Selective Production of Benzylamine via Gas Phase Hydrogenation of Benzonitrile over Supported Pd Catalysts

Article abstract of DOI:10.1007/s10562-015-1655-8

The continuous gas phase hydrogenation of benzonitrile has been studied over Pd/C and Pd/Al₂O₃ (mean Pd size = 2.5-3.0 nm from STEM). Toluene was the predominant product over Pd/C whereas benzylamine was preferred over Pd/Al₂O₃. This is linked to differences in Pd electronic character (from XPS) that impact on nitrile adsorption. Greater surface acidity and hydrogen availability (based on chemisorption/TPD) for Pd/C resulted in higher conversion rates. Operation at lower contact time for multiple Pd/Al₂O₃ beds in series promoted exclusive formation of benzylamine at an appreciably higher rate than reported in the literature.

Full text of DOI:10.1007/s10562-015-1655-8

Catal Lett
DOI 10.1007/s10562-015-1655-8
Selective Production of Benzylamine via Gas Phase
Hydrogenation of Benzonitrile over Supported Pd Catalysts
1
1
1
1
Yufen Hao
•
Maoshuai Li
•
Fernando C a´ rdenas-Lizana
•
Mark A. Keane
Received: 20 August 2015 / Accepted: 10 November 2015
Ó Springer Science+Business Media New York 2015
Abstract The continuous gas phase hydrogenation of
benzonitrile has been studied over Pd/C and Pd/Al O3
Keywords Selective hydrogenation Á Benzonitrile Á
Benzylamine Á Pd/C Á Pd/Al O Á Catalyst beds in series
2
2 3
(
mean Pd size = 2.5–3.0 nm from STEM). Toluene was
the predominant product over Pd/C whereas benzylamine
was preferred over Pd/Al O . This is linked to differences
1 Introduction
2
3
in Pd electronic character (from XPS) that impact on nitrile
adsorption. Greater surface acidity and hydrogen avail-
ability (based on chemisorption/TPD) for Pd/C resulted in
higher conversion rates. Operation at lower contact time for
multiple Pd/Al O beds in series promoted exclusive for-
Benzylamine is a versatile chemical used in the production
of Vitamin H (Biotin), as an intermediate for photographic
chemicals, in the manufacture of cationic paints and as an
active ingredient in the synthesis of nylon fibres [1]. The
hydrogenation of benzonitrile as the principal route to
benzylamine [2, 3] is challenging due to undesired con-
densation with the benzylideneimine intermediate to give
dibenzylamine (with NH3 elimination) and amine
hydrogenolysis to toluene [4–6]; see Fig. 1. Benzonitrile
2
3
mation of benzylamine at an appreciably higher rate than
reported in the literature.
Graphical Abstract
hydrogenation has been established at elevated H pressure
2
(
6–100 bar) [4, 6–11] over supported metal (Ni [6, 12], Pt
10], Rh [10] and Pd [4, 7, 9–11] ) catalysts. Palladium has
[
delivered the highest combined activity and selectivity [10]
where biphasic separation (for reactions in CO ? H O)
2
2
under supercritical conditions (70 bar) served to elevate
benzylamine selectivity (up to 98 %) by preventing con-
densation [7]. The energy demands associated with super-
critical CO lowers process efficiency. Conventional batch
2
&
Mark A. Keane
m.a.keane@hw.ac.uk
liquid phase operation is a less energy intensive alternative.
Benzylamine has been generated as principal product over
supported Pd [9] where the factors proposed to inhibit
dibenzylamine formation and increase benzylamine yield
include support [4], solvent [12] and incorporation of (acid)
additives to the reaction medium [6]. The requirement for
separation and purification unit operations to extract the
target product represents a decided drawback in terms of
sustainable/clean synthesis.
