*4.4. Electrochemical Testing Results*

*4.4. Electrochemical Testing Results*

in Figure 12a.

Figure 12a shows the potentiodynamic polarization curves of tungsten in different pH solutions. The composition of electrolyte prepared at different pH values was consistent with that of the polishing slurry except that there was no oxidant. Figure 12b shows the corrosion potential (Ecorr) and the corrosion current density (Icorr) of tungsten at different pH solutions, which could be obtained from the potentiodynamic polarization curves in Figure 12a.

Figure 12a shows the potentiodynamic polarization curves of tungsten in different pH solutions. The composition of electrolyte prepared at different pH values was con‐ sistent with that of the polishing slurry except that there was no oxidant. Figure 12b shows the corrosion potential (Ecorr) and the corrosion current density (Icorr) of tungsten at differ‐ ent pH solutions, which could be obtained from the potentiodynamic polarization curves

**Figure 12.** (**a**) Dynamic potential polarization curves at different pH solutions. (**b**) The corrosion

potential Ecorr and the corrosion current density Icorr at different pH solutions.

**Figure 11.** The tungsten samples before and after polishing.

vol.% H2O2; (**b**) 1.5 vol.% H2O2; (**c**) 2.0 vol.% H2O2.

vol.% H2O2; (**b**) 1.5 vol.% H2O2; (**c**) 2.0 vol.% H2O2.

*Micromachines* **2022**, *13*, x FOR PEER REVIEW 10 of 17

pits, or other defects.

pits, or other defects.

**Figure 11.** The tungsten samples before and after polishing. **Figure 11.** The tungsten samples before and after polishing. in Figure 12a.

**Figure 12.** (**a**) Dynamic potential polarization curves at different pH solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different pH solutions. **Figure 12.** (**a**) Dynamic potential polarization curves at different pH solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different pH solutions.

of the tungsten sample after C‐SDP achieved a mirror effect without obvious scratches,

**Figure 10.** Surface morphologies of polished tungsten under different H2O2 concentrations: (**a**) 1.0

of the tungsten sample after C‐SDP achieved a mirror effect without obvious scratches,

**Figure 10.** Surface morphologies of polished tungsten under different H2O2 concentrations: (**a**) 1.0

**Figure 12.** (**a**) Dynamic potential polarization curves at different pH solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different pH solutions. The more negative the corrosion potential is, the more easily the tungsten surface is corroded [30]. The Ecorr gradually tended to negative values with increasing pH values, as shown in Figure 12b. Among the six pH values, the Ecorr was the largest (−268 mV) at pH = 7, which indicated that tungsten was the least susceptible to corrosion at this time. The Ecorr was the smallest (−590 mV) at pH = 12, and tungsten was most easily corroded at this time. These were consistent with the polishing results shown in Figure 6; the surface quality of tungsten was the worst when pH = 12. In electrochemical experiments, the Icorr is used to represent the corrosion rate of the workpiece, which can often reflect the change of *MRR* in the polishing process. As shown in Figure 12b, the Icorr increased with increases in the pH values, which meant the corrosion rate of tungsten increased gradually. The maximum Icorr was 15.01 µA when pH = 12, which was two orders of magnitude higher than that at pH = 7 (0.1185 µA). This phenomenon corresponded to the maximum *MRR* under the condition of pH = 12 shown in Figure 5a. According to the values of Icorr and Ecorr, with increases in pH values, tungsten was more easily corroded, and the corrosion rate increased in the alkaline solution. This phenomenon was consistent with the effect of different pH values on the polishing performance of C-SDP.

Figure 13a shows the potentiodynamic polarization curves of tungsten in different H2O<sup>2</sup> concentration solutions, and four typical H2O<sup>2</sup> concentrations of 0, 0.5 vol.%, 1 vol.%, and 2 vol.% were selected. Except for the different concentrations of H2O2, the other components of solution were the same as those of the polishing slurry. Figure 13b shows the corrosion potential (Ecorr) and the corrosion current density (Icorr) of tungsten in different H2O<sup>2</sup> concentration solutions, which could be obtained from the potentiodynamic polarization curves shown in Figure 13a. As shown in Figure 13b, the Ecorr of tungsten under the H2O<sup>2</sup> concentration of 1 vol.% (68 mV) was higher than that under the H2O<sup>2</sup> concentrations of 0 vol.% (−372 mV), 0.5 vol.% (−9 mV), and 2 vol.% (−172 mV). This indicated that when

