Amino-Modiﬁed Silica as E ﬀ ective Support of the Palladium Catalyst for 4-Nitroaniline Hydrogenation

: The article describes the synthesis of aminoorgano-functionalized silica as a prospective material for catalysis application. The amino groups have electron donor properties which are valuable for the metal chemical state of palladium. Therefore, the presence of electron donor groups is important for increasing catalysts’ stability. The research is devoted to the investigation of silica amino-modified support influence on the activity and stability of palladium species in 4-nitroaniline hydrogenation process. A series of catalysts with different supports such as SiO 2 , SiO 2 -C 3 H 6 -NH 2 (amino-functionalized silica), γ -Al 2 O 3 and activated carbon were studied. The catalytic activity was studied in the hydrogenation of 4-nitroaniline to 1,4-phenylenediamine. The catalysts were characterized by scanning electron microscopy, transmission electron microscopy, X-ray photoelectron spectroscopy, Fourier transform infrared spectroscopy and chemisorption of hydrogen by the pulse technique. The5wt.%Pd / SiO 2 -C 3 H 6 -NH 2 catalystexhibitedthehighestcatalyticactivityfor4-nitroaniline hydrogenation with 100% conversion and 99% selectivity with respect to 1,4-phenylenediamine. observed. solution was stirred for 24 h at room temperature.

Possible ways to increase catalytic activity and selectivity include the development of catalysts synthesis methods, the optimization of reaction conditions, catalyst modification by different chemical elements and compounds, the application of supports and modification of the catalyst surface [4,[11][12][13][14][16][17][18][19][20]. Modification of catalyst support surface is one possible solution to increase catalysts' activity and stability. The presence of functional groups on the support surface has great influence on the electronic state of active metal and metal dispersion [21]. The aminopropyl groups provide a unique possibility for further surface modification. Amino groups' modified supports are very useful for further surface nucleophilic substitution and obtaining the highest metal dispersion and study of this effect is a sufficiently novel trend in catalysis [4,19,20,[22][23][24][25]. A lot of articles are dedicated to the application of amino-functionalized mesoporous silica in medicine, pharmaceutics and smart materials, however, publications on their influence on active metals' catalytic properties and smart materials, however, publications on their influence on active metals' catalytic properties are fragmental [26][27][28][29][30][31]. The present study reports the synthesis of aminoorgano-functionalized silica immobilized by palladium nanoparticles using the sol-gel method. This allows the synthesization of various hybrid composition materials by the one-pot method. For comparison, amorphous silica, activated carbon and gamma-alumina were used as supports for palladium particles.
The catalytic activity of synthesized catalysts was tested in 4-nitroaniline hydrogenation ( Figure  1). Liquid phase hydrogenation of substituted nitrobenzenes is an essential technology for the production of various aromatic amines-key intermediates for manufacturing agrochemicals, isocyanates, pharmaceuticals and dyes [32][33][34][35]. The most commonly used catalysts for different hydrogenation processes contain platinum group metals (palladium, platinum, iridium, etc.) attached to different supports [33][34][35][36][37][38]. Palladium nanoparticles were found to be the most active and have a high potential for the catalytic application of essential chemical products [33][34][35]. However, the catalysts' stability is an open issue for such types of catalyst, therefore the development of active and stable catalysts is the focus of the current article. Compounds such as 4-nitroaniline are used as an industrial raw material to produce agricultural chemicals, rubber compounding agents, synthetic resin additives, polyamides, pharmaceuticals and dyes. The conductive polyamines can be used in electronics [39], and also as antioxidants and preservatives [40][41][42]. The polyamines can find a wide application in films, materials and membranes due to their biodegradable and electro-conductive properties.

FTIR Spectroscopy of Silica Supports
Silica modification with aminopropyl groups with an NH2 content of 30 wt.% was verified by Fourier transform infrared spectroscopy ( Figure 2). The analysis of the FTIR spectra of modified and pure supports revealed the broad bands centered around 3441 cm −1 and correspond to O-H groups' vibrations. On the other hand, the spectrum of amino-functionalized silic peaks at 2940 and 2886 cm −1 and is assigned to the C-H stretching vibrations of CH-and -CH2-groups, which can be attributed to the incorporation of the amino group [22,43]. The peak of the C-N stretching vibration at 1140 cm −1 overlaps with the Si-O-Si stretching band in the range 1000-1200 cm −1 . The peak of the N-H stretching vibration at ~ 3295 and 3358 cm −1 overlaps with a wide peak of stretching vibration at O-H at 3080 cm −1 [22,43]  Compounds such as 4-nitroaniline are used as an industrial raw material to produce agricultural chemicals, rubber compounding agents, synthetic resin additives, polyamides, pharmaceuticals and dyes. The conductive polyamines can be used in electronics [39], and also as antioxidants and preservatives [40][41][42]. The polyamines can find a wide application in films, materials and membranes due to their biodegradable and electro-conductive properties.

