TUNG Ho Kwan

Document Sample
TUNG Ho Kwan Powered By Docstoc
					  CITY UNIVERSITY OF HONG KONG
                DEPARTMENT OF
 PHYSICS AND MATERIALS SCIENCE



BACHELOR OF ENGINEERING (HONS) IN MATERIALS ENGINEERING

                             2010-2011

                          DISSERTATION




         Deposition of cubic boron nitride and diamond films




                                 By




                           Tung Ho Kwan




                             March 2011
Deposition of cubic boron nitride and diamond films




                        By




                  Tung Ho Kwan




       Submitted in partial fulfilment of the

          Requirements for the degree of

    BACHELOR OF ENGINEERING (HONS)

                        IN

          MATERIALS ENGINEERING

                       From

          City University of Hong Kong




                   March 2011




 Project Supervisor:            Dr. Wenjun Zhang
                                                                                         i



Abstract:


      Cubic boron nitride (cBN) has attracted great attention among materials


scientists due to its excellent physical and chemical properties, which are similar to


or even superior to diamond. Yet, the development of cBN has been held back by the


limited qualities of the material available. In this dissertation, we will mainly focus on


(i) the background and historical development of cBN films, (ii) the deposition of cBN


films by using electron cyclotron resonance microwave plasma chemical vapor


deposition (ECR MPCVD), (iii) the synthesis of boron nitride films by utilizing BCl3


instead of BF3 in ECR MPCVD, and (iv) characterization and analysis of cBN films by


using Raman spectroscopy, Fourier-transform infrared spectroscopy (FTIR), X-ray


photoelectron spectroscopy (XPS) and Scanning electron microscopy (SEM).



      The cBN films deposited by different conditions or parameters are presented in


this report. It was found that the decrease of the substrate bias leads to improved


crystal quality of cBN films, which has relatively less crevasses that caused by residual


film stress. Also, the syntheses of cBN films onto different substrates were compared,


and, B-doped poly-diamond substrate was discovered to be the best substrate for


growing the cBN films in this report. Increasing the deposition duration can form a


thicker cBN films but it may also increase the film stress of the cBN films. Increasing
                                                                                     ii



hydrogen flow rate during the deposition processing will induce the formation of


hexagonal boron nitride (hBN), which lowers the cBN content and qualities of the


films. Moreover, it was also revealed that BN films deposited by using BCl3 instead of


BF3 in this experiment mainly composed of hBN phase.
                                                                                  iii



Objective:

  1. Study of basic knowledge and historical development of cubic boron nitride
     (cBN).


  2. Deposition of cBN films by using an electron cyclotron resonance microwave
     plasma chemical vapor deposition (MPCVD).

  3. Synthesis of high quality cBN films by applying different gas ratios, duration,
     substrates, and substrate bias.

  4. Synthesis of cubic boron nitride films by utilizing BCl3 instead of BF3 in ECR
     MPCVD.

  5. Characterization of cBN films by using Fourier-transform infrared
     spectroscopy (FTIR), Raman spectroscopy, scanning electron microscopy
     (SEM) and/or transmission electron microscopy (TEM).
                                                                                   iv



Acknowledgements:


     This final year project was supported by my supervisor, Dr. Wenjun Zhang, who is


very friendly and always helps me when I am in trouble. I would also like to thank my


second assessor, Prof. Joseph Lai, who usually gives me useful suggestion in this


project. Special thanks are given to my laboratory tutor, Mr. YE Qing, who teaches me

in any respect during the completion of this final year project.
                                                                                  v



                              Contents:
Abstract……………………………………………………………………………………………………………..i

Objectives……………………………………………………………………………………………….……….iii
Acknowledgements……………………………………………………………………………………………....vi

Chapter 1 – Historical development and background information of
Cubic Boron Nitride films
1.1 – Introduction information of cubic boron nitride.………………………………………….P.1

1.3 – Properties of cubic boron nitride……………………………………………………………...….P.5

         1.3.1 – Mechanical properties……………………………………………………….….………P.5

         1.3.2 – Chemical reactivity and thermal stability……………………………….….…….P.5

         1.3.3 – Electrical & electronic properties………………………………………..…….…...P.6

1.4 – Factors hampering the development of cBN films & Solution…………..…..….……P.7


Chapter 2 - Synthesis of cubic boron nitride films
2.1 – Synthesis by Chemical Vapor Deposition……………………………………………….…..…P.9

2.2 – Electron cyclotron resonances microwave plasma CVD…………………………....…P.10

2.3 – Substrate selection……………………………………………………………………………………...P.11

2.4 – Common methods for synthesis of cBN films……………………………………….………P.13


Chapter 3 – Techniques for characterization of cBN films

3.1 – Raman spectroscopy……………………………………………………………….…………………..P.15

3.2 – Fourier transform infrared spectroscopy………………………………………….………….P.16

3.3 – Scanning electron microscopy………………………………………….………………………….P.18

3.4 – Common techniques used for characterization…………………………………………...P.19
                                                                                        vi




Chapter 4 – Experimental procedures for synthesis and characterization
of cBN films

4.1 – Synthesis of cBN films………………………………………………………………………….…..P.21

         4.1.1 –Preparation of sample substrate for ECR MPCVD………..………..….…P.21

         4.1.2 –Operation of ECR MPCVD……………………………………….…………….....P.21

4.2 – Characterization of cBN films………………………………………………………………..….P.23

         4.2.1 –Raman spectroscopy………………………………………….……..….…...……P.23

         4.2.2 –Fourier transform infrared spectroscopy…………………………………...P.23

         4.2.3 – Scanning electron microscopy…………………………………………………..P.24


Chapter 5 – Experimental results and discussion

5.1 – Experimental results and analysis……………………………………………………….……..P.25

     5.1.1 – The effect of applying different bias voltages...…………………….……….……P.26

     5.1.2 – Synthesis of BN films by using various substrates…………..……………..……P.29

     5.1.3 – The study of cBN film formed by different duration……………………………P.33

     5.1.4 –Gas ratio dependence for deposition of cBN films……………....……………...P.35

     5.1.5 – Synthesis of BN films by utilizing BCl3 instead of BF3 ……...……………………P.38


Chapter 6 – Conclusion…………………………………………………………………………….….P.40
References…………………………………………………………………………………………………..…P.41
                                                                                   vii




List of Figures

 Figure 1. Crystal structure of hBN, wBN, rBN and cBN. The black colour one is
            Boron atoms, the white is nitrogen


 Figure 2.    A diagram of ECR MPCVD system for the deposition of cBN films


 Figure 3. HRTEM image of the interface region between cBN and silicon substrate


 Figure 4. HRTEM image for cBN grown on CVD diamond substrate


 Figure 5. Raman spectra of (a) hBN and (b) cBN crystallites


 Figure 6.    Transmission infrared spectra of (a) hBN and (b) cBN films


 Figure 7.    (a) cross section view of cBN film deposited by DC jet plasma CVD, (b)
               plan view of cBN film deposited by a two-step process and by
               performing the fluorine chemistry


 Figure 8.    Typical FTIR background signal of the substrates


 Figure 9.    (a) SEM cross-section, (c) plan view image and (e) IR spectrum of cBN
               film deposited by -25V. (b) SEM cross-section,(d) plan view image and
              (f) IR spectrum of cBN film deposited by -20V.


