Hydrogenation of nitrobenzene to 4-aminophenol over supported platinum catalysts

Article abstract of DOI:10.1021/op700049p

The conversion of nitrobenzene (NB) to p-aminophenol (PAP) takes place by way of an initial partial hydrogenation to produce phenylhydroxyl amine (PHA) which then undergoes an in situ acid-catalyzed rearrangement to PAP. This reaction is most commonly run using Pt/C catalysts in the presence of aqueous sulfuric acid and a surfactant to assist in dispersing the NB throughout the reaction medium. The yield of PAP is closely related to those reaction parameters which facilitate first the partial hydrogenation step and second the acid-promoted rearrangement before further hydrogenation to aniline can take place. The effect which a number of reaction parameters such as hydrogen pressure, reaction temperature, stirring rate, and the amounts of NB, the catalyst, and the surfactant present in the reaction mixture had on the rate and selectivity of the hydrogenation was examined. Optimization of these parameters led to the formation of PAP at a selectivity (PAP/AN) of 5.4 with a productivity of over 80,000 g PAP/g Pt/h.

Full text of DOI:10.1021/op700049p

Organic Process Research & Development 2007, 11, 681−688
Hydrogenation of Nitrobenzene to 4-Aminophenol over Supported Platinum
Catalysts
Setrak K. Tanielyan,† Jayesh J. Nair,† Norman Marin,† Gabriela Alvez,† Robert J. McNair,‡ Dingjun Wang,‡ and
Robert L. Augustine*,†
Center for Applied Catalysis, Seton Hall UniVersity, South Orange, New Jersey 07079, U.S.A., and Johnson Matthey,
West Deptford, New Jersey 08066, U.S.A.
Abstract:
While most of the information regarding the effect of the
catalyst and catalyst modifiers, the nature of the surfactant,
or specific process improvements are well documented in
patent sources,10-17 surprisingly little has been published in
the open literature.18-24 The commonly accepted reaction
mechanism involves the initial hydrogenation of NB to give
the intermediate PHA which desorbs from the catalyst surface
into the aqueous phase and is further converted to PAP
through an acid-catalyzed rearrangement21,25,26 (eq 1). Some
of the PHA. though, can be further hydrogenated to aniline
(AN). The ratio of the products from these two pathways
determines the reaction selectivity toward PAP.
The conversion of nitrobenzene (NB) to p-aminophenol (PAP)
takes place by way of an initial partial hydrogenation to produce
phenylhydroxyl amine (PHA) which then undergoes an in situ
acid-catalyzed rearrangement to PAP. This reaction is most
commonly run using Pt/C catalysts in the presence of aqueous
sulfuric acid and a surfactant to assist in dispersing the NB
throughout the reaction medium. The yield of PAP is closely
related to those reaction parameters which facilitate first the
partial hydrogenation step and second the acid-promoted
rearrangement before further hydrogenation to aniline can take
place. The effect which a number of reaction parameters such
as hydrogen pressure, reaction temperature, stirring rate, and
the amounts of NB, the catalyst, and the surfactant present in
the reaction mixture had on the rate and selectivity of the
hydrogenation was examined. Optimization of these parameters
led to the formation of PAP at a selectivity (PAP/AN) of 5.4
with a productivity of over 80,000 g PAP/g Pt/h.
Introduction
The catalytic hydrogenation of nitrobenzene (NB) to
p-aminophenol (PAP) under phase transfer conditions is an
industrially important process first reported by Henke and
Vaughen in 1940.1 PAP is an important intermediate which
has been used in the production of analgesic drugs such as
acetaminophen,2 photographic developers, and dyes.3 This
hydrogenation is generally conducted in a four-phase sys-
tem: an organic phase of the NB, an aqueous phase of dilute
sulfuric acid, a supported platinum catalyst, and a hydrogen
atmosphere. A significant rate enhancement and improvement
in the selectivity to PAP was achieved when the hydrogena-
tion was carried out in the presence of a phase transfer agent
(PTA) such as a quaternary ammonium salt or a polyether
polyol surfactant4-9 which facilitated the dispersion of the
NB throughout the aqueous reaction mixture.
Details regarding the influence of some reaction variables
on the hydrogenation of NB to PAP in a multiphase system
in the presence of a platinum catalyst and N,N-dimethyl-
dodecylamine as a PTA are presented below. The variables
(10) Benwell, N. R. W.; Buckland, I. J. British Patent 1,181,969, 1970.
