CC BY
12Т 61 (4-5)
ИЗВЕСТИЯ ВЫСШИХ УЧЕБНЫХ ЗАВЕДЕНИИ. Серия «ХИМИЯ И ХИМИЧЕСКАЯ ТЕХНОЛОГИЯ»
2018
V 61 (4-5)
IZVESTIYA VYSSHIKH UCHEBNYKH ZAVEDENIY KHIMIYA KHIMICHESKAYA TEKHNOLOGIYA
RUSSIAN JOURNAL OF CHEMISTRY AND CHEMICAL TECHNOLOGY
2018
DOI: 10.6060/tcct.20186104-05.5695
УДК: 537.525
ПАРАМЕТРЫ ПЛАЗМЫ И КИНЕТИКА АКТИВНЫХ ЧАСТИЦ В СМЕСИ CF4+C4F8+Ar
Александр Михайлович Ефремов*, Дмитрий Борисович Мурин
Кафедра технологии приборов и материалов электронной техники, Ивановский государственный химико-технологический университет, Шереметевский пр., 7, Иваново, Российская Федерация, 153000 E-mail: efremov@isuct.m*
Институт термодинамики и кинетики химических процессов Кванг Хо Квон
Лаборатория применения плазмы, Департамент разработки средств и методов контроля, Университет Кореи, 208 Сеочанг-Донг, Чочивон, Республика Корея, 339-800.
В данной работе обсуждаются взаимосвязи между начальным составом смеси CF4 + C4F8 + Ar, характеристиками газовой фазы и кинетикой гетерогенных процессов в условиях плазмы индукционного разряда пониженного давления. Целью работы являлось исследование зависимостей внутренних параметров плазмы (температура электронов, концентрация электронов, энергия ионной бомбардировки) и кинетики активных частиц от соотношения компонентов CF4/C4F8 в плазмообразующей смеси, а также выявление механизмов влияния указанных параметров на такие характеристики «сухого» травления, как скорость травления и селективность. Исследования проводились методами диагностики плазмы двойным зондом Лангмюра и моделирования плазмы в условиях планар-ного индукционного плазмохимического реактора. Эксперименты и расчеты проводились при постоянном давлении газа (10 мтор), вкладываемой мощности (800 Вт) и мощности смещения (150 Вт), при этом отношение CF4/C4F8 варьировалось изменением парциальных скоростей потока этих газов. Было найдено, что замещение CF4 на C4F8 в плазмооб-разующем газе CF4 + C4F8 + Ar приводит к снижению скорости генерации атомов фтора и плотности их потока на обрабатываемую поверхность из-за снижения концентрации в объеме плазмы. Предположено, что рост концентрации и плотности потока ненасыщенных радикалов CFX (x=1,2) в смесях с большим содержанием C4F8 при близких к постоянным значениях плотности потока энергии ионов (т.е. близкой к постоянной эффективности ионной бомбардировки) способствует снижению эффективной вероятности взаимодействия атомов фтора за счет увеличения толщины фторуглеродной полимерной пленки на обрабатываемой поверхности.
Ключевые слова: CF4, C4F8, скорость реакции, энергия ионов, концентрация, поток, травление, полимеризация
А.М. Ефремов, Д.Б. Мурин, К.Х. Квон
PLASMA PARAMETERS AND ACTIVE SPECIES KINETICS IN CF4+C4F8+Ar GAS MIXTURE
A.M. Efremov, D.B. Murin, K.H. Kwon
Alexander M. Efremov*, Dmitriy B. Murin
Department of Electronic Devices and Materials Technology, Ivanovo State University of Chemistry and Technology, Sheremetevskiy ave., 7, Ivanovo, 153000, Russia.
E-mail: efremov@isuct.ru*
Kwang H. Kwon
Plasma Application Lab., Dept. of Instrumentation and Control Engineering, Korea University, 208 Seochang-
Dong, Chochiwon, Korea, 339-800.
