Novel Cyclosilazane-Type Silicon Precursor and Two-Step Plasma for

Subscriber access provided by UNIV OF SCIENCES PHILADELPHIA

Article

Novel cyclosilazane-type silicon precursor and two-step plasma for plasma-enhanced atomic layer deposition of silicon nitride Jae-Min Park, Se Jin Jang, Sang-Ick Lee, and Won-Jun Lee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b19741 • Publication Date (Web): 20 Feb 2018 Downloaded from http://pubs.acs.org on February 23, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Novel cyclosilazane-type silicon precursor and two-step plasma for plasma-enhanced atomic layer deposition of silicon nitride Jae-Min Park, Se Jin Jang†, Sang-Ick Lee†,a), and Won-Jun Lee a) Department of Nanotechnology and Advanced Materials Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul, 05006, Republic of Korea †

DNF Co.Ltd., 142 Daehwa-ro 132 beon-gil, Daedeok-gu, Daejeon, 34366, Republic of Korea

KEYWORDS silicon nitride, plasma-enhanced atomic layer deposition (PEALD), cyclosilazane, three-step PEALD, step coverage, wet etching rate

ABSTRACT: We designed cyclosilazane-type silicon precursors and proposed a three-step plasma-enhanced atomic layer deposition (PEALD) process to prepare silicon nitride films with high quality and excellent step coverage. The cyclosilazane-type precursor, 1,3-di-isopropylamino-2,4-dimethylcyclosilazane (CSN-2) has a closed ring structure for good thermal stability and high reactivity. CSN-2 showed thermal stability up to 450°C and a sufficient vapor pressure of 4 Torr at 60°C. The energy for the chemisorption of CSN-2 on the undercoordinated silicon nitride surface was calculated to be -7.38 eV by density functional theory method. The PEALD process window was between 200 and 500°C with a growth rate of 0.43 Å/cycle. The best film quality was obtained at 500°C with hydrogen impurity of ~7 at.%, oxygen impurity less than 2 at.% and low wet etching rate and excellent step coverage of ~95%. At 300°C and lower temperatures, the wet etching rate was high especially at the lower sidewall of the trench pattern. We introduced the three-step PEALD process to improve the film quality and the step coverage on the lower sidewall. The sequence of the three-step PEALD process consists of the CSN-2 feeding step, the NH3/N2 plasma step, and the N2 plasma step. The H radicals in NH3/N2 plasma efficiently remove the ligands from the precursor, and the N2 plasma after the NH3 plasma removes the surface hydrogen atoms to activate the adsorption of the precursor. The films deposited at 300°C using the novel precursor and the three-step PEALD process showed a significantly improved step coverage of ~95% and excellent wet etching resistance at the lower sidewall, which is only twice as high as the blanket film prepared by low-pressure chemical vapor deposition (LPCVD).

INTRODUCTION The complexity of semiconductor chip continues to increase, the minimum feature size decreases, and threedimensional device structures have been adopted. The state-of-the-art technology of semiconductor fabrication has required thin films deposition techniques with a more accurate control of film thickness at the low-temperature process. Especially, Silicon nitride thin films for gate spacer of FinFET devices1 or the charge trap layer of threedimensional vertical NAND flash memory2,3 require excellent film quality and superior step coverage. Conventional chemical vapor deposition (CVD) has limitations to fulfill these requirements, and atomic layer deposition (ALD) technique is the most promising solution. ALD is an atomic layer-by-layer growth process based on selflimiting surface reactions and has been studied to produce uniform and conformal thin films on high-aspectratio patterns at low temperature with high quality. There have been reports on the ALD of silicon nitride for last two decades, and two review papers have been recently published.4,5 Various silicon chloride precursors,

such as SiCl46,7,8, SiH2Cl27, 9 and Si2Cl610,11 were used in the thermal ALD of silicon nitride thin films, while NH3 was used as the nitriding agent except for the report11 which used hydrazine. Relatively high process temperature and the corrosive reaction byproduct, hydrochloric acid limit the application of thermal ALD for gate spacer applications.12 Plasma-enhanced ALD (PEALD) process can lower the deposition temperature by lowering the activation energies for the surface reaction13. The PEALD of silicon nitride have been reported using SiH414, SiH2Cl215, Si2Cl616, trisilylamine (TSA)17,18, bis(tertiary-butyl-amino)silane (BTBAS)19,20, bis(dimethylaminomethylsilyl)trimethylsilylamine (DTDN2-H2)21, and di(sec-butylamino)silane (DSBAS)22 as the silicon precursor. Nitriding agent was NH3 plasma or N2 plasma. NH3 plasma was used for the PEALD using SiH4, TSA, and silicon chlorides, however, the PEALD using aminosilane precursors and NH3 plasma showed significantly low growth rates because the chemisorption of silicon precursors on the NH2-terminated surface is difficult.20 N2 plasma produces under-coordinated silicon

