Hydrogenation of butanal over silica-supported Shvo's catalyst and its use for the gas-phase conversion of propene to butanol via tandem hydroformylation and hydrogenation

Article abstract of DOI:10.1016/j.jcat.2013.11.012

The objective of the present study was to develop a heterogeneous catalyst for the hydrogenation of butanal that could function in the presence of CO and propene and, hence, could be used in a tandem reactor to enable the gas-phase conversion of propene and synthesis gas to butanol. To this end, we investigated the activity of silica-supported Shvo's catalyst (Shvo/SiO₂) for the gas-phase hydrogenation of butanal. Experiments were performed to determine the kinetics of n- and iso-butanal hydrogenation. The apparent activation energies and the apparent partial pressure dependencies of n- and iso-butanal, H ₂, and CO on the rates of n- and iso-butanol formation were determined. A mechanism for butanal hydrogenation was proposed to rationalize the observed kinetics and some of the reaction intermediates were observed by in situ infrared and ³¹P MAS NMR spectroscopy. It was found that Shvo/SiO₂ was inhibited by SX (SX = sulfoxanthphos) and CO, and is inactive for alkene hydrogenation. The tandem catalytic conversion of propene and synthesis gas to butanol was then carried out using a SX-Rh supported ionic liquid phase (SILP) catalyst to promote the hydroformylation of propene to butanal and Shvo/SiO₂ to promote the hydrogenation of butanal to butanol. The rate expressions describing the kinetics of each of the catalysts were then used to predict operating conditions required to achieve high conversion of propene to butanol. Under the most favorable conditions examined (H₂/CO = 10), an overall yield of 13% to butanol was achieved with 15% propene conversion and 90% aldehyde conversion at a temperature of 413 K.

Full text of DOI:10.1016/j.jcat.2013.11.012

Journal of Catalysis 311 (2014) 52–58
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Hydrogenation of butanal over silica-supported Shvo’s catalyst and its
use for the gas-phase conversion of propene to butanol via tandem
hydroformylation and hydrogenation
⇑
David G. Hanna, Sankaranarayanapillai Shylesh, Pedro A. Parada, Alexis T. Bell
Department of Chemical and Biomolecular Engineering, University of California, Berkeley, CA 94720, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The objective of the present study was to develop a heterogeneous catalyst for the hydrogenation of but-
anal that could function in the presence of CO and propene and, hence, could be used in a tandem reactor
to enable the gas-phase conversion of propene and synthesis gas to butanol. To this end, we investigated
the activity of silica-supported Shvo’s catalyst (Shvo/SiO2) for the gas-phase hydrogenation of butanal.
Experiments were performed to determine the kinetics of n- and iso-butanal hydrogenation. The appar-
ent activation energies and the apparent partial pressure dependencies of n- and iso-butanal, H2, and CO
on the rates of n- and iso-butanol formation were determined. A mechanism for butanal hydrogenation
was proposed to rationalize the observed kinetics and some of the reaction intermediates were observed
by in situ infrared and 31P MAS NMR spectroscopy. It was found that Shvo/SiO2 was inhibited by SX
(SX = sulfoxanthphos) and CO, and is inactive for alkene hydrogenation. The tandem catalytic conversion
of propene and synthesis gas to butanol was then carried out using a SX-Rh supported ionic liquid phase
(SILP) catalyst to promote the hydroformylation of propene to butanal and Shvo/SiO2 to promote the
hydrogenation of butanal to butanol. The rate expressions describing the kinetics of each of the catalysts
were then used to predict operating conditions required to achieve high conversion of propene to buta-
nol. Under the most favorable conditions examined (H2/CO = 10), an overall yield of 13% to butanol was
achieved with 15% propene conversion and 90% aldehyde conversion at a temperature of 413 K.
Ó 2013 Elsevier Inc. All rights reserved.
Received 27 August 2013
Revised 1 November 2013
Accepted 12 November 2013
Available online 15 December 2013
Keywords:
Butanal
Hydrogenation
Butanol
Shvo
Kinetics
1. Introduction
of propene and synthesis gas to butanol. Achievement of this goal
requires identification of a heterogeneous hydroformylation cata-
Roughly 2 ꢀ 106 tons of butanol are produced annually [1] for
use as a plasticizer, an industrial solvent, and an intermediate in
the production of butyl acetate, a key ingredient in lacquers and
varnishes [2]. The demand for butanol is expected to increase in
the future as a consequence of recent studies showing that butanol
is a viable alternative to ethanol as an additive to gasoline [3,4].
While many processes exist for the production of butanol, such
as the aldol condensation of ethanal [5], oxidation of butane [6], or
the enzymatic fermentation of sugars [3,7], the overwhelming
majority of butanol is produced in a three-step process involving
the homogeneously catalyzed hydroformylation of propene, sepa-
ration of the resulting butanal, and subsequent hydrogenation of
butanal. Separation of butanal is necessary because aldehyde
hydrogenation catalysts such as supported Pt and Pt–Sn [8], Rh–
Sn [9], and Ni [10] are poisoned by CO, as well as being active for
alkene hydrogenation. Therefore, it would be highly desirable to
develop a catalyst system for the tandem, gas-phase conversion
lyst and a heterogeneous aldehyde hydrogenation catalyst that
functions in the presence of CO and is inactive for alkene
hydrogenation.