FTIR Spectroscopy of Silica Supports
Silica modification with aminopropyl groups with an NH 2 content of 30 wt.% was verified by Fourier transform infrared spectroscopy ( Figure 2). The analysis of the FTIR spectra of modified and pure supports revealed the broad bands centered around 3441 cm −1 and correspond to O-H groups' vibrations. On the other hand, the spectrum of amino-functionalized silic peaks at 2940 and 2886 cm −1 and is assigned to the C-H stretching vibrations of CH-and -CH 2 -groups, which can be attributed to the incorporation of the amino group [22,43]. The peak of the C-N stretching vibration at 1140 cm −1 overlaps with the Si-O-Si stretching band in the range 1000-1200 cm −1 . The peak of the N-H stretching vibration at 3295 and 3358 cm −1 overlaps with a wide peak of stretching vibration at O-H at 3080 cm −1 [22,43]. The intense bands appearing at 1059 and 1176 cm −1 were assigned to the asymmetric stretching The band observed at 467 cm −1 is assigned to the plane-stretching vibrations of Si-C [43].  However, the use of tetraethoxysilane as a silica precursor led to the formation of surface methylfunctionalized silica. This was certified by a broad peak in the range 1338-1500 cm −1 which can be assigned to the C-H deformation vibrations of aliphatic bands [43].

The Pulse Chemisorption
The hydrogen pulse chemisorption data are presented in Table 1. Silica surface modification by tetraethoxysilane results in some increase in metal dispersion, and a concentration of active sites and a metallic surface area. However, metal dispersion remains considerably low.

XPS of Catalysts before the Reaction
The surface chemical composition of silica supports determined by X-ray photoelectron spectroscopy. The high-resolution XPS spectra of C 1s, N 1s, Si 2p are shown in the appropriate figure ( Figure 3). According to XPS data, the silica contains O, C, Si elements. The synthesized aminoorganomodified silica contains O, C, Si and N elements.
The presence of quantities of aliphatic carbon compounds in silica is a result of the inclusion of an aminopropyl substituent in the silica structure, and the incomplete hydrolysis of silane and carbon pollution. The C-N and C-Si bonds prove the inclusion of aminopropyl substituents in the silicon oxide matrix. According to the published data [43,44], the peaks at 399.5 and 401.6 eV in the spectrum of N 1s can be attributed to C-N and N-H bonds that correlate with the IR-spectroscopy data ( Figure  2) [43,44]. The bonds -C-ONa belong to molecules of sodium carbonate ( Figure 3) [43,44]. However, the use of tetraethoxysilane as a silica precursor led to the formation of surface methyl-functionalized silica. This was certified by a broad peak in the range 1338-1500 cm −1 which can be assigned to the C-H deformation vibrations of aliphatic bands [43].

The Pulse Chemisorption
The hydrogen pulse chemisorption data are presented in Table 1. Silica surface modification by tetraethoxysilane results in some increase in metal dispersion, and a concentration of active sites and a metallic surface area. However, metal dispersion remains considerably low.