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nitride surface by removing H atoms from the surface,14,23 and the PEALD using aminosilane precursors and N2 plasma showed reasonable growth rates.20 Carbon incorporation in the deposited film was an issue because the reaction byproducts were redeposited by N2 plasma. The carbon content could be reduced by reducing the residence time of the reaction byproducts.19,24 In the previous work, we reported a PEALD process using a newly designed silicon precursor, DTDN2-H2.21 High-quality silicon nitride films were deposited with a growth rate of 0.036 nm/cycle in ALD temperature window between 250 and 400°C, and the RF power and the substrate temperature were optimized to deposit high purity silicon nitride films with no carbon impurity.21 Still, step coverage needs improvement for next-generation semiconductor devices. In this work, we designed cyclosilazane-type silicon precursors and proposed a three-step PEALD process to improve step coverage. The cyclosilazane-type precursors have a closed ring structure for good thermal stability and high reactivity. Among them, 1,3-di-isopropylamino-2,4dimethylcyclosilazane (CSN-2) was selected as the silicon precursor of the present work due to its high vapor pressure. The growth kinetics and the physical properties of the deposited films, such as refractive index, composition, and wet etching rate, were characterized at different deposition temperatures. The three-step PEALD process composed of precursor pulse, NH3/N2 plasma and then N2 plasma was also developed to improve step coverage as well as film quality. The H radicals in NH3/N2 plasma efficiently remove the ligands from the precursor25 and are much more efficient than the N radical of the N2 plasma. The N2 plasma following the NH3 plasma removes the surface hydrogen atoms to activate the adsorption of the precursor.23 The film thickness and the wet etching rate of the deposited film were examined at different positions on the sidewall of a high-aspect-ratio trench-pattern.

EXPERIMENTAL DFT Calculations. We performed the first principle density functional theory (DFT) calculations with Dmol3 package26,27 of Material Studio 7.0 (Accelrys, USA). We used the generalized gradient approximation (GGA) with the Perdew-Burke-Ernzerhof (PBE)28 as the exchangecorrelation functional and the double numerical polarization (DNP) as the basis set. We obtained the structure with the lowest system energy by the geometry optimization.29,30 The case with the lowest adsorption energy was selected as the most stable configuration for the surface reaction. The details of the procedure to calculate the bond dissociation energy (BDE) values, the energy of reaction, and the energy barrier are described in elsewhere.21,23,31 To study the 1st half reaction of the PEALD process, that is the chemisorption of CSN-2, we constructed under-coordinated bare β-Si3N4 surface (>Si=N-) to simulate the N2 plasma treated silicon nitride surface.23 We expected the path of the first half reaction by calculating the energy and the structure of unbound, physisorbed, chemisorbed, and transition states of the precursor on the under-coordinated bare β-Si3N4 surface.

Page 2 of 34

Modeling amorphous silicon nitride with dangling bonds and hydrogen impurities allow a more realistic simulation of the ALD process at 300°C. However, modeling amorphous materials by DFT method needs extremely high computing power because the large unit cell with random structures is required to mimic amorphous materials. Therefore, we modeled and simulated based on the βSi3N4 structure. PEALD of Silicon Nitride. Silicon nitride thin films were deposited using a showerhead-type cold-wall reactor (Atomic–Premium, CN1 Co., Ltd., Korea). The SiNx ALD experiments were carried out in a commercial plasma reactor manufactured by CN1 Co., Ltd. The silicon precursor, CSN-2, was heated to 60°C, and the vapor was delivered into the chamber with an N2 carrier gas of 50 sccm. Direct plasma was generated by an RF generator (27.12 MHz) under 2.7 Torr. We injected 6000 sccm of N2 for the N2 plasma and co-injected 50 sccm of NH3 with 6000 sccm of N2 for the NH3/N2 plasma. Characterization of the Deposited Films. The thickness and the refractive index of the deposited films were measured by ellipsometry (M2000D (RCT), J.A. Woollam, USA). The film thickness and the step coverage were examined by cross-sectional transmission electron microscopy (TEM) (JEM-2100F HR, JEOL, Japan). The composition of the silicon nitride films was analyzed by Auger electron spectroscopy (AES) (PHI-700, ULVAC-PHI, Japan). The [N]/[Si] ratio of the PEALD film was calibrated using bulk β-silicon nitride. Hydrogen concentration was obtained by secondary ion mass spectrometry (IMS 7f, Cameca, France) with a Cs+ primary ion beam, rastered over an area of 250 × 250 um2. The wet etching rate of the films was evaluated in a diluted HF solution (H2O/HF = 300:1).

