Probing Surface Information of Alloy by Time of Flight-Secondary Ion Mass Spectrometer

: In recent years, time of ﬂight-secondary ion mass spectrometer (ToF-SIMS) has been widely employed to acquire surface information of materials. Here, we investigated the alloy surface by combining the mass spectra and 2D mapping images of ToF-SIMS. We found by surprise that these two results seem to be inconsistent with each other. Therefore, other surface characteristic tools such as SEM-EDS were further used to provide additional supports. The results indicated that such differences may originate from the variance of secondary ion yields, which might be affected by crystal orientation.


Introduction
The rapid development of advanced experimental methods has made surface analysis a popular research topic. Obtaining the surface/interface information of materials/devices is extremely important, as the quality of surface/interface could directly determine the material/device performance. Solar cells based on hybrid perovskite have been considered as one of the most promising candidates for next generation of photovoltaics [1][2][3]. However, perovskite solar cells exhibited instability problems even under encapsulation [4][5][6][7], i.e., the surface of function layer began to degrade within only hundreds of hours, which limited their practical applications. To resolve this, understanding the degradation mechanism at the surface and improving its stability become imperative and important [8,9]. Moreover, in the battery research, investigating the surface components of cathodes or electrodes could provide kinetic information about the charging/discharging process [10,11], which may help with improving the cell efficiency and developing new materials for batteries. Surface analysis has also been applied into other fields including bioelectronics [12], polymers [13], catalysts [14][15][16], etc.
Nowadays, X-ray photoelectron spectroscopy (XPS) and ToF-SIMS have been considered as the two most powerful tools to probe surface characteristics of different materials. The working principle of XPS is quite straightforward, where the sample is bombarded by X-rays to emit photoelectrons. The binding energy of the photoelectron could then be determined by subtracting its kinetic energy from the total energy [17]. The working principle of ToF-SIMS is also complicated. Typically, a pulsed primary ion beam is employed to bombard the sample surface, which leads to the yields of positive and negative second ions. These secondary ions are then collected into the analyzer by applying high voltage potential and their mass is determined by measuring their time of flight from the sample to the detector [17,18]. The ion mass and 2D imaging could thus be obtained to probe the component and its distribution at the surface of the sample [19,20]. In addition, the excellent sensitivity [17], variable depth probing [21], high spatial resolution and mass range [22] further make ToF-SIMS a popular technique to acquire surface information of materials. In recent years, ToF-SIMS has been widely employed in nano-and bio-materials to investigate their surface and bulk properties [20,[22][23][24][25][26][27][28]. In contrast, employing the mass spectra and 2D mapping to analyze the alloy surface has been rarely reported [19,29], which needs to be investigated.
It is, therefore, the purpose of the current report, to investigate the alloy surface characteristics by combining the mass spectra and 2D mapping functions of ToF-SIMS. Both surface components and their spatial distribution could be evaluated. In addition, various regions of interest (ROI) were selected and extracted to compare their compositional difference. However, the results obtained from the mass spectrum, which suggested there is no obvious difference of composition among ROIs, seems to be contradictory with the mapping images. To reveal its reason, SEM-EDS was further conducted, and the results indicated an even distribution of all elements, corresponding well with the mass spectrum. Such phenomenon could be ascribed to their various crystal orientations, which might induce different ion yields of secondary ions, thereby appearing color differences in mapping of various crystals. This work may expand the application of ToF-SIMS to other fields such as probing the sample crystal orientation.

Materials
254. SMO austenitic stainless steel (batch: 00Cr20Ni18Mo6CuN) (Jiangsu Haixin steel Inc., Taizhou, China). The specimen was polished with 2000 grit mesh and then electrically polished by the twin-jet electropolishing method in a solution of 5% perchloric acid. Before EDS and ToF-SIMS measurements, sputtering was performed to clean the sample surface. The cleaning was conducted using a 3 kV 100 nA Argon gas beam (area: 600 µm × 600 µm; time: 2 min).

