Introduction

Precious metals nanoparticles on catalyst supports have attracted considerable interest because they have been used in a number of catalysis applications, including transformation of organic1,2,3,4,5 and fuel cells6,7,8,9,10,11. Reports indicate that the catalytic performance could be greatly enhanced by nanotube supports. Halloysite nanotube (HNT) is a typical catalyst support material. Halloysite (Al2Si2O5(OH)4·2H2O), a hydrated layered aluminosilicate of the kaolinite group, containing octahedral gibbsite Al(OH)3 and tetrahedral SiO4 sheets (i.e., halloysite nanotubes), possesses a hollow cylinders formed by multiply rolled layers12,13. Because of its structural features, it offers potential application as support for catalytic composites with remarkable improved mechanical properties and dispersibility14,15,16,17. However, natural untreated halloysite has a very low adhesion to bond metal nanoparticles due to the absence of chemical conjunction, leading to the catalyst particles easily leaching from the support surface during the reactions.

It is well-known that the aminofunctional silanes can be applied to improve the adhesion between the metal nanoparticles and the support, thus increase the amounts of the metal particles loaded18,19,20,21. In recent years, there are many literatures about the immobilization of metal nanoparticles on clay minerals with organosilane, especially for tubular aluminosilicate clay (e.g. imogolite22,23, chrysotile24,25 and halloysite26,27,28), which plays an important role in catalytic area. The modification of halloysite surfaces by coupling with functionalized organosilanes is applicable to the fields of enzymes immobilization29,30, anti-corrosion31,32,33 and nanocomposite materials14,34.

In this paper, we reported the immobilization of palladium nanoparticles on silanized HNTs (Pd/NH2-HNTs) with the assistance of organosilane. The Pd/NH2-HNTs catalyst exhibits an excellent activity in the hydrogenation of styrene because of the strong synergistic interaction between palladium nanoparticles and HNTs. We investigated the interfacial characteristics of palladium, organosilane and halloysite and its possible enhanced catalytic activity in detail.

Results

The XRD patterns of HNTs, NH2-HNTs and Pd/NH2-HNTs are shown in Figure 1a. For the HNTs sample, all of the observed peaks are close to the characteristic data of halloysite (JCPDS card No. 29-1487). For the NH2-HNTs sample, all of the observed peaks are also consistent with HNTs, indicating that the existence of APTES in the NH2-HNTs nanocomposites doesn't affect the structure of HNTs. Full range XPS spectra of samples are exhibited in Figure S1, the semiquantitative analysis of the relative atomic concentration of different elements are shown in Table S1. Si and Al are detected in both types of samples and the concentration ratio of Si/Al is 1.46 in NH2-HNTs, significantly higher than the value of 1.14 in HNTs, which confirms that the silane coupling agent is present in NH2-HNTs. The Si in NH2-HNTs partly comes from the APTES. Compared with HNTs, N 1s signal is detected in NH2-HNTs, but the peak is not obvious especially for the sample with low APTES concentration. Those results indicate that the N species are belonging to APTES. For the Pd/NH2-HNTs sample, a weak peak observed at 40°, corresponding to (111) diffraction of palladium. Those results indicate that palladium salts have been reduced to metallic Pd, which is deposited on NH2-HNTs as nanoparticles. The characteristic peak of palladium in Pd/NH2-HNTs with high APTES concentration is not obvious (Figure S2), which should be due to only a partial reduction of Pd(II) to Pd(0). The crystal size of the Pd nanoparticles in Pd/NH2-HNTs sample calculated by the Scherrer's equation is ~ 3 nm.

Figure 1
figure 1

Crystallization and spectra of the samples.

(a) XRD patterns and (b) FTIR spectra of HNTs, NH2-HNTs and Pd/NH2-HNTs, (c) amplification of the indicated part in (b).

Full size image

The FTIR spectra of HNTs, NH2-HNTs and Pd/NH2-HNTs are shown in Figure 1b. For HNTs, the peak at 3698 cm−1 can be assigned to the O-H stretching vibration of the inner-surface Al-OH groups. The peak at 3622 cm−1 can be ascribed to the O-H stretching vibration of the inner Al-OH groups (between the interface of the Si-O tetrahedron and the Al-O octahedron). The broad band between 1633 and 3448 cm−1 arises from adsorbed water. The peak at 1037 cm−1 corresponds to the in-plane stretching vibration of the Si-O network (Si-O-Si and O-Si-O). The peak at 912 cm−1 is attributed to the O-H deformation vibration of inner Al-OH groups. The deformation vibration of Al-O-Si at 553 cm−1 is attributed to Al-O tetrahedral sheets. The Si-O-Si deformation vibration observed at 471 cm−1 is attributed to Si-O tetrahedral sheets.

