CC BY
19DOI: 10.6060/ivkkt.20226510.6604 УДК: 537.525
СОСТАВ ПЛАЗМЫ И КИНЕТИ^ ТРАВЛЕНИЯ SiO2 В СМЕСИ CF4/C4F8/Ar/He: ВЛИЯНИЕ СООТНОШЕНИЯ CF4/C4F8 И МОЩНОСТИ СМЕЩЕНИЯ
А.М. Ефремов, K.-H. Kwon
Александр Михайлович Ефремов (ORCID 0000-0002-9125-0763)*
Ивановский государственный химико-технологический университет, Шереметевский просп., 7, Иваново, Российская Федерация, 153000 E-mail: amefremov@mail.ru*
Kwang-Ho Kwon (ORCID 0000-0003-2580-8842)
Korea University, 208 Seochang-Dong, Chochiwon, Korea, 339-800 E-mail: kwonkh@korea.ac.kr
Исследованы характеристики газовой фазы и кинетика реактивно-ионного травления диоксида кремния в плазме CF/CiFs/Ar/He при варьировании соотношения CF4/C4F8 и потенциала смещения в режиме малой (~ 0,05 Вт/см3) вкладываемой мощности. Интерес к такому режиму обусловлен возможностью получения высокой анизотропии травления при малых радиационных повреждениях поверхности. Схема исследования включала диагностику плазмы с помощью зондов Ленгмюра и оптической эмиссионной спектроскопии в варианте внутренней (без использования стандартной добавки) актинометрии. Показано, что замещение C4F8 на CF4 не приводит к существенным изменениям параметров электронной и ионной компонент плазмы, но сопровождается незначительным ростом концентрации атомов фтора. Напротив, увеличение мощности (а, следовательно, и потенциала) смещения не возмущает состава плазмы, но характеризуется пропорциональным изменением энергии бомбардирующих ионов. Таким образом, выбранные варьируемые параметры представляют классические «химический» и «физический» механизмы воздействия на кинетику гетерогенных стадий процесса травления. Установлено, что основной вклад в процесс травления SiO2 вносит химическая составляющая, при этом при мощностях смещения выше 400 Вт вероятность химической реакции Si(s.) + + xF ^ SiFx (где индекс (s.) отвечает частице, локализованной на поверхности) не зависит от эффективности образования центров адсорбции для атомов фтора под действием ионной бомбардировки SiOx(s.) ^ Si(s.) + xO. В условиях малых мощностей смещения наличие такой зависимости подтверждается симбатным поведением эффективной вероятности взаимодействия и интенсивности ионной бомбардировки, характеризуемой произведением плотности потока ионов на корень квадратный из их энергии. Сделаны предположения об особенностях кинетики объемных и гетерогенных процессов в условиях низкой плотности плазмы.
Ключевые слова: C4F8, CF4, плазма, параметры, активные частицы, ионизация, диссоциация, травление
PLASMA COMPOSITION AND SiO2 ETCHING KINETICS IN CF4/C4F8/Ar/He MIXTURE: EFFECTS OF CF4/C4F8 MIXING RATIO AND BIAS POWER
A.M. Efremov, K.-H. Kwon
Alexander M. Efremov (ORCID 0000-0002-9125-0763)*
Ivanovo State University of Chemistry and Technology, Sheremetevskiy ave., 7, Ivanovo, 153000, Russia E-mail: amefremov@mail.ru*
Kwang-Ho Kwon (ORCID 0000-0003-2580-8842)
Korea University, 208 Seochang-Dong, Chochiwon, Korea, 339-800
E-mail: kwonkh@korea.ac.kr
Gas-phase characteristics as well as reactive-ion etching kinetics of silicon dioxide in CF/CiFs/Ar/He plasma with variable CF4/C4F8 mixing ratio and bias potential were investigated under conditions of low (~ 0.05 W/cm3) input power regime. The interest to such regime is due to the possibility to obtain higher etching anisotropy with lower surface damages. The research scheme included plasma diagnostics by Langmuir probes and optical emission spectroscopy in the internal (with no use of standard additives) actinometry approach. It was shown that the substitution of C4F8 for CF4 does not produce sufficient changes in both electrons- and ions-related plasma parameters, but causes a weak increase in fluorine atom density. On the contrary, an increase in the bias power (and thus, in the bias potential) does not disturb plasma composition, but is characterized by proportional changes in the ion bombardment energy. As such, selected variable parameters represent somewhat classical "chemical" and "physical" factors influencing heterogeneous stages of the etching process. It was found that the dominant contribution to the SiO2 etching process belongs to its chemical component while bias powers above 400 Wprovide no dependence of Si(s.) + xF ^ SiFx (where index (s.) points out the particle situated on the surface) reaction probability on the efficiency of ion-induced production of adsorption sites for fluorine atoms, as SiOx(s.) ^ Si(s.) + xO. At lower bias powers, the presence of such dependence is confirmed by similar changes of effective reaction probability and ion bombardment intensity, traced by the multiplication of ion flux on square root of ion energy. Some suggestions concerning peculiarities of both gas-phase and heterogeneous process kinetics at low plasma densities were made.