Yufen Hao
y.hao927@gmail.com
Maoshuai Li
ml297@hw.ac.uk
Fernando C a´ rdenas-Lizana
f.cardenaslizana@hw.ac.uk
1
Chemical Engineering, School of Engineering and Physical
Sciences, Heriot-Watt University, Edinburgh EH14 4AS,
Scotland, UK
A ‘‘white paper’’ by the American Chemical Society in
conjunction with the Green Chemistry Institute and global
123

Y. Hao et al.
2.2 Catalyst Characterisation
The Pd content was measured by inductively coupled plasma-
optical emission spectrometry (ICP-OES, Vista-PRO, Varian
Inc.) from the diluted extract in HF. Hydrogen and NH3
chemisorption, temperature programmed desorption (TPD) and
specific surface area (SSA) measurements were conducted on
the commercial CHEM-BET 3000 (Quantachrome) unit
equipped with a thermal conductivity detector (TCD) and TPR
TM
Win
software for data acquisition/manipulation. Samples
were loaded into a U-shaped Quartz cell (3.76 mm i.d.) and
3
-1
heated in 17 cm min (Brooks mass flow controlled) 5 % v/v
-
1
H /N at 2 K min to 573 K where the effluent gas passed
2
2
through a liquid N trap. The reduced samples were maintained
2
Fig. 1 Reaction pathways associated with the hydrogenation of
benzonitrile (BN) to target (framed) benzylamine (BA, solid arrows),
condensation to dibenzylamine (DBA, dotted arrows) and
hydrogenolysis to toluene (TOL, dashed arrow)
at the final temperature in a flow of H /N until the signal
2
2
-1
3
returned to baseline, swept with 65 cm min N /He for 1.5 h,
2
cooled to ambient temperature and subjected to H (BOC,
2
9
9.99 %) or NH (BOC, 99.98 %) pulse (50–1000 ll) titration.
3
3
-1
pharmaceutical corporations identified the move from
batch to continuous processing as #1 priority to achieve
sustainable chemical production and related research was
given the highest strategic importance [13]. Low activity
has been a common feature of the reported work on gas-
phase continuous benzonitrile hydrogenation, which has
been limited to the use of Cu–MgO [14] and Ni/Al O [15].
Samples were thoroughly flushed in N
/He (65 cm min ) to
2
remove weakly bound H or NH and subjected to TPD (at
2
3
-
1
10–50 K min ) to 950–1200 K. The resultant profile was
corrected using the TPD recorded in parallel directly following
TPR to explicitly determine H
recorded with a 30 % v/v N /He flow using N
internal standard. Three cycles of N adsorption–desorption in
the flow mode were employed to determine SSA using the
standard single point BET method. SSA and H /NH uptake/
or NH
release. SSA was
2
3
(99.9 %) as
2
2
2
3
2
Palladium has exhibited higher activity in liquid phase
operation but low benzylamine yield [10, 11]. Gas phase
continuous operation facilitates control over contact time
by varying the volume and number of catalyst beds, which
can influence product selectivity [16, 17]. The use of
multiple catalyst beds in catalytic processes has been
considered to a limited extent [18–20] and we could not
find any reported application in catalytic hydrogenation. In
this work we examine for the first time the possible role of
support (C and Al O ) on the catalytic action of Pd in the
2
3
desorption were reproducible to ± 5 % and values quoted rep-
resent the mean. X-ray photoelectron spectroscopy (XPS)
analyses were conducted on an Axis Ultra instrument (Kratos
-
8
Analytical) under ultra-high vacuum conditions (\10 Torr)
using a monochromatic Al Ka X-ray source (1486.6 eV). The
source power was maintained at 150 W and the emitted pho-
2
toelectrons were sampled from a 750 9 350 lm area at a
take-off angle = 90°. The analyser pass energy was 80 eV for
survey spectra (0–1000 eV) and 40 eV for high resolution
spectra (Pd 3d5/2 and 3d3/2). The adventitious carbon 1s peak
was calibrated at 284.5 eV and used as an internal standard to
compensate for charging effects. Palladium particle morphol-
ogy (shape and size) was determined by scanning transmission
electron microscopy (STEM, JEOL 2200FS field emission
unit) employing Gatan DigitalMicrograph 1.82 for data
acquisition/manipulation. Samples for analysis were crushed
and deposited (dry) on a holey carbon/Cu grid (300 Mesh). Up
to 800 individual Pd particles were counted for each catalyst
and the surface area-weighted mean Pd diameter (dSTEM) cal-
culated from
2
3
gas phase continuous hydrogenation of benzonitrile to
benzylamine and evaluate the use of multiple beds in series
as a means to fine-tune contact time and promote selective
amine production.
