XPS Investigation of the Oxidation States of the As-Deposited Ta Films Prepared by Magnetron Sputtering Technology

Due to their versatile and unique properties, tantalum-based thin films have been extensively studied. However, tantalum is susceptible to oxidation due to its higher chemical activity, which is crucial regardless of whether oxidations of Ta are beneficial or detrimental. Therefore, the oxidation of Ta during material processing, especially without conscious means, should be taken seriously. In this study, pure Ta films were fabricated by magnetron sputtering under set procedure parameters. The effects of base pressure and substrate temperature on the degree of oxidation of Ta films were investigated. The results revealed that the magnitude of the base pressure directly affects the oxidation state of the as-deposited Ta films. When preferably avoiding the oxidation of sputtered Ta films, the base pressure should be controlled below 4.4 × 10−4 Pa. The substrate temperature has little effect on the oxidation state of the as-deposited Ta films under a base pressure ranging from about 10−2 Pa to 10−4. We hope that this study can provide some references for controlling the oxidation states of Ta involved in relevant film preparation.


Introduction
In recent years, tantalum-based materials have attracted considerable attention due to their high corrosion resistance [1][2][3][4][5][6], unique electrical properties [7,8], and excellent in vitro and in vivo biocompatibility [9][10][11].These properties endow Ta-based materials with the possibility for extended applications.Due to the limitations arising from high melting points, manufacturing costs, and resource scarcity, Ta-based films may be the most important form for the application of tantalum compared to the bulk material of tantalum.Furthermore, it is possible to achieve some fantastic combinations of properties by preparing Ta-based films on the surface of other substrate materials.
It is well known that Ta has high chemical activity, and it easily bonds with oxygen and carbon atoms to form oxides/carbides.For Ta-based films, a superficial layer formed with a stable oxidized status on the film's surface is a key factor in their good mechanical properties and corrosion resistance.The oxidation status of tantalumin Ta-based films also has a major impact on the electrical, optical, and photocatalytic performance of related films.For example, D. Cristea et al. [12] prepared Ta oxynitride thin films with various configurations by magnetron sputtering.Structures ranging from tetragonal β-Ta to fcc-Ta (O,N) to tantalum-based (Ta x O y , Ta x N y , and/or Ta x O y N z ) crystallites in an amorphous matrix have been obtained.The electrical resistivity varies in a very large domain, starting from 5.29 × 10 −4 Ωcm up to 1.93 × 10 6 Ωcm.However, on the other hand, the formation Materials 2023, 16, 7405 2 of 10 of oxide layers is not always desired for applications of Ta-based films.It is reasonable, in some cases, to protect Ta from oxidation.
So far, several techniques used for preparing Ta-based films have been reported, involving magnetron sputtering [13,14], ion beam assisted deposition (IBAD) [15], filtered cathodic vacuum arc deposition (FCVAD) [16], and chemical vapor deposition (CVD) [17].Either way, the residual oxygen in the deposition chamber may lead to partial oxidation of the as-deposited Ta films to some extent.This oxidation induced by residual gas could be negligible for some Ta films' applications, while for other applications, it must be strictly controlled.Although the unintentional result of partial oxidation with respect to as-deposited Ta-based films was often reported in the related studies, there have been few reports elaborating on the effects of deposition parameters on the oxidation status of as-deposited Ta films.
In this study, we deposited pure Ta films by magnetron sputtering.The base pressures and temperatures of the substrate were elaborately controlled during the films' deposition.The crystal structure and phase compositions of the as-grown sputtered Ta films were examined by X-ray diffraction (XRD).The surface and sub-surface chemistry status of the prepared samples was analyzed using X-ray photoelectron spectroscopy (XPS) in detail.Then the effect of depositing parameters on the oxidation state of as-deposited Ta films was studied.The purpose of this study is to provide a basis for regulating process parameters aiming to control the oxidation state of Ta-based films prepared by vacuum vapor deposition.

