Next Article in Journal
Study on Adhesion Properties and Process Parameters of Electroless Deposited Ni-P Alloy for PEEK and Its Modified Materials
Next Article in Special Issue
Cutting Energy Consumption Modelling of End Milling Cutter Coated with AlTiCrN
Previous Article in Journal
Photocatalytic Oxidation of Amoxicillin in CPC Reactor over 3D Printed TiO2-CNT@PETG Static Mixers
Previous Article in Special Issue
Characterization of Structure, Morphology, Optical and Electrical Properties of AlN–Al–V Multilayer Thin Films Fabricated by Reactive DC Magnetron Sputtering
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Fluorination of TiN, TiO2, and SiO2 Surfaces by HF toward Selective Atomic Layer Etching (ALE)

Department of Chemical Engineering, Hongik University, Seoul 04066, Republic of Korea
*
Author to whom correspondence should be addressed.
Coatings 2023, 13(2), 387; https://doi.org/10.3390/coatings13020387
Submission received: 15 January 2023 / Revised: 5 February 2023 / Accepted: 7 February 2023 / Published: 8 February 2023
(This article belongs to the Special Issue Surface Modification of Engineering and Functional Materials)

Abstract

:
As semiconductor devices become miniaturized, the importance of the molecular-level understanding of the fabrication processes is growing. Titanium nitride (TiN) is an important material utilized in various architectural components of semiconductor devices requiring precise control over size and shape. A reported process for atomic layer etching (ALE) of TiN involves surface oxidation into titanium oxide (TiO2) and selective oxidized layer removal by hydrogen fluoride (HF). However, the chemical selectivity of these Ti-based materials in the etching process by HF remains unclear. In this study, computational chemistry methods utilizing density functional theory (DFT) calculations were applied to the fluorination reactions of TiN, TiO2, and SiO2 to identify and compare the surface chemical reactivity of these substrates toward etching processes. It is shown that the materials can be etched using HF, leaving TiF4 and SiF4 as the byproducts. However, while such a TiN reaction is thermodynamically hindered, the etching of TiO2 and SiO2 is suggested to be favorable. Our study provides theoretical insights into the fluorination reactivity of TiN, which has not been reported previously regardless of technological importance. Furthermore, we explore the etching selectivity between TiN, TiO2, and SiO2, which is a crucial factor in the ALE process conditions of TiN.

1. Introduction

Titanium nitride (TiN) is an important industrial material with various widespread applications because of its excellent mechanical stability and good electrical and thermal conductivity. In the semiconductor industry, it is used as a diffusion barrier between interconnect metals such as copper (Cu) with dielectric layers. In addition, it is employed as metal gate electrodes due to its chemical stability and good adhesion to the substrates [1,2,3,4,5]. TiN is also highly resistant to corrosion and can withstand high temperatures, making it suitable for use in the aerospace and automotive industries, such as in wear-resistant coating and other high-stress applications [6,7,8]. Despite the excellent chemical resistance of the bulk TiN material, its surface can be oxidized under exposure to an oxidative atmosphere [9].
Atomic layer etching (ALE) is a technique used in microfabrication and nanofabrication to precisely etch thin layers of material from a substrate [10,11,12,13,14]. One of the key advantages of ALE over other etching techniques is its ability to achieve high precision levels and control over the etch depth. The thickness of the material removed in each cycle of the ideal ALE process is a constant value, thus the etched material’s total thickness can be precisely controlled by adjusting the number of ALE cycles performed. In addition, the highly conformal nature of ALE makes it possible to etch inside features with high aspect ratios [15]. ALE involves sequential substrate exposure to a series of gases, reacting with the surface to remove a material layer [16]. Such a process mainly consists of two chemical reactions. First, a converted layer is formed by the reaction between the first etchant and the substrate. As the reaction between the second etchant and the converted layer proceeds, volatile byproducts are generated, thereby etching the surface.
The ALE process of TiN has been developed following the miniaturization of semiconductor devices. In a reported ALE process of TiN [17], the first conversion reaction forms a TiO2 layer using oxidizing agents such as O3 or H2O2, whose thickness is quasi-limited by slow O diffusion into TiO2. Then, hydrogen fluoride (HF) etchant reacts with TiO2 in the second reaction to produce the volatile byproducts H2O and TiF4. However, because HF does not etch TiN, the etch process effectively stops when TiO2 is completely removed and TiN becomes exposed. Therefore, a constant etching rate, often called etch per cycle (EPC), can be achieved, which is a requirement for ALE. In other words, the difference in etching tendency between TiN and TiO2 by HF plays a decisive role in the ALE process of TiN. A few other alternatives that reported the ALE processes of TiN also involve the liberation of Ti via halogenation either by F or Cl [18,19,20]. In addition, SiO2 is ubiquitously used in the semiconductor industry, which is known to be etched through a fluorination reaction with HF [21,22]. However, in the aforementioned report on thermal ALE of TiN, an optimized TiN ALE process showed high selectivity against SiO2, so that the etching rate was significantly higher for TiN than SiO2 [17].
Using atomistic computational chemistry simulations, the pathways of each reaction in the etching processes can be explored, and volatile byproducts can be inferred [23,24]. However, since the thermal ALE processes present various theoretical challenges, such as self-limiting reactions and the generation of volatile reaction products, only limited studies have been conducted to identify the reaction mechanisms [25,26,27,28,29,30,31,32]. In particular, while the process conditions for fluorination and removal of Ti atoms from TiO2 by HF were previously molecularly elucidated [33,34], its selectivity against TiN or SiO2 substrates is yet to be known.
In this study, the reactivity toward the etching of TiN, TiO2, and SiO2 substrates by HF is comparatively investigated using density functional theory (DFT) calculations. The molecular adsorption mechanism of HF and possible etching products, such as TiF4, NH3, H2O, and SiF4, were identified. It was observed that the dissociative HF adsorption and the subsequent TiF4 desorption are preferable on TiO2 but not on TiN. The etching of SiO2 by HF was also found to be possible.