2
Experimental
2
.1 Materials and Catalyst Activation
Commercial (1 % w/w) Pd/C and (1.2 % w/w) Pd/Al O3
2
catalysts were obtained from Sigma-Aldrich. Prior to use,
P
nidi3
i
the samples were sieved to 75 lm mean diameter and
activated in 60 cm min H at 2 K min to 573 K,
3
-1
-1
2
dSTEM ¼ P
ð1Þ
nidi2
which was maintained for 1 h to ensure complete reduction
to zero valent Pd [21]. Samples for off-line characterisation
i
were passivated in 1 % v/v O /He at ambient temperature.
2
where n is the number of particles of diameter d .
i i
1
23

Selective Production of Benzylamine via Gas Phase Hydrogenation of Benzonitrile over…
2
.3 Catalytic Procedure
3 Results and Discussion
Reactions of benzonitrile/benzylamine (Sigma-Aldrich,
C99 %) were conducted in situ, immediately following
catalyst activation, under atmospheric pressure at 353 K in
a fixed bed vertical glass reactor (i.d. = 15 mm). The
reactant was delivered as a 2-propanol (Sigma-Aldrich,
C99.5 %) solution at a fixed calibrated flow rate (Model
3.1 Pd/C: Characterisation and Catalyst Test
Results
The physicochemical properties of Pd/C are summarised in
Table 1. Total surface area (SSA) is consistent with the
literature [22]. Palladium particle size was determined by
H chemisorption assuming dissociative adsorption (H /Pd
1
00 kd scientific microprocessor-control infusion pump) to
2
2
the reactor via a glass/Teflon air-tight syringe and Teflon
adsorption stoichiometry = 1/2 [23]) and validated by
STEM analysis. The representative STEM image given in
Fig. 2a reveals pseudo-spherical Pd nano-particles with a
narrow size distribution (\5 nm, Fig. 2b) and mean of
2.5 nm that is in good agreement with that (2.2 nm)
obtained from H2 titration. Hydrogen temperature pro-
grammed desorption (TPD) generated the profile in Fig. 2c
line in a co-current flow of H (BOC, [99.98 %). The 2-
2
propanol serves as a carrier and did not undergo any
reaction. A layer of borosilicate glass beads served as
preheating zone where the organic reactant was vaporised
and reached reaction temperature before contacting the
catalyst. Isothermal conditions (± 1 K) were maintained by
diluting the catalyst bed with ground glass (75 lm) which
was mixed thoroughly with catalyst before insertion in the
reactor. The reaction temperature was continuously moni-
tored using a thermocouple inserted in a thermowell within
the catalyst bed. Catalyst performance in a multiple bed
arrangement was investigated where each catalyst bed was
separated by glass wool (10 mm); catalyst activation fol-
lowed the procedure described above. The inlet molar
where a two stage H release is evident with temperature
2
maxima (Tmax) at 789 and 1148 K. Desorption at
T C 770 K can be attributed to loss of spillover hydrogen
generated during thermal activation [24], i.e. migration of
atomic hydrogen to the support following dissociation at
Pd sites. Reported surface acidity measurements of sup-
ported metal systems have relied on NH desorption [25–
3
27]. In this study, NH3 chemisorption matched that
released in TPD (Table 1) with an associated
Tmax = 490 K (Fig. 2d) within the range (444–573 K)
reported for activated carbon supported Pd [28]. In order to
gain insight into the electronic character of the Pd phase,
XPS analysis over the Pd 3d binding energy (BE) region
was conducted and the profile is given in Fig. 2e. The Pd
3d5/2 signal (at 335.9 eV) is 0.7 eV higher than that char-
acteristic of metallic Pd (335.2 eV) [29], suggesting elec-
tron transfer to the carbon support with the generation of
Pdd? as proposed elsewhere for nano-scale (4–12 nm) Pd
on carbon [21, 30].
-
1
reactant flow rate was in the range 0.18–0.36 mmol h
where H content was up to 900 times in excess of the
2
stoichiometric requirement for benzylamine production.