Recent studies have shown that the gas-phase hydroformyla-
tion of propene can be carried out on a heterogeneous catalyst pre-
pared by impregnating a Rh–phosphine complex dissolved in ionic
liquid onto a silica support [11,12]. Such catalysts, referred to as
supported ionic liquid phase (SILP) catalysts, are stable and active
under continuous gas-phase reaction conditions. We have recently
shown that the exceptional stability of these catalysts is due to the
interaction of the Rh–phosphine complex with silanol groups pres-
ent on the support surface and have discussed the mechanism and
kinetics propene hydroformylation over this catalyst [13,14].
Several attempts to perform the tandem conversion of Cn al-
kenes and synthesis gas to the Cn+1 alcohol in a single reactor sys-
tem have been reported. Drent and coworkers have shown that
chloride anion promoted palladium-phosphine complexes could
be used to convert propene to n- and iso-butanol with 85% selectiv-
ity [15]. High turnovers (1000 hꢁ1) and selectivity (99%) were also
observed in the conversion of C8–C10 internal alkenes into their
⇑
Corresponding author. Fax: +1 510 642 4778.
E-mail address: bell@cchem.berkeley.edu (A.T. Bell).
0021-9517/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2013.11.012

D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
53
corresponding alcohols [15]. A halogen-free system studied by
Zakzeski et al. demonstrated that the conversion of hexene to hept-
anol could be achieved utilizing HRh(PPh3)2(CO) as the hydrofor-
mylation catalyst and H2Ru(PPh3)3CO as the hydrogenation
catalyst [16]. It was noted, though, that the hydrogenation of the
aldehyde would only occur after the CO was consumed or removed
from the reactor because the Ru complex was inhibited by CO [16].
More recently, Nozaki and coworkers have demonstrated that it is
possible to carry out the hydroformylation of propene to butanal
and the subsequent hydrogenation of butanal to butanol in a solu-
tion containing a Rh complex, Shvo’s catalyst, and bidentate phos-
phine ligand, sulfoxantphos (SX) [17]. Using this approach, butanol
yields of up to 85% were obtained [17]. The authors suggest that
the presence of excess SX influences the activity and/or the che-
moselectivity of Shvo’s catalyst to preferentially hydrogenate alde-
hydes [17]. These findings motivated us to explore the possibility
of supporting Shvo’s catalyst on silica in order to produce a heter-
ogeneous catalyst that could be used in a tandem, gas-phase
hydroformylation–hydrogenation process. In this work, we report
the successful preparation of a silica-supported Shvo’s catalyst,
present the kinetics of n- and iso-butanal hydrogenation, and dis-
cuss the effects of CO on the hydrogenation activity. The activity of
SX-modified silica-supported Shvo’s catalyst was also investigated
to test whether the presence of a phosphine ligand enhances the
activity and/or improves the CO tolerance. We also show that sil-
ica-supported Shvo’s catalyst can be used in tandem with a SX-
Rh SILP catalyst to carry out the direct synthesis of butanol from
propene and synthesis gas.
equimolar mixture of n- and iso-butanal was placed in a 5-mL syr-
inge connected to a syringe pump (Cole-Palmer, 74900 series). The
liquid mixture was injected into a heated port subjected to contin-
uous flow of He. A molar ratio of butanal to H2 of 1:1 was used un-
less specified otherwise. Experiments were carried out at a total
gas pressure of 1 atm. The total gas flow rate was typically
100 cm3 minꢁ1 at STP. Using these conditions, the conversion of
aldehyde was always less than 10%. Reaction products were ana-
lyzed using an Agilent 6890N gas chromatograph containing a
bonded and crosslinked (5%-Phenyl)-methylpolysiloxane capillary
column (Agilent, HP-1) connected to a flame ionization detector
(FID).
The continuous gas-phase hydrogenation of butene was also
examined in order to determine the activity of Shvo/SiO2 for alkene
hydrogenation. Prior to reaction, 0.3 g of Shvo/SiO2 catalyst was
heated to 363 K at a rate of 2 K minꢁ1 in pure He (Praxair,
99.999%) flowing at 20 cm3 minꢁ1 at STP. The feed to the reactor
was then switched to one containing 1-butene (Praxair, 99.9%),
H2 (Praxair, 99.999%), and He (Praxair, 99.999%). A 1:1 stoichiom-
etric ratio of 1-butene to H2 was introduced to reactor at 1 atm. The
partial pressure of 1-butene and He was 0.05 atm and 0.9 atm,
respectively. The total gas flow rate was 100 cm3 minꢁ1 at STP.
Reaction products were analyzed using a gas chromatograph–mass
spectrometer (Varian, Model 320) equipped with a 14-port sam-
pling valve and three sample loops. One sample loop was injected
into an Alumina PLOT column for the FID and mass spectrometer,
and the other sample loop was injected into a Hayesep and Mol
Sieve packed columns.
