This article was published in Chemical Engineering and Technology, 38(9), 1625-1636,

2015
http://dx.doi.org/10.1002/ceat.201400468

Hydrogenation of Nitrobenzene over a Pd/Al2O3 Catalyst –

Mechanism and Effect of the Main Operating Conditions

Clara Sá Couto1 Luis M. Madeira2

Clemente Pedro Nunes3 Paulo Araújo1

1CUF – Químicos Industriais, S.A., Quinta da Indústria, Estarreja, Portugal.

2LEPABE, Departamento de Engenharia Química, Faculdade de Engenharia, Universidade do
Porto, Porto, Portugal.
3CERENA, Departamento de Engenharia Química, Instituto Superior Técnico, Universidade de
Lisboa, Lisboa, Portugal

The catalytic hydrogenation of nitrobenzene (NB) was studied in a three-phase basket reactor with a commercial
Pd/Al2O3 sample as catalyst. The kinetic experiments allowed a better understanding of the mechanism behind the
formation of aniline (ANL) and by-products, a topic not yet well comprehended. The effect of some operating
conditions was studied and the existence of more by-products than mentioned in the literature was stated;
specifically, benzene formation was verified. Both the reaction kinetics and selectivity were found to be strongly
dependent on the temperature, while the effect of total pressure is not that pronounced. Moreover, the high selectivity
of the catalyst used in the present work was proved, and as such the deep hydrogenation of ANL to form by-products
only occurs in considerable extension when the NB concentration in the reaction mixture becomes negligible.

Introduction
The catalytic hydrogenation of nitrobenzene (NB) is an important industrial reaction used in
the commercial production of aniline (ANL) that is in turn a major intermediate in the poly-
urethane industry. ANL is mainly consumed in the production of methylene
diphenyldiisocyante (MDI) [1].
ANL production can be carried out in gaseous or in liquid phase [2]. For the vapor phase
reaction, fluidized- and fixed- bed reactors are used, typically at temperatures of 200 ºC to
400 ºC and pressures in the range of 1–10 bar, whereas for the liquid phase the common reactor
used is a slurry reactor operating in the temperature range of 100–300 ºC and at pressures
between 1.5 and 22 bar [3–8]. The choice of the catalyst to be used will depend on the type of
reactor and also on the possibility of being poisoned by nitro compounds. For instance, it
was detected that in the case of Ni-based catalysts the presence of a large excess of NB may cause
some kind of inhibition since NB will be adsorbed preferentially on the most active sites and
consequently only a few sites remain available to promote the hydrogenation reaction [2].
At industrial scale, for the reaction in liquid phase the main catalysts reported are based on