Gas flow rate was monitored using a Humonics (Model
5
20) digital flowmeter. The molar palladium to inlet
-
3
reactant ratio (n/F) spanned the range 1.3 9 10
–
-
.6 9 10 h. In blank tests passage of benzonitrile or
2
3
benzylamine in a stream of H through the empty reactor
2
did not result in any detectable conversion. The reactor
effluent was frozen in liquid N for subsequent analysis
2
using a Perkin-Elmer Auto System XL chromatograph
equipped with a programmed split/splitless injector and
flame ionisation detector, employing a DB-1 capillary
column (i.d. = 0.33 mm, length = 50 m, film thick-
ness = 0.20 lm). Benzonitrile conversion (X) is given by
Conversion (X) and product selectivity (S ) as a
i
function of contact time (n/F) are presented in Fig. 2f
where an increase in benzonitrile conversion was recor-
ded at higher n/F. Toluene was generated as the principal
product (S [ 80 %) with secondary benzylamine forma-
i
-
3
½
BenzonitrileŠin À ½BenzonitrileŠout
tion at lower n/F (\8 9 10 h). At extended contact
time, toluene was the sole product via amine
hydrogenolysis (dashed arrow in Fig. 1). Under the same
reaction conditions, use of benzylamine as feed generated
toluene. There was no detectable dibenzylamine, a result
that deviates from liquid phase reaction systems where
significant by-product formation (S \ 14 %) was
observed over Pd/C [6, 8]. Condensation of the imine
intermediate with benzylamine can be circumvented due
to the shorter contact time in continuous gas phase
operation. Nitrile reduction to primary amine proceeds
via a nucleophilic mechanism where hydrogen (as a weak
X ð%Þ ¼
with selectivity (S) to product ‘‘i’’
ProductŠi; out
 100
ð2Þ
½
BenzonitrileŠin
½
Si ð%Þ ¼
 100
½
BenzonitrileŠin À ½BenzonitrileŠout
ð3Þ
where subscripts ‘‘in’’ and ‘‘out’’ represent the reactor inlet
and outlet effluent streams. Repeated reactions with dif-
ferent samples of catalyst delivered raw data repro-
ducibility and a mass balance better than ± 5 %.
123

Y. Hao et al.
Table 1 Palladium loading, specific surface area (SSA), Pd particle
size (from H chemisorption and STEM analysis), H and NH
chemisorption and release during TPD. Benzonitrile hydrogenation
rate (R) and selectivity (S
i
) to benzylamine (BA) and toluene (TOL) at
2
2
3
2 3
X = 20 % over Pd/C and Pd/Al O
2
Pd (% w/w) SSA (m g
-1
)
-1
(mmol gPd
-1
)
-1 -1
R (mol h molPd
Catalyst
Pd/C
Pd size (nm)
H
2
)
NH
3
(mmol g
)
Si, X = 20 % Product (%)
a
2.2 /2.5
b
c
2.5 /80
d
c
0.94 /0.92
d
e
1.0
1.2
870
145
230 (527)
BA (10)
TOL (90)
BA (96)
TOL (4)
a
2.4 /3.0
b
c
2.1 /10
d
c
0.52 /0.51
d
e
2
Pd/Al O
3
99 (272)
a
from H
2
chemisorption
b
c
d
e
from STEM
/NH
/NH
H
2
3
uptake from titration at ambient temperature
H
2
3
release from TPD
-1
TOF (h ) based on Pd dispersion from STEM analysis
Fig. 2 a Representative STEM
image with b associated Pd
particle size distribution, c H
and d NH TPD profiles, e XPS
2
3
spectrum over the Pd 3d region
and f variation of conversion (X,
open triangle) and selectivity
i
(S ) to benzylamine (filled
square) and toluene (filled
circle) with contact time (n/F)
for reaction over Pd/C
1
23

Selective Production of Benzylamine via Gas Phase Hydrogenation of Benzonitrile over…
nucleophile) attacks activated –C:N to generate the
reactive imine (–CH=NH) that undergoes further hydro-
genation (to –CH –NH ). A number of adsorption modes
2
2
(i.e. parallel, perpendicular and side-on) [4] on Pd sur-
faces has been proposed for nitrile hydrogenation. Nitrile
interaction should be favoured via the lone pair electron
on nitrogen with electron deficient Pdd? in a perpendic-
ular orientation (Fig. 3a), where the adsorbed nitrile
undergoes attack by surface dissociated hydrogen with
cleavage of C–N to generate toluene. A perpendicular
adsorption of the basic nitrile group on support acid sites
[
31–34] is also possible with hydrogenolytic bond scis-
Fig. 3 Surface benzonitrile orientation for reaction over a Pd/C and
b Pd/Al
sion by spillover hydrogen.