(11) Greco, N. P. U.S. Patent 3,953,509, 1976.
(12) Dunn, T. J. U.S. Patent 4,264,529, 1981.
(13) Gubelmann, M.; Maliverney, C. U.S. Patent 5,288,906, 1994.
(14) Chaudhari, R. V.; Divekar, S. S.; Vaidya, M. J.; Rode, C. V. U.S. Patent
6,028,227, 2000.
(15) Rylander, P. N.; Karpenko, I. M.; Pond, G. R. U.S. Patent 3,715,397, 1973.
(16) Caskey, D. C.; Chapmann, D. W. U.S. Patent 4,571,437, 1986.
(17) Mais, F.-J.; Marhold, A.; Steffan, G. U.S. Patent 6,504,059, 2003.
(18) Jiang, T. M.; Hwang, J. C.; Ho, H. O.; Chen, C. Y. J. Chin. Chem. Soc.
1988, 35, 135.
* To whom correspondence should be addressed. E-mail: augustro@shu.edu.
† Seton Hall University.
(19) Hwang, J. C.; Chang, K. A.; Chen, C. Y.; Juang, T. M. Chin. Pharm. J.
1992, 44, 475.
‡ Johnson Matthey.
(1) Henke, C. O.; Vaughen, J. V. U.S. Patent 2,198,249, 1940.
(2) Kirk-Othmer Mitchell, S. Encyclopedia of Chemical Technology, 4th ed.;
Wiley-Interscience: New York, 1992; p 481.
(3) Raavichandran, C.; Chellammal, S.; Analtharman, P. N. J. Appl. Electro-
chem. 1989, 19, 45.
(20) Gao, Y.; Wang, F.; Liao, S.; Yu, D. React. Kinet. Catal. Lett. 1998, 64,
351.
(21) Rode, C. V.; Vaidya, M. J.; Chaudhari, R. V. Org. Process Res. DeV. 1999,
3, 465.
(22) Liu, Z.-Q.; Hu, A.-L.; Wang, G.-Y. J. Nat. Gas Chem. 1999, 8, 305.
(23) Rode, C. V.; Vaidya, M. J.; Jaganathan, R.; Chaudhari, R. V. Chem. Eng.
Sci. 2001, 56, 1299.
(24) Komatsu, T.; Hirose, T. Appl. Catal., A 2004, 276, 95.
(25) Ternery, L.T. Contemporary Organic Chemistry; W.B. Saunders Co. Ltd.:
Philadelphia, PA, 1976; p 661.
(4) Spiegler, L. U.S. Patent 2,765,342, 1956.
(5) Brenner, R. G. U.S. Patent 3,383,416, 1968.
(6) Brown, B. B.; Schilling, F. A. E. U.S. Patent 3,535,382, 1970.
(7) Sathe, S. S. U.S. Patent 4,176,138, 1979.
(8) Derrenbacker, E. L. U.S. Patent 4,307,249, 1981.
(9) Miller, D. C. U.S. Patent 5,312,991, 1994.
(26) Stratz, A. M. Chem. Ind. (Dekker) 1984, 5 (Catal. Org. React.), 335.
10.1021/op700049p CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/13/2007
Vol. 11, No. 4, 2007 / Organic Process Research & Development
•
681

Table 1. Physical properties of the 1.5% Pt/C catalysts used
for the hydrogenation of AN to PAP under standard
reaction conditions
was accurately maintained at the desired value by using a
microprocessor controlled magnetic stirrer (VWR brand
model 400S). Once the desired temperature was reached the
stirring was stopped, the nitrogen was replaced with hydrogen
by multiple flush cycles, and the flask was pressurized to
10 psig with hydrogen. The stirring was reinitiated and
computer monitoring of the reaction parameters begun. The
progress of the reaction was followed by the hydrogen
uptake, and once the required conversion level was reached,
the reaction flask was cooled to ambient and a sample
extracted for HPLC analysis. In some instances the reaction
was run to full conversion with small portions of the reaction
mixture extracted via gastight syringe at increments of 10%
conversion, and the samples were analyzed.
metal
mean
catalyst
acidity/
% metal
cat. #
JM#
surface
particle
dispersiona
area (m2/g) size (µm) basicity
1
2
3
4
5
6
7
8
9
C-5105
C-5113
C-5118
C-5117
C-5109
C-5126
C-5112
C-5106
C-5114
C-5110
53
62
73
74
79
23
25
65
79
81
1.96
2.39
2.71
2.76
2.94
0.87
0.91
2.40
2.94
3.00
36.0
36.0
36.0
36.0
36.0
20.0
20.0
34.0
34.0
34.0
acidic
acidic
acidic
acidic
acidic
neutral
neutral
neutral
neutral
neutral
10
Multiple Catalyst Use. As commonly done in most of
commercial applications, the catalyst was separated after each
reaction by terminating the hydrogenation when about 85-
90% of the NB has reacted. The remaining NB was sufficient
to hold the catalyst in a suspended form and allowed an easy
extraction of the products dissolved in the aqueous phase.