 Figure 10.    SEM plan-view image of BN film deposited on (a) Si and (b)
                poly-Diamond/Si substrate, (c)XPS spectrum of BN film deposited
                on poly-Diamond/Si substrate


 Figure 11.   IR spectra of cBN deposited by B-doped (a) poly-diamond and (b)
              nano-diamond substrate.


 Figure 12.    The FTIR spectrums of cBN films made with (a)1 hr 9 and (b)1.5hr.
                                                                                      viii




 Figure 13.    FTIR spectra of cBN films deposited by gas ratio (a) H2/BF3 = 1:5
               and (b) H2/BF3 = 1:4


 Figure 14.    Raman spectra of cBN films deposited by gas ratio (a) H2/BF3 = 1:5
               and (b) H2/BF3 = 1:4


 Figure 15. SEM Plan view (a) 20000X and (b) 5000X image of BN films
               deposited by using BCl3


Figure 16.    IR spectra of BN films deposited by using BCl3 and different hydrogen
              ratio. (a) H2:BCl3 = 1:5, (b) H2:BCl3 = 1.5:10
                                                                   ix



List of Tables


Table 1. Electrical and electronic properties of cBN films



Table 2. Experiment parameters of BN films produced by ECR MPCVD
                                                                                        1



Chapter 1 – History and background information of cBN films
1.1 - Introduction of cubic Boron Nitride



   Cubic boron nitride (cBN), which is not a natural material, is a synthetic material


that has many superexcellent characteristics, such as the second hardest material, the


second most thermally conductive material, high resistance to oxidation, chemical


stability, both p-type or n-type doping available and so on. Due to these superior


material properties, it has been activity investigated for thin film applications [1]. The


Vickers hardness and thermal conductivity of cBN is about 5000 kgmm -2 and 13


Wcm-1 K-1 respectively, which is the second highest next to diamond. Also, cBN has


very good stability, for example, the chemical inertness and thermal stability of cBN

are superior to diamond [2]. While diamond burns at ~ 600oC and dissolves in iron


at elevated temperatures, cBN is chemically stable and does not react with ferrous

materials at temperature up to ~1200oC[2,3]. The band gap of cBN is ~6.3eV,


which is quite wide and larger than that of diamond ~5.5eV. [4,5]. Furthermore,


cBN can be potentially a better semiconducting material than diamond because it can


be doped to be both p- and n-type conductivity, while there are some hindering factors


for n-type conductivity of diamond [6-8]. Because of these desirable advantages of


cBN, it often becomes a candidate material for fabrication of some tools or devices


designed to operate in extreme conditions.
                                                                                       2



     Nowadays, cBN powders, which can be synthesized by High Pressure High


Temperature (HPHT) methods, are commercially available as abrasives, and they are


also molded and cemented by metal binders to manufacture cutting tools with some


desirable shapes [9]. Yet, it is difficult to cement cBN powder into complex shapes, so


it is necessary to develop a new technology for cBN synthesizing. Virtually, there are


two popular techniques that can be used for cBN synthesizing, which are Chemical


vapor deposition (CVD) and Physical vapor deposition (PVD) methods. The method


of using CVD will be deeply discussed in this project. Also, since the structure,


properties, evolution and the application of cBN and diamond are similar, it is of great


significance to consider several points related to the synthetic methods of these two


materials. In the paragraphs that follow, we will introduce the crystal structure and


properties of BN, and then the testing methods and the properties of cubic boron


nitride will be discussed.



                       1.2 The crystal structure of Boron Nitride



   The structure of cBN is similar to diamond, with sp3 atomic bonding in zinblende


structure. Actually, boron nitride (BN) has primarily four phases, which include cubic


boron nitride (cBN), hexagonal boron nitride, wurtzite boron nitride (wBN) and


rhombohedral boron nitride (rBN), as shown Figure 1 [2,6]. It is known that cBN and
                                                                                                   3



wBN are sp3 bonded structure, but hBN and rBN are sp2 bonded structure. Besides,


during the preparing processing of cBN, it is found that amorphous BN (aBN) and


turbostratic BN (tBN) are usually present. However, cBN is the most attractive of


these phases because of what I mentioned before.




  Figure 1. Crystal structure of hBN, wBN, rBN and cBN. The black colour one is Boron atoms, the

                                      white is nitrogen[10].


      The hBN has an sp2-bonded layered structure with 2D, six-membered rings of


alternating boron and nitrogen atoms (stacked in a AA’AA’…), and, rBN is quite


similar to the graphite-like modification of hBN (stacked in ABCABC…). Also, the


metastable wBN is sp3-bonded and hexagonal in structure [10].
                                                                                    4



      However, it is discovered that when applying direct compression along the


hexagonal axis, these stacking differences can cause the production of a diamond-like


tetrahedral structure. The applied compression of hBN and rBN and splitting of basal


planes will cause strong chemical bond formation between (0002) planes, which will


consequently induce a tetrahedral coordination of atoms and their stacking sequences


are preserved, for instance, hBN change to wBN, rBN change to cBN [11].




      Since our topic is read up on cBN, we will mainly focus on the observation of


the cBN. The structure of cBN can be seen as two interpenetrating face centre cubic


(FCC) sub-lattices, each containing one type of atom and shifted over 1/4 of the


lattice along the diagonal direction. The bonding of cBN is covalent, and slightly


ionic in nature, which is related to the presence of longitudinal optical (LO) and


transverse optical (TO) phonon modes in the spectra that will be discussed later.


Besides, it is found that turbostratic BN and amorphous BN may also have a chance to


form during the cBN deposition processing. Actually, tBN is characterized by an


accidental stacking of the layers rotated randomly around the c-axis, it also has


extensive lattice curvature, normally a larger interplanar spacing by around 15% than


the basal planes of the ordered hBN [12]. Yet, the disordered BN form is aBN that has


non-datectable order on microscopic observation.
                                                                                       5



                          1.3 - Properties of cubic boron nitride


   1.3.1 – Mechanical properties


   To determine the mechanical properties of cBN, we usually measure their


hardness and elastic modulus. However, if we use different methods for synthesis of


cBN films, the hardness and elastic modulus will be different. As mentioned


previously, the Vickers hardness of cBN films can be up to about 5000 kgmm-2, and,


the elastic modulus up to about 500 GPa[2,13], which are extremely high, making it


the second hardest material in the world. Therefore, cBN films are superior materials


in the field of cutting tools. Experts nowadays want to develop thicker cBN films in


order to increase the tool lifetimes and cutting speeds.

   1.3.2 – Chemical properties


     cBN has very high chemical stability, which is chemically inert to a lot of metals,


especially molten ferrous materials up to 1300oC. In addition, DeVires discovered that


cBN reacts with Fe, Ni and Co at 1620-1670K and with Ni- and/or Fe based alloys at


1520-1570K in Ar atmosphere [14]. Besides, usual acid and alkali solutions cannot


solubilize the cBN, whereas cBN is soluble in molten alkalis and nitrides. Therefore,


metal nitrides are widely utilized as catalysts or solvents in HPHT synthesis of cBN.


