- Email: [email protected]
Effect of water vapor and hydrogen treatments on the surface structure of Ni3Al foil
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Applied Surface Science xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil Ya Xu a,∗ , Yan Ma b , Junya Sakurai a , Yuden Teraoka c , Akitaka Yoshigoe c , Masahiko Demura a , Toshiyuki Hirano a a
Hydrogen Materials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan North China Electric Powder University, 2 Beinong Road, Huilongguan, Changping District, Beijing 102206, China c Quantum Beam Science Directorate, Japan Atomic Energy Agency, 1-1-1 Kouto, Sayo-cho, Sayo-gun, Hyogo 679-5148, Japan b
a r t i c l e
i n f o
Article history: Received 30 December 2013 Received in revised form 22 February 2014 Accepted 24 February 2014 Available online xxx Keywords: Ni3 Al foil catalyst Synchrotron radiation X-ray photoemission spectroscopy Water vapor treatment Hydrogen reduction
a b s t r a c t We have developed a water vapor treatment followed by hydrogen reduction to modify the surface structure of Ni3 Al foils in order to obtain high catalytic activity. The Ni3 Al foils were heat treated in water vapor at 873 K for 1 h followed by H2 reduction at 873 K for 1 h. The effects of the water vapor treatment and the H2 reduction on the surface structure of the Ni3 Al foils were investigated by means of scanning electron microscopy and synchrotron radiation X-ray photoemission spectroscopy. Both Ni and Al in the surface layer of the Ni3 Al foil were oxidized during the water vapor treatment; fine NiO particles with a high density were formed on the outermost surface, accompanied by the formation of oxide layers of Al(OH)3 and NiAl2 O4 /Al2 O3 beneath the NiO particles. The NiO particles were reduced to metallic Ni and the Al(OH)3 was decomposed to Al2 O3 , whereas the NiAl2 O4 and Al2 O3 remained unchanged during the subsequent H2 reduction, forming a Ni-enriched porous structure on the surface layer of NiAl2 O4 /Al2 O3 . © 2014 Elsevier B.V. All rights reserved.
1. Introduction The Ni3 Al intermetallic compound is known as a promising high-temperature structural material because of its excellent hightemperature strength and corrosion/oxidation resistance [1–3]. However, its poor room-temperature ductility has been a serious problem. We have overcome this problem and successfully developed thin foils of Ni3 Al with a thickness of less than 30 m by cold rolling unidirectionally solidified ingots [4–6]. Recently, we investigated the catalytic properties of the Ni3 Al foils for methanol decomposition in the temperature range from 513 to 793 K and found that such flat foils show high catalytic activity and selectivity for methanol decomposition into H2 and CO, despite the small surface area of the foils, demonstrating that the Ni3 Al foils can be used as plate type catalysts [7], and thus serve both functionalities of catalytic and structural materials, by which one can make a more efficient reactor for hydrogen production [8,9]. The high catalytic activity for methanol decomposition was attributed to the formation of fine Ni particles on the foil surface through a selective
∗ Corresponding author at: Hydrogen Materials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. Tel.: +81 298592573; fax: +81 298592501. E-mail address: [email protected] (Y. Xu).
oxidation and/or hydroxylation of Al by the small amount of H2 O produced during the reaction [9,10]. This result suggests that water vapor oxidation is effective for modifying the surface morphology and enhancing the catalytic activity of Ni3 Al foils. Many studies have been carried out on the oxidation of Ni3 Al in air or pure O2 [11–17], whereas there have been very few studies on the oxidation of Ni3 Al in water vapor [18,19]. Schumann et al. [18] studied the oxidation behavior of Ni3 Al single crystals under low oxygen partial pressure realized by exposing the Ni3 Al in a flowing H2 /H2 O mixture at 1223 K. Fine Ni particles were observed on the surface of the Ni3 Al after being exposed for 1 min, accompanied by formation of a continuous thin ␥-Al2 O3 oxide scale. Continued oxidation resulted in thickening of the ␥-Al2 O3 scale and coalescing of the Ni particles. However, they did not report the effect of water vapor at temperatures lower than 1223 K and at high water vapor partial pressures. Garza et al. [19] studied the interaction of an oxide film on Ni3 Al with the water vapor, and they revealed that the water vapor can significantly affect the oxide film even at low H2 O pressures. However, they did not mention the oxidation behavior of Ni3 Al caused by water vapor. In this study, we carried out a water vapor treatment of Ni3 Al foils at 873 K, followed by H2 reduction at 873 K, which is commonly used as a pre-reduction process for Ni3 Al catalysts [20,21]. The effect of the water vapor treatment and H2 reduction on the surface structure of the Ni3 Al foils was investigated by means of
http://dx.doi.org/10.1016/j.apsusc.2014.02.144 0169-4332/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model APSUSC-27360; No. of Pages 6
ARTICLE IN PRESS Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
2
scanning electron microscopy (SEM) and synchrotron radiation Xray photoemission spectroscopy (SR-XPS).
2. Experimental 2.1. Foil sample preparation Ni3 Al (Ni-24 at% Al) foils with a thickness of 30 m were fabricated using 98% cold rolling of single crystals [5,6]. Samples for SR-XPS measurement were cut into 7 mm × 7 mm squares and then mechanically polished to get a surface roughness height of less than 0.5 m. After polishing, the samples were washed with ethanol and de-ionized water in an ultrasonic bath.
2.2. Water vapor and H2 reduction treatments Water vapor and H2 reduction treatments were carried out in a conventional fixed-bed flow reactor composed of a quartz tube (internal diameter 8 mm) in an electric furnace connected with gas and liquid supply units [7,8]. The samples were first heated up to 873 K in flowing N2 (30 mL/min) and then water vapor was introduced using a micro-pump at a liquid flow rate of 50 L/min for 1 h with continually flowing N2 (101 kPa total pressure). The partial pressure of the water vapor in the reactor was 68 kPa during the water vapor treatment. After the water vapor treatment, the H2 reduction was carried out at 873 K for 1 h in a flow of mixed H2 (30 mL/min) and N2 (5 mL/min). After the treatments, the samples were cooled down to room temperature in flowing N2 (30 mL/min). Then the samples were taken out and preserved in a desiccator at room temperature for several days until SEM and SR-XPS analyses.
2.3. SR-XPS analysis The SR-XPS measurements were performed on the samples in the as-polished state, after the water vapor treatment, and after the H2 reduction using monochromatic synchrotron radiation with a photon energy of 1486.6 eV at the high-energy resolution soft Xray beam line BL23SU at the synchrotron radiation facility SPring-8 (Japan). A detailed description of the BL23SU and the experimental station used can be found in previous reports [22,23]. The base pressure of the analysis chamber was maintained below 5 × 10−8 Pa during the measurements. The core level spectra of Ni 2p, Al 2p, Ni 3p, O 1s, and C 1s were measured for each sample. In order to obtain the depth profile of the surface structure, the measurements were carried out at different take-off angles (TOA) from 0◦ to 60◦ with respect to the surface normal. The photon energy scale was calibrated by using the peak position of the Au 4f7/2 core level at 84.0 eV [24].
2.4. Surface morphology analysis The morphology of the surface and cross section of the foils in the as-polished state, after the water vapor treatment, and after the H2 reduction was examined by means of scanning electron microscopy (SEM) (JEOL, JSM-7000F) with a field emission gun. The chemical composition of the surface products was analyzed by an energy dispersive X-ray spectroscopy (EDS) system equipped in the SEM. The cross-sectional samples were prepared by means of an argon ion beam cross-section polisher (JEOL, SM09010).
Fig. 1. Backscattered electron image of the cross section of as-polished Ni3 Al foil.