2. Material and methods
2.3. Gas-phase hydroformylation and hydrogenation of propene to
butanol
2.1. Catalyst synthesis
Tandem hydroformylation and hydrogenation experiments
were conducted using a procedure similar to that described in Sec-
tion 2.2. A bi-layered catalyst bed consisting of Rh-SILP and Shvo/
SiO2 catalysts was utilized. The SILP catalyst was prepared as de-
scribed in Refs. [13,14]. Shvo/SiO2 and Rh-SILP were loaded
sequentially into the reactor (so that SILP catalyst was at the top)
along with an intermediary layer of quartz wool to minimize the
mixing of the two catalysts. Prior to reaction, the reactor was
heated to 413 K at a rate of 2 K minꢁ1 in pure He (Praxair,
99.999%) flowing at 20 cm3 minꢁ1 at STP. Experiments were car-
ried out at total gas pressures varying from 1 to 3 atm and the feed
to the reactor consisted of propene (Praxair, 99.9%), CO (Praxair,
99.99%), H2 (Praxair, 99.999%), and He (Praxair, 99.999%). The total
gas flow rate varied from 20 to 60 cm3 minꢁ1 at STP in order to
maintain a constant residence time. The reaction products were
analyzed by a Agilent 6890N gas chromatograph equipped with a
HP-1 column connected to a FID.
Approximately, 0.06 g of 1-hydroxytetraphenylcyclopentadie-
nyl (tetraphenyl-2,4-cyclopentadien-1-one)-l-hydrotetracarbon-
yldiruthenium(II), Shvo’s catalyst, (Strem, 98%) was dissolved in
10 mL of anhydrous methanol (Aldrich, 99.8%). After 10 min of stir-
ring, 1 g of silica (Silicycle, 500 m2 gꢁ1, average pore diameter 60 Å)
stored under vacuum at 353 K was added to solution and stirred
for an additional 1 h. Methanol was then slowly removed under
vacuum in a rotary evaporator. The resulting catalyst is a yellow-
ish-orange colored powder containing 0.4 wt.% Ru. SX-Shvo/SiO2
was prepared in a similar manner except that in this case, sulfox-
antphos (SX) [18] was added to the solution prior to the addition
of SiO2. Approximately 434 mg of SX was used to achieve a molar
ratio of SX/Ru = 5.
2.2. Measurements of catalyst activity
Measurements of reaction rates were performed in a 6.35 mm
OD quartz tube containing an expanded section (ꢂ12.7 mm OD,
ꢂ20 mm length). A plug of quartz wool was placed below the cat-
alyst bed to hold the powder in place. The reactor was heated by a
ceramic furnace with external temperature control and the catalyst
bed temperature was measured with a K-type thermocouple
sheathed in a quartz capillary placed in direct contact with the cat-
alyst bed.
Prior to reaction, 0.3 g of catalyst was heated to the reaction
temperature at a rate of 2 K minꢁ1 in pure He (Praxair, 99.999%)
flowing at 20 cm3 minꢁ1 at STP. The feed to the reactor consisted
of n-butanal (Aldrich, 98%), iso-butanal (Aldrich, 99%), H2 (Praxair,
99.999%), and He (Praxair, 99.999%). CO (Praxair, 99.99%) was
passed through a trap packed with 3.2 mm pellets of 3 Å molecular
sieve in order to remove iron pentacarbonyl formed within the cyl-
inder [19] and was then co-fed with the other reactants. An
2.4. In situ infrared spectroscopy
Infrared spectra were acquired using a Thermo Scientific Nicolet
6700 FTIR spectrometer equipped with a liquid nitrogen cooled
MCT detector. Each spectrum was obtained by averaging 32 scans
taken with 1 cmꢁ1 resolution. 0.015 g of Shvo/SiO2 or SX-Shvo/SiO2
was pressed into a 20 mm-diameter pellet (<1 mm thick) and
placed into a custom-built transmission cell equipped with CaF2
windows, a K-type thermocouple for temperature control, and
resistive cartridge heaters similar to that described in Ref. [20].
All scans were acquired at 363 K and 1 atm. All absorption spectra
were taken relative to the empty transmission cell. The spectrum
of the catalyst under He flow was subtracted from all the results
reported.

54
D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
2.5. Solid-state 31P MAS NMR
Table 1
Apparent activation energies (kJ molꢁ1) for the formation of n- and iso-butanol using
Shvo/SiO2 or SX-Shvo/SiO2. T = 343–373 K, PTotal = 1 atm, PH ¼ Pbutanal ¼ 0:05 atm,
Solid-state 31P MAS NMR experiments were performed on a
Bruker Avance I-500 MHz spectrometer. Data were obtained by
measuring the samples using a frequency of 202.5 MHz, 90° pulse
2
balance He. Total gas flow rate = 100 cm3 minꢁ1, catalyst mass = 0.3 g.
Catalyst
Apparent activation energy (kJ molꢁ1
)
n-Butanol
iso-Butanol
in 4.2 ls, and a delay of 60 s.
Shvo/SiO2
SX-Shvo/SiO2
49
46
50
51
3. Results and discussion
3.1. Catalyst stability and kinetics of butanal hydrogenation
partial pressure dependencies of H2, iso-butanal, and CO on the
rate of iso-butanol formation are 0.45, 0.95, and ꢁ0.31, respec-
tively over SX-Shvo/SiO2 catalysts.