Key words: C4F8, CF4, plasma, parameters, active species, ionization, dissociation, etching
Для цитирования:
Ефремов А.М., Kwon K.-H. Состав плазмы и кинетика травления SiO2 в смеси CF4/C4Fs/Ar/He: влияние соотношения CF4/C4F8 и мощности смещения. Изв. вузов. Химия и хим. технология. 2022. Т. 65. Вып. 10. С. 47-53. DOI: 10.6060/ivkkt.20226510.6604.
For citation:
Efremov A.M., Kwon K.-H. Plasma composition and S1O2 etching kinetics in CF4/C4Fs/Ar/He mixture: effects of CF4/C4F8 mixing ratio and bias power. ChemChemTech [Izv. Vyssh. Uchebn. Zaved. Khim. Khim. Tekhnol.]. 2022. V. 65. N 10. P. 47-53. DOI: 10.6060/ivkkt.20226510.6604.
INTRODUCTION
Fluorocarbon gases with a general formula of CxFy play a remarkable role in modern electronic device technology being used for dry (plasma-assisted) patterning of silicon family compounds [1-3]. The most effective tool to obtain nano-scale feature sizes is the reactive-ion etching (RIE) process which combines physical (physical sputtering) and chemical (ion-assisted heterogeneous reactions) etching pathways. Accordingly, the adjustment of these two components in most of cases provides the desirable balance between various output RIE characteristics, such as etching rate, etching selectivity in respect to the mask material as well as the degree of anisotropy which controls shapes of etching profiles. From many published works (see, for example, Refs. [4-6]), it is known also that the etching performance for the given fluorocarbon gas strongly depends on the y/x ratio that pre-determines densities of F atoms and polymerizing CFx (x < 2) radicals [5, 6]. In particular, the CF4 plasma (y/x = 4) represents the system with low polymerizing ability and thus, provides the domination of etching over the surface polymerization. The fea-
tures of such plasmas are high etching rates and surface cleanness, nearly-isotropic etching profiles and bad SiO2/Si etching selectivity [6, 7]. Oppositely, the C4F8 plasma (y/x = 2) is the case of high polymerizing gas systems. Accordingly, it is characterized by lower etching rates, but provides the decent etching anisot-ropy (due to the protection of sidewalls by the deposited polymer layer) and SiO2/Si selectivity (due to the thinner polymer film on the oxygen-containing surface) [3, 4, 8].