Recharging the reactor with a fresh portion of the starting
reaction mixture permitted the multiple use of the catalyst
without exposing it to air.
a (Number of metal atoms on the surface of the catalyst/total number of metal
atoms) × 100.
studied were the hydrogen pressure, the stirring efficiency,
the temperature, and the amounts of catalyst, NB, sulfuric
acid, and the PTA. The effects of each of these variables on
the initial rate of hydrogenation, the reaction productivity,
and the selectivity to PAP formation were investigated.
Analytical Methods. The analytical method used was
similar to that reported.19 A reversed phase Eclips XDB C18
column (150 mm × 4.6 mm ID), particle size 3.5 µk (Agilent
Technologies) was used. The mobile phase A was a 0.084M
NH4OAc buffer at pH ) 6.8 and mobile phase B was a 10:1
solution of acetonitrile/i-PrOH. The flow rate was set at 1
mL/min at 45% mobile phase A in isocratic mode. To
prepare the sample solutions, 0.5 mL of the reaction mixture
was pipetted into a 50 cm3 volumetric flask, and the sample
was diluted to the mark with a 0.5% H2SO4 solution of the
internal standard, N-phenyl-1,2-phenylenediamine (0.5 mg/
mL).
Data Processing and Analysis. At the end of each run
the hydrogen uptake data were plotted against time, and the
initial reaction rates were calculated from the slope of the
initial linear portion of the hydrogen uptake curve. On the
basis of this and on the results of the HPLC analysis of the
aqueous phase, the following parameters were used to
evaluate the efficiency of the reaction:
Experimental Section
Materials. The materials used were all certified ACS
grade. The nitrobenzene, aniline, N,N-dimethyldodecylamine,
and p-aminophenol were obtained from Sigma-Aldrich and
were used without further purification. Ten 1.5% Pt/carbon
catalysts were received from Johnson Matthey, Precious
Metals Division. Each had a uniform distribution of Pt on
the support and was supplied in a wet form. The properties
of these catalysts are listed in Table 1. In screening these
catalysts the amount of each used corresponded to a dry
weight of 13.5 mg of catalyst (1.04 µmol of Pt).
Apparatus. Experiments were carried out in a low-
pressure glass volumetric system described elsewhere.27 This
apparatus allowed for the simultaneous running of four
reactions with independent temperature and pressure control
as well as the monitoring of the hydrogen uptake of each
reaction separately. Each reactor was composed of an Ace
Glass reaction flask capped with a Teflon plug which
accommodated injection and thermocouple ports. The Teflon
plug was additionally equipped with a suspended stirrer also
made of Teflon.
Nitrobenzene Hydrogenation. In a representative hy-
drogenation the reactor was charged with sufficient wet
catalyst equivalent to 13.5 mg on a dry basis (1.04 µmol
Pt), 30 mL of 10% w/w aqueous H2SO4, 0.75 mL of the
phase transfer agent (PTA) solution (0.145 M N,N-dimeth-
yldodecylamine dissolved in 10% aqueous sulfuric acid) and
4.86 g (39.3 mmol) of freshly distilled NB. The reactor was
connected to the system and was then alternately filled with
nitrogen to 15 psig and purged five times. The reactor
temperature was then slowly raised to the set value under
constant stirring. The stirring rate, which appeared to be the
most critical requirement for obtaining reproducible results,
conversion of NB: ([PAP] + [AN]) × 100/([NB]), %
selectivity to PAP: g PAP/g AN
rate of H2 uptake from the H2 uptake curve to
30% conversion, mmol/min
productivity to PAP: g PAP/g Pt/
Results and Discussion
The reaction conditions described in previous reports
concerning the partial hydrogenation of NB in acid media
to form p-aminophenol are listed in Table 2. As seen from
these data the platinum-catalyzed hydrogenation of NB in
aqueous sulfuric acid has generally been carried out using
relatively low concentrations of NB, a small amount of a
Pt/C catalyst, temperatures of 80-90 °C, and low hydrogen
pressures, generally below atmospheric. One publication21
(27) Augustine, R. L.; Tanielyan, S. K. Chem. Ind. (Dekker) 2003, 89 (Catal.