RESULTS AND DISCUSSION The chemical structure of CSN-2, 1,3-diisopropylamino-2,4-dimethylcyclosilazane, is shown in Fig. 1 (a).

Figure 1. (a) The chemical structure, (b) vapor pressure, (c) TGA curve (10°C/10 min), and (d) DSC curve (10°C/10 min) of CSN-2.


Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The CSN-2 was synthesized through several steps such as amino-substitution reactions and redistribution reactions. The detailed procedure for the synthesis of CSN-2 is described in the US patant32. CSN-2 shows a sufficient vapor pressure of 4 Torr at 60°C as shown in the vapor pressure versus temperature plot of Fig. 1 (b). Figure 1 (c) shows the dynamic thermogravimetric analysis (TGA) at a ramp rate of 10°C/10 min. The temperature at which half of CSN-2 evaporates is 180°C, which is lower than 220°C of DTDN2-H2 in the previous work21. Therefore, it is easier to deliver CSN-2 to the reactor than with DTDN2-H2 at the same canister temperature. The differential scanning calorimetric (DSC) analysis in Fig. 1 (d) shows that the precursor is thermally stable up to 450°C. The small bump near 300°C is due to boiling of the precursor at atmospheric pressure.

We modeled and simulated the reaction of CSN-2 with the under-coordinated silicon nitride surface. The total energy of the system was calculated along the possible reaction pathways to find the reaction product with the lowest energy, as shown in figure 3. The unbound state was modeled assuming the injection of CSN-2 into undercoordinated silicon nitride surface. The transition state (TS) between the physisorption and the chemisorption shows the dissociation of a hydrogen from the precursor with an energy barrier of 1.32 eV. The final structure shows the chemisorption of CSN-2 on the silicon nitride surface, which is an exothermic reaction with total reaction energy of – 7.38 eV. The energy diagram in figure 3 shows that the reaction of CSN-2 with the surface is more exothermic than the reaction of DTDN2-H2 in the previous work21, and the activation energy of CSN-2 is lower than that of DTDN2-H2. Therefore, CSN-2 would show faster chemisorption on the silicon nitride surface.

Figure 2. (a) Geometry-optimized structure and (b) electrophilicity of a CSN-2 molecule. (c) Geometry optimized structure and (d) nucleophilicity of the under-coordinated bare silicon nitride surface.

Figure 2 shows the geometry-optimized structure and the electrophilicity of a CSN-2 molecule and the silicon nitride surfaces, respectively. The cyclosilazane ring of CSN-2 has a planar structure, and the silicon and nitrogen atoms in CSN-2 are more exposed to active sites on the surface than other precursors, allowing the direct interaction of the silicon and nitrogen atoms of the precursor with the silicon nitride surface. In contrast, the tetrahedral geometry of the DTDN2-H2 precursor of the previous work suggests that a direct interaction of the silicon atom with the surface can be sterically hindered. The most electrophilic atom is susceptible to nucleophilic attack, and the most nucleophilic atom is susceptible to electrophilic attack33. The charge analysis using Fukui function shows that the silicon atom of CSN-2 with the highest electrophilicity (blue) would react with the nitrogen atom of the substrate with the highest nucleophilicity (red). Also, the nitrogen atoms of CSN-2 with the lowest electrophilicity (red) would react with the silicon atoms of the substrate with the lowest nucleophilicity (blue).

Figure 3. Reaction of CSN-2 on the under-coordinated bare βSi3N4 surfaces during the first half cycle of PEALD.

The deposition rate of PEALD using CSN-2 and N2 plasma was investigated at 300°C. Figure 4 shows the growth rate of the deposited film as a function of the CSN-2 feeding time with a fixed N2 plasma time of 60 s and an RF power of 75 W. The growth rate per cycle increased with increasing CSN-2 feeding time and saturated at 0.43 Å/cycle when the feed time exceeded 2.5 s, as shown in Fig. 4 (a). The saturation behaviors indicate



that the surface reaction between CSN-2 and the substrate surface is self-limiting and there is no thermal decomposition of the precursor34. The growth rate also saturated when the N2 plasma time was 30 s or more, as shown in Fig. 4 (b). Films produced by the deposition conditions with fully-saturated surface reactions showed a refractive index of about 1.98, which is similar to 2.01 of the stoichiometric silicon nitride. The refractive index values of the thin films produced with short N2 plasma times were lower, which means that the quality of the deposited films is reduced.

Figure 4. The growth rate of the PEALD SiN films at 300°C as a function of (a) CSN-2 feeding time or (b) N2 plasma time. N2 plasma time was set at 60 s in (a), and CSN-2 feeding time was set at 2.5 s in (b), with a fixed RF power of 75 W.

