`(cid:14)
`Surface and Coatings Technology 131 2000 222]227
`
`A study on the high rate deposition of CrN films with
`x
`controlled microstructure by magnetron sputtering
`
`Kyung H. NamU, Min J. Jung, Jeon G. Han
`Plasma Applied Materials Laboratory, Department of Ad¤anced Materials, Sung Kyun Kwan Uni¤ersity, 300 Chunchun-dong, Jangan-ku,
`Suwon 440-746, South Korea
`
`Abstract
`
`High rate deposition of CrN films with control of microstructure was carried out by magnetron sputtering. For these purposes,
`x
`the deposition processes parameters were varied: N flow rate and especially substrate bias voltage, duty cycle and frequency
`2
`(cid:14)
`.
`using a pulsed DC power supply. The microstructure was analyzed by X-ray diffraction XRD and scanning electron microscopy
`(cid:14)
`.
`SEM , and mechanical properties were evaluated by a microhardness test and adhesion test. The maximum deposition rate for
`CrN compound films could be reached to nearly 90% compared with that for pure Cr coating due to the increase of ionization
`x
`efficiency caused by a negative-pulsed DC bias. As N flow rate is increased, the microstructure of CrN films was changed from
`2
`x
`CrqCr N to CrN. Also, a phase transformation occurred between Cr NqCrN multi-phase and CrN mono-phase by control of a
`2
`2
`negative DC andror pulsed DC bias voltage, duty cycle and frequency. Microhardness for CrN films were measured to be up to
`1600 kgrmm2 and the maximum hardness value of 2250 kgrmm2 was obtained for CrN film deposited with a N flow rate of 20
`sccm at a negative DC bias of y100 V. Q 2000 Elsevier Science B.V. All rights reserved.
`
`2
`
`x
`
`x
`
`Keywords: CrN ; Magnetron; Deposition rate; Microstructure; Pulsed DC bias
`x
`
`1. Introduction
`
`High rate deposition processes such as high current
`arc, laser arc, hollow cathode discharge ion plating and
`magnetron sputtering methods have been developed
`w
`x
`for cost effective industrial applications 1]3 . Espe-
`cially magnetron sputtering is emerging as a very effi-
`cient method for the synthesis of high rate deposited
`dense films. In the early years of the 1990s a deposition
`rate of 1 ;3 mmrmin has been reached using an
`unbalanced magnetron, but these have been restricted
`to pure metal films such as Cu, Ag, etc., of high
`sputtering yield. Overcoming a poisoning effect between
`(cid:14)
`.
`(cid:14)
`metallic targets Ti, Cr, etc. and reactive gases N ,2
`.
`O , etc. , high rate deposition of reactively sputtered
`2
`
`U Corresponding author. Tel.: q82-331-290-7381; fax: q82-331-
`290-7386.
`.
`(cid:14)
`E-mail address: khnam@nature.skku.ac.kr K.H. Nam .
`
`nitride and oxide films by reactive magnetron sputter-
`ing have been realized to the deposition rate of 60 ;
`70% compared with that of pure metals by precise
`w
`x
`partial pressure control of the reactive gas 4,5 . In the
`mean time, the correlations between process parame-
`ters, microstructure and film properties have also been
`intensively studied to ensure the reproducibility of films
`w
`x
`with pre-defined properties 6]8 . In case of nitride
`films, particularly, it has been reported that the main
`parameters controlling the film microstructure are de-
`position temperature, N partial pressure and bias volt-
`2
`w
`x
`age 6,7,9 .
`In this study, an unbalanced magnetron sputtering
`was employed to synthesize CrN films for high rate
`x
`deposition with a control of microstructure. Deposition
`processes for such purpose were varied with N flow
`2
`rate and specially substrate bias voltage, duty cycle and
`frequency using pulsed DC power supply. The mi-
`(cid:14)
`.
`crostructure was analyzed by X-ray diffraction XRD
`
`0257-8972r00r$ - see front matter Q 2000 Elsevier Science B.V. All rights reserved.
`(cid:14)
`.
`PII: S 0 2 5 7 - 8 9 7 2 0 0 0 0 8 1 3 - 6
`
`Page 1 of 6
`
`APPLIED MATERIALS EXHIBIT 1031
`
`
`
`K.H. Nam et al. r Surface and Coatings Technology 131 2000 222]227
`)
`(
`
`223
`
`.
`(cid:14)
`and scanning electron microscopy SEM , and mechan-
`ical properties were evaluated by microhardness and
`adhesion tests.