This work discusses the relationships between the initial composition of the CF4 + C4F8 + Ar gas mixture, gas-phase characteristics and heterogeneous process kinetics under the condition of low-pressure inductively coupled plasma. The goals were to investigate how the CF4/C4F8 mixing ratio influences internal plasma parameters (electron temperature, electron density and ion bombardment energy) and kinetics of plasma active species as well as to analyze how the changes in above parameters may influence the dry etching characteristics, such as etching rates and selectivi-ties. The investigation was carried out using the combination of plasma diagnostics by double Lang-muir probes and 0-dimensional plasma modeling. Both experiments and calculations were carried out at constant gas pressure (10 mTorr), input power (800 W) and bias power (150 W) while the CF4/C4F8 mixing ratio was varied through the partial flow rates for corresponding gases. It was shown that the substitution of CF4 for C4F8 in the CF+CF+Ar feed gas lowers F atom formation rates and causes the decreasing F atom flux to the treated surface due to decreasing their volume density. It was proposed that an increase in the densities and fluxes of unsaturated CFx (x=1,2) radicals toward C4F8-rich plasmas at the nearly constant ion energy flux (i.e. at the nearly constant efficiency of ion bombardment) causes a decrease in the effective reaction probability for F atoms through the increasing thickness of the fluorocarbon polymer film on the treated surface.
Key words: CF4, C4F8, reaction rate, ion energy, density, flux, etching, polymerization
Для цитирования:
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 For citation:
Efremov A.M., Murin D.B., Kwon K.H. Plasma parameters and active species kinetics CF4+C4F8+Ar gas mixture. Izv. Vyssh. Uchebn.
Zaved. Khim. Khim. Tekhnol. 2018. V. 61. N 4-5. P. 31-36
INTRODUCTION
Fluorocarbon gases are widely used in the microelectronic industry for dry patterning of silicon wafers and dielectric (SiO2, Si3N4) thin films [1,2]. Among these, the CF4 is characterized by the highest F/C ratio and provides the domination of etching over the surface polymerization process under the typical reactive ion etching conditions [3, 4]. The more polymerizing fluo-rocarbons (C4F6, C4F8, CHF3 and CH2F2) are normally used for the etching processes which require as much as possible SiO2/Si etching selectivity. When both Si and SiO2 etching mechanisms are sufficiently affected by the solid-state diffusion of etchant species through the fluorocarbon polymer film, the etching rates are rather sensitive to the film thickness than to the flux of F atoms coming from bulk plasma. Accordingly, since the thickness of the polymer film on the SiO2 appears to be lower
(due to the destruction of polymer in the reactions with surface oxygen atoms [2,5]), the SiO2 etching rate exceeds that for Si. At the same time, together with increasing SiO2/Si etching selectivity, the decrease in absolute Si and SiO2 etching rates as well as an increase in etching residues take place [3-5]. These facts make difficulties for etching process optimization.
One can expect that the reasonable balance between SiO2/Si etching selectivity, absolute etching rates and etching residues may be achieved by the use of two fluorocarbon gases in one gas mixture. In such gas system, the less polymerizing component (for example, CF4) provides the effective generation of etchant species while the more polymerizing component creates the favorable conditions for obtaining high SiO2/Si etching selectivity. From the other side, the mixing of CF4 with other fluorocarbon gas mandatory results in more
Изв. вузов. Химия и хим. технология. 2018. Т. 61. Вып. 4-,
complicated reaction scheme, changes plasma parameters as well as influences the formation/decay kinetics and fluxes for all types of plasma active species. Therefore, an understanding of plasma chemistry in the gas systems with two fluorocarbon components is the important tasks for the correct determination of etching mechanisms and optimal choice of gas chemistry for the given etching process.
The goals of current work were 1) to investigate how the CF4/C4F8 mixing ratio in CF4+C4F8+Ar inductively coupled plasma influences internal plasma characteristics (electron temperature, electron density and ion bombardment energy) as well as the kinetics of plasma active species, their densities and fluxes; and 2) to analyze how the changes in above parameters may influence the dry etching characteristics, such as etching rates and selectivities.