1
noble metals like Pd, Pt, and Ru [5, 6]. Nevertheless, many studies have already been made, in
order to obtain the best catalyst for this reaction (Tab. 1).
Although there is a large volume of literature available de- scribing this reaction, there is not
much information about the reaction mechanism itself. This mechanism is complex and can be
divided into two main steps: the first one is the formation of ANL and the second one is the
production of secondary products from the deep hydrogenation of ANL.
The first step was explained by the mechanism for ANL formation through NB hydrogenation
that was first proposed by Haber in 1898 [20]. However, this mechanism does not fully
explain all the experimental results. Consequently, more studies were done, and another
mechanism was proposed by Wisniak and Klein [21] that is slightly more complicated than
the Haber’s one. Some years later, Gelder et al. [22] suggested that the number of steps
involved in ANL formation was higher and substantially different from those previously
reported.
There is more information available for the mechanism of ANL production than for the
formation of secondary products in the second step. Nagata et al. [23] proposed a mechanism
to explain the formation of some of these compounds (Fig. 1 a). However, this mechanism
does not include the formation of cyclohexanol (CHOL) and dicyclohexylamine (DICHA)
that are observed in experimental tests carried out by several authors and also in those
performed in the present work.
In 1995, Narayanan et al. [10] identified the following reaction products: cyclohexane
(CH), cyclohexylamine (CHA), DICHA, and N-phenylcyclohexylamine (NPCHA). They
observed that in the case of supported nickel catalysts, depending upon the conditions of the
experiment and catalyst metal content, products such as CHA and DICHA are formed. In a
latter article, Narayanan and Unnikrishnan [24] compared the catalytic properties of Co/
Al2O3 and Ni/Al2O3 materials. It was con- cluded that by increasing the contact time of ANL,
the conversion increased and CHA and NPCHA were the two major products. CHA
formation was slightly favored at short contact time and NPCHA was formed in roughly
similar amounts at all contact times. However, under long con- tact times, DICHA (DCHA
in their notation) and CH were also formed. Based on these results they proposed the
mechanism for the formation of secondary products as indicated in Fig. 1 b.
In 2008, Relvas [25], using a Ni catalyst, proposed another mechanism based on the one of
Nagata et al. [23]. For this proposal, several laboratorial tests were made to study operating
condition effects. DICHA and CHOL (through cyclohexanone (CHONA) hydrogenation)
were included in the proposed mechanism (Fig. 1 c) since they were detected in the
experimental tests, as well as N-cyclohexylaniline (CHANIL) and cyclohexyldeneaniline
(CHENO). Nevertheless, as recognized by Relvas [25], this mechanism does not fully explain
the formation of all secondary products in NB hydrogenation. So, further work in this topic is
still required.
Summarizing, the information available in the literature about the overall reaction
mechanism involved in the NB hydrogenation is not sufficient. It is important to mention
that the knowledge and understanding of this mechanism has implications for both catalyst
and reactor design. Consequently, the present study has the objective of trying to determine
all the compounds obtained from this reaction, and also the overall mechanism that may
explain both the formation of ANL and secondary products during NB hydrogenation. The
experiments were performed using a Pd-based catalyst and under different experimental

2
conditions.

2 Materials and Methods

Hydrogenation of NB in liquid phase was carried out in a 1-L Parr batch reactor provided
with an air-impelled stirrer to pro- mote appropriate mixing and gas distribution during
the experiments. The stirrer was equipped with a basket, where the catalyst was placed. The
catalyst used was a commercial 1 wt % Pd/Al2O3 material in spherical form with a diameter
of 2–4 mm.
A known amount of Pd/Al2O3 catalyst was loaded into the reactor. The material was
pretreated in situ as described in the Supporting Information. NB was charged in a vessel
connected to the reactor and subsequently pushed into the reactor by using a high-pressure
hydrogen stream; cf. Fig. S1 in the Sup- porting Information. The procedures used in the
catalytic studies and in the analyses along the reaction time were the same as described by
Relvas et al. [2] and are summarized in the Supporting Information. The reference values for
tempera- ture, pressure, and NB concentration during the parametric study are T = 150 ºC,
P = 14 barg, and Cref = 10 wt% NB (100 000 ppm), respectively. The experiments performed
and the conditions used are given in Tab. 2. Test B12 was carried out only with ANL in order
to study the formation of secondary products. The catalyst-to-liquid volume ratio was in all
cases 80.0 g L–1.
The concentration of all compounds is presented as a relative dimensionless value that was
calculated as follows:

where Ci is the dimensionless concentration of component i, Ci,t is the concentration of

component i at time t, and Cref stands for the NB reference concentration.

3 Results and Discussion

In order to better understand the mechanism for the formation of aniline and secondary
products in the reaction of NB hydrogenation, experiments were done in batch mode where
the influence of NB initial concentration, temperature, and pressure was studied in detail.
The liquid phase analysis con- firmed the presence of the following products: NB, ANL,
CHA, CHOL, CHONA, CHANIL, DICHA, CHENO, and benzene (Bz); cf. nomenclature
list at the end of the paper. Relatively to the carbon mass balance, all the compounds
identified allowed to close the balance within the analytical uncertainty (± 6 %).
Some tests were performed under the same operating conditions, namely, 10 wt % NB, 14 barg,
and 150 ºC, which allowed for confirming the reproducibility of the experiments (data not
shown). It was also concluded that the catalyst is stable during the runs made and within the
conditions studied.
A comparison between tests B4 and B12 as indicated in Fig. S2 of the Supporting
Information, clearly confirms that the formation of secondary products, i.e., sum of all species
other than ANL, only begins to occur in a considerable extension after the significant