After the functionalization of HNTs by APTES, the broad band at 3360 cm−1 is attributed to the N-H2 asymmetric stretching vibration. The peak at 2930 cm−1 is assigned to the symmetric stretching vibration of C-H2. The peak at 1343 cm−1 is attributed to the deformation (scissoring) vibration of Si-CH. The peak at 1304 cm−1 can be assigned the stretching vibration of C-N band. The peak at 1257 cm−1 is ascribed to deformation (scissoring) vibration of C-H3. All of these observations prove the presence of the APTES in the functionalized HNTs. The signals of adsorbed water become increased with the increase of the APTES concentration. (Figure S3 & Figure S4).

After NH2-HNTs being assembled by PVP-Pd, some peaks have shifted and some are increased, indicating the existence of Pd in Pd/NH2-HNTs. The C = O stretching vibration of PVP at 1659 cm−1 is attributed to the interaction between PVP and metal species. The C-H3 broad stretching band around 1257 cm−1 and the stretching vibration of C-N band at 1304 cm−1 shift to 1240 cm−1 and 1296 cm−1, respectively. These shifts of peaks in the Pd/NH2-HNTs are associated with the formation of bonds between PVP-Pd and NH2-HNTs.

TEM images of HNTs and Pd/NH2-HNTs are presented in Figure 2. The white HNTs are short cylindrical hollow tubes averaging 0.7–1.5 μm in length, with an external diameter of 50–75 nm and an internal diameter of 10–30 nm (Figure 2a). After assembly with Pd nanoparticles, the characteristic tube morphology of the original minerals have been retained (Figure 2b). According to the additional EDS spectra (Figure S5a,c, inset), compared to those deposited on HNTs without silanization (1.0 wt.%, Figure S5c,d), more Pd nanoparticles (11.3 wt.%) are densely dispersed on the external and internal surfaces of the NH2-HNTs (Figure S5a,b, Figure 2c), attributed to the amino groups on the surface of NH2-HNTs catching the metal colloidal more easily through coordinate bond, which is beneficial to the enhanced catalytic activity of Pd/NH2-HNTs. The size of Pd nanoparticle is observed as ~ 3 nm (Figure 2d), which is shows very nice consistency with the particle size estimated by XRD.

Figure 2
figure 2

Morphologies of the samples.

TEM images of (a) HNTs and (b,c) Pd/NH2-HNTs, (d) enlargement of the indicated part in (c), inset is the particle size distribution of Pd on NH2-HNTs.

Full size image

Figure 3a shows the 29Si CP/MAS NMR spectra of HNTs, NH2-HNTs and Pd/NH2-HNTs. The spectrum of HNTs indicates an obvious peak at −92 ppm, assigned to the Si(OSi)3(OAl) of halloysite. The spectrum of NH2-HNTs and Pd/NH2-HNTs displays two more resonance peaks at −65 ppm corresponding to tridentate and at −57 ppm, attributed to bidentate. The tridentate form of Si implies that all three of the ethoxyl groups of APTES are hydrolyzed and condensed with surface hydroxyl groups of halloysite or with other hydrolyzed APTES. Observation of the bidentate form of Si means that some APTES species possess one ethoxyl (or hydroxyl) group that did not hydrolyze or condense.

Figure 3
figure 3

Spectra analysis of the samples.

(a) Solid-state 29Si and (b) 13C NMR spectra of HNTs, NH2-HNTs and Pd/NH2-HNTs, respectively. (c) Pd 3d XPS spectra of Pd/NH2-HNTs.

Full size image

Figure 3b shows the 13C CP/MAS NMR spectra of HNTs, NH2-HNTs and Pd/NH2-HNTs. For HNTs, no signals are resolved by Gaussian deconvolution. For the spectra of NH2-HNTs and Pd/NH2-HNTs, the signal at 12.8 ppm is assigned to the carbon atoms at position Si-C in ethoxyl groups. The signal at 28.1 ppm is attributed to the carbon atoms at position C-C in the propyl groups, the intensities of corresponding signals become increased as the APTES concentration increased (Figure S6).

Figure 3c shows the Pd 3d spectra and their peak positions from the sample. Pd/NH2-HNTs has photoelectron peaks corresponding to both Pd(0) and Pd(II) species. The Pd 3d5/2 and Pd 3d3/2 electronic states for Pd(0) are observed at 335.3 and 340.6 eV, respectively. Peaks are also detected at 337.0 and 342.3 eV, corresponding to the 3d5/2 and 3d3/2 electronic states, of Pd(II) species, respectively, which suggests that the reduction process led to only a partial reduction of Pd(II) to Pd(0) for Pd/NH2-HNTs. As in this investigation, the atomic concentration of Pd 3d increases with increasing the atomic concentration of the N 1s (Table S1), which could be related to APTES-Pd(II) or APTES-Pd(0) complex. Figure S7a shows that Pd2+ is not reduced until the hydrazine hydrate is added. Based on the peak integration (Figure S7b), the proportion of Pd(0)/(Pd(0) + Pd(II)) in Pd/NH2-HNTs with high APTES concentration is 23.5%, lower than the Pd/NH2-HNTs with low APTES concentration (31.2%), indicating that the coordination between Pd(II) and APTES makes Pd(II) difficult to be reduced.