The overall trend in the modern RIE technology is the use of multi-component gas mixtures combining one or several fluorocarbon components with noble and/or molecular additives. The reason is that the variation of gas mixing ratio offers the additional tool to optimize output RIE characteristics through changes in both gas-phase and heterogeneous reaction kinetics. In our previous works [9-13], we carried out detailed investigations of gas mixing effects in binary and ternary gas systems composed by CF4, C4F8 and Ar under conventional RIE conditions (gas pressures below 50 mtorr, input power densities of 0.50.1 W/cm3 that correspond to the plasma density of ~ 1010-1011 cm-3, and ion bombardment energies above
200 eV). Corresponding results may briefly be summarized as follows:
1) An increase in Ar fraction in both CF4 +Ar and C4F8 + Ar plasmas changes electrons- and ions-related plasma parameters and thus, influences both gas-phase and heterogeneous reaction kinetics. Most principal effect here are a) the elevation of electron temperature and plasma density; b) the increasing efficiency of electron-impact processes that results in slower decrease of F atoms density [9, 10]; and c) the growth of ion bombardment intensity that accelerates ion-driven heterogeneous processes, such as surface bond breaking, the desorption of low volatile reaction products and the destruction of fluorocarbon polymer film. As such, the change in the effective probability of ion-assisted chemical reaction represents the superposition of all above effects [11, 12].
2) The substitution of C4F8 for CF4 in the C4F8 + + CF4 + Ar plasma exhibits much weaker effect on both electron temperature and plasma density, but directly influences densities of F atoms and CFx radicals through corresponding source species. In this case, the ion-assisted chemical reaction occurs at the nearly constant ion bombardment intensity while the effective reaction probability traces the polymer film thickness.
Unfortunately, the existing data cannot be applied directly to explain plasma chemical phenomena in more complicated gas systems and/or under different processing conditions. The reason is that the introduction of new component as well as the change in discharge excitation regime disturbs both electron energy distribution and charged species balance. This causes the mandatory response from chemical kinetics of all gas-phase components and thus, results in the multi-channel influence on densities of plasma active species. Therefore, the study of plasma parameters and composition in new gas systems and/or processing regimes is the mandatory thing to understand the performance and the technological potential for given RIE process.
The subject of this work was the CF4 + C4F8 + + Ar + He plasma excited in the low (~ 0.05 W/cm3) input power mode. The interest to such regime is due to the possibility to obtain higher etching anisotropy with lower surface damages [14-16]. As variable parameters, we selected the CF4/C4F8 mixing ratio and the bias power that directly influence the negative bias potential. According to previously published data, it can be expected that these ones represent somewhat classical "chemical" and "physical" factors influencing heterogeneous stages of the etching process. The silicon dioxide, as the etched material, was a kind
of test object with the well-known etching mechanism under conventional RIE conditions. Therefore, the analysis of corresponding etching kinetics allows one better understanding the features of the low-power RIE process.
EXPERIMENTAL DETAILS
Experiments were carried out in the inductively coupled plasma (ICP) reactor known from our previous works [10-13]. Plasma was excited using the 13.56 MHz power supply connected to the flat copper coil on the top side of reactor chamber. Another 13.56 MHz RF generator powered the bottom electrode to produce the negative bias potential, -Udc. Constant processing parameters were gas pressure (p = 10 mtorr, or 1.33 Pa), input power (W = 50 W that corresponded to ~ 0.05 W/cm3) and total gas flow rate (q = 114 sccm). Accordingly, variable inputs were represented by bias power (Wdc = 200-500 W) and initial composition of CF4 + C4F8 + Ar + He gas mixture. The latter was set by partial flow rates for CF4 and C4F8 within the total value of 14 sccm at constant q(Ar) = 25 sccm and q(He) = 75 sccm. Therefore, an increase in q(CF4) in the range of 0-14 sccm corresponded to the full substitution of C4F8 for CF4, and the fraction of CF4 in a feed gas y(CF4) = q(CF4)/q was changed as 0-12%.
Gas-phase plasma parameters were measured using the double Langmuir probe tool (DLP2000, Plasmart Inc.). The treatment of raw I-V curves with accounting for well-known statements of Langmuir probe theory in low pressure plasmas [17, 18] yielded data on electron temperature (Te) and ion current density (J+). In order to reduce experimental errors due to the deposition of fluorocarbon polymer on probe tips, these were conditioned in 50% Ar + 50% O2 plasma for ~ 5 min before and after each experiment.