We investigated the growth rate and quality of PEALD films at temperatures between 250°C and 500°C, as shown in Fig. 5. The thin films were deposited by repeating 300 cycles of supplying CSN-2 for 5 s and N2 plasma for 60 s. The RF power was fixed at 75 W. As shown in Fig. 5 (a), the growth rate per cycle is almost constant at 0.43 Å/cycle between 250°C and 500°C, which is regarded as the ALD temperature window. The growth rate per cycle and the ALD temperature window obtained in the present work are superior to those using DTDN2-H2 or BTBAS in the previous literature19,21. The composition of the PEALD thin film was analyzed by Auger electron spectroscopy after removing the surface layer by Ar ion beam etching as shown in Fig. 5 (b). The deposited films exhibited a slightly higher [N] / [Si] ratio than 1, which is nearly identical to that of the LPCVD films fabricated at 770°C. Carbon impurities were not detected, and the concentration of oxygen impurities was reduced with increasing deposition temperature and was measured to be less than 2% at over 300°C. It is a typical phenomenon for all silicon nitride films to be oxidized in air, and the oxygen content is a good measure for estimating the film quality. The oxygen content of less than 4% of this work is excellent, as compared with ~5% of the ALD film prepared by BTBAS19 and minimum 6% of the film prepared by DTDN2-H221. The hydrogen concentration was 6.5~7.0 at.% by SIMS, which is superior to the values in the PEALD literature35.

Page 4 of 34

Figure 5. Effect of deposition temperature. PEALD silicon nitride films were prepared with an RF power of 75W at various deposition temperatures: (a) growth rate per cycle, (b) the concentration ratio of nitrogen to silicon ([N]/[Si]) and the concentration of impurities.

The step coverage of silicon nitride film was investigated at temperatures between 250 and 500°C, as shown in Fig. 6. The aspect ratio of the trench-patterned wafer is 5.5, and the entrance width of the trench is 75 nm. The step coverage of PEALD films improved with increasing deposition temperature. The bottom coverage increased from 91% at 250°C to 95% at 500°C, and the sidewall coverage increased from 87% at 250°C to 98% at 500°C at the center of the trench. These values are superior to those of PEALD film produced using BDEAS. (Bottom coverage, 80%; sidewall coverage, 57%; not shown here). The area with the most significant improvement in conformality was the lower sidewall, and the coverage of the lower sidewall increased from 81% to 93% as the deposition temperature increased.

Figure 6. (a) Step coverage of the PEALD silicon nitride film as a function of process temperature and the cross-section transmission TEM images of the film prepared at (b) 250°C, (c) 350°C, and (d) 500°C. (CSN-2 pulse, 5 s; N2 plasma, 60 s; rf power, 75 W; cycle number, 240)


Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. (a) Wet etching rate of the PEALD silicon nitride film as a function of process temperature and the crosssectional transmission TEM images of the film prepared at (b) 250°C, (c) 350°C, and (d) 500°C. (CSN-2 pulse, 5 s; N2 plasma, 60 s; rf power, 75 W; cycle number, 240)

To compare the integrity of the deposited film on the trench-patterned wafer, we evaluated the wet etching rate of the film in a diluted HF solution at different locations over the trench-pattern. The wet etching rate of a blanket LPCVD film deposited at 770°C was used as a reference. As shown in Fig. 7, the wet etching rate decreased significantly with increasing deposition temperature. The PEALD films deposited on the upper horizontal surface exhibited low wet etching rates and were much better than the blanket LPCVD films. The wet etching rate of the thin film deposited on the bottom horizontal surface was 1.8 Å/min at 250°C and decreased to 0.4 Å / min at 500°C, which is better than LPCVD film (1.5 Å/min). However, the wet etching rate at the lower sidewall was significantly higher than that of the LPCVD film. Especially, it was 24.9 Å/min at 250°C. When the process temperature increased to 500°C, it decreased to 2.9 Å/min but was still higher than that of LPCVD thin film.

Figure 8. Growth rate and refractive index of three-step PEALD SiN film with different times of (a) the first NH3/N2 plasma time and (b) the second N2 plasma step time.