Characterizations
The scanning electron microscopy (SEM) measurements were carried out on a Regulus 8230 instrument (Hitachi Inc., Naka, Ibaraki, Japan) with EDS function. X-ray diffraction (XRD) patterns were obtained using a Rigaku Smartlab diffractometer with Cu Kα as the X-ray source. ToF-SIMS measurements were conducted with a PHI nano ToF II (ULVAC-PHI Inc., Chigasaki, Kanagawa, Japan), where a 30 kV Bi + pulsed primary ion beam was used for analysis.

XRD
The crystal structure of the alloy sample was investigated by XRD measurements. As shown in Figure 1, the alloy crystallized into a face-centered cubic (FCC) structure. The diffraction peaks could be assigned to (111), (200), (220) and (311) planes of austenite, respectively, as confirmed by its corresponding Joint Committee on Powder Diffraction Standards (JCPDS) card.

Mass Spectra
Next, TOF-SIMS was employed to investigate the components of the sample. It is worth mentioning here that selecting proper ion-beam parameters to conduct the measurements is necessary and important. Both the primary and sputtering ion beams could influence the acquired data [30][31][32] by varying their damage depth, cluster sizes, sputtering rate, etc. For instance, C 60 + and Ar + ion beams could reveal the component distribution of perovskite solar cells more accurately than O 2 + beam and Ar-gas cluster ion beam [31] (Ar-GCIB). In the current work, Bi + was employed as the primary ion source due to its higher Crystals 2021, 11, 1465 3 of 9 monatomic ion yields and better mapping spatial resolution than Bi 3 + . The sample was sputtered employing an Ar + ion beam to remove oxides and contaminants on the surface.

Mass Spectra
Next, TOF-SIMS was employed to investigate the components of the sample. It worth mentioning here that selecting proper ion-beam parameters to conduct the mea urements is necessary and important. Both the primary and sputtering ion beams cou influence the acquired data [30][31][32] by varying their damage depth, cluster sizes, sputte ing rate, etc. For instance, C60 + and Ar + ion beams could reveal the component distributio of perovskite solar cells more accurately than O2 + beam and Ar-gas cluster ion beam [3 (Ar-GCIB). In the current work, Bi + was employed as the primary ion source due to higher monatomic ion yields and better mapping spatial resolution than Bi3 + . The samp was sputtered employing an Ar + ion beam to remove oxides and contaminants on t surface.
The sample surface was etched before ToF-SIMS measurements to remove any o ide/contaminant at the sample surface. Since the probing depth of ToF-SIMS is only with 2 nm, the existence of oxides and contaminants may affect the results. Both positive an negative ions were collected, and the mass spectrum was presented in Figure 2. To ca brate the ToF-SIMS spectra, m/z peaks at 15.0235 (CH3), 27.0235 (C2H3) and 41.03 (C3H5) were used for the calibration of positive ions spectra, while m/z peaks at 13.00 (CH), 17.0027 (OH) and 25.0078 (C2H) were selected for the calibration of negative io spectra. In general, the spectra of both positive ions and negative ions should be measur to acquire complete/full compositional information. From the negative spectral data of t sample, elements including F/N/Cl/S and oxides such as Cr-O/Fe-O/Mo-O were detecte while for positive spectral data, Fe/Ni/Cr, etc., were probed. The spectrum collected fro positive ions shows more information about metallic elements, while non-metallic el ments were observed from the negative ion mass spectrum. The alloy mainly contain metallic elements including iron (Fe), nickel (Ni), chromium (Cr), molybdenum (Mo), s dium (Na) and calcium (Ca), as well as fluorine (F) and chlorine (Cl) as non-metallic el ments. Moreover, the compound information may also be revealed, which is another a vantage of ToF-SIMS. As can be observed, chemical compounds such as CrO, FeO an HSiO3 were also formed at the surface of the sample. The formation of oxides could attributed to the adsorption of carbon and oxygen atoms on the sample surface when was placed in air. It should be mentioned here that the intensity of each ion in the spect can reveal their contents qualitatively. The sample surface was etched before ToF-SIMS measurements to remove any oxide/contaminant at the sample surface. Since the probing depth of ToF-SIMS is only within 2 nm, the existence of oxides and contaminants may affect the results. Both positive and negative ions were collected, and the mass spectrum was presented in Figure 2. To calibrate the ToF-SIMS spectra, m/z peaks at 15.0235 (CH 3 ), 27.0235 (C 2 H 3 ) and 41.0391 (C 3 H 5 ) were used for the calibration of positive ions spectra, while m/z peaks at 13.0078 (CH), 17.0027 (OH) and 25.0078 (C 2 H) were selected for the calibration of negative ions spectra. In general, the spectra of both positive ions and negative ions should be measured to acquire complete/full compositional information. From the negative spectral data of the sample, elements including F/N/Cl/S and oxides such as Cr-O/Fe-O/Mo-O were detected, while for positive spectral data, Fe/Ni/Cr, etc., were probed. The spectrum collected from positive ions shows more information about metallic elements, while non-metallic elements were observed from the negative ion mass spectrum. The alloy mainly contained metallic elements including iron (Fe), nickel (Ni), chromium (Cr), molybdenum (Mo), sodium (Na) and calcium (Ca), as well as fluorine (F) and chlorine (Cl) as non-metallic elements. Moreover, the compound information may also be revealed, which is another advantage of ToF-SIMS. As can be observed, chemical compounds such as CrO, FeO and HSiO 3 were also formed at the surface of the sample. The formation of oxides could be attributed to the adsorption of carbon and oxygen atoms on the sample surface when it was placed in air. It should be mentioned here that the intensity of each ion in the spectra can reveal their contents qualitatively.