3. Results and discussion 3.1. As-polished samples Fig. 1 shows the SEM backscattered electron (BE) image of the cross section of the as-polished foil. The surface of the foil was macroscopically smooth. No particles or precipitates were observed near the surface. The image contrast shows no obvious change from the surface to the depths of the foil, indicating a uniform elemental composition across the cross section of the foil. Fig. 2 shows the Ni 2p (Fig. 2a), Al 2p and Ni 3p (Fig. 2b) SRXPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ for the as-polished foil. At TOA = 0◦ , a Ni 2p3/2 peak at a binding energy of 853.6 eV and a Ni 2p1/2 peak at 870.8 eV were detected; they were accompanied by satellite peaks. These peaks correspond to metallic Ni [24]. With increasing TOA, the shape of the satellite peaks changed, and the spectral intensity ratio of the satellites relative to the main Ni 2p peaks increased. This change matches the features of Ni oxides [11,19,24], indicating that a small amount of Ni oxide was formed near the surface even though the oxides were not detected by SEM observation. The Ni 3p spectra in Fig. 2b also show a shape change with the increase of TOA. At TOA = 0◦ , a Ni 3p main peak is observed at a binding energy of 67.0 eV with a shoulder on the high binding energy side, which agrees well with the Ni 3p spectra of metallic Ni [10,12]. However, the intensity of the shoulder relative to the main Ni 3p peak increased with the increase of TOA, becoming even larger than the main Ni 3p peak at TOA = 60◦ , which is typical of Ni oxides [14]. This result is in agreement with that of Ni 2p, indicating the formation of Ni oxides on the surface. These Ni oxides are thought to be native Ni oxides (NiOx ) formed during polishing or exposure to air at room temperature. The Al 2p core level showed two chemical states, as shown in Fig. 2b. The major peak located at 73.0 eV corresponds to metallic Al, whereas the minor peak at around 74.5 eV corresponds to Al oxide, probably a native amorphous AlOx [10,14]. With increasing TOA, the intensity of the AlOx peak relative to the metallic Al peak increased, indicating the presence of AlOx on the surface. 3.2. Samples after water vapor treatment Fig. 3a shows the SEM secondary electron (SE) image of the Ni3 Al foil surface after water vapor treatment. A high density of fine particles on the surface was observed. Most of the particles were
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
3
Intensity (Arbitrary units)
Intensity (Arbitrary Units)
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
890
880
870 860 Binding Energy (eV)
850
80
75
70 65 Binding Energy (eV)
60
Fig. 2. SR-XPS spectra of the as-polished Ni3 Al foils measured at TOA = 0◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra.
Table 1 EDS analysis of the chemical compositions of the positions marked in Fig. 3b for the foil after water vapor treatment. Element (At%)
Ni
Al
O
Position A Position B Position C
51.0 26.6 69.1
3.9 28.3 26.5
45.1 45.1 4.5
irregularly shaped, and the size of the particles ranged from several tens to several hundreds of nanometers. The large particles seem to be aggregates of small particles. Fig. 3b shows the SEM BE image of a cross section of the foil after water vapor treatment. Three surface layers were observed. The outermost layer composed of the particles was discontinuous. The second layer just beneath the particles was thin but continuous with a relatively dark contrast. The third layer beneath the dark thin layer was relatively thick and clear grain or phase boundaries can be observed in it. The chemical composition in each layer was examined by EDS analysis. Table 1 shows the chemical compositions of the positions marked in Fig. 3b after the water vapor treatment. The particles on the outermost surface (position A) were composed of Ni and O, with a very small amount of Al, indicating that the particles were NiO. The third layer (position B) had a higher Al concentration and lower Ni concentration as compared to the other layers, suggesting that NiAl2 O4 and/or Al2 O3 might have been formed in this layer. The substrate beneath the third layer (position C) was mainly composed of Ni and Al, with very small amounts of O, and its composition was close to Ni3 Al. The second layer was not examined by EDS analysis, but it is thought to be mainly composed of Al and O since it was relatively dark in
the BE image, considering that the brightness in the BE image arises mainly from the heavier element (Ni). Fig. 4a shows the Ni 2p, and Fig. 4b shows the Ni 3p, and Al 2p SRXPS spectra obtained at TOA = 0◦ , 20◦ , 40◦ , and 60◦ for the foil after water vapor treatment. Both the Ni 2p and the Ni 3p spectra were much different from the corresponding metallic Ni spectra of the as-polished samples shown in Fig. 2. The main peaks of both Ni 2p and Ni 3p appeared at higher binding energies than those of metallic Ni 2p and Ni 3p. In addition, strong satellite peaks of Ni 2p3/2 and Ni 2p1/2 were also observed. These features were related to the presence of Ni oxides. As shown in Fig. 4a, the shape of the main Ni 2p3/2 spectra changed with increasing TOA. At TOA = 0◦ , the main Ni 2p3/2 spectra were composed of three components at 857.5, 855.7, and 853.6 eV. The major component at 857.5 eV corresponds to the spinel NiAl2 O4 , the second component at 855.7 eV corresponds to NiO, and the third component at 853.6 eV corresponds to metallic Ni according to previous reports [25,26]. With increasing TOA, the relative intensity of the NiAl2 O4 component decreased, whereas that of the NiO component increased and became the largest at TOA = 60◦ . This result indicates that NiO was a surface species and NiAl2 O4 was a subsurface species. In addition, the relative intensity of the metallic Ni component decreased with increasing TOA and almost disappeared at TOA = 60◦ , suggesting that this Ni component was not a surface species. Instead, it corresponded to the metallic Ni in the substrate. The Al 2p core level showed a major peak at 75.5 eV at TOA = 0◦ , which corresponds to Al oxides, hydroxides, and spinel [10,24,25], and no metallic Al peak was detected (Fig. 4b). This means that the Al in the surface layer was oxidized and/or hydroxylated during the water vapor treatment. This major peak is thought to correspond to
Fig. 3. SEM images of the Ni3 Al foils after the water vapor treatment. (a) Secondary electron image of the foil surface; (b) backscattered electron image of the cross section with the marked positions for EDS analysis shown in Table 1.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
Intensity (Arbitrary units)
4
890
880
870 860 Binding Energy (eV)
850
80
75
70 65 Binding Energy (eV)
60
Fig. 4. SR-XPS spectra of the Ni3 Al foils after the water vapor treatment measured at TOA = 0◦ , 20◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra. Table 2 EDS analysis of the chemical compositions of the positions marked in Fig. 5b for the foil after the H2 reduction. Element (At%)
Ni
Al
O
Position A Position B Position C
87.3 39.8 69.9
4.4 39.3 24.8
8.4 20.9 5.3
a mixture of an Al2 O3 component at 75.8 eV, an Al(OH)3 component at 75.2 eV, an NiAl2 O4 component and a native AlOx component at 74.6 eV. As shown in Fig. 4b, the major peak slightly shifted to a lower binding energy with increasing TOA. Because the NiAl2 O4 was a subsurface species, the slight shift is probably due to the formation of Al(OH)3 and AlOx , especially Al(OH)3 above Al2 O3 . In addition, the intensity of the Al 2p spectra relative to the Ni 3p spectra decreased with increasing TOA, suggesting that the Al2 O3 and Al(OH)3 were beneath the NiO. 3.3. Samples after H2 reduction The samples treated by water vapor were reduced by flowing H2 at 873 K for 1 h. Fig. 5a shows the SE image of the foil surface after H2 reduction. The particles locally coalesced with each other, forming a porous surface structure. Fig. 5b shows the BE image of the cross section of the foil after H2 reduction. Two layers were formed: an outermost layer, which was much brighter than other regions, and a sub-surface layer, which had a nonuniform thickness. The chemical composition in each layer was measured by EDS (Table 2). The results show that the outermost layer was mainly composed of Ni with a small amount of Al2 O3 , which means that the NiO particles that formed during the water vapor treatment were reduced to metallic Ni by the H2 reduction. The subsurface layer was composed of Ni, Al, and O, suggesting that this layer might be a mixture of NiAl2 O4 and Al2 O3 . On the other hand, the dark thin layer beneath the outermost layer, which was present after the water vapor treatment, was not identified. Fig. 6a shows the Ni 2p SR-XPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ for the foil after H2 reduction. The main Ni 2p3/2 spectra were composed of two components at 857.5 and 853.6 eV. The component at 857.5 eV corresponded to NiAl2 O4 and the other component at 853.6 eV corresponded to metallic Ni. In contrast, a NiO peak at 855.7 eV was extremely weak. This result reveals that most of the NiO was reduced to metallic Ni by the H2 reduction, whereas NiAl2 O4 was not. Fig. 6b shows the Al 2p and Ni 3p SRXPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ . The Al 2p spectra at TOA = 0◦ showed a major peak at 75.8 eV, which might correspond to a mixture of Al2 O3 , NiAl2 O4 , and AlOx , and no obvious shift of the peak was observed with increasing TOA. This result suggests that