The poor deposition rate and wet etching rates at the lower sidewalls are due to incomplete removal of the ligands of the precursor by the N2 plasma. In general, PEALD is challenging to deposit high-quality conformal films on the lower sidewalls due to the recombination loss of radicals. In particular, the quality of the film on the lower sidewall is more problematic in PEALD of silicon nitride using N2 plasma compared to ALD of silicon oxide using O2 plasma, because N radicals have a shorter lifetime than O radicals.36 Since the amount of N radical reaching the trench lower sidewall is smaller compared with the upper horizontal surface, the reaction of N2 plasma with the precursor adsorbed on the substrate surface was not fully saturated, resulting in low deposition rates and high wet etching rates at the bottom sidewall. Poor film quality at bottom sidewall can be deduced from the low refractive index values when the N2 plasma time is short as shown in Fig. 2 (b). Higher carbon concentration with short N2 plasma time was reported for the PEALD of silicon nitride using BTBAS and N2 plasma,35 and higher wet etching rate at bottom sidewall was also reported for the PEALD of silicon nitride using di(sec-butylamino)silane (DSBAS) and N2 plasma.22 As the deposition temperature increased, the reaction between the N radical and the ligand was activated, efficiently removing the ligand, thereby improving the growth rate and wet etching rate. However, even at 500°C, more N radicals are needed to lower the wet etching rate at the bottom sidewall. The role of the N2 plasma stage in PEALD not only supplies N radicals but also removes ligands from the surface, facilitating the chemisorption of the silicon precursor in the next precursor supply step. H radical in NH3 plasma is very efficient in removing ligands from the surface25 and also has a longer lifetime than N radical,25 but it passivates the surface with hydrogen atoms, inhibiting the chemisorption of the precursor20. According to the previous work, N2 plasma can remove hydrogen atoms of silicon nitride surface to form under-coordinate bare silicon nitride surface20,23. The growth rate of ALD silicon nitride using Si2Cl6 and NH3 was improved by adding an N2 plasma step after NH3 pulse because the N2 plasma step enhanced the adsorption of Si2Cl6 on the surface. Therefore, we introduced the three-step PEALD process to increase the film quality and the step coverage on the lower sidewall. The sequence of the three-step PEALD process consists of the precursor feeding step, the NH3/N2 plasma step, and the N2 plasma step. Better material properties and better step coverage were obtained at temperatures at 500°C. However, we studied the ALD at 300°C using different plasma conditions to improve the properties and step coverage, because the gate spacer of logic transistors requires low-temperature processing. We investigated the growth rate of the thin film of the three-step PEALD process by fixing the CSN-2 feed time to 10 s and varying the NH3/N2 plasma time or the N2 plasma time. Figure 8 (a) shows that growth decreased slightly by adding the NH3/N2 plasma step to the two-step ALD process. The decrease in growth rate is because the –



NH2 or –SiH surface groups, which are difficult for the precursor to adsorb on, were formed by the NH3/N2 plasma. For the same reason, the growth rate was significantly reduced when only NH3/N2 plasma was used without N2 plasma in Fig. 8 (b). The N2 plasma increased the growth rate as the N2 plasma time increases because the H atoms can be removed from –NH2 and –SiH to form an undercoordinated silicon nitride surface. Treatment with 75W N2 plasma for more than 30 s will restore the growth rate to 80% of that of the two-step PEALD using N2 plasma.

Page 6 of 34

plasma step. The films deposited at 300°C using the three-step PEALD process showed a significantly improved step coverage of ~95% and excellent wet etching resistance at the lower sidewall, which is only twice as high as the blanket LPCVD film.

AUTHOR INFORMATION Corresponding Author a) C0-corresponding authors (Won-Jun Lee) E-mail: [email protected] (Sang-Ick Lee) E-mail: [email protected]

ACKNOWLEDGMENTS This work was supported by Industrial Strategic Technology Development Program, 10041792, Key technology development of low-temperature PEALD equipment using inorganic precursor for 1x/2x semiconductor device, funded by the Ministry of Trade, Industry & Energy (MOTIE) of Korean Government.

ABBREVIATIONS USED Figure 9. The cross-sectional transmission TEM images at 300°C, (a) before and after wet etching of CSN-2 and N2 plasma process, and (b) CSN-2 and three-step PEALD process. (CSN-2 pulse, 5 s; NH3/N2 plasma, 30 s; N2 plasma, 30 s; rf power, 75 W; cycle number, 240)

We investigated the step coverage and the integrity of the film prepared by the three-step PEALD process, as shown in Fig. 9. The step coverage and the wet etching rate at the lower sidewall were improved by introducing the three-step PEALD process. When the NH3/N2 plasma time and N2 plasma time were 30 s, the lower sidewall coverage at 300°C was 95%, which is significantly improved from 81% when only N2 plasma was used for 60 s. Moreover, the wet etching rate of lower sidewall was significantly reduced from 13 Å/min to 3 Å/min, which is only twice as high as the blanket LPCVD film.