Org. React.), 73.
682
•
Vol. 11, No. 4, 2007 / Organic Process Research & Development

Table 2. Conditions reported for the hydrogenation of
nitrobenzene to PAP
ref
H2
T
4
7
8
9
12
21
400
80
3
this
paper
10
3.9
88
1
0.9a
85
3
0.4b
85
10
80
1.5
30
0.4c
90
(psig)
90
(°C)
Pt/C
3
Pt/Al2O3
7.5
1.5
(%)
Pt
16
7.5
7.5
185
110
650
1
0.2
(mg)
NB
(g)
66.2 108
28.2 80
1,000 640
1200 108
665 80
4,000 650
2 mL
23
11
4.86
3.84
H2SO4
(g)
H2O
to 100 30
23 mg
(mL)
PTA
NB/Pt
0.75 g 0.5 mL 2 mL 4 g
4100 14,400 25,000 40,000 14,400 22,000 24,000
Figure 1. Comparison between metal surface area and reaction
productivity and selectivity in the hydrogenation of NB to PAP
over the catalysts listed in Table 1.
(w/w)
NB/acid 2.3
1.35
5.4
1.68
4.1
1.8
6.6
1.35
5.5
2.0
1.4
1.27
6.7
(w/w)
PAP/AN 7.5
Table 3. Initial hydrogenation rate, reaction productivity,
and selectivity observed in the multiple use of catalyst 1
under the standard reaction conditions listed in the last
column of Table 1a
a 6-25 in. of water (0.2-0.9 psig). b 10 in. of water (0.4 psig). c 6-10 in. of
water.
reported the use of much higher hydrogen pressures, but the
resulting reaction selectivity was very low. Since this is a
four-phase system, an efficient and accurately controlled
mixing of the two liquid phases is crucial for obtaining
reproducible and meaningful results. On the basis of these
published data the “standard” reaction conditions to be used
in this study were selected as being fairly consistent with
those listed in Table 1 while still allowing for a reason-
able amount of variation when examining the effect which
each parameter has on the outcome of the reaction. These
standard conditions are listed in the last column of Table 2.
As mentioned above, the degree and type of agitation of
the reaction mixture is critical to the success of this
hydrogenation. It is, however, difficult to compare stirring
rates with the efficiency of agitation since the design and
size of the reactor are also factors influencing mass transfer.
On the basis of some preliminary studies it was determined
that with our reactor system a stirring rate of 1700 rpm was
sufficient in order to have a reaction reasonably free of
diffusion limitations as shown by the data presented in
Figure 4.
The Catalyst. The catalysts listed in Table 1 were each
used for the hydrogenation of NB to PAP under our
“standard” reaction conditions. Figure 1 shows the relation-
ship between the metal surface areas of the catalyst and the
product selectivity and productivity obtained with each. Most
of these catalysts produced fairly high selectivities and
productivities so the selection as to which one to use for the
reaction parameter study was not straightforward. It was
decided, though, that catalyst 1, which had only a moderate
metal surface area (and dispersion) but still gave good
selectivity and productivity, was the most reasonable to use
in the further study. It was felt that with this moderate metal
surface area this catalyst might be more influenced by
changes in the various reaction parameters.
rate
mmol/min
0.592
0.592
0.555
productivity
g PAP/g Pt /h
15,600
selectivity
PAP/AN
5.7
first use
second use
third use
15,300
14,900
15,000
5.7
5.9
5.9
fourth use
0.533
a The hydrogenation was carried out at [NB] ) 1.13 M, stir rate 1700 rpm,
pressure 10 psi, catalyst 1 ) 33.7 mg (wet), temperature 90 °C, [PTA] ) 3.1 ×
10-3 M.
successive NB hydrogenations with the resultant initial rates,
productivities, and selectivities listed in Table 3. These data
show that the catalyst performance during these repeated
cycles remained relatively stable as measured by the
hydrogen uptake rate and the PAP selectivity and productiv-
ity. The cumulative turnover number for the catalyst after
these four cycles was 132,000 mol PAP/mol Pt at an average
selectivity to PAP in the 87-88% range and a productivity
of about 15,000 g PAP/g Pt/h.
