Investigation of the Oxidation of Plasma Sprayed Silicon Coating

doi:10.21203/rs.3.rs-864711/v1

The quality of plasma sprayed silicon coating determined by density and spreading condition of lamella greatly influences its performance. The oxidation of silicon coating deteriorates its performance. However, the investigators mostly focus on the oxidation and mechanism of amorphous silicon, porous silicon and specific crystal planes on single crystal. The factors which influence the quality and oxidation of silicon coating has never be studied.

The helium secondary gas flow has more influence on the quality of silicon coating than other spraying parameters. So, we prepare the silicon coating by plasma spraying technology with different secondary gas flow. The relationship between silicon coating’s quality and secondary gas flow is investigated. Furthermore, the oxidation of plasma sprayed silicon coating is discussed. We find that the secondary gas has an adverse effect on quality and oxidation degree of silicon coating. With the decrease of secondary gas flow, both quality and oxygen content of coatings increase. Besides, the oxygen atoms heterogeneously concentrate at the outermost layer of silicon lamellas. Components of oxygen enrichment area (OEA) from outside to inside are Si + SiO_x+SiO₂→Si + Si₂O + SiO₂→Si + SiO_x→Si (1 < x < 1.5). The width of OEA in lamellas at top layer of silicon coating is about 180nm, obviously thicker than that in inner lamellas. The results obtained from research can provide support to better understand the behavior of silicon coating in the service process.

Ceramics

silicon coating

plasma spray

oxidation

Due to the excellent chemical and physical properties, silicon has been widely used in many fields such as lithium ion battery[1–3], solar cell[4, 5] and so on[6, 7]. Some of these researches have made great breakthroughs. However, considerable researches focus on single crystal silicon and silicon material with nanoscale instead of silicon coating[7, 8]. he silicon coating has broad application prospect in the fields of biomedical application[9, 10] and environmental barrier coatings[11–13]. For instance, Wadley et al.[8] prepared an environmental barrier coating system containing a silicon bond coating and an ytterbium disilicate top coating on the SiC composite by plasma spraying technology to resist the erosion of hot steam. When hot steam flows into the inside of coating along the cracks and pores in top layer, the silicon coating can stop hot steam to contact with SiC substrate, prolonging the service time of substrate. The silicon coating can also alleviate the thermal mismatch between substrate and top coating. Ding et al. [9] successfully fabricated silicon coating on titanium alloy substrate by vacuum plasma spraying, which had great potential in orthopedics due to its improved bioactivity and biocompatibility treated in deionized water for several hours.

Plasma spraying technology is a high-efficient and low-cost approach to prepare silicon coatings. The quality of plasma sprayed silicon coating determined by density and spreading condition of lamella greatly influences its performance. The oxidation of silicon coating deteriorates its performance. But oxidation of silicon droplets is inescapable during spraying process, which restricts the popularity of plasma sprayed silicon coating. Some works have been carried out to investigate the oxidation mechanism and behavior of silicon materials such as silicon wafer and specific crystal planes on single crystal[14–21]. B.E. Deal and A. Grove[20] have established the classical Deal-Grove “linear-parabolic” model as far back as 1965 to investigate the thermal oxidation rule of silicon over a wide range of temperature (700 ~ 1300 ℃), partial pressure (0.1 ~ 1.0 atm) when environment atmosphere is oxygen. Recently, Li et al.[22] investigated the oxidation of amorphous silicon and crystal silicon via reactive molecular dynamics simulations. The interfacial structure between Si and oxide is also revealed. But the researches about oxidation behavior of plasma sprayed silicon coating have not get attention. Liu et al.[21] just investigated the phase composition and microstructure of plasma sprayed silicon coating. The report didn’t involve the further discuss about silicon oxide and the influence factor of coating’s quality.

According to our pre-experiment, the quality and oxygen content of silicon coating are most affected by helium secondary gas flow compared with other spraying parameters. In this paper, the silicon coatings are prepared on high alloy substrate with different secondary gas flow. The relationship between quality of silicon coatings and secondary gas flow is investigated. The existing location and form of oxygens in silicon coatings are also discussed.