The catalytic properties of the HNTs, Pd/HNTs and Pd/NH2-HNTs are tested in the hydrogenation of styrene to ethylbenzene in an autoclave, as shown in Figure 4. It can be seen that the hydrogenation rate of Pd/NH2-HNTs is higher than that of Pd/HNTs under the same reaction conditions. Correlation of the catalytic activities of the Pd nanoparticles with their distribution indicates that a higher loading content and dispersion led to higher catalytic activity (Figure S8). Our experimental findings suggest that the surface property of the NH2-HNTs is beneficial to improving the dispersion and loading content, leading to a higher catalytic activity.

Figure 4
figure 4

Catalytic property of the samples.

The change of concentration of styrene and ethylbenzene with the reaction proceeding.

Full size image

Discussion

The present process is schematically illustrated in Figure 5. Firstly, the surface of HNTs is functionalized with -NH2 by APTES. The functionalized HNTs possess a large number of -NH2 groups with positive charge and can catch the PVP-Pd2+ colloidal with negative charge through coordinate bonds35,36,37. Due to the bond formation between PVP-Pd2+ and -NH2 groups, most of the PVP-Pd2+ are combined with the -NH2 groups. Then, the addition of hydrazine hydrate could result in the formation of Pd-NH2 complexes; meanwhile the Pd ions are reduced by hydrazine hydrate. The bond formation between Pd and -NH2 groups could be testified.

Figure 5
figure 5

Pd nanoparticles assembled on silanized HNTs.

Schematic of the fabrication process for Pd/NH2-HNTs nanocomposites.

Full size image

The peak fitting program on XPS spectra of Al 2p and Si 2p is performed to detect the possible interaction between functional APTES and HNTs27,38. The XPS spectra, together with their fitting curves, are showed in Figure 6. The content of aluminol (Al-OH) at 74.8 eV on HNTs inner surface is quite low and the aluminumeoxygen (Al-O) bond is dominant in the existing forms of Al atoms. The binding energy (BE) value of Si 2p in silanol (Si-OH) is at 103.3 eV. The content of Si-OH, located at the outer side of HNTs lumens, is lower than that of siliconeoxygen (Si-O) bond (102.7 eV). When HNTs is treated with APTES, the XPS spectra of Al 2p show that the BE value of Al atom in Al-O and Al-OH are shifted to 74.6 and 75.1 eV, respectively. According to the spectra of Si 2p, the BE value of Si in Si-OH partially moves to 103.4 eV. The other part of Si-O bonding moves to 102.8 eV. The obviously changed BE values for Al and Si atoms may be related to coupling reaction between ethoxyl groups group in APTES and hydroxyls (Al-OH and Si-OH) on HNTs. The covalently bonding between APTES and HNTs is possibly responsible for the obvious BE shifts (Figure S9). After NH2-HNTs being assembled by PVP-Pd, the XPS spectra of Al 2p show that the BE value of Al atom in Al-O and Al-OH are shifted to 74.4 and 74.9 eV, respectively. According to the spectra of Si 2p, The BE value of Si in Si-OH partially shifted to lower binding energies with a higher value for the full width at a half-maximum (FWHM). The BE value of Si in Si-O moves to 103.1 eV. The obviously changed BE values for Al and Si atoms may be related to coupling reaction between carbonyl group in PVP-Pd and amino on NH2-HNTs.

Figure 6
figure 6

XPS spectra of the samples.

(a) Al 2p and (b) Si 2p and XPS spectra of HNTs, NH2-HNTs and Pd/NH2-HNT, (c) O 1s of HNTs and NH2-HNTs and (d) N 1s XPS spectra of NH2-HNTs and Pd/NH2-HNT.

Full size image

To prove that the APTES were chemically bonded onto HNTs, the O 1s spectra of the HNTs and NH2-HNTs are presented in Figure 6c. As observed in Figure 6c, the intense characteristic peak that emerged at 532.0 eV corresponds well with O 1s of HNTs binding energy. After the functionalization of HNTs by APTES, the binding energy of O 1s increases to 532.4 eV, indicating that the APTES has successfully grafted on the HNTs. To further confirm the Pd nanoparticles are chemically bonded onto NH2-HNTs, Figure 6d shows the N 1s spectra of the NH2-HNTs and Pd/NH2-HNTs, the peak occurred at 400.4 eV, standing for the chemical states of nitrogen in NH2-HNTs. However, the binding energies of N 1s observed in Pd/NH2-HNTs shifts to 401.2 eV, demonstrating the strong interaction between Pd and -NH2 groups. The fitting line at lower binding energy (Peaks at 399.6 eV) is attributed to an N species in PVP-Pd. So, all of the above results can confirm the bond formation between the Pd and -NH2 groups and the functional groups can combine Pd with chemical bond, which shows an exciting prospect for immobilization of Pd.