Reaction Details. The hydrogen uptake curve obtained
from a hydrogenation run under our “standard” conditions
using catalyst 1 is shown in Figure 2. Examination of this
curve along with a careful observation of the reaction mixture
shows that the hydrogenation is composed of four distinct
regions. The first, A, is characterized by the linear rate of
hydrogen uptake which continues to about the 70-75%
conversion level. In the second region, B, the rate of
hydrogen uptake slows, but the acid/NB/water emulsion is
still intact (in these two regions it takes an average of 10-
15 min after the stirring has stopped to see a phase separation
with a bottom organic phase and a top aqueous phase with
the catalyst being in the organic phase). During the third
stage, C, when most of the NB has been consumed and there
is not enough NB present to keep the catalyst in the dispersed
state, the emulsion collapses, and the catalyst separates with
the remaining NB into a few isolated organic droplets. The
fourth stage, D, finishes the reaction with the disappearance
of the remaining organic phase, redispersion of the catalyst
In order to determine the stability of this catalyst under
our standard reaction conditions it was used for four
Vol. 11, No. 4, 2007 / Organic Process Research & Development
•
683

Figure 3. Representation of the four-phase system in the NB
hydrogenation over a Pt/C catalyst.
PTA. The hydrophobic catalyst is contained exclusively
within the NB droplets produced as a result of the stirring
action and the presence of the PTA. The hydrogen migrates
through the aqueous/organic interface into the droplet and
onto a catalyst surface. Here, the initial hydrogenation of
NB to PHA takes place. In an appropriate “hydrogen poor”
environment, the PHA migrates out of the droplet and into
the acidic aqueous phase to rearrange to PAP. However, as
the reaction proceeds, the amount of NB present decreases,
and most likely, the size of the NB droplets also decreases.
Nevertheless, since the amount of catalyst remains the same,
the catalyst concentration within each droplet increases, thus
producing droplets such as that depicted by B in Figure 3.
In such a “catalyst rich” environment the PHA will find it
more difficult to migrate out of the droplet without encoun-
tering more catalyst with the potential for promoting further
hydrogenation. This is apparently one reason why the PAP
selectivity decreases during the course of the reaction,
especially in the later stages of region A and throughout
region B in Figure 1. There is, however, over 6 times more
PAP formed than AN in regions C and D so that further
hydrogenation of PHA is still slower than the rearrangement
by an appreciable amount.
Figure 2. Hydrogen uptake curve and reaction selectivity
recorded during the hydrogenation of NB under the phase
transfer conditions listed in the Experimental Section.
into the aqueous phase, and the slow hydrogenation of the
NB still remaining in the aqueous phase.
The PAP/AN ratios for samples withdrawn at different
times in these four regions are also shown in Figure 2. The
most interesting changes were taking place in the transition
from region A to region B. Here the emulsion was still stable
in appearance, and the hydrogenation was proceeding at a
relatively constant rate, but the PAP selectivity showed a
continuous deterioration. In regions C and D the hydrogena-
tion took place at the expense of the NB dissolved in the
aqueous phase. In these regions the system produced AN
and PAP at relatively constant rates which led to a
consistency in the PAP/AN ratios. In order to eliminate the
effect of the variation in the selectivity to PAP and to have
a reliable basis for comparison in all further experiments,
all subsequent reactions were quickly quenched at the end
of region B, and the product composition was analyzed after
cooling to ambient.
Stirring Rate and Catalyst Quantity. One of the reaction
parameters considered to be critical in achieving a rapid
reaction and high PAP selectivity was the stirring rate. This
should be sufficient to enhance the diffusion of the gaseous
and liquid substrates to and from the catalyst but not too
vigorous to keep an optimum “hydrogen-poor” environment
so that high PAP selectivity can be attained.
Another observation from the analysis of the samples
taken within the same run was that the only products formed
during the hydrogenation were PAP and AN. No other
intermediates or byproducts accumulated in measurable
amounts, a fact which supports the idea that under this set
of conditions the rate-determining step is probably the
formation of the PHA intermediate.6 Once this intermediate
is formed, it can be hydrogenated further to produce AN or
migrate to the acidic aqueous phase to rearrange to PAP
(eq 1).
Since further hydrogenation requires a hydrogen-rich
environment, all reaction variables which would increase the
hydrogen availability to the platinum site should lead to the
formation of more AN and a lower PAP selectivity.
Alternately, creating a “hydrogen-poor” environment with
a decreased hydrogen availability to the catalyst and an
efficient transport of the PHA intermediate to the acidic phase
will promote a higher PAP selectivity.