3
decrease of NB concentration in the reaction mixture, ca. 90–100 min, although both NB and
ANL were initially loaded into the reactor (run B4). On the other hand, if there is initially only
ANL (run B12) in the reactor, the formation of secondary products starts from the beginning
of the reaction. So, when the NB is also present in the reactor, this formation will be delayed
and only happens when the NB con- centration becomes very small, which means that this
catalyst is very selective for NB hydrogenation.

3.1 Influence of Initial Nitrobenzene Concentration

Tests B4 (10 wt % NB), B7 (30 wt % NB), and B11 (3 wt % NB) were realized under the same
operating conditions, 150 ºC and 14 barg, and only the NB concentration was varied, cf. Tab.
2 and Fig. 2. Comparing the three tests, it is observed especially in the case of CHONA (Fig. 2
d) that in the beginning of the re- action the concentration is higher when the NB concentration
is 30 wt %, unlike what is observed for the other compounds where the concentration is very
similar even for high NB concentrations.
According to the mechanism illustrated in Fig. 1 c, CHA, CHANIL, and DICHA are formed
from ANL hydrogenation, which was also observed by Nagata et al. [23], although they did
not identify the formation of DICHA. The results obtained are thus in agreement with Relvas’s
mechanism; for short reaction times, their concentration is small but when the NB con-
centration starts to decrease along each experiment, which means an increase of ANL
concentration and high NB conversion (see also Fig. 1 c), their concentration also becomes
higher; this is particularly noticeable for CHA and CHANIL, as demonstrated in Fig. 2
because DICHA is a final hydrogenation product. CHENO shows a similar trend but its
formation is anticipated by another route; cf. Fig. 1 c.
As indicated in Fig. 3 a, at the beginning of the reaction the ANL concentration decreases due
to the entrance of NB in the reactor. Then, NB starts to convert mainly into ANL and con-
sequently its concentration increases steadily, the hydrogenation of NB being the preferential
reaction, thus confirming the selectivity of the catalyst used in this reaction. In fact, ANL
hydrogenation is verified mostly when the NB concentration inside the reactor is quite low.
Although this is not very clear in these tests because the runs were stopped before the
hydrogenation of ANL becomes the preferential reaction, in other tests a subsequent decrease
of ANL concentration is clearly seen, e.g., in Fig. 8 e, at higher temperatures.
Narayanan and Unnikrishnan [24] observed under vapor phase conditions the formation of
cyclohexane (CH) and ammonia (NH3) through hydrogenolysis of CHA at 250 ºC. In the
current case, CHA conversion was not observed in some additional studies that were done (data
not shown for brevity), and CH was not detected. This could mean that CH and DICHA are
not formed from CHA conversion as it was proposed by those authors, but from ANL
hydrogenation as it is proposed by Nagata et al. [23] and Relvas [25]. Nevertheless, Bz was
herein identified and quantified. To the authors’ knowledge, the quantification of Bz obtained
as a by-product is not referred nor presented in any other previous studies. The presence of
Bz and also of NH3, which was detected during sample collection, could be related to the
hydrodenitrogenation of ANL.
Fig. 4 demonstrates the time evolution of the total of secondary products formed, i.e.,