2 3
O
Fig. 4 a Representative STEM
image with b associated Pd
particle size distribution, c H
and d NH TPD profiles, e XPS
2
3
spectrum over the Pd 3d region
and f dependence of conversion
(
X, open triangle) and
selectivity (S ) to benzylamine
filled square) and toluene
filled circle) on contact time (n/
i
(
(
2 3
F) for reaction over Pd/Al O
123

Y. Hao et al.
-
3
3
.2 Pd/Al O : Characterisation and Catalyst Test
2
as feed (n/F = 3 9 10 h) over Pd/Al O generated
2 3
3
-
1
Results
toluene as product but at a much lower rate (3 mol h
-1
mol ) than Pd/C (70 mol h molPd ). Taking a common
-
-
1
-1
Pd
The SSA of Pd/Al O was appreciably lower than Pd/C
2
fractional nitrile conversion (X = 20 %, see Table 1), Pd/
3
2
Table 1) and is comparable with values (94–185 m g
-1
(
)
Al O delivered an appreciably higher selectivity to ben-
2 3
reported in literature for commercial c-Al O supported Pd
zylamine but at a lower rate compared with Pd/C. This
deviation in selectivity can be attributed to differences in
Pd electronic character (from XPS) that impact on the
mode of adsorption of benzonitrile. The higher nitrile
consumption rate over Pd/C can be linked to greater
availability of surface reactive hydrogen (from H2
chemisorption/TPD) and possible enhanced nitrile activa-
tion at support acid sites (with higher acidity based on NH3
chemisorption/TPD).
2
3
[
35, 36]. The Pd phase on Al O took the form of discrete
2 3
pseudo-spherical particles (Fig. 4a) with a similar size
distribution (Fig. 4b) and mean (3.0 nm) to Pd/C that was
again in good agreement with H chemisorption (Table 1).
2
Hydrogen TPD from Pd/Al O resulted in a broad signal
2
3
(
Fig. 4c) over the 610–950 K range. Total H desorbed
2
exceeded uptake during pulse titration but was significantly
lower than that measured for Pd/C (Table 1). Spillover is
influenced by the concentration of initiating and acceptor
sites, degree of contact between participating phases and
metal-support interaction(s) [37]. Greater H desorption
2
3.3 Use of Catalyst Beds in Series: Enhanced
Benzylamine Production Over Pd/Al O
2 3
from Pd/C can be linked to the higher available surface
area relative to Pd/Al O which can accommodate more
2
3
spillover. Ammonia desorption from Pd/Al O (Fig. 4d)
2
The results presented in Fig. 4f established full selectivity to
benzylamine over Pd/Al at low contact time, where
3
exhibited a lower Tmax (470 K) relative to Pd/C, suggesting
O
2 3
weaker interaction of NH with surface acid sites. Total
3
X \ 15 %. This falls short in terms of the productivity
required for practical applications where reaction exclusiv-
ity must be targeted at high conversion. An increase in n/
F served to increase conversion (Fig. 4f) but this was
accompanied by undesired toluene formation. Hydrogenol-
NH release from Pd/Al O coincided with chemisorption
3
2 3
-
and is close to that (0.54 mmol g ) reported for Pd/Al O
1
2
3
by Nam et al. [27]. Ammonia adsorption/desorption mea-
surements confirm a greater level of surface acidity for Pd/
C relative to Pd/Al O . The XPS profile for Pd/Al O ,
ysis to toluene was suppressed over Pd/Al O at low contact
2 3
2
3
2 3
given in Fig. 4e, is characterised by a Pd 3d5/2 BE that is
.3 eV lower than the metallic Pd [29], suggesting partial
time and a configuration based on discrete catalyst beds in
series, each operated at low n/F, should serve to increase
overall nitrile conversion while retaining hydrogenation
selectivity. Unreacted nitrile is converted to target product in
subsequent catalyst beds and given the low rate of benzy-
0
electron transfer from support to metal phase. This is in
accordance with reported electron-rich Pdd- (2–10 nm) on
Al O [21].
2
3
In contrast to Pd/C, Pd/Al O exhibited full selectivity
2
lamine hydrogenolysis over Pd/Al O the amine can be
2 3
3
to the target benzylamine at low contact time (n/
carried through without further reaction. Reaction rates and
product selectivities obtained in the multi-bed catalyst
arrangement (see Fig. 5) are recorded in Table 2. The same
total mass of catalyst was divided equally into N (= 1–4)
-
3
F \ 3 9 10 ) as shown in Fig. 4f. This result is signifi-
cant in terms of clean synthesis of benzylamine when
compared with existing literature where dibenzylamine
production (up to 16 % selectivity) has been reported over
Pd/Al O in liquid phase reaction [4, 7, 9]. From a con-
beds with the same inlet benzonitrile and H flow rate.