To investigate the stability of SX-Shvo/SiO2 and Shvo/SiO2 with
time-on-stream, experiments were conducted using an equimolar
gas mixture of H2 and n-butanal. As shown in Fig. 1a, both catalysts
were stable, although the activity of SX-Shvo/SiO2 was nearly 50%
lower than that of Shvo/SiO2. Fig. 1b shows that the addition of CO
to the feed after 2 h of time on stream decreased the butanal
hydrogenation activity of both catalysts quite substantially. The
turnover frequency of SX-Shvo/SiO2 and Shvo/SiO2 dropped from
roughly 8 hꢁ1 to 3 hꢁ1 and 4 hꢁ1 to 2 hꢁ1, respectively.
Table 1 lists the apparent activation barriers for Shvo/SiO2 and
SX-Shvo/SiO2 catalysts measured in the temperature range 343–
373 K. The apparent activation energy for aldehyde hydrogenation
to n-butanol and iso-butanol with Shvo/SiO2 are 50 kJ molꢁ1 and
49 kJ molꢁ1, respectively. Similar activation energies were also
measured for SX-Shvo/SiO2, 46 kJ molꢁ1 and 51 kJ molꢁ1 for n-
and iso-butanol, respectively.
To determine the apparent reaction orders of n- and iso-but-
anal, H2, and CO on the rate of butanol formation, the partial pres-
sures of two components were fixed while the partial pressures of
the third reactant and He were varied. The total gas flow rate was
kept at 100 cm3 minꢁ1 at STP in order to maintain a constant resi-
dence time. Table 2 shows the effects of varying the partial pres-
sures of H2, n- and iso-butanal, and CO on the rate of formation
of n- and iso-butanol over Shvo/SiO2 and SX-Shvo/SiO2. The partial
pressure dependencies and catalytic rates are nearly equal for n-
and iso-butanol formation. The orders with respect to H2, n-but-
anal, and CO on the rate of formation of n-butanol are 0.26, 1.0,
and ꢁ0.59, respectively. Similarly, the orders with respect to H2,
iso-butanal, and CO on the rate of formation of iso-butanol are
0.29, 1.1, and ꢁ0.58, respectively. The apparent reaction orders of
H2 and CO increased in the presence of SX. The partial pressure
dependencies of H2, n-butanal, and CO on the rate of formation
of n-butanol are 0.42, 0.98, and ꢁ0.29, respectively, while the
3.2. Proposed mechanism of butanal hydrogenation
The kinetics of butanal formation can be interpreted using the
mechanism proposed by Shvo and coworkers [21,22] shown in
Scheme 1. Shvo’s catalyst operates through a concerted, outer-
sphere mechanism in which the aldehyde is reduced in one ele-
mentary step despite not being bound directly to the metal [23].
The reaction sequence begins with the dissociation of the dimeric
Ru complex (Shvo’s catalyst, species A) into the constitutive mono-
mers species B and C (Reaction 1). Butanal coordinates to the outer
sphere of the complex by aligning the polarity of the acetyl group
(partially positively charged carbon and partially negatively
charged oxygen) with the hydridic (H–Ru) and acidic hydrogen
(H–O) present on species C. The bound aldehyde is then reduced
to alcohol, which converts species C into species B (Reaction 3).
The catalytic cycle is closed by reaction of H2 with species B to
yield species C (Reaction 2). Species B has a vacant coordination
site and therefore can react with either ligand, CO or SX, via Reac-
tions 4 or 5 respectively to fulfill the vacancy. Species D and E are
both saturated species which must dissociate a ligand to be cata-
lytically active.
Evidence for the existence of several of the intermediates
shown in Scheme 1 was provided by in situ infrared spectroscopy.
Fig. 2 shows IR spectra of the pristine Shvo’s complex and Shvo/
SiO2 under exposure of He at 298 K, He at 363 K, an equimolar mix-
ture of H2 and He at 363 K, and an equimolar mixture of H2, CO,
and He at 363 K. The peak in spectra A and B at 1713 cmꢁ1 is as-
cribed to the bridging O–H–O bond present in the Ru dimer
[21,24], while the broad peak at 1875 cmꢁ1 in spectrum B is from
the SiO2 support as shown by Fig. S1 in the Supporting information.
Spectrum C shows that as the reaction temperature was increased
to 363 K, the band at 1713 cmꢁ1 disappeared while bands at 1965,
2005, and 2030 cmꢁ1 became better resolved. The disappearance of
the 1713 cmꢁ1 band with increased temperature indicates that the
Ru dimer complex has dissociated into its constitutive monomers.
The bands appearing at 1965, 2005, and 2030 cmꢁ1 are character-
istic of carbonyl ligands bound to species B and species C [25,26]
and are sharper in spectrum B than in spectrum A due to the re-
moval of physisorbed water on the SiO2 surface at elevated tem-
perature. The appearance of
a
broad band at 1630 cmꢁ1 is
attributed to the oxo functionality present on the cyclopentadienyl
ligand of species B which is another indication that the Ru dimer
has dissociated into Ru monomer units. Once the catalyst is treated
with H2 (spectrum D), a negative band appears at 1630 cmꢁ1 upon
subtraction from spectrum B signifying the conversion of species B
into species C via Reaction 2. Exposure of the catalyst to CO, as
shown in spectrum E, causes two new bands to appear at 2100
and 2050 cmꢁ1, while the band at 1965 cmꢁ1 becomes negative
upon subtraction from spectrum B. The new bands in the carbonyl
stretching region are attributed to the transformation of species B
into a tricarbonyl Ru complex species E.