4
excluding ANL, for the three runs as well as the NB concentration. Analyzing this figure, it is
observed that the total of secondary products exhibits in general a greater increase when the NB
concentration starts decreasing significantly, and so it can be concluded that their formation is
mostly due to the ANL hydrogenation reaction. This can be also related with the adsorption
affinity because NB adsorbs much easier on the catalyst surface than ANL. Moreover, it is seen
that in general the higher the initial concentration of NB, the higher is the formation of secondary
products, although the maximum in their concentration is shifted towards longer reaction times.
The reason is that a longer reaction time is required to reduce the NB to a low level for
increasingly initial loads of this reactant in the reactor.
In the results of NB consumption, illustrated in Fig. 4 and as will be shown below, it is
noteworthy that the maximum expectable value of its concentration is never reached because
the compound is not inside the reactor at the initial instant as de- tailed in the experimental/
Supporting Information section. Consequently, when it is loaded into the reactor, there is
immediately a consumption of NB and so the maximum expectable concentration is not
achieved, e.g., CNB/Cref of 3 for run B7 is not reached, neither CNB/Cref of 1 for run B4, etc.
Fig. 5 presents the same results as the ones in Fig. 2, but in this case the species
concentrations are represented as a function of the dimensionless NB concentration. Thus, it is
possible to confirm and better understand the discussion above and also to compare the results
at the same level of NB concentration. The results should be analyzed taking into account
that the evolution of the reaction time is done from the right to the left, which corresponds to
the decrease of NB concentration. Consequently, it is observed that the secondary products
are formed in greater quantities when the NB concentration is very low. Because ANL is also
initially loaded into the reactor (see Supporting Information), this means that the Pd/Al2O3
catalyst is selective towards NB hydrogenation, but the undesired reactions become more
significant when the NB concentration is very low. Moreover, it is noted that some of the
compounds are already present in the reactor at the start of the reaction, which is due to the fact
that the ANL used, namely, industrial-grade ANL, contains already traces of secondary
products. It is finally noted that, in general, the amount of secondary products formed is
increased for higher initial NB concentrations: run B7 30 wt % NB > run B4 10 wt % NB > run
B11 3 wt % NB.

3.2 Influence of Pressure

The tests in this section were performed at the same temperature, 150 ºC, and with the same
initial concentration of nitro- benzene, 10 wt % NB. The total pressure in the reactor was in-
creased from 14 barg (run B4) by 6 barg (20 barg in run B2) and 16 barg (30 barg in run B3).
Pressure variation is due to the increase of hydrogen pressure; so there will be more hydrogen
in the gas phase and also solubilized in the reaction mixture. Fig. 6 presents the results obtained
as a function of the reaction time. It is noticed that Bz formation is almost independent of the
total pressure (Fig. 6 a). For the other secondary products, it seems that for higher pressures
higher quantities of these compounds are obtained, although this effect varies from product to
product. Even so, the effect of pressure on the total formation of secondary products is not
very pronounced (Fig. 7 b).

5
 Since these by-products are mostly formed after ANL formation/NB complete consumption
(Fig. 7), it is concluded that the results obtained are in agreement with Relvas’ [25] mechanism.
Moreover, at the same reaction time and when the pressure is higher, the NB concentration is
lower and, consequently, the ANL concentration rises. Therefore, ANL hydrogenation will be
favored and the concentration of the secondary products will increase sooner, as indicated in
Fig. 7. Although for higher pressures it seems that more secondary products are formed, the
effect of an increase in pressure from 20 to 30 barg is less significant.
The influence of pressure is also visible in the reaction rate. When the total pressure in the
reactor rises, there is also an increase in the reaction rate and consequently, NB is consumed
in a shorter period of time and the ANL concentration will increase more rapidly.
Fig. S3 of the Supporting Information presents the same data but as a function of the
dimensionless NB concentration.