XPS of Catalysts before the Reaction
The surface chemical composition of silica supports determined by X-ray photoelectron spectroscopy. The high-resolution XPS spectra of C 1s, N 1s, Si 2p are shown in the appropriate figure ( Figure 3). According to XPS data, the silica contains O, C, Si elements. The synthesized amino-organomodified silica contains O, C, Si and N elements.  Figure 3d shows the model decomposition of the XPS high-resolution spectra of Pd 3d energy core-level of 5% wt. Pd/SiO2-C3H6-(30%)NH2 catalyst before the reaction. The XPS data of the other catalysts were published in reference 52 of this paper. The atomic concentration of palladium compounds is given in Table 2. The model decomposition of Pd 3d spectra for 5 wt.% Pd/γ-Al2O3, 5 wt.% Pd/SiO2, and 5 wt.% Pd/SiO2-C3H6-(30%)NH2 catalysts showed the presence of two palladium chemical states (Table 2, Figure 3d) [45]. Different states of palladium were observed in the catalyst of activated carbon, particularly metallic palladium Pd 0 , oxidized PdO on metallic Pd 0 and palladium oxides PdO, PdO2 [45]. PdO/Pd 0 0.15 An interesting observation was a shift in the binding energy by ~0.5 eV of palladium metal in the 5 wt.% Pd SiO2-C3H6-(30%)NH2 catalyst in comparison to the standard binding energy of metal palladium (335 eV). There was an assumption that this may be the result of the displacement of the electron cloud of amino groups to palladium atoms. This effect can take place in case of Pd nanoparticles' coordination by amino groups. A comparatively similar interaction between palladium nanoparticles and nitrogen-containing supports was reported in several articles devoted to palladium nanostructured materials synthesis and characterization [24,46,47]. Typically, a Pd charging by amino groups has a positive influence on hydrogen chemisorption on active metal that can have a positive effect on the hydrogenation of the substrate [24,46,47].  The presence of quantities of aliphatic carbon compounds in silica is a result of the inclusion of an aminopropyl substituent in the silica structure, and the incomplete hydrolysis of silane and carbon pollution. The C-N and C-Si bonds prove the inclusion of aminopropyl substituents in the silicon oxide matrix. According to the published data [43,44], the peaks at 399.5 and 401.6 eV in the spectrum of N 1s can be attributed to C-N and N-H bonds that correlate with the IR-spectroscopy data ( Figure 2) [43,44]. The bonds -C-ONa belong to molecules of sodium carbonate ( Figure 3) [43,44]. Figure 3d shows the model decomposition of the XPS high-resolution spectra of Pd 3d energy core-level of 5% wt. Pd/SiO 2 -C 3 H 6 -(30%)NH 2 catalyst before the reaction. The XPS data of the other catalysts were published in reference 52 of this paper. The atomic concentration of palladium compounds is given in Table 2. The model decomposition of Pd 3d spectra for 5 wt.% Pd/γ-Al 2 O 3 , 5 wt.% Pd/SiO 2 , and 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 catalysts showed the presence of two palladium chemical states (Table 2, Figure 3d) [45]. Different states of palladium were observed in the catalyst of activated carbon, particularly metallic palladium Pd 0 , oxidized PdO on metallic Pd 0 and palladium oxides PdO, PdO 2 [45]. An interesting observation was a shift in the binding energy by~0.5 eV of palladium metal in the 5 wt.% Pd SiO 2 -C 3 H 6 -(30%)NH 2 catalyst in comparison to the standard binding energy of metal palladium (335 eV). There was an assumption that this may be the result of the displacement of the electron cloud of amino groups to palladium atoms. This effect can take place in case of Pd nanoparticles' coordination by amino groups. A comparatively similar interaction between palladium nanoparticles and nitrogen-containing supports was reported in several articles devoted to palladium nanostructured materials synthesis and characterization [24,46,47]. Typically, a Pd charging by amino groups has a positive influence on hydrogen chemisorption on active metal that can have a positive effect on the hydrogenation of the substrate [24,46,47].

XPS of Used Catalysts
The atomic surface concentrations of palladium and palladium oxide decrease in 5 wt.% Pd/γ-Al 2 O 3 , 5 wt.% Pd/SiO 2 and 5 wt.% Pd/C catalysts (Tables 2 and 3) which can be attributed to metal particles' diffusion into a matrix of support [45]. However, for amino-modified sample 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 , a change in active metal surface concentration was not observed. The quantity of zero valence metal in 5 wt.% Pd/SiO 2 and 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 samples before and after reaction did not change significantly ( Table 3). The changes in Pd 0 surface concentrations in 5 wt.% Pd/C and Pd/γ-Al 2 O 3 catalysts were the most noticeable [45]. In the case of the 5 wt.% Pd/C catalyst, the quantity of PdO decreased on 0.7 at.% and the content of PdO 2 was increased on 0.7 at.% (Tables 2 and 3). The contents of PdO and Pd 0 in 5 wt.% Pd/SiO 2 and 5 wt.% Pd/γ-Al 2 O 3 catalysts were decreased.