Also, the temperature-dependent oxidation of diamond, graphite, hBN and cBN in air


stream have been studied by thermogravimetry within the range from 30 to 1300oC
                                                                                       6



[15]. At 600 to 700oC, the weight of diamond and graphite are observed to start


decreasing, and, when the temperature up to 800oC, the weight were dropped rapidly


then completely vanished at 900 to 1000oC because of pyrolysis. Yet, hBN and cBN


has good oxidation resistance that never changes will be observed in the weight up to


1200oC. Just ~10% gain weight of hBN and cBN was found and it is due to the


formation of B2O3 [16].

1.3.3 Electrical & electronic properties of cubic-BN




         The band gap of cBN is about 6.2eV and 5.5eV for hBN [4, 5]. One of the


benefits of cBN is that it can be both n- and p-type doping. Intrinsic or unintentionally


doped cBN can exists p-type or n-type [17-20] semiconducting character which


depends on the preparation methods. The origin of conductivity is not clearly


determined yet. In addition, Be and Mg have been known to be p-type dopants; while


Si and S shown as n-type donors in cubic boron nitride, shown as table 1 [21].


However, a number of studies reported that electronic properties of Si-doped cBN are
                                                                                          7



better that the Sulfur doped one [22], because the electron mobility of Si-doped cBN


can be attained to about 825cm2V-1s-1, comparing with 1 cm2V-1s-1 of Sulfur-doped.


When using Be and Si for p- and n-type doping of high-pressure high-temperature


cBN crystals, the cBN p-n junction diode can be             fabricated and operated normally


between 25 to 650oC.[7, 21].

             1.4 – Hampering factors for the development of cBN films


A) The stress of the films


   The stress produced in the cBN films can create a great number of defects like


interstitial, cracks. Actually, the stress can be observed to have magnitudes around


30GPa [23]. The origin of the stress is mainly due to ion bombardment. Ion


bombardment is essential during the growth of cBN, but it can destroy the growing


film when the bombardment is strong. Even with bombardment by low energy ions,


defects can be induced and cause stress compression. In order to solve this problem,


Reinke et al. [24] reported a balance of defect creation and relaxation, the stress can


be related to the ion energy by:



                                                  -------------------(1)

              where   is the Young’s modulus, v is the Poisson ratio of the cBN film.


Also, recent literature pointed out using the ion energy bombardment of the range


between 10keV to 1MeV, which can considerably decrease the stress of cBN films
                                                                                      8



after deposition. Besides that, it is discovered that there are some measures for


decreasing the stress of cBN film, such as, annealing, ion implantation, third element


addition. On the other hand, some studies also found that considerable stress reduction


can be obtained if after nucleation of cBN the ion energy is decreased [25].

B) The adhesion of the films


   The poor adhesion of the cBN films is mainly caused by the formation of


interlayers such as aBN and tBN. An early research pointed out a method that avoids


this layer in order to improve the adhesion, but it seems very hard to achieve. In


addition, some studies suggested the use of rough interfaces between substrate (or


adhesion interlayer) and the hBN nucleation layer, on the one hand, and hBN layer


and cBN film on the other [25]. By using this method, it considerably improved the


adhesion of the cBN films. In fact, such adhesion layers are typical techniques to


improve the adhesion and widely used in thin film application.

C) Other difficulties


   Apart from the film stress and adhesion, other problems like limited film thickness,


poor crystallinity and random orientation also affected the development of cBN films


application. Therefore, there are a lot of reports trying to overcome these difficulties


by using different substrates, different parameters for smoothing the nucleation and


growth of cBN thin films, and apply different new ideas.
                                                                                      9



         Chapter 2 - Synthesis of cubic boron nitride films
        2.1 – Introduction of microwave plasma chemical vapor deposition



   Chemical vapor deposition (CVD) is an advance technique widely used to coat


tools, prepare nano-size samples like carbon nano-tubes and cubic boron nitride thin


films. In CVD process, the substrate is exposed to the volatile precursor gases that


react and decompose on its surface to induce the desirable deposit, CVD also supplies


ion bombardment during the growth of cBN. This process always produces


by-products, and they can be removed by applying gas flow through the reaction


chamber. Considering ordinary CVD, the reactions are in ground states. Also, the


reactions are driven by purely thermochemical processes and progress under nearly


thermodynamic equilibrium. For the synthesis of cubic boron nitride films by using


CVD, it allow the cBN thin films growth via chemical route, which is a little bit better


than that of the cBN films producing with physical vapor deposition (PVD) way. It is


because the physical route for the growth of cBN will usually contain relative thick


layers like aBN and tBN, and these two layers are not desirable[26]. However, it is


always use of plasma enhanced chemical vapor deposition to deposit cBN films. For


example, Electron cyclotron resonances microwave plasma chemical vapor deposition


(ECR MPCVD). In the paragraphs that follow, we will introduce more about the ECR


MPCVD and how it can make the cBN thin films.
                                                                                       10



               2.2 – Electron cyclotron resonances microwave plasma CVD




         Figure 2. A diagram of ECR MPCVD system for the deposition of cBN films[31]




   There are several CVD methods being utilized, and, electron cyclotron resonance


microwave plasma chemical vapor deposition (ECR MPCVD), shown as Figure 2, is


mainly used for studying in this project. By applying microwave plasma in CVD, it


can enhance the chemical reactions so that the cBN film can form easily. In addition,


in order to obtain highly activated microwave plasma, electron cyclotron resonance at


relatively low gas pressure (0.1 to 10 Torr) will be utilized.



   In fact, the gas inlets including He-Ar-N2-BF3-H2 are necessary for the deposition


of cBN films. The BF3 acted as a boron source, which decomposes into BFx clusters


and fluorine atom/ions. Fluorine is chosen to be used because it was discovered to be
                                                                                   11



workable precursor for cBN deposition, and it can improve the quality of cBN films.


The ratio of each gas is important to consider, for instance, Dr. Zhang and Matsumoto


in an early experiment discovered that the hydrogen-to-fluorine ratio can control the


deposition rate and phase purity of the cBN films. Therefore, the hydrogen flow rate


was comparable to BF3 one, if there is very high hydrogen flow rate, the deposition of


hexagonal boron nitride (hBN) will be form [27]. It is noted that the substrate


temperature plays an essential role in the CVD methods; for example, a high critical


substrate temperature (above 700oC) is necessary to grow a pure and well-adherent


cBN film by using fluorine assisted CVD.



