Lennon, D., McFarlane, A., Geller, H., Silverwood, I. P., Cooper, R. I.

,
Watkin, D. J., Parker, S. F., and Winfield, J. M. (2016) The application of
inelastic neutron scattering to investigate the interaction of methyl
propanoate with silica. Physical Chemistry Chemical Physics, 18(26), pp.
17210-17216. (doi:10.1039/C6CP01276K)

This is the author’s final accepted version.

There may be differences between this version and the published version.
You are advised to consult the publisher’s version if you wish to cite from
it.

http://eprints.gla.ac.uk/119093/

Deposited on: 16 May 2016

Enlighten – Research publications by members of the University of Glasgow

http://eprints.gla.ac.uk
Draft: 19th April 2016
Paper for submission to:-
Physical Chemistry Chemical Physics, themed issue on Neutron Scattering in Catalysis and Energy Materials

The application of inelastic neutron scattering to investigate the interaction of

methyl propanoate with silica

Andrew R McFarlane,a Hannah Geller,a Ian P. Silverwood,a Richard I. Cooper,b David J. Watkin,b
Stewart F. Parker,b John M. Winfielda and David Lennon*a

a
School of Chemistry, Joseph Black Building, University of Glasgow, Glasgow, G12 8QQ, UK.
E-mail: David.Lennon@glasgow.ac.uk
b
Chemical Crystallography, Chemistry Research Laboratory, Mansfield Road, Oxford OX1 3TA, UK
c
ISIS Facility, STFC Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, UK.

Proofs to:–
Professor David Lennon,
School of Chemistry,
Joseph Black Building,
The University of Glasgow,
Glasgow, G12 8QQ.
U.K.

Email: David.Lennon@glasgow.ac.uk
Telephone: (+44)-(0)-141-330-4372

1
Abstract

A modern industrial route for the manufacture of methyl methacrylate involves the reaction of methyl
propanoate and formaldehyde over a silica-supported Cs catalyst. Although the process has been
successfully commercialised, little is known about the surface interactions responsible for the forward
chemistry. This work concentrates upon the interaction of methyl propanoate over a representative silica. A
combination of infrared spectroscopy, inelastic neutron scattering, DFT calculations, X-ray diffraction and
temperature-programmed desorption is used to deduce how the ester interacts with the silica surface.

Keywords: methyl propanoate; silica; adsorption; inelastic neutron scattering, infrared spectroscopy.

2
1. Introduction
The synthesis of acrylic compounds represents a major component of the modern chemical manufacturing
sector. The products of this industry include furniture, automobile components, mobile phone screens and
high definition LCD panels for use in TV and computer screens. Methyl methacrylate (CH2C=C(CH3)-C(=O)-
OCH3, MMA) is the essential ingredient for all these materials and has a global demand in excess of 2.2
million tonnes.1 In 2005 there were over 30 manufacturing plants worldwide,2 demonstrating the economic
significance of chemistry related to the synthesis of methyl methacrylate. Many companies have been
involved in developing the synthesis of MMA and over 17 possible synthetic routes are reported.1

The conventional route to MMA is the acetocyanohydrin (ACH) process. This reaction produces HCN from
CH4 and NH3, which is then reacted with acetone to form ultimately MMA. However, throughout the world,
new routes using a variety of reagents are currently being developed.1,3 One of the main drivers for this
development is the avoidance of manufacturing, storing and transporting the highly toxic hydrogen cyanide;
the risk potential for this chemical is dependent on the specific process operation. Another problem with the
ACH reaction is the formation of bisulfate by-products, which reduce the atom economy of the overall
process. Lucite International, a global leader in the design, development and manufacture of acrylic-based
products,4 has developed the Alpha process 2,4
(see Scheme 1) which, in addition to representing dramatic
improvements in operational performance, makes uses of readily available raw materials (ethene, methanol,
and carbon monoxide) and combines a homogeneous catalytic stage followed by a heterogeneous catalytic
stage to produce MMA in a more efficient and sustainable manner.2 Such innovation in the chemical
manufacturing industry has the potential to revolutionise operational practices. As part of that process,
there needs to be an increasing awareness of the chemical interactions that favour the chemical
transformations under consideration.