Running the hydrogenations using the standard conditions
listed in Table 2 but varying the stirring rate from 300 to
2000 rpm produced the results shown in Figure 4A. The most
dramatic changes occurred when the stirring rate was varied
between 300 and 1400 rpm during which the PAP selectivity
declined to a PAP/AN ratio of 7 with the hydrogen uptake
rate increasing and following a second-order relationship.
When the stirring rate was increased from 1400 to 1800 rpm,
the selectivity remained essentially constant, and the hydro-
genation rate appeared to be leveling off, indicating that
mass-transfer processes were becoming less important to the
reaction with this range of agitation. All further reactions
were run using 1700 rpm as the standard stirring rate.
The effect of the catalyst quantity on the rate of hydrogen
uptake and the PAP selectivity was studied over the range
of 10-60 mg of wet catalyst, corresponding to substrate-
to-Pt mol ratios of 127,000:1 to 21,200:1. As shown in Figure
4B, varying the quantity of catalyst while maintaining the
other standard conditions resulted in an initial rate increase
with an increase in the amount of catalyst with a trend toward
Figure 3 shows a cartoon depicting the various processes
occurring during the hydrogenation of NB in presence of a
684
•
Vol. 11, No. 4, 2007 / Organic Process Research & Development

Figure 4. Dependence of the hydrogen uptake rate (O), the selectivity to PAP (4), and the productivity (0) on the stirring rate (A)
and on the catalyst quantity (B); other reaction parameters are as given in the Experimental Section.
saturation observed using the higher catalyst amounts. This
tendency of the hydrogen uptake to deviate from the first-
order reaction kinetics indicated that some diffusion-related
phenomenon may be taking place at the higher catalyst
loadings within the limited space provided by the organic
phase (Figure 2). The PAP selectivity and productivity both
dropped off considerably at higher catalyst loadings.
The data shown in graphs A and B of Figure 4 comple-
ment each other and fit well into the model depicted in Figure
2 in which the PHA intermediate once formed in the organic
phase will have to migrate from within the NB droplet
towards the exterior acidic interface to rearrange or remain
and be hydrogenated further to AN. According to this model,
increasing the stirring rate at given substrate and PTA
concentrations should lead to a reduction in the size of the
NB particle and, respectively, to an increase in the contact
area between the organic and the aqueous phases. The
reduction in the size of the NB droplet is expected to act in
two opposite directions. From one side, this will increase
the contact area between the organic and the aqueous phase
and will improve the rate at which the PHA can migrate out
of the droplet and into the acidic aqueous phase. It will also
increase the rate of hydrogen transport to the interior of the
particle. Higher hydrogen availability will facilitate the PHA
formation but will also enhance the rate of AN formation.
This aspect is supported by the observed second-order
relationship between the stirring rate and the hydrogen uptake
rate and the reduction in the PAP selectivity for stirring rates
in the 300-1400 rpm region. Above that value the rate of
hydrogen diffusion likely remains constant, and the size of
the NB particles probably does not change significantly, a
factor which would also explain the independence of the
hydrogenation rate and the PAP selectivity at stirring rates
above 1400 rpm.
difficult to migrate into the acidic aqueous phase because
of the amount of catalyst present in the NB droplet, further
hydrogenation of the PHA to AN has a high probability of
taking place.
Temperature and Hydrogen Pressure Effect. The effect
which the reaction temperature had on the hydrogenations
is shown in Figure 5A in which a maximum in the reaction
rate and PAP productivity is found at a reaction temperature
of 90 °C. The relatively narrow temperature range studied
(80-100 °C) was used because outside of this temperature
interval the rate of hydrogenation was either too low or the
reaction showed decreased selectivity.
The effect of the hydrogen pressure on the initial rate and
on the selectivity under otherwise standard conditions is
shown in Figure 5B. With an increase in the hydrogen
pressure both the initial rate and productivity increased
almost linearly up to about 30 psig beyond which point they
reached constant levels of about 1.4 mmol/min and about
36,000 g PAP/g Pt/h respectively. The PAP productivity,
which is a parameter relating the catalyst performance to
the selectively in forming PAP, increased 2-fold in the
hydrogen pressure range shown in Figure 5B. Increasing the
hydrogen pressure and consequently the local hydrogen
concentration in the interior of the organic droplets appears
to affect primarily the rate of the first hydrogenation step to
give PHA and to a lesser extent the further hydrogenation
to give AN. This suggests that even under higher hydrogen
availability conditions the further hydrogenation of PHA to
AN remains relatively slow compared with the rearrangement
of PHA to PAP. This is not too surprising given that the
initial step involves the hydrogenation of an unsaturated
species while the conversion of PHA to AN requires the
hydrogenolytic cleavage of a N-O single bond.