no Al(OH)3 remained on the surface after H2 reduction. Considering that the thin dark layer that formed during the water vapor treatment (Fig. 3b) disappeared after the H2 reduction (Fig. 5b), it is thought that Al(OH)3 was probably present in this dark layer. In contrast, Al2 O3 might not show an isolated layer; it may instead coexist with NiAl2 O4 in the subsurface layer or with Ni in the outermost layer, which was suggested by the SEM observation (Fig. 5b) and EDS analysis results (Table 2). The Al 2p spectra relative to the Ni 3p spectra showed no obvious change with increasing TOA, indicating that there was no obvious separation between Al2 O3 and NiAl2 O4 (Ni) along the depth direction. This result is in agreement with the above SEM and EDS results, suggesting that Al2 O3 coexisted with NiAl2 O4 in the subsurface layer and with Ni in the outermost layer. 3.4. Surface structure evolution during the water vapor and H2 treatments The above results show that both Ni and Al in the surface layer of the Ni3 Al foil were oxidized or hydroxylated by the water vapor treatment, resulting in the formation of NiO particles on the outermost surface, accompanied by oxide layers of Al(OH)3 , NiAl2 O4 , and Al2 O3 beneath the particles. By the subsequent H2 reduction, the NiO was reduced to metallic Ni, forming a porous structure of Ni on the outermost surface. On the basis of these results, the evolution of the surface structure caused by water vapor treatment and H2 reduction is discussed below. For the as-polished samples, a thin layer of native Al oxide, AlOx , with a small amount of native Ni oxide (NiOx ) was present on the surface of the foil (Fig. 2). These native oxides are thought to have formed when the samples were exposed in air. The thickness of such native AlOx had been estimated to be approximately 13 A˚ [10]. This surface structure in the as-polished state is illustrated in Fig. 7a. After water vapor treatment, NiO, NiAl2 O4 , Al2 O3 , and Al(OH)3 were detected (Figs. 3 and 4). Al2 O3 can be formed more easily than NiO, considering that the dissociation pressure of Al2 O3 is approximately 10−40 Pa, much lower than that of NiO, which is approximately 10−14 Pa at 873 K. It is thought that the Al in Ni3 Al was preferentially oxidized to Al2 O3 at the beginning of the water vapor treatment in accordance with reaction (1). 2Ni3 Al + 3H2 O ⇔ 6Ni + Al2 O3 + H2
(1)
This process also led to the formation of metallic Ni. Then, the metallic Ni diffused toward the outside, forming the fine Ni particles on the outermost surface. The origin of the out-diffusion of Ni has been discussed by Bobeth et al. [27] and was attributed to a relaxation of internal stresses produced by the formation of Al2 O3 . We assume that not all the metallic Ni diffused outside; some of
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
5
Intensity (Arbitrary units)
Intensity (Arbitrary units)
Fig. 5. SEM images of the Ni3 Al foils after the H2 reduction. (a) Secondary electron image of the foil surface; (b) backscattered electron image of the cross section with the marked positions for EDS analysis shown in Table 2.
890
880
870
860
850
Binding Energy (eV)
80
75
70 65 Binding Energy (eV)
60
Fig. 6. SR-XPS spectra of the Ni3 Al foils after the H2 reduction measured at TOA = 0◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra.
it is thought to remain and coexist with Al2 O3 within the surface layer in order to form spinel NiAl2 O4 , as described below. With the further progress of the water vapor treatment, both the Ni particles and the remaining Ni were oxidized to NiO in accordance with reaction (2), resulting in the formation of NiO particles and a layer of a
NiOx AlOx
3
Ni3Al (a) As-polished NiO Al(OH)3 Al2O3 +NiAl2O4
Ni3Al (b) Aer water vapor
Ni + Al2O3 Al2O3 +NiAl2O4
Ni3Al (c) Aer H2 reducon
Fig. 7. Schematic illustration of the surface structure evolution of Ni3 Al foil.
mixture of NiO and Al2 O3 . The NiO and Al2 O3 reacted and formed NiAl2 O4 by solid-state diffusion reaction (3) [28]. Furthermore, the Al2 O3 at the top of the surface layer might also react with H2 O in accordance with reaction (4), resulting in the formation of Al(OH)3 , which was observed as the thin dark layer beneath the NiO particles but above the mixed NiAl2 O4 and Al2 O3 layer (Fig. 3b). The surface structure after water vapor treatment is schematically shown in Fig. 7b. Ni + H2 O ⇔ NiO + H2
(2)
NiO + Al2 O3 ⇔ NiAl2 O4
(3)
Al2 O3 + 3H2 O ⇔ 2Al(OH)3
(4)
After the H2 reduction at 873 K, Ni, NiAl2 O4 , and Al2 O3 were identified (Figs. 5 and 6). The NiO particles are thought to have been reduced to Ni particles during the H2 reduction by reaction (5). It was confirmed by a temperature-programmed reduction experiment in our previous study that most of the NiO formed on the Ni3 Al can be reduced by H2 at 873 K [26]. An agglomeration of the Ni particles was thought to have occurred during the H2 reduction, forming the porous-like structure. In addition, the Al(OH)3 is thought to have decomposed to Al2 O3 and H2 O in accordance with reaction (6) during the H2 reduction, considering that Al(OH)3 is not stable at high temperatures without a water vapor atmosphere. The produced Al2 O3 probably coexisted with the Ni particles, resulting in a Ni-enriched porous structure at the outermost surface, as indicated in Fig. 5 and Table 2. In contrast, NiAl2 O4 and Al2 O3 remained unchanged, coexisting in the subsurface layer
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model APSUSC-27360; No. of Pages 6
ARTICLE IN PRESS Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
6
during H2 reduction. The mixture of NiAl2 O4 and Al2 O3 is supposed to serve as the support for metallic Ni particles and at the same time to prevent the further oxidation of the foil. The surface structure after the H2 reduction is schematically shown in Fig. 7c. NiO + H2 ⇔ Ni + H2 O
(5)
2Al(OH)3 ⇔ Al2 O3 + 3H2 O
(6)
As shown above, the surface structure was significantly modified by the water vapor treatment followed by the H2 reduction. The water vapor treatment resulted in the formation of fine NiO particles in high density distributed on the surface layers of Al(OH)3 , Al2 O3 , and NiAl2 O4 . This structure is much different from that of Ni3 Al foils oxidized at 973 K or above in air, in which a continuous granular NiO layer was formed [21]. This difference is attributed to the difference of the oxygen partial pressure and temperature. The oxygen partial pressure in the water vapor treatment was much lower than that in oxidation in air, which should be favorable for the formation of fine NiO particles in high density. A lower oxidation temperature might also be favorable for the formation of fine NiO particles. The formation of fine NiO particles led to the formation of a Ni-enriched porous-like surface structure by the subsequent H2 reduction. The present study reveals that it is possible to modify the surface structure by the water vapor treatment followed by a H2 reduction, forming a Ni-enriched porous-like surface structure which is expected to have enhanced catalytic activity. 4. Conclusions The effect of water vapor treatment at 873 K for 1 h and the subsequent H2 reduction at 873 K for 1 h on the surface structure of Ni3 Al foils has been examined by using SR-XPS and SEM analyses. The results are summarized as follows: (1) Fine NiO particles in high density distributed on the surface layers of Al(OH)3 and Al2 O3 /NiAl2 O4 were formed by the water vapor treatment. A thin Al(OH)3 layer was present beneath the NiO particles and a layer consisting of Al2 O3 and NiAl2 O4 was present beneath the Al(OH)3 layer. (2) The NiO particles were reduced to metallic Ni and the Al(OH)3 was decomposed to Al2 O3 by the subsequent hydrogen reduction, whereas Al2 O3 and NiAl2 O4 remained unchanged, forming a Ni-enriched porous structure on the layer consisting of Al2 O3 and NiAl2 O4 . Acknowledgments The synchrotron XPS experiments were performed using SUREAC 2000 at BL23SU at the SPring-8 facilities under the approval of the Japan Synchrotron Radiation Research Institute (JASRI) and the Japan Atomic Energy Agency (JAEA) (proposals no. 2011B3806, 2012A3806, 2013A3873, and 2013B3873). This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant number 22560769.