EXPERIMENTAL AND MODELING DETAILS
Plasma diagnostics experiments were performed in a planar inductively coupled plasma (ICP) reactor described in our previous works [6,7]. The experiments were performed at a fixed total gas flow rate (q = 60 sccm), gas pressure (p = 10 mTorr), input power (W = 800 W) and bias power (Wdc = 150 W). The initial compositions of CF4+C4F8+Ar gas mixture were set by adjusting the flow rates of the corresponding gases. Particularly, the Ar flow rate qAr was fixed at 20 sccm, so that the fraction of Ar in the feed gas yAr = qAr/q was always 33%. The fluorocarbon gases were mixed at various ratios within qCp4 + qc4F8 = 40 sccm while the maximum flow rate for C4F8 did not exceed 15 sccm. Accordingly, the maximum fraction of C4F8 in the CF4+C4F8+Ar gas mixture reached 25%.
Plasma parameters were measured by double Langmuir probe (DLP2000, Plasmart Inc., Korea). The treatment of I - V curves aimed at obtaining electron temperature (Te) and ion saturated current density (j+) was carried out using the software supplied by the equipment manufacturer. The calculations were based on the Johnson & Malter's double probes theory [8] with the one-Maxwellian approximation for the electron energy distribution function (EEDF). The total positive ion density (n+) was extracted from the measured j+ using the Allen-Boyd-Reynolds (ABR) approximation [9].
In order to obtain the densities of neutral species, we developed a simplified zero-dimensional kinetic model [6, 10, 11] with using the data of Te and n+ as input parameters. The set of chemical reactions was taken from our previous work [6]. Since the latter provides the detailed discussion on both kinetic schemes and sources of chemical kinetics data, these issues were not the subjects of current study. The model used following assumptions: 1) The electron energy distribution function
5
(EEDF) is close to Maxwellian one. The applicability of Maxwellian EEDFs for the description of the electron-impact kinetics for CF4-based low-pressure (p < 50 mTorr) ICPs has been confirmed by the reasonable agreement between the diagnostic results and modeling [11,12]; 2) Under the given set of process conditions, the electronegativity of CF4+C4F8+A plasma with more than 30% fraction of electropositive components is low enough to assume n_ << ne » n+. The reasonability of such approach for the CF4- and C4Fg-based ICPs was confirmed in several works by both modeling and experiment [6, 12, 13]; 3) The heterogeneous chemistry of atoms and radicals can be described in terms of the conventional first-order recombination kinetics [6,11]; and 4) The temperature of the neutral ground-state species (Tgas) is independent on the feed gas composition. Since the experimental data on gas temperature were not available in this study, we took Tgas = 600 K as the typical value for the ICP etching reactors with similar geometry under the close range of experimental conditions [11, 13]. At least, our previous works [6, 7, 11] made with the same set of experimental equipment showed the reasonable agreement between measured and model-predicted plasma parameters for Tgas = 600 K. Also, the test model runs indicated no principal differences in gas-phase densities of neutral species obtained from the calculations within the Tgas uncertainty ~ 100 K. Such result looks quite reasonable since the dominant decay channels for atoms and radicals under the low pressure conditions are the heterogeneous processes.
For the analysis of heterogeneous chemistry, the fluxes for each king of neutral species with the volume density n were calculates as r « 0.25nvT, where
vT = (8kBTgas/nm% . The total flux of positive ions was simply evaluated as r+ « j+/e.
RESULTS AND DISCUSSION
The substitution of CF4 for C4F8 at p, W and Wdc= const results in slightly increasing Te (3.7-3.9 eV 0-25% C4F8), but suppresses the values of both j+ (1.11-0.97 mA/cm2 for 0-25% C4F8) and n+ « ne (4.5-1010-3.8-1010 cm-3 for 0-25% C4F8). The corresponding data are shown in Fig. 1. The behavior of electron temperature is probably connected with the decrease in electron energy loss for the electronic excitation and ionization of the dominant neutral species. As for the decreasing tendencies for n+ and
j+, it may
be caused by the combination of following reasons. First, since the addition of C4F8 results in only weak change of Te, the ionization rate coefficients, kiz, for all types of neutral species may be assumed to be independent on CF4/C4F8 mixing ratio. According to Ref. [14], the absolute values of kiz for CF4, CF2, CF2 and C2F4, which are the dominant neutral species in CF4
and C4F8 plasmas, are rather close. That is why, one cannot expect an increase in the effective ionization frequency toward increasing C4F8 content in a feed gas. And secondly, under one and the same operating conditions the electronegativity of the C4F8 plasma is lower than that for CF4 one [12,14]. Therefore, the substitution of CF4 for C4F8 in the CF4+C4F8+Ar gas mixture leads to increasing electron diffusion coefficient and thus, to increasing electron decay rate on the reactor walls. Accordingly, the decreasing ne results in the same behavior of n+ in order to keep the plasma quasi-neutrality. Physically, the last effect is supported by the decreasing ionization rate.