Concentration of H2SO4 in the Aqueous Phase. The
effect of the concentration of the H2SO4 in the aqueous phase
was studied over the range of 5-15% w/w under the standard
reaction conditions. The results are shown in Figure 6.
Increasing the acid concentration over this range led to a
slight reduction in the hydrogen uptake rate and, at the same
The drop in PAP productivity and selectivity observed at
higher catalyst quantities (Figure 4B) is probably another
example of the effect of having large amounts of catalyst
present in the NB droplets. As discussed above (Figure 2B)
after the initial formation of the PHA, if this species finds it
Vol. 11, No. 4, 2007 / Organic Process Research & Development
•
685

Figure 5. Dependence of the hydrogen uptake rate (O), the selectivity to PAP (4), and the productivity (0) on the temperature (A)
and on hydrogen pressure (B); other reaction parameters are as given in the Experimental Section.
At the lower acid concentration increasing the substrate
concentration over the range of interest resulted in a near
linear increase in the initial hydrogen uptake which reached
a saturation level at a substrate concentration of 1.14 M. The
PAP productivity passed through a maximum at [NB] ≈ 1
M, while the selectivity to PAP deteriorated continuously
to a PAP/AN ratio of 5 at high substrate concentrations
(Figure 7A). The same trend in the PAP selectivity was
observed under the “high” acid concentrations as well (Figure
7B). In contrast to the “low” acid system, in the presence of
the more concentrated acid the initial rate of hydrogen uptake
remained relatively constant, and as a result, the productivity
showed a distinct deterioration with increasing [NB].
In general, an increase in the initial concentration of NB
at constant PTA concentration should probably lead to the
formation of a suspension in which the NB particles have a
larger average particle size. The immediate result of such
changes would be that the PHA intermediate will have to
Figure 6. Dependence of the hydrogen uptake rate (O), the
traverse longer distances to reach the exterior of the particle
to form PAP while the probability for the secondary
hydrogenation to form AN will be favored in the larger NB
droplets. This can also explain the sharp decline in the PAP
selectivity at high NB concentration under both acid con-
centrations.
Phase Transfer Agent Concentration Effect. The PTA
concentration was also varied over the range of 1 - 5 ×
10-3 M with the results shown in Figure 8A-C. The effect
of the [PTA] was studied at three different sets of condi-
tions: (1) at low substrate and low acid concentrations (8A),
(2) at low substrate and at high acid concentrations (8B),
and (3) at high substrate and high acid concentrations (8C).
As shown, the hydrogen uptake rate in the first case (8A)
increased linearly with increasing PTA concentrations to
about 4 × 10-3 M after which point the rate remained
relatively constant. Surprisingly, the selectivity to PAP in
this region was not affected. It is worth noting that the molar
selectivity to PAP (4), and the productivity (0) on the
concentration of H2SO4 in the aqueous phase; other reaction
parameters are as given in the Experimental Section.
time, to an increase in the PAP selectivity. A further increase
in the acid concentration did not produce significant changes
in the reaction efficiency. The initial hydrogenation rate and
the PAP selectivity remained fairly constant at the higher
concentration of H2SO4. These findings are in agreement with
the results reported earlier for the NB hydrogenation carried
out under similar conditions but at higher hydrogen pres-
sures.18,21
Nitrobenzene Concentration Effect. Figure 7 shows the
effect of the initial NB concentration in 10% aqueous H2-
SO4 (A) and 15% aqueous H2SO4 (B) on the hydrogenation
rate, PAP selectivity and reaction productivity. The [NB]
was varied over the range 0.5-1.6M. A significant difference
was found in the performance of the system under the “low”
acid vs “high” acid regimes.
686
•
Vol. 11, No. 4, 2007 / Organic Process Research & Development

Figure 7. Dependence of the hydrogen uptake rate (O), the selectivity to PAP (4), and the productivity (0) on the NB concentration
using 10% aqueous H2SO4 (A) and 15% H2SO4 (B); other reaction parameters are as given in the Experimental Section.