The negative bias potential, -Udc, was monitored using the high-voltage probe (AMN-CTR, Youngsin Eng.).
Steady-state densities of F atoms were determined using plasma diagnostics by the optical emission spectroscopy (OES) (AvaSpec-3648, JinYoung Tech). For this purpose, we measured emission intensities of well-known actinometrical lines F 703.8 nm and Ar 750.4 nm which are featured by a) direct electron impact excitation mechanisms with known process cross-sections [19]; and b) low radiational lifetimes. Taking into account the constant and known density of Ar atoms in both feed gas and under plasma conditions, one can apply the standard actinomet-rical approach in a form of
I(F)/I(Ar) = CF,Ar[F]/[Ar], (1)
where I is the measured emission intensity, and CF,Ar
is the coefficient combining corresponding excitation rate coefficients kex = f(Te) and transition probabilities [19]. The last work demonstrated also that a) Cf,at « const at Te = 3-4 eV; and b) the F atoms density determined from Eq. (1) has a good agreement with mass-spectrometry experiments.
Etching kinetics of SiO2 was studied using fragments of thermally oxidized Si (111) wafers. Treated samples with an average size of ~ 2^2 cm were placed in the middle part of the bottom electrode. The bottom electrode has the built-in water-flow cooling system which allows one to stabilize its
temperature at the nearly constant value of ~ 17 °C. In order to determine SiO2 etching rates, we covered a part of sample surface by the photoresist mask as well as measured the height of the step Ah between masked and non-masked areas using the surface profiler Alpha-Step 500 (Tencor). In preliminary experiments, it was found that all kinetic curves Ah = f(x) exhibited nearly linear shapes within processing times x up to 5 min. Since this surely points out on the steady-state etching regime, etching rates were simply calculated as R = Ah/x.
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
y(cf4), % y(cf4), %
Fig. 1. Electrons- and ions-related plasma characteristics as functions of CF4 fraction in a feed gas (at Wdc = 400 W, curve 1) and bias power (at y(CF4) = 10%, curve 2): a) electron temperature; b) ion current density; c) negative bias potential; and d) parameter Si1/2r+,
characterizing the ion bombardment intensity Рис. 1. Характеристики электронной и ионной компонент плазмы в зависимости от доли CF4 в плазмообразующем газе (при Wdc = 400 Вт, кривая 1) и мощности смещения (при y(CF4) = 10%, кривая 2): а) температура электронов; б) плотность ионного тока; с) отрицательный потенциал смещения; и д) параметр Si1/2r+, характеризующий интенсивность ионной бомбардировки
RESULTS AND DISCUSSION
Fig. 1 represents effects of CF4/C4F8 mixing ratio and bias power on electrons- and ions-related gas-phase characteristics. The analysis of these data with accounting for the previous research experience of CF4 + Ar [9, 13], C4F8 + Ar [9-11, 13] and CF4 + + C4F8 + Ar [10-12] plasmas allows one to formulate several important features which may be useful for
the optimization of corresponding RIE process. These are as follows:
1) Similarly to conventional RIE conditions [10, 11], an increase in CF4 fraction in a feed gas causes rather weak changes in both electron temperature (Fig. 1(a)) and ion current density (Fig. 1(b)). Perhaps, the general reason is that the substitution of C4F8 for CF4 does not results in global changes in the electron energy loss channels (and thus, in the
electron energy distribution function) as well as does not disturb formation/decay balances for both ion and electrons. On this background, the weak decrease in Te may be associated with increasing fractions of more saturated and bigger-sized CFx species (for instance, CF4 and CF3, instead of CF2 and CF which dominate in C4F8-based plasmas [10-12]) characterized by higher cross-sections and lower threshold energies for electron-impact excitation and ionization processes. Accordingly, this also causes the growth of total ionization rate, densities of charged species and the parameter J+, as mentioned in Fig. 1(b). It can be seen also, that the combination of nearly constant bias potential (Fig. 1(c)) and increasing ion flux r+ « J+/e provide the similar monotonic change in the parameter Si1/2r+ (Fig. 1(d)), where Si = e(-Udc-Uf) is the ion bombardment energy, and Uf = f(Te) [4] is the floating potential. As the sputtering yield is generally proportional to the momentum transferred from the incident ion to the surface atom [4], the multiplication of Si1/2r+ traces the ion bombardment intensity [9, 10]. Therefore, the somewhat activation of ion-driven heterogeneous processes also takes place.