2. Computational Methods

DFT calculations were performed using the Vienna ab initio simulation package version 5.4.4. The PBE exchange-correlation functional and D3BJ dispersion correction were applied with projector-augmented wave methods. Surface models of rocksalt cubic TiN (001), anatase TiO2 (101), and quartz SiO2 (0001), which are known to be the most stable crystalline orientations of each material, were constructed (Figure 1). The top layers close to the surface were optimized, while the bottom layers were fixed at their ideal crystalline positions. Cut-off energy values of (450, 450, 500 eV), and slab sizes in (length × width × height) of (4.21 × 4.21 × 10.94), (10.46 × 7.61 × 10.30), and (4.93 × 4.93 × 11.86) Å3 were used for TiN, TiO2, and SiO2, respectively. The vacuum space between the slabs was 15 Å to prevent unphysical interlayer interactions. All calculations applied Monkhorst-Pack k-point mesh settings of 5 × 5 × 1.
Calculations were performed to compare the reaction pathways and energies of HF for the three substrates. For the TiN and the SiO2 substrates, 1 to 4 HF molecules were added. In the case of the TiO2 substrate with a larger size, the calculation was performed by adding 4, 8, 12, and 16 HF, respectively, according to the size ratio. Each energy was divided by the number of HF to confirm the reaction tendency for each substrate.
The electronic energies of adsorption and desorption reactions (Eads) are expressed by the following equation:
E a d s = ( E a d s o r p t i o n + E b y p r o d u c t ) ( E s l a b + E H F ) Here , E a d s = ( E a d s o r p t i o n + E b y p r o d u c t ) ( E s l a b + E p r e c u r s o r )
Here, E a d s o r p t i o n is the energy of the substrate after adsorption of HF molecules; E b y p r o d u c t is the sum of the energy of the gaseous byproducts; E s l a b is the energy of the initial substrate without HF; and E H F is the sum of the energy of HF molecules. For Gibbs free energy (G) calculations, pressure and volume terms are ignored for the solid substrates and only considered for the gaseous molecules, using the following equations.
U ( T ) = U ( 0 ) + Z P E + Δ U ( 0 T )
H ( T ) = U ( T ) + P V
G ( T ) = H ( T ) T S = U ( 0 ) + Z P E + Δ U ( 0 T ) + P V T S
U ( 0 ) , zero-temperature internal energy, can be approximated as electronic energy; ZPE is zero-point vibrational energy; Δ U ( 0 T ) is internal energy change at finite temperature T; H ( T ) is enthalpy; and S is entropy. For the calculation of entropy of the surface structures, the rotational and translational degrees of freedom are neglected.