                              2.3 – Substrate selection


      A number of materials can be chosen for the deposition of cBN films, such as


silicon substrate, diamond substrate, wurtzite aluminum nitride substrate, and WC


substrate. Actually, there are three main factors suggested to select the substrate


materials for the deposition of cBN films, which include: (i) the compatibility of


surface energy, (ii) the lattice match, and (iii) structural stability when ion


bombardment applied in the early stages of the deposition processing. Silicon


substrate is focused here because we will mainly use silicon substrate to operate our


experiment. Also, some information about the selection of diamond substrate will also


be studied.
                                                                                             12



I) Silicon substrates




       Figure 3. HRTEM image of the interface region between cBN and silicon substrate[33]


   Two hindering factors appear when using silicon substrates for deposition of cBN


films. First, the surface energy of silicon is different from the cBN, which is much


lower. Second, the lattice of cBN and silicon do not match (acBN = 3.62 Å while asi =


5.43 Å). Therefore, ion bombardment is needed to solve the problem of different of


surface energy and enhance the nucleation of cBN on the silicon substrate. The impact


energy for starting cBN nucleation is found to be about 100eV, which is higher than


diamond nucleate on silicon. Also, lattice mismatch can be alleviated by 3:2 registries


for cBN and silicon lattices, and it is also quite similar to the situation of diamond


nucleate on silicon [28, 29]. Besides, as shown in Figure 3, cBN grows through the


formation of aBN and tBN interlayer from the silicon substrate, these formations are


caused by the ion bombardment and depend on the particle kinetic energy and


substrate temperature. Yet, the formation of soft aBN and tBN interlayer lead to the


cBN difficult to epitaxial grows on the silicon substrate.
                                                                                        13



II) Diamond substrates




             Figure 4. HRTEM image for cBN grown on CVD diamond substrate [30]


   Although there are several disadvantages of using silicon to deposit the cBN films,


it is quite a excellent material due to its low cost and convenience. Therefore, it was


enlightened that use the diamond powders to grind the silicon samples surface, which


can obtain the diamond substrates. Diamond is the best substrates for the growth of


cBN because of its (i)crystal structure, (ii)lattice matching and (iii)the suitable surface


energy. By using HRTEM, it was found that the phase purity of the films is better and


no aBN and/or tBN were gained at the interface, as seen in figure 4. In addition, the


direct contacts of cBN nanoparticles (111) on Si are easier to obtain. For the


deposition of cBN films on these diamond substrates, it can be up to 2μm.




                    2.5 – Other methods for synthesis of cBN films


       Apart from CVD, there are a number of methods for synthesis of cBN, for


example, physical vapor deposition (PVD). By using usual PVD techniques, the ion
                                                                                     14



bombardment was known with the energy range from 60 to 1k eV; and it is essential


for the formation of cBN. The minimum energy 60 eV mentioned is the threshold for


cBN nucleation, and the threshold for cBN growth is 40 eV. It is noted that the lower


the ion bombardment energy, the better is the quality of the cBN films due to the


reduction of film stress and higher crystallinity. If the ion bombardment energy is


higher than 1k eV, strong re-sputtering will be induced, which cannot deposit the cBN


film. Also, by using PVD for cBN deposition on the foreign substrates excluding


diamond always show a layered structure like substrate/aBN/tBN/cBN.




       High-Pressure High-Temperature method (HPHT), is also a popular

technique used to commercially synthesize cBN. Just as the name implies, this


method forms the cBN by applying very high pressure (in GPa unit) and temperature


(in kT unit) in a certain short time. As a result, it can produce large amounts of cBN


powders that are in the size range from submicron to millimeter, which is obviously


thicker than the cBN produced by other methods; also, the crystallinity and quality are


better. Yet, this technique is extremely costly, so, experts nowadays want to find a way


for synthesis of cBN that is relatively low cost and convenient.
                                                                                          15



    Chapter 3 – Techniques for characterization of cBN films

                                 3.1 – Raman spectroscopy




           Figure 5. (a) and (b) show the Raman spectra of hBN and cBN crystallites[32]


   The different structures of BN phase have different vibration characteristics, and,


this can be used to identify the BN phase due to these different vibration


characteristics. Normally, Raman spectral analysis can be used to detect the


frequencies of vibration of BN crystals. For example, in figure 5(a), we can see there


is one Raman active vibration peak at about 1370 cm-1. In fact, It should have two


peaks, at 55 and 1370 cm-1. However, the cut -off of the Raman-shift wave-number


that used in the previous studies are 700cm-1, thus it cannot see the 55 cm-1 peak in the


figure [33]. On the other hand, figure 5(d), shows that the Raman spectrum of


single-crystal cBN is separated into TO and LO vibration modes at about 1056 and


1305 cm-1 respectively. However, the LO phonons for both hBN and cBN are not
                                                                                      16



detected for normal incident angle IR light and they are often resolved by changing


the incident angle of IR irradiation [34, 35].


      Actually, the Raman signals of cBN are quite dependent on the grain size and


defect density [36]. For instance, if the defect states increase and /or the crystal size


decrease, it will lead to the Raman peaks become asymmetrically broadened and


downshifted to lower wave-numbers, whereas the compressive stress in the films


causes a blue shift of the peaks. Besides that, the Raman signals may only appear on


high quality films. They may not appear if the films are highly defective and


nanocrystalline. In addition, the full width at half maximum (FWHM) and position


shift of these peaks can be used to determine the residual stress and crystallinity of


cBN films, i.e.: the smaller the FWHM imply the higher the crystallinity of the films


[36, 37].

                   3.2 – Fourier transform infrared spectroscopy


      Besides the Raman spectroscopy, Fourier transform infrared spectroscopy


(FTIR) is also a important for characterizing BN films. In fact, FTIR is the more


common method used to characterize the BN films because of its high sensitivity


compared to Raman spectral analysis. FTIR can be utilized to evaluate the


crystallinity and measure relative amounts of cBN in BN films. The above two figures


show the transmission infrared spectra of hBN and cBN films. In figure 6(a), there are
                                                                                         17



two peaks at 780 and 1380 cm-1. These two peaks in the FTIR spectrum belong to


hBN and correspond to out-of-plane bending and in-plane stretching. On the other


hand, from figure 6(b), it shows the FTIR spectrum of a cBN thin film, and there is


one peak located at 1075cm-1.




 Figure 6. (a) and (b) show the transmission infrared spectra of hBN and cBN films[32]


 To measure the amount of cBN content, there is an equation assumed that the cBN
 volume fraction is:

  cBN(%cBN) = IcBN/( IcBN + IhBN)----------------------------(2)


, where IcBN and IhBN are the normalized reflected or transmitted intensities of the IR


absorbencies (cBNTO and hBNstretching modes), cBN (%cBN) is the volume fraction of


the deposited films. However, due to the different absorptions of highly crystalline


cBN and hBN, this estimation is susceptible to errors. For example, the quantitative


analysis of the FTIR spectra of thick cBN films, which is over 700nm, is very difficult


because of the following reasons: (i) the cBN peak broadens when the thickness of the
                                                                                                  18



film increase, and the transmittance may drop to zero, yet, peaks also broaden when


cBN films are very thick; (ii) with the optical interference, the spectra produce a good


deal of non-linear backgrounds which will make it difficult to measure of peak


intensities [38]; (iii) defective materials may also broaden the IR peak because of the


overall signal results from a broad range of vibration states. Therefore, the


information given by FTIR spectra cannot be reliable if the films are too thick, and,


the alternative method is to use Raman spectroscopy.



                             3.3 – Scanning electron microscopy




Figure 7 (a) cross section view of cBN film deposited by DC jet plasma CVD, (b) plan view of cBN film
deposited by a two-step process and by performing the fluorine chemistry[39]



    Scanning electron microscopy (SEM) is also a useful technique that can determine


the thickness, surface topography, orientation and phase of Boron nitride films. The


result of SEM can be straightforward that we can directly identify a lot of information


immediately by seeing the plan view and cross section view of the samples, such as
                                                                                  19



figure 7 (a) and (b). We can obtain from Figure 7a that the largest thickness of the


cBN film deposited by DC jet plasma CVD can be about 200 μm [40-42]. Also, the


surface topography of deposited cBN film is shown in figure 7b, it can seen that the


surface is bumpy, so, previous reports point that the cBN film deposited present a


columnar structure and each column in the film was seen as a cBN single crystal [43].