In summary, the surface of HNTs has been functionalized with silane coupling agent and Pd nanoparticles are then deposited on the surface of NH2-HNTs to fabricate Pd/NH2-HNTs composites. The distribution of Pd nanoparticles deposited on silanized HNTs are much more uniform and the particle size is smaller, compared to those deposited on HNTs without silanization. Moreover, the Pd nanoparticles deposited on NH2-HNTs show a higher activity in the liquid-phase hydrogenation of styrene to ethylbenzene.

Methods

Material preparation

Halloysite nanotubes (HNTs) were obtained from Hunan, China. All chemicals were analytical grade and used without further purification. HNTs were pretreated via emulsion dispersion, filtering, washing with distilled water and drying for 8 h at 313 K. (1) PVP-PdCl2 was prepared as follows: 18 mg of sodium tetrachloride palladate (Na2PdCl4) was dissolved in 50 mL of methanol containing 19 mg of poly (vinyl pyrrolidone) (PVP). The solution was refluxed for 2 h until a yellowish-brown color was observed. The solution was stored in a dark bottle at 4°C for further use. (2) NH2-HNTs nanocomposites were synthesized as follows: 4 mL of APTES was dissolved in 250 mL of dry toluene. Approximately 5 g of clay powder was added and the suspension was dispersed ultrasonically for 30 min. Evacuation pretreatment may be carried out at this stage. The suspension was then refluxed at 120°C for 20 h under constant stirring. In the refluxing system, a calcium chloride drying tube was attached to the end to ensure a dry environment. The solid phase in the resultant mixture was filtered and extensively washed six times with fresh toluene to remove the excess organosilane, then dried overnight at 120°C for further curing. (3) PVP-Pd/NH2-HNTs nanocomposites were synthesized as follows: 0.3 g NH2-HNTs sample was kept in 150 mL PVP-PdCl2 solution for 12 h and was then rinsed thoroughly with distilled water and dried at room temperature. After impregnation, the color of the sample changed from white to yellowish-brown. Finally, the sample was reduced by 60 mL of a 0.015 M alkaline solution (pH = 9.5) of hydrazine hydrate (N2H4·H2O), the color of the sample changed from yellowish-brown to dark-brown, suggesting the formation of metallic palladium nanoparticles on the surface. (4) Pd/HNTs nanocomposites were synthesized as follows: 0.3 g HNTs was kept in 150 mL PVP-PdCl2 solution for 12 h, then thoroughly washed with distilled water and dried at room temperature. After impregnation, the color of the sample changed from white to yellowish-brown. Finally, the sample was reduced by 60 mL of a 0.015 M alkaline solution (pH = 9.5) of hydrazine hydrate (N2H4·H2O), the color of the sample changed from yellowish-brown to dark-brown, suggesting the formation of metallic palladium nanoparticles on the surface.

Characterization

The crystalline phases were identified by XRD analysis using a RIGAKU D/max-2550VB+ 18-kW powder diffractometer with Cu Kα radiation (λ = 1.5418 Å). The data were collected in the scanning range 2θ = 10–80°, with a scanning speed of 2°/min. FTIR spectra were recorded using a Nicolet 5700 spectrophotometer. TEM images were obtained using a JEOL JEM-200CX instrument equipped with an energy dispersive X-ray spectroscopy (EDS) at an accelerating voltage of 200 kV. Solid state 29Si MAS NMR measurements were recorded using a Bruker AMX400 spectrometer in a static magnetic field of 9.4 T at a resonance frequency of 79.49 MHz. X-ray photoelectron spectroscopy (XPS) analysis was performed on Thermo Fisher Scientific K-Alpha 1063 spectrometer equipped with an Al Ka monochromator X-ray source. The test chamber pressure was maintained below 10−9 mbar during spectral acquisition. The XPS binding energy (BE) was internally referenced to the C 1s peak (BE = 284.1 eV). The catalytic hydrogenation of styrene to ethylbenzene (Figure S10) was carried out in a 300 mL stainless steel autoclave equipped with a magnetically driven impeller. For each experiment, 1 mL of styrene dissolved in 200 mL of ethanol was first charged into the autoclave reactor. Once the Pd/NH2-HNTs were mixed inside the autoclave, hydrogen was introduced into the reactor to a set level and the stirring rate was maintain at 60 rpm.