3. Results and Discussions

The chemical reactivity of oxide or nitride surfaces can be strongly affected by the presence and coverage of OH (hydroxyl) or NH (amino) groups on the surfaces [35,36,37,38,39,40,41]. At process-relevant temperatures (ca. 400–600 K) and under vacuum environments, a low OH group density is expected for TiO2 surfaces [41,42], thus an OH-free bare surface model of TiO2 is considered. In contrast, SiO2 surfaces are expected to possess relatively high OH group coverage even at high temperatures and under vacuum [43], and thus were modeled as fully hydroxylated. In the TiN surface case, to the best of our knowledge, there has not been conclusive experimental observation or a thermodynamic model for the coverage of NH or NH2 groups under fabrication process conditions other than that reporting (100) as the most stable surface orientation [44]. In the kinetic analysis of chemical vapor deposition of TiN, it is often assumed that some NH2 and NH groups are present on the substrate’s surface [45]. Still, it can be suspected that H adatoms on TiN surfaces may easily desorb as H2, considering the low stability and small diffusion barrier of H adatoms on TiN surfaces [46,47,48]. In addition, most previous theoretical studies on various surface reactions of TiN assumed bare surface models without terminating functional groups [49,50,51,52,53]. Therefore, a bare TiN(100) surface is assumed as the model for the TiN fluorination.
Figure 2 presents the surface structures of the TiN, TiO2, and SiO2 substrates according to the number of HF molecules reacted. Adsorption of HF is assumed to initially occur dissociatively so that F and H adatoms form on the (Ti, Si) and (N, O) atoms of the substrate surfaces, respectively. Then, additional adsorption of HF leads to the formation and desorption of TiF4 or SiF4, which creates a vacancy on the surface. It was assumed that the reaction of multiple HF molecules would proceed step-by-step so that the subsequent surface structure from additional HF adsorption originates from the previous reaction’s product.
First, when one HF molecule reacts per one Ti atom on the TiN surface, Ti-F(*) and N-H(*) are formed by dissociative HF adsorption, where (*) denotes the surface-adsorbed species. For the adsorption of the second HF per Ti atom, we assumed two F and H form bonds with each surface Ti and N atoms so that Ti-F2(*) and N-H2(*) are formed. Afterward, the formation and dissociation of one NH3 are assumed to occur upon exposure to the third HF per Ti atom, leaving N vacancy and Ti-F3(*) on the surface. In this structure, some portion of the F atoms is in a bridging configuration between multiple Ti atoms, partially occupying the N vacancy formed by the NH3 desorption. Thereafter, the formation and dissociation of one TiF4 occur in the step of reacting four HF molecules per Ti atom, creating another N-H(*) on the surface. The gaseous NH3 and TiF4 molecules do not participate in the reaction after being generated as volatile reactants. The considered reactions are summarized in Table 1.
TiO2 was also considered as a substrate for the HF adsorption and etching of Ti. Because there are four Ti atoms exposed on the TiO2 surface model, 4, 8, 12, and 16 HF molecules were sequentially considered to react with the TiO2 substrate. Similar to TiN, Ti-F(*) and O-H(*) are formed by dissociative adsorption of the first HF molecule per Ti atom. Then, the second HF molecule per Ti (8 HF molecules on the calculation cell) is assumed to undergo a disproportionation reaction so that Ti-F(*) and Ti-F3(*) are formed, and 1/2 H2O molecule per surface Ti form from the O-H(*) and desorbs at this stage. Here, F atoms bridging between Ti atoms are observable, especially for those in Ti-F3(*) configurations. After exposure to the third HF per Ti atom (12 HF molecules on the calculation cell), the Ti-F(*) is converted to Ti-F3(*), leaving one H2O molecule per Ti. Finally, adsorption of the fourth HF per surface Ti atom would produce TiF4 and H2O as gaseous byproducts, leaving Ti and O vacancies and Ti-F(*) and Ti-F3(*) on the surface. Therefore, we observe the TiO2 surface becoming rougher at the atomic level upon initial exposure to HF. However, such a locally roughened substrate region can be expected to have a higher chemical etching rate upon further exposure to an additional etchant [54,55], eventually creating a smoother surface.
In the SiO2 case, the substrate is assumed to be fully hydroxylated, so there are two OH groups per surface Si atom before exposure to HF. Thus, for the first two HF molecules per Si atom, dissociative adsorption of each HF molecule results in the desorption of one H2O molecule and adsorption of one F on the surface Si atom, first to Si-F(*) and then to Si-F2(*). Then, the third HF is assumed to break the bond between the surface Si and O, thus a Si-OH is formed below Si-F3(*). Finally, adsorption of the fourth HF per surface Si atom would produce SiF4 and H2O as gaseous byproducts, re-generating two OH groups per now-exposed Si atom.