Yet, there is one main drawback of this method because the use of ion-beam in the


SEM will induce the damage of the samples.

     3.4 – Other common techniques that can apply to investigate the cBN films


   Transmission electron microscopy (TEM) is helpful for identifying phase and

providing detailed microstructural information about the BN films. For instance,


Dark-field TEM method, it is useful for measuring the morphology, grain size and the


phase distribution of cBN films. The segments of the tBN (0002) and cBN (111) ring


are commonly used in the BN system [10]. In addition, High-resolution TEM has also


been critical to investigate the microstructure of BN films.


   Moreover, applying electron energy loss spectroscopy (EELS) in the TEM can

supply very useful data regarding the bonding information of the BN polymorphs. In


addition, EELC in reflection mode can also be utilized to check the near surface


regions of cBN coating films.


     Also, Auger electron spectroscopy (AES) and X-ray photoelectron
                                                                                       20



spectroscopy (XPS) can be utilized to obtain a chemical analysis of the near surface

of BN films. Yet, there are some disadvantages of using these techniques, for example,


the different compositions between the near surface and the bulk [43], the applied ion


sputtering may alter the composition, or even the electron-beam currents may damage


the boron nitride. Therefore, the uses of electron beam currents in these tests are


usually very low [44].


     Wavelength-dispersive X-ray spectroscopy (WDS) applied with SEM can

used to determine the stoichiometry of the BN films. But it also has some


disadvantages like damage of the films and shifting peaks due to matrix effects [45].


In addition, Rutherford backscattering spectroscopy (RBS) or elastic recoil


detection (ERD) and neutron depth profiling (NDP) have also supplied the

compositional information of cBN films, and the film thickness of the specimens can


be obtained more accurately than all the techniques that mentioned above, however,


the equipment is very expensive, and there are complicated procedures needed, and


the result are quite dependent on the materials of the substrates for RBS [10].


     Moreover, Ellipsometry is helpfully used to determine the frequency-dependent

real imaginary parts of the index of refraction or the dielectric constant. It is revealed


that the real refractive index of cBN at visible frequencies is about 2.1, and the


refractive index of cBN coating films are found to be about 1.9 to 2.1 [1,10].
                                                                                    21



Chapter 4 – Experimental procedures for synthesis and

characterization of cBN films

                             4.1 – Synthesis of cBN films



4.1.1 Preparation of sample substrate for ECR MPCVD.


   A silicon seed with single-crystal structure was dipped into molten mass of high


purity polycrystalline, a single-crystal ingot was formed. The ingot was sawed into


thin wafers from 0.2 to 0.4mm thickness. The orientation arrangement of these silicon


wafers were indicated by different shapes, and, the desirable orientation was selected.


Each of these wafers was divided into four tablets. Diamond powders were used to


polish the surface of the tablets by around 20 to 30 minutes. After polishing, the


tablets were seen as substrates. The substrates were held by sample holder and placed


inside Electron cyclotron resonance microwave plasma chemical vapor deposition


(ECR MPCVD).

2.4.2 – The operation of Electron cyclotron resonance microwave plasma chemical


vapor deposition (ECR MPCVD)


     ECR MPCVD was used to prepare the cBN thin films. The gas mixtures, which


were He-Ar-N2-BF3-H2 and He-Ar-N2-BCl3-H2, were used to deposit the cBN films


respectively in order to investigate the differences. The plasma was produced by an
                                                                                       22



ASTeX 1.5-kW microwave system, and the magnetic field of about 875 G was used to


the central area of the reaction chamber. The silicon substrate was fixed on a tantalum


plate that is DC-biased and placed on a hexagonal BN holder. A cleaning step was


applied and it was followed by producing H2 into the chamber for nucleation and


growth of BN films. The chamber was evacuated to a pressure of about 10-6 Torr and


the sample was then cleaned by using He/Ar/N2 and positive biases to etch off the


impurities and oxides on the sample surface [31].

       Table 2, Experiment parameters of BN films produced by ECR MPCVD
         Parameters                         Unit                            Range

             H2                             sccm                           2.0 – 6.0

             N2                             sccm                              50
             Ar                             sccm                              10
             He                             sccm                             131
         BF3/ BCl3                          sccm                         1 to 10 / 10
                                             o
   Substrate temperature                      C                           330 - 900
          Duration                          hour                             1-8
        Bias voltage                          V                            0 to -35
          Pressure                          mTorr                       1.489 – 1.635
         MW power                             W                          1395 - 1500
      Sample Position                        mm                           -25 to +33



   The gas ratio, pressure, substrate temperature, substrate bias and time used in the


experiment are stated in Table 2. All of these parameters were changed over wide


range in order to study their effect on the quality of the BN films, such as purity,


crystallinity and thickness. For each experiment, each parameter was changed
                                                                                    23



independently so that it can satisfy the factorial experiment method. After each


experiment, a cleaning step, like scrubbing, was taken place so that the formation of


hBN cannot block the microwave next time.

                         4.2 – Characterization of cBN films


4.2.1 Raman spectroscopy


      The model of the Raman spectrometer used was RENISHAW in-Va Raman


microscope. Samples which contain cBN content like cw946, cw998 and yh003


(which are deposited under different conditions) were tested. In this spectroscopy,


514nm laser wavelength was used to excite the molecular vibration of the samples.


The samples were placed onto the XYZ displacement and the monochrome light was


applied. The suitable image of the sample was adjusted by using coarse and fine


adjustment knobs. The laser power was set to be 10 units, and then the test was started


after all the adjustments mentioned above.


4.2.2 – Fourier transform infrared spectroscopy


      The holder with mid-size-hole was selected to hold the deposited films. Then,


the holder was placed inside the Perkin Elmer PC 16 (FTIR). The related computer


program was opened and the scan unit was changed to %T. The scan number was set


to be 30 times. In this experiment, both reflective and transmission mode were used to
                                                                                    24



analyze the samples, but transmission was the most used because it is easier to operate


and more commonly used to test BN films. The background signal was checked first


before testing each film and all special regions of the films were scanned. The


crystallinity and relative amounts of cBN in the films were evaluated.


4.2.3 – Scanning electron microscopy


      The    model     of   the    Scanning     Electron    microscope     used    was


PHILIPS-XL30-FEG.The samples were cut into 5mm x 5mm size and then mounted


firmly onto a clean holder (no water and solvents). The position of the each specimen


was marked. Then, the samples was coated with a very thin gold film in order to make


it conductive. The holder was put inside the SEM; it was adjusted to high vacuum


state. The electron beam position and focusing were controlled, and, the observation


positions of the samples were selected by using computer software. Also, the


magnification range was from 5000X to 40000X, which depends on the samples. The


significant image was taken by the computer software.
                                                                                     25



          Chapter 5 – Experimental results and Discussion


      In this chapter, we will show and analyze the BN films made under different


conditions, such as different substrate materials, substrate biases, gas components, gas


ratio, duration, and deposited position.