2) An increase in bias power does not disturb both electron temperature and plasma density (as it follows from the change of J+) that surely suggests no influence on densities of neutral species. At the same time, it results in the nearly proportional growth of negative bias potential (Fig. 1(c)), so that one can obtain the enforcement of the ion bombardment (Fig. 1(d)) even under the condition of r+ « const. It is important to note that the corresponding change in Si1/2r+ value (and thus, in the effect on ion-driven heterogeneous processes) exceeds that for the CF4/C4F8 mixing ratio.
From plasma emission spectra, it was found that emission intensities for F 703.8 nm and Ar 750.4 nm remain to be nearly constant vs. Wdc, but demonstrate the monotonic increase with increasing y(CF4) (Fig. 2(a)). In particular, the IAr = f(Wdc) curve surely means the constancy of corresponding excitation function kexne, as it would be expected from mentioned changes in Te and J+. Accordingly, the same behavior of IF, as well as the condition IF/IAr = const allow one to assume no effect of bias power on the F atom density. Obviously, such situation results from the nearly constant F atom formation rate in R1: CFx + + e ^ CF x-1 + F + e due to the quite close k1 values for various CFx species. It should be noted that the weak increase in [F] toward higher bias powers, as shown in Fig. 2(b), is within the standard experimental error. However, the physical reason may be the slightly increasing electron density, as follows
from the behavior of J+. On the contrary, the If = f(y(CF4)) curve reflects the overall effect from both F atom density and excitation function. The real [F] value extracted from Eq. (1) demonstrates the non-linear growth toward CF4-rich plasmas as well as tends to be saturated at maximum y(CF4). The identical behavior of F atom density have been obtained by plasma modeling of CF4 + C4F8 + Ar gas system under conventional RIE conditions [11, 12]. In fact, this suggests the similarity of general plasma chemical phenomena in high- and low- input power modes as well as assumes the same explanation for the change in [F]. According to Refs. [11-13], this is a combination of the nearly constant F atom formation rate in R1 and the decreasing F atom loss frequency in R2: C2F4 + F ^ CF2 + CF3 which dominates over the heterogeneous loss in C4F8-rich plasmas. Therefore, the main feature of the low input power mode is only the lower F atom density.
From etching experiments, it was found that the SiO2 etching rate increases slightly vs. y(CF4) as well as demonstrates the rapid growth toward higher bias powers (Fig. 2(c)). From Refs. [12, 20], it can be understood that the total rate of RIE process may be represented in a form of two summands, Rphys + Rchem. These components characterize parallel etching pathways, such as physical sputtering and ion-assisted chemical reaction. The evaluation of Rphys « YsF+ using the experimental data on SiO2 sputtering yields (Ys « 0.4 atom/ion for 0-12% CF4 and 0.3-0.45 atom/ion for Wdc = 200-500 W [21, 22]) indicated that the maim contribution to the measured etching rate belongs to Rchem (Fig. 2(c)). The effective probability of ion-assisted chemical reaction yR = Rchem/rF (Fig. 2(d)), where Tf is the flux of fluorine atoms, exhibits no correlation with the fluorocarbon film thickness, as the latter decreases toward CF4-rich plasmas. The last fact evidently follows from decreasing polymer deposition rate (due to much lower polymerizing ability of CF4 compared with C4F8 [4, 5, 9]) and increasing polymer destruction rate (due to increasing ion bombardment intensity). Therefore, the obtained situation contradicts with the general rule "the thinner film, the higher reaction probability" mentioned by both experiment and modeling for many fluorocarbon gas plasmas, including C4F8 + Ar [10, 12] and C4F8 + CF4 + + Ar [11, 12] gas systems. In our opinion, the reason is not the feature of low-power plasma excitation mode, but simply the low content of fluorocarbon components in the gas mixture. Accordingly, this produces the very thin or even the non-continuous polymer film which does not influence the SiO2 etching kinetics. Another remarkable fact is that the