2
Selectivity is assessed at two representative nitrile conver-
sions (X = 45 and 60 %). An increase in overall reaction
rate and benzylamine selectivity (to 100 %) was observed
with increasing bed number up to the quadruple bed
arrangement. Our results demonstrate the beneficial effect of
multi-beds of the same catalyst in series to enhance selective
2
3
sideration of Pd electronic character, a partial negative
charge (Pdd-) should favour interaction with the polarised
d?
carbon (C ) of the –C:N group through a side-on
d-
adsorption due to repulsion of Pd with the nitrogen lone
pair (Fig. 3b). Hydrogenation of unsaturated –C:N to
saturated amine –CH –NH reduces carbon polarity,
-
1
amine formation. We have achieved a rate (99 mol h
-
2
2
-
molPd ) with full selectivity to benzylamine that is signifi-
1
weakening adsorption with the result that the primary
amine product desorbs without further reaction. In common
with Pd/C, toluene selectivity was increased at higher n/F,
-
1
-1
cantly higher than that reported (ca. 0.01 mol h molCu
calculated from data provided) for reaction over Cu–MgO
(at 513 K) with the highest reported selectivity to benzy-
lamine (99 %) in gas phase continuous reaction [14].
indicative of
a
sequential pathway, i.e. benzoni-
trile ? benzylamine ? toluene. Reaction of benzylamine
1
23

Selective Production of Benzylamine via Gas Phase Hydrogenation of Benzonitrile over…
Fig. 5 Schematic diagram for
the catalytic system based on
2 3
single and multiple Pd/Al O
catalyst beds in series with
reactant and product
distribution. Note: ‘‘m’’
represents the total mass of
catalyst; BN benzonitrile
reactant, BA benzylamine, TOL
toluene
Table 2 Selectivity (S
benzylamine (BA) and toluene
TOL) at X = 45 and 60 % and
benzonitrile hydrogenation rate
R) using multiple Pd/Al
catalyst beds in series
i
) to
-1 -1
R (mol h molPd )
Catalyst beds
S
i, X = 45 % (%)
S
i, X = 60 % (%)
(
BA
TOL
BA
TOL
(
2
O
3
Single-bed
78
93
22
7
62
90
38
10
4
48
75
84
99
Double-bed
Triple-bed
97
3
96
Quadruple-bed
100
0
100
0
Acknowledgments We acknowledge financial support to Y. Hao
and M. Li through the Overseas Research Students Award
Scheme (ORSAS).
4
Conclusions
Gas phase continuous hydrogenation of benzonitrile over
Pd/C (mean Pd size = 2.5 nm) generated toluene as prin-
cipal product with higher nitrile consumption rates than Pd/
Al O (mean Pd size = 3.0 nm). This is ascribed to greater
References
2
3
available surface reactive hydrogen and a contribution of
1. Krupka J, Pasek J (2012) Curr Org Chem 16:988
2
. Nishimura S (2001) Handbook of heterogeneous catalytic
hydrogenation for organic synthesis. Wiley, New York
. Lawrence SA (2004) Amines: synthesis, properties and applica-
tions, vol 1. Cambridge University Press, Cambridge
surface acidity to nitrile activation. Electron transfer to the
carbon support with the formation of Pdd? (from XPS
3
analysis) favours a perpendicular adsorption of benzonitrile
through the nitrogen lone pair with hydrogen attack to
4. Bakker JJW, van der Neut AG, Kreutzer MT, Moulijn JA,
Kapteijn F (2010) J Catal 274:176
5. Bawane SP, Sawant SB (2004) Chem Eng J 103:13
generate consecutively, imine, amine and toluene. The
occurrence of Pdd- on Al O results in a side-on activation
2
3
6
7
. Heged }u s L, M a´ th e´ T (2005) Appl Catal A 296:209
. Yoshida H, Wang Y, Narisawa S, Fujita S-I, Liu R, Arai M
(2013) Appl Catal A 456:215
of benzonitrile and desorption of benzylamine as exclusive
-
3
product at n/F \ 3 9 10 h; hydrogenolysis to toluene
was observed at higher n/F. Utilisation of multiple Pd/
Al O beds in series serves to retain full selectivity to
8
9
. Yap AJ, Masters AF, Maschmeyer T (2012) ChemCatChem
4
:1179
. Jes u´ s YL-D, Johnson C, Monnier J, Williams C (2010) Top Catal
3:1132
2
3
benzylamine, delivering a far higher production rate than
previously reported. We have established the viability of
operating catalyst beds in series under isothermal condi-
tions as an effective means of delivering high and exclusive
amine throughput.