CONCLUSION CSN-2, a new cyclosilazane-type silicon compound, was designed and synthesized for PEALD of silicon nitride films with high quality and excellent step coverage. CSN-2 showed thermal stability up to 450°C and a sufficient vapor pressure of 4 Torr at 60°C. The energy for the chemisorption of CSN-2 on the undercoordinated silicon nitride surface was calculated to be – 7.38 eV by DFT. The PEALD process window was between 200 and 500°C with a growth rate of 0.43 Å/cycle. The best film quality was obtained at 500°C with hydrogen impurity of ~7 at.%, oxygen impurity less than 2 at.% and low wet etching rate and excellent step coverage of ~95%. At 300°C and lower temperatures, the wet etching rate was high especially at the lower sidewall of the trench pattern. We introduced the three-step PEALD process to increase the film quality and the step coverage on the lower sidewall. The sequence of the three-step PEALD process consists of the CSN-2 feeding step, the NH3/N2 plasma step, and the N2

CSN-2, 1,3-di-isopropylamino-2,4-dimethylcyclosilazane; DFT, density functional theory; BDE, bond dissociation energy.

REFERENCES (1) Kaneko, A.; Yagishita, A.; Yahashi, K.; Kubota, T.; Omura, M.; Matsuo, K.; Mizushima, I.; Okano, K.; Kawasaki, H.; Inaba, S.; Izumida, T.; Kanemura, T.; Aoki, N.; Ishimaru, K.; Ishiuchi, H.; Suguro, K.; Eguchi, K.; Tsunashima, Y. Sidewall Transfer Process and Selective Gate Sidewall Spacer Formation Technology for Sub-15nm Finfet with Elevated Source/drain Extension. In IEEE Int. Devices Meet. 2005. IEDM Technical Dig.; IEEE, 2005; 844–847. (2) Tanaka, H.; Kido, M.; Yahashi, K.; Oomura, M.; Katsumata, R.; Kito, M.; Fukuzumi, Y.; Sato, M.; Nagata, Y.; Matsuoka, Y.; Iwata, Y.; Aochi, H.; Nitayama, A. Bit Cost Scalable Technology with Punch and Plug Process for Ultra High Density Flash Memory. In IEEE Symposium on VLSI Technology; IEEE, 2007; 14–15. (3) Jang, J.; Kim, H.-S.; Cho, W.; Cho, H.; Kim, J.; Shim, S. Il; Jang, Y.; Jeong, J.-H.; Son, B.-K.; Kim, D. W.; Kim, K.; Shim, J.J.; Lim, J. S.; Kim, K.-H.; Yi, S. Y.; Lim, J.-Y.; Chung, D.; Moon, H.C.; Hwang, S.; Lee, J.-W.; Son, Y.-H.; Chung, U.-I.; Lee, W.-S. Vertical Cell Array Using TCAT(Terabit Cell Array Transistor) Technology for Ultra High Density NAND Flash Memory. IEEE Symp. VLSI Technol. 2009, 192–193. (4) Meng, X.; Byun, Y.-C.; Kim, H.; Lee, J.; Lucero, A.; Cheng, L.; Kim, J. Atomic Layer Deposition of Silicon Nitride Thin Films: A Review of Recent Progress, Challenges, and Outlooks. Materials (Basel). 2016, 9 (12), 1007. (5) Kaloyeros, A. E.; Jové, F. A.; Goff, J.; Arkles, B. Review—Silicon Nitride and Silicon Nitride-Rich Thin Film Technologies: Trends in Deposition Techniques and Related Applications. ECS J. Solid State Sci. Technol. 2017, 6 (10), P691– P714. (6) Klaus, J. W.; Ott, A. W.; Dillon, A. C.; George, S. M. Atomic Layer Controlled Growth of Si3N4 Films Using Sequential Surface Reactions. Surf. Sci. 1998, 418 (1), L14–L19. (7) Lee, W.; Lee, J.; Park, C. O.; Lee, Y. A Comparative Study on the Si Precursors for the Atomic Layer Deposition of


Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Silicon Nitride Thin Films. J. Korean Phys. Soc. 2004, 45 (5), 1352–1355. (8) Nakajima, A.; Khosru, Q. D. M.; Yoshimoto, T.; Kidera, T.; Yokoyama, S. NH3-Annealed Atomic-Layer-Deposited Silicon Nitride as a High-K Gate Dielectric with High Reliability. Appl. Phys. Lett. 2002, 80 (7), 1252–1254. (9) Lee, W. J.; Kim, U. J.; Han, C. H.; Chun, M. H.; Rha, S. K.; Lee, Y. S. Characteristics of Silicon Nitride Thin Films Prepared by Using Alternating Exposures of SiH2Cl2 and NH3. J. Korean Phys. Soc. 2005, 47, S598–S602. (10) Park, K.; Yun, W. D.; Choi, B. J.; Kim, H. Do; Lee, W. J.; Rha, S. K.; Park, C. O. Growth Studies and Characterization of Silicon Nitride Thin Films Deposited by Alternating Exposures to Si2Cl6 and NH3. Thin Solid Films 2009, 517, 3975–3978. (11) Morishita, S.; Sugahara, S.; Matsumura, M. AtomicLayer Chemical-Vapor-Deposition of Silicon-Nitride. Appl. Surf. Sci. 1997, 112, 198–204. (12) Koehler, F.; Triyoso, D. H.; Hussain, I.; Antonioli, B.; Hempel, K. Challenges in Spacer Process Development for Leading-Edge High- K Metal Gate Technology. Phys. status solidi 2014, 11 (1), 73–76. (13) Profijt, H. B.; Potts, S. E.; van de Sanden, M. C. M.; Kessels, W. M. M. Plasma-Assisted Atomic Layer Deposition: Basics, Opportunities, and Challenges. J. Vac. Sci. Technol. A Vacuum, Surfaces, Film. 2011, 29 (5), 50801. (14) King, S. W. Plasma Enhanced Atomic Layer Deposition of SiNx:H and SiO2. J. Vac. Sci. Technol. A 2011, 29 (4), 41501. (15) Goto, H.; Shibahara, K.; Yokoyama, S. Atomic Layer Controlled Deposition of Silicon Nitride with Self-limiting Mechanism. Appl. Phys. Lett. 1996, 68 (23), 3257–3259. (16) Ovanesyan, R. a; Hausmann, D. M.; Agarwal, S. LowTemperature Conformal Atomic Layer Deposition of SiN X Films Using Si 2 Cl 6 and NH 3 Plasma. ACS Appl. Mater. Interfaces 2015, 7 (20), 10806–10813. (17) Triyoso, D. H.; Hempel, K.; Ohsiek, S.; Jaschke, V.; Shu, J.; Mutas, S.; Dittmar, K.; Schaeffer, J.; Utess, D.; Lenski, M. Evaluation of Low Temperature Silicon Nitride Spacer for HighK Metal Gate Integration. ECS J. Solid State Sci. Technol. 2013, 2 (11), N222–N227. (18) Jang, W.; Jeon, H.; Kang, C.; Song, H.; Park, J.; Kim, H.; Seo, H.; Leskela, M.; Jeon, H. Temperature Dependence of Silicon Nitride Deposited by Remote Plasma Atomic Layer Deposition. Phys. Status Solidi 2014, 211 (9), 2166–2171. (19) Knoops, H. C. M.; Braeken, E. M. J.; de Peuter, K.; Potts, S. E.; Haukka, S.; Pore, V.; Kessels, W. M. M. Atomic Layer Deposition of Silicon Nitride from Bis( Tert -Butylamino)silane and N 2 Plasma. ACS Appl. Mater. Interfaces 2015, 7 (35), 19857– 19862. (20) Ande, C. K.; Knoops, H. C. M.; de Peuter, K.; van Drunen, M.; Elliott, S. D.; Kessels, W. M. M. Role of Surface Termination in Atomic Layer Deposition of Silicon Nitride. J. Phys. Chem. C 2015, 6 (18), 3610–3614. (21) Park, J.-M.; Jang, S. J.; Yusup, L. L.; Lee, W.-J.; Lee, S.-I. Plasma-Enhanced Atomic Layer Deposition of Silicon Nitride Using a Novel Silylamine Precursor. ACS Appl. Mater. Interfaces 2016, 8 (32), 20865–20871. (22) Faraz, T.; van Drunen, M.; Knoops, H. C. M.; Mallikarjunan, A.; Buchanan, I.; Hausmann, D. M.; Henri, J.; Kessels, W. M. M. Atomic Layer Deposition of Wet-Etch Resistant Silicon Nitride Using Di( Sec -Butylamino)silane and N 2 Plasma on Planar and 3D Substrate Topographies. ACS Appl. Mater. Interfaces 2017, 9 (2), 1858–1869. (23) Yusup, L. L.; Park, J.-M.; Noh, Y.-H.; Kim, S.-J.; Lee, W.-J.; Park, S.; Kwon, Y.-K. Reactivity of Different Surface Sites with Silicon Chlorides during Atomic Layer Deposition of Silicon Nitride. RSC Adv. 2016, 6 (72), 68515–68524.