change of yR does not correlate with the ion bombardment intensity at Wdc > 400 W. In fact, this means that the ion-induced production of adsorption sites for F atoms R3: SiOx(s.) ^ Si(s.) + xO (where index (s.) points out the particle situated on the surface) does not limit the rate of chemical reaction R4: Si(s.) + xF ^ SiFx. Perhaps, the latter occurs in the F-atom-deficient regime due to high ion bombardment intensity and low F atom flux, as a result of low plasma density in a low input power mode. Accordingly, the small decrease in yR toward higher y(CF4) values may be due to the ion-induced desorption of F atoms prior their interaction with silicon in R4. Therefore, the low input power RIE mode somewhat equal-
ized etching mechanisms for Si and SiO2. At the same time, at bias powers below 400 W, the effective reaction probability follows the ion bombardment intensity (Fig. 2(d)). Such situation is quite typical for conventional SiO2 RIE process in low polymerizing plasmas. It was found also that the parameter rF/Si1/2r+ (the so-called neutral/charged ratio) characterizing the etching anisotropy exhibits the weak sensitivity to y(CF4) (1.4-1.6 for 0-12% CF4 at Wdc = 400 W), but decreases by ~ 1.5 times with increasing bias power. As such, similarly to conventional RIE conditions, Wdc represents the most effective tool to increase the "directional" component of the etching process and thus, to adjust the shape of etching profile.
Bias power, W 300 400
Bias power, W 300 400
S 10
ii
E 1.1
Bias power, W 300 400
Bias power, W 300 400
^ 1 "VT^-y ■ --' 1 """" t /
2 /
■/2"
c)
■ 1 r^ -uff-*1 '—-
2 / d)
0 2 4 6 8 10 12 14 y(CF4), %
468 y(CF4), %
10 12 14
4 6 8 y(CF„), %
468 y(CF„), %
10 12 14
Fig. 2. Densities of F atoms and SiO2 etching kinetics as functions of CF4 fraction in a feed gas (at Wdc = 400 W, curve 1) and bias power (at y(CF4) = 10%, curve 2): a) emission intensities for Ar 750.4 nm (solid lines) and F 703.8 nm (dashed lines) maxima; b) F atom density; c) SiO2 etching rate (dashed satellite curves represent Rchem); and d) effective reaction probability Рис. 2. Концентрации атомов фтора и кинетика травления SiO2 в зависимости от доли CF4 в плазмообразующем газе (при Wdc = 400 Вт, кривая 1) и мощности смещения (при y(CF4) = 10%, кривая 2): а) интенсивности эмиссионных максимумов Ar 750,4 нм (сплошные линии) и F 703,8 нм (пунктир); б) концентрации атомов F; в) скорость травления SiO2 (пунктирные кривые-спутники представляют Rchem); и г) эффективная вероятность взаимодействия
200
500
200
500
200
500
200
500
5
40
30
20
0
0
0
2
0
2
0 12 14
0
2
CONCLUSIONS
In this work, we investigated gas-phase characteristics and reactive-ion etching (RIE) kinetics of SiO2 in CF4/C4Fs/Ar/He plasma with variable CF4/C4F8 mixing ratio and bias power in the low (~ 0.05 W/cm3) input power mode. It was found that the substitution of C4F8 for CF4 does not produce sufficient changes in both electrons- and ions-related plasma parameters, but causes a weak increase in fluorine atom density. The last effect is the same with that obtained for conventional RIE conditions and probably results from decreasing loss frequency for F atoms in C2F4 + F ^ CF2 + CF3. The increasing bias power has no influence on gas-phase plasma characteristics, but enforces the ion bombardment due to the change in the negative bias potential. From etching experiments, it can be understood that a) the measured SiO2 etching rate is mainly composed by its chemical component; b) the thin or even the non-continuous fluorocarbon polymer film does not limit
the access of F atoms to the etched surface; and c) the condition Wdc > 400 W sufficiently reduces the influence of ion-induced process SiOx(s.) ^ Si(s.) + xO on the Si(s.) + xF ^ SiFx reaction kinetics. Perhaps, the last feature is due to the combination of high ion bombardment intensity and low F atom flux, as a result of low plasma density in a low input power mode. On the contrary, at lower bias powers, the effective reaction probability correlates with the ion bombardment intensity.