It was found that, under the given set of operating conditions, the main sources of F atoms in 67% CF4 + 33% Ar plasma are the electron-impact dissociations of CF4 (R1: CF4 + e ^ CF3 + F + e, R2: CF4 + e ^ CF3+ + F + 2e) and CF3 (R3: CF3 + e ^ CF2 + F + e).
4,4
4,2
'4,0
3,8
4,0 -
h"
3,8 -
3,6
3,4
0,00
0,05
0,10
0,15
0,20
0,25
§ 1,0 5!
0,9
0,05 0,10 0,15 0,20 C.F„ fraction in CF.+C.F„+Ar
0,25
b
Fig. 1. Measured plasma parameters: a) - electron temperature (1), total positive ion and electron density (2), б) - negative dc bias (3), ion current density (4) as functions of C4F8 fraction in
CF4+C4F8+Ar gas mixture Рис. 1. Измеренные параметры плазмы: а) - температура электронов (1), суммарная концентрация положительных ионов и электронов (2), б - напряжение отрицательного смещения (3), плотность ионного тока (4) как функции доли C4F8 в плазмообразующей смеси CF4+C4F8+Ar
These processes constitute approximately 85% of the total F atom formation rate while the contribution from the CF2 and CF radicals through R4: CF2 + e ^
CF + F + e, R5: CF2 + e ^ C + 2F + e and R6: CF + e ^ C + F + e does not exceed 5% due to the much lower densities of corresponding species. The remaining 10% comes from R7: F2 + e ^ 2F + e, which is supported by the high F ^ F2 recombination rate on the reactor walls. Accordingly, the decay of F atoms is mainly caused by their heterogeneous recombination while the rate of the fastest bulk process R8: CF3 + F ^ CF4 is ~ 10 times less. As can be seen from Fig. 2, the substitution of CF4 for C4F8 in the CF4+C4F8+Ar gas mixture results in rapidly increasing densities of both C2F4 Oc2f4 = 2.11011-3.4-1013 cm-3, or by ~ 160 times for 0-25% C4F8) and CF2 (nCF2 = 2.8-1012-4.1-1013 cm-3, or by ~ 15 times for 0-25% C4F8) radicals. The reason is that the C2F4 directly appears from C4F8 through R9: C4F8 + e ^ 2C2F4 + e while the CF2 is the main dissociation product of C2F4 in R10: C2F4 + e ^ 2CF2 + e. The increasing density of CF2 radicals lifts up the density of CF3 (nCF/ = 1.3-1013-2.9-1013 cm-3, or by ~ 2.2 times for 0-25% C4F8) mainly due to the increasing CF2 + F ^ CF3 recombination rate both in bulk plasma and on reactors walls. As a result, the contributions of R3 and R4 to the total F atom formation rate exceeds the level of R1 and R2 after 20% C4F8 in CF4+C4F8+Ar gas mixture. Thought the rate coefficients for R3 (8.9-10-101.1-10-9 cm3/s for 0-25% C4F8) and R4 (1.2-10-91.610-9 cm3/s for 0-25% C4F8) exceed the sum of 0k1 + k2) = 5.8-10-10-8.3-10-10 cm3/s, an increase in C4F8 fraction in a feed gas does no lead to increasing F atom formation rate. The reason is that a weak increase in the F atom formation efficiency toward C4F8-rich plasmas is overcome-pensated by decreasing ne. Simultaneously, the substitution of CF4 for C4F8 accelerated the decay of fluorine atoms through CFx+ F ^ CFx+1 (x = 1, 2) as well as introduces the new effective decay channel in R11: C2F4 + F ^ CF2 + CF3 (k&& ~ 4.0-10-11 cm3/s). That is why the F atom density decreases monoton-ically (nF = 1.6-1013-2.5-1012 cm-3, or by ~ 6 times) toward higher contents of C4F8 in CF4+C4F8+Ar gas mixture.