3.3 Influence of Temperature

The influence of temperature – runs B4 (150 ºC), B9 (180 ºC), B10 (210 ºC), and B5 (240 ºC)
– was also studied at the same pressure, 14 barg, and initial NB concentration, 10 wt %. It was
concluded that temperature is the parameter with greater influence in the production of
secondary products, as shown below (Fig. 8 and Fig. S4 of the Supporting Information).
Fig. 8 presents the results obtained in these experiments, showing the concentration of by-
products and ANL as a func- tion of the reaction time. Concentration profiles show the same
trend for all the tests; indeed, it is possible to see that for higher temperatures the
concentration of secondary products increases in the order 240 ºC > 210 ºC > 180 ºC >
150 ºC.
Bz belongs to the compounds that seem to be more influenced by temperature variation.
In the temperature range of 150–210 ºC, the concentration profiles obtained are similar to those
shown above. While at lower temperatures Bz is detected only in small amounts, at 240 ºC, when
the dimensionless NB concentration is < 0.012 (after ca. 1 h), its concentration starts to increase
considerably (Fig. 8 a) which had not been observed before. Some tests of ANL hydrogenation,
which were per- formed at different operational conditions (data not shown), confirmed the
Bz formation from ANL and that its concentration is higher when the conditions are more
severe, namely the temperature; see also Fig. S4 of the Supporting Information.
In the case of CHA, DICHA, and CHANIL, it is shown that when the temperature rises, there
is an increase in their con- centration when the NB concentration is low (high conversion).
As it was observed in the tests with pressure variation, in the first hour of reaction the
concentration profile is described by a plateau. This plateau corresponds to the period when NB
is present in high quantities and when the conversion of NB in- to ANL mainly takes place. After
that period, although the NB conversion still proceeds, ANL hydrogenation becomes more
pronounced and, consequently, the concentration of secondary products increases. This
reaction, ANL hydrogenation, starts to be more important with higher temperature.
Both CHONA and CHENO generally present concentration profiles similar to the compounds
discussed above, but at the end of the reaction their concentration seems to start to stabilize,

6
reaching a plateau as indicated in Figs. 8 d and 8 g. For CHOL, similar profiles are
anticipated; however, longer reaction times are probably required to observe such a pattern.
In relation to NB, there is, effectively, an important growth of the rate of the hydrogenation
reaction with the temperature increase, which means that for higher temperatures the same
NB conversion is achieved in much less time, as shown in Fig. 9.
Once more it is noticed that the concentration of secondary products is strongly dependent on
the temperature of the reaction (Fig. 9 b). Indeed, the formation of secondary products is
tremendously accelerated as a consequence of the exponential temperature effect in the kinetics
according to the Arrhenius law. The different activation energies for each reaction in the
complex reaction scheme also explain the various effects of temperature in each product
formed. For those steps with higher activation energies, reaction rates are more favored for the
same increase in the temperature.
Narayanan and Unnikrishnan [24] also studied the influence of temperature on a Co/Al2O3
and a Ni/Al2O3 catalyst and concluded that with higher temperature the ANL conversion
increases over Co/Al2O3 but over Ni/Al2O3 the increase in temperature reduced the
conversion, due to the decrease of the hydrogen adsorption. Besides, they also observed that
above a certain temperature, hydrodenitrogenation is favored leading to the formation of CH.
In the current case, instead of CH formation, the formation of Bz was observed and the same
tendency was noticed, namely, a higher temperature favors the Bz formation; see Fig. 8 a.
Relatively to the catalyst herein used, its behavior was similar to the Co/Al2O3 one; with higher
temperatures the ANL conversion increased.
Analyzing all the results obtained, it can be stated that the obtained data are in general
agreement with the results described by Relvas [25], and all compounds mentioned were
identified in our experiments. Furthermore, even using a different type of catalyst, Pd in this
work vs. Ni in Relvas’ work, and different experimental conditions, the mechanism was found to
be very similar. Another important issue which is worth mentioning in the results herein
presented, is the formation of Bz, which is not considered by Relvas nor by any other authors
in the literature, except by Králik et al. [26]; however, the authors did not quantify Bz.
Consequently, and reflecting the results obtained, the Bz formation should be considered in the
mechanism through the reaction of ANL hydrodenitrogenation, as detailed in Fig. 10.