Deposition Conditions
A series of pure Ta films with different thicknesses were deposited by the RF magnetron sputtering technique.The vacuum system affiliated with this device is mainly composed of a cylindrical vacuum chamber with a diameter of 500 mm and a height of 300 mm, a mechanical pump, and a molecular pump in series.Si (100) wafers with dimensions of about 1 cm × 1 cm, which have a thin native oxide surface layer, were used as substrates.Before film preparation, all substrates were cleaned ultrasonically in alcohol and acetone for 15 min in turn, then blow-dried with nitrogen and quickly transferred into the deposition chamber.
Firstly, the vacuum chamber was evacuated to the target base pressure using a mechanical and a molecular pump in series.After that, all samples were sputter-etched under a negative bias of 300 V in an Ar plasma with a pressure of 8.0 Pa for 20 min to remove surface contaminants.Then, the target was pre-sputtered for 10 min with the shutter closed to clean the target surface for every deposition batch.Finally, opening the shutter, the Ta films were deposited.Additionally, the substrate temperature was set to room temperature and 150 • C, respectively.The substrates were stationary during film deposition, and the argon used in all processes of sample preparation had a purity of 99.9%.A Ta target of 3 inches in diameter and a purity better than 99.9 wt.% was used.All depositions were carried out for 1 h.The main parameters corresponding to the samples (labeled for T x ) are detailed in Table 1.

Analyses
The crystal structure and phase composition of the as-deposited thin films were analyzed using an X-ray diffraction instrument (XRD; Philips, X'Pert Rro, Philips, Amsterdam, The Netherlands) using Cu Kα incident radiation.Grazing-incidence XRD inspection of the film samples was performed in the 2 theta mode in the range between 20 • and 90 • in the step of 0.05 • .The angle of incidence was fixed at 2 • .The excitation voltage and current were set at 45 kV and 40 mA, respectively.
The chemistry and oxidation states of the as-deposited films were investigated using X-ray photoelectron spectroscopy (XPS; PHI5702 Quantera Scaning X-ray Microprobe, Physical Electronics (PHI), Chanhassen, MN, USA) with monochromatic Al Kα radiation as the excitation source at 14.9 keV and 10 mA.All XPS data were calibrated to the adventitious C1S peak present at 284.8 eV as the reference.The pressure during XPS analysis was better than 2.0 × 10 −7 Pa.The XPS peak fitting was performed using the Thermo Scientific (Waltham, MA, USA) Avantage v5.9678 software, which models the Gauss-Lorenzian contributions after background subtraction (Shirely).
It is noted that, in order to reduce the impact of atmospheric oxygen as much as possible, two measures were taken.Firstly, each sample was transferred immediately from the deposition vacuum chamber to the XPS analysis vacuum chamber after film deposition was completed.Secondly, the surface and sub-surface were analyzed, respectively, to obtain more accurate information regarding the oxidation states of the Ta films.Here, the surface refers to the topmost surface of the as-deposited films.The sub-surface refers to the region ~15 nm beneath the topmost surface of the as-deposited films, which were prepared by an ion-etching step using an Argon gun of 3 kV for each XPS measurement.Additionally, the morphologies of the films were characterized using a field emission scanning electron microscope (FESEM; Hitachi SU8020, Hitachi, Tokyo, Japan).

Structural Analysis
The microstructures of the Ta films deposited under different conditions on Si wafer substrates were characterized using FESEM.The cross-sectional images are shown in Figure 1.Relatively packed columnar structures were observed for all films.The values of thickness for T 1 , T 2 , T 4 , and T 5 were 1.75 µm, 1.74 µm, 1.63 µm, and 1.51 µm, respectively.It is noted that the thickness of T 3 has considerably decreased to 0.86 µm.This may be attributed to an increase in the mean free path of the molecular structure due to lower base pressure, which enhances the resputtering effect of the deposited particles.