The work was supported by the Russian Science Foundation grant № 22-29-00216, https://rscf.ru/ ■ project/22-29-00216/.
The authors declare the absence a conflict of interest warranting disclosure in this article.
Исследование выполнено за счет гранта Российского научного фонда № 22-29-00216, https://rscf.ru/project/22-29-00216/.
Авторы заявляют об отсутствии конфликта интересов, требующего раскрытия в данной статье.
REFERENCES ЛИТЕРАТУРА
1. Wolf S., Tauber R.N. Silicon Processing for the VLSI Era. V. 1. Process Technology. New York: Lattice Press. 2000. 416 p.
2. Nojiri K. Dry etching technology for semiconductors. Tokyo: Springer Internat. Publ. 2015. 116 p. DOI: 10.1007/9783-319-10295-5.
3. Advanced plasma processing technology. New York: John Wiley & Sons Inc. 2008. 479 p.
4. Lieberman M.A., Lichtenberg A.J. Principles of plasma discharges and materials processing. New York: John Wiley & Sons Inc. 2005. 757 p. DOI: 10.1002/0471724254.
5. Standaert T.E.F.M., Hedlund C., Joseph E.A., Oehrlein G.S., Dalton T.J. Role of fluorocarbon film formation in the etching of silicon, silicon dioxide, silicon nitride, and amorphous hydrogenated silicon carbide. J. Vac. Sci. Technol. A. 2004. V. 22. P. 53-60. DOI: 10.1116/1.1626642.
6. Schaepkens M., Standaert T.E.F.M., Rueger N.R., Sebel P.G.M., Oehrlein G.S., Cook J.M. Study of the SiO2-to-Si3N4 etch selectivity mechanism in inductively coupled flu-orocarbon plasmas and a comparison with the SiO2-to-Si mechanism. J. Vac. Sci. Technol. A. 1999. V. 17. P. 26-37. DOI: 10.1116/1.582108.
7. Kastenmeier B.E.E., Matsuo P.J., Oehrlein G.S. Highly selective etching of silicon nitride over silicon and silicon dioxide. J. Vac. Sci. Technol. A. 1999. V. 17. P. 3179-3184. DOI: 10.1116/1.58209.
8. Donnelly V.M., Kornblit A. Plasma etching: Yesterday, today, and tomorrow. J. Vac. Sci. Technol. 2013. V. 31. P. 050825-48. DOI: 10.1116/1.4819316.
9. Efremov A., Murin D., Kwon K.-H. Concerning the Effect of Type of Fluorocarbon Gas on the Output Characteristics of the Reactive-Ion Etching Process. Russ. Microelectronics. 2020. V. 49. N 3. P. 157-165. DOI: 10.1134/S1063739720020031.
10. Efremov A., Murin D., Kwon K.-H. Plasma parameters, densities of active species and etching kinetics in C4F8+Ar gas mixture. ChemChemTech [Izv. Vyssh. Uchebn. Zaved. Khim. Khim. Tekhnol.]. 2019. V. 62. N 2. P. 31-37. DOI: 10.6060/ivkkt.20196202.5791.