References [1] D.P. Pope, S.S. Ezz, Mechanical properties of Ni3 Al and nickel-based alloys with high volume fraction of ␥’, Int. Mater. Rev. 29 (1984) 136–167. [2] N.S. Stoloff, Physical and mechanical metallurgy of Ni3 Al and its alloys, Int. Mater. Rev. 34 (1989) 153–184. [3] M. Yamaguchi, Y. Umakoshi, The deformation behavior of intermetallic superlattice compounds, Prog. Mater. Sci. 34 (1990) 1–148. [4] T. Hirano, T. Mawari, Unidirectional solidification of Ni3 Al by a floating zone method, Acta Metall. Mater. 41 (1993) 1783–1789. [5] M. Demura, Y. Suga, O. Umezawa, K. Kishida, E.P. George, T. Hirano, Fabrication of Ni3 Al thin foil by cold-rolling, Intermetallics 9 (2001) 157–167. [6] M. Demura, K. Kishida, Y. Suga, M. Takanashi, T. Hirano, Fabrication of thin Ni3 Al foils by cold rolling, Scr. Mater. 47 (2002) 267–272. [7] L. Zhou, Y. Gou, M. Yagi, M. Sakurai, H. Kameyama, Investigation of a novel porous anodic alumina plate for methane steam reforming: hydrothermal stability, electrical heating possibility and reforming reactivity, Int. J. Hydrogen Energy 34 (2009) 844–858. [8] D.H. Chun, Y. Xu, M. Demura, K. Kishida, M.H. Oh, T. Hirano, D.M. Wee, Catalytic properties of Ni3 Al foils for methanol decomposition, Catal. Lett. 106 (2006) 71–75. [9] D.H. Chun, Y. Xu, M. Demura, K. Kishida, M.H. Oh, T. Hirano, D.M. Wee, Spontaneous catalytic activation of Ni3 Al thin foils in methanol decomposition, J. Catal. 243 (2006) 99–107. [10] Y. Xu, H. Yoshikawa, J.H. Jang, M. Demura, K. Kobayashi, S. Ueda, Y. Yamashita, D.M. Wee, T. Hirano, Characterization of surface structure evolution in Ni3 Al foil catalysts by hard X-ray photoelectron spectroscopy, J. Phys. Chem. C 114 (2010) 6047–6053. [11] A.M. Venezia, C.M. Loxton, Oxidation of Ni3 Al at low and high oxygen pressures, Surf. Sci. 194 (1988) 136–148. [12] U. Bardi, A. Atrei, G. Rovida, Initial stages of oxidation of the Ni3 Al alloy: structure and composition of the aluminum oxide overlayer studied by XPS, LEIS and LEED, Surf. Sci. 268 (1992) 87–97. [13] R.T. Haasch, A.M. Venezia, C.M. Loxton, The relationship between mass transport and oxide chemistry in oxidation of Ni3 Al alloys, J. Mater. Res. 7 (1992) 1341–1349. [14] M. Haerig, S. Hofmann, Mechanisms of Ni3 Al oxidation between 500 ◦ C and 700 ◦ C, Appl. Surf. Sci. 125 (1998) 99–114. [15] C. Backer, J. Kandler, H. Raaf, R. Linke, T. Pelster, M. Drager, M. Tanemura, K. Wandelt, Oxygen adsorption and oxide formation on Ni3 Al (111), J. Vac. Sci. Technol. A 16 (1998) 1000–1005. [16] A. Rosenhahn, J. Schneider, C. Becker, K. Wandelt, The formation of Al2 O3 -layer on Ni3 Al (111), Appl. Surf. Sci. 142 (1999) 169–173. [17] S.G. Addepalli, B. Ekstrom, N.P. Magtoto, J.S. Lin, J.A. Kelber, STM atomic-scale characterization of the ␥ A-Al2 O3 film on Ni3 Al (111), Surf. Sci. 442 (1999) 385–399. [18] E. Schumann, G. Schnotz, K.P. Trumble, M. Ruhle, Microstructural observations on the oxidation of ␥ -A-Ni3 Al at low oxygen partial pressure, Acta Metall. Mater. 40 (1992) 1311–1319. [19] M. Garza, N.P. Magtoto, J.A. Kelber, Characterization of oxidized Ni3 Al (110) and interaction of the oxide film with water vapor, Surf. Sci. 519 (2002) 259–268. [20] Y. Ma, Y. Xu, M. Demura, T. Hirano, Catalytic stability of Ni3 Al powder for methane steam reforming, Appl. Catal. B 80 (2008) 15–23. [21] J.H. Jang, Y. Xu, M. Demura, D.M. Wee, T. Hirano, Catalytic activity improvement of Ni3 Al foils for methanol decomposition by oxidation–reduction pretreatment, Appl. Catal. A 398 (2011) 161–167. [22] Y. Saitoh, Y. Fukuda, Y. Takeda, H. Yamagami, S. Takahashi, Y. Asano, T. Hara, K. Shirasawa, M. Takeuchi, T. Tanaka, H. Kitamura, Performance upgrade in the JAEA actinide science beamline BL23SU at Spring-8 with a new twin-helical undulator, J. Synchrotron Radiat. 19 (2012) 388–393. [23] Y. Teraoka, A. Yoshigoe, Design of surface chemistry end-station of BL23SU in Spring-8, Jpn. J. Appl. Phys. 38 (Suppl. 38-1) (1999) 642–645. [24] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, Physical Electronics, Inc., Chanhassen, MN, 1995. [25] A. Velon, I. Olefjord, Oxidation behavior of Ni3 Al and Fe3 Al: I. XPS calibrations of pure compounds and quantification of the results, Oxid. Met. 56 (2001) 415–424. [26] A. Velon, I. Olefjord, Oxidation behavior of Ni3 Al and Fe3 Al: II. Early stage of oxide growth, Oxid. Met. 56 (2001) 425–452. [27] M. Bobeth, W. Pompe, E. Schumann, M. Ruhle, Out-diffusion of the noble component during the initial stage of the oxidation of ␥ -Ni3 Al, Acta Metall. Mater. 40 (1992) 2669–2676. [28] J.D. Kuenzly, D.L. Douglass, The oxidation mechanism of Ni3 Al containing yttrium, Oxid. Met. 8 (1974) 139–178.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Applied Surface Science xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil Ya Xu a,∗ , Yan Ma b , Junya Sakurai a , Yuden Teraoka c , Akitaka Yoshigoe c , Masahiko Demura a , Toshiyuki Hirano a a
Hydrogen Materials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan North China Electric Powder University, 2 Beinong Road, Huilongguan, Changping District, Beijing 102206, China c Quantum Beam Science Directorate, Japan Atomic Energy Agency, 1-1-1 Kouto, Sayo-cho, Sayo-gun, Hyogo 679-5148, Japan b
a r t i c l e
i n f o
Article history: Received 30 December 2013 Received in revised form 22 February 2014 Accepted 24 February 2014 Available online xxx Keywords: Ni3 Al foil catalyst Synchrotron radiation X-ray photoemission spectroscopy Water vapor treatment Hydrogen reduction
a b s t r a c t We have developed a water vapor treatment followed by hydrogen reduction to modify the surface structure of Ni3 Al foils in order to obtain high catalytic activity. The Ni3 Al foils were heat treated in water vapor at 873 K for 1 h followed by H2 reduction at 873 K for 1 h. The effects of the water vapor treatment and the H2 reduction on the surface structure of the Ni3 Al foils were investigated by means of scanning electron microscopy and synchrotron radiation X-ray photoemission spectroscopy. Both Ni and Al in the surface layer of the Ni3 Al foil were oxidized during the water vapor treatment; fine NiO particles with a high density were formed on the outermost surface, accompanied by the formation of oxide layers of Al(OH)3 and NiAl2 O4 /Al2 O3 beneath the NiO particles. The NiO particles were reduced to metallic Ni and the Al(OH)3 was decomposed to Al2 O3 , whereas the NiAl2 O4 and Al2 O3 remained unchanged during the subsequent H2 reduction, forming a Ni-enriched porous structure on the surface layer of NiAl2 O4 /Al2 O3 . © 2014 Elsevier B.V. All rights reserved.