The reaction energies of TiN, TiO2, and SiO2 with HF were calculated and compared (Figure 3a) to compare the tendency toward surface fluorination reactions of various substrates. The energy per HF was calculated by dividing the reaction energy by the number of HFs to compare the reactivity of the three substrates. TiN and SiO2 substrates reacted with 1, 2, 3, and 4 HF molecules, while TiO2 reacted with 4, 8, 12, and 16 HF molecules. The calculated energy change values were divided into the respective number of HF molecules, as shown in Figure 3a.
First, in the case of TiN, the adsorption of the first HF into H and F adatoms is exothermic and can be expected to occur. However, because subsequent adsorption of additional HF, resulting in TiF2(*) and NH2(*), is highly endothermic (2.0 eV above the initial coverage), it can be predicted that the adsorption of two HF molecules per surface Ti atom would be difficult. Such sudden destabilization by adsorption of additional HF may originate from lateral steric hindrance between the TiF2(*) and NH2(*) moieties (Figure 2b). Although the overall energy is stabilized by the adsorption of the third and fourth HF per surface Ti atom to form gaseous NH3 and TiF4, respectively, these steps would need to undergo sequential addition of H and F atoms to the N and Ti atoms on the surface. Thus, it can be predicted that the formation of NH3 and TiF4 volatile byproducts will be difficult through the fluorination reaction of TiN. Furthermore, the etching of the TiN substrate through fluorination is expected to be difficult to occur.
In contrast, for TiO2 and SiO2, each step in the surface reactions with HF is generally exothermic. Furthermore, although some F atoms are suggested to remain after exposure of the computational unit cell to 16 HF molecules, adsorption of more than 16 HF molecules accompanying TiF4 desorption was found to be continuously exothermic (not shown). Therefore, it is predicted that the chemical etching of TiO2 and SiO2 through fluorination by HF is a preferred reaction.
However, it is notable that a fully hydroxylated surface was assumed for SiO2, allowing for easier H2O byproduct formation and facilitating the SiF4 removal; zero OH coverage was assumed on TiO2. As the etching of SiO2 by gaseous HF is known to be significantly enhanced by forming a mixture with H2O [56], similar enhancements may occur for TiO2 as well. In such a case, the H2O generated by the HF reaction may further self-catalytically facilitate the etching reaction of TiO2. In contrast, the less-than-full actual OH coverage on SiO2 may result in lower reactivity toward the HF etching in reality.
The Gibbs energy difference for each reaction temperature between the substrate and HF was calculated (Figure 3b) to more accurately identify the reaction preference according to temperature. The temperature was compared from 250 to 400 °C, and the partial HF pressure was calculated as 80 mTorr, similar to the conditions in the previous experimental study on TiN ALE [17]. A negative value of the Gibbs free energy change predicts a spontaneous reaction, while a positive value predicts the opposite. The Gibbs free energy change value had a positive value in all reaction steps of TiN and HF. Therefore, it can be predicted that the reaction is nonspontaneous. As the reaction involves the adsorption of multiple gaseous molecules onto a surface of solid material, translation entropy loss of the HF molecules results in nonspontaneous adsorption even for the lowest coverage considered; such an effect is more significant at higher temperatures.
The current observation of preference toward fluorination etching partially agrees with the previous experimental report on the TiN ALE process [17]. For the self-limiting etching conditions to hold so that a constant EPC value can be achieved in such an oxidation-fluorination sequence, the HF pulse should selectively etch only TiO2 layers on top of the TiN surface but not the bulk of the TiN materials. Therefore, our theoretical observation of selectivity toward initial fluorination showed that HF readily etches TiO2, but TiN is resistant toward HF adsorption, which explains the core nature of the material properties of TiN versus TiO2 that govern the ALE process.
The projected density of states (PDOS) of TiN(100) surfaces upon sequential reactions with HF are obtained in order to guide future experimental studies. The PDOS of the bare TiN(100) surface (Figure 4a) shows good correspondence with the reported properties of the bulk rocksalt TiN in the literature, exhibiting the characteristic metallic nature of which the EF (Fermi level) is dominated by the Ti atomic orbitals [47,57]. Adsorption of F and H adatoms results in characteristic valence level states, which are F2p at ca. −4 eV and H1s at −7.0 eV below the EF, respectively (Figure 4b–d). In contrast, the creation of Ti vacancy after removal of TiF4 (Figure 4e) does not significantly alter the PDOS compared to the initial TiN(100) surface.