`
`2. Experimental details
`
`2.1. Film deposition
`
`CrN films were deposited on AISI 304 stainless
`x
`steel and Si wafers by magnetron sputtering of a rect-
`angular Cr target with a moving magnet designed for
`high erosion efficiency,
`in our laboratory. The dis-
`charges of this magnetron, which are elliptically shaped
`with the longer axis perpendicular to the longer axis of
`the target, are generated by separated magnetic units
`placed behind the target. Non-sputtered regions inside
`the individual racetracks are eliminated by the simulta-
`neous sweeping of all magnetron discharges along the
`longer axis of the target, which is achieved by moving
`the magnetic means behind the target. All specimens
`were cleaned following conventional cleaning process
`prior to deposition. The deposition process was per-
`(cid:14) .
`(cid:14) .
`formed in the following steps: 1 radiation heating; 2
`DC glow discharge cleaning in an Ar atmosphere for 10
`(cid:14) .
`min; 3 sputter deposition of a 0.2 mm Cr interlayer
`(cid:14) .
`film; 4 deposition of CrN films at various conditions
`x
`listed in Table 1.
`2.2. E¤aluation of films
`
`For the evaluation of phase and texture formation
`for CrN films XRD analyses were performed with an
`x
`incident angle of 38. By using SEM fracture cross-sec-
`
`Table 1
`Conditions for CrN coating process
`x
`
`Deposition parameters
`
`Base pressure
`Ar pressure
`Target power density
`Distance between target and
`substrate
`Temperature
`N flow rate
`2
`.
`(cid:14)
`Substrate bias pulsed DC
`(cid:14)
`.
`Voltage V
`.
`(cid:14)
`Duty cycle %
`.
`(cid:14)
`Frequency kHz
`
`Conditions
`y5
`3 =10
`torr
`y3
`1.8=10
`torr
`13 "1 Wrcm DC
`2 (cid:14)
`.
`80 mm
`
`400"108C
`0 ;45 sccm
`
`y50, y100, y200
`50, 70, 100
`5, 10, 20
`
`tional morphologies were investigated and the deposi-
`tion rate of coated samples was calculated. Micro
`Knoop hardness was measured at a normal
`load of
`0.025 N. The adhesion strength was compared by
`observing the propensity for cracks and the degree of
`delamination near the indentation periphery using an
`optical microscope after Rockwell C indentation test.
`
`3. Results and discussion
`
`3.1. Influence of N flow rate
`2
`
`Fig. 1 shows XRD patterns of CrN films deposited
`x
`on Si wafer with various N flow rates at a negative DC
`2
`bias of y100 V. At a N flow rate of 20 sccm, a mixed
`2
`(cid:14)
`.
`(cid:14)
`.
`(cid:14)
`.
`phase containing Cr 110 , CrN 200 and Cr N 111 was
`2
`observed. As N flow rate is further increased upon
`2
`
`(cid:14) .
`(cid:14) .
`(cid:14) .
`(cid:14) .
`(cid:14) .
`Fig. 1. XRD patterns of CrN films deposited on Si wafer with various N flow rates. a 0 sccm, b 20 sccm, c 30 sccm, d 40 sccm and e 45
`2
`x
`sccm.
`
`Page 2 of 6
`
`
`
`224
`
`K.H. Nam et al. r Surface and Coatings Technology 131 2000 222]227
`)
`(
`
`(cid:14) .
`(cid:14) .
`(cid:14) .
`Fig. 2. Cross-sectional scanning electron micrographs of CrN films deposited on Si wafer with various N flow rates. a 0 sccm, b 20 sccm, c
`2
`x
`(cid:14) .
`(cid:14) .
`30 sccm, d 40 sccm and e 45 sccm.
`
`deposition, CrN films tend to change from the hexago-
`x
`nal Cr N phase to the cubic CrN phase. The CrN film
`2
`x
`deposited with N flow rate of 30 sccm was formed
`2
`mostly with Cr N mono-phase and then transformed to
`2
`
`CrN mono-phase with a further increase of the N flow
`2
`to 45 sccm.
`The SEM micrographs of fractured cross-sections of
`the films are illustrated in Fig. 2. It shows that the
`
`Page 3 of 6
`
`
`
`K.H. Nam et al. r Surface and Coatings Technology 131 2000 222]227
`)
`(
`
`225
`
`in terms of the German short form of adhesion strength
`w
`x
`13,14 .
`
`3.2. Influence of substrate bias
`
`For the understanding of substrate bias effect influ-
`enced CrN film properties CrN films were deposited
`x
`x
`with various substrate bias voltage, duty cycle and
`frequency using pulsed DC power supply at constant
`N flow rate of 40 sccm. Table 2 illustrates sample
`2
`name and summary of the substrate bias effect on the
`deposition rate, microhardness and adhesion strength
`of CrN films. Moreover, the microstructure of each
`x
`coated sample was identified in Fig. 4 by XRD analy-
`ses. The microstructure of CrN film deposited with a
`x
`negative DC bias voltage of y100 V CrN-2 was
`(cid:14)
`.