1. Introduction The Ni3 Al intermetallic compound is known as a promising high-temperature structural material because of its excellent hightemperature strength and corrosion/oxidation resistance [1–3]. However, its poor room-temperature ductility has been a serious problem. We have overcome this problem and successfully developed thin foils of Ni3 Al with a thickness of less than 30 m by cold rolling unidirectionally solidified ingots [4–6]. Recently, we investigated the catalytic properties of the Ni3 Al foils for methanol decomposition in the temperature range from 513 to 793 K and found that such flat foils show high catalytic activity and selectivity for methanol decomposition into H2 and CO, despite the small surface area of the foils, demonstrating that the Ni3 Al foils can be used as plate type catalysts [7], and thus serve both functionalities of catalytic and structural materials, by which one can make a more efficient reactor for hydrogen production [8,9]. The high catalytic activity for methanol decomposition was attributed to the formation of fine Ni particles on the foil surface through a selective
∗ Corresponding author at: Hydrogen Materials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. Tel.: +81 298592573; fax: +81 298592501. E-mail address: [email protected] (Y. Xu).
oxidation and/or hydroxylation of Al by the small amount of H2 O produced during the reaction [9,10]. This result suggests that water vapor oxidation is effective for modifying the surface morphology and enhancing the catalytic activity of Ni3 Al foils. Many studies have been carried out on the oxidation of Ni3 Al in air or pure O2 [11–17], whereas there have been very few studies on the oxidation of Ni3 Al in water vapor [18,19]. Schumann et al. [18] studied the oxidation behavior of Ni3 Al single crystals under low oxygen partial pressure realized by exposing the Ni3 Al in a flowing H2 /H2 O mixture at 1223 K. Fine Ni particles were observed on the surface of the Ni3 Al after being exposed for 1 min, accompanied by formation of a continuous thin ␥-Al2 O3 oxide scale. Continued oxidation resulted in thickening of the ␥-Al2 O3 scale and coalescing of the Ni particles. However, they did not report the effect of water vapor at temperatures lower than 1223 K and at high water vapor partial pressures. Garza et al. [19] studied the interaction of an oxide film on Ni3 Al with the water vapor, and they revealed that the water vapor can significantly affect the oxide film even at low H2 O pressures. However, they did not mention the oxidation behavior of Ni3 Al caused by water vapor. In this study, we carried out a water vapor treatment of Ni3 Al foils at 873 K, followed by H2 reduction at 873 K, which is commonly used as a pre-reduction process for Ni3 Al catalysts [20,21]. The effect of the water vapor treatment and H2 reduction on the surface structure of the Ni3 Al foils was investigated by means of
http://dx.doi.org/10.1016/j.apsusc.2014.02.144 0169-4332/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model APSUSC-27360; No. of Pages 6
ARTICLE IN PRESS Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
2
scanning electron microscopy (SEM) and synchrotron radiation Xray photoemission spectroscopy (SR-XPS).
2. Experimental 2.1. Foil sample preparation Ni3 Al (Ni-24 at% Al) foils with a thickness of 30 m were fabricated using 98% cold rolling of single crystals [5,6]. Samples for SR-XPS measurement were cut into 7 mm × 7 mm squares and then mechanically polished to get a surface roughness height of less than 0.5 m. After polishing, the samples were washed with ethanol and de-ionized water in an ultrasonic bath.
2.2. Water vapor and H2 reduction treatments Water vapor and H2 reduction treatments were carried out in a conventional fixed-bed flow reactor composed of a quartz tube (internal diameter 8 mm) in an electric furnace connected with gas and liquid supply units [7,8]. The samples were first heated up to 873 K in flowing N2 (30 mL/min) and then water vapor was introduced using a micro-pump at a liquid flow rate of 50 L/min for 1 h with continually flowing N2 (101 kPa total pressure). The partial pressure of the water vapor in the reactor was 68 kPa during the water vapor treatment. After the water vapor treatment, the H2 reduction was carried out at 873 K for 1 h in a flow of mixed H2 (30 mL/min) and N2 (5 mL/min). After the treatments, the samples were cooled down to room temperature in flowing N2 (30 mL/min). Then the samples were taken out and preserved in a desiccator at room temperature for several days until SEM and SR-XPS analyses.
2.3. SR-XPS analysis The SR-XPS measurements were performed on the samples in the as-polished state, after the water vapor treatment, and after the H2 reduction using monochromatic synchrotron radiation with a photon energy of 1486.6 eV at the high-energy resolution soft Xray beam line BL23SU at the synchrotron radiation facility SPring-8 (Japan). A detailed description of the BL23SU and the experimental station used can be found in previous reports [22,23]. The base pressure of the analysis chamber was maintained below 5 × 10−8 Pa during the measurements. The core level spectra of Ni 2p, Al 2p, Ni 3p, O 1s, and C 1s were measured for each sample. In order to obtain the depth profile of the surface structure, the measurements were carried out at different take-off angles (TOA) from 0◦ to 60◦ with respect to the surface normal. The photon energy scale was calibrated by using the peak position of the Au 4f7/2 core level at 84.0 eV [24].
2.4. Surface morphology analysis The morphology of the surface and cross section of the foils in the as-polished state, after the water vapor treatment, and after the H2 reduction was examined by means of scanning electron microscopy (SEM) (JEOL, JSM-7000F) with a field emission gun. The chemical composition of the surface products was analyzed by an energy dispersive X-ray spectroscopy (EDS) system equipped in the SEM. The cross-sectional samples were prepared by means of an argon ion beam cross-section polisher (JEOL, SM09010).
Fig. 1. Backscattered electron image of the cross section of as-polished Ni3 Al foil.