4. Conclusions

The fluorination and etching reactions of TiN, TiO2, and SiO2 surfaces by HF were investigated using DFT calculations. In the case of TiN, it was confirmed that as the adsorption and etching reactions progressed, the reaction was not favored. It was observed that in the TiO2 and SiO2 cases, energy was stabilized as volatile byproducts such as TiF4, SiF4, and H2O were produced. Therefore, it was found that TiO2 and SiO2 can be selectively etched by HF, while TiN is resistant to etching in exposure to HF. Therefore, this study clarified the initial surface reaction mechanism of the ALE process on TiN. Future studies on elucidating the surface chemistry of the emerging ALE processes will further help develop unit semiconductor fabrication processes.

Author Contributions

Conceptualization, B.S.; methodology, B.S. and H.O.; software, J.H.J. and H.O.; validation, J.H.J. and H.O.; formal analysis, J.H.J. and H.O.; investigation, J.H.J. and H.O.; resources, B.S.; data curation, J.H.J. and H.O.; writing—original draft preparation, J.H.J.; writing—review and editing, B.S.; visualization, J.H.J.; supervision, B.S.; project administration, B.S.; funding acquisition, B.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by: Korea Institute for Advancement of Technology (KIAT) grant funded by the Korean Government (MOTIE) (P0012451, The Competency Development Program for Industry Specialist); by the National Research Foundation (NRF) grant funded by the Korean Government (NRF-2021M3H4A6A01048300); and by 2022 Hongik University Research Fund.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Kwak, M.Y.; Shin, D.H.; Kang, T.W.; Kim, K.N. Characteristics of TiN Barrier Layer against Cu Diffusion. Thin Solid Films 1999, 339, 290–293. [Google Scholar] [CrossRef]
  2. Fillot, F.; Morel, T.; Minoret, S.; Matko, I.; Maîtrejean, S.; Guillaumot, B.; Chenevier, B.; Billon, T. Investigations of Titanium Nitride as Metal Gate Material, Elaborated by Metal Organic Atomic Layer Deposition Using TDMAT and NH3. Microelectron. Eng. 2005, 82, 248–253. [Google Scholar] [CrossRef]
  3. Lee, C.; Kuo, Y.-L. The Evolution of Diffusion Barriers in Copper Metallization. JOM 2007, 59, 44–49. [Google Scholar] [CrossRef]
  4. Lima, L.P.B.; Moreira, M.A.; Diniz, J.A.; Doi, I. Titanium Nitride as Promising Gate Electrode for MOS Technology. Phys. Status Solidi C 2012, 9, 1427–1430. [Google Scholar] [CrossRef]
  5. Lee, B.-J.; Kim, Y.-S.; Seo, D.-W.; Choi, J.-W. The Effect of Deposition Temperature of TiN Thin Film Deposition Using Thermal Atomic Layer Deposition. Coatings 2023, 13, 104. [Google Scholar] [CrossRef]
  6. Zhang, S.; Zhu, W. TiN Coating of Tool Steels: A Review. J. Mater. Process. Technol. 1993, 39, 165–177. [Google Scholar] [CrossRef]
  7. Santecchia, E.; Hamouda, A.M.S.; Musharavati, F.; Zalnezhad, E.; Cabibbo, M.; Spigarelli, S. Wear Resistance Investigation of Titanium Nitride-Based Coatings. Ceram. Int. 2015, 41, 10349–10379. [Google Scholar] [CrossRef]
  8. Datta, S.; Das, M.; Balla, V.K.; Bodhak, S.; Murugesan, V.K. Mechanical, Wear, Corrosion and Biological Properties of Arc Deposited Titanium Nitride Coatings. Surf. Coat. Technol. 2018, 344, 214–222. [Google Scholar] [CrossRef]
  9. Saha, N.C.; Tompkins, H.G. Titanium Nitride Oxidation Chemistry: An X-ray Photoelectron Spectroscopy Study. J. Appl. Phys. 1992, 72, 3072–3079. [Google Scholar] [CrossRef]
  10. Kanarik, K.J.; Lill, T.; Hudson, E.A.; Sriraman, S.; Tan, S.; Marks, J.; Vahedi, V.; Gottscho, R.A. Overview of Atomic Layer Etching in the Semiconductor Industry. J. Vac. Sci. Technol. A 2015, 33, 020802. [Google Scholar] [CrossRef] [Green Version]
  11. Fang, C.; Cao, Y.; Wu, D.; Li, A. Thermal Atomic Layer Etching: Mechanism, Materials and Prospects. Prog. Nat. Sci. Mater. Int. 2018, 28, 667–675. [Google Scholar] [CrossRef]
  12. Kim, D.S.; Kim, J.E.; Gill, Y.J.; Jang, Y.J.; Kim, Y.E.; Kim, K.N.; Yeom, G.Y.; Kim, D.W. Anisotropic/Isotropic Atomic Layer Etching of Metals. Appl. Sci. Converg. Technol. 2020, 29, 41–49. [Google Scholar] [CrossRef]
  13. Fischer, A.; Routzahn, A.; George, S.M.; Lill, T. Thermal Atomic Layer Etching: A Review. J. Vac. Sci. Technol. A 2021, 39, 030801. [Google Scholar] [CrossRef]
  14. Hwang, I.-H.; Cha, H.-Y.; Seo, K.-S. Low-Damage and Self-Limiting (Al)GaN Etching Process through Atomic Layer Etching Using O2 and BCl3 Plasma. Coatings 2021, 11, 268. [Google Scholar] [CrossRef]
  15. Fischer, A.; Mui, D.; Routzahn, A.; Gasvoda, R.; Sims, J.; Lill, T. Surface Reaction Modelling of Thermal Atomic Layer Etching on Blanket Hafnium Oxide and Its Application on High Aspect Ratio Structures. J. Vac. Sci. Technol. A 2023, 41, 012601. [Google Scholar] [CrossRef]