5
10. Chatterjee M, Kawanami H, Sato M, Ishizaka T, Yokoyama T,
Suzuki T (2010) Green Chem 12:87
1. Dom ´ı nguez-Quintero O, Mart ´ı nez S, Henr ´ı quez Y, D’Ornelas L,
Krentzien H, Osuna J (2003) J Mol Catal A 197:185
1
123

Y. Hao et al.
1
1
2. Cheng H, Meng X, Wu C, Shan X, Yu Y, Zhao F (2013) J Mol
Catal A 379:72
3. Jim e´ nez-Gonz a´ lez C, Poechlauer P, Broxterman QB, Yang B-S,
Ende D, Baird J, Bertsch C, Hannah RE, Dell’Orco P, Noorrnan
H, Yee S, Reintjens R, Wells A, Massonneau V, Manley J (2011)
Org Proc Res Dev 15:900
4. Marella RK, Koppadi KS, Jyothi Y, Rao KSR, Burri DR (2013)
New J Chem 37:3229
5. Rode CV, Arai M, Shirai M, Nishiyama Y (1997) Appl Catal A
23. Prelazzi G, Cerboni M, Leofanti G (1999) J Catal 181:73
24. C a´ rdenas-Lizana F, Keane MA (2015) Phys Chem Chem Phys
17:28088
25. Braos-Garc ´ı a P, Maireles-Torres P, Rodr ´ı guez-Castell o´ n E,
Jim e´ nez-L o´ pez A (2003) J Mol Catal A 193:185
26. Turco M, Bagnasco G, Cammarano C, Senese P, Costantino U,
Sisani M (2007) Appl Catal B 77:46
27. Nam I, Cho KM, Seo JG, Hwang S, Jun K-W, Song IK (2009)
Catal Lett 130:192
28. Pusk a´ s R, S a´ pi A, Kukovecz A, K o´ nya Z (2012) Top Catal
55:865
1
1
1
1
48:405
6. Taylor PD, Thomas WD, Buchanan DW (1994) Process of vapor
phase catalytic hydrogenation of maleic anhydride to c-butyro-
lactone in high conversion and high selectivity using an activated
catalyst. United States Patent 5347021, 13 Sept 1994
29. Brun M, Berthet A, Bertolini JC (1999) J Electron Spectrosc
104:55
30. Jiang L, Gu H, Xu X, Yan X (2009) J Mol Catal A 310:144
31. Strunk MR, Williams CT (2003) Langmuir 19:9210
32. Ortiz-Hernandez I, Williams CT (2007) Langmuir 23:3172
33. Rask o´ J, Kiss J (2006) Appl Catal A 298:115
34. Kim S, Byl O, Yates JT (2005) J Phys Chem B 109:6331
35. de Pedro ZM, Diaz E, Mohedano AF, Casas JA, Rodriguez JJ
(2011) Appl Catal B 103:128
36. Balla A, Marcu C, Axente D, Borodi G, Laz a˘ r D (2012) Cent Eur
J Chem 10:1817
37. Prins R (2012) Chem Rev 112:2714
1
7. Arakawa H, Dubois JL, Sayama K (1992) Energy Convers Mgmt
3
3:521
1
1
8. Hu C, Wu J, Zhang H, Qin S (2007) AIChE J 53:2925
9. Gawade P, Alexander A-MC, Clark R, Ozkan US (2012) Catal
Today 197:127
0. P e´ rez-Ram ´ı rez J, Garc ´ı a-Cort e´ s JM, Kapteijn F, Ill a´ n-G o´ mez MJ,
Ribera A, Lecea CS-Mnd, Moulijn JA (2000) Appl Catal B
2
2
5:191
2
2
1. C a´ rdenas-Lizana F, Hao Y, Crespo-Quesada M, Yuranov I, Wang
X, Keane MA, Kiwi-Minsker L (2013) ACS Catal 3:1386
2. Janiak T, Okal J (2009) Appl Catal B 92:384
1
23