Fig. 1. (a and b) Time on stream study with (
ꢃ
) Shvo/SiO2 and (s) SX-Shvo/SiO2
catalysts, T = 363 K, PTotal = 1 atm, catalyst mass = 0.3 g, total gas flow rate at
STP = 100 cm3 minꢁ1
.
(a)
PH ¼ Pbutanal ¼ 0:05
atm,
balance
He.
(b)
2
PCO ¼ PH ¼ Pbutanal ¼ 0:05 atm, balance He.
2

D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
55
Table 2
Apparent reaction orders of n-butanol and iso-butanol synthesis using Shvo/SiO2 and SX-Shvo/SiO2 catalysts. T = 363 K, PTotal = 1 atm, catalyst mass = 0.3 g, total gas flow
rate = 100 cm3 minꢁ1
.
Reaction product
Catalyst
Apparent reaction order
a
H2
Butanalb
COc
n-Butanol
iso-Butanol
a
Shvo/SiO2
SX-Shvo/SiO2
0.26
0.42
1.0
0.98
ꢁ0.59
ꢁ0.29
ꢁ0.58
ꢁ0.31
Shvo/SiO2
SX-Shvo/SiO2
0.29
0.45
1.1
0.95
PH ¼ ½0:05—0:15 atmꢄ; Pn-butanal ¼ Piso-butanal ¼ 0:025 atm, balance He.
2
b
c
Pn-butanal ¼ Piso-butanal ¼ ½0:025—0:075 atmꢄ; PH ¼ 0:05 atm, balance He.
2
PCO ¼ ½0:05—0:15 atmꢄ; PH 0:05 atm; Pn-butanal ¼ Piso-butanal ¼ 0:025 atm, balance He.
2
Scheme 1. Proposed mechanism for aldehyde hydrogenation using Shvo/SiO2 catalyst.
Further evidence for the coordination of SX to Shvo’s catalysts
under reaction conditions was obtained from 31P NMR spectros-
copy. Fig. 3 shows the results of solid-state 31P MAS NMR measure-
ments of spent SX-Shvo/SiO2 which exhibits the same spectral
features of the fresh SX-Shvo/SiO2. The broad peak observed at
ꢁ13 ppm can be ascribed to the free SX adsorbed on the silica sur-
face [13], while the broad peak centered at 34 ppm is due to the
phosphine bonded to Ru as shown in species E [17]. Although
many of species shown in Scheme 1 could undergo ligand ex-
change with SX, Reaction 5 is the most facile because of the vacant
coordination site on species B. Previous reports that have shown by
solution 31P NMR that singly coordinated phosphorus to Ru
Fig. 2. In situ infrared spectra of (a) Shvo’s complex in He at 298 K and Shvo/SiO2
catalyst under, (b) He at 298 K, (c) He at 363 K, (d) H2 at 363 K, and (e) H2/CO = 1 at
363 K.
The spectra for SX-Shvo/SiO2 exhibit similar bands to those ob-
served for Shvo/SiO2. The most notable difference between the cat-
alysts is that the carbonyl stretching region is red-shifted by 5–
10 cmꢁ1 in the presence of SX because phosphines are better sigma
donors than carbonyl ligands [27]. The higher electron density on
Ru enables greater pi-backbonding to the carbonyl groups, thereby
reducing the frequency of the carbonyl stretching frequency. A
comparison of the spectra for Shvo/SiO2 and SX-Shvo/SiO2 is pro-
vided in the Supporting information.
Fig. 3. Solid-state 31P MAS NMR spectra of spent SX-Shvo/SiO2 catalyst.

56
D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
(species E of Scheme 1) results from the reaction of species B with
SX [17]. However, the possibility that both phosphorus atoms of SX
are coordinated to Ru cannot be ruled out on the basis of the pres-
ent data. While there have reports indicating that Shvo’s catalyst
modified with PPh3 can complete the catalytic cycle, this occurs
at much slower rates than with the unmodified catalysts [28];
thus, all of the catalytic activity is attributed to unmodified species
C. The data and the literature, therefore, indicate that SX-Shvo/SiO2
is less active than Shvo/SiO2 because coordination of SX reduces
the concentration of species C available for the catalytic cycle.
3.4. Butene hydrogenation over Shvo/SiO2
The hydrogenation of 1-butene was investigated in order to
determine activity of Shvo/SiO2 for hydrogenation of alkenes. It
was observed that the 1-butene hydrogenation activity was
0.4 hꢁ1, which is more than an order of magnitude lower than
the activity for butanal hydrogenation under comparable condi-
tions. Consistent with our results, Shvo and coworkers demon-
strated that the alkenes are unreactive for hydrogenation in the
presence of Shvo’s Catalyst and formic acid [29]. The difference
in hydrogenation activity between 1-butene and butanal is attrib-
utable to the non-polarity of 1-butene. Accordingly, 1-butene is
unable to coordinate to the hydridic and acidic hydrogen atoms
on the Ru cluster as effectively as the polar carbonyl of butanal.