3. Results and discussion 3.1. As-polished samples Fig. 1 shows the SEM backscattered electron (BE) image of the cross section of the as-polished foil. The surface of the foil was macroscopically smooth. No particles or precipitates were observed near the surface. The image contrast shows no obvious change from the surface to the depths of the foil, indicating a uniform elemental composition across the cross section of the foil. Fig. 2 shows the Ni 2p (Fig. 2a), Al 2p and Ni 3p (Fig. 2b) SRXPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ for the as-polished foil. At TOA = 0◦ , a Ni 2p3/2 peak at a binding energy of 853.6 eV and a Ni 2p1/2 peak at 870.8 eV were detected; they were accompanied by satellite peaks. These peaks correspond to metallic Ni [24]. With increasing TOA, the shape of the satellite peaks changed, and the spectral intensity ratio of the satellites relative to the main Ni 2p peaks increased. This change matches the features of Ni oxides [11,19,24], indicating that a small amount of Ni oxide was formed near the surface even though the oxides were not detected by SEM observation. The Ni 3p spectra in Fig. 2b also show a shape change with the increase of TOA. At TOA = 0◦ , a Ni 3p main peak is observed at a binding energy of 67.0 eV with a shoulder on the high binding energy side, which agrees well with the Ni 3p spectra of metallic Ni [10,12]. However, the intensity of the shoulder relative to the main Ni 3p peak increased with the increase of TOA, becoming even larger than the main Ni 3p peak at TOA = 60◦ , which is typical of Ni oxides [14]. This result is in agreement with that of Ni 2p, indicating the formation of Ni oxides on the surface. These Ni oxides are thought to be native Ni oxides (NiOx ) formed during polishing or exposure to air at room temperature. The Al 2p core level showed two chemical states, as shown in Fig. 2b. The major peak located at 73.0 eV corresponds to metallic Al, whereas the minor peak at around 74.5 eV corresponds to Al oxide, probably a native amorphous AlOx [10,14]. With increasing TOA, the intensity of the AlOx peak relative to the metallic Al peak increased, indicating the presence of AlOx on the surface. 3.2. Samples after water vapor treatment Fig. 3a shows the SEM secondary electron (SE) image of the Ni3 Al foil surface after water vapor treatment. A high density of fine particles on the surface was observed. Most of the particles were
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
3
Intensity (Arbitrary units)
Intensity (Arbitrary Units)
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
890
880
870 860 Binding Energy (eV)
850
80
75
70 65 Binding Energy (eV)
60
Fig. 2. SR-XPS spectra of the as-polished Ni3 Al foils measured at TOA = 0◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra.
Table 1 EDS analysis of the chemical compositions of the positions marked in Fig. 3b for the foil after water vapor treatment. Element (At%)
Ni
Al
O
Position A Position B Position C
51.0 26.6 69.1
3.9 28.3 26.5
45.1 45.1 4.5
irregularly shaped, and the size of the particles ranged from several tens to several hundreds of nanometers. The large particles seem to be aggregates of small particles. Fig. 3b shows the SEM BE image of a cross section of the foil after water vapor treatment. Three surface layers were observed. The outermost layer composed of the particles was discontinuous. The second layer just beneath the particles was thin but continuous with a relatively dark contrast. The third layer beneath the dark thin layer was relatively thick and clear grain or phase boundaries can be observed in it. The chemical composition in each layer was examined by EDS analysis. Table 1 shows the chemical compositions of the positions marked in Fig. 3b after the water vapor treatment. The particles on the outermost surface (position A) were composed of Ni and O, with a very small amount of Al, indicating that the particles were NiO. The third layer (position B) had a higher Al concentration and lower Ni concentration as compared to the other layers, suggesting that NiAl2 O4 and/or Al2 O3 might have been formed in this layer. The substrate beneath the third layer (position C) was mainly composed of Ni and Al, with very small amounts of O, and its composition was close to Ni3 Al. The second layer was not examined by EDS analysis, but it is thought to be mainly composed of Al and O since it was relatively dark in
the BE image, considering that the brightness in the BE image arises mainly from the heavier element (Ni). Fig. 4a shows the Ni 2p, and Fig. 4b shows the Ni 3p, and Al 2p SRXPS spectra obtained at TOA = 0◦ , 20◦ , 40◦ , and 60◦ for the foil after water vapor treatment. Both the Ni 2p and the Ni 3p spectra were much different from the corresponding metallic Ni spectra of the as-polished samples shown in Fig. 2. The main peaks of both Ni 2p and Ni 3p appeared at higher binding energies than those of metallic Ni 2p and Ni 3p. In addition, strong satellite peaks of Ni 2p3/2 and Ni 2p1/2 were also observed. These features were related to the presence of Ni oxides. As shown in Fig. 4a, the shape of the main Ni 2p3/2 spectra changed with increasing TOA. At TOA = 0◦ , the main Ni 2p3/2 spectra were composed of three components at 857.5, 855.7, and 853.6 eV. The major component at 857.5 eV corresponds to the spinel NiAl2 O4 , the second component at 855.7 eV corresponds to NiO, and the third component at 853.6 eV corresponds to metallic Ni according to previous reports [25,26]. With increasing TOA, the relative intensity of the NiAl2 O4 component decreased, whereas that of the NiO component increased and became the largest at TOA = 60◦ . This result indicates that NiO was a surface species and NiAl2 O4 was a subsurface species. In addition, the relative intensity of the metallic Ni component decreased with increasing TOA and almost disappeared at TOA = 60◦ , suggesting that this Ni component was not a surface species. Instead, it corresponded to the metallic Ni in the substrate. The Al 2p core level showed a major peak at 75.5 eV at TOA = 0◦ , which corresponds to Al oxides, hydroxides, and spinel [10,24,25], and no metallic Al peak was detected (Fig. 4b). This means that the Al in the surface layer was oxidized and/or hydroxylated during the water vapor treatment. This major peak is thought to correspond to
Fig. 3. SEM images of the Ni3 Al foils after the water vapor treatment. (a) Secondary electron image of the foil surface; (b) backscattered electron image of the cross section with the marked positions for EDS analysis shown in Table 1.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
Intensity (Arbitrary units)
4
890
880
870 860 Binding Energy (eV)
850
80
75
70 65 Binding Energy (eV)
60
Fig. 4. SR-XPS spectra of the Ni3 Al foils after the water vapor treatment measured at TOA = 0◦ , 20◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra. Table 2 EDS analysis of the chemical compositions of the positions marked in Fig. 5b for the foil after the H2 reduction. Element (At%)
Ni
Al
O
Position A Position B Position C
87.3 39.8 69.9
4.4 39.3 24.8
8.4 20.9 5.3
a mixture of an Al2 O3 component at 75.8 eV, an Al(OH)3 component at 75.2 eV, an NiAl2 O4 component and a native AlOx component at 74.6 eV. As shown in Fig. 4b, the major peak slightly shifted to a lower binding energy with increasing TOA. Because the NiAl2 O4 was a subsurface species, the slight shift is probably due to the formation of Al(OH)3 and AlOx , especially Al(OH)3 above Al2 O3 . In addition, the intensity of the Al 2p spectra relative to the Ni 3p spectra decreased with increasing TOA, suggesting that the Al2 O3 and Al(OH)3 were beneath the NiO. 3.3. Samples after H2 reduction The samples treated by water vapor were reduced by flowing H2 at 873 K for 1 h. Fig. 5a shows the SE image of the foil surface after H2 reduction. The particles locally coalesced with each other, forming a porous surface structure. Fig. 5b shows the BE image of the cross section of the foil after H2 reduction. Two layers were formed: an outermost layer, which was much brighter than other regions, and a sub-surface layer, which had a nonuniform thickness. The chemical composition in each layer was measured by EDS (Table 2). The results show that the outermost layer was mainly composed of Ni with a small amount of Al2 O3 , which means that the NiO particles that formed during the water vapor treatment were reduced to metallic Ni by the H2 reduction. The subsurface layer was composed of Ni, Al, and O, suggesting that this layer might be a mixture of NiAl2 O4 and Al2 O3 . On the other hand, the dark thin layer beneath the outermost layer, which was present after the water vapor treatment, was not identified. Fig. 6a shows the Ni 2p SR-XPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ for the foil after H2 reduction. The main Ni 2p3/2 spectra were composed of two components at 857.5 and 853.6 eV. The component at 857.5 eV corresponded to NiAl2 O4 and the other component at 853.6 eV corresponded to metallic Ni. In contrast, a NiO peak at 855.7 eV was extremely weak. This result reveals that most of the NiO was reduced to metallic Ni by the H2 reduction, whereas NiAl2 O4 was not. Fig. 6b shows the Al 2p and Ni 3p SRXPS spectra obtained at TOA = 0◦ , 40◦ , and 60◦ . The Al 2p spectra at TOA = 0◦ showed a major peak at 75.8 eV, which might correspond to a mixture of Al2 O3 , NiAl2 O4 , and AlOx , and no obvious shift of the peak was observed with increasing TOA. This result suggests that