2D Mapping Image
To acquire the surface composition distribution, the 2D mapping images of the sample were further analyzed. To investigate the distribution of each ion, individual color scale bars were employed for both positive and negative modes. Figure 3 illustrates the mapping results collected from positive ions, where grains with different colors could be observed. This color difference might be originated from the variation of surface components, as reported by previous literature [19,23]. In this work, a brighter color of same ion indicates a higher content of the component. In addition, impurities such as Na was found to be concentrated at certain spots, while other components (Fe, Cr, Ni, Mo) were uniformly distributed in the grain. Such distribution difference may vary their properties, which need to be investigated in future.

2D Mapping Image
To acquire the surface composition distribution, the 2D mapping images of ple were further analyzed. To investigate the distribution of each ion, individ scale bars were employed for both positive and negative modes. Figure 3 illust mapping results collected from positive ions, where grains with different colors observed. This color difference might be originated from the variation of surfac nents, as reported by previous literature [19,23]. In this work, a brighter color of indicates a higher content of the component. In addition, impurities such as Na w to be concentrated at certain spots, while other components (Fe, Cr, Ni, Mo) w formly distributed in the grain. Such distribution difference may vary their pr which need to be investigated in future.

2D Mapping Image
To acquire the surface composition distribution, the 2D mapping images of the sample were further analyzed. To investigate the distribution of each ion, individual color scale bars were employed for both positive and negative modes. Figure 3 illustrates the mapping results collected from positive ions, where grains with different colors could be observed. This color difference might be originated from the variation of surface components, as reported by previous literature [19,23]. In this work, a brighter color of same ion indicates a higher content of the component. In addition, impurities such as Na was found to be concentrated at certain spots, while other components (Fe, Cr, Ni, Mo) were uniformly distributed in the grain. Such distribution difference may vary their properties, which need to be investigated in future. ToF-SIMS negative ion maps of the sample were also presented ( Figure 4) to reveal more component distribution information. As can be observed, negative ion components ToF-SIMS negative ion maps of the sample were also presented ( Figure 4) to reveal more component distribution information. As can be observed, negative ion components (F, O, Cl) exhibited high ion intensity (bright color), while positive ion such as Fe was barely detected when employing negative ion collectors. To investigate the component variation of grains with different colors, ROI measurements were conducted and discussed in the next section.
(F, O, Cl) exhibited high ion intensity (bright color), while positive ion such as Fe was barely detected when employing negative ion collectors. To investigate the component variation of grains with different colors, ROI measurements were conducted and discussed in the next section.

