries 1-3), suggesting that cleavage of addition products may yield these products. At high
temperatures and high conversions, only traces of fission products were observed (Entries 5-6, 8-9).
Selectivities to secondary aldol condensation products were less than 5% at all tested temperatures.
22
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
Table 4. Effect of conversion and temperature on product selectivity in excess butanone
Entry Temp. Benzaldehyde
Addition
Condensation Fission (%) φw-p
Catalyst
(°C)
Conversiona
A (%) B (%) A (%) B (%)
(%)
1
2
25
3.5
5.4
4.85
1.6
0.3
0
0.6
1.9
0.2
0.2
0.1
0.2
0
70
72
73
11
15
13
14
14
17
16
19
25
6
14
5
-
-
-
-
-
-
34.1
10.5
50.7
77
3
50
12
1
Benzene
sulfonic
acid
4
83
5
80
110
140
140
170
50
84
80
76
78
71
80
76
81
77
62
42
6
0.3
0.1
4
6
79
0
7
16
0.7
0
8
80
0.1
0
0.2
0.3
0.5
2
-
-
9
78
0
10
11
12
13
14
15
16
17
18
71
13
9
0.3
0.9
0
0.6
17
32
54
97
242
6
80
54
12
13
17
19
19
59
72
82
Amberlyst 36
95
62
2
3
110
140
170
110
140
170
74
1
0.2
0.1
0.2
2
5
81
0.4
0.3
33
14
4
17
37
0
70
26.5
31
Sn-BEA
4
9
1
19
79
25.5
4
9
1
aBenzaldehyde conversion was varied by adjusting the reaction time between 1 to 36 h
3.2.3. Assessment of diffusion limitations
Possible diffusion limitation was evaluated in several ways, by measuring rates as a function of particle
size and associated calculations of the Weisz-Prater criterion, and by determining the activation energies.
The Weisz-Prater criterion (φw-p) is a critical metric used to determine if a reaction is limited by
intraparticle diffusion. The effective diffusivity in the Weisz-Prater criterion incorporates the Knudsen
23
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
number λ, which is the ratio of radius of the diffusing product molecule to the radius of the pore. Hence,
the criterion accounts for variations in the kinetic diameter depending on the product molecule
considered, as well as for variations in pore size when different materials are compared. In general, for
φw-p<<1, there is no intraparticle concentration gradient and hence the reaction is not limited by diffusion,
whereas φw-p>>1 indicates strong diffusion limitation. The Weisz-Prater criterion was calculated for all
families of catalysts. For the resins, it was assumed that the pore size remains constant. For the SBA-15
series, using mesopore dimensions to evaluate φw-p resulted in a value much smaller than one, suggesting
no diffusion limitation in mesopores. Hence, micropore dimensions were used to calculate the Weisz-
Prater criterion and these values are reported. For the zeotypes, the effective diffusivities were very
conservatively estimated by scaling measured diffusivities[54] rather than by calculating diffusivities from
molecular properties and pore dimensions. To evaluate diffusion limitations, the largest species in the
reaction network was assessed, and reported dimensions of benzosuberone were used in lieu of the
(unknown) dimensions of the addition products. The full calculations are outlined in Section 2 of the
supplementary information.
3.2.4. Resin catalysts performance and regimes of diffusion control
The resins are characterized by sulfonic acid groups attached to a benzene ring; thus, their activity
would be expected to be in the same range as that of benzenesulfonic acid. With Dowex WX2 200 mesh,
characterized by the lowest site density and the smallest particle size, a benzaldehyde turnover frequency
(TOF) of 33 h-1 was observed versus a TOF of 44 h-1 with the free acid (at a temperature of 140 °C in
toluene). This result suggests that the resin sites were either slightly less active than benzenesulfonic acid,
which could be explained by the +I effect of the attached alkyl chain (from the polymer), or were not fully
accessible. Like their soluble counterparts, the resins produced predominantly the Pathway A
condensation product (Figure 6) and only a small amount of the Pathway B condensation product.
24
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
The rate of conversion of benzaldehyde was compared among commercially obtained sulfonated
cross-linked polystyrene resin beads with identical –SO3H groups but different particle sizes (0.074 mm -
0.8 mm). Figure 6 shows similar conversions with increasing number of sites present in the reaction
mixture, indicating a decrease in apparent turnover frequency and increasingly severe diffusion limitation.