Figure 8. Dependence of the hydrogen uptake rate (O), the selectivity to PAP (4), and the productivity (0) on the PTA concentration
using 4.86 g of NB and 10% aqueous H2SO4 (A), 4.86 g of NB and 15% aqueous H2SO4 (B), and 6.06 g of NB and 15% aqueous
H2SO4; other reaction parameters are as given in the Experimental Section.
ratio of NB/PTA varied from 1000:1 at the low PTA
concentration to 200:1 for the highest PTA concentration.
In an attempt to further understand how the surfactant could
affect the overall performance of this complex system one
should consider the following simplified structure of small,
water-immiscible organic droplets uniformly dispersed in the
aqueous phase. In a model such as this a surfactant molecule
interacts with the organic phase in the droplet with the alkyl
chain being oriented towards the interior of the particle and
the hydrophilic part forming a protective film around the
dispersed phase.28 It is also known that ionic surfactant
molecules would form a more dense population on the
surface of the particle than nonionic species due to electro-
static repulsion between their head groups. Such a particle
forms a fairly immobile layer of ions that stick tightly to
the surface and that may include water molecules. The
densely charged particle attracts an oppositely charged ionic
atmosphere. The inner shell of charge and the outer
atmosphere are called the electric double layer the density
of which is a function of the concentration of the surfactant.
With such a model in mind, one can hypothesize that an
increase in the surfactant concentration could lead to a
reduction in the size of the droplet and, as such, would lead
to an increase in the rate of hydrogen transport into the
interior of the particle. This, in turn, would bring about a
reaction rate enhancement. When the PTA concentration is
further increased above a certain level, the surface charge
may become strong enough to hinder the hydrogen flow
(28) Atkins, P. W. Physical Chemistry, 4th ed.; W.H. Freeman and Company:
New York, 1990; p 708.
Vol. 11, No. 4, 2007 / Organic Process Research & Development
•
687

through the interface and cause a leveling off of the reaction
rate. This surface charge, though, does not appear to affect
the migration of the PHA into the acidic aqueous phase since
the PAP selectivity remained constant over the whole range
of PTA concentrations (Figure 8A).
Similar conclusions can be drawn for the reactions run
with low NB and high acid concentration. Here, because of
the higher proton concentration in the aqueous phase, the
higher surface charge around the NB droplet may further
hinder the hydrogen flow through the interface and cause
an additional reduction in the reaction rate (Figure 8B).
Increasing simultaneously the NB concentration and the acid
concentration led to further deterioration in the PAP selectiv-
ity (Figure 8C).
Optimized Reaction Conditions. Examination of the data
presented in Figures 3-7 shows that modification of our so-
called “standard” reaction conditions can lead to an increase
in PAP formation. The data indicate that such an increase
could be accomplished by increasing the hydrogen pressure
and the NB concentration while decreasing the amount of
catalyst used. A small increase in the PTA concentration
could also be of advantage. Changes in reaction temperature,
agitation rate, and sulfuric acid concentration do not appear
to result in a significant improvement in the reaction
selectivity or productivity. As shown in Table 3, using the
standard reaction conditions produces PAP with an initial
hydrogen rate of 0.59 mmol/min, a productivity of 15,600 g
PAP/g Pt/h and a selectivity of 5.7. To ascertain the extent
to which the PAP productivity would increase when “opti-
mized” reaction conditions were used, the hydrogenation was
run with the amount of NB increased from 4.86 to 6.88 g,
the hydrogen pressure increased from 10 to 35 psig, the PTA
concentration increased from 3.1 × 10-3 to 3.5 × 10-3 M,
and the amount of catalyst decreased from 33.5 to 20 mg
(wet). This reaction proceeded at an initial rate of 1.73 mmol
H2/min with a productivity of 81,700 g PAP/g Pt/h and a
selectivity of 5.4. Under these conditions the NB/Pt ratio
increased to about 57,000, considerably higher than those
ratios listed in Table 2. Granted, these particular reaction
conditions apply only to the reactor used in this study, but
these data do show that the hydrogen pressure and the NB/
Pt ratio are the most important factors in this reaction. They
also indicate that even though almost all of the previously
reported NB to PAP hydrogenations (Table 2) were run under
low hydrogen pressure the use of somewhat higher pressure,
smaller amounts of catalyst, and a higher amount of NB
appears to be of advantage in increasing the productivity of
the reaction.
Acknowledgment
We acknowledge financial support for this project from
Johnson Matthey.
Received for review February 27, 2007.
OP700049P
688
•
Vol. 11, No. 4, 2007 / Organic Process Research & Development