  16. George, S.M. Mechanisms of Thermal Atomic Layer Etching. Acc. Chem. Res. 2020, 53, 1151–1160. [Google Scholar] [CrossRef]
  17. Lee, Y.; George, S.M. Thermal Atomic Layer Etching of Titanium Nitride Using Sequential, Self-Limiting Reactions: Oxidation to TiO2 and Fluorination to Volatile TiF4. Chem. Mater. 2017, 29, 8202–8210. [Google Scholar] [CrossRef]
  18. Shinoda, K.; Miyoshi, N.; Kobayashi, H.; Izawa, M.; Ishikawa, K.; Hori, M. Rapid Thermal-Cyclic Atomic-Layer Etching of Titanium Nitride in CHF3/O2 Downstream Plasma. J. Phys. Appl. Phys. 2019, 52, 475106. [Google Scholar] [CrossRef]
  19. Miyoshi, N.; McDowell, N.; Kobayashi, H. Atomic Layer Etching of Titanium Nitride with Surface Modification by Cl Radicals and Rapid Thermal Annealing. J. Vac. Sci. Technol. A 2022, 40, 032601. [Google Scholar] [CrossRef]
  20. Shim, D.; Kim, J.; Kim, Y.; Chae, H. Plasma Atomic Layer Etching for Titanium Nitride at Low Temperatures. J. Vac. Sci. Technol. B 2022, 40, 022208. [Google Scholar] [CrossRef]
  21. Miki, N.; Kikuyama, H.; Kawanabe, I.; Miyashita, M.; Ohmi, T. Gas-Phase Selective Etching of Native Oxide. IEEE Trans. Electron Devices 1990, 37, 107–115. [Google Scholar] [CrossRef]
  22. Kim, D.H.; Kwak, S.J.; Jeong, J.H.; Yoo, S.; Nam, S.K.; Kim, Y.; Lee, W.B. Molecular Dynamics Simulation of Silicon Dioxide Etching by Hydrogen Fluoride Using the Reactive Force Field. ACS Omega 2021, 6, 16009–16015. [Google Scholar] [CrossRef] [PubMed]
  23. Neyts, E.C.; Brault, P. Molecular Dynamics Simulations for Plasma-Surface Interactions. Plasma Process. Polym. 2017, 14, 1600145. [Google Scholar] [CrossRef]
  24. Vanraes, P.; Parayil Venugopalan, S.; Bogaerts, A. Multiscale Modeling of Plasma–Surface Interaction—General Picture and a Case Study of Si and SiO2 Etching by Fluorocarbon-Based Plasmas. Appl. Phys. Rev. 2021, 8, 041305. [Google Scholar] [CrossRef]
  25. Kondati Natarajan, S.; Elliott, S.D. Modeling the Chemical Mechanism of the Thermal Atomic Layer Etch of Aluminum Oxide: A Density Functional Theory Study of Reactions during HF Exposure. Chem. Mater. 2018, 30, 5912–5922. [Google Scholar] [CrossRef]
  26. Mullins, R.; Kondati Natarajan, S.; Elliott, S.D.; Nolan, M. Self-Limiting Temperature Window for Thermal Atomic Layer Etching of HfO2 and ZrO2 Based on the Atomic-Scale Mechanism. Chem. Mater. 2020, 32, 3414–3426. [Google Scholar] [CrossRef]
  27. Basher, A.H.; Krstić, M.; Fink, K.; Ito, T.; Karahashi, K.; Wenzel, W.; Hamaguchi, S. Formation and Desorption of Nickel Hexafluoroacetylacetonate Ni(Hfac)2 on a Nickel Oxide Surface in Atomic Layer Etching Processes. J. Vac. Sci. Technol. A 2020, 38, 052602. [Google Scholar] [CrossRef]
  28. Clancey, J.W.; Cavanagh, A.S.; Smith, J.E.T.; Sharma, S.; George, S.M. Volatile Etch Species Produced during Thermal Al2O3 Atomic Layer Etching. J. Phys. Chem. C 2020, 124, 287–299. [Google Scholar] [CrossRef]
  29. Cheng, E.; Hwang, G.S. Dissociative Chemisorption of Methyl Fluoride and Its Implications for Atomic Layer Etching of Silicon Nitride. Appl. Surf. Sci. 2021, 543, 148557. [Google Scholar] [CrossRef]
  30. Konh, M.; Janotti, A.; Teplyakov, A. Molecular Mechanism of Thermal Dry Etching of Iron in a Two-Step Atomic Layer Etching Process: Chlorination Followed by Exposure to Acetylacetone. J. Phys. Chem. C 2021, 125, 7142–7154. [Google Scholar] [CrossRef]
  31. Hu, X.; Schuster, J. Chemical Mechanism of AlF3 Etching during AlMe3 Exposure: A Thermodynamic and DFT Study. J. Phys. Chem. C 2022, 126, 7410–7420. [Google Scholar] [CrossRef]
  32. Mullins, R.; Gutiérrez Moreno, J.J.; Nolan, M. Origin of Enhanced Thermal Atomic Layer Etching of Amorphous HfO2. J. Vac. Sci. Technol. A 2022, 40, 022604. [Google Scholar] [CrossRef]
  33. Kondati Natarajan, S.; Cano, A.M.; Partridge, J.L.; George, S.M.; Elliott, S.D. Prediction and Validation of the Process Window for Atomic Layer Etching: HF Exposure on TiO2. J. Phys. Chem. C 2021, 125, 25589–25599. [Google Scholar] [CrossRef]
  34. Wang, Y.; Zhang, H.; Han, Y.; Liu, P.; Yao, X.; Zhao, H. A Selective Etching Phenomenon on {001} Faceted Anatase Titanium Dioxide Single Crystal Surfaces by Hydrofluoric Acid. Chem. Commun. 2011, 47, 2829–2831. [Google Scholar] [CrossRef] [PubMed]
  35. 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. Lett. 2015, 6, 3610–3614. [Google Scholar] [CrossRef] [PubMed]
  36. 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, 68515–68524. [Google Scholar] [CrossRef]