3.3. Kinetics of butanal hydrogenation
If it is assumed that at reaction temperature, species A dissoci-
ates completely to species B and C, that Reactions 2, 4, and 5 are
quasi-equilibrated, and that Reaction 3 is the irreversible, rate-lim-
iting step, then the rate of butanol formation is given by the follow-
ing relation:
3.5. Tandem hydroformylation and hydrogenation using Rh-SILP and
Shvo/SiO2 catalysts
The direct conversion of propene to butanol was investigated
using a bi-layered catalyst bed of SX-Rh SILP and Shvo/SiO2. The
SX-Rh SILP catalyst was selected because of its high activity and
stability for propene hydroformylation [13]. Our previous study
of Rh-SILP catalysts showed that catalysts containing the SX ligand
were almost six times more active than catalysts prepared with tri-
phenylphosphine or sulfonated triphenyl phosphine ligands [13].
We have also demonstrated that at 363 K, the apparent partial
pressure dependencies of propene, H2, and CO are 0.93, 0.1, and
ꢁ0.43 [14]. However, the apparent partial pressure dependencies
of propene, H2, and CO become 0.76, 0.84, and zero, respectively
at 413 K. Since the kinetics of butanal hydrogenation exhibit a
nearly first-order dependence on H2 and an inverse dependence
on the partial pressure of CO (see Table 2), a high ratio of H2 to
CO partial pressures is desired. The reaction temperature was set
to 413 K to exploit first-order and zero-order dependence of H2
and CO on the rates of propene hydroformylation and to lower
the molar ratio of n-butanal to iso-butanal (n/i) formation. A low
n/i ratio is desirable for fuel production because the anti-knock in-
dex of iso-butanol is much greater than n-butanol [4]. The CO and
propene partial pressure were fixed at 0.25 atm, while the H2 par-
tial pressure was varied from 0.25 to 2.5 atm. As shown in Fig. 5,
the propene conversion to butanal and the aldehyde conversion
to butanol rises as the H2/CO molar ratio is increased. Also shown
in Fig. 5 is the predicted propene and aldehyde conversion as cal-
culated from the rate laws presented in this work and in Ref.
[14]. For convenience, the values of the parameters in the hydro-
formylation rate expression measured at 413 K are provided in Ta-
ble 4. As expected from our previous work on hydroformylation
and this work, the molar ratios of n-butanal to iso-butanal and n-
butanol to iso-butanol were both approximately 7. Under the most
favorable conditions examined (H2/CO = 10), 15% propene conver-
sion and 90% aldehyde conversion were observed; thus, a 13% yield
to butanol was achieved. The results presented in Fig. 5 illustrate,
to the best of our knowledge, the first example of the direct conver-
sion of propene to butanol by means of the tandem hydroformyla-
tion and hydrogenation reactions carried out in the gas phase using
heterogeneous catalysts.
K2k3PbutanalPH ½Ruꢄ
2
rbutanol
¼
ð1Þ
1 þ K2PH þ K4PCO þ K5½SXꢄ
2
where k3 is the rate coefficient for Reaction 3, Ki is the equilibrium
constant for Reaction i, Pj is the partial pressure of species j, and [Ru]
is the total surface concentration of Ru. The derivation of Eq. (1) is
provided in the Supporting information.
Eq. (1) is consistent with the experimentally observed partial
pressure dependencies and those found in literature. Casey and
coworkers have reported that the order with respect to benzalde-
hyde and H2 in a homogenous solution of Shvo’s Catalyst to be 1
and 0, respectively [28]. The rate of butanal synthesis is first order
in butanal due to the Pbutanal term in the numerator of Eq. (1). The
0–1 order in the partial pressure of H2 depends on the magnitude
of the K2PH in the denominator of Eq. (1). As the value of K2PH in-
2
2
creases, the order with respect to H2 tends toward zero. Thus, the
value for K2PH in solution must be high in order to explain the
2
zero-order dependence observed in Ref. [28]. The inhibitory effect
of CO and SX is represented by the K4PCO and the K5[SX] terms in
the denominator, respectively. The order with respect to CO be-
comes less negative in presence of SX due to the term K5[SX] also
present in the denominator of Eq. (1).
The parameters appearing in Eq. (1) for Shvo/SiO2 and SX-Shvo/
SiO2 were estimated using the data fitting routine (i.e. ‘‘nlinfit’’
function) in MATLAB with the initial guesses of the parameters
set to zero. Values for the parameters are listed in Table 3 for n-
and iso-butanol formation. As seen in Fig. 4a–d, the derived rate
expression given in Eq. (1) describes the experimental data effec-
tively. The coefficient of determination, R2, is above 90% for all
cases considered. Values of R2, the root-mean-square error (RMSE),
and a parity plot are presented in the Supporting information.