the H2O<sup>2</sup> concentration was 1 vol.%, the corrosion tendency of the solution for tungsten was the smallest. Between 0 and 1 vol.%, the oxidizing properties of the solution became stronger with the increase of H2O<sup>2</sup> concentration, and the Ecorr tended to be more positive. It seems that a dense and thick passivation film gradually formed on the tungsten surface, which reduced the corrosion tendency of tungsten. The Ecorr decreased at 2 vol.% H2O2. This may be due to the high concentration of 2 vol.% H2O2, leading to the destruction of the passivation film, which accelerated the dissolution rate of the passivation film in the solution, resulting in a decrease in the corrosion resistance of tungsten. In the four solutions with different H2O<sup>2</sup> concentrations, the Icorr increased with increasing H2O<sup>2</sup> concentrations, which were 1.003 µA, 4.094 µA, 8.855 µA, and 12.55 µA, respectively, corresponding to the trend of *MRR* in Figure 8a. When the H2O<sup>2</sup> concentration in solution was 2 vol.%, the Icorr was the largest, and it was an order of magnitude higher than that without H2O2, which indicated that the passivation film was more likely to be formed on the tungsten surface after the addition of H2O2. the H2O2 concentration was 1 vol.%, the corrosion tendency of the solution for tungsten was the smallest. Between 0 and 1 vol.%, the oxidizing properties of the solution became stronger with the increase of H2O2 concentration, and the Ecorr tended to be more positive. It seems that a dense and thick passivation film gradually formed on the tungsten surface, which reduced the corrosion tendency of tungsten. The Ecorr decreased at 2 vol.% H2O2. This may be due to the high concentration of 2 vol.% H2O2, leading to the destruction of the passivation film, which accelerated the dissolution rate of the passivation film in the solution, resulting in a decrease in the corrosion resistance of tungsten. In the four solu‐ tions with different H2O2 concentrations, the Icorr increased with increasing H2O2 concen‐ trations, which were 1.003 μA, 4.094 μA, 8.855 μA, and 12.55 μA, respectively, corre‐ sponding to the trend of *MRR* in Figure 8a. When the H2O2 concentration in solution was 2 vol.%, the Icorr was the largest, and it was an order of magnitude higher than that without H2O2, which indicated that the passivation film was more likely to be formed on the tung‐ sten surface after the addition of H2O2.

*Micromachines* **2022**, *13*, x FOR PEER REVIEW 11 of 17

different pH values on the polishing performance of C‐SDP.

The more negative the corrosion potential is, the more easily the tungsten surface is corroded [30]. The Ecorr gradually tended to negative values with increasing pH values, as shown in Figure 12b. Among the six pH values, the Ecorr was the largest (−268 mV) at pH = 7, which indicated that tungsten was the least susceptible to corrosion at this time. The Ecorr was the smallest (−590 mV) at pH = 12, and tungsten was most easily corroded at this time. These were consistent with the polishing results shown in Figure 6; the surface qual‐ ity of tungsten was the worst when pH = 12. In electrochemical experiments, the Icorr is used to represent the corrosion rate of the workpiece, which can often reflect the change of *MRR* in the polishing process. As shown in Figure 12b, the Icorr increased with increases in the pH values, which meant the corrosion rate of tungsten increased gradually. The maximum Icorr was 15.01 μA when pH = 12, which was two orders of magnitude higher than that at pH = 7 (0.1185 μA). This phenomenon corresponded to the maximum *MRR* under the condition of pH = 12 shown in Figure 5a. According to the values of Icorr and Ecorr, with increases in pH values, tungsten was more easily corroded, and the corrosion rate increased in the alkaline solution. This phenomenon was consistent with the effect of

Figure 13a shows the potentiodynamic polarization curves of tungsten in different H2O2 concentration solutions, and four typical H2O2 concentrations of 0, 0.5 vol.%, 1 vol.%, and 2 vol.% were selected. Except for the different concentrations of H2O2, the other com‐ ponents of solution were the same as those of the polishing slurry. Figure 13b shows the corrosion potential (Ecorr) and the corrosion current density (Icorr) of tungsten in different H2O2 concentration solutions, which could be obtained from the potentiodynamic polari‐ zation curves shown in Figure 13a. As shown in Figure 13b, the Ecorr of tungsten under the H2O2 concentration of 1 vol.% (68 mV) was higherthan that under the H2O2 concentrations of 0 vol.% (−372 mV), 0.5 vol.% (−9 mV), and 2 vol.% (−172 mV). This indicated that when

**Figure 13.** (**a**) Dynamic potential polarization curves at different H2O<sup>2</sup> concentration solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different H2O<sup>2</sup> concentration solutions.