Scheme 1: The Alpha Process; the manufacture of methyl methacrylate from ethene, methanol and CO via
methyl propanoate.2,4,5

3
An important part of the Alpha Process involves a reaction between methyl propanoate (CH3COOCH2CH3),
and formaldehyde (H2CO); special handling procedures are required for the large scale use and storage of
formaldehyde. Typically this heterogeneously catalysed reaction takes place over a modified silica catalyst,
e.g. caesium nitrate supported on high surface area silica.5 The reaction is thought to be an example of base
catalysis, with the catalyst presenting basic sites that control the chemisorption and subsequent reaction of
the two reagents. However, a survey of the literature indicates a paucity of work examining the interaction
of oxygenates and esters in particular on oxide surfaces. Preliminary studies on silicas similar to that used
with the Alpha process are believed to indicate the presence of both weak Brønsted acid and Brønsted base
sites,2 although the evidence for these assertions is not in the public domain.

This paper is divided into two parts. To our knowledge, methyl propanoate has been neither structurally nor
spectroscopically characterised; thus, Part 1 provides a comprehensive description of the solid state
structure and vibrational spectroscopy of methyl propanoate in the gas, liquid and solid phases, with
assignments supported by periodic density functional theory (DFT). In Part 2 we characterise the adsorption
of methyl propanoate on a representative silica (Fuji Q-10 silica spheres). Previous work by Jackson and co-
workers examined the adsorption of acetic acid on this particular silica, as well as Cs-doped Q-10, with the
latter constituting model methyl methacrylate synthesis catalysts.6 That work followed on from a patent
filed in 1999 by Jackson and co-workers at ICI (UK) entitled ‘novel catalyst for manufacture of ethylenically
unsaturated acids or esters, especially for manufacture of methyl methacrylate’.7 Silica spheres, as opposed
to silica powder, are attractive to industrial use as bulk handling properties are simplified. Against this
background, Fuji Q-10 silica is examined in this communication. The combined structural and spectroscopic
investigations establish the adsorption geometry for methyl propanoate adsorbed on this silica. The work is
not extended to consider Cs-doped silica samples.

2. Experimental section
2.1 X-ray crystallography
A sample of methyl propanoate (Aldrich, 99%, used as received) was sealed in a 0.5 mm Lindeman tube and
flash-frozen in liquid nitrogen to confirm that the material did not form a glass. The sample was then
mounted on a goniometer head with both arcs and translational adjustment so that the tube could be made
co-linear with the diffractometer phi axis. The Enraf–Nonius KCCD diffractometer was set to θ =20°, ω=180°,
χ =90° and Dx =165 so that the φ axis was horizontal and perpendicular to the X-ray beam, and the locally
constructed computer-controlled sample heating element installed.

4
The sample was chilled to a polycrystalline mass at 150 K using an Oxford Cryosystems Series 600
Cryostream.8 The sample was zone-refined until it looked clear, but a short φ rotation image indicated that
the sample still contained several large crystals lying parallel to the Lindeman tube axis. Further zone
refinement failed to produce a single-crystal.

A new sample was prepared in a 0.1 mm Lindeman tube. This also produced aligned poly-crystals, but
prolonged zone refinement converted the sample to a major component and several minor components.
This size differentiation enabled diffraction images collected at κ=60° (to offset the tube axis from the scan

axis) to be indexed. All three cell angles were quite close to 90°. A triclinic data set was collected in P1 using
four different scan regions. Because of the possibility that the cell was really monoclinic, reflections from
each scan region were indexed independently. All gave essentially the same triclinic cell. The whole data set
was then processed using the initial indexing, and the cell parameters refined from all of the data.