It was found that under the given set of operating conditions the main sources of F atoms in 67% CF4 + 33% Ar plasma are the electron-impact dissociations of CF4 (R1: CF4 + e ^ CF3 + F + e, R2: CF4 + e ^ CF3+ + F + 2e) and CF3 (R3: CF3 + e ^ CF2 + F + e). These processes constitute approximately 85% of the total F atom formation rate while the contribution from the CF2 and CF radicals through R4: CF2 + e ^ CF + F + e, R5: CF2 + e ^ C + 2F + e and R6: CF + e ^ C + F + e does not exceed 5% due to the much lower densities of corresponding species. The remaining 10% comes from R7: F2 + e ^ 2F + e, which is supported by the high F ^ F2 recombination rate on the reactor walls.
C.K fraction in CK+C.K+Ar
a
1,1
0,8
0,00
Изв. вузов. Химия и хим. технология. 2018. Т. 61. Вып. 4-5
Accordingly, the decay of F atoms is mainly caused by their heterogeneous recombination while the rate of the fastest bulk process R8: CF3 + F ^ CF4 is ~ 10 times less.
As can be seen from Fig. 2, the substitution of CF4 for C4F8 in the CF4+C4F8+Ar gas mixture results in rapidly increasing densities of both C2F4 (nC.F/l = =2.1 ■ 1011-3.4-1013 cm-3, or by ~ 160 times for 0-25% C4F8) and CF2 (nCF. = 2.81012-4.11013 cm-3, or by ~ 15 times for 0-25% C4F8) radicals. The reason is that the C2F4 directly appears from C4F8 through R9: C4F8+ + e ^ 2C2F4 + e while the CF2 is the main dissociation product of C2F4 in R10: C2F4 + e ^ 2CF2 + e. The increasing density of CF2 radicals lifts up the density of CF3 (nCF/ = 1.3-1013-2.9-1013 cm-3, or by ~ 2.2 times for 0-25% C4F8) mainly due to the increasing CF2 + F ^ ^ CF3 recombination rate both in bulk plasma and on reactors walls. As a result, the contributions of R3 and R4 to the total F atom formation rate exceeds the level of R1 and R2 after 20% C4F8 in CF4+C4F8+Ar gas mixture. Thought the rate coefficients for R3 (8.9-10-101.110-9 cm3/s for 0-25% C4F8) and R4 (1.210-91.610-9 cm3/s for 0-25% C4F8) exceed the sum of 0k& + k21 = 5.8-10-10-8.3-10-10 cm3/s, an increase in C4F8 fraction in a feed gas does no lead to increasing F atom formation rate. The reason is that a weak increase in the F atom formation efficiency toward C4F8-rich plasmas is overcome-pensated by decreasing ne. Simultaneously, the substitution of CF4 for C4F8 accelerated the decay of fluorine atoms through CFx+ F ^ CFx+1 (x = 1, 2) as well as introduces the new effective decay channel in R11: C2F4 + F ^ CF2 + CF3 (k&& ~ 4.0-10-11 cm3/s). That is why the F atom density decreases monotonically (nF = 1.6-1013-2.5-1012 cm-3, or by ~ 6 times) toward higher contents of C4F8 in CF4+C4F8+Ar gas mixture.