no Al(OH)3 remained on the surface after H2 reduction. Considering that the thin dark layer that formed during the water vapor treatment (Fig. 3b) disappeared after the H2 reduction (Fig. 5b), it is thought that Al(OH)3 was probably present in this dark layer. In contrast, Al2 O3 might not show an isolated layer; it may instead coexist with NiAl2 O4 in the subsurface layer or with Ni in the outermost layer, which was suggested by the SEM observation (Fig. 5b) and EDS analysis results (Table 2). The Al 2p spectra relative to the Ni 3p spectra showed no obvious change with increasing TOA, indicating that there was no obvious separation between Al2 O3 and NiAl2 O4 (Ni) along the depth direction. This result is in agreement with the above SEM and EDS results, suggesting that Al2 O3 coexisted with NiAl2 O4 in the subsurface layer and with Ni in the outermost layer. 3.4. Surface structure evolution during the water vapor and H2 treatments The above results show that both Ni and Al in the surface layer of the Ni3 Al foil were oxidized or hydroxylated by the water vapor treatment, resulting in the formation of NiO particles on the outermost surface, accompanied by oxide layers of Al(OH)3 , NiAl2 O4 , and Al2 O3 beneath the particles. By the subsequent H2 reduction, the NiO was reduced to metallic Ni, forming a porous structure of Ni on the outermost surface. On the basis of these results, the evolution of the surface structure caused by water vapor treatment and H2 reduction is discussed below. For the as-polished samples, a thin layer of native Al oxide, AlOx , with a small amount of native Ni oxide (NiOx ) was present on the surface of the foil (Fig. 2). These native oxides are thought to have formed when the samples were exposed in air. The thickness of such native AlOx had been estimated to be approximately 13 A˚ [10]. This surface structure in the as-polished state is illustrated in Fig. 7a. After water vapor treatment, NiO, NiAl2 O4 , Al2 O3 , and Al(OH)3 were detected (Figs. 3 and 4). Al2 O3 can be formed more easily than NiO, considering that the dissociation pressure of Al2 O3 is approximately 10−40 Pa, much lower than that of NiO, which is approximately 10−14 Pa at 873 K. It is thought that the Al in Ni3 Al was preferentially oxidized to Al2 O3 at the beginning of the water vapor treatment in accordance with reaction (1). 2Ni3 Al + 3H2 O ⇔ 6Ni + Al2 O3 + H2
(1)
This process also led to the formation of metallic Ni. Then, the metallic Ni diffused toward the outside, forming the fine Ni particles on the outermost surface. The origin of the out-diffusion of Ni has been discussed by Bobeth et al. [27] and was attributed to a relaxation of internal stresses produced by the formation of Al2 O3 . We assume that not all the metallic Ni diffused outside; some of
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model
ARTICLE IN PRESS
APSUSC-27360; No. of Pages 6
Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
5
Intensity (Arbitrary units)
Intensity (Arbitrary units)
Fig. 5. SEM images of the Ni3 Al foils after the H2 reduction. (a) Secondary electron image of the foil surface; (b) backscattered electron image of the cross section with the marked positions for EDS analysis shown in Table 2.
890
880
870
860
850
Binding Energy (eV)
80
75
70 65 Binding Energy (eV)
60
Fig. 6. SR-XPS spectra of the Ni3 Al foils after the H2 reduction measured at TOA = 0◦ , 40◦ , and 60◦ . (a) Ni 2p spectra; (b) Al 2p and Ni 3p spectra.
it is thought to remain and coexist with Al2 O3 within the surface layer in order to form spinel NiAl2 O4 , as described below. With the further progress of the water vapor treatment, both the Ni particles and the remaining Ni were oxidized to NiO in accordance with reaction (2), resulting in the formation of NiO particles and a layer of a
NiOx AlOx
3
Ni3Al (a) As-polished NiO Al(OH)3 Al2O3 +NiAl2O4
Ni3Al (b) Aer water vapor
Ni + Al2O3 Al2O3 +NiAl2O4
Ni3Al (c) Aer H2 reducon
Fig. 7. Schematic illustration of the surface structure evolution of Ni3 Al foil.
mixture of NiO and Al2 O3 . The NiO and Al2 O3 reacted and formed NiAl2 O4 by solid-state diffusion reaction (3) [28]. Furthermore, the Al2 O3 at the top of the surface layer might also react with H2 O in accordance with reaction (4), resulting in the formation of Al(OH)3 , which was observed as the thin dark layer beneath the NiO particles but above the mixed NiAl2 O4 and Al2 O3 layer (Fig. 3b). The surface structure after water vapor treatment is schematically shown in Fig. 7b. Ni + H2 O ⇔ NiO + H2
(2)
NiO + Al2 O3 ⇔ NiAl2 O4
(3)
Al2 O3 + 3H2 O ⇔ 2Al(OH)3
(4)
After the H2 reduction at 873 K, Ni, NiAl2 O4 , and Al2 O3 were identified (Figs. 5 and 6). The NiO particles are thought to have been reduced to Ni particles during the H2 reduction by reaction (5). It was confirmed by a temperature-programmed reduction experiment in our previous study that most of the NiO formed on the Ni3 Al can be reduced by H2 at 873 K [26]. An agglomeration of the Ni particles was thought to have occurred during the H2 reduction, forming the porous-like structure. In addition, the Al(OH)3 is thought to have decomposed to Al2 O3 and H2 O in accordance with reaction (6) during the H2 reduction, considering that Al(OH)3 is not stable at high temperatures without a water vapor atmosphere. The produced Al2 O3 probably coexisted with the Ni particles, resulting in a Ni-enriched porous structure at the outermost surface, as indicated in Fig. 5 and Table 2. In contrast, NiAl2 O4 and Al2 O3 remained unchanged, coexisting in the subsurface layer
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144
G Model APSUSC-27360; No. of Pages 6
ARTICLE IN PRESS Y. Xu et al. / Applied Surface Science xxx (2014) xxx–xxx
6
during H2 reduction. The mixture of NiAl2 O4 and Al2 O3 is supposed to serve as the support for metallic Ni particles and at the same time to prevent the further oxidation of the foil. The surface structure after the H2 reduction is schematically shown in Fig. 7c. NiO + H2 ⇔ Ni + H2 O
(5)
2Al(OH)3 ⇔ Al2 O3 + 3H2 O
(6)
As shown above, the surface structure was significantly modified by the water vapor treatment followed by the H2 reduction. The water vapor treatment resulted in the formation of fine NiO particles in high density distributed on the surface layers of Al(OH)3 , Al2 O3 , and NiAl2 O4 . This structure is much different from that of Ni3 Al foils oxidized at 973 K or above in air, in which a continuous granular NiO layer was formed [21]. This difference is attributed to the difference of the oxygen partial pressure and temperature. The oxygen partial pressure in the water vapor treatment was much lower than that in oxidation in air, which should be favorable for the formation of fine NiO particles in high density. A lower oxidation temperature might also be favorable for the formation of fine NiO particles. The formation of fine NiO particles led to the formation of a Ni-enriched porous-like surface structure by the subsequent H2 reduction. The present study reveals that it is possible to modify the surface structure by the water vapor treatment followed by a H2 reduction, forming a Ni-enriched porous-like surface structure which is expected to have enhanced catalytic activity. 4. Conclusions The effect of water vapor treatment at 873 K for 1 h and the subsequent H2 reduction at 873 K for 1 h on the surface structure of Ni3 Al foils has been examined by using SR-XPS and SEM analyses. The results are summarized as follows: (1) Fine NiO particles in high density distributed on the surface layers of Al(OH)3 and Al2 O3 /NiAl2 O4 were formed by the water vapor treatment. A thin Al(OH)3 layer was present beneath the NiO particles and a layer consisting of Al2 O3 and NiAl2 O4 was present beneath the Al(OH)3 layer. (2) The NiO particles were reduced to metallic Ni and the Al(OH)3 was decomposed to Al2 O3 by the subsequent hydrogen reduction, whereas Al2 O3 and NiAl2 O4 remained unchanged, forming a Ni-enriched porous structure on the layer consisting of Al2 O3 and NiAl2 O4 . Acknowledgments The synchrotron XPS experiments were performed using SUREAC 2000 at BL23SU at the SPring-8 facilities under the approval of the Japan Synchrotron Radiation Research Institute (JASRI) and the Japan Atomic Energy Agency (JAEA) (proposals no. 2011B3806, 2012A3806, 2013A3873, and 2013B3873). This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant number 22560769.