`defined to be Cr NqCrN multi-phase. However, this
`2
`(cid:14)
`multi-phase was changed to Cr N mono-phase CrN-1,
`2
`.3 when the substrate bias voltage was varied. Also, the
`variation of pulse frequency at a duty cycle of 70% led
`to the phase transformation from Cr N mono-phase
`2
`CrN-5 to Cr NqCrN multi-phase CrN-4, 6 , and
`(cid:14)
`.
`(cid:14)
`.
`2
`(cid:14)
`.
`then Cr N mono-phase CrN-5 was changed to Cr N
`2
`2
`qCrN multi-phase CrN-7 with the decrease of duty
`(cid:14)
`.
`cycle at the same frequency. These phase transforma-
`tions with the change of substrate bias is due to nearly
`equal energy of formation between Cr N y122.88
`(cid:14)
`2
`kJrmol and CrN y123.98 kJrmol at 4008C with
`.
`(cid:14)
`.
`w
`x
`constant N partial pressure 10 . Two different phases,
`2
`CrN and Cr N, which have very closed value of free
`2
`energy of formation have almost same probability to
`nucleate and grow. Thus, these two phases might inde-
`pendently nucleate depending on the adatom energy
`state which is strongly influenced by substrate bias
`when other deposition parameters such as power den-
`sity of target, substrate temperature and N flow rate
`2
`were the same.
`At a negative bias voltage with sufficient duty cycle
`(cid:14)
`.
`and frequency CrN-5, 6 , respectively, the deposition
`rate was increased. It is estimated that the ionization
`efficiency was increased by repetitive impact and stag-
`nation between adatoms caused by a negative pulsed
`w
`x
`DC bias 15 . The maximum deposition rate of 210
`
`Deposition
`rate nmrmin
`.
`(cid:14)
`
`Microhardness
`kgrmm
`2
`(cid:14)
`.
`
`174
`165
`194
`162
`210
`180
`163
`
`1631
`1930
`2099
`2044
`2037
`2063
`1599
`
`Adhesion
`strength
`
`HF3;4
`HF1;2
`HF1;2
`HF2;3
`HF1;2
`HF1;2
`HF3;4
`
`Fig. 3. Microhardness changes of CrN films measured at normal
`x
`load of 0.025 N for various N flow rates.
`2
`
`increasing of N flow rate leads to an increase of film
`2
`density and the decrease of its deposition rate from 236
`to 165nmrmin except for CrN film deposited with a
`x
`N flow rate of 45 sccm. This reduction of deposition
`2
`rate is due to the formation of chromium nitride at the
`w
`x
`target surface, the so called ‘poisoning effect’ 4,5 .
`Fig. 3 illustrates the microhardness of CrN films
`x
`deposited with various N flow rates. The maximum
`2
`hardness value of 2250 kgrmm2 was obtained for CrNx
`film deposited with a N flow rate of 20 sccm while
`2
`further increase of the N flow rate up to 45 sccm
`2
`tended to reduce the hardness
`in the range of
`1800]2000 kgrmm2. These results are somewhat dif-
`ferent from other reports in which the maximum hard-
`w
`x
`ness value was obtained for Cr N mono-phase 10,11 .
`2
`w
`x
`It is estimated that the mixing effect 12 between Cr,
`Cr N and CrN plays a role to improve hardness of
`2
`CrN film deposited with N flow rate of 20 sccm. And
`2
`x
`also the reason why the similar hardness value was
`obtained for Cr N film deposited with N flow rate of
`2
`2
`30 sccm compared with CrN films is predicted that this
`film was consist of not only Cr N phases but also CrN
`2
`phases as shown in Fig. 1.
`After results of Rockwell C indentation adhesion
`tests, all CrN films deposited with various N flow
`2
`x
`rates persisted in fairly good adhesion with a little
`crack and delamination corresponding to HF1;HF3
`
`Table 2
`Sample identification and summary of the substrate bias effect
`
`Sample
`
`CrN-1
`CrN-2
`CrN-3
`CrN-4
`CrN-5
`CrN-6
`CrN-7
`
`Duty
`(cid:14)
`cycle %
`
`.
`
`Frequency
`(cid:14)
`.
`kHz
`
`100
`100
`100
`70
`70
`70
`50
`
`]
`]
`]
`5
`10
`20
`10
`
`Bias
`.
`(cid:14)
`voltage V
`y50
`y100
`y200
`y100
`y100
`y100
`y100
`
`Page 4 of 6
`
`
`
`226
`
`K.H. Nam et al. r Surface and Coatings Technology 131 2000 222]227
`)
`(
`
`(cid:14) .