The changes in plasma parameters and composition described above allow one to make some approaches concerning the changes in etching rates, se-lectivities and etching residues. According to Refs. [6,15], the rate of chemical etching pathway in the fluorine-containing plasmas can be expressed through the flux of F atoms as yRrF, where yR is the effective reaction probability. The data of Tab. 1 indicate that the flux of F atoms follows the behaviors of nF and show the same decreasing tendency toward C4F8-rich plasmas (3.5-1017-5.6-1016 cm-2s-1 for 0-25% C4F8). Assuming yR » const at constant surface temperature, this fact obviously points out on the depletion of the chemical etching pathway. However, the above conclusion is valid only when the etching rate is not limited by the transport of F atoms from the fluorocarbon polymer
film. In most of cases, the fluorocarbon polymer film is thick enough to reduce the amount of F atoms on the "film-etched surface" interface compared with that coming from bulk plasma [4,5]. In such situation, the etching kinetics should be described in the terms of the film-thickness-dependent reaction probability which is determined by the balance between polymer deposition and destruction rates [5].
s
Q
0.00
0.05
0.10
0.15
0.20
0.25
Fig. 2. Model-predicted densities of neutral species as functions
of C4F8 fraction in CF4+C4F8+Ar gas mixture Рис. 2. Расчетные концентрации нейтральных частиц как функции доли C4F8 в плазмообразующей смеси CF4+C4Fs+Ar
From Refs. [3-5, 16], it can be understood that 1 the formation of the fluorocarbon polymer films on the etched surface is mainly provided by the radicals with more than one free bonds; and 2) the polymer film growths better in the fluorine-poor plasmas where the polymer surface is composed by the less saturated flu-orocarbon groups and thus, appears to be more reactive for radicals coming from bulk plasma. Accordingly, one can expect that the change in the polymer deposition rate may be adequately characterized by the rpoi/TF ratio, where Vpol is the total flux of CF2 and CF radicals. From Table, it can be seen that the substitution of CF4 for C4F8 lifts up both Гро1 (3.110166.3-1017 cm-2s-1 for 0-25% C4F8) and rpol/VF (0.1-11.3 for 0-25% C4F8). Therefore, an evident increase in the polymer deposition rate takes place. The destruction of the fluorocarbon polymer film in the non-oxygenized plasmas is mainly provided by the physical etching pathway due to the ion bombardment [4,5]. This process has the rate of 78Г+ where Ys is the ion-type-averages sputtering yield. Since Ys is determined by the
C4F8 traction in CF4+C4F8+Ar
momentum transferred from the incident ion to the surface atom in a single collision [5], one can assume Y ~ 9M(<î , where < ~ e|—U@ — Udc| is the ion bombardment energy, M; is the effective ion molar mass, U@ ^ 0.5reln(me/2.3mj) is the floating potential
and -Udc is the negative bias provided by Wdc. The data of Fig. 1 show that -Udc = 209-220 V for 0-25% C4F8 that results in < » 232-245 eV. The weak changes in < together with mi » const allow one to apply the linear correlation for Ys = f(£;) and use the
parameter < ;r+, the so-called ion energy flux, to characterize the rate of the physical etching pathway for the fluorocarbon polymer film. From Tab. 1, it can be seen that the opposite changes of £ ; and r+ maintain the ion energy flux on the near-to-constant level up to 22% C4F8 in a feed gas while the overall decrease in r+ for 0-25% C4F8 does not exceed 10%. Accordingly, the parameter rpoi/£;r+rF characterizing the amount of deposited polymer exhibits the sufficient increase toward C4F8-rich plasmas and occupies that range of 6.010-20-7.610-18 eV-1cm2s for 0-25% C4F8.