SEM
The morphology of the amino-functionalized mesoporous silica was studied by scanning electron microscopy ( Figure 4). The synthesized silica particles have a spherical shape. Average particle size diameter was found to be 1 µm before and after catalysts' synthesis.

SEM
The morphology of the amino-functionalized mesoporous silica was studied by scanning electron microscopy ( Figure 4). The synthesized silica particles have a spherical shape. Average particle size diameter was found to be 1 µm before and after catalysts' synthesis.

TEM
The images of transmission electron microscope were obtained for the reliable determination of palladium particle size ( Figure 5). The images show that the particles are quite different in size.

TEM
The images of transmission electron microscope were obtained for the reliable determination of palladium particle size ( Figure 5). The images show that the particles are quite different in size.  The particle size in the 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 catalyst varies from 3 to 10 nm; in the 5 wt.% Pd/SiO 2 catalyst, particle size varies from 2 to 10 nm; in the 5 wt.% Pd/γ-Al 2 O 3 catalyst, particle size varies from 5 to 70 nm; and in the 5 wt.% Pd/C catalyst, particle size varies from 30 to 100 nm. In general, the data agree with the result of hydrogen pulse chemisorption.

TPD of Matrices
According to the results of the TPD NH 3 study, the modification of silica by amino groups increased the amount of acid sites compared with silica ( Figure 6).
Catalysts 2020, 10, x FOR PEER REVIEW 7 of 15 The particle size in the 5 wt.% Pd/SiO2-C3H6-(30%)NH2 catalyst varies from 3 to 10 nm; in the 5 wt.% Pd/SiO2 catalyst, particle size varies from 2 to 10 nm; in the 5 wt.% Pd/γ-Al2O3 catalyst, particle size varies from 5 to 70 nm; and in the 5 wt.% Pd/C catalyst, particle size varies from 30 to 100 nm. In general, the data agree with the result of hydrogen pulse chemisorption.

TPD of Matrices
According to the results of the TPD NH3 study, the modification of silica by amino groups increased the amount of acid sites compared with silica ( Figure 6). The alumina contains less acidic properties than amino-modified silica. Thus, according to the quantity and acid sites' strength, the catalyst supports can be given in the following order: SiO2-C3H6-NH2 > γ-Al2O3 > SiO2.

The Activity of Synthesized Catalysts
In all cycles for all catalysts, the conversion of 4-nitroaniline to 1,4-phenylenediamine was 100%.  Table 4).  The alumina contains less acidic properties than amino-modified silica. Thus, according to the quantity and acid sites' strength, the catalyst supports can be given in the following order: SiO 2 -C 3 H 6 -NH 2 > γ-Al 2 O 3 > SiO 2 .