                      Figure 8 – Typical FTIR background signal of the substrates


    However, what we want to present first is the background signal of FTIR, as


shown in Figure 8. Figure 8 shows the typical background signal of all substrates


tested in this project, which implies all the background signals of the substrates are


under the similar condition. The checking of the background signal must be


performed because we should ensure that the films do not contain other materials,


which may have an effect on the result. From Figure 8, there are mainly three regions,


which represent different substances of the substrate. For region A, it is virtually


showing some hydrogen bonding information. For region B, it stands for the moisture
                                                                                              26



or hydrosphere existed on the films. For region C, it represents the carbon dioxides of


the BN films. Yet, the background signal of the films does not have significant


influence on the test results of the films because the magnitude of Transmittance (a.u)


is not large. In the analysis of the test results, the background signals can actually be


ignored.



               5.1.1 – Synthesis of cBN films with different bias voltages




              Figure 9 – SEM cross-section, of cBN film deposited by (a) -25V and (b) -20V.
                                                                                                                                     27




                        35                                                                    160
                        30                                                                    140
                                                                       Transmittance (a.u.)
  Transmittance (a.u)




                        25                                                                    120
                                                                                              100
                        20
                                                                                              80
                        15
                                                                                              60
                        10                                                                    40
                         5                                                                    20
                         0                                                                      0
                             0        1000      2000       3000                                     0     1000   2000         3000
                                 Wavenumber (cm-1)          (e)                                         Wavenumber   (cm-1)      (f)

Figure 9 – (c) plan view image and (e) IR spectrum of cBN film deposited by -25V.,(d) plan view
                                       image and (f) IR spectrum of cBN film deposited by -20V.
                                                                                    28



         The differences between the cBN films deposited by -20 V and -25V bias


voltages were studied. By using SEM, it was found that a new film was formed onto


the both substrates and we assume that this is cBN film. Besides, we can see that the


growth direction of cBN is almost parallel to the direction of nano-diamond substrate.


The thicknesses of cBN layer of these two films are quite close, which are about 1 μm,


as shown in Figure 9(a) and 9(b). In fact, the deposition rate can be determined.


Because these two samples were deposited in about 270 minutes, so the deposition


rates of these two films are:




       However, the qualities between the films are different. From Figure 9(c) and


9(d), the surface topography of the film prepared by -25V bias voltage is rougher than


that of -20V, which show the cBN growth steadily at lower bias voltage. And, there


are some crevasses that appear in Figure 9(c), these crevasses may induced by the


film stress because of the higher energy ion bombardment. In addition, there is some


white-light presents in Figure 9(c) and (d) which is caused by the sharp edge of the


film surface. Actually, FTIR spectrums shown as Figure 9(e) and 9(f) support the


above presentation. Figure 9(e) and 9(f) show that there are strong cBN absorption


peak appeared in both specimens, which are around 1056 cm-1.By using equation (2),


it shows that the cBN content in sample (a) is little bit lower that sample (b), which
                                                                                                29



are 58.9% and 63.7% respectively. Furthermore, the cBN absorption peak of the


sample that deposited with higher voltage was up-shift to 1082 cm-1, which implies


there is film stress within the film. Hence, increasing the bias voltage will have a


negative effect on the film quality. Yet, it is impossible to set the voltage to zero


because it will not attract the ion to react the film. Therefore, it is best to optimize


other parameters so that we can decrease the bias voltage to as low as possible.


                 5.1.2 – Synthesis of cBN films with different substrate




            Figure 10 – SEM plan-view image of BN film deposited on (a) Si and (b) poly-Diamond/Si
                                                  substrate.
                                                                                                30




               Figure 10 –(c) XPS spectrum of BN film deposited on poly-Diamond/Si substrate



      The   differences    between     the    BN    films    deposited     on    Silicon       and


poly-diamond/Silicon substrates are shown as Figure 10. In Figure 10 (a), it was


found that there are a number of holes on the film, possibly caused by the handling


problems of deposition processing, or the film is actually very thin. Also, the grain


sizes shown in Figure 10(a) are very uneven, so we assume that the film may not have


much cBN content. Actually, just as mentioned in chapter 2, synthesis of cBN films


by using Silicon substrate can have several problems, such as the inevitable formation


of aBN and tBN. It is believed that the film shown in Figure 10(a) may have aBN or


tBN content. However, from Figure 10(b), it can see that the grain sizes of this film is


more even, and, early studies point out that cBN is more easy to deposit on diamond
                                                                                      31



substrate, thus it is assumed that the film shown in Figure 10(b) may be the cBN film,


but there are some fovea-areas shown in the picture, which might also caused by


handling problems (such as parameters control and sample preparation) or formation


of thin film. Therefore, by these observations, it can be seen that the quality of cBN


films deposited onto diamond substrates are better than that onto silicon substrates.


Actually, in order to do the chemical analysis of the film deposited on


poly-Diamond/Si substrate, XPS was used, as shown Figure 10(c). This figure shows


there are mainly four elements, which are Boron, Nitrogen, Carbon, and Oxygen. The


peak of Boron is three times higher than that of Nitride, which is cause by the


absorption sensitivity differences of XPS, in fact, the content of B and N in the film is


virtually 1:1. The present of carbon is due to the existence of diamond substrate.


     Moreover, we also studied the differences between the cBN film deposited onto


B-doped poly-diamond and B-doped nano-diamond substrate. The B-doped substrates


were become p-type materials, which can enhance the conductivities. With the better


conductivity, the substrate bias can be applied uniformly on the substrates during the


deposition processing, and make a better result.
                                                                                                32



                       250

                       200
         Reflectance
                       150

                       100                                                            (a)
                                                                                      (b)
                        50

                         0
                             0   500     1000     1500       2000       2500

                                        Wavenumber (cm-1)


Figure 11 –IR spectra of cBN deposited by B-doped (a) poly-diamond and (b) nano-diamond substrate.




       The spectra in Figure 11 were obtained by FTIR of Reflective mode. From


Figure 11, it shows that the feature of the IR spectra is similar. To measure their cBN


content, the equation (2) mentioned in chapter 3 was used:


For sample A:                    cBN (% cBN) =                  = 74.86%

For sample B:                    cBN (% cBN) =                  = 63.58%

       This calculation indicate that the cBN content of the film deposited on B-doped


poly-diamond substrate is higher than that of B-doped nano-diamond substrate. In


addition, the peak of sample A is broader that sample B, which may imply the


thickness of the sample A is thicker [46]. Therefore, it can assume that the cBN films


deposit on poly-diamond substrate may better that nano-diamond substrate. However,


SEM should be used to carefully confirm the prediction stated above.
                                                                                             33



                5.1.3 – Synthesis of cBN films with different duration

                               800



                 Reflectance
                               600
                               400
                               200
                                 0
                                     0    1000       2000       3000   4000     5000

                                            Wavenumber (cm-1)                      (a)

                               250
                               200
                               150
                               100
                                50
                                 0
                                     0    1000       2000   3000       4000     5000

                                            Wavenumber(cm-1)                       (b)