4 Conclusions
The hydrogenation of NB to ANL over a Pd catalyst was studied. A parametric study was
performed in which the effects of the initial NB concentration, temperature, and pressure were
tested. It was found that temperature is the parameter with greater influence in the
production of secondary products, for the ranges analyzed.
It was also observed that the formation of secondary products is strongly dependent on the
concentration of NB in the reaction mixture: their concentration rises more when the NB
concentration starts decreasing significantly, or becomes almost negligible, and consequently it
can be concluded that the formation of secondary compounds is mostly due to the ANL
hydrogenation. This highlights the high selectivity of the used Pd-based catalyst.
Some mechanisms were already proposed in the literature; nevertheless, the one proposed by
Relvas, using a Ni/SiO2 catalyst, is the one that better describes the majority of the experimental
results herein obtained for both ANL and formation of secondary products, despite the fact that

7
a Pd-based catalyst was now used. Furthermore, the formation of Bz was also identified, which
was not considered in a quantitative manner by any previous authors in the literature.

Acknowledgment
C. Sá Couto thanks the Fundação para a Ciência e Tecnologia (FCT) for the Ph. D. Grant (ref.
SFRH/BDE/51794/2012) and CUF – Químicos Industriais, S. A., for the support provided
for experimental work.

The authors have declared no conflict of interest.

Abbreviations
ANL aniline
Bz benzene
CH cyclohexane
CHA cyclohexylamine
CHENO cyclohexyldeneaniline
CHOL cyclohexanol
CHONA cyclohexanone
DICHA / DCHA dicyclohexylamine
MDI methylenediphenyldiisocyanate
NB nitrobenzene
NPCHA N-phenylcyclohexylamine

References

[1] R. Lawrence, W. J. Marshall, Aniline, in Ullmann’s Encyclopedia of Industrial

Chemistry, 6th ed., John Wiley & Sons, New York 1998.
[2] J. Relvas, R. Andrade, F. Gama Freire, F. Lemos, P. Araújo, M. Pinho, C. Pedro
Nunes, F. Ramoˆa Ribeiro, Catal. Today 2008, 163, 133–135. DOI:
10.1016/j.cattod.2007.11.050
[3] H. D. Burge, D. J. Collins, B. H. Davis, Ind. Eng. Chem. Prod. Res. Dev. 1980,
19, 389–391. DOI: 10.1021/i360075a019
[4] D. J. Collins, A. D. Smith, B. H. Davis, Ind. Eng. Chem. Prod. Res. Dev. 1982,
21, 279–281. DOI: 10.1021/i300006a016
[5] F. Figueras, B. Coq, J. Mol. Catal. A: Chem. 2001, 173, 223– 230. DOI:
0.1016/S1381-1169(01)00151-0
[6] X. Yu, M. Wang, H. Li, Appl. Catal., A 2002, 202, 17–22. DOI: 10.1016/S0926-
860X(00)00454-3
[7] E. Klemm, B. Amon, H. Redlingsho¨fer, E. Dieterich, G. Emig, Chem. Eng. Sci.
2001, 56, 1347–1353. DOI:10.1016/S0009-2509(00)00357-2
[8] R. Downing, P. Kunkeler, H. van Bekkum, Catal. Today 1997, 37, 121–136. DOI:
10.1016/S0920-5861(97)00005-9
[9] D. Rihani, T. Narayanan, L. Doraiswamy, Ind. Eng. Chem. Prod. Res. Dev. 1965,
4, 403–410. DOI: 10.1021/i260016a012