The Activity of Synthesized Catalysts
In all cycles for all catalysts, the conversion of 4-nitroaniline to 1,4-phenylenediamine was 100%. A sufficient quantity of hydrogen for completely reduce to 4-nitroaniline is 3.62 mmol. Thus, according to the data of the catalysts, activity can arrange in the order: 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2  Table 4). A drastic increase in the catalytic activity for Pd/SiO2-C3H6-(30%)NH2 sample compared to Pd/SiO2 can be subscribed to an increase in the active surface as a result of some decrease in Pd nanoparticles' average diameter. However, Pd nanoparticles' surface modification by aminogroups and a sift in electron density to Pd nanoparticles can be the main reason for the increase in Pd activity [24,46,47].
Comparison of the catalysts' activity after a catalyst recycle showed that the highest activity after five reaction cycles was observed for 5 wt.% Pd/SiO2-C3H6-(30%)NH2 (Table 4, Figure 7). The catalysts' stability can be arranged in the following order: 5 wt.% Pd/SiO2-C3H6-(30%)NH2 > 5 wt.% Pd/SiO2 > 5 wt.% Pd/C ≈ 5 wt.% Pd/γ-Al2O3. Gas chromatographic analysis made it possible to calculate the yield of 1,4-phenylenediamine in the hydrogenation reaction after each repeated addition of 4-nitroaniline (Table 5). In all samples, at the end of the reaction, in each input of 4-nitroaniline, traces of the initial compound were not detected, and the hydrogenation conversion of 4-nitroaniline was considered to be close to full conversion.  [45,52] 0.8 0.8 0.8 0.7 0.6 Analysis of 4-nitroaniline conversion rate and the formation of 1,4-phenylenediamine showed a similar tendency to decrease in the values of the rates, as for the consumption of hydrogen ( Table 4).
The second catalyst with a lower content of amino groups was obtained to confirm the effect of the number of amino groups on the surface.  A drastic increase in the catalytic activity for Pd/SiO 2 -C 3 H 6 -(30%)NH 2 sample compared to Pd/SiO 2 can be subscribed to an increase in the active surface as a result of some decrease in Pd nanoparticles' average diameter. However, Pd nanoparticles' surface modification by aminogroups and a sift in electron density to Pd nanoparticles can be the main reason for the increase in Pd activity [24,46,47].
Comparison of the catalysts' activity after a catalyst recycle showed that the highest activity after five reaction cycles was observed for 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 (Table 4, Figure 7). The catalysts' stability can be arranged in the following order: 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 > 5 wt.% Pd/SiO 2 > 5 wt.% Pd/C ≈ 5 wt.% Pd/γ-Al 2 O 3 . Gas chromatographic analysis made it possible to calculate the yield of 1,4-phenylenediamine in the hydrogenation reaction after each repeated addition of 4-nitroaniline (Table 5). In all samples, at the end of the reaction, in each input of 4-nitroaniline, traces of the initial compound were not detected, and the hydrogenation conversion of 4-nitroaniline was considered to be close to full conversion.
Analysis of 4-nitroaniline conversion rate and the formation of 1,4-phenylenediamine showed a similar tendency to decrease in the values of the rates, as for the consumption of hydrogen (Table 4). Table 5. Data of gas chromatographic analysis in hydrogenation of 4-nitroaniline to 1,4-phenylenediamine.
The second catalyst with a lower content of amino groups was obtained to confirm the effect of the number of amino groups on the surface.
A kinetic experiment showed that the activity of the 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 catalyst with an NH 2 content of 30 wt.% is two times higher than the 5 wt.% Pd/SiO 2 -C 3 H 6 -(10%)NH 2 catalyst with a content of 10 wt.% NH 2 groups: 10.1·10 −5 mole (H 2 )/sec and 5.5·10 −5 mole (H 2 )/sec. Kinetic curves of the hydrogen uptake of the catalytic hydrogenation of 4-nitroaniline on organically modified catalysts are shown in Figure 8. The higher the concentration of amino groups on the matrix surface was fixed, the more active the palladium deposited on the silica matrix [24,46,47].
Catalysts 2020, 10, x FOR PEER REVIEW 9 of 15 A kinetic experiment showed that the activity of the 5 wt.% Pd/SiO2-C3H6-(30%)NH2 catalyst with an NH2 content of 30 wt.% is two times higher than the 5 wt.% Pd/SiO2-C3H6-(10%)NH2 catalyst with a content of 10 wt.% NH2 groups: 10.1•10 −5 mole (H2)/sec and 5.5•10 −5 mole (H2)/sec. Kinetic curves of the hydrogen uptake of the catalytic hydrogenation of 4-nitroaniline on organically modified catalysts are shown in Figure 8. The higher the concentration of amino groups on the matrix surface was fixed, the more active the palladium deposited on the silica matrix [24,46,47].
According to gas chromatography analysis, the conversion of 4-nitroaniline and the yield of 1.4phenylenediamine also amounted to 100%.  According to gas chromatography analysis, the conversion of 4-nitroaniline and the yield of 1.4-phenylenediamine also amounted to 100%.

Synthesis of Silica
To synthesize spherical silica nanoparticles, 700 mL of distilled water was added to 500 mL of 2-propanol in a flat-bottomed flask and the solution was stirred to obtain a uniform solution. Then, 200 mL of tetraethoxysilane was added to the solution. The reaction mixture was kept under vigorous stirring for 30 min with a mixing rate of 1500 rpm. Subsequently, 50 mL of aqueous ammonia solution (60 wt.%) was added dropwise to the reaction mixture over 30 min. The mixture was stirred for 2 h with a mixing rate of 1000 rpm. Then, the precipitate was filtered, washed with distilled water and dried at 200 • C for 3 h on air.