`(cid:14) .
`Fig. 4. XRD patterns of CrN films deposited on Si wafer at N flow rate of 40 sccm with a various substrate bias voltages and b various
`2
`x
`substrate bias duty cycles and frequencies.
`
`nmrmin was obtained for CrN-5, which is 89% com-
`pared with the deposition rate of pure Cr coating under
`the same conditions except for a negative-pulsed DC
`duty cycle.
`The microhardness of CrN films were measured to
`x
`be similar values independent of the microstructure
`except for CrN-1 and CrN-7 which were deposited with
`low bias voltage or duty cycle. It has been reported
`w
`x
`16,17 that the low bias voltage or duty cycle leads to a
`decrease of microhardness of films due to decreasing
`adatom mobility. After results of adhesion tests by
`Rockwell C indentation, CrN-1 and CrN-7 were proved
`
`to be HF3;4 while other films have a good adhesion
`strength corresponding to HF1;3. The low adhesion
`of CrN-1 and CrN-7 is understood by the decrease of
`ion bombardment caused by low adatom mobility dur-
`ing processes.
`
`4. Summary
`
`The high rate deposition of CrN films was carried
`x
`out by magnetron sputtering with controlled micro-
`
`Page 5 of 6
`
`
`
`K.H. Nam et al. r Surface and Coatings Technology 131 2000 222]227
`)
`(
`
`227
`
`structure. In this study, the following results have been
`obtained:
`
`the present study by the Ministry of Commerce, Indus-
`try and Energy of Korea.
`
`1. The maximum deposition rate for CrN compound
`x
`films was reached to 89% compared with that for
`pure Cr coating due to the increase of ionization
`efficiency caused by a negative pulsed DC bias.
`2. The microstructure of CrN films could be success-
`x
`fully controlled with by variation of the negative
`bias voltage, duty ratio and frequency as well as the
`N flow rate.
`2
`3. Microhardness for CrN films were measured to be
`x
`more than two times of that for Cr coating and the
`maximum hardness value of 2250 kgrmm2 was
`obtained for CrN film deposited with the N flow
`2
`x
`rate of 20 sccm at a negative DC bias of y100 V.
`
`Acknowledgements
`
`The authors are grateful for the financial support of
`
`References
`
`w x
`.
`(cid:14)
`1 B. Schultrich, P. Siemroth, Surf. Coat. Technol. 93 1997 64.
`w x
`P. Siemroth, T. Witke, Surf. Coat. Technol. 68r69 1994 314.
`(cid:14)
`.
`2
`w x
`.
`(cid:14)
`3
`J. Musil, A. Rajsky, J. Vac. Sci. Technol. A 14 1996 2187.
`w x
`4 M.S. Wong, W.J. Chia, Surf. Coat. Technol. 86r87 1996 381.
`(cid:14)
`.
`w x
`P. Yashar, J. Rechner, Surf. Coat. Technol. 94r95 1997 333.
`(cid:14)
`.
`5
`w x
`(cid:14)
`.
`6 C. Meunier, G. Bertrand, Surf. Coat. Technol. 107 1998 149.
`w x
`7 C. Gautier, J. Machet, Surf. Coat. Technol. 86r87 1996 254.
`(cid:14)
`.
`w x
`S.J. Bull, D.S. Rickerby, Surf. Coat. Technol. 43r44 1990 732.
`(cid:14)
`.
`8
`w x
`(cid:14)
`.
`9
`J.P. Terrat, A. Gaucher, Surf. Coat. Technol. 45 1991 59.
`w
`x
`(cid:14)
`.
`10 M. Pakala, R.Y. Lin, Surf. Coat. Technol. 81 1996 233.
`w
`x
`P. Hones, R. Sanjines, Surf. Coat. Technol. 94r95 1997 398.
`.
`(cid:14)
`11
`w
`x
`(cid:14)
`.
`12
`J. Musil, J. Vlcek, Czech. J. Phys. 48 1998 10.
`w
`x
`13 R. Bantle, A. Matthew, Surf. Coat. Technol. 74r75 1995 857.
`.
`(cid:14)
`w
`x
`.
`(cid:14)
`14 W. Heinke, A. Matthew, Thin Solid Films 270 1995 431.
`w
`x
`(cid:14)
`.
`15 K.T. Rie, F. Schnatbaum, Mater. Sci. Eng. A 160 1991 448.
`w
`x
`16 O. Piot, J. Machet, Surf. Coat. Technol. 94r95 1997 409.
`(cid:14)
`.
`w
`x
`17 E. Lugscheider, O. Knotek, Surf. Coat. Technol. 76r77 1995
`.
`(cid:14)
`705.
`
`Page 6 of 6
`
`