Analyses
The crystal structure and phase composition of the as-deposited thin films were analyzed using an X-ray diffraction instrument (XRD; Philips, X'Pert Rro, Philips, Amsterdam, Nederland) using Cu Kα incident radiation.Grazing-incidence XRD inspection of the film samples was performed in the 2 theta mode in the range between 20° and 90° in the step of 0.05°.The angle of incidence was fixed at 2°.The excitation voltage and current were set at 45 kV and 40 mA, respectively.
The chemistry and oxidation states of the as-deposited films were investigated using X-ray photoelectron spectroscopy (XPS; PHI5702 Quantera Scaning X-ray Microprobe, Physical Electronics (PHI), Chanhassen, MN, USA) with monochromatic Al Kα radiation as the excitation source at 14.9 keV and 10 mA.All XPS data were calibrated to the adventitious C1S peak present at 284.8 eV as the reference.The pressure during XPS analysis was better than 2.0 × 10 −7 Pa.The XPS peak fitting was performed using the Thermo Scientific (Waltham, MA, USA) Avantage v5.9678 software, which models the Gauss-Lorenzian contributions after background subtraction (Shirely).
It is noted that, in order to reduce the impact of atmospheric oxygen as much as possible, two measures were taken.Firstly, each sample was transferred immediately from the deposition vacuum chamber to the XPS analysis vacuum chamber after film deposition was completed.Secondly, the surface and sub-surface were analyzed, respectively, to obtain more accurate information regarding the oxidation states of the Ta films.Here, the surface refers to the topmost surface of the as-deposited films.The sub-surface refers to the region ~15 nm beneath the topmost surface of the as-deposited films, which were prepared by an ion-etching step using an Argon gun of 3 kV for each XPS measurement.Additionally, the morphologies of the films were characterized using a field emission scanning electron microscope (FESEM; Hitachi SU8020, Hitachi, Tokyo, Japan).

Structural Analysis
The microstructures of the Ta films deposited under different conditions on Si wafer substrates were characterized using FESEM.The cross-sectional images are shown in Figure 1.Relatively packed columnar structures were observed for all films.The values of thickness for T1, T2, T4, and T5 were 1.75 µm, 1.74 µm, 1.63 µm, and 1.51 µm, respectively.It is noted that the thickness of T3 has considerably decreased to 0.86 µm.This may be attributed to an increase in the mean free path of the molecular structure due to lower base pressure, which enhances the resputtering effect of the deposited particles.Figure 2 shows the X-ray diffractograms of Ta films deposited on silicon substrates with different deposition parameters.The XRD patterns contain distinct (110), ( 200), (211), and (220) α-Ta reflections, indicating that all the as-deposited films are composed of the α-Ta phase, which is the stable body-centered cubic crystal structure (bcc).The β-phase, which is the tetragonal metastable structure of Ta, was not detected in the as-deposited Ta films in our study.The presented XRD result is consistent with that of the thick Ta-based coating prepared by chemical vapor deposition (CVD) on polished AISI 316L stainless steel [3].No peaks corresponding to oxide phases were detected from the XRD scan in all cases, indicating that no obvious oxidation occurred for the deposited Ta films when the base pressure ranged from 10 −2 to 10 −4 Pa.In addition, the relative intensities of the diffraction peaks for the T 3 sample obviously decreased compared to the patterns of other samples.The reason for the decrease has not been fully determined, and further research may be required.Figure 2 shows the X-ray diffractograms of Ta films deposited on silicon substrates with different deposition parameters.The XRD patterns contain distinct (110), ( 200), (211), and (220) α-Ta reflections, indicating that all the as-deposited films are composed of the α-Ta phase, which is the stable body-centered cubic crystal structure (bcc).The β-phase, which is the tetragonal metastable structure of Ta, was not detected in the as-deposited Ta films in our study.The presented XRD result is consistent with that of the thick Ta-based coating prepared by chemical vapor deposition (CVD) on polished AISI 316L stainless steel [3].No peaks corresponding to oxide phases were detected from the XRD scan in all cases, indicating that no obvious oxidation occurred for the deposited Ta films when the base pressure ranged from 10 −2 to 10 −4 Pa.In addition, the relative intensities of the diffraction peaks for the T3 sample obviously decreased compared to the patterns of other samples.The reason for the decrease has not been fully determined, and further research may be required.

XPS Characterization
Figure 3 presents the Ta 4f spectra measured from the surfaces and sub-surfaces of the as-deposited films.Using the known binding energies for the metallic tantalum doublet and the well-reported peak positions of the Ta2O5 doublet, both pairs of peaks shown in Figure 3a,b can be identified.The contributions (positioned at around 21 eV/23 eV) at the lower binding energy doublet are assigned to the Ta 4f7/2 and Ta 4f5/2 peaks of metallic tantalum Ta 0 .The contributions (positioned at around 26 eV/28 eV) at the higher binding energy doublet correspond to the Ta 4f7/2 and Ta 4f5/2 peaks of tantalum pentoxide (Ta2O5) Ta 5+ .The two pairs of Ta 0 and Ta 5+ peaks discerned clearly from the Ta 4f spectra shown in Figure 3a indicate that both the oxidation and metallic states of tantalum are involved in the surfaces of all the as-deposited Ta films, i.e., partial oxidation of Ta occurred for the surfaces of the Ta films deposited under conditions differing by base pressures and substrate temperatures.But the intensities of the Ta 0 state relative to those of the Ta 5+ state increase gradually with decreasing base pressure.Although the effect of exposure to the atmosphere cannot be completely avoided, the variations with respect to the relative intensities of different chemical states of Ta for the T1, T2, and T3 spectra suggest that the content of residual oxygen existing in the films' deposition chamber would directly have an effect on the surface oxidation degree of the as-deposited Ta films.Interestingly, there

XPS Characterization
Figure 3 presents the Ta 4f spectra measured from the surfaces and sub-surfaces of the as-deposited films.Using the known binding energies for the metallic tantalum doublet and the well-reported peak positions of the Ta 2 O 5 doublet, both pairs of peaks shown in Figure 3a,b can be identified.The contributions (positioned at around 21 eV/23 eV) at the lower binding energy doublet are assigned to the Ta 4f 7/2 and Ta 4f 5/2 peaks of metallic tantalum Ta 0 .The contributions (positioned at around 26 eV/28 eV) at the higher binding energy doublet correspond to the Ta 4f 7/2 and Ta 4f 5/2 peaks of tantalum pentoxide (Ta 2 O 5 ) Ta 5+ .The two pairs of Ta 0 and Ta 5+ peaks discerned clearly from the Ta 4f spectra shown in Figure 3a indicate that both the oxidation and metallic states of tantalum are involved in the surfaces of all the as-deposited Ta films, i.e., partial oxidation of Ta occurred for the surfaces of the Ta films deposited under conditions differing by base pressures and substrate temperatures.But the intensities of the Ta 0 state relative to those of the Ta 5+ state increase gradually with decreasing base pressure.Although the effect of exposure to the atmosphere cannot be completely avoided, the variations with respect to the relative intensities of different chemical states of Ta for the T 1 , T 2 , and T 3 spectra suggest that the content of residual oxygen existing in the films' deposition chamber would directly have an effect on the surface oxidation degree of the as-deposited Ta films.Interestingly, there were no significant changes in the ratio of Ta 0 peak intensities to Ta 5+ peak intensities in the samples deposited under differing substrate temperatures.Therefore, it can be inferred that, with a comparable base pressure level, the substrate temperature would not have a noticeable effect on the oxidation degree of the surface of the as-deposited Ta film.
were no significant changes in the ratio of Ta 0 peak intensities to Ta 5+ peak intensities in the samples deposited under differing substrate temperatures.Therefore, it can be inferred that, with a comparable base pressure level, the substrate temperature would not have a noticeable effect on the oxidation degree of the surface of the as-deposited Ta film.For the Ta 4f spectra measured from the sub-surfaces with a ~15 nm depth, there is a pronounced decrease in Ta 5+ peak intensities compared to those from the surfaces of corresponding samples.This implies that the oxidation degree of all the samples decreases significantly with increasing depth in all cases.It is important to note that, for the T3 sample, the Ta 5+ doublet located at a binding energy ranging from about 25 eV to 29 eV was no longer observed.This reveals that once the base pressure is lower than 4.4 × 10 −4 Pa, the residual oxygen in the deposition chamber may have no effect on the deposited Ta films.On the other hand, based on the consistent transfer procedure for all the samples, it could be deduced that the effects of atmospheric exposure are likely to be limited to a depth of ~15 nm.
In order to further explore the detailed differences of the oxidation states of the asdeposited Ta films, the deconvolutions of all Ta 4f spectra were conducted with a fixed condition where the spin-orbit doublet separation was maintained at 1.9 ± 0.1 eV for all Ta chemical states and the intensity ratios between Ta 4f5/2 and Ta 4f7/2 peak at 3/4. Figure 4 shows the typical fitting results (corresponding to the T3 and T5 samples).The other deconvolution results of the Ta 4f spectra relating to the remaining samples are presented in an attached document.The chemical information of all samples, including peak position and peak area, is also given in the attachment (Supplementary Table S1).According to the deconvolution results, the chemical composition of Ta atoms with different valence states for all the studied samples is reported in Table 2.For the Ta 4f spectra measured from the sub-surfaces with a ~15 nm depth, there is a pronounced decrease in Ta 5+ peak intensities compared to those from the surfaces of corresponding samples.This implies that the oxidation degree of all the samples decreases significantly with increasing depth in all cases.It is important to note that, for the T 3 sample, the Ta 5+ doublet located at a binding energy ranging from about 25 eV to 29 eV was no longer observed.This reveals that once the base pressure is lower than 4.4 × 10 −4 Pa, the residual oxygen in the deposition chamber may have no effect on the deposited Ta films.On the other hand, based on the consistent transfer procedure for all the samples, it could be deduced that the effects of atmospheric exposure are likely to be limited to a depth of ~15 nm.
In order to further explore the detailed differences of the oxidation states of the asdeposited Ta films, the deconvolutions of all Ta 4f spectra were conducted with a fixed condition where the spin-orbit doublet separation was maintained at 1.9 ± 0.1 eV for all Ta chemical states and the intensity ratios between Ta 4f 5/2 and Ta 4f 7/2 peak at 3/4. Figure 4 shows the typical fitting results (corresponding to the T 3 and T 5 samples).The other deconvolution results of the Ta 4f spectra relating to the remaining samples are presented in an attached document.The chemical information of all samples, including peak position and peak area, is also given in the attachment (Supplementary Table S1).According to the deconvolution results, the chemical composition of Ta atoms with different valence states for all the studied samples is reported in Table 2.    [15] is the most stable oxide state.In addition to Ta 5+ , the other possible sub-oxide reduction states relating to tantalum may also be present, including Ta 4+ (TaO2) [25.4-26.0eV]/[27.3-27.9eV], Ta 3+ (Ta2O3) [23.0-23.6 eV]/[25.5-24.9eV], Ta 2+ (TaO) [24.1-24.5 eV]/[26.0-26.4eV], and Ta 1+ (Ta2O) [22.4-22.6 eV]/[24.3-24.5 eV] [18], respectively.In order to gain a more accurate peak fitting, all chemical states may be taken into consideration when dealing with Ta 4f spectra as a function of the oxidation states of tantalum.However, the lower oxidation states of tantalum can be generated during Ar + bombardment of XPS analysis resulting from the preferential sputtering of O from Ta2O5 [18].On the other hand, good fittings of Ta 4f envelopes for all the studied samples were obtained when only Ta 5+ and Ta 0 doublets were involved.Therefore, in order to avoid overinterpreting the related data, Ta n+ corresponding to sub-oxidation states of Ta were excluded in our analyses of the chemical states.
According to the peak fitting results of the Ta 4f spectra, the chemical states of tantalum from surfaces and sub-surfaces of all the films were analyzed quantitatively.The nor- , respectively.In order to gain a more accurate peak fitting, all chemical states may be taken into consideration when dealing with Ta 4f spectra as a function of the oxidation states of tantalum.However, the lower oxidation states of tantalum can be generated during Ar + bombardment of XPS analysis resulting from the preferential sputtering of O from Ta 2 O 5 [18].On the other hand, good fittings of Ta 4f envelopes for all the studied samples were obtained when only Ta 5+ and Ta 0 doublets were involved.Therefore, in order to avoid overinterpreting the related data, Ta n+ corresponding to sub-oxidation states of Ta were excluded in our analyses of the chemical states.
According to the peak fitting results of the Ta 4f spectra, the chemical states of tantalum from surfaces and sub-surfaces of all the films were analyzed quantitatively.The normalized oxidation degree was defined as the atomic percentage of Ta 5+ atoms.The oxidation degrees of the surface and sub-surface of each sample are plotted in Figure 5, where the total tantalum atoms are normalized at at. 100%.There are three points to note: (i) for sputtered Ta films deposited under different parameters, the variation trend of oxidation degrees of tantalum on the surface is basically consistent with that of the sub-surface; (ii) a base pressure lower than 4.4 × 10 −4 Pa should be maintained for sputtered Ta films to be free of oxidation caused by residual oxygen; (iii) the oxidation degrees of the surfaces and sub-surfaces of the samples deposited under heating conditions are comparable to those of the un-heating samples deposited under the same base pressure, respectively.
further studied in more detail.Additionally, the oxidation degrees of the sub-surfaces of the T1, T2, T4, and T5 samples were 13.34, 12.21, 12.14, and 10.86 at.%, respectively.There are no considerable differences between the oxidation degrees of these films.Furthermore, these values are much lower compared to those of the surfaces.This indicates that the overall oxidation levels of the films are not high, which is consistent with the results of the XRD scan showing no reflections corresponding to oxides.6a in stack mode, it was observed that the intensity of the O 1s peak at the sub-surface of T3 exhibited a very low intensity.This is consistent with the metallic state (Ta 0 ) for the sub-surface of the T3 sample verified by XPS analysis, as shown in Figure 4d.The O 1s spectra of the surfaces of T3 and T5, as shown in Figure 6b,c, differ by the presence of a high-energy shoulder.The peak located at ~532.1 eV is assigned to hydroxylic groups and/or H2O adsorbed at the sample surface, and the peak located at ~530.3 eV is assigned to Ta-O bonds in tantalum oxides.The characteristics of the O 1s spectra are very similar to those obtained in previous studies [7].In addition, for the Ta 4f and O 1s spectra of each sample, one can see that all peaks were shifted slightly towards a high binding energy after Ar + etching was conducted for sub-surface XPS measurements.During the XPS experiments, the negative charge continuously removed from the surface region as a result of a photoelectric effect has to be replenished at a sufficiently high rate to preserve charge neutrality.If this condition is not fulfilled, the surface acquires positive It is unexpected that the temperature rise of the substrate had no obvious effects on the oxide states of sputtered Ta films.Temperature increases may give rise to the desorption of materials' surfaces or interfaces.But on the other hand, the chemical reactivity of residual oxygen species should be enhanced by increased temperature.The two effects on the oxidation of deposited films may inhibit each other.Thus, the effects of the substrate temperature on the oxidation states of sputtered Ta films are not definitive and should be further studied in more detail.Additionally, the oxidation degrees of the sub-surfaces of the T 1 , T 2 , T 4 , and T 5 samples were 13.34, 12.21, 12.14, and 10.86 at.%, respectively.There are no considerable differences between the oxidation degrees of these films.Furthermore, these values are much lower compared to those of the surfaces.This indicates that the overall oxidation levels of the films are not high, which is consistent with the results of the XRD scan showing no reflections corresponding to oxides.
Figure 6 shows the corresponding O 1s spectra of the T 3 and T 5 samples.Comparing the O 1s peaks shown in Figure 6a in stack mode, it was observed that the intensity of the O 1s peak at the sub-surface of T 3 exhibited a very low intensity.This is consistent with the metallic state (Ta 0 ) for the sub-surface of the T 3 sample verified by XPS analysis, as shown in Figure 4d.The O 1s spectra of the surfaces of T 3 and T 5 , as shown in Figure 6b,c, differ by the presence of a high-energy shoulder.The peak located at ~532.1 eV is assigned to hydroxylic groups and/or H 2 O adsorbed at the sample surface, and the peak located at ~530.3 eV is assigned to Ta-O bonds in tantalum oxides.The characteristics of the O 1s spectra are very similar to those obtained in previous studies [7].In addition, for the Ta 4f and O 1s spectra of each sample, one can see that all peaks were shifted slightly towards a high binding energy after Ar + etching was conducted for sub-surface XPS measurements.During the XPS experiments, the negative charge continuously removed from the surface region as a result of a photoelectric effect has to be replenished at a sufficiently high rate to preserve charge neutrality.If this condition is not fulfilled, the surface acquires positive potential, which decreases the kinetic energy of escaping photoelectrons and, in consequence, leads to the apparent shift of all core-level peaks towards higher binding energy [19].Grzegorz Greczynski and Lars Hultman discussed in detail the relevant issues with using the C 1s peak for binding energy calibration [19,20].As for our experiments, the perfectly accurate determination of the binding energy of Ta and O from XPS spectra is of little significance to this work and is beyond the scope of this work.potential, which decreases the kinetic energy of escaping photoelectrons and, in consequence, leads to the apparent shift of all core-level peaks towards higher binding energy [19].Grzegorz Greczynski and Lars Hultman discussed in detail the relevant issues with using the C 1s peak for binding energy calibration [19,20].As for our experiments, the perfectly accurate determination of the binding energy of Ta and O from XPS spectra is of little significance to this work and is beyond the scope of this work.

Conclusions
In this study, Ta films with different thicknesses were prepared using RF magnetron sputtering equipment under experimental conditions differing by base pressure and substrate temperature.The crystal structures of the as-deposited Ta films were studied.Especially, the chemical valence states of tantalum in the surface and sub-surface layers of the prepared films were analyzed in detail using XPS.The results show that: 1.The as-deposited Ta films sputtered under different conditions are composed of the α-Ta phase, and none of the films undergo significant oxidation.2. The residual oxygen concentration in the deposition chamber directly affects the oxidation levels of the Ta films' surfaces, but the oxidation depth of the film is generally limited to about 15 nm below the surface.As the residual oxygen concentration in the chamber decreases, the oxidation degrees of the surfaces of Ta films also show a significant decrease.When the base pressure of the deposition chamber is lower than 4.4 × 10 −4 Pa, the Ta films are almost unaffected by the residual oxygen.In addition, it was shown that the substrate temperature has little effect on the oxidation states of the as-deposited Ta films under a base pressure ranging from about 10 −2 to 10 −4 Pa.

Conclusions
In this study, Ta films with different thicknesses were prepared using RF magnetron sputtering equipment under experimental conditions differing by base pressure and substrate temperature.The crystal structures of the as-deposited Ta films were studied.Especially, the chemical valence states of tantalum in the surface and sub-surface layers of the prepared films were analyzed in detail using XPS.The results show that: 1.
The as-deposited Ta films sputtered under different conditions are composed of the α-Ta phase, and none of the films undergo significant oxidation.

2.
The residual oxygen concentration in the deposition chamber directly affects the oxidation levels of the Ta films' surfaces, but the oxidation depth of the film is generally limited to about 15 nm below the surface.As the residual oxygen concentration in the chamber decreases, the oxidation degrees of the surfaces of Ta films also show a significant decrease.When the base pressure of the deposition chamber is lower than 4.4 × 10 −4 Pa, the Ta films are almost unaffected by the residual oxygen.In addition, it was shown that the substrate temperature has little effect on the oxidation states of the as-deposited Ta films under a base pressure ranging from about 10 −2 to 10 −4 Pa.

Figure 1 .
Figure 1.FESEM micrographs of the cross-section of Ta films deposited under different conditions, (a) T 1 , (b) T 2 , (c) T 3 , (d) T 4 , and (e) T 5 .The thickness of films is revealed by yellow line.

Figure 2 .
Figure 2. XRD patterns of the Ta films deposited under set parameters.The characteristic lattice planes are marked by asterisk "*".

Figure 2 .
Figure 2. XRD patterns of the Ta films deposited under set parameters.The characteristic lattice planes are marked by asterisk "*".

Figure 3 .
Figure 3. XPS spectra of the Ta 4f spectrum acquired from (a) the surfaces and (b) the sub-surfaces of the Ta films deposited under different parameters.

Figure 3 .
Figure 3. XPS spectra of the Ta 4f spectrum acquired from (a) the surfaces and (b) the sub-surfaces of the Ta films deposited under different parameters.

Figure 4 .
Figure 4. Fitting of the Ta 4f spectra for the (a) T 5 surface, (b) T 3 surface, (c) T5 sub-surface, and (d) T3 sub-surface.The fitting line are colored.It is well known that Ta 5+ (Ta 2 O 5 ) [26.5-27.5 eV]/[28.4-29.4eV] [15] is the most stable oxide state.In addition to Ta 5+ , the other possible sub-oxide reduction states relating to tantalum may also be present, including Ta 4+ (TaO 2 ) [25.4-26.0eV]/[27.3-27.9eV], Ta 3+ (Ta 2 O 3 ) [23.0-23.6 eV]/[25.5-24.9eV], Ta 2+ (TaO) [24.1-24.5 eV]/[26.0-26.4eV], and Ta 1+ (Ta 2 O) [22.4-22.6 eV]/[24.3-24.5 eV] [18], respectively.In order to gain a more accurate peak fitting, all chemical states may be taken into consideration when dealing with Ta 4f spectra as a function of the oxidation states of tantalum.However, the lower oxidation states of tantalum can be generated during Ar + bombardment of XPS analysis resulting from the preferential sputtering of O from Ta 2 O 5[18].On the other hand, good fittings of Ta 4f envelopes for all the studied samples were obtained when only Ta 5+ and Ta 0 doublets were involved.Therefore, in order to avoid overinterpreting the related data, Ta n+ corresponding to sub-oxidation states of Ta were excluded in our analyses of the chemical states.According to the peak fitting results of the Ta 4f spectra, the chemical states of tantalum from surfaces and sub-surfaces of all the films were analyzed quantitatively.The normalized oxidation degree was defined as the atomic percentage of Ta 5+ atoms.The oxidation degrees of the surface and sub-surface of each sample are plotted in Figure5, where the total tantalum atoms are normalized at at. 100%.There are three points to note: (i) for sputtered Ta films deposited under different parameters, the variation trend of oxidation degrees of tantalum on the surface is basically consistent with that of the sub-surface; (ii) a base pressure lower than 4.4 × 10 −4 Pa should be maintained for sputtered Ta films to be free of oxidation caused by residual oxygen; (iii) the oxidation degrees of the surfaces and sub-surfaces of the samples deposited under heating conditions are comparable to those of the un-heating samples deposited under the same base pressure, respectively.

Figure 5 .
Figure 5.The oxidation degrees of the surfaces and sub-surfaces of the as-deposited Ta films.

Figure 6
Figure6shows the corresponding O 1s spectra of the T3 and T5 samples.Comparing the O 1s peaks shown in Figure6ain stack mode, it was observed that the intensity of the O 1s peak at the sub-surface of T3 exhibited a very low intensity.This is consistent with the metallic state (Ta 0 ) for the sub-surface of the T3 sample verified by XPS analysis, as shown in Figure4d.The O 1s spectra of the surfaces of T3 and T5, as shown in Figure6b,c, differ by the presence of a high-energy shoulder.The peak located at ~532.1 eV is assigned to hydroxylic groups and/or H2O adsorbed at the sample surface, and the peak located at ~530.3 eV is assigned to Ta-O bonds in tantalum oxides.The characteristics of the O 1s spectra are very similar to those obtained in previous studies[7].In addition, for the Ta 4f and O 1s spectra of each sample, one can see that all peaks were shifted slightly towards a high binding energy after Ar + etching was conducted for sub-surface XPS measurements.During the XPS experiments, the negative charge continuously removed from the surface region as a result of a photoelectric effect has to be replenished at a sufficiently high rate to preserve charge neutrality.If this condition is not fulfilled, the surface acquires positive

Figure 5 .
Figure 5.The oxidation degrees of the surfaces and sub-surfaces of the as-deposited Ta films.

Figure 6 .
Figure 6.O 1s spectra of (a) T3 and T5 with a stack mode, (b) T5 at the surface, (c) T3 at the surface, and (d) T5 at the sub-surface.The fitting lines of O 1s spectra are colored.

Figure 6 .
Figure 6.O 1s spectra of (a) T 3 and T 5 with a stack mode, (b) T 5 at the surface, (c) T 3 at the surface, and (d) T 5 at the sub-surface.The fitting lines of O 1s spectra are colored.

Table 1 .
Deposition conditions of the sputtered tantalum films.

Table 2 .
The chemical composition (at.%) of Ta atoms with different valence states.

Table 2 .
The chemical composition (at.%) of Ta atoms with different valence states.