Region of Interest (ROI)
One of the advantages of ToF-SIMS technique is that it can extract ROIs from the mapping images to compare their component variations. Three ROIs derived from positive ion mapping were first demonstrated in Figure 5a. Although the ion intensity (color) is different in three regions, no obvious differences of substrate elemental composition (Cr, Fe, Ni, Mo, etc.) among ROI-1/2/3 were detected, as revealed by their mass spectrum (Figure 5b-d). The above contradiction between mapping images and the mass spectrum was also found in negative ion results ( Figure 6). We therefore resolved to other surface characteristic techniques to obtain its element distribution maps.

Region of Interest (ROI)
One of the advantages of ToF-SIMS technique is that it can extract ROIs from the mapping images to compare their component variations. Three ROIs derived from positive ion mapping were first demonstrated in Figure 5a. Although the ion intensity (color) is different in three regions, no obvious differences of substrate elemental composition (Cr, Fe, Ni, Mo, etc.) among ROI-1/2/3 were detected, as revealed by their mass spectrum (Figure 5b-d). The above contradiction between mapping images and the mass spectrum was also found in negative ion results ( Figure 6). We therefore resolved to other surface characteristic techniques to obtain its element distribution maps.

Region of Interest (ROI)
One of the advantages of ToF-SIMS technique is that it can extract ROIs from the mapping images to compare their component variations. Three ROIs derived from positive ion mapping were first demonstrated in Figure 5a. Although the ion intensity (color) is different in three regions, no obvious differences of substrate elemental composition (Cr, Fe, Ni, Mo, etc.) among ROI-1/2/3 were detected, as revealed by their mass spectrum (Figure 5b-d). The above contradiction between mapping images and the mass spectrum was also found in negative ion results ( Figure 6). We therefore resolved to other surface characteristic techniques to obtain its element distribution maps.

SEM-EDS
To make a comparison, the morphology and element distribution of the sample were measured via SEM-EDS. As shown in Figure 7, no grain was observed from planar SEM image. The EDS results (Figure 8) indicated that the sample is mainly composed of Fe, Ni,

SEM-EDS
To make a comparison, the morphology and element distribution of the sample were measured via SEM-EDS. As shown in Figure 7, no grain was observed from planar SEM image. The EDS results (Figure 8) indicated that the sample is mainly composed of Fe, Ni, Mo and Cr, which is consistent with the mass spectrum results. In addition, the existence of Na was not detected in EDS results, which indicated the excellent detective sensitivity of ToF-SIMS (Figure 3). The distribution of elements was demonstrated in Figure 7. All elements were distributed evenly, which is inconsistent with 2D mapping results. To verify, 2D mapping images of positive ions normalized to total ion density were measured and presented in Figure S1, which showed the same distribution as Figure 3 results. To investigate the possible topographical effects of the sample, AFM measurements were further conducted. As shown in Figure S2, the sample exhibited a smooth surface with an average roughness of~2.23 nm, which could exclude the topographic effect of the sample.