Calculation of the Weisz-Prater criterion for Amberlyst-36 under these conditions gave a value of 156,
confirming diffusion limitation. With increasing particle size and more severe transport limitations, the
selectivity towards the branched condensation product (III) decreased with an associated increase in the
fission product (V) selectivity as shown in Figure 6.
100
20.2
4.5
33
80
60
40
20
0
22.5
Dowex wx2 (200 mesh)
(0.074 mm)
Dowex wx2 (50-100 mesh) Dowex 50wx8 (50-100 mesh)/
Amberlyst-36
(0.6-0.8 mm)
(0.15-0.3 mm)
(0.15-0.3 mm)
Benzaldehyde Conversion
Condensation A
Addition A
Condensation B
Addition B
Fission A
Secondary products
Figure 6. Effect of catalyst particle diameter on product selectivities with sulfonated cross-linked
polystyrene resins. Reaction conditions: solvent; toluene, 3 M butanone, 1 M benzaldehyde, 0.1 M
hexadecane (internal standard); 140 °C; 10 hours and catalyst mass 50 mg. Site densities are 0.6, 1.7
and 1.95 eq/l for Dowex WX2, Dowex WX8 and Amberlyst-36, respectively. The turnover frequency
(h-1) is indicated on the benzaldehyde conversion bars.
To further explore the range of diffusion limitations, the apparent activation energy was determined.
The rates measured by in situ ATR-FTIR analysis were used to generate the Arrhenius plot shown in Figure
7. It is evident that there are two different slopes, resulting in different activation energies Ea. The
apparent Ea is 48 kJ mol-1 in the 50–95 °C range, and 23 kJ mol-1 in the 110–140 °C range. The decrease in
apparent activation energy with temperature suggests an increased diffusive resistance and further
25
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
confirms pore diffusion limitation. The Weisz-Prater criterion also exceeded a value of one at high
temperatures (Table 4). Moreover, fission product selectivity increased as temperature and the values for
the Weisz-Prater criterion increased, suggesting that the fission products are favored over condensation
products as diffusion limitations become more severe.
Figure 7. Arrhenius plot of benzaldehyde disappearance rate vs temperature for Amberlyst 36
between 50 °C and 140 °C. Reaction conditions: Excess 2-butanone, 1 M benzaldehyde, 0.1 M
hexadecane (internal standard) and catalyst mass 50 mg. The reaction profiles at each temperature
are shown in Figure S8.
3.2.5. Reaction pathways investigated using resin catalysts
3.2.5.1. Origin of the fission product
The significant amounts of fission products that were observed with the resins under diffusion control
made possible the measurement of selectivities over a wide range of conversions to identify pathways in
the reaction network. Cleavage products equivalent to those observed here were observed in other aldol
condensations[10,13] and several pathways were formulated, a decomposition of the aldol,[10] or a protolytic
or hydrolytic cleavage of the condensation product.[7,8,10,11] To distinguish these possibilities, the Pathway
A products were analyzed with one of the benchmark catalysts, Amberlyst 36. Figure 8 shows the
26
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
benzaldehyde conversion and the yields of condensation and fission products as a function of reaction
time. Benzaldehyde conversion linearly increased with time. Both products yields increased steadily over
time, indicating that neither is an intermediate. The yield of the fission product was characterized by a
rapid initial rise and moderate growth at long reaction times, while the yield of condensation product was
initially low and then grew more significantly over time. These trends suggest parallel formation of these
products from the same intermediate (as illustrated in Scheme 1), rather than formation of the fission
product from the condensation product.
Figure 8. Dependence of product yields in Pathway A on benzaldehyde conversion using
Amberlyst 36 in 2-butanone. Reaction conditions: 1 M benzaldehyde, 0.1 M hexadecane
(internal standard); excess butanone, 140 °C, catalyst mass 25 mg. Error bars are
standard deviations above and below mean obtained from three repetitions of the same
test.