(24) Bosch, R. H. E. C.; Cornelissen, L. E.; Knoops, H. C. M.; Kessels, W. M. M. Atomic Layer Deposition of Silicon Nitride from Bis(tertiary-Butyl-Amino)silane and N2 Plasma Studied by in Situ Gas Phase and Surface Infrared Spectroscopy. Chem. Mater. 2016, 28 (16), 5864–5871. (25) Aydil, E. S. Real Time in Situ Monitoring of Surfaces during Glow Discharge Processing: NH3 and H2 Plasma Passivation of GaAs. J. Vac. Sci. Technol. B Microelectron. Nanom. Struct. 1995, 13 (2), 258. (26) Delley, B. An All-Electron Numerical Method for Solving the Local Density Functional for Polyatomic Molecules. J. Chem. Phys. 1990, 92 (1), 508–517. (27) Delley, B. From Molecules to Solids with the DMol3 Approach. J. Chem. Phys. 2000, 113 (18), 7756–7764. (28) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77 (18), 3865–3868. (29) Baker, J.; Kessi, A.; Delley, B. The Generation and Use of Delocalized Internal Coordinates in Geometry Optimization. J. Chem. Phys. 1996, 105 (1), 192–212. (30) Andzelm, J.; King-Smith, R. D.; Fitzgerald, G. Geometry Optimization of Solids Using Delocalized Internal Coordinates. Chem. Phys. Lett. 2001, 335 (3–4), 321–326. (31) Yusup, L. L.; Park, J.-M.; Mayangsari, T. R.; Kwon, Y.K.; Lee, W.-J. Surface Reaction of Silicon Chlorides during Atomic Layer Deposition of Silicon Nitride. Appl. Surf. Sci. 2017, 432, 127–131. (32) Jang, S. J.; Yang, B.; Kim, S. G.; Kim, J. H.; Kim, D. Y.; Lee, S.; Seok, J. H.; Lee, S. I.; Kim, M. W. Novel Cyclodisilazane Derivative, Method for Preparing the Same and SiliconContaining Thin Film Using the Same. US 20160326193 A1, 2016. (33) Huang, L.; Han, B. Density Functional Theory Study on the Full ALD Process of Silicon Nitride Thin Film Deposition via BDEAS or BTBAS and NH 3. Phys. Chem. … 2014, 16, 18501–18512. (34) George, S. M. Atomic Layer Deposition: An Overview. Chem. Rev. 2010, 110, 111–131. (35) Knoops, H. C. M.; de Peuter, K.; Kessels, W. M. M. Redeposition in Plasma-Assisted Atomic Layer Deposition: Silicon Nitride Film Quality Ruled by the Gas Residence Time. Appl. Phys. Lett. 2015, 107 (1), 14102. (36) Kovalgin, A. Y.; Yang, M.; Banerjee, S.; Apaydin, R. O.; Aarnink, A. A. I.; Kinge, S.; Wolters, R. A. M. Hot-Wire Assisted ALD: A Study Powered by In Situ Spectroscopic Ellipsometry. Adv. Mater. Interfaces 2017, 4 (18), 1700058.



Page 8 of 34

Table of Contents


8

Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig 1(a) 41x33mm (600 x 600 DPI)



fig 1(b) 49x49mm (600 x 600 DPI)


Page 10 of 34

Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig 1(c) 49x49mm (600 x 600 DPI)



fig 1(d) 49x49mm (600 x 600 DPI)


Page 12 of 34

Page 13 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig 2(a) 34x23mm (600 x 600 DPI)



fig 2(b) 24x12mm (600 x 600 DPI)


Page 14 of 34

Page 15 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig 2(c) 61x76mm (600 x 600 DPI)



fig 2(d) 59x58mm (600 x 600 DPI)


Page 16 of 34

Page 17 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig 3 90x136mm (600 x 600 DPI)



fig4(a) 49x49mm (600 x 600 DPI)


Page 18 of 34

Page 19 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig4(b) 49x49mm (600 x 600 DPI)



fig5(a) 49x49mm (600 x 600 DPI)


Page 20 of 34

Page 21 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig5(b) 44x40mm (600 x 600 DPI)



fig6(a) 49x49mm (600 x 600 DPI)


Page 22 of 34

Page 23 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig6(b) 75x141mm (600 x 600 DPI)



fig6(c) 75x143mm (600 x 600 DPI)


Page 24 of 34

Page 25 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig6(d) 88x194mm (600 x 600 DPI)



fig7(a) 49x49mm (600 x 600 DPI)


Page 26 of 34

Page 27 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig7(b) 73x133mm (600 x 600 DPI)



fig7(c) 73x133mm (600 x 600 DPI)


Page 28 of 34

Page 29 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig7(d) 76x146mm (600 x 600 DPI)



fig8(a) 54x58mm (600 x 600 DPI)


Page 30 of 34

Page 31 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig8(b) 54x58mm (600 x 600 DPI)



fig9(a) 56x63mm (600 x 600 DPI)


Page 32 of 34

Page 33 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fig9(b) 64x83mm (600 x 600 DPI)



Table of content 29x17mm (600 x 600 DPI)


Page 34 of 34

Novel Cyclosilazane-Type Silicon Precursor and Two-Step Plasma for

Recommend Documents