Table 3
Values of the rate parameters appearing in Eq. (1) obtained by
experimental data.
a best fit to
The rate expressions of hydroformylation and hydrogenation
were also used to predict the conditions necessary to achieve a
high yield of butanol from propene over SX-Rh SILP and Shvo/
SiO2 catalysts. Fig. 6 shows the effects of varying the total pressure
on the minimum amount of hydroformylation catalyst and hydro-
genation catalyst necessary to achieve 99% conversion in each
reaction. In all of the cases examined, the initial molar flow rate
of the feed is set to 54 mmol hꢁ1 and the inlet feed consists of a
1:1:2 M ratio of C3H6:CO:H2. Under these conditions, Shvo/SiO2
operates in the absence of CO because 99% of the propene and
K2k3 PbutanalPH2 ½Ruꢄ
rbutanol
¼
1þK2PH2 þK4PCO þK5 ½SXꢄ
Parameter
Value
Unit
K2
K4
K5
k3
30
55
atmꢁ1
atmꢁ1
7.3 ꢀ 106
255a
325b
m2 atmꢁ1
atmꢁ1 hꢁ1
a
n-Butanol.
iso-Butanol.
b

D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
57
Fig. 4. (a–d) Non-linear regression of partial pressure data using (a,b) Shvo/SiO2 and (c,d) SX-Shvo/SiO2. Rate of (a,c) n- and (b,d) iso-butanol synthesis vs. partial pressure of
(
ꢃ
,s) butanal, (
N,4) H2, and (j,h) CO. Experimental points are indicated by data markers. Spline curves indicate predicted values. T = 363 K, PTotal = 1 atm, catalyst
mass = 0.3 g, total gas flow rate = 100 cm3 minꢁ1
.
Fig. 5. Results from tandem
T = 413 K. (s,4,h) Propene conversion to butanal. (
experiment.
, ,j) Butanal conversion to
hydroformylationꢃ–hydrogenation
N
Fmige.nt6..(ꢃM)oMdealsesdofreSsXu-lRtsh SoIfLPthceattaalnysdteamndh(ysd)romfoarsms yolfaSthiovno–/ShiyOd2rocagteanlaytsitonneceexspsearriy-
butanol. PC
¼ PCO ¼ 0:25 atm. (s,
ꢃ
) PH ¼ PH6 ¼ 0:25 atm, (4,
N
) PH ¼ 1:25 atm,
3 H6
2
2
to achieve 98% yield to butanol as
a function of total pressure. T = 413 K,
(h,j) PH ¼ 2:5 atm: SX-Rh SILP catalyst mass = 0.4 g (0.2 wt.% Rh), Shvo/
2
NT ¼ 54 mmol hꢁ1; 2PC
¼ 2PCO ¼ PH
.
_
3 H6
2
SiO2 = 0.8 g (1 wt.% Ru). The solid lines represent conversions determined from
the rate expressions for propene hydroformylation and butanal hydrogenation
(Eq. (1)).
that a high yield of butanol could be achieved starting with a stoi-
chiometric feed of propene and synthesis gas.
Table 4
Rate parameters determined for propene hydroformylation over SX-Rh SILP Ref. [14].
4. Conclusion
aPC3 6 PH2 ½Rhꢄ
H
rbutanal
¼
1þbð1þK5ÞPC3
H
6
The kinetics of butanal hydrogenation over Shvo/SiO2 and SX-
Shvo/SiO2 have been investigated. The activity of SX-Shvo/SiO2
was lower than the activity of Shvo/SiO2, but the apparent activa-
tion energies for hydrogenation of n- and iso-butanol were nearly
identical (ꢂ50 kJ molꢁ1) for both catalysts. This result suggests that
the catalytically active species is the same for both catalysts. The
partial pressure dependence on butanal is roughly first order for
both catalysts, while the partial pressure dependence on H2 is
ꢂ0.3 and 0.4 for Shvo/SiO2 and SX-Shvo/SiO2, respectively. Simi-
larly, the partial pressure dependence on CO increases from ꢁ0.6
in the absence of SX to ꢁ0.3 in the presence of SX. The differences
in the orders in the partial pressures of H2 and CO are ascribed to
the coordination of SX to Ru, which was confirmed by in situ
Parameter
Value
Unit
n-Butanal
iso-Butanal
589
85.6
0.22
1.8
atmꢁ2 hꢁ1
atmꢁ1
a ¼ K1K2K3K4K5k6
b ¼ K1K2K3K4
K5
k6 ¼ aðbK5Þꢁ1
0.89
0.20
3287
–
210
atmꢁ1 hꢁ1
CO are consumed in the hydroformylation reaction. Fig. 6 shows
the amount of catalyst necessary to achieve a 98% yield of butanol
decreases substantially as the total pressure is increased. While the
span of experimental conditions predicted are beyond the limita-
tions of the apparatus available for the present study, it is evident

58
D.G. Hanna et al. / Journal of Catalysis 311 (2014) 52–58
infrared spectroscopy and 31P MAS NMR. Infrared spectroscopy
also provided evidence for several of the intermediates appearing
in the mechanism shown in Scheme 1. Based upon the proposed
mechanism, the ability of Shvo’s catalyst to function in the pres-
ence of CO is attributed to the competition between a vacant coor-
dination site and the addition of a H2 to the Ru complex. A rate law
was derived assuming that the reduction of aldehyde is the rate-
determining step (Reaction 3 in Scheme 1) and that the other ele-
mentary steps are quasi-equilibrated. The rate coefficients appear-
ing in Eq. (1) were obtained by non-linear regression (see Table 3)
and found to describe the experimental data very well.