SEM-EDS
To make a comparison, the morphology and element distribution of the sample were measured via SEM-EDS. As shown in Figure 7, no grain was observed from planar SEM image. The EDS results (Figure 8) indicated that the sample is mainly composed of Fe, Ni, Mo and Cr, which is consistent with the mass spectrum results. In addition, the existence of Na was not detected in EDS results, which indicated the excellent detective sensitivity of ToF-SIMS (Figure 3). The distribution of elements was demonstrated in Figure 7. All elements were distributed evenly, which is inconsistent with 2D mapping results. To verify, 2D mapping images of positive ions normalized to total ion density were measured and presented in Figure S1, which showed the same distribution as Figure 3 results. To investigate the possible topographical effects of the sample, AFM measurements were further conducted. As shown in Figure S2, the sample exhibited a smooth surface with an average roughness of ~2.23 nm, which could exclude the topographic effect of the sample.  Since the SIMS intensity could be described by the following equation [23]: = where is the intensity of primary ion, Y is the ion sputter yield, is the ionizatio probability, is the transmission efficiency of the detection system, and refer t the species isotopic abundance and its fractional concentration in the material, respec tively. During the measurements, and should remain unchanged since the exper Since the SIMS intensity could be described by the following equation [23]: where I p is the intensity of primary ion, Y is the ion sputter yield, α A is the ionization probability, η A is the transmission efficiency of the detection system, θ A and X A refer to the species isotopic abundance and its fractional concentration in the material, respectively. During the measurements, I p and η A should remain unchanged since the experimental conditions were not altered. Previous literature [23,30] suggested that the sputtering yield and ionization probability could be affected by the matrix, i.e., different matrixes may lead to various secondary ion intensities even though they possess the same composition. As mentioned earlier, topographical effects of 2D images could be excluded, as confirmed by SEM and AFM results. Meanwhile, ROI and EDS results indicated that there is no compositional difference, which indicated the same θ A and X A . In addition, normalized data of Fe and Ni were also analyzed to exclude topographic effects. Similar mapping results were obtained. We thus hypothesized that such a difference of 2D imaging and EDS results might be ascribed to various crystal orientations of the polycrystalline sample, which would affect bonding strengths and the arrangement of surface atoms, resulting in different secondary ion intensities. Figure 9 schematically illustrate such process. Future investigation on its specific mechanism needs to be conducted. Since the SIMS intensity could be described by the following equation [23]: = where is the intensity of primary ion, Y is the ion sputter yield, is the ionization probability, is the transmission efficiency of the detection system, and refer to the species isotopic abundance and its fractional concentration in the material, respectively. During the measurements, and should remain unchanged since the experimental conditions were not altered. Previous literature [23,30] suggested that the sputtering yield and ionization probability could be affected by the matrix, i.e., different matrixes may lead to various secondary ion intensities even though they possess the same composition. As mentioned earlier, topographical effects of 2D images could be excluded, as confirmed by SEM and AFM results. Meanwhile, ROI and EDS results indicated that there is no compositional difference, which indicated the same and . In addition, normalized data of Fe and Ni were also analyzed to exclude topographic effects. Similar mapping results were obtained. We thus hypothesized that such a difference of 2D imaging and EDS results might be ascribed to various crystal orientations of the polycrystalline sample, which would affect bonding strengths and the arrangement of surface atoms, resulting in different secondary ion intensities. Figure 9 schematically illustrate such process. Future investigation on its specific mechanism needs to be conducted.

Discussion
In conclusion, ToF-SIMS was employed to investigate the surface characteristics of an alloy sample. By acquiring its mass spectra and 2D mapping images, the surface components and their inhomogeneous distribution were revealed. To make comparison, three ROIs with different chromatic aberrations were selected. However, contradictory results of component distribution were obtained from their mass spectrum and mapping images. To resolve it, SEM-EDS was further performed, and the results suggested a homogeneous distribution of all elements at the surface, which is consistent with the mass spectrum results. We hypothesized that other possible factors such as crystal orientation may cause

Discussion
In conclusion, ToF-SIMS was employed to investigate the surface characteristics of an alloy sample. By acquiring its mass spectra and 2D mapping images, the surface components and their inhomogeneous distribution were revealed. To make comparison, three ROIs with different chromatic aberrations were selected. However, contradictory results of component distribution were obtained from their mass spectrum and mapping images. To resolve it, SEM-EDS was further performed, and the results suggested a homogeneous distribution of all elements at the surface, which is consistent with the mass spectrum results. We hypothesized that other possible factors such as crystal orientation may cause any influence on the mapping image appearance by affecting the ion yields of secondary ions. Our findings provide deeper insights into the working principle of ToF-SIMS, and may expand its application in other fields such as crystal orientation.