References [1] D.P. Pope, S.S. Ezz, Mechanical properties of Ni3 Al and nickel-based alloys with high volume fraction of ␥’, Int. Mater. Rev. 29 (1984) 136–167. [2] N.S. Stoloff, Physical and mechanical metallurgy of Ni3 Al and its alloys, Int. Mater. Rev. 34 (1989) 153–184. [3] M. Yamaguchi, Y. Umakoshi, The deformation behavior of intermetallic superlattice compounds, Prog. Mater. Sci. 34 (1990) 1–148. [4] T. Hirano, T. Mawari, Unidirectional solidification of Ni3 Al by a floating zone method, Acta Metall. Mater. 41 (1993) 1783–1789. [5] M. Demura, Y. Suga, O. Umezawa, K. Kishida, E.P. George, T. Hirano, Fabrication of Ni3 Al thin foil by cold-rolling, Intermetallics 9 (2001) 157–167. [6] M. Demura, K. Kishida, Y. Suga, M. Takanashi, T. Hirano, Fabrication of thin Ni3 Al foils by cold rolling, Scr. Mater. 47 (2002) 267–272. [7] L. Zhou, Y. Gou, M. Yagi, M. Sakurai, H. Kameyama, Investigation of a novel porous anodic alumina plate for methane steam reforming: hydrothermal stability, electrical heating possibility and reforming reactivity, Int. J. Hydrogen Energy 34 (2009) 844–858. [8] D.H. Chun, Y. Xu, M. Demura, K. Kishida, M.H. Oh, T. Hirano, D.M. Wee, Catalytic properties of Ni3 Al foils for methanol decomposition, Catal. Lett. 106 (2006) 71–75. [9] D.H. Chun, Y. Xu, M. Demura, K. Kishida, M.H. Oh, T. Hirano, D.M. Wee, Spontaneous catalytic activation of Ni3 Al thin foils in methanol decomposition, J. Catal. 243 (2006) 99–107. [10] Y. Xu, H. Yoshikawa, J.H. Jang, M. Demura, K. Kobayashi, S. Ueda, Y. Yamashita, D.M. Wee, T. Hirano, Characterization of surface structure evolution in Ni3 Al foil catalysts by hard X-ray photoelectron spectroscopy, J. Phys. Chem. C 114 (2010) 6047–6053. [11] A.M. Venezia, C.M. Loxton, Oxidation of Ni3 Al at low and high oxygen pressures, Surf. Sci. 194 (1988) 136–148. [12] U. Bardi, A. Atrei, G. Rovida, Initial stages of oxidation of the Ni3 Al alloy: structure and composition of the aluminum oxide overlayer studied by XPS, LEIS and LEED, Surf. Sci. 268 (1992) 87–97. [13] R.T. Haasch, A.M. Venezia, C.M. Loxton, The relationship between mass transport and oxide chemistry in oxidation of Ni3 Al alloys, J. Mater. Res. 7 (1992) 1341–1349. [14] M. Haerig, S. Hofmann, Mechanisms of Ni3 Al oxidation between 500 ◦ C and 700 ◦ C, Appl. Surf. Sci. 125 (1998) 99–114. [15] C. Backer, J. Kandler, H. Raaf, R. Linke, T. Pelster, M. Drager, M. Tanemura, K. Wandelt, Oxygen adsorption and oxide formation on Ni3 Al (111), J. Vac. Sci. Technol. A 16 (1998) 1000–1005. [16] A. Rosenhahn, J. Schneider, C. Becker, K. Wandelt, The formation of Al2 O3 -layer on Ni3 Al (111), Appl. Surf. Sci. 142 (1999) 169–173. [17] S.G. Addepalli, B. Ekstrom, N.P. Magtoto, J.S. Lin, J.A. Kelber, STM atomic-scale characterization of the ␥ A-Al2 O3 film on Ni3 Al (111), Surf. Sci. 442 (1999) 385–399. [18] E. Schumann, G. Schnotz, K.P. Trumble, M. Ruhle, Microstructural observations on the oxidation of ␥ -A-Ni3 Al at low oxygen partial pressure, Acta Metall. Mater. 40 (1992) 1311–1319. [19] M. Garza, N.P. Magtoto, J.A. Kelber, Characterization of oxidized Ni3 Al (110) and interaction of the oxide film with water vapor, Surf. Sci. 519 (2002) 259–268. [20] Y. Ma, Y. Xu, M. Demura, T. Hirano, Catalytic stability of Ni3 Al powder for methane steam reforming, Appl. Catal. B 80 (2008) 15–23. [21] J.H. Jang, Y. Xu, M. Demura, D.M. Wee, T. Hirano, Catalytic activity improvement of Ni3 Al foils for methanol decomposition by oxidation–reduction pretreatment, Appl. Catal. A 398 (2011) 161–167. [22] Y. Saitoh, Y. Fukuda, Y. Takeda, H. Yamagami, S. Takahashi, Y. Asano, T. Hara, K. Shirasawa, M. Takeuchi, T. Tanaka, H. Kitamura, Performance upgrade in the JAEA actinide science beamline BL23SU at Spring-8 with a new twin-helical undulator, J. Synchrotron Radiat. 19 (2012) 388–393. [23] Y. Teraoka, A. Yoshigoe, Design of surface chemistry end-station of BL23SU in Spring-8, Jpn. J. Appl. Phys. 38 (Suppl. 38-1) (1999) 642–645. [24] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, Physical Electronics, Inc., Chanhassen, MN, 1995. [25] A. Velon, I. Olefjord, Oxidation behavior of Ni3 Al and Fe3 Al: I. XPS calibrations of pure compounds and quantification of the results, Oxid. Met. 56 (2001) 415–424. [26] A. Velon, I. Olefjord, Oxidation behavior of Ni3 Al and Fe3 Al: II. Early stage of oxide growth, Oxid. Met. 56 (2001) 425–452. [27] M. Bobeth, W. Pompe, E. Schumann, M. Ruhle, Out-diffusion of the noble component during the initial stage of the oxidation of ␥ -Ni3 Al, Acta Metall. Mater. 40 (1992) 2669–2676. [28] J.D. Kuenzly, D.L. Douglass, The oxidation mechanism of Ni3 Al containing yttrium, Oxid. Met. 8 (1974) 139–178.
Please cite this article in press as: Y. Xu, et al., Effect of water vapor and hydrogen treatments on the surface structure of Ni3 Al foil, Appl. Surf. Sci. (2014), http://dx.doi.org/10.1016/j.apsusc.2014.02.144