It is evident that different rate equations must apply for formation of fission and condensation
products if they stem from the same aldol intermediate. The stronger curvature of the fission yield
indicates that fission to β-methyl styrene and acetic acid is characterized by a smaller rate constant and a
higher order in the aldol than dehydration to the condensation product.
27
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
3.2.5.2. Effect of adding water on activity and product selectivity
To further corroborate the formation of the fission product from the aldol, the effect of adding water
to the reaction medium prior to the reaction was investigated. Water formed in the condensation step
has been proposed to participate in the side reaction to fission products.[9,10,15,55] The addition of water in
stoichiometric amounts (1:1 to benzaldehyde) strongly suppressed the rate of benzaldehyde conversion
and the rate of formation of the fission products at both 140 °C and 170 °C, as can be seen from the vastly
different reaction times (13 h and 4 h at 140 °C, 13 h and 1 h at 170 °C for reactions with and without
water respectively) needed to obtain similar conversions in Figure 9.
100
80
60
40
20
0
Amberlyst 36 + (1M water)
13 h, 140 °C
Amberlyst 36
4 h, 140 °C
Amberlyst 36 + (1M water)
13 h, 170 °C
Amberlyst 36
1 h, 170 °C
Benzaldehyde Conversion
Condensation A
Addition A
Condensation B
Addition B
Fission A
Secondary products
Figure 9. Effect of adding water to the reaction mixture on product selectivities with Amberlyst 36.
Reaction conditions: Excess 2-butanone, 1 M benzaldehyde, 0.1 M hexadecane (internal standard),
catalyst mass 50 mg, and optionally 1 M water.
3.2.6. Catalytic performance of pore-expanded functionalized SBA-15 materials
The catalyst pore structure, which will have a strong influence on effective diffusivity of reactants and
products, is different for the functionalized SBA-15 materials than for the resins. The series of the four
PrSO3H/SBA-15 catalysts is characterized by increasing mesopore diameter with concomitantly decreasing
micropore volume, and likely by a preferential location of the acid sites in the micropores as deduced by
the presence of more H-bonded species (3520 cm-1) in the IR spectrum for the sample with higher
micropore volume in Figure 3. The effect of changing the pore architecture on the catalytic performance
28
This article is protected by copyright. All rights reserved.

10.1002/cphc.201701219
ChemPhysChem
is shown in Figure 10. The sites in the samples with larger mesopore diameters were characterized by
slightly smaller turnover frequencies, that is, the conversion was slightly lower at equal reactant-to-site
ratios and similar reaction times.
Comparison of selectivities at similar conversion provides insights into the reaction pathways leading
to different products. All these catalysts preferentially yielded the branched (Pathway A) condensation
product, the associated fission products, and undesired secondary aldol condensation products formed
through nucleophilic attack of another molecule of benzaldehyde by α,β-unsaturated ketones. The
selectivity toward the fission product decreased in the order PrSO3H/SBA-15 (50 °C) > PrSO3H/SBA-15 (80
°C) > PrSO3H/SBA-15 (100 °C) > PrSO3H/SBA-15 (120 °C).
191
56
59
5
100
80
60
40
20
0
PrSO₃H/
SBA-15(50 °C)
PrSO₃H/
SBA-15(80 °C)
PrSO₃H/
SBA-15(100 °C)
PrSO₃H/
SBA-15(120 °C)
Benzaldehyde Conversion
Condensation A
Addition A
Condensation B
Addition B
Fission A
Secondary products
Figure 10. Effect of increasing mesopore diameter and decreasing micropore volume on product
selectivity for aldol condensation over PrSO3H/SBA-15. Reaction conditions: 3 M butanone, 1 M
benzaldehyde, 0.1 M hexadecane (internal standard); 140 °C, and ratio of benzaldehyde to acid sites
of 517:1. Reaction time 8 h. The Weisz-Prater number is indicated on the benzaldehyde conversion
bar. Error bars represent one standard deviation above and below mean obtained from three replicate
tests. (The error bars for selectivites in stacked bars here are shown in Figure S9.)
PrSO3H/SBA-15 (120 °C) yielded a relatively higher amount of secondary condensation products. The
mesopore channels in all samples are wide enough to accommodate these products, and the pore size
should not affect the yield. An increase in secondary condensation products with increasing aging
29
This article is protected by copyright. All rights reserved.