Ефремов А.М., Мурин Д.Б., Квон К.Х. Параметры плазмы, концентрации активных частиц и кинетика травления в смеси C4F8+Ar. Изв. вузов. Химия и хим. технология. 2019. Т. 62. Вып. 2. С. 31-37. DOI: 10.6060/ivkkt.20196202.5791.
11. Efremov A.M., Murin D.B., Kwon K.-H. Plasma parameters and active species kinetics in CF4+C4F8+Ar gas mixture.
ChemChemTech [Izv. Vyssh. Uchebn. Zaved. Khim. Khim. Tekhnol.]. 2018. V. 61. N 4-5. P. 31-36. DOI: 10.6060/tcct.20186104-05.5695.
Ефремов А.М., Мурин Д.Б., Квон К.Х. Параметры плазмы и кинетика активных частиц в смеси CF4+C4F8+Ar. Изв. вузов. Химия и хим. технология. 2018. Т. 61. Вып. 4-5. С. 31-36. DOI: 10.6060/tcct.20186104-05.5695.
12. Efremov A., Lee B. J., Kwon K.-H. On Relationships Between Gas-Phase Chemistry and Reactive-Ion Etching Kinetics for Silicon-Based Thin Films (SiC, SiO2 and SixNy) in Multi-Component Fluorocarbon Gas Mixtures. Materials. 2021. V. 14. P. 1432(1-27). DOI: 10.3390/ma14061432.
13. Lim N., Efremov A., Kwon K.-H. A comparison of CF4, CHF3 and C4F8 + Ar/O2 Inductively Coupled Plasmas for Dry Etching Applications. Plasma Chem. Plasma Process. 2021. V. 41. P. 1671-1689. DOI: 10.1007/s11090-021-10198-z.
14. Veselov D.S., Bakun A.D., Voronov Yu.A. Reactive ion etching of silicon using low-power plasma etcher. J. Phys.: Conf. Ser. 2016. V. 748. P. 012017(1-4). DOI: 10.1088/1742-6596/748/1/012017.
15. Ashraf M., Sundararajan S.V., Grenc G. Low-power, low-pressure reactive-ion etching process for silicon etching with vertical and smooth walls for mechanobiology application. J. Micro/Nanolith. Mems Moems. 2017. V. 16. P. 034501(1-8). DOI: 10.1117/1.JMM.16.3.034501.
16. Osipov A.A., Aleksandrov S.E., Solov'ev Yu.V., Uvarov A.A. Etching of SiC in low power inductively-coupled plasma. Russ. Microelectronics. 2018. V. 47. N 6. P. 427-433. DOI: 10.1134/S1063739719010074.
17. Johnson E.O., Malter L. A floating double probe method for measurements in gas discharges. Phys. Rev. 1950. V. 80. P. 58-70. DOI: 10.1103/PhysRev.80.58.
18. Shun'ko E.V. Langmuir probe in theory and practice. Boca Raton: Universal Publ. 2008. 245 p.
19. Lopaev D.V., Volynets A.V., Zyryanov S. M., Zotovich A.I., Rakhimov A.T. Actinometry of O, N and F atoms. J. Phys. D: Appl. Phys. 2017. V. 50. P. 075202 (1-17). DOI: 10.1088/1361 -6463/50/7/075202.
20. Gray D.C., Tepermeister I., Sawin H.H. Phenomenological modeling of ion-enhanced surface kinetics in fluorine-based plasma-etching. J. Vac. Sci. Technol. B. 1993. V. 11. P. 1243-1257. DOI: 10.1116/1.586925.
21. Seah M.P., Nunney T.S. Sputtering yields of compounds using argon ions. J. Phys. D: Appl. Phys. 2010. V. 43. P. 253001 (1-24). DOI: 10.1088/0022-3727/43/25/253001.
22. Chapman B. Glow Discharge Processes: Sputtering and Plasma Etching. New York: John Wiley & Sons Inc. 1980. 432 p.
Поступила в редакцию 15.03.2022 Принята к опубликованию 15.06.2022
Received 15.03.2022 Accepted 15.06.2022