The H atoms were all located in a difference map, but those attached to carbon atoms were repositioned
geometrically. The H atoms were initially refined with soft restraints on the bond lengths and angles to
regularize their geometry (C—H in the range 0.93–0.98 Å) and Uiso(H) (in the range 1.2–1.5 times Ueq of the
parent atom), after which the positions were refined freely.

Data collection: COLLECT 9.; data reduction and cell refinement: DENZO/SCALEPACK 10; programme used to
11 12
solve structure: Superflip ; programme used to refine structure: CRYSTALS ; molecular graphics:
CAMERON 13. The data has been deposited with the Cambridge Structural Database (CSD) with deposition
number: CCDC 1453909.

2.2 Computational methods

Periodic density functional theory (periodic-DFT) calculations of the crystal structure were carried out using
the plane-wave pseudopotential method implemented in the CASTEP code.14,15 Exchange and correlation
were approximated using the PBE functional. Norm-conserving pseudopotentials with a plane-wave cut-off
energy of 830 eV were used. Brillouin-zone sampling of electronic states was performed on a 4×4×2
Monkhorst-Pack grid. The equilibrium structure, an essential prerequisite for lattice dynamics calculations
was obtained by BFGS geometry optimization, after which the residual forces were converged to zero within
|0.0008| eV Å-1. Phonon frequencies were obtained by diagonalisation of dynamical matrices computed
using density functional perturbation theory (DFPT).15 An analysis of the resulting eigenvectors was used to
map the computed modes to the corresponding irreducible representations of the point group and assign
IUPAC symmetry labels. DFPT was also used to compute the dielectric response and the Born effective

5
charges and from these the mode oscillator strength tensor and infrared absorptivity were calculated.
Conformational analysis of the isolated molecule was carried out with Gaussian 03.16 The B3LYP functional
with the 6-311g(d) basis set was used to generate a 2D relaxed potential energy scan about the central C2–
C3 and the C3–O2 bonds (see Fig. 1 for the numbering scheme). To model methyl propanoate on silica, the
B3LYP functional with the 6-31g basis set was used. INS spectra were generated from the output of CASTEP
and Gaussian with ACLIMAX.17 We emphasise that the transition energies determined by either CASTEP or
Gaussian have not been scaled.

Fig. 1 Numbering scheme for methyl propanoate.

2.3 Temperature-programmed desorption and infrared spectroscopy

The silica used in all experiments was provided by Fuji (Cariact Q10). As received spheres of 2-3 mm
diameter exhibited a surface area of 359 m2 g-1 ,a pore volume of 1.01 cm3 g-1 and a hydroxyl group density of
2.12 mmol g-1.6 Samples were prepared by grinding the silica into particles of size fraction 250-500 µm.
Adsorption measurements were performed using a facility described elsewhere.18 Briefly, silica samples
were loaded in to a ¼″ od stainless steel tubular reactor. Methyl propanoate (Aldrich, 99 % purity) was
contained in a bubbler arrangement in advance of the reactor so that methyl propanoate vapour could be
exposed to the catalyst in a controlled manner. Prior to adsorption measurements, activation of the silica
was carried out by heating the silica under flowing helium (10 ml min-1) for one hour at 623 K.6 Post-
activation, the sample was allowed to cool to 353 K and maintained at this temperature during exposure to a
vapour stream of methyl propanoate. After a defined exposure period, the methyl propanoate flow was
terminated whilst maintaining the helium flow for 30 min at 353 K. Thereafter, the sample was allowed to
cool to 303 K in a continuous stream of helium. These arrangements disfavoured formation of physisorbed
ester.