            Figure 12 – The FTIR spectrums of cBN films made with (a)1 hr 9 and (b)1.5hr 9




       The spectrums in Figure 12 were obtained by FTIR of Reflective mode. Sample


A was deposited one hour, while sample B was deposited one-and-half hour. However,


these 2 graphs already show the significant differences. For instance, it was


discovered that the Reflectance height of cBN absorption peak (at around 1075cm-1)


of sample A is about two times higher than sample B, which are 522.3 and 230.5


respectively. Also, there is an unknown-peak shown in Figure 12(a) at around 868


cm-1. To find out their cBN content, the equation (2) mentioned before was used:


For sample A:                        cBN (% cBN) =                            = 35.60%

(The value of the unknown-peak should be added)
                                                                                        34



For sample B:           cBN (% cBN) =                      = 74.86%

     It shows that the content of cBN of sample A is lower than sample B. This result


shows obviously that higher the duration can enhance the cBN content of the films. It


is mainly because of the cancellation of the unknown-material in Figure 12(a). In


addition, the thickness difference can also be predicted by seeing the width of the


cBN absorption peak of the IR spectra. From Figure 12, the width of the peak in


sample A is broader than that of sample B, and, it is known that the cBN absorption


peak will be broaden with increasing thickness [46]. Therefore, it can know that the


cBN thickness of the sample A is thicker than the sample B. In fact, this result may


imply that the increasing of the deposition time can allow the more cBN growth onto


the substrate. Surely, it has a “critical-duration” that cannot further avail the growth of


cBN films, for example, the deposition rate may gradually decrease. Also improperly


increasing the deposition time may also increase the impurity of the films. If there are


any defective materials in the films, the peak can also be broadened. Moreover, we


discovered that the cBN absorption peak in Figure 12(b) was up-shifted from 1052


cm-1 to 1136 cm-1 comparing with Figure 12(a). The shift of cBN peak means that the


film stress or some defective materials may be induced during the process. Therefore,


higher duration may also produce negative effect on the cBN films.
                                                                                                                           35


                                                   5.1.4 – Synthesis of cBN films with different gas ratio

                                                   160
                                                   140


                            Transmittance (a.u.)
                                                   120
                                                   100
                                                    80
                                                    60
                                                    40
                                                    20
                                                     0
                                                         0           500        1000       1500   2000   2500
                                                                       Wavenumber (cm-1)                       (a)


                                                   140
                           Transmittance (a.u.)




                                                   120
                                                   100
                                                    80
                                                    60
                                                    40
                                                    20
                                                     0
                                                         0           500       1000      1500     2000    2500
                                                                           Wavenumber (cm-1)                 (b)


 Figure 13 – FTIR spectra of cBN films deposited by gas ratio (a) H2/BF3 = 1:5 and (b) H2/BF3 = 1:4


                       60000


                       50000


                       40000
    Intensity (a.u.)




                       30000
                                                                                                                     (a)
                                                                                                                     (b)
                       20000


                       10000


                           0
                                          0              200   400     600    800   1000 1200 1400 1600 1800
                                                                           Raman shift (cm-1)

Figure 14 – Raman spectra of cBN films deposited by gas ratio (a) H2/BF3 = 1:5 and (b) H2/BF3 = 1:4
                                                                                    36



      Synthesis of cBN films with different gas ratio was studied in this project.


However, we only changed the gas ratio between H2 and BF3, because these two gases


are of great significance that react together during the deposition processing and


provide Boron source. Ar, He, and N2 were kept constant in this experiment, which are


10 sccm, 131 sccm and 50 sccm respectively. The IR spectra of cBN films deposited


by H2/BF3 = 1 : 5 and H2/BF3 = 1 : 4 are plotted in Figure 13. It shows some


differences between two specimens, for example, the cBN absorption peaks in 13(b)


are weaker than that of 13(a), which implies the cBN content in sample B relative low.


To determine their cBN content, the equation (2) in chapter 3 was used:


For sample A:          cBN (% cBN) =                   = 68.62%

For sample B:          cBN (% cBN) =                   = 62.84%

      In spite of the fact that the cBN content of sample A is higher than sample B, it


is important to compare the film quality and thickness between them. From Figure 13,


the cBN absorption peak of sample B is broader that sample A; this observation may


indicate either the presence of thicker cBN film or more defective materials. In order


to find out the true reason, Raman spectroscopy was used and the Raman spectrum is


shown as Figure 14. From Figure 14, there is a diamond peak shown at 1330 cm-1


because the substrates were prepared by poly-Diamond powers. In fact, there should


be a cBNLO peak at around 1305 cm-1, yet, it may be covered by the strong diamond
                                                                                     37



peak. However, the cBNTO peak is presented in Figure 14, but the peak is quite weak


in both sample A and B, which implies that the crystallites of these two films are very


small that close to nano-size as well as the thickness. Nevertheless, the cBNTO peak in


sample A is more obvious that sample B, which shows that the film of sample A has


fewer defects, larger crystallites. Therefore, it was shown that higher H2/BF3 ratio not


only decrease the cBN content of the films but also have some negative effect on the


film quality. This observation agrees with the review stated in chapter 2, which point


out that higher hydrogen flow rate may cause the deposition of hexagonal borno


nitride (hBN), as a result, the cBN content of the film will be reduced and affect the


film quality.
                                                                                                                             38



                                 5.1.5 – Synthesis of cBN films by utilizing BCl3 instead of BF3




   Figure 15 – SEM Plan view (a) 20000X and (b) 5000X image of BN films deposited by using BCl3



                           250                                                         200
    Transmittance (a.u.)




                                                                 Transmittance (a.u)




                           200
                                                                                       150
                           150
                                                                                       100
                           100
                                                                                       50
                            50

                             0                                                           0
                                 0     1000   2000     3000                                  0     1000    2000      3000
                                     Wavenumber (cm-1)    (a)                                    Wavenumber (cm-1)     (b)


Figure 16 – IR spectra of BN films deposited by using BCl3 and different hydrogen ratio.(a) H2:BCl3 = 1:5,
(b) H2:BCl3 = 1.5:10
                                                                                    39



      Since the use of fluorine is very dangerous in the industries, BCl3 is more


commonly used because Cl is the same group of F and Cl is relatively safe. Yet, in this


experiment, no cBN film was successfully formed onto the samples. From 16(a) and


(b), the IR spectra showed that only hBN was formed because there are two hBN


absorption peaks appeared around 780 cm-1 and 1378 cm-1 at the spectrums. Although


we tried to change the gas ratio (H2/ BCl), no any cBN was formed. In addition, in


order to study the hBN film, SEM plan view image was taken, as shown in Figure 15.


Figure 15 shows the surface topography of the film. The irregular ball-like grains may


be the hBN, and its average size is about 2 μm. Also, there are some positions that are


not covered by the hBN film, as shown Figure 15(b). The failure of the formation of


cBN film may have several reasons, which including (i) the use of Cl may not be a


workable precursor for cBN deposition, (ii) the discord of other parameters may affect


the film formation and quality, and (iii) improper substrate may be used. However,


there is early report shown by applying low density supersonic plasma flows, the cBN


can be synthesis through BCl3[47]. Therefore, the synthesis of cBN films by utilizing


BCl3 instead of BF3 is not impossible. Surely, further study should be needed so that


the best condition for using BCl3 can be found out.
                                                                                   40



                            Chapter 6 –Conclusion


      To conclude, the cBN films made by ECR MPCVD in this experiment were


tested and analyzed by using Raman spectroscopy, Fourier transform infrared


spectroscopy, X-ray photoelectron spectroscopy and Scanning electron spectroscopy.