Figure 14a shows the potentiodynamic polarization curves of tungsten in solutions with different components. Three solutions with different components were an NaOHbased solution (pH = 9), H2O2-based solution (1 vol.% H2O2), NaOH, and H2O2-based solution (pH = 9, 1 vol.% H2O2). Figure 14b shows the corrosion potential (Ecorr) and corrosion current density (Icorr) of tungsten in solutions with different components. As shown in Figure 14b, in the solution without H2O<sup>2</sup> addition, the Icorr decreased from 8.855 µA to 1.003 µA, and the Ecorr changed from 68 V to −372 mV compared with the NaOH and H2O2-based solutions. This indicated that H2O<sup>2</sup> participated in the chemical reaction and could reduce the corrosion of the tungsten surface. After removing NaOH from the solution, the corrosion current Icorr on the tungsten surface was the smallest at only 0.454 µA. The result shows that the combined action of NaOH and H2O<sup>2</sup> can enhance the corrosion rate of tungsten, thereby improving the material removal rate in C-SDP.

It can be seen from the above electrochemical experiments that the chemical corrosion rate and corrosion resistance of tungsten are significantly affected by the chemical agents in the polishing slurry. The etch rate of tungsten can be improved by NaOH and H2O<sup>2</sup> in the polishing slurry, which is consistent with the material removal rate results shown in Figures 5a and 8a. The chemical composition of the tungsten surface in different solutions is analyzed below, and the chemical reactions that occur during the tungsten C-SDP process are discussed.

solutions.

tungsten, thereby improving the material removal rate in C‐SDP.

**Figure 14.** (**a**) Dynamic potential polarization curves at different solutions. (**b**) The corrosion poten‐ tial Ecorr and the corrosion current density Icorr at different solutions. **Figure 14.** (**a**) Dynamic potential polarization curves at different solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different solutions.

**Figure 13.** (**a**) Dynamic potential polarization curves at different H2O2 concentration solutions. (**b**) The corrosion potential Ecorr and the corrosion current density Icorr at different H2O2 concentration

Figure 14a shows the potentiodynamic polarization curves of tungsten in solutions with different components. Three solutions with different components were an NaOH‐ based solution (pH = 9), H2O2‐based solution (1 vol.% H2O2), NaOH, and H2O2‐based so‐ lution (pH = 9, 1 vol.% H2O2). Figure 14b shows the corrosion potential (Ecorr) and corrosion current density (Icorr) of tungsten in solutions with different components. As shown in Fig‐ ure 14b, in the solution without H2O2 addition, the Icorr decreased from 8.855 μA to 1.003 μA, and the Ecorr changed from 68 V to −372 mV compared with the NaOH and H2O2‐ based solutions. This indicated that H2O2 participated in the chemical reaction and could reduce the corrosion of the tungsten surface. After removing NaOH from the solution, the corrosion current Icorr on the tungsten surface was the smallest at only 0.454 μA. The result shows that the combined action of NaOH and H2O2 can enhance the corrosion rate of

### *4.5. XPS Testing Results*

It can be seen from the above electrochemical experiments that the chemical corro‐ sion rate and corrosion resistance of tungsten are significantly affected by the chemical agents in the polishing slurry. The etch rate of tungsten can be improved by NaOH and H2O2 in the polishing slurry, which is consistent with the material removal rate results shown in Figures 5a and 8a. The chemical composition of the tungsten surface in different solutions is analyzed below, and the chemical reactions that occur during the tungsten C‐ Figure 15 shows the XPS full spectra of the tungsten surface under different conditions. As seen in Figure 15, the main elements on the surface of tungsten samples were C, O, and W. The C element mainly came from the pollutants adsorbed on the tungsten surface during XPS. The presence of the O element under the four different conditions indicates that oxides were produced on the tungsten surface after soaking or polishing. The XPS fine spectra of the tungsten surface in Figure 16 were further analyzed in terms of the element valence. *Micromachines* **2022**, *13*, x FOR PEER REVIEW 13 of 17

**Figure 15.** X‐ray photoelectron full spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O2 solution immersion; (**c**) polishing slurry immersion; (**d**) after C‐ SDP. **Figure 15.** X-ray photoelectron full spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O<sup>2</sup> solution immersion; (**c**) polishing slurry immersion; (**d**) after C-SDP.