6
A quadrupole mass spectrometer (MKS Microvision Plus) was mounted downstream of the reactor, sampling
the eluting gases via a differentially pumped capillary line and a sintered metal filter. Temperature-
programmed desorption measurements were performed at a heating rate of 12 K min-1 up to 653 K whilst
the mass spectrometer monitored the methyl propanoate signal intensity. In situ infrared spectroscopy of
methyl propanoate adsorption was performed using a diffuse reflectance Smart Collector (Spectra-Tech)
environmental cell housed inside a Nicolet Nexus FTIR spectrometer. Silica activation and methyl
propanoate exposure were performed in the same manner as that employed for the tubular reactor. Spectra
presented are background subtracted, using a spectrum of the activated silica as a background.

Transmission infrared spectroscopic measurements of methyl propanoate as a gas, liquid and solid utilised
the same FTIR instrument but respectively coupled to the following sample cells: a Graseby-Specac 5660
heated gas cell, a Specac GS01800 liquid cell and a Specac GS21252 variable temperature cell.

2.4 Inelastic neutron scattering spectroscopy

Inelastic neutron scattering (INS) spectroscopy is a complementary form of vibrational spectroscopy that
emphasises hydrogen motion.19 This has the consequence that silica is essentially transparent to neutrons
and the entire 0 – 4000 cm-1 range is accessible, in particular there is no spectral cut-off at ~1400 cm-1 as
there is for infrared spectroscopy. INS spectra were recorded using the MAPS 20 and TOSCA 21 spectrometers
at ISIS.22 On TOSCA the resolution is ~ 1.25% of the energy transfer across the entire energy range, while on
MAPS, under the conditions used here, it is ~1.5% of the incident energy at the largest energy transfer and
degrades with decreasing energy transfer. Thus TOSCA provides excellent energy resolution at energy
transfers <1600 cm-1, at larger energy transfer MAPS provides better resolution by virtue of the access to low
momentum transfer.20 TOSCA and MAPS are highly complementary and enable the complete range of
interest, 0 – 4000 cm-1, to be covered with reasonable resolution.

For the INS experiments, silica (Fuji Q10 silica spheres) was dried using a previously described gas manifold 23
in a flow-through Inconel cell under flowing helium at 623 K for 90 mins and left to cool overnight. This
sample was then measured as a background. A second sample was similarly dried, allowed to cool to 353 K
and methyl propanoate introduced as a vapour stream entrained in helium gas via a Dreschel bottle. The INS
spectra of the samples were then recorded.

7
3. Results and Discussion
Part 1: Characterisation of methyl propanoate
The structure of methyl propanoate as determined by a single crystal X-ray diffraction study is shown in Fig.

2. The crystal is triclinic, space group P1 with Z' = 2 (Z= 4) (see Table S1 for the crystal data). One of the
independent molecules is almost planar with Cs symmetry, the other has a torsion angle of 159.7º about the
central C2–C3 bond and C1 symmetry. There is no evidence of disorder or pseudo-symmetry. A periodic-DFT
calculation (CASTEP) of the structure using the crystal structure as the initial input gives results in good
agreement with the observed structure. A comparison of selected structural parameters is given in Table S2.

Fig. 2 P1 crystal structure of methyl propanoate at 150 K. The molecules at the bottom right and top left
have approximate Cs symmetry and those at the bottom left and top right have C1 symmetry.

The conformational isomerism was also investigated with Gaussian 03 by starting from the planar Cs
structure and carrying out a relaxed potential energy scan about the central C2–C3 bond and also about the
C–OCH3 (C3–O2) bond. The results are shown in Fig. 3. It can be seen that the planar Cs conformation is the
lower energy structure. The C1 conformation found in the solid state is ~5 kJ mole-1 higher in energy.

8
Fig. 3 2D potential energy scan of methyl propanoate. The four corners correspond to methyl propanoate
with Cs symmetry, the arrow indicates the structure of the molecule with C1 symmetry.

The infrared spectra of the gas, liquid and solid are shown in Fig. 4a-c and the inelastic neutron scattering
(INS) spectrum of the solid is presented in Fig. 5a. Calculation of the spectra using the energy minimised
structure finds all real transition energies, confirming that this is a stable structure. The calculated infrared
and INS spectra of the solid are also shown in Figs. 4d and 5b. It can be seen that while the peaks are shifted
lower energy, generally the relative intensities of the modes are well-reproduced. (The discrepancy in
energy is apparent because we do not scale the spectra). Note that the calculated infrared spectrum
assumes a constant linewidth; this is not correct for the C–O and C=O stretch modes at 1160 and 1700 cm-1
(they are broadened by the hydrogen bonding). The agreement in the C–H stretch region is poorer because
these modes are strongly affected by anharmonicity that is not included in our model. In the low energy
region, the discrepancy in the calculated positions is emphasised by the large x-scale used. In Fig. 5b, the
pattern in the range 100 – 300 cm-1 is repeated by the calculated spectrum. Below 100 cm-1 the acoustic
modes occur, these are seen in the INS spectrum but are not present in the calculation because it is a Γ-point

9
only calculation where the acoustic modes have zero energy. To reproduce these would require an
expensive full dispersion calculation, which we could not justify for what would be a relatively minor
improvement. Table 1 lists modes that are of relevance to the investigation of the adsorbed species
described in Part 2; a complete assignment is given in Table S3.

Table 1 hereabouts

Fig. 4 Infrared spectra of methyl propanoate: (a) gas, (b) liquid (both at room temperature), (c) solid at 130 K
and (d) CASTEP calculated solid state spectrum.

10
Fig. 5 INS spectra of: (a) solid methyl propanoate at 20 K and (b) CASTEP calculated solid state spectrum. The
range 300 – 1800 cm-1 is ordinate expanded ×4 relative to the 0 – 300 cm-1 region.

11
Part 2: The adsorption of methyl propanoate on silica
Methyl propanoate was adsorbed on silica (activated by drying at 623 K) at 353 K (to minimise physisorption)
then purged at 303 K for two hours. The sample was then heated to 673 K in 50 K steps, cooled to 303 K
after each step and the infrared spectrum recorded, Fig. 6. Methyl propanoate is adsorbed intact, Fig. 6a,
and also apparently desorbs intact, as there is no evidence of decomposition products remaining on the
surface, Fig. 6f. On initial adsorption, the O–H stretch region clearly shows a conversion of isolated hydroxyls
(negative-going peak at 3737 cm-1) to a hydrogen-bonded species (Fig. 6a, intense broad peak at 3416 cm-1)
that progressively shifts up to 3555 cm-1, Fig. 6f, as methyl propanoate desorbs (indicated by the arrow). As
the isolated hydroxyls are regenerated, so the peak at 3737 cm-1 disappears. This is because Fig. 6 is a
difference spectrum, the background spectrum has the isolated hydroxyls present, thus we are looking at
changes in their numbers. The disappearance of the feature at 3737 cm-1 is consistent with essentially
complete regeneration of the initial population of isolated hydroxyls. Temperature-programmed desorption
on a similarly prepared sample showed maximum desorption at 473 K and essentially complete desorption
by 573 K (see Fig. S2), consistent with Fig. 6.

Fig. 6 Diffuse reflectance infrared difference spectra of the desorption of methyl propanoate on silica.
Spectra were recorded at 303 K after heating to the stated temperature. (a) 323 K, (b) 373 K, (c) 423 K, (d)
473 K, (e) 523 K and (f) 573 K. (c) is ordinate expanded ×2, (d) and (e) ×4 and (f) ×16 all relative to (a) and (b).
The arrow shows the shift to higher energy of the hydrogen-bonded O–H peak as the methyl propanoate
desorbs.

12
Fig. 7 compares the infrared spectrum of solid methyl propanoate (7a) with that of the adsorbed species
(7b). The carbonyl stretch is at 1727 cm-1 in both cases, although the band is much broader for the adsorbed
species. However, there are noticeable differences in the 1300 – 1500 cm-1 region of the spectrum. The
complexity in the spectrum of the solid material is the result of the presence of two conformers in the crystal
structure. The much simpler spectrum of the adsorbed species would suggest that there is predominantly a
single conformer. The periodic-DFT calculation of the solid state material was repeated but with all but one
molecule removed from the unit cell. This was carried out for both conformers and the results are shown in
Fig. 7 for the Cs (7c) and C1 (7d) conformers. Neither exactly reproduces the spectrum of the adsorbed
species but the Cs conformer (7c) is the better match, suggesting that this is the dominant structure.

13
Fig. 7 Infrared spectra: (a) solid methyl propanoate at 113 K, (b) of methyl propanoate on silica after
desorption at 50 °C, (c) calculated spectrum of the Cs conformer and (d) calculated spectrum of the C1
conformer. (c) and (d) are plotted on the same scale.

The INS spectra provide a complementary view of the system. Fig. 8 shows the C–H and O–H stretch region
of activated silica (8a) and after adsorption of methyl propanoate (8b). Figure 8 clearly shows the O–H
stretch downshift from 3750 to 3485 cm-1 as a result of methyl propanoate hydrogen-bonding to the
hydroxyls. This is in good agreement with the infrared results which show the peak at 3416 cm-1, however,
the precise location of the stretch mode in the infrared is hampered by the strong electrical anharmonicity
that results from the hydrogen-bonding, this is irrelevant to INS and the location of the fundamental is easily
seen.

14
Fig. 8 INS spectra of: (a) silica after activation at 623 K and (b) methyl propanoate on silica.

As stated earlier, one of the advantages of INS spectroscopy is that common catalyst supports, carbon,
alumina, silica, are essentially transparent to neutrons. The benefits of this are shown in Fig. 9 which
compares the solid state spectrum of methyl propanoate (9a) with the difference spectrum, generated by
subtraction of the activated silica from that of the adsorbed material (9b). Over the range 400 – 1600 cm-1
the spectra are very similar, confirming molecular adsorption. This is not the case in the 0 – 400 cm-1 region;
in the solid state spectrum there are many more bands. This arises from four molecules with two distinct
conformations in the unit cell, whereas the infrared spectra indicate (predominantly) a single conformer
adsorbed on the surface.

15
Fig. 9 INS spectra of: (a) methyl propanoate and (b) methyl propanoate on silica, after subtraction of the
activated silica.

Both the infrared and the INS spectra show that methyl propanoate is hydrogen bonded to the silica
hydroxyls. The 2D PES scan, Fig. 3, shows that the planar Cs conformer is the lower energy structure in the
gas phase and the infrared spectrum suggests that this situation is retained on interaction with the surface.
Methyl propanoate has two possible hydrogen-bond acceptor sites: the carbonyl oxygen, Fig. 10a, or the
ester oxygen, Fig. 10b. To discriminate between the two possibilities the two systems were modelled with
Gaussian 03 with the results shown in Table 2.

Fig. 10 Possible adsorption modes of methyl propanoate on silica: (a) via the carbonyl oxygen or (b) via the
ester oxygen.

16
Table 2 Selected calculated vibrational transition energies for silica, free and adsorbed methyl propanoate.
Values in brackets are the difference between the free and the adsorbed species.
SiO2 Methyl Adsorbed via Adsorbed via Experimental
propanoate carbonyl O ester O
O–H stretch 3765 3508 (-257) 3576 (-189) (-265)
/ cm-1
C=O stretch 1730 1699 (-31) 1746 (+16) (-36)
-1
/ cm
C2–C3 + C3–O1 836 842 (+6) 808 (-28) (+12)
in-phase stretch
-1
/ cm

The absolute values are in modest agreement only with the data, however, the differences are likely to be
much more reliable. For the species adsorbed via the carbonyl oxygen, the O–H stretch undergoes a large
downshift, the carbonyl stretch downshifts and the skeletal stretch mode slightly upshifts, for adsorption via
the ester oxygen, there is a somewhat smaller downshift in the O–H stretch mode, the carbonyl upshifts and
the skeletal mode downshifts. The experimentally observed pattern is in almost quantitative agreement with
the predictions of adsorption via the carbonyl oxygen.

It should be noted that the spectra of adsorbed methyl propanoate on silica resembles that of the solid in
some respects, in particular, the carbonyl stretch occurs at the same transition energy (± a few
wavenumbers). Inspection of the solid state structure of methyl propanoate shows that there are weak
hydrogen-bonding interactions of the type C–HO=C present (HO = 2.3 – 2.9 Å, ∠ C–HO ~ 169°).
Thus, hydrogen-bonding to the silica hydroxyls mimics to some extent the solid state, which results in the
carbonyl transition energies being similar.

The possibility that there is solid methyl propanoate on the silica surface can be discounted for several
reasons. The melting and boiling points of methyl propanoate are 185 and 353 K respectively; all the sample
dosing was carried out at 353 K or above to minimise physisorption and the infrared spectra were recorded
at 303 K, well within the liquid phase. The INS spectra in the lattice mode region, (see the 0 – 400 cm-1 region
of Fig. 9), is distinctly different as only the internal modes of methyl propanoate are observed; those
assigned (Table S3) to whole body motions are absent. The characteristic downshift and increase in intensity
of the silica hydroxyls on addition of methyl propanoate is also inconsistent with the presence of solid methyl
propanoate.

17
Figure 10(a) defines the adsorption geometry for methyl propanoate on the silica, with the silica providing
Brønsted acid sites that bind the ester via the carbonyl group. Further work examining the adsorption and
reaction of reagents on a Cs modified silica would be helpful in understanding the specific interactions
associated with the industrial process. For example, INS would be insightful for determining how the
addition of a Cs salt modifies the silica hydroxyl groups.

4. Conclusions
Methyl propanoate has been characterised by a combination of vibrational spectroscopy (IR spectroscopy
and INS, backed up by DFT calculations) and single crystal X-ray diffraction (Section 3, Part 1). The interaction
of the ester with silica was then investigated by IR, INS and temperature-programmed desorption. The
following conclusions can be drawn.

• The structure of methyl propanoate as determined by X-ray diffraction is triclinic, space group P1 . DFT
calculations show the planar Cs conformation to be the lower energy structure. Experimental IR and INS
spectra of methyl propanoate are combined with associated DFT calculations to provide a complete
vibrational assignment.
• Adsorption experiments of methyl propanoate on silica employing temperature-programmed IR
spectroscopy and temperature-programmed desorption show the methyl propanoate to be molecularly
adsorbed on the silica. Both IR and INS spectra of the adsorption complex show methyl propanoate to
be hydrogen-bonded to silica hydroxyls, with adsorption occurring via the carbonyl group of methyl
propanoate.
• Silica hydroxyl groups constitute weak Brønstead acid sites that bind methyl propanoate.

Acknowledgements
The STFC Rutherford Appleton Laboratory is thanked for access to neutron beam facilities. Computing
resources (time on the SCARF compute cluster for the CASTEP calculations and the Gaussian 03 calculations)
was provided by STFC’s e-Science facility. Dr David Johnson and Dr Sabina Ziemian (Lucite International) are
thanked for helpful discussions.

18
References
1. K. Nagai, Appl. Catal. A: Gen., 2001, 221, 367.
2. A. Tullo, Chem. Engineering News, 2009, 87(42), 22-23.
3. J. Tai and R. J. Davis, Catal. Today, 2007, 123, 42.
4. http://www.luciteinternational.com/ (accessed 20/02/2016).
5. K. AlGhamdi, J. S. J. Hargreaves and S. D. Jackson in Metal Oxide Catalysis, Eds. S.D. Jackson and J.S.J.
Hargreaves, Wiley-VCH, Weinheim, 2009, p.819-843.
6. S.D. Jackson, G.J. Kelly and D. Lennon, React. Kinet. Catal. Lett., 2000, 70, 207.
7. S.D. Jackson, D.W. Johnson, J.D. Scott, G.J. Kelly, B.P. Williams (Imperial Chemical Industries Plc, UK),
Patent PCT Int. Appl. WO9952628-A1 (1999).
8. J. Cosier and A. M. Glazer, J. Appl. Cryst., 1986, 105, 107.
9. Nonius (2001). COLLECT. Nonius BV, Delft, The Netherlands.
10. Z. Otwinowski and W. Minor, Methods in Enzymology, 276, Eds. C. W. Carter Jr and R. M. Sweet, pp.
307–326. Academic Press, New York, 1997.
11. L. Palatinus and G. Chapuis, J. Appl. Cryst., 2007, 40, 786–790.
12. P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and D. J. Watkin, J. Appl. Cryst., 2003, 36,
1487.
13. D. J. Watkin, C. K. Prout and L. J. Pearce,. CAMERON, Chemical Crystallography Laboratory, Oxford,
UK, 1996.
14. S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. J. Probert, K. Refson and M. C. Payne, Z. Krist.,
2005, 220, 567.
15. K. Refson, P. R. Tulip, S. J. Clark, Phys. Rev. B, 2006, 73, 155114.
16. Gaussian 03, Revision B.05, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala,
K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels,
M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W.
Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Wallingford CT,
2004.

19
17. A. J. Ramirez-Cuesta, Comput. Phys. Commun., 2004, 157, 226.
18. T. Lear, R. Marshall, J.A. Lopez-Sanchez, S.D. Jackson, T. M. Klapötke, G. Rupprechter, M. Bäumer, H-J.
Freund and D. Lennon, J. Chem. Phys., 2005, 123, 174706.
19. P.C.H. Mitchell, S.F. Parker, A.J. Ramirez-Cuesta and J. Tomkinson, Vibrational spectroscopy with
neutrons, with applications in chemistry, biology, materials science and catalysis, World Scientific,
Singapore, 2005.
20. S.F. Parker, D. Lennon and P.W. Albers, Applied Spectroscopy, 2011, 65, 1325-1341.
21. D. Colognesi, M. Celli, F. Cilloco, R.J. Newport, S.F. Parker, V. Rossi-Albertini, F. Sacchetti, J.
Tomkinson and M. Zoppi, Appl. Phys. A 74 [Suppl.] (2002) S64–S66.
22. http://www.isis.stfc.ac.uk/ (accessed 19th April 2016)
23. I.P. Silverwood , N.G. Hamilton , A. McFarlane , R.M. Ormerod, T. Guidi, J. Bones, M.P. Dudman, C.M.
Goodway, M. Kibble, S.F. Parker and D. Lennon, Rev. Sci. Inst., 2011, 82, 034101.

20
Table 1 Vibrational transition energies of methyl propanoate in the solid state and adsorbed on silica.
Mode Solid state / cm-1 Adsorbed on SiO2 / cm-1
C=O stretch 1728 1726
C4 Methyl asym def 1469, 1456 1468
C4 Methyl sym deformation 1445, 1432 1442
Methylene scissors 1417 1427
C1 Methyl sym deformation 1384 1384
Methylene wag 1356 1364
C1 methyl rock 1092 1090
C1 methyl rock 1024 1029
C1–C2 + O1–C4 out-of-phase 962 953
stretch
C2–C3 + C3–O1 in-phase stretch 858 859
Methylene rock 809 802
C=O out-of-plane bend 581 577
C2–C3–O1 in-plane bend 443 445
C1–C2–C3 + C3–C1–C4 339 343
out-of-phase, in-plane bend
C1 methyl torsion 263, 244, 231 225
C1 methyl torsion and 204, 192, 163 170
C3–O1 torsion

21