Synthesis of Amino-Functionalized Silica
The first step of the reaction procedure was to prepare the oil/water emulsion. For this, 104 mL of distillated water and 16 mL of pure ethanol were mixed with a stirrer with a mixing rate 1500 rpm. Then, 2 mL of cyclohexane was added as a co-template. After the addition of 4.8 mL of dodecyldimetylamine N-oxyde to the solution, the formation of emulsion was observed. To achieve a higher emulsion dispersion, the mixture was treated by ultrasound for 5 min in ultrasonic bath 36-42 kHz, 200 W. In order to homogenize the reaction mixture, an extra 0.2 mL of surfactant was added. Then, 0.8 mL for SiO 2 -C 3 H 6 -(30%)NH 2 and 0.3 mL for SiO 2 -C 3 H 6 -(10%)NH 2 of aminopropyltrimetoxysilane were adjusted in the system for particles' formation. The solution was stirred for 10 min at a reaction temperature 60 • C. Then first portion of tetraethoxysilane (1 mL) was added to the mixture and solution opalescence was immediately observed. The solution was stirred for 24 h at room temperature. The final solution was a viscous colloid of white color, which was sufficiently stable. A vacuum filtration of solution with 0.2 micrometers membrane filter was used to separate the dispersed phase from the solution. The material was dried in vacuum oven at pressure of 0.4 bar and temperature of 100 • C.

Catalyst Preparation
The synthesis of Pd-based catalysts was made by the precipitation of palladium hydroxide due to hydrolysis reaction by the addition of 3 mM H 2 PdCl 4 solution in water to 1g of support (SiO 2 , γ-Al 2 O 3 or activated carbon) suspended in a 40 mL aqueous solution of 0.1 M Na 2 CO 3 and 0.1M solution of sodium dodecyl sulfate using an ultrasonic bath (Bandelin, 10P, Mannheim, Germany). The solution of H 2 PdCl 4 was prepared in 0.2 M hydrochloric acid. The suspension was mixed for 3 h at 35 • C in an ultrasonic bath. The catalysts sample was filtered, thoroughly washed with distilled water and dried at 65 • C under air. Prior to kinetic experiments, catalysts' sample was reduced at 250 • C for 60 min in the hydrogen flow in tube furnace than cooled to ambient temperature and flashed with nitrogen and immediately transferred to the catalytic reactor. The following samples were synthesized by this methodology and denoted as 5 wt.% Pd/γ-Al 2 O 3 , 5 wt.% Pd/SiO 2 , 5 wt.% Pd/C, 5 wt.% Pd/SiO 2 -C 3 H 6 -NH 2, with an NH 2 content of 30 wt.%.

Fourier Transform Infrared Spectroscopy (FTIR)
The diffuse reflectance IR spectroscopy was carried out using the FTIR spectrometer IRPrestige-21 (Shimadzu, Kyoto, Japan) equipped with a diffuse reflection attachment. DRS-8000 was used for the qualitative composition of catalyst surface. The resolution of all spectra was 4 cm −1 and spectra were registered in the wave number range 490-4000 cm −1 .

Hydrogen Pulse Chemisorption
The metal dispersion and the chemically active surface area were determined due to pulse chemisorption analysis by applying a pulsed titration of the catalyst with a hydrogen. The spectra were registered on an automatic analyzer of chemosorption AutoChem HP 2950 (Micromeritics, Norcross, GA, USA).

XPS
X-ray photoelectron spectroscopy (XPS) data were obtained using ES-2403 spectrometer (manufacturer: Institute for Analytic Instrumentation of RAS, St. Petersburg, Russia) with anode Mg Kα (hν = 1253.6 eV), energy analyzer PHOIBOS 100-MCD5 (SPECS, Berlin, Germany) and X-Ray source XR-50 (SPECS, Berlin, Germany). All the data were acquired at an X-ray power of 250 W. Survey spectra were recorded at an energy step of 0.5 eV with the analyzer pass energy 40 eV, and high resolution spectra were recorded at an energy step of 0.05 eV with the analyzer pass energy 7 eV. Samples were degassed within 180 min before analysis and were stable during the treatment. The obtained spectra were analyzed using CasaXPS software taking sensitivity factors in the quantity analysis of data into account.

SEM
The morphological characteristics of the amino-functionalized mesoporous silica and catalyst were examined by scanning electron microscopy (SEM, TESCAN, Vega-LSU) equipped with X-ray microanalysis (OXFORD INCA PentaFETx3). Scanning electron microscope images were acquired at a magnification of 66.1 kX at 20 kV with an SE detector.