  37. Choi, W.; Lee, S.; Han, D.-H.; Lim, H.T.; Park, H.; Lee, G.-D. Reaction Mechanisms of Chlorine Reduction on Hydroxylated Alumina in Titanium Nitride Growth: First Principles Study. Appl. Surf. Sci. 2021, 550, 149391. [Google Scholar] [CrossRef]
  38. Kim, H.-M.; Lee, J.-H.; Lee, S.-H.; Harada, R.; Shigetomi, T.; Lee, S.; Tsugawa, T.; Shong, B.; Park, J.-S. Area-Selective Atomic Layer Deposition of Ruthenium Using a Novel Ru Precursor and H2O as a Reactant. Chem. Mater. 2021, 33, 4353–4361. [Google Scholar] [CrossRef]
  39. Lee, J.; Lee, J.-M.; Oh, H.; Kim, C.; Kim, J.; Kim, D.H.; Shong, B.; Park, T.J.; Kim, W.-H. Inherently Area-Selective Atomic Layer Deposition of SiO2 Thin Films to Confer Oxide Versus Nitride Selectivity. Adv. Funct. Mater. 2021, 31, 2102556. [Google Scholar] [CrossRef]
  40. Ta, H.T.T.; Bui, H.V.; Nguyen, V.-H.; Tieu, A.K. Reactions between SiCl4 and H2O on Rutile TiO2 Surfaces in Atomic Layer Deposition of SiO2 by First-Principles Calculations. Surf. Interfaces 2023, 36, 102454. [Google Scholar] [CrossRef]
  41. Kamphaus, E.P.; Shan, N.; Jones, J.C.; Martinson, A.B.F.; Cheng, L. Selective Hydration of Rutile TiO2 as a Strategy for Site-Selective Atomic Layer Deposition. ACS Appl. Mater. Interfaces 2022, 14, 21585–21595. [Google Scholar] [CrossRef] [PubMed]
  42. Walle, L.E.; Borg, A.; Johansson, E.M.J.; Plogmaker, S.; Rensmo, H.; Uvdal, P.; Sandell, A. Mixed Dissociative and Molecular Water Adsorption on Anatase TiO2(101). J. Phys. Chem. C 2011, 115, 9545–9550. [Google Scholar] [CrossRef]
  43. Zhuravlev, L.T. The Surface Chemistry of Amorphous Silica. Zhuravlev Model. Colloids Surf. Physicochem. Eng. Asp. 2000, 173, 1–38. [Google Scholar] [CrossRef]
  44. Wang, C.; Dai, Y.; Gao, H.; Ruan, X.; Wang, J.; Sun, B. Surface Properties of Titanium Nitride: A First-Principles Study. Solid State Commun. 2010, 150, 1370–1374. [Google Scholar] [CrossRef]
  45. Imhoff, L.; Bouteville, A.; Remy, J.C. Kinetics of the Formation of Titanium Nitride Layers by Rapid Thermal Low Pressure Chemical Vapor Deposition from TiCl4-NH3-H2. J. Electrochem. Soc. 1998, 145, 1672. [Google Scholar] [CrossRef]
  46. You, M.S.; Nakanishi, N.; Kato, E. The Equilibrium of the Chemisorption of TiCl4, H2, and N2 on Titanium Nitride. J. Electrochem. Soc. 1991, 138, 1394. [Google Scholar] [CrossRef]
  47. Marlo, M.; Milman, V. Density-Functional Study of Bulk and Surface Properties of Titanium Nitride Using Different Exchange-Correlation Functionals. Phys. Rev. B 2000, 62, 2899–2907. [Google Scholar] [CrossRef]
  48. Siodmiak, M.; Govind, N.; Andzelm, J.; Tanpipat, N.; Frenking, G.; Korkin, A. Theoretical Study of Hydrogen Adsorption and Diffusion on TiN(100) Surface. Phys. Status Solidi B 2001, 226, 29–36. [Google Scholar] [CrossRef]
  49. Phung, Q.M.; Vancoillie, S.; Pourtois, G.; Swerts, J.; Pierloot, K.; Delabie, A. Atomic Layer Deposition of Ruthenium on a Titanium Nitride Surface: A Density Functional Theory Study. J. Phys. Chem. C 2013, 117, 19442–19453. [Google Scholar] [CrossRef]
  50. Seifitokaldani, A.; Savadogo, O.; Perrier, M. Density Functional Theory (DFT) Computation of the Oxygen Reduction Reaction (ORR) on Titanium Nitride (TiN) Surface. Electrochim. Acta 2014, 141, 25–32. [Google Scholar] [CrossRef]
  51. Kura, C.; Kunisada, Y.; Tsuji, E.; Zhu, C.; Habazaki, H.; Nagata, S.; Müller, M.P.; De Souza, R.A.; Aoki, Y. Hydrogen Separation by Nanocrystalline Titanium Nitride Membranes with High Hydride Ion Conductivity. Nat. Energy 2017, 2, 786–794. [Google Scholar] [CrossRef]
  52. Wang, Y.; Fu, Z.; Zhang, X.; Yang, Z. Understanding the Correlation between the Electronic Structure and Catalytic Behavior of TiC(001) and TiN(001) Surfaces: DFT Study. Appl. Surf. Sci. 2019, 494, 57–62. [Google Scholar] [CrossRef]
  53. Sharma, V.; Kondati Natarajan, S.; Elliott, S.D.; Blomberg, T.; Haukka, S.; Givens, M.E.; Tuominen, M.; Ritala, M. Combining Experimental and DFT Investigation of the Mechanism Involved in Thermal Etching of Titanium Nitride Using Alternate Exposures of NbF5 and CCl4, or CCl4 Only. Adv. Mater. Interfaces 2021, 8, 2101085. [Google Scholar] [CrossRef]
  54. Fang, Z.; Zhang, Y.; Li, R.; Liang, Y.; Deng, H. An Efficient Approach for Atomic-Scale Polishing of Single-Crystal Silicon via Plasma-Based Atom-Selective Etching. Int. J. Mach. Tools Manuf. 2020, 159, 103649. [Google Scholar] [CrossRef]
  55. Gerritsen, S.H.; Chittock, N.J.; Vandalon, V.; Verheijen, M.A.; Knoops, H.C.M.; Kessels, W.M.M.; Mackus, A.J.M. Surface Smoothing by Atomic Layer Deposition and Etching for the Fabrication of Nanodevices. ACS Appl. Nano Mater. 2022, 5, 18116–18126. [Google Scholar] [CrossRef]