Table
Model-predicted fluxes and flux-to-flux ratios as functions of C4F8 fraction in CF4+C4F8+Ar gas mixture Таблица. Расчетные потоки частиц и их отношения как функции доли C4F8 в плазмообразующей смеси
CF4+C4F8+Ar
Fractions in a feed gas rF, 1017 cm-2s-1 Гро;, 1017 cm- 2s-1 <г+, 1018 eVcm-2s-1 г lpol "ГТ rpoi 10-i8 eV-
CF4 C4F8 Ar £;Г+г> 1cm2s1
0.67 0.00 0.33 3.49 0.31 1.61 0.09 0.06
0.58 0.08 0.33 0.69 2.81 1.60 4.27 2.66
0.53 0.13 0.33 0.62 3.41 1.59 5.61 3.49
0.50 0.17 0.33 0.59 4.36 1.57 7.36 4.60
0.45 0.22 0.33 0.58 5.70 1.55 9.74 6.10
0.42 0.25 0.33 0.56 6.28 1.49 11.3 7.59
The results described above allow one to formulate some conclusions concerning the features of the etching kinetics in the given gas system. First, the substitution of CF4 for C4F8 in the CF4+C4F8+Ar gas mixture suppresses the chemical etching pathway not only due to the decrease in F atom flux, but also through the change in effective reaction probability because of the increasing thickness of the fluorocarbon polymer film on the etched surface. Therefore, one can expect the decreasing absolute etching rates. And secondly, the addition of C4F8 probably provides the film-thickness-limited etching regime and thus, creates the favorable conditions for obtaining higher SiO2/Si etching selectivity. Accordingly, the reasonable balance between etching rate and selectivity may be effectively adjusted by the CF4/C4F8 mixing ratio.
REFERENCES
1. Rossnagel S.M., Cuomo J.J., Westwood W.D. Handbook of plasma processing technology. Park Ridge: Noyes Publications 1990. 338 p.
2. Rooth J.R. Industrial Plasma Engineering. Philadelphia: IOP Publishing LTD. 1995. 620 p.
3. Roosmalen A.J., Baggerman J.A.G., Brader S.J.H. Dry etching for VLSI. New-York: Plenum Press. 1991. 490 p.
4. Wolf S., Tauber R.N. Silicon Processing for the VLSI Era. V. 1. Process Technology. New York: Lattice Press. 2000. 416 p.
5. Lieberman M.A., Lichtenberg A.J. Principles of plasma discharges and materials processing. New York: John Wiley & Sons Inc. 1994. 757 p.
6. Chun I., Efremov A., Yeom G.Y., Kwon K.-H. A comparative study of CF4/O2/Ar and C4F8/O2/A plasmas for dry etching applications. Thin Solid Films. 2015. V. 579. P. 136-143.
7. Kwon K.-H., Efremov A., Kim M., Min N.K., Jeong J.,
Kim K. A model-based analysis of plasma parameters and composition in HBr/X (X=Ar, He, N2) inductively coupled plasmas. J. Electrochem. Soc. 2010. V. 157. P. H574-H579.
8. Johnson E.O., Malter L. A floating double probe method for measurements in gas discharges. Phys. Rev. 1950. V. 80. P. 58-70.
9. Sugavara M. Plasma etching: Fundamentals and applications. New York: Oxford University Press. 1998. 469 p.
10. Lee C., Lieberman M.A. Global model of Ar, O2, Cl2, and Ar/O2 high-density plasma discharges. J. Vac. Sci. Technol. A. 1995. V. 13. P. 368-389.
11. Efremov A.M., Kim D.-P., Kim C.-I. Effect of gas mixing ratio on gas-phase composition and etch rate in an inductively coupled CF4/Ar plasma. Vacuum. 2004. V. 75. P. 133-142.
12. Kimura T., Ohe K. Probe measurements and global model of inductively coupled Ar/CF4 discharges. Plasma Sources Sci. Technol. 1999. V. 8. P. 553-560.
13. Rauf S., Ventzek P.L.G. Model for aninductively coupled Ar-C4F8 plasma discharge. J. Vac. Sci. Technol. A. 2002. V. 20. P. 14-23.
14. Kokkoris G., Goodyear A., Cooke M., Gogolides E. A global model for C4F8 plasmas coupling gas phase and wall surface reaction kinetics. J. Phys. D: Appl. Phys. 2008. V. 41. P. 195211.
15. 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.
16. Matsui M., Tatsumi T., Sekine M. Relationship of etch reaction and reactive species flux in C4F8-Ar-O2 plasma for SiO2 selective etching over Si and Si3N4. J. Vac. Sci. Technol. A. 2001. V. 19. P. 2089-2096.
Поступила в редакцию (Received) 23.01.2018 Принята к опубликованию (Accepted) 19.03.2018