The direct gas-phase conversion of propene to butanol was
demonstrated for the first time using a bi-layered, packed-bed
reactor containing SX-Rh SILP, and Shvo/SiO2 catalysts. Under the
conditions investigated, the overall yield a 13% to n- and iso-buta-
nol (n/i = 7) was achieved using a feed molar ratio of H2/CO = 10.
This proof-of-concept experiment demonstrates the feasibility of
converting propene to butanol in a single reactor. The rate expres-
sions describing the kinetics of propene hydroformylation and but-
anal hydrogenation were also used to predict the amount of each
catalyst required to achieve a 98% yield of butanol as a function
of total pressure for an assumed stoichiometric feed and initial mo-
lar flow rate.
[2] P. Durre, Ann. N. Y. Acad. Sci. 1125 (2008) 353.
[3] S. Lee, M.O. Cho, C.H. Park, Y. Chung, J.H. Kim, B. Sang, Y. Um, Energy Fuels 2
(2008) 3459.
[4] E. Christensen, J. Yanowitz, M. Ratcliff, R.L. McCormick, Energy Fuels 25 (2011)
4723.
[5] J.E. Rekoske, M.A. Barteau, Ind. Eng. Chem. Res. 50 (2011) 41.
[6] S. Marengo, P. Comotti, G. Galli, Catal. Today 81 (2003) 205.
[7] Y. Lee, J.H. Park, S.H. Jang, L.K. Nielsen, J. Kim, K.S. Jung, Biotechnol. Bioeng. 101
(2008) 209.
[8] F. Coloma, A. Sepulveda-Escribano, J.L.G. Fierro, F. Rodriguez-Reinoso, Appl.
Catal., A 136 (1996) 231.
[9] P. Reyes, M.C. Aguirre, J.L.G. Fierro, G. Santori, O.J. Ferretti, Mol. Catal. A: Chem.
184 (2002) 431.
[10] J.K. Jeon, J.H. Yim, Y.K. Park, Chem. Eng. J. 140 (2008) 555.
[11] A. Riisager, K.M. Eriksen, P. Wasserscheid, R. Fehrmann, Catal. Lett. 90 (2003)
149.
[12] A. Riisager, R. Fehrmann, M. Haumann, B.S.K. Gorle, P. Wasserscheid, Ind. Eng.
Chem. Res. 44 (2005) 9853.
[13] S. Shylesh, D.G. Hanna, S. Werner, A.T. Bell, ACS Catal. 4 (2012) 487.
[14] D.G. Hanna, S. Shylesh, S. Werner, A.T. Bell, J. Catal. 292 (2012) 166.
[15] D. Konya, K.Q. Almeida Lereno, E. Lereno, Organometallics 23 (2006) 3166.
[16] J. Zakzeski, H.R. Lee, Y.L. Leung, A.T. Bell, Appl. Catal., A 374 (2010) 201.
[17] K. Takahashi, M. Yamashita, T. Ichihara, K. Nakano, K. Nozaki, Angew. Chem.
Int. Ed. 49 (2010) 4488.
[18] W.P. Mul, K. Ramkisoensing, P.C.J. Kamer, J.N.H. Reek, A.J. van der Linden, A.
Marson, P.W.N.M. van Leeuwen, Adv. Synth. Catal. 344 (2002) 293.
[19] L.Q. Xu, V.L. Zholobenko, L.M. Kustov, W.M.H. Sachtler, J. Mol. Catal. 83 (1993)
391.
[20] J.F. Joly, N. Zanier-Szydlowski, S. Colin, F. Raatz, J. Saussey, J.C. Lavalley, Catal.
Today 9 (1991) 31.
[21] Y. Shvo, D. Czarkie, Y. Rahamim, J. Am. Chem. Soc. 108 (1986) 7400.
[22] N. Menashe, Y. Shvo, Organometallics 10 (1991) 3885.
[23] J.S.M. Samec, J.E. Backvall, P.G. Andersson, P. Brandt, Chem. Soc. Rev. 35 (2006)
237.
Acknowledgment
This work was supported by the XC2 program funded by BP.
[24] J.E. Caton, C.V. Banks, Inorg. Chem. 6 (1967) 1670.
[25] M.J. Mays, M.J. Morris, P.R. Raithby, Y. Shvo, D. Czarkie, Organometallics 8
(1989) 1162.
[26] C.P. Casey, S.E. Beetner, J.B. Johnson, J. Am. Chem. Soc. 130 (2008) 2285.
[27] R.H. Crabtree, Carbonyls, phosphine complexes, and ligand substitution
reactions, in: The Organometallic Chemistry of the Transition Metals, fourth
ed. John Wiley & Sons, Inc., Hoboken, 2005, p. 101.
[28] C.P. Casey, N.A. Strotman, S.E. Beetner, J.B. Johnson, D.C. Priebe, T.E. Vos, B.
Khodavandi, I.A. Guzei, Organometallics 25 (2006) 1230.
[29] N. Menashe, E. Salant, Y. Shvo, J. Organomet. Chem. 514 (1996) 97.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2013.11.012.
References
[1] A. Chauvel, G. Lefebvre, in: Petrochemical Processes, second ed., Editions
Technip, Paris, 1989, pp. 81–90.