8
 [10] S. Narayanan, R. Unnikrishnan, V. Vishwanathan, Appl. Catal., A 1995, 129, 9–
19. DOI: 10.1016/0926-860X(95)00087-9
[11] V. Mohan, C. V. Pramod, M. Suresh, K. Hari Prasad Reddy, B. David Raju, K. S.
Rama Rao, Catal. Commun. 2012, 18, 89–92. DOI: 10.1016/j.catcom.2011.11.030
[12] P. Sangeetha, P. Seetharamulu, K. Shanthi, S. Narayanan, K. S. Rama Rao, J.
Mol. Catal. A: Chem. 2007, 273, 244–249. DOI: 10.1016/j.molcata.2007.03.020
[13] P. Sangeetha, K. Shanthi, K. S. Rama Rao, B. Viswanathan, P. Selvam, Appl.
Catal., A 2009, 353, 160–165. DOI: 10.1016/j.apcata.2008.10.044
[14] C. Li, Z. Yu, K. Yao, S. Ji, J. Liang, J. Mol. Catal. A: Chem. 2005, 226, 101–105.
DOI: 10.1016/j.molcata.2004.09.046
[15] V. Ho¨ller, D. Wegricht, I. Yuranov, L. Kiwi-Minsker, A. Renken, Chem. Eng.
Technol. 2000, 23, 251–255. DOI: 10.1002/(SICI)1521-4125(200003)23:3<251::AID-
CEAT251>3.0.CO;2-S
[16] F. A. Harraz, S. E. El-Hout, H. M. Killa, I. A. Ibrahim, J. Catal.2012, 286, 184–
192. DOI: 10.1016/j.jcat.2011.11.001
[17] N. Bouchenafa-Saı¨b, P. Grange, P. Verhasselt, F. Addoun, V. SNM>Dubois,
Appl. Catal., A 2005, 286, 167–174. DOI:10.1016/j.apcata.2005.02.022
[18] E. Gelder, S. Jackson, C. Lok, Catal. Lett. 2002, 84, 205–208.DOI:
10.1023/A:1021432104496
[19] C. Mitchell, D. Stewart, WO Patent 113491, 2011.
[20] F. Haber, Z. Elektrochem. 1898, 4, 506–514. DOI: 10.1002/bbpc.18980042204
[21] J. Wisniak, M. Klein, Ind. Eng. Chem. Prod. Res. Dev. 1984,23, 44–50. DOI:
10.1021/i300013a009
[22] E. Gelder, S. Jackson, C. Lok, Chem. Commun. 2005, 522–524. DOI:
10.1039/B411603H
[23] T. Nagata, K. Watanabe, Y. Kono, A. Tamaki, T. Kobayashi,US Patent 5283365,
1994.
[24] S. Narayanan, R. Unnikrishnan, J. Chem. Soc., FaradayTrans. 1997, 93, 2009–
2013. DOI: 10.1039/A608074J
[25] J. Relvas, Ph. D. Thesis, Technical University of Lisbon 2008.
[26] M. Kra´lik, M. Tura´kova, I. Maca´k, S. Wenchich, J. Chem. Chem. Eng. 2012, 6,
1074–1082.

9
 Figure 1. Reaction network for the formation of ANL and secondary products as pro- posed by
(a) Nagata et al. [23]; (b) Narayanan and Unnikrishnan [24]; (c) Relvas [25].
*Very reactive and unstable compound

10
 Figure 2. Influence of initial nitrobenzene concentration in the formation of secondary
products and NB consumption vs. time; runs B4, B7, and B11.

11
 Figure 3. Influence of initial NB concentration in the (a) ANL formation and (b) NB conversion
vs. time; runs B4, B7 and B11.

Figure 4. Comparison between total formation of secondary products (closed symbols) and NB
consumption (open symbols) as a function of reaction time for different initial NB
concentrations; runs B4, B7, and B11.

12
 Figure 5. Influence of NB initial concentration in the formation of secondary products; runs B4,
B7, and B11.

13
 Figure 6. Influence of reaction pressure in the formation of secondary products and ANL vs.
time; runs B2, B3, and B4.

14
 Figure 7. Comparison between (a) ANL formation and (b) total formation of secondary
products (closed symbols) and NB consumption (open symbols) as a function of reaction time;
runs B2, B4, and B5.

15
 Figure 8. Influence of reaction temperature in the formation of ANL and by-products vs.
reaction time; runs B4, B5, B9, and B10.

Figure 9. Comparison between( a) NB conversion and (b) total formation of secondary products
(closed symbols) and NB consumption (open symbols) as a function of reaction time for different
reaction temperatures; runs B4, B5, B9, and B10.

16
 Figure 10. Reaction network proposed for formation of ANL and and secondary products
including Bz. *Very reactive and unstable compound.

Table 1. Main catalysts studied for NB hydrogenation.

17
Table 2. Initial conditions of the experiments performed.

18