Figure 16 shows the XPS fine spectra of the tungsten surface under different conditions. In Figure 16, the deconvolution of the W (4f) spectrum shows four peaks: the two peaks located around 35.5 eV and 37.7 eV can be indexed, respectively, to W 4f7/2 and W 4f5/2 of W6+, while the other two peaks at 31.0 eV and 33.2 eV refer, respectively, to W 4f7/2 and W 4f5/2 of W [31]. The peaks of the W 4f orbital appeared in pairs, with the peak at the higher binding energy being W 4f5/2 and the lower one being W 4f7/2.

**Figure 16.** X‐ray photoelectron fine spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O2 solution immersion; (**c**) polishing slurry solution immersion; (**d**)

higher binding energy being W 4f5/2 and the lower one being W 4f7/2.

Figure 16 shows the XPS fine spectra of the tungsten surface under different condi‐ tions. In Figure 16, the deconvolution of the W (4f) spectrum shows four peaks: the two peaks located around 35.5 eV and 37.7 eV can be indexed, respectively, to W 4f7/2 and W 4f5/2 of W6+, while the other two peaks at 31.0 eV and 33.2 eV refer, respectively, to W 4f7/2 and W 4f5/2 of W [31]. The peaks of the W 4f orbital appeared in pairs, with the peak at the

after C‐SDP.

SDP.

**Figure 15.** X‐ray photoelectron full spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O2 solution immersion; (**c**) polishing slurry immersion; (**d**) after C‐

**Figure 16.** X‐ray photoelectron fine spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O2 solution immersion; (**c**) polishing slurry solution immersion; (**d**) after C‐SDP. **Figure 16.** X-ray photoelectron fine spectra of the tungsten surface under different conditions: (**a**) pH 9 solution immersion; (**b**) H2O<sup>2</sup> solution immersion; (**c**) polishing slurry solution immersion; (**d**) after C-SDP.

Figure 16 shows the XPS fine spectra of the tungsten surface under different condi‐ tions. In Figure 16, the deconvolution of the W (4f) spectrum shows four peaks: the two peaks located around 35.5 eV and 37.7 eV can be indexed, respectively, to W 4f7/2 and W 4f5/2 of W6+, while the other two peaks at 31.0 eV and 33.2 eV refer, respectively, to W 4f7/2 and W 4f5/2 of W [31]. The peaks of the W 4f orbital appeared in pairs, with the peak at the higher binding energy being W 4f5/2 and the lower one being W 4f7/2. Anik et al. investigated the anodic behavior of tungsten at different pH values. When the pH < 1, the main dissolution pathway of oxide is H<sup>+</sup> -assisted dissolution. When the pH is between 4 and 6.5, the dissolution of oxide mainly depends on OH−-assisted dissolution. Under strong alkalinity (pH > 12.5), the dissolution is achieved by the slow diffusion ofOH– to the W surface [32]. Lillard et al. believed that under acidic conditions, the oxides formed on tungsten surface could be divided into an inner barrier layer of WO<sup>3</sup> and an outer hydrated layer consisting of WO<sup>3</sup> (H2O) [26].

It was found that tungsten underwent chemical reactions in pH 9 solution and H2O<sup>2</sup> solution, and W6+ was formed on the tungsten surface, as determined by XPS spectroscopic analysis. Therefore, it is speculated that WO<sup>3</sup> may exist on the tungsten surface. It is worth noting that the equilibrium information in the Pourbaix diagram shows that WO<sup>3</sup> can only be formed stably when pH < 4 [26]. However, this diagram reacts the final product at different pH values and does not involve intermediate products. In addition, it was found by XPS characterization that WO<sup>3</sup> existed in a weakly acidic environment (4.5 < pH < 6.5), and the electrochemical study of tungsten in a weakly alkaline solution (such as pH = 9) showed that the anodic reaction was independent of pH [32]. Weidman et al. believed that the tungsten surface could show limited passivation behavior at neutral and slightly alkaline pH values [33]. Kneer et al. mentioned that a thin protective oxide layer exists on the tungsten surface over almost the entire pH range. When the pH is 4–9, a thin metastable WO<sup>3</sup> film is formed on the tungsten surface [34]. Generally, the oxide film on the tungsten surface will dissolve rapidly under strong alkaline conditions [33]. Combined with the XPS spectra in Figure 16, it can be inferred that NaOH and H2O<sup>2</sup> react with tungsten, and the main chemical reactions in alkaline C-SDP polishing slurry [35,36] are:

$$\rm{W} + 6\rm{OH}^- \rightarrow \rm{WO}\_3 + 3\rm{H}\_2\rm{O} \tag{2}$$

$$\rm{W} + \rm{3H}\_{2}\rm{O}\_{2} \rightarrow \rm{WO}\_{3} + \rm{3H}\_{2}\rm{O} \tag{3}$$

$$\rm{WO\_3 + 2OH^- \to WO\_4^{2-} + H\_2O} \tag{4}$$

Table 3 analyzes the peak areas of different valence states of tungsten shown in Figure 16. As shown in Table 3, the peak area representing the W element is much larger than that of W6+, which indicates that the content of hexavalent compounds on the tungsten surface is very small, so the oxide film is very thin. Compared with the tungsten surface immersed in pH 9 solution, the tungsten surface immersed in the polishing slurry has a higher W element content. This may be due to a relatively dense passivation film formed by the reaction between H2O<sup>2</sup> and tungsten in the polishing slurry, preventing further oxidation of the internal tungsten. On the other hand, the passivation film can react with OH−, leading to a reduction in its thickness. The proportion of the W element on the tungsten surface was also increased after C-SDP, which indicates that the addition of mechanical action can reduce the thickness of the passivation film to a certain extent. In general, the hardness of tungsten oxides tends to decrease as the oxidation state increases. Therefore, in the actual C-SDP process, both the dissolution mode and the chemical state of the oxide play a role in the material removal of the tungsten surface. At present, there are few studies on the chemical corrosion mechanisms of tungsten polishing in alkaline environments, and it is necessary to carry out further systematic research in the follow-up.

**Table 3.** Analysis results of XPS peak area of tungsten.


### **5. Conclusions**

In this paper, a novel high-efficiency C-SDP method was proposed to obtain high surface quality tungsten, and the effects of pH values and H2O<sup>2</sup> concentrations on the polishing performance of tungsten were investigated. The experimental results showed that tungsten C-SDP was significantly affected by pH values and H2O<sup>2</sup> concentrations. The *MRR* gradually increased with increasing pH values from 6.69 µm/h to 13.67 µm/h. With increasing H2O<sup>2</sup> concentrations, the *MRR* increased from 9.71 µm/h to 34.95 µm/h. When the pH value was 9 and the H2O<sup>2</sup> concentration was 1 vol.%, the optimal *R*<sup>a</sup> was 1.87 nm, and the corresponding *MRR* was 26.46 µm/h. This indicates that the C-SDP polishing technique was an effective method to obtain high surface quality tungsten.

The mechanism influences of pH values and H2O<sup>2</sup> concentrations on tungsten C-SDP were clarified by electrochemical and XPS tests. In alkaline polishing slurries containing H2O2, the tungsten surface mainly undergoes an oxidation reaction to form relatively soft tungsten trioxide, which can be quickly removed by abrasives. Then a new surface of the tungsten workpiece is exposed, and the chemical reaction continues, thereby increasing the material removal rate. In the C-SDP polishing slurry, when the alkalinity is too strong or the concentration of oxidant too high, excessive chemical corrosion and poor surface quality will result.

**Author Contributions:** Conceptualization, L.X. and H.C.; methodology, X.W.; software, L.W.; validation, L.W., W.H. and B.L.; formal analysis, X.W.; investigation, L.W.; resources, L.W.; data curation, H.C.; writing—original draft preparation, L.X. and H.C.; writing—review and editing, W.Z., F.C. and H.C.; visualization, X.W. and L.W.; supervision, J.Y.; project administration, L.X.; funding acquisition, H.C., X.W. and W.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work is supported by the National Natural Science Foundation of China (Grant Nos. 51905485 and 51805485), and the Joint Funds of the National Natural Science Foundation of China (Grant No. U20A20293) and the Natural Science Foundation of Zhejiang Province (LY21E050011 and LY21E050010).

**Conflicts of Interest:** The authors declare no conflict of interest.