The cBN films deposited under different conditions or parameters were observed. To


summarize the results, it is suggested that decrease the substrate bias to as low as


possible and properly increase the deposition duration can form a relative good cBN


film, which has less residual film stress and become thicker. Also, increasing of


H2/BF3 gas ratio reduce the cBN content of the films due to the formation of


hexagonal boron nitride (hBN), and also lower the film qualities. Furthermore, it was


proved that the quality of cBN films deposited onto diamond substrates are better than


that onto silicon substrates. Furthermore, the substrate of B-dope poly-diamond is a


relatively good choice for depositing cBN films compared with B-dope nano-diamond


substrate. In addition, it was found that the BN films prepared by using BCl3 instead


of BF3 in this experiment are mainly the formation of hBN with no any cBN content,


thus, more studies and experiments will be needed to solve the difficulties.
                                                                                         41



References:
[1] L.Vel.G. Demazeau, J. Etoumeau, Mater.Sci. Engin. B 10 (1991) 149
[2] Solozhenko V L, Hausermann D, Mezouar M and Kunz M 1998 Appl. Phys. Lett. 72
    1691
[3] Yoshida T 1996 Diamond Relat. Mater. 5 501
[4] Miyata N, Moriki K, Mishima O, Fujisawa M and Hattori T 1989 Phys. Rev. B 40
    12028
[5] Davies G, 1988 in Diamond Optics ed A Feldman and S Holly (New Mexico:
    SPIE)
[6] Mohammad S N 2002 Solid State Electron. 46 203
[7] Mishima O, Tanaka J, Yamaoka S and Fukunaga O 1987 Science 238 181
[8] Kalish R 1999 Carbon 37 781
[9] T. Taniguchi, K. Watanabe, S. Koizumi, Phys. Stat. Sol. A 201 (2004) 2573.
[10] Lee S T and Lifshitz Y 2003 Nature 424 500
[11] W.J. Zhang, Y.M. Chong, I. Bello, S.T. Lee, J. Phys. D-Appl. Phys. 40 (2007) 6159.
[12] S. Matsumoto, W.J. Zhang, New Diam. Front. Carbon Technol. 11 (2001) 1.
[13] C.Ziebert, J.Ye, K. Sell, S.Ulrich Forschungszentrum Karlsruhe, Institut für
     Materialforschung I, Hermann-von-Helmholtz-Platz 1, 76344Eggenstein
     Leopoldshafen, Germany.
[14] DeVries R C 1972 Cubic boron nitride: handbook of properties Report No.
     72CRD178 (Schenectady, NY:General Electric Company)
[15] Feldermann H, Ronning C, Hof¨ass H, Huang Y L and Seibt M 2001 J. Appl. Phys.
     90 3248
[16] Pechentkovskaya L E and Nazarchuk T N 1970 Sov. PowderMetall. 6 83
[17] Taniguchi T, Teraji T, Koizumi S, Watanabe K and Yamaoka S 2002 Japan. J.
     Appl. Phys. Part 2—Lett. 41 L109
[18] Phani A R, Manorama S and Rao V J 1995 Semicond. Sci. Technol. 10 1520
[19] Lu M, Bousetta A, Bensaoula A, Waters K and Schultz J A 1996 Appl. Phys. Lett
     68 622
[20] Nose K, Tachibana K and Yoshida T 2003 Appl. Phys. Lett. 83 943
[21] W J Zhang, Y M Chong, I Bello, S T Lee, Nucleation, groeth and characterization
     of cubic boron nitride (cBN) films
[22] Taniguchi T, Tanaka J, Mishima O, Ohsawa T and Yamaoka S 1993 Appl. Phys.
     Lett. 62 576
[23] S. Ulrich, J. Scherer, I. Schwan, I. Barzen, K. Jung, H. Ehrhardt, Diamond Relat.
     Mater. 4 (1995) 288.
[24] S. Reinke, M. Kuhr, W. Kulisch, Diamond Relat. Mater. 5 (1996) 508.
                                                                                    42


[25] 222 W. Kulisch, R. Freudenstein / Thin Solid Films 516 (2007) 216–222
[26] Hofs¨ass H, Eyhusen S and Ronning C 2004 Diamond Relat. Mater. 13 1103
[27] ZhangWJ and Matsumoto S 2000 Chem. Phys. Lett. 330 243
[28] Jiang X and Jia C L 2000 Phys. Rev. Lett. 84 3658
[29] Ishibashi K and Furukawa S 1985 Japan. J. Appl. Phys. 24 912
[30] X. W. Zhang, H.G. Boyen, P. Ziemann, M. Ozawa, F. Banhart and M. Schreck,
      Growth mechanism for epitaxial cubic boron nitride films on diamond
      Substrates by ion beam assisted deposition
[31] Y M Chong, Deposition and characterization of cubic boron nitride diamond
composite films, 43-46
[32] W J Zhang, Y M Chong, I Bello and S T Lee Nucleation, growth and
     Characterization of cubic boron nitride (cBN) films
[33] Nemanich R J, Solin S A and Martin R M 1981 Phys. Rev. B 23 6348
[34] Ulrich S, Theel T, Schwan J and Ehrhardt H 1997 Surf. Coat. Technol. 97 45
[35] Scheible P and Lunk A 2000 Thin Solid Films 364 40
[36] Zhang W J and Matsumoto S 2001 Phys. Rev. B 63 073201
[37] Zhang W J and Matsumoto S 2001 J. Mater. Res. 16 3430
[38] Turrel G 1972 Infrared and Raman Spectra of Crystrals (London: Academic)
[39] Zhang W J, Matsumoto S, Li Q, Bello I and Lee S T 2002 Adv. Funct. Mater. 12250
[40] Zhang W J and Matsumoto S 2000 Appl. Phys. A 71 469
[41] Matsumoto S and Zhang W J 2000 Japan. J. Appl. Phys.
[42] Matsumoto S and Zhang W J 2001 Diamond Relat. Mater.10 1868
[43] 0. Burat, D. Bouchier, V. Stambouli. G. Gautherin. J. Appl. Phps. 68 (1990) 2780.
[44] J R. Trehan. Y. Lifshitz, J.W. Rabalais. J. Vat. Sci. Technol. A 8 (1990) 4026.
[45] V.Y. Kulikovsky, L.R. Shaginyan, V.M. Vereschaka, N.G. Hatyenko, Diamond Relat.
     Mater. 4 (1995) 113.
[46] T.A. Friedmann, P.B, Mirkarimi, D.L. Medlin, K.F. McCarty, E.J. Klaus, D. Boehme,
     H.A. Johnsen, M.J. Mills,D.K. Ottrsen, J. Appl. Phys. 76 (1994) 3088.
[47] D. H. Bernsa) and M. A. Cappelli, Cubic boron nitride synthesis in
     low-density supersonic plasma flows

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:3
posted:2/9/2012
language:
pages:53