  56. Helms, C.R.; Deal, B.E. Mechanisms of the HF/H2O Vapor Phase Etching of SiO2. J. Vac. Sci. Technol. A 1992, 10, 806–811. [Google Scholar] [CrossRef]
  57. Yu, S.; Zeng, Q.; Oganov, A.R.; Frapper, G.; Zhang, L. Phase Stability, Chemical Bonding and Mechanical Properties of Titanium Nitrides: A First-Principles Study. Phys. Chem. Chem. Phys. 2015, 17, 11763–11769. [Google Scholar] [CrossRef] [Green Version]
Figure 1. Slab models of cubic TiN(001) (a,b), anatase TiO2(101) (c,d), and quartz SiO2(0001) (e,f) surfaces, showing side view (a,c,e) and top view (b,d,f). Ti = gray, Si = brown, O = red, N = blue, and H = white.
Figure 1. Slab models of cubic TiN(001) (a,b), anatase TiO2(101) (c,d), and quartz SiO2(0001) (e,f) surfaces, showing side view (a,c,e) and top view (b,d,f). Ti = gray, Si = brown, O = red, N = blue, and H = white.
Coatings 13 00387 g001
Figure 2. DFT-calculated surface structures of the TiN, TiO2, and SiO2 substrates according to the number of HF molecules (1–4) reacted; (a) the side view and (b) the top view. The removed surface Ti and Si atomic sites are marked with yellow. Ti = gray, Si = brown, F = green, O = red, N = blue, and H = white. A part of the sub-surface atoms is hidden for presentation clarity.
Figure 2. DFT-calculated surface structures of the TiN, TiO2, and SiO2 substrates according to the number of HF molecules (1–4) reacted; (a) the side view and (b) the top view. The removed surface Ti and Si atomic sites are marked with yellow. Ti = gray, Si = brown, F = green, O = red, N = blue, and H = white. A part of the sub-surface atoms is hidden for presentation clarity.
Coatings 13 00387 g002
Figure 3. Energy change according to the number of HF molecules adsorbed on the substrates: (a) electronic energy of TiN, TiO2, and SiO2, and (b) Gibbs free energy changes of TiN at 250 °C and 400 °C.
Figure 3. Energy change according to the number of HF molecules adsorbed on the substrates: (a) electronic energy of TiN, TiO2, and SiO2, and (b) Gibbs free energy changes of TiN at 250 °C and 400 °C.
Coatings 13 00387 g003
Figure 4. Projected density of states (PDOS) plots of the TiN surface structures upon reaction with HF with varying numbers of HF molecules; (a) bare TiN, (be) 1–4 HF per surface Ti, as defined in Figure 2.
Figure 4. Projected density of states (PDOS) plots of the TiN surface structures upon reaction with HF with varying numbers of HF molecules; (a) bare TiN, (be) 1–4 HF per surface Ti, as defined in Figure 2.
Coatings 13 00387 g004
Table 1. The sequential and overall surface reactions of each substate material with HF.
Table 1. The sequential and overall surface reactions of each substate material with HF.
TiN TiN ( s ) + HF ( g )   TiF ( ) +   NH ( ) (reaction 1)
TiF ( ) +   NH ( ) + HF ( g )   TiF 2 ( ) +   NH 2 ( ) (reaction 2)
TiF 2 ( ) +   NH 2 ( ) + HF ( g )   TiF 3 ( ) + NH 3 ( g ) (reaction 3)
TiF 3 ( ) + HF ( g )   TiF 4 ( g ) + NH ( ) (reaction 4)
Overall :   2 TiN ( s ) + 4 HF ( g ) TiF 4 ( g ) + NH 3 ( g ) + NH ( ) (reaction 5)
TiO2 TiO 2 ( s ) + HF ( g ) TiF ( ) + OH ( ) (reaction 6)
TiF ( ) + OH ( ) +   HF ( g ) 1 2 Ti F ( ) + 1 2   TiF 3 ( ) + OH ( ) + 1 2   H 2 O ( g ) (reaction 7)
1 2 TiF ( ) + OH ( ) +   HF ( g ) 1 2   TiF 3 ( ) + H 2 O ( g ) (reaction 8)
TiF 3 ( ) + HF ( g ) 1 2 TiF 4 ( g ) + 1 2 TiF ( ) + 1 2 H 2 O ( g ) (reaction 9)
Overall :   TiO 2 ( s ) + 4 HF ( g ) 1 2 TiF 4 ( g ) + 1 2 TiF ( ) + 1 2   TiF 3 ( ) + 2 H 2 O ( g ) (reaction 10)
SiO2 SiO 2 ( s ) + 2   OH ( ) + HF ( g ) SiF ( ) + OH ( ) + H 2 O ( g ) (reaction 11)
SiF ( ) + OH ( ) + HF ( g ) SiF 2 ( ) + H 2 O ( g ) (reaction 12)
SiF 2 ( ) + HF ( g ) SiF 3 ( ) + OH ( )   (reaction 13)
SiF 2 ( ) + OH ( ) + HF ( g ) SiF 4 ( g ) + 2 OH ( ) (reaction 14)
Overall :   SiO 2 ( s ) + 4 HF ( g ) SiF 4 ( g ) + 2 H 2 O ( g ) (reaction 15)
(*) denotes the surface-adsorbed species.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jung, J.H.; Oh, H.; Shong, B. Fluorination of TiN, TiO2, and SiO2 Surfaces by HF toward Selective Atomic Layer Etching (ALE). Coatings 2023, 13, 387. https://doi.org/10.3390/coatings13020387

AMA Style

Jung JH, Oh H, Shong B. Fluorination of TiN, TiO2, and SiO2 Surfaces by HF toward Selective Atomic Layer Etching (ALE). Coatings. 2023; 13(2):387. https://doi.org/10.3390/coatings13020387

Chicago/Turabian Style

Jung, Ju Hyeon, Hongjun Oh, and Bonggeun Shong. 2023. "Fluorination of TiN, TiO2, and SiO2 Surfaces by HF toward Selective Atomic Layer Etching (ALE)" Coatings 13, no. 2: 387. https://doi.org/10.3390/coatings13020387

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop