Journal Home > Volume 10 , issue 3

Chemical vapor deposition is an important method for the preparation of boron carbide. Knowledge of the correlation between the phase composition of the deposit and the deposition conditions (temperature, inlet gas composition, total pressure, reactor configuration, and total flow rate) has not been completely determined. In this work, a novel approach to identify the kinetic mechanisms for the deposit composition is presented. Machine leaning (ML) and computational fluid dynamic (CFD) techniques are utilized to identify core factors that influence the deposit composition. It has been shown that ML, combined with CFD, can reduce the prediction error from about 25% to 7%, compared with the ML approach alone. The sensitivity coefficient study shows that BHCl2 and BCl3 produce the most boron atoms, while C2H4 and CH4 are the main sources of carbon atoms. The new approach can accurately predict the deposited boron–carbon ratio and provide a new design solution for other multi-element systems.


menu
Abstract
Full text
Outline
About this article

Machine learning and a computational fluid dynamic approach to estimate phase composition of chemical vapor deposition boron carbide

Show Author's information Qingfeng ZENGa,b( )Yong GAOaKang GUANc( )Jiantao LIUdZhiqiang FENGe,f
Science and Technology on Thermostructural Composite Materials Laboratory, School of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an 710072, China
MSEA International Institute for Materials Genome, Gu’an 065500, China
School of Materials Science and Engineering, South China University of Technology, Guangzhou 510640, China
School of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031, China
School of Mechanics and Engineering, Southwest Jiaotong University, Chengdu 610031, China
LMEE-UEVE, Université Paris-Saclay, Evry 91020, France

Abstract

Chemical vapor deposition is an important method for the preparation of boron carbide. Knowledge of the correlation between the phase composition of the deposit and the deposition conditions (temperature, inlet gas composition, total pressure, reactor configuration, and total flow rate) has not been completely determined. In this work, a novel approach to identify the kinetic mechanisms for the deposit composition is presented. Machine leaning (ML) and computational fluid dynamic (CFD) techniques are utilized to identify core factors that influence the deposit composition. It has been shown that ML, combined with CFD, can reduce the prediction error from about 25% to 7%, compared with the ML approach alone. The sensitivity coefficient study shows that BHCl2 and BCl3 produce the most boron atoms, while C2H4 and CH4 are the main sources of carbon atoms. The new approach can accurately predict the deposited boron–carbon ratio and provide a new design solution for other multi-element systems.

Keywords:

machine learning (ML), computational fluid dynamic (CFD), chemical vapor deposition, boron carbide, B/C ratio, kinetic mechanisms
Received: 10 August 2020 Revised: 24 December 2020 Accepted: 06 January 2021 Published: 26 April 2021 Issue date: June 2021
References(40)
[1]
Ohnabe H, Masaki S, Onozuka M, et al. Potential application of ceramic matrix composites to aero-engine components. Compos Part A-Appl S 1999, 30: 489-496.
[2]
Inghels E, Lamon J. An approach to the mechanical behaviour of SiC/SiC and C/SiC ceramic matrix composites. J Mater Sci 1991, 26: 5411-5419.
[3]
Christin F. Design, fabrication, and application of thermostructural composites (TSC) like C/C, C/SiC, and SiC/SiC composites. Adv Eng Mater 2002, 4: 903-912.
[4]
Katoh Y, Snead LL, Henager CH, et al. Current status and recent research achievements in SiC/SiC composites. J Nucl Mater 2014, 455: 387-397.
[5]
Naslain R, Guette A, Rebillat F, et al. Boron-bearing species in ceramic matrix composites for long-term aerospace applications. J Solid State Chem 2004, 30: 489-496.
[6]
Liu YS, Cheng LF, Zhang LT, et al. Oxidation protection of multilayer CVD SiC/B/SiC coatings for 3D C/SiC composite. Mat Sci Eng A-Struct 2007, 466: 172-177.
[7]
Sezer AO, Brand JI. Chemical vapor deposition of boron carbide. Mater Sci Eng B-Adv 2001, 79: 191-202.
[8]
Jacques S, Guette A, Langlais F, et al. C(B) materials as interphases in SiC/SiC model microcomposites. J Mater Sci 1997, 32: 983-988.
[9]
Ruggles-Wrenn MB, Wallis TA. Creep in interlaminar shear of an Hi-Nicalon™/SiC–B4C composite at 1300 ℃ in air and in steam. J Compos Mater 2019, 54: 1819-1829.
[10]
Deshpande SV, Gulari E, Harris SJ, et al. Filament activated chemical vapor deposition of boron carbide coatings. Appl Phys Lett 1994, 65: 1757-1759.
[11]
Karaman M, Sezgi NA, Doğu T, et al. Kinetic investigation of chemical vapor deposition of B4C on tungsten substrate. AICHE J 2006, 52: 4161-4166.
[12]
Karaman M, Sezgi NA, Doğu T, et al. Mechanism studies on CVD of boron carbide from a gas mixture of BCl3, CH4, and H2 in a dual impinging-jet reactor. AICHE J 2009, 55: 701-709.
[13]
Berjonneau J, Langlais F, Chollon G, et al. Understanding the CVD process of (Si)–B–C ceramics through FTIR spectroscopy gas phase analysis. Surf Coat Tech 2017, 201: 7273-7285.
[14]
Berjonneau J, Chollon G, Langlais F, et al. Deposition process of Si–B–C ceramics from CH3SiCl3/BCl3/H2 precursor. Thin Solid Films 2008, 516: 2848-2857.
[15]
Liu YS, Zhang LT, Cheng LF, et al. Uniform design and regression analysis of LPCVD boron carbide from BCl3–CH4–H2 system. Appl Surf Sci 2009, 255: 5729-5735.
[16]
Zeng B, Feng ZD, Li SW, et al. Microstructural study of oxidation of carbon-rich amorphous boron carbide coating. Front Mater Sci 2008, 2: 375-380.
[17]
Mollick PK, Venugopalan R, Srivastava D. CFD coupled kinetic modeling and simulation of hot wall vertical tubular reactor for deposition of SiC crystal from MTS. J Cryst Growth 2017, 475: 97-109.
[18]
Ni H, Lu S, Chen C. Modeling and simulation of silicon epitaxial growth in siemens CVD reactor. J Cryst Growth 2014, 404: 89-99.
[19]
Deck CP, Khalifa HE, Sammuli B, et al. Fabrication of SiC–SiC composites for fuel cladding in advanced reactor designs. Prog Nucl Energy 2012, 57: 38-45.
[20]
Reinisch G, Patel S, Chollon G, et al. Methyldichloroborane evidenced as an intermediate in the chemical vapour deposition synthesis of boron carbide. J Nanosci Nanotechnol 2011, 11: 8323-8327.
[21]
Li J, Qin H, Liu Y, et al. Effect of the SiCl4 flow rate on SiBN deposition kinetics in SiCl4–BCl3–NH3–H2–Ar environment. Materials 2017, 10: 627-637.
[22]
Kleijn CR. Modeling of Chemical Vapor Deposition of Tungsten Films. Boston, USA: Birkhäuser Basel, 1993.
[23]
Beek WJ, Muttzall KMK, van Heuven JW. Transport Phenomena, 2nd edn. New York, USA: John Wiley & Sons, 1999.
[24]
Cuadros F, Cachadiña I, Ahumada W. Determination of Lennard-Jones interaction parameters using a new procedure. Mol Eng 1996, 6: 319-325.
[25]
Ge Y, Gordon M S, Battaglia F, et al. Theoretical study of the pyrolysis of methyltrichlorosilane in the gas phase. 1. Thermodynamics. J Phys Chem A 2007, 111: 1462-1474.
[26]
Ge Y, Gordon M S, Battaglia F, et al. Theoretical study of the pyrolysis of methyltrichlorosilane in the gas phase. 2. Reaction paths and transition states. J Phys Chem A 2007, 111: 1475-1486.
[27]
Ge Y, Gordon M S, Battaglia F, et al. Theoretical study of the pyrolysis of methyltrichlorosilane in the gas phase. 3. Reaction Rate Constant Calculations. J Phys Chem A 2010, 114: 2384-2392.
[28]
Liu Y, Su KH, Zeng QF, et al. Reaction paths of BCl3 + CH4 + H2 in the chemical vapor deposition process. Struct Chem 2012, 23: 1677-1692.
[29]
Liu Y, Su KH, Zeng QF, et al. Decomposition reaction rate of BCl3–CH4–H2 in the gas phase. Theor Chem Acc 2015, 134: 1-9.
[30]
Reinisch G, Leyssale JM, Vignoles GL. Theoretical study of the decomposition of BCl3 induced by a H radical. J Phys Chem A 2011, 115: 4786-4797.
[31]
Lee J H, Shin J, Realff M J. Machine learning: Overview of the recent progresses and implications for the process systems engineering field. Comput Chem Eng 2018, 114: 111-121.
[32]
Moreno R, Corona F, Lendasse A, et al. Extreme learning machines for soybean classification in remote sensing hyperspectral images. Neurocomputing 2014, 128: 207-216.
[33]
Basu A, Shuo S, Zhou H, et al. Silicon spiking neurons for hardware implementation of extreme learning machines. Neurocomputing 2013, 102: 125-134.
[34]
Benoît F, Heeswijk M, Miche Y, et al. Feature selection for nonlinear models with extreme learning machines. Neurocomputing 2013, 102: 111-124.
[35]
Feng S, Zhou H, Dong H, et al. Using deep neural network with small dataset to predict material defects. Mater Des 2019, 162: 300-310.
[36]
Kushvaha V, Kumar S A, Madhushri P, et al. Artificial neural network technique to predict dynamic fracture of particulate composite. J Compos Mater 2021, 54: 3099-3108.
[37]
Liu X, Gao C, Li P, et al. A comparative analysis of support vector machines and extreme learning machines. Neural Netw 2012, 33: 58-66.
[38]
Berjonneau J, Chollon G, Langlais F, et al. Deposition process of amorphous boron carbide from CH4/BCl3/H2 precursor. Proc Electrochem Soc 2006, 153: C795-C800.
[39]
Vandenbulcke L G. Theoretical and experimental studies on the chemical vapor deposition of boron carbide. Ind Eng Chem Res 2002, 24: 568-575.
[40]
Lartigue S, Cazajous D, Nadal M, et al. Study of boron carbides vapor-deposited under low pressure. In Proceedings of the Fifth European Conference on Chemical Vapor Deposition, 1985: 413-419.
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 10 August 2020
Revised: 24 December 2020
Accepted: 06 January 2021
Published: 26 April 2021
Issue date: June 2021

Copyright

© The Author(s) 2021

Acknowledgements

We thank the National Key R&D Program of China (Grant No. 2017YFB0703200), National Natural Science Foundation of China (Grant Nos. 51702100 and 51972268), and China Postdoctoral Science Foundation (Grant No. 2018M643075) for the financial support.

Rights and permissions

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.

The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.

Reprints and Permission requests may be sought directly from editorial office.

Return