2.1 Preparation of silicon powders and coatings

The raw material used in this work was irregular silicon powders (≥ 99.9%, median particle diameter 75µm, Forsman, Beijing, China). The spherical silicon powders were prepared by using induced plasma processing equipment (PL-35, TEKNA Plasma Systems Inc., Canada) to process irregular powders. The processing parameters are shown in Table 1. The as-prepared spherical powders were cleaned thoroughly using absolute ethyl alcohol. Dry spherical powders were obtained by keeping them in an air oven at 100℃. After drying, the spraying silicon powders in the range of 30 − 80µm were obtained by sieving. The flowability of spraying silicon powders is 72s/50g which satisfies the demand of plasma spraying.

Table 1

Parameters of the induction plasma spheroidization procedure.
Parameter	Value
Power, kW	30
Primary gas (Ar) flow, (L min^-1)	50
Secondary gas (H₂) flow, (L min^-1)	4
Dispersed gas (Ar) flow, (L min^-1)	2
Central gas (Ar) flow, (L min^-1)	20
Carrier gas (Ar) flow, (L min^-1)	4.5 ~ 5
Powder feed rate, RPM	5

The GH3128 Nickle-based high temperature alloy substrates with the size of φ25 mm × 3 mm used in experiment were provided by Iron and Steel Research Institute. The substrates were grit-blasted to gain rough surface which can increase the bonding strength between coating and substrate. The spraying silicon powders were deposited on the substrate by air plasma spraying equipment (GTS-5500, Praxair, America) equipped with SG-100 spraying gun and 1264 powder feeder. The substrates were fixed on the steel wire gauze on the surface of steel shelf. The moving velocity of spraying gus is 500 mm/s. The thickness of silicon coating was about 200 µm. The silicon coatings were prepared with three groups of spraying parameters. The plasma spraying parameters are listed in Table 2. The three groups of spraying parameters only differ in secondary gas flow. The silicon coatings without substrates were obtained by using aqua regia to dissolve substrates.

Table 2

Plasma spraying parameters used for the deposition of silicon coatings.
Parameter	Value
Series No.	1	2	3
Spraying current (A)	550	550	550
Primary gas (Ar) flow (L min^− 1)	35.4	35.4	35.4
Secondary gas (He) flow (L min^− 1)	4.7	2.4	0
Carrier gas (Ar) flow (L min^− 1)	5.7	5.7	5.7
Powder feed rate (RPM)	3	3	3
Spray distance (mm)	75	75	75

2.2 Analysis and characterization

The surface and fracture morphology of silicon coatings were observed by scanning electron microscope (SEM, Philips S-4800, Hitachi Ltd., Yokohama, Japan). The oxygen contents of silicon coatings were detected by oxygen and nitrogen analyzer (TC500, Laboratory Equipment Corp., America). The oxygen content of spherical spraying silicon powder with (CSP) and without (SP) corrosion by aqua regia were tested to tell whether aqua regia will influence oxygen content of silicon coating. The component of silicon lamella was determined by an X-ray photoelectron spectrometer (XPS spectra, 250Xi, Thermo Fisher Scientific Inc., America) with an X-ray source of Al Kα. Each XPS testing result contains three spectra which are survey spectra from 0 ~ 1300 eV, Si 2p spectra and O 1s spectra. The C1s peak at 284.8 eV from adventitious carbon was used to calibrate the photoelectron binding energy (BE). Before the XPS test of inner surface of silicon lamella, the sample was subjected to Ar sputtering to exposure inner surface of silicon lamella. The spectra obtained from XPS test were analyzed by Thermal Avantage software. The microstructure analysis and elemental mapping of coating sample were performed using scanning transmission electron microscope (STEM, Tecnai G2 F30, FEI, USA) with attached high angle annular dark field probe (HADDF) and energy dispersive spectrometer (EDS). The specimen used in TSEM was taken by focused ion beam instrument (FIB, Crossbeam 540, Carl Zeiss, German) to from coating. The thickness of specimen is less than 50 nm. The data from TEM were analyzed by DigitalMicrography software.

3.1 Microstructure of silicon coating

The surface and fracture morphologies of silicon coatings prepared with 4.7 L/min, 2.4 L/min and 0 L/min secondary gas flow are shown in Fig. 1(a) ~ 1(d), Fig. 1(e) ~ 1(h), Fig. 1(i) ~ 1(l), respectively. It can be seen that silicon coatings prepared with different secondary gas flow all show the typical lamellar morphology of plasma sprayed coating and the fractures of coatings are dense. The surfaces of all coatings are relatively flat and most of the silicon powders completely melt and spread out (Fig. 1(a), 1(e) and 1(i)). Some fragments appear on the surfaces of silicon lamellas (Fig. 1(c), 1(g) and 1(k)) and some pores can be seen at the nonoverlapping area of silicon lamellas (Fig. 1(a), 1(e) and 1(i)). The splashing particles approximately 10µm in diameter attach around the silicon lamellas, which can be seen from the surface and fracture of silicon coatings. The differences between coatings prepared with different secondary gas flow are the numbers of splashing particles and fragments. When secondary gas flow is 4.7 L/min, the number of splashing particles shown in the surface and fracture of silicon coating is the maximum. The number of fragments in coating prepared with 4.7 L/min secondary gas flow is also the maximum among these coatings. With the decrease of secondary gas flow, the numbers of splashing particles and fragments decrease. When the secondary gas flow is 0 L/min, there are scarcely splashing particles attaching around the well-spread lamellas and the surfaces of silicon lamellas are smooth without any fragments.

According to the analysis results from Fig. 1, we find that the secondary gas can deteriorate the quality of plasma sprayed silicon coating. The quality of silicon coating prepared without secondary gas is the best. The main gas and secondary gas used in plasma spraying are argon and helium. The heat conductivity of helium is much bigger than argon. When we provide helium secondary gas in plasma spraying process, it will transfer more heat to silicon powders than argon when main gas is ionized. With the help of secondary gas, the temperature of silicon droplets is higher than the temperature of droplets without the help of secondary gas. And the morphology of silicon coating is closely related with the temperature of silicon droplets. Because the higher the temperature of silicon droplets is, the smaller the surface tension and viscosity of silicon droplets are, the better the fluidity of silicon droplet is[23–25]. When providing secondary gas during plasma spraying, excessive heat leads to the excessive temperature of silicon droplets. When over-heated silicon droplets hit the substrate, small particles easily splash from the droplets due to lower viscosity and smaller surface tension. So as shown in Fig. 1(b) and 1(f) that some splashing powders attaching around the silicon lamellas when secondary gas flow is 2.4 L/min and 4.7 L/min. Besides, the melting point and boiling point of crystal silicon are only 1414℃ and 2900 ℃. The silicon powders smaller than 30 µm account for approximately 10% among the spraying silicon powders. When excessive heat is transferred to silicon powders during its in-flight process, small powders are much easier to evaporate. The increase of secondary gas flow will accelerate the flying velocity and shorten the flying time of silicon droplets in spraying gas. But the heat transferred by secondary gas is enough to get smaller silicon powders evaporate. When silicon droplets hit the substrate, the silicon vapor around the droplets will solidify on the surface of spread lamella in the form of fragments[26–28]. Both the numbers of splashing particles and fragments in coating prepared with 4.7 L/min secondary gas flow are slightly larger than that in coating prepared with 2.4 L/min secondary gas flow because of the higher temperature of silicon droplets. Without secondary gas, heat transferred by argon can just melt silicon powders. Silicon droplets spread well when they are deposited on the substrate, forming a smooth and dense coating. And there are rarely splashing particles and fragments on the surface of silicon coating prepared without secondary gas. Although small splashing particles and fragments have adverse influence on the surface quality of silicon coatings, the silicon coatings prepared with different secondary gas flow all show the dense fractures, which can also be attributed to the secondary gas flow. With the help of smaller surface tension and viscosity, the fluidity of over melted silicon droplets is good enough to penetrate intervals between splashing particles, successfully sealing the intervals. So, the fractures of all coatings are dense. According to the analysis results, we find that the secondary gas has an adverse influence on the quality of silicon coatings. And the quality of silicon coating prepared without secondary gas is the best.

3.2 Oxygen content of silicon coating

The oxygen contents of silicon powders and coatings are shown in Fig. 2. The oxygen contents in SP and CSP are 0.089%, 0.101%. And the oxygen contents in coatings prepared with 4.7 L/min, 2.4 L/min, 0 L/min secondary gas flow are 0.489%, 0.35%, 0.387%, respectively. According to the calculation results, the oxygen content of SP is 11.88% less than oxygen content of CSP. The oxygen content of silicon coating prepared with 4.7 L/min is 39.7%, 26.36% higher than that in coatings prepared with 2.4 L/min and 0 L/min secondary gas flow. By the comparison above, we could think that the incidence of aqua regia to oxygen content of silicon coatings is negligible compared with the incidence of secondary gas flow. Decreasing the secondary gas flow can significantly reduce the oxygen content in silicon coating.

As presented in Fig. 2 that oxygen contents in coatings are much higher than that in silicon powders because oxidation between silicon sprayed powders and oxygen is inevitable during plasma spraying process. Although silicon powders are melted in inert atmosphere, some silicon powders still be blown out from inert atmosphere. The high-speed plasma gases will incur air entrainment[29]. Besides, helium has a very high viscosity compared to argon which will aggravate the air entrainment. The reasons above-mentioned enhance the contacting likelihood of silicon droplets and oxygen, leading to the oxidation of silicon droplets. Besides, the higher secondary gas flow will transfer more heat to silicon powders when plasma gas is ionized. The higher the temperature of silicon droplets is, the severer oxidation happens in plasma spraying process. So, the oxygen content in silicon coating prepared with 4.7 L/min secondary gas flow is higher than that in coatings prepared in other two situations.

3.3 Component of silicon lamella

Figure 3(a) and 3(b) show the XPS spectra of silicon coating tested at the same location under room temperature and 400℃. The sample used in XPS is as-sprayed silicon coating prepared without secondary gas. According to the XPS results shown in Fig. 3(a) and 3(b), both Si 2p spectra are deconvoluted into four bases of 99.6 eV, 100.2 eV, 102.5 eV and 103.6 eV. The BE peaks at 99.6 eV and 100.2 eV are assigned to Si 2p3/2 and 2p1/2 with a spin-orbit split of 0.6 eV. The BE peak at 102.5 eV corresponds to Si-O bond in SiO_x (1 < x < 1.5). SiO_x is formed by alternant arrangement of SiO and Si₂O₃ tetrahedrons. The BE peak at 103.6eV is associated with Si-O bond in SiO₂. Both O 1s fine spectra in Fig. 3(a) and 3(b) are deconvoluted into two bases of 532.8 eV and 533.9 eV. The BE peaks at 532.8 eV and 533.9 eV are attributed to O-Si bonds in SiO_x and SiO₂, which are consistent with the results from Si 2p spectra. The differences between the analysis results from Fig. 3(a) and 3(b) are the proportions of atoms and components. Atomic percentage of oxygen detected at room temperature and 400℃ are 59.25% and 54.99% respectively. At room temperature and 400℃, the proportions of Si, SiO_x, SiO₂ contained on the surface of silicon lamella are 46.11%, 5.87%, 48.02% and 44.43%, 18.00%, 37.57% (as shown in Table 3). It is obvious that with the increase of temperature, the proportion of SiO₂ detected on the surface of silicon lamella decreases and the proportion of SiO_x increases, which can be can be attributed to the weakly bonded O-Si bond in SiO₂. With the increase of temperature, weakly bonded O-Si bond break, leading to the decrease of oxygen element. So, the changing tendencies of the proportion of SiO_x and SiO₂ are opposite. 9

Table 3

Components proportions of silicon lamella^a on different conditions
Component	Room temperature	400℃	0 nm	60 nm	120 nm	180 nm
Si	46.11%	44.43%	67.97%	68.64%	90.80%	100.00%
Si₂O	—	—	—	23.78%	—	—
SiO_x	5.87%	18.00%	—	—	9.20%	—
SiO₂	48.02%	37.57%	32.03%	7.58%	—	—
^aThe Chosen lamella was prepared without argon secondary gas.

Figure 3(c) ~ 3(f) show the XPS outcomes detected at the same location but different depths of one silicon lamella. The testing location here is different from the testing location depicted in Fig. 3(a) and 3(b). And the testing depths shown in Fig. 3(c) ~ 3(f) are the surface of silicon lamella and 60 nm, 120 nm, 180 nm below the surface of lamella, respectively. It can be obviously seen from Fig. 3(c) ~ 3(f) that the categories and proportions of silicon oxides tested in different depths are different. As shown in Fig. 3(c), SiO₂ can be detected on the surface of silicon lamella except to Si, which is different from the results obtained from Fig. 3(a). The difference in components at different positions can be attributed to the inhomogeneous distribution of oxygen element on lamella surface. The results from Fig. 3(d) indicate Si, Si₂O and SiO₂ existing beneath 60 nm of lamella surface. The analysis results of XPS spectra in Fig. 3(e) and 3(f) are basically same. Both Si and SiO_x can be detected 120 nm and 180 nm below the surface of silicon lamella. But the proportion of SiO_x in 180 nm depth is less than that in 120 nm depths. As shown in Table 4, the intensity ratio of Si 2p peaks and O 1s peaks detected on the surface and 60 nm, 120 nm, 180 nm beneath surface of silicon lamella are 0.47, 2.72, 4.14 and 5.14, respectively. According to the intensity ratios of Si 2p peaks and O 1s peaks and analysis results obtained from Si 2p spectra in Fig. 3(f), we could think only Si exists beneath 180 nm of silicon lamella. From the surface to the 180 nm below the surface of silicon lamella, the total proportions of silicon oxides decrease, and the proportions of Si increase from 67.97–100% as shown in Table 3.

Table 4

Parameters of Si 2p and O 1s XPS peaks tested in different depths
Depth (nm)	Element	Binding energy (eV)	Peak intensity (counts s^− 1)	Intensity ratio I_Si/I_O
0	O	532.18	44507.24	0.47
0	Si	98.98	20885.11	0.47
60	O	532.88	13873.57	2.72
60	Si	99.68	37742.96	2.72
120	O	533.08	9746.38	4.14
120	Si	99.78	40368.81	4.14
180	O	532.58	8010.6	5.14
180	Si	99.88	41213.53	5.14

Combining the analysis results above, we can find that oxygen atoms concentrate on the outermost layer of silicon lamella, existing in the form of different silicon oxides. The thickness of OEA is about 180 nm. The components from surface to inner of OEA change as following order: Si + SiO_x+SiO₂→Si + Si₂O + SiO₂→Si + SiO_x→Si (1༜x༜1.5). The reason why silicon oxides distribute as above will be discussed in Sect. 3.4.

3.4 Microstructure and element distribution of silicon coating

Figure 4(a) reveals the microscopic features of as-sprayed silicon coating prepared without secondary gas. A and B represent the two silicon lamellas deposited in succession. Lamella A is deposited first, followed by lamella B. There is an obvious boundary between lamella A and B. A nanocrystal zone of 100 nm in thickness can be observed in the lamella A. The nanocrystal zone closely next to the interface. The columnar grains above nanocrystal zone are perpendicular to the interface of silicon lamellas. Relatively big equiaxed crystals, as shown in lamella B in Fig. 4(a), appear near the upper surface of silicon lamella. The difference in size and morphology of grains can be attributed to the cooling rate and stress on different locations. In the process of plasma spraying, melted silicon droplets obtain a great amount of kinetic energy. When droplet A hits the as-deposited and solidified lamella B, the stress is generated at the bottom of solidified lamella A due to kinetic energy. The temperature gradient at interface between lamella A and B is higher than that in the interior of silicon lamella. The comprehensive effect of temperature gradient and stress causes the formation of nanocrystal. The high-resolution transaction electron micrography (HRTEM) image in nanocrystal zone shown in Fig. 4(j) further illustrates the existence of nanocrystals. Due to the high cooling rate, grains near nanocrystal zone grow in the form of columnar crystals, silicon atoms nearing the nanocrystals arrange in complete disorder, as shown in Fig. 4(k). Equiaxed grains are formed near the upper surface of silicon lamella affected by a medium temperature gradient. As shown in Fig. 4(i), the silicon atoms orderly arrange and the interplanar spacing of (111) is 0.32 nm.

Figure 4(b) is the HAADF image of region marked with blue box in Fig. 4(a). There is an obvious dark irregular strip between lamellas A and B. Figure 4(c) and 4(d) are enlarged views of regions C and D marked with blue boxes in Fig. 4(b). According to the line scanning results shown in Fig. 4(e), there is an obvious vibration of elements proportions along the dotted line in Fig. 4(d). Oxygen proportion obviously increases in dark irregular strip region, but silicon proportion decreases at the same position. This opposite changing tendency means dark irregular strip is OEA. As illustrated in Fig. 4(f) and 4(g), the mapping results of region shown in Fig. 4(c) also prove that the dark strip made up by white dotted line is the OEA. Furthermore, the contrast of OEA is high. The light gray area and dark gray area exist at the same time in OEA. Light gray area corresponds to silicon oxides with less oxygen, and the oxygen proportion in dark gray is higher, corresponding the silicon oxygen with more oxygen. This also demonstrates oxygen heterogeneously distribute in OEA of silicon lamellas.

The results obtained from EDS mapping images in Fig. 4 and XPS spectra in Sect. 3.3 both illustrate that oxygen atoms concentrate at the outermost layer of silicon lamellas. But the thickness of OEA depicted in Fig. 4 and Sect. 3.3 are greatly different. According to the XPS analysis results, we could find that the thickness of OEA is about 180 nm, but the maximum thickness of OEA depicted in EDS mapping images is only 30 nm. And this 30 nm thick OEA belongs to two adjacent lamellas. We know that the silicon coating is formed by lamellas piling one atop another. And the difference in thickness of OEA is attributed to the location of silicon lamellas. The results obtained from XPS spectra in Sect. 3.3 reflect the information of silicon lamellas at the top layer of coating. But EDS mapping images reveal the information about lamellas in inner coating. During the plasma spraying process, silicon powders are heated in inert gas flow. The high-speed inert gas flow incurs air entrainment as shown in Fig. 5. The silicon atoms at the surface of silicon droplets react with entrained oxygen, forming the silicon oxides. When hit the substrate, the solidification of silicon droplets always accompanies with the oxidation. Both in-flight and solidification are moments every silicon powder goes through in coating formation process. So OEA can be detected at any lamella, both on the top of coating and inside coating. But for lamellas on the top of silicon coating, they contact with oxygen not only in the process of plasma spraying, but also in the cooling process after removing the spraying gun[30, 31]. The in-flight time of silicon powders only lasts several microseconds, but the cooling time of top lamellas continues for several seconds. The longer contacting time between oxygen and lamellas leads to the thicker OEA of lamellas. So, the OEA in lamellas on the top layer of silicon coating is much thicker than that in lamellas inside the coating. Besides, oxygen heterogeneously distribute at the OEA of silicon lamellas due to the different oxygen permeability in different components[32]. When silicon atoms directly contacting with oxygen, SiO₂ is formed covering the surface of silicon lamellas. With the penetration of oxygen in lamella, the SiO₂ layer gets thicker. But the oxygen permeability in SiO₂ is worse than that in Si. Oxygen partial pressure in inner lamella is not enough to generate SiO₂, so less-oxygen oxides such as Si₂O and SiO_x forms inside the OEA. As depicted in Fig. 5 that Si and silicon oxides distribute from outside to inner OEA are Si + SiO_x+SiO₂→Si + Si₂O + SiO₂→Si + SiO_x→Si (1<x<1.5).

The silicon coatings are prepared by plasma spraying with 4.7 L/min, 2.4 L/min and 0 L/min secondary gas flow. The relationship between quality of plasma sprayed silicon coatings and the secondary gas flow is investigated. The oxidation behavior of silicon coatings is also explained.

(a) With the decrease of secondary gas flow from 4.7 L/min to 0 L/min, the quality of coatings persistently increases. And the quality of silicon coating fabricated without secondary gas is the best. Almost no pores, splashing spherical and fragments can be observed on the surface and fracture of coating. This can be attributed to the secondary gas flow. When we provide secondary gas in plasma spraying process, it will transfer more heat when plasma gas is ionized, leading to the excessive temperature of silicon droplets. Splashing particles and fragments are easily formed on the surface of over melted silicon lamellas.

(b) The oxygen content of silicon coating prepared with 4.7 L/min secondary gas flow is much higher than others. Because excessive secondary gas makes the temperature of silicon droplets rise. The oxidation degree of silicon droplets and lamellas would increase with increase of temperature.

(c) The grain morphology of silicon lamella contains nanocrystals, columnar crystals and equiaxed grains. Due to the effort of temperature gradient in silicon lamellas and impacting force between silicon droplets and solidified lamellas during spraying, nanocrystals are formed at the bottom of lamella. The high and medium temperature gradient leading to the formation of columnar grains and equiaxed grains, respectively.

(d) Components in OEA from outside to inside are Si+SiO_x+SiO₂→Si+Si₂O+SiO₂→Si+SiO_x→Si (1<x<1.5). The heterogeneously distribution of oxygen atoms and components can be attributed to the different oxygen permeability in different components. SiO₂ is formed at the outermost of OEA due to the enough oxygen. With the thicker of SiO₂ layer, the oxygen partial pressure in inner lamellas and droplets decrease due to the lower oxygen permeability in silicon oxides. So, the less-oxides are formed in inner lamellas.

(e) The thickness of OEA in silicon lamellas at top layer of coating is much thicker than that in inner lamellas. Because the thickness of OEA correlates with the time of oxygen which silicon droplets and lamellas contact. In the process of plasma spraying, every silicon droplet is likely to contract with oxygen. But only silicon lamellas on the top layer of coating contact oxygen after removing the spraying gun. The longer contacting time with oxygen makes the thickness of OEA in lamellas on the top coating bigger.

(f) The conclusions obtained in this research provide the reference to prepare high quality plasma sprayed silicon coating. The research also provides a new perspective to understand the performance silicon coating in service.

Acknowledgments

This work was supported by the National Natural Science Foundation of China [No. 51772027].

Appendices

None.

Sun R, Xu H, Hang S et al (2020) Effect of acid leaching on silicon nanoparticles and their electrochemical performance in lithium-ion batteries. J Nanosci Nanotechno 20:2543–2549. http://doi.org/10.1166/jnn.2020.17378
Ette PM, Bhargav PB, Ahmed N et al (2019) Nanocrystalline silicon embedded highly conducting phosphorus doped silicon thin film as high power lithium ion battery anode. Electrochim Acta 330:135318. http://doi.org/10.1016/j.electacta.2019.135318
Tao B, He J, Miao F, A hybrid sandwich structure of TiO₂/N-graphene/Ag supported by ordered silicon nanowires and its application as lithium-ion battery electrodes, Mater. Lett. 262 1–4. http://doi.org/10.1016/j.matlet.2019.127046
Zheng J, Mehrvarz H, Liao C et al (2019) Large-area 23%-efficient monolithic perovskite/homojunction-silicon tandem solar cell with enhanced UV stability using down-shifting material. ACS Energy Lett 2019:2623–2631. http://doi.org/10.1021/acsenergylett.9b01783
Mehmood H, Nasser H, Tauqeer T et al (2019) Simulation of silicon heterostructure solar cell featuring dopant-free carrier-selective molybdenum oxide and titanium oxide contacts. Renew Energ 143C:359–367
Graeve ID, Schoukens I, Lanzutti A et al (2013) Mechanism of corrosion protection of hot-dip aluminium-silicon coatings on steel studied by electrochemical depth profiling. Corros Sci 76:325–336. http://doi.org/10.1016/j.corsci.2013.07.005
Shirvani K, Mastali S, Rashidghamat A et al (2013) The effect of silicon on thermal shock performance of aluminide-thermal barrier coatings. Corros Sci 75:142–147. http://doi.org/10.1016/j.corsci.2013.05.025
Richards BT, Young KA, de Francqueville F et al (2016) Response of ytterbium disilicate-silicon environmental barrier coatings to thermal cycling in water vapor. Acta Mater 106:1–14. http://doi.org/10.1016/j.actamat.2015.12.053
Niu Y, Liu X, Ding C (2008) Vacuum-plasma-sprayed silicon coatings for biomedical application. Mat Sci Eng C 28:1132–1137. http://doi.org/10.1016/j.msec.2007.07.001
Ma L, Niu J (2019) Facile fabrication of silicon coating for improving biodegradation and cytocompatibility of pure magnesium by hydrothermal method. Metall Res Technol 116:416. http://doi.org/10.1051/metal/2019002
Richards BT, Wadley H (2014) Plasma spray deposition of tri-layer environmental barrier coatings. J Eur Ceram Soc 34:3069–3083. http://doi.org/10.1016/j.jeurceramsoc.2014.04.027
Rohbeck N, Morrell P, Xiao P (2019) Degradation of ytterbium disilicate environmental barrier coatings in high temperature steam atmosphere. J Eur Ceram Soc 39:3153–3163. http://doi.org/10.1016/j.jeurceramsoc.2019.04.034
Richards BT, Begley MR, Wadley HNG (2015) Mechanisms of ytterbium monosilicate/mullite/silicon coating failure during Thermal Cycling in water vapor. J Am Ceram Soc 98:4066–4075. http://doi.org/10.1111/jace.13792
Kathleen C, Chu PW, Marquis EA (2018) Early oxidation behavior of Si-coated titanium. Corros Sci 140:297–306. http://doi.org/10.1016/j.corsci.2018.05.035
Jayachandran S, Delabie A, Billen A et al (2015) Deposition of O atomic layers on Si(100) substrates for epitaxial Si-O superlattices: investigation of the surface chemistry. Appl Surf Sci 324:251–257. http://doi.org/10.1016/j.apsusc.2014.10.086
Feng T, Li H, Fu Q et al (2011) The oxidation behavior and mechanical properties of MoSi₂-CrSi₂-SiC-Si coated carbon/carbon composites in high-temperature oxidizing atmosphere. Corros Sci 53:4102–4108. http://doi.org/10.1016/j.corsci.2011.08.018
Lamontagne B, Roy D, Sporken R et al (1995) Identification of the adsorption sites of molecular oxygen on Si(111) using XPS. Prog Surf Sci 50:315–324. http://doi.org/10.1016/0079-6816(95)00065-8
Engstrom JR, Bonser DJ, Nelson MM et al (1991) The reaction of atomic oxygen with Si(100) and Si(111). Surf Sci 256:317–343. http://doi.org/10.1016/0039-6028(91)90875-S
Morita M, Ohmi T, Hasegawa E et al (1990) Growth of native oxide on a silicon surface. J Appl Phys 68:1272–1281. http://doi.org/10.1063/1.347181
De Al BE, Grove AS (1965) General relationship for the thermal oxidation of silicon. J Appl Phys 36:3770–3778. http://doi.org/10.1063/1.1713945
Niu Y, Liu X, Ding C (2006) Phase composition and microstructure of silicon coatings deposited by air plasma spraying. Surf Coat Tech 201:1660–1665. http://doi.org/10.1016/j.surfcoat.2006.02.040
Zhang X, Duan Y, Dai X et al (2019) Atomistic origin of amorphous-structure-promoted oxidation of silicon. Appl Surf Sci 504:144437. http://doi.org/10.1016/j.apsusc.2019.144437
Fujii H, Matsumoto T, Hata N et al., Surface tension of molten silicon measured by the electromagnetic levitation method under microgravity, Matall Mater Trans A 31A (2000) 1585 – 1583. http://doi.org/10.1007/s11661-000-0168-1
Millot F, V S, J-C R et al (2008) The surface tension of liquid silicon at high temp. Mat Sci Eng A 495:8–13. http://doi.org/10.1016/j.msea.2007.10.108
Sasaki H, Tokizaki E, Huang XM et al., Temperature dependence of the viscosity of molten silicon measured by the oscillating cup method, Jpn J Appl Phys Pt 1, Regular papers & short notes. (1995) http://doi.org/10.1143/JJAP.34.3432
Buchner P, Schubert H, Uhlenbusch J et al (2001) Evaporation of zirconia powders in a thermal radio-frequency plasma. J Therm Spray Techn 10:666–672. http://doi.org/10.1361/105996301770349196
Shinoda K, Murakami H (2010) Splat morphology of yttria-stabilized zirconia droplet deposited via hybrid plasma spraying. J Therm Spray Techn 19:602–610. http://doi.org/10.1007/s11666-009-9460-9
Matthews S, Taliana F, James B (2012) Heat treatment of plasma-sprayed Al₂O₃ and Al₂O₃–WO₃ coatings between 500 and 1000°C. Surf Coat Tech 212:109–118. http://doi.org/10.1016/j.surfcoat.2012.09.030
Vardelle M, Vardelle A, Li KI et al (1996) Coating generation: Vaporization of particles in plasma spraying and splat formation. Pure Appl Chem 68:1093–1099. http://doi.org/10.1351/pac199668051093
Sobolev VV, Guilemany JM (1998) Oxidation of coatings in thermal spraying. Mater Lett 37:231–235. http://doi.org/10.1016/S0167-577X(98)00097-4
Kolman D, Voleník K (2002) Modeling of oxidation during plasma spraying of iron particles, Plasma Chem. Plasma P 22:437–450. http://doi.org/10.1023/A:1015373201069
Lee KN (2019) Yb₂Si₂O₇ Environmental barrier coatings with reduced bond coat oxidation rates via chemical modifications for long life. J Am Ceram Soc 102:1507–1521. http://doi.org/10.1111/jace.15978

Investigation of the Oxidation of Plasma Sprayed Silicon Coating

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Material And Methods