TEM
The structural properties of the samples were examined using an electron microscope (JEM-2200FS, accelerating voltage 200 kV) in the transmission high-resolution electron microscopy mode (PFEM, high-resolution transmission electron microscopy-HRTEM). A model of the microscope JEOL JEM-2200FS showed an accelerating voltage of 200 kV. Resolution: by points-0.19 nm; on the lattice-0.1 nm; in the mapping mode-0.2 nm; in the HAADF mode-0.14 nm. For the analysis of the elemental composition, electron microscopy was used in the energy dispersive X-ray spectroscopy mode (EMF, energy-dispersive X-ray spectroscopy (EDS)).

Temperature Programmed Desorption (TPD) of NH 3
The study of support acid properties was provided by AutoChem HP 2950 automatic chemisorption analyzer (Micromeritics, Norcross, GA, USA). The high purity grade helium was used by way of carrier gas. The signal was recorded by a heat conductivity detector. The studied sample was pre-processed in mixture 3% NH 3 -He flow at room temperature. The excess of NH 3 gas was desorbed by flushing the system with helium upon gradual heating with a rate of 10 • /min. The desorption of NH 3 was carried by heating the sample up to 950 • C with a constant heating rate of 10 • C per minute. Then, the sample was cooled to ambient temperature.

Catalysts Activity Experiments Description
The stirred reactor with the temperature-control was used to carry out 4-nitroaniline hydrogenation to 1,4-phenylenediamine. The catalysts 5 wt.% Pd/SiO 2 -C 3 H 6 -NH 2 , 5 wt.% Pd/SiO 2 , 5 wt.% Pd/γ-Al 2 O 3 and 5 wt.% Pd/C were tested in the hydrogenation of 4-nitroaniline in the environment of a water solution of 2-propanol (0.68 mole fraction). Before experiments, the reactor was flushed three times with hydrogen for air removal. The reaction rates were defined by volumetric measurements of hydrogen consumption. The rates of reaction were calculated as an inclination angle tangent of kinetic curves of hydrogen consumption. The test for the contribution of homogeneous catalysis to the reaction rate was performed by Sheldon's filtration test methodology. After the partial conversion of 4-nitroaniline, the reaction stopped and the catalyst was extracted by filtration. Further to this, the reaction was continued without catalyst, but chemical transformation wasn't observed. Thus, the hydrogenation was catalyzed only by heterogeneous catalysts.
Chromatographic analysis was performed using a gas chromatograph (Crystal, manufacturer Chromatek, Yoshkar-Ola, Russia). The reaction rate was controlled through the 4-nitroaniline conversion. After completion, the catalyst was separated from the reaction mixture by centrifugation.

Deactivation Experiments Using Recovered Catalysts
The deactivation of catalysts was studied by the hydrogenation of 4-nitroaniline to 1,4-phenylenediamine in a water solution of 2-propanol with the repeated injection of 4-nitroaniline at the end of reaction. Five repeated injections of 4-nitroaniline were carried out on each catalyst.

Conclusions
Silica surface modification with aminopropyltrimetoxysilane allows the effective synthesis of catalytically active Pd nanoparticles with diameters varying from 3 to 10 nm. Pd species mainly present in the form of metal palladium and palladium oxide(II). Silica surface modification with amino groups results in a higher palladium surface concentration (2.12 at%) compared to unmodified silica (1.47 at%) according to XPS data. This results in an increase in sample activity in p-nitroaniline hydrogenation reaction up to 10.1 mole(H 2 )/sec.·10 −5 for 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 compared to 0.8 mole(H 2 )/sec.·10 −5 for the unmodified sample 5 wt.% Pd/SiO 2 .
The comparison of TPD and kinetics data shows a certain correlation. The higher the acidity of the support, the higher the activity of supported palladium.
An increase in surface aminogroups concentration in the case of samples 5 wt.% Pd/SiO 2 -C 3 H 6 -(30%)NH 2 and 5 wt.% Pd/SiO 2 -C 3 H 6 -(10%)NH 2 results in an appropriate increase in catalytic activity by 1.9 times. This confirms the hypothesis that the electron-donating properties of the amino group have a positive effect on the catalytic activity of the supported metal. This also confirms the kinetic experiments using an unmodified, commonly used Pd supported on activated carbon, alumina and Raney nickel samples. Silica surface modification with amino aminopropyltrimetoxysilane has a positive influence on catalyst stability because of palladium nanoparticles' strong interaction with silica compare to unmodified support.

Conflicts of Interest:
The authors declare no conflicts of interest.