Surface and Coatings Technology 138 2001.

14 22

Sputter-deposited nanocrystalline Cr and CrN coatings on steels

J.W. Seok, N.M. Jadeed, R.Y. LinU
Department of Materials Science and Engineering, M.L. n12, Uni ersity of Cincinnati, Cincinnati, OH 45221-0012, USA Received 21 April 2000; received in revised form 25 September 2000; accepted 3 October 2000

Abstract Chromium and chromium-nitride coatings on steels have been deposited with a magnetron sputter-deposition system. The deposition power was 200 W pulsed DC with a frequency of 185 kHz and 96% deposition duty. The substrate temperature was maintained at 200C. The sputtering gas was argon mixed with 0, 3, 5 or 7% nitrogen. Results show that the average deposition rate was 1.2 mrh and was not affected by the nitrogen content in the sputtering gas. Using X-ray diffraction analysis, it was observed that the deposited Cr under pure argon condition, was nanocrystalline bcc chromium with a particle size in the range of 7 8 nm. With an increasing nitrogen content in the sputtering gas the amount of CrN increased. The measured microhardness of the chromium-coated steel increases with the increasing nitrogen content. With less than a 2- m Cr coating, the steel hardness increases from 129 to 255 HV when the nitrogen content in the gas is 7%. The microhardness of CrrCrN-coated steel prepared with sputter deposition is superior to that prepared with electroplating. The hardness of the coating layer calculated from these data is 1270 HV. Scratch tests were used to characterize coating adhesion. The critical load was determined to be the applied load under which an acoustic noise was found and cracks in the scratch track were rst observed. The critical loads of deposited lms with 0, 3, 5 and 7% N2 in the gas, were 1.57, 5.68, 8.33 and 20.29 N, respectively. 2001 Elsevier Science B.V. All rights reserved.
Keywords: Cr and CrN; Hard coatings; Sputter deposition; Pulsed DC magnetron; X-Ray diffraction; Phase stability; Microhardness; Scratch test; Adhesion of coatings

1. Introduction Coatings impart to engineering components properties such as improved wear resistance, good corrosion resistance, low friction coefcients and attractive appearances. Coating with thin chromium layers has been an important engineering practice for years in automotive, aerospace and decorative industries for various reasons. They are used as wear andror corrosion-resistant coatings, coatings for bright and non-tarnishing surfaces, and high temperature coatings.
U

Corresponding author. Tel.: q1-513-556-3116; fax: q1-513-5562569. E-mail address: ray.lin@uc.edu R.Y. Lin..

The conventional way of producing these coatings is by electrodeposition. An estimated billion tons of coatings are produced, and electroplating of chromium is a multi-billion dollar industry nationwide w1,2x. However, during electrodeposition, carcinogenic vapors of hexavalent chromium Cr 6q . are released from the CrO 3 used in the bath. The US environmental protection agency EPA. has recently imposed very stringent requirements on the emission control for such industries. The approximate cost for industry to operate under such enforcement is predicted to be in the region of hundreds of million dollars w2x. In the past decade, alternative methods have been developed to deposit these coatings in an economical yet environmentally safe process. Sputter deposition has essentially no Cr 6q emission since it involves transfer of vapor atoms in-

0257-8972r01r$ - see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S 0 2 5 7 - 8 9 7 2 0 0 . 0 1 1 2 0 - 8

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

15

stead of ions. Hence, sputter deposition is a strong candidate to produce low-cost chromium coatings. However, the hardness of sputter-deposited pure chromium coatings are not as high as electrodeposited coatings, where the hardness is derived from the presence of complex hydrides. This problem can be solved by incorporating nitrogen in the sputtering environment in a process known as reactive sputter deposition w2,3x. Depending on the nitrogen content, the coating consists of either a chromium Cr. phase with nitrogen dissolved in the lattice, or chromium-nitride CrN. phases. Either way, they enhance the hardness of these coatings. During the past decade, investigations on alternative techniques for producing chromium and chromiumnitride coatings and lms have been concentrating on sputter deposition w4 20x while only a few were on arc evaporation or ion-plating w21 25x. The applications include the underlayer for thin magnetic recording media to stimulate epitaxial growth of the cobalt-based magnetic media, diffusion barriers in electronic devices, oxidation resisting coatings, corrosion resisting coatings w15,19x and wear resisting coatings w17,18x. Sputter-deposited chromium or chromium-nitride lms have been characterized for their structure and morphology w12 19x by different workers on various substrates such as glass, silicone, single-crystal rock salt, and aluminum. Although steels with TiN coatings are now commonly available and widely used w13,14,16 20x, some researchers suggest that chromium-nitride coatings can improve the corrosion and wear resistance of steels to a greater extent than TiN. In our previous studies, the Cr lms, which were sputter-deposited at substrate temperatures up to 400C, were found to have strong 1104 preferred orientation w6 8x. The preferred orientation of the deposited lms changes to 1004 with increasing substrate temperatures. Straight-columnar structures with dome tops are found extending outwards from the substrate surface. The width of these columns increases with increasing thickness of the lms. The present paper reports results of our continuing effort in the development of alternative hard chromium-coating techniques. Our previous work focused on the composite, hard, chromium-coatings using a RF magnetron sputter deposition method w6 8x. In this study, we are developing the composite hard chromium-coatings using a pulsed DC magnetron deposition technique.

Fig. 1. Schematic drawing of the vacuum chamber for the four-gun pulsed DC magnetron sputter deposition unit.

2. Experimental procedure 2.1. Sample preparation Cr and CrN were deposited on commercially avail-

able 1010 steel 267 ml Steel, Zinc-coated, Kocour company, Chicago, IL. and Si 100. wafers. The reason for using the zinc-coated steel was to have a well-controlled fresh surface with no pits or scratches for coating after removing the zinc layer. The dimensions of steel specimens were approximately 25 = 25 = 0.3 mm. The Zn layer on the steel substrates was removed by using 15% HCl solution and the steel substrates were, then, rinsed with distilled water. The Si 100. wafer and steel substrate were further cleaned with acetone, and methanol in an ultrasonic cleaner for 20 min, and dried by blowing with pressurized oxygen. The coatings were deposited using a DC magnetron sputter deposition unit Cooke vacuum product, Norwalk, CT.. This unit has four US-II guns for mounting targets and is shown in Fig. 1. During this study, only one gun was used for depositing CrrCrN coatings. In a separate study, another gun was used for depositing an interlayer prior to depositing CrrCrN coatings without breaking the vacuum. The pulsed DC plasma generator RPG-50 ENI, Rochester, NY. with a frequency of 185 kHz and duty cycle of 96% was used to deposit coatings in this study. The plasma generator has a power output of 10 kW maximum output voltage of 500 V.. To minimize potential effect of warm-target on deposition, the gun was cooled by deionized water with a minimum specic resistivity of 15 M rcmy2 . After placing the substrates in the chamber, the vacuum was drawn in the chamber. Typically, the base vacuum was approximately 1.99 = 10y4 Pa 1.5 = 10y6 torr.. Before each deposition, pre-sputtering was done for one hour with ultrahigh-purity argon as the sputtering gas to remove any nitridesroxides from the target surface. For all depositions, the sputtering power was maintained at 200 W. A target-to-substrate distance of 8.4 cm was maintained during all depositions. Deposition was carried out at a total pressure argonrargon

16

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

Table 1 Deposition parameters and deposition rates for Cr and CrN coatings on steels prepared at 200C with pulsed DC magnetron-sputter deposition Sputter gas N2 content 0 vol.% 3 vol.% 5 vol.% 7 vol.% Power W. 200 200 200 200 Deposition time min. 92 93 97 93 Thickness m. 1.92 1.85 1.93 1.89 Deposition rate mrh. 1.25 1.19 1.19 1.22

plus reactive gas. of 1.333 Pa 10 mtorr. with the substrate being at 200C, which was maintained by a substrate heater with nichrome wire-resistance heating. The sputtering gas compositions were controlled by two ow meters, one each for argon and nitrogen, and the nitrogen composition is reported in terms of percentage volume. In this study, the nitrogen content in the sputtering gas was varied from 0 to 7%. The deposition parameters are presented in Table 1. 2.2. Characterizations The thickness of the lms was measured with a Dektak II prolometer Veeco, UK.. During deposition, a dummy sample with a small, masked area was placed along with the sample on the substrate holder. The dummy sample consisted of a large piece of silicon wafer with a small piece of silicon wafer glued on top to mask deposition in the small area. The top piece was removed after deposition producing a step, which consisted of uncoated silicon wafer and the coating. The prolometer was used to measure the step height, which represented the thickness of the deposited lm. An average of four readings was taken for each sample. X-Ray diffraction XRD. analysis was performed to determine the phases presented in the coatings using a Phillips diffractometer with CuK radiation, Ni lter, and a scanning speed 2 sy1 . of 0.04 per s. Microhardness of the lms was determined using a M-400-H1 microhardness tester from LECO. An average of ten readings was taken for each sample. A dwelling time of 20 s, and loads between 0.245 N and 1.96 N were used for the measurement. The indent diagonals were measured with an optical microscope at a magnication of 400 times. The adhesion between the coating layer and the substrate and the critical load, at which the coating fails, were examined by using a Sebastian-V scratch tester Quad Group, Spokane, WA.. The sample stage moves horizontally along with the sample at a predetermined scanning rate. An increasing normal load, starting from zero, is applied on the stylus diamond tip with a 533- m in radius. at a predetermined loading rate. Each sample was scratched three times. Computer-logged data obtained from the scratch test in-

clude applied load, transverse force, friction coefcient, and acoustic emission signal, which is detected by a piezoelectric acoustic transducer mounted on the stylus arm. The acoustic noise generated during the scratch test is recorded in order to identify the critical force. A burst in the acoustic signal indicates either debonding or cracking of the coatings. The scratch tracks on the lms were examined by using an optical microscope. During the scratch test, the scan and loading rates were maintained at 0.04 cm sy1 and 0.294 N sy1 , respectively. The maximum load and travel distance was xed at 39 N and at 3 cm, respectively.

3. Results 3.1. Sputter deposition rate To determine possible incubation period of deposition, a series of Cr depositions were done on the steel substrate for 60, 90, 180 and 300 min. The thickness of the deposited Cr lms was found to vary linearly with the deposition time indicating that there was no incubation time for deposition under the deposition condition of this study. The deposition rate was determined as the slope of the coating thickness vs. the deposition time. For coating with 0% nitrogen in the sputtering gas, the deposition rate was found to be 1.24 mrh. The sputter deposition rate was determined as a function of nitrogen content in the sputtering gas. As also shown in Table 1, the deposition rate was independent of the nitrogen content. 3.2. Structure of coatings The XRD patterns of the Cr coatings are shown in the Fig. 2. Although there is only one peak each for the Cr and CrN phases presented in Fig. 2, more Cr and CrN peaks were observed at angles beyond 65 which was omitted in this gure. Body-centered cubic BCC.Cr phase is detected in all coatings and face-centered cubic FCC.-CrN phase is detected only when nitrogen gas exists in the sputtering gas. Hexagonal closedpacked HCP.-Cr2 N phase was not detected in any coatings. The potential presence of the Cr2 N phase is

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

17

Fig. 3. Cross-section images of Cr and CrrCrN coatings deposited on Si 100. wafer with a. 0, b. 3, c. 5 and d. 7% N2 , at 200C.

Fig. 2. X-Ray diffraction patterns of Cr and CrrCrN coatings deposited at 200C with 0, 3, 5 and 7% N2 in the gas.

to be analyzed in the discussion section. With increasing nitrogen contents in the sputtering gas, the intensity of CrN 200. peak was increased. In Fig. 2, the peak-broadening effect of Cr and CrN coatings was caused by the small grain size and the lattice strain w26x. The peak full width at half maximum FWHM. was measured for Cr 110. and CrN 200. phases. The measured FWHM was found to be 0.02 of 2 for the Cr 110. peak. The average grain size of Cr and its nitride coatings can be calculated using Scherrers equation w27x. d s 0.94 rB 2 . cos 1.

used to observe the cross-sectional image of the coatings, shown in Fig. 3, deposited in various nitrogen concentrations. All coatings clearly show the columnar microstructure. The coating thickness determined by the prolometer agrees with that measured from the SEM images. 3.3. Microhardness of coatings The composite microhardness of Cr and CrN on 1010 steel were measured using Vickers diamond indenter. The applied loads were 0.245, 0.49, 0.98 and 1.96 N. The hardness of the composite was twice as high as that for the steel substrate when the nitrogen content in the sputtering is 7% when a load of 0.245 N is applied. The composite microhardness is summarized in Table 2. The lm hardness, Hf , was calculated using Jonsson and Hogmark expression w28x, Hf s Hs q Hc y Hs t t 2C yC 2 D D
2

where d is the crystal dimension perpendicular to the reecting plane, is the wavelength, and B 2 . is FWHM. Using this equation, the calculated grain size of Cr 100. phase for the sputter-deposited Cr coating was in the range of 7 to 8 nm. However, the grain size for the CrN phase in the coating cannot be calculated using Scherrers equation because the peak is too diffused to correctly calculate the FWHM of the diffraction pattern. Scanning electron microscopy SEM. was
Table 2 Microhardness HV. of the substrate and coated samples Load N. Microhardness of substrates HV. 129.32 132.21 132.07 125.72

2.

where, Hf is the lm hardness calculated, Hc is the composite hardness measured, Hs is the substrate hardness measured, t is the lm thickness, D is the

Microhardness of coated samples HV. 0% N2 211.05 167.40 167.10 143.52 3% N2 196.70 168.72 161.03 148.86 5% N2 204.93 180.75 161.32 150.37 7% N2 255.94 185.15 165.01 154.64

0.245 0.49 0.98 1.96

18

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

Fig. 5. Optical micrographs of a typical indent on CrrCrN coating at 1.96 N load.

Fig. 4. Hardness of the Cr and CrrCrN coatings sputter-deposited on 1010 steel substrates with 0, 3, 5 and 7% N2 in the gas and electrodeposited Cr-coating. Hardness of the Cr target was measured from the target used in this study.

indentation depth and C is 2sin 2 11. The calculated lm hardness was shown in Fig. 4. The lms were much harder than the substrate and the hardness increased with increasing nitrogen content in the sputtering gas.

An optical micrograph of a typical indent was shown in Fig. 5. Cracks were observed around the rim of the indentation. The presence of cracks could be responsible for low calculated lm hardness. The reported bulk hardness values of CrN and Cr2 N are 1100 and 1600 HV, respectively w6x. The calculated microhardness of the coatings showed that the microhardness at a 0.245 N load increased from 1229 HV at 0% N2 , to 1664 HV at 7% N2 in the sputtering gas. This corresponds to the formation of the hard CrN phase in Cr phase as well as the solid solution strengthening of the Cr lattice by nitrogen. It also shows that the hardness of the sputter-deposited Cr lm with 0% N2 was higher than that of the Cr target used in this study due to the grain-boundary effect. Since the sputter-deposited Cr

Fig. 6. Scratches of coatings deposited with various amounts of N2 in the sputtering gas.

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

19

and CrN lms show a grain size of 7 8 nm, the hardness of Cr lms is likely to be higher than that of the bulk Cr or Cr target. Microhardness measurements of the Cr target indeed reveal a low microhardness as presented in Fig. 4. The microstructure of the bulk Cr and Cr target show that the grain size of the bulk Cr to be approximately 300 m, and that of Cr target to be approximately 50 m. 3.4. Adhesion of the lms Scratch tests were performed for all coatings to determine the lm integrity and the lm adhesion. All scratch tracks were examined under an optical microscope. The scratch of coating with 0% N2 content in the sputtering gas exhibits transverse cracks inside the scratch track and chips along the scratch track at a relatively low load. At a relatively high load, coatings were completely removed and the substrate was exposed. The coatings with 3, 5 and 7% N2 contents in the sputtering gas show a smooth scratch track instead of a transverse crack at a relatively low load, and little broken chips at a relatively high load. Fig. 6 shows the scratch tracks of these four conditions. To determine the critical load, the computer-logged acoustic signal was examined along with the scratch

track image. Typical computer-logged data and an optical micrograph of scratches on Cr coating were shown in Fig. 7. The critical load increases with increasing nitrogen contents in the sputtering gas. The critical loads of deposited lms with 0, 3, 5 and 7% N2 in the gas were 1.57, 5.68, 8.33 and 20.29 N, respectively.

4. Discussion 4.1. X-Ray diffraction analysis of the relati e amount of Cr and CrN in the coating The relative intensities of Cr 110. and CrN 200. peaks for each coating were presented in Fig. 8. The amount of CrN was found to increase with increasing N2 content in the sputtering gas while Cr decreases with it. From the relative intensities of Cr 110. and CrN 200. peaks, the amount of CrN phase in the coating has been calculated and is also included in Fig. 8. The amount of the CrN phase in the coating increased from 31.6 vol.% at 3% N2, to 49.3 vol.% at 7% N2 in the gas. As shown in the result section, the coating microhardness increased with increasing N2 contents in the gas. This quantitative analysis of CrN phase in the coating suggests that the amount of the

Fig. 7. Typical computer-logged data and optical micrograph of scratched CrrCrN coating deposited on 1010 steel substrate with 7% N2 in the gas.

20

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

2 Cr s . q N2 g . s 2 CrN s . G O s y242462 y 48.1T log T q 359.1T ; J r mol

3.

3 - 1. 4 Cr s . q N2 g . s 2 Cr 2 N s . 4.

G O s y217149 y 48.1T log T q 317.7T ; J r mol 4 - 1 . For stability comparison, the Gibbs free energy of mixing from elements Crs. and N2g . based on per mole of the binary system must be calculated for each phase. Fig. 10 shows the calculated free energy of mixing, G M , for composition N s 0.33 and 0.5 at 200, 400, 600, 800 and 1000C where N is the atomicmole. fraction of nitrogen in the Cr-N binary system. At 200C, G M for Cr2 N is higher than the combination of Cr q CrN., indicating that Cr2 N is not stable with respect to the Cr q CrN. combination. Hence, the phases occurring at 200C are Cr and CrN. A similar situation holds for the stability of phases at 400C although the free energy data suggest that Cr2 N is approaching equilibrium with Cr q CrN.. At 600, 800 and 1000C, all Cr, CrN, and Cr2 N phases are stable. As shown by the relative G M values of each compound, such thermodynamic analysis matches well with the results obtained in this study and those reported previously w6 8x. The phase diagrams of the Cr-N binary system reported in the literature w30 33x do not include the region below 500C. Based on thermodynamic stability analysis shown above and experimental results in the present study, it is believed that the Cr2 N phase is not

Fig. 8. Relative intensities of Cr 110. and CrN 200. peaks deposited with 0, 3, 5 and 7% N2 in the gas at 200C. The calculated volume fraction of CrN in the coating is shown on top of the CrN bar.

CrN phase in the coating may enhance coating hardness. Fig. 9 shows the dependence of coating hardness as a function of the CrN content in the lm. 4.2. Stability of Cr, Cr2 N and CrN As reported previously, no Cr2 N phase appeared in X-ray diffraction patterns although the phase diagram of the Cr-N binary system shows the existince of Cr2 N. To determine the stability of Cr, Cr2 N and CrN phases, thermodynamic analysis was performed. The standard Gibbs free energy of formation of CrN and Cr2 N are obtained from Kubaschewski and Alcock w29x.

Fig. 9. The dependence of coating hardness at 0.98 N load as a function of the CrN content in the lm.

Fig. 10. The variation of G M with X N , mole fraction of nitrogen in the Cr-N binary system.

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22

21

stable below 400C. At approximately 400C, the eutectoid reaction Crs. q CrNs . s Cr2 Ns . exists. By allowing G M for Crs. , G M for Cr2 Ns . , and G M for CrNs . into a straight line in the G M vs. composition diagram, the corresponding temperature was found to be at 430C. This is proposed to be the eutectoid temperature. From this, the portion of the Cr-N phase diagram below 500C is proposed as shown in Fig. 11. Results by Almer et al. w21x, Creus et al. w22x and Su et al. w25x agree with this proposed phase diagram. Analysis made here was based on the thermodynamic equilibrium consideration. Deposition with extraordinarily high perturbations, such as high voltage bias on the substrate w5x or high deposition power w19x, might result in the formation of otherwise unstable phases such as Cr2 N. Results reported in the literature also showed that RF magnetron sputter-deposition might have facilitated the formation of Cr2 N at temperatures lower than that predicted by the equilibrium consideration discussed here w6,12 14,20x. 4.3. Comparison of sputter deposited Cr r CrN coatings with electroplated Cr coatings Electroplated hard Cr coatings are industrially popular in numerous applications due to their excellent wear and corrosion resistance. As mentioned above, a serious drawback of electroplated hard Cr coatings is the hexavalent Cr emission during production. Sputter-deposition, on the other hand, is environmentally safe. There is no hexavalent chromium emission at all. For industrial applications, sputter Cr coatings must perform as good as electroplated Cr coatings in wear resistance. Microhardness is an indication of the mechanical strength of coatings. To determine the quality of sputter-deposited CrrCrN coatings in this study, microhardness of Cr

Fig. 12. Optical micrographs of the surface of Cr-coatings on 1010 steels; a. Sputter-deposited CrrCrN lm with 7% N2 in the gas, b. Electrodeposited Cr lm.

lms coated with electrodeposition was determined and compared with those for sputter deposited CrrCrN lms. Commercially available electroplated Cr coatings on 1010 steel were obtained from the Porter-Guertin company, Cincinnati, OH. The coating hardness of the electroplated Cr lms were calculated using the Jonsson and Hogmark expression w20x, and also shown in Fig. 4. At 0.245 and 0.49 N loads, the coating hardness of electroplated Cr lms is higher than that of the sputter-deposited CrrCrN lms with 0, 3 and 5% N2 in the gas. At 0.98 and 1.96 N loads, the coating hardness of electroplated Cr decreased signicantly. It is noted that since the hydrogen gas was generated during the electroplating of Cr, the complex hydrides and cavities were formed in the Cr coatings. Fig. 12 shows the optical micrographs of the surface of a sputter deposited CrrCrN and an electroplated Cr coating. Due to the cavities in the Cr coatings, widely-scattered data were observed from the electrodeposited Cr coatings. With 7% N2 in the gas, CrrCrN coatings show higher hardness at all loads than the electroplated coating. These results suggest that sputter-deposition can produce CrrCrN coatings, mechanically superior to electroplated hard Cr coatings. 5. Conclusions Pulsed DC magnetron sputter deposition of CrrCrN lms was carried out at 200C for different nitrogen contents in the gas. This study indicates that sputterdeposition is a feasible method for producing Cr coatings. From results found in this study, the following conclusions can be obtained. 1. CrrCrN coatings on steels have been successfully prepared with sputter-deposition. 2. XRD analysis of the coatings shows that the volume fraction of CrN in the coating increases with the increasing nitrogen contents in the sputtering gas. 3. The grain size of the Cr phase was found to be 7 8 nm.

Fig. 11. Phase diagram of the Cr-N binary system. The phase diagram below 500C is proposed in this study.

22

J.W. Seok et al. r Surface and Coatings Technology 138 (2001) 14 22 w11x A. Ehrlich, M. Kuhn, F. Richter, W. Hoyer, Surf. Coat. Technol. 76 77 1995. 280 286. w12x G. Bertrand, C. Savall, C. Meunier, Properties of reactively r.f. magnetron-sputtered chromium nitride coatings, Surf. Coat. Technol. 96 1997. 323 329. w13x C. Meunier, S. Vives, G. Bertrand, X-ray diffractometry analysis of r.f.-magnetron-sputtered chromiumrchromium nitride coatings, Surf. Coat. Technol. 107 1998. 149 158. w14x G. Bertrand, H. Mahdjoub, C. Meunier, A study of the corrosion behavior and protective quality of sputtered chromium nitride coatings, Surf. Coat. Technol. 126 2000. 199 209. w15x L. Cunha, M. Andritschky, Residual stress, surface defects and corrosion resistance of CrN hard coatings, Surf. Coat. Technol. 111 1999. 158 162. w16x B. Navinsek, P. Panjan, A. Cvelbar, Characterization of low temperature CrN and TiN PVD. hard coatings, Surf. Coat. Technol. 74 75 1995. 155 161. w17x B. Navinsek, P. Panjan, I. Milosev, Industrial applications of CrN PVD. coatings, deposited at high and low temperatures, Surf. Coat. Technol. 97 1997. 182 191. w18x G. Aldrich-Smith, D.G. Teer, P.A. Dearnley, Corrosion wear response of sputtered CrN and S-phase coated austenitic stainless steel, Surf. Coat. Technol. 116 119 1999. 1161 1165. w19x F. Cosset, G. Contoux, A. Celerier, J. Machet, Deposition of corrosion-resistant chromium and nitrogen-doped chromium coatings by cathodic magnetron sputtering, Surf. Coat. Technol. 79 1996. 25 34. w20x C. Friedrich, G. Berg, E. Broszeit, K.-H. Kloos, X-ray diffractometry analysis of r.f.-sputtered hard coatings based on nitrides of Ti, Cr, Hf, Surf. Coat. Technol. 74 75 1995. 279 285. w21x J. Almer, M. Oden, L. Hultman, G. Hakansson, Microstructural evolution during tempering of arc-evaporated Cr-N coatings, J. Vac. Sci. Technol. A 18 1. 2000. 121 130. w22x J. Creus, H. Idrissi, H. Mazille, F. Sanchette, P. Jacquot, Improvement of the corrosion resistance of CrN coated steel by an interlayer, Surf. Coat. Technol. 107 1998. 183 190. w23x A. Kawana, H. Ichimura, Y. Iwata, S. Ono, Development of PVD ceramic coatings for valve seats, Surf. Coat. Technol. 86 87 1996. 212 217. w24x F.D. Lai, J.K. Wu, Structure, hardness and adhesion properties of CrN lms deposited on nitrided and nitrocarburized SKD 61 tool steels, Surf. Coat. Technol. 88 1996. 183 189. w25x Y.L. Su, S.H. Yao, C.T. Wu, Comparisons of characterizations and tribological performance of TiN and CrN deposited by cathodic arc plasma deposition process, Wear 199 1996. 132 141. w26x R.S. Smith, IBM J. Res. 4 1960. 205 207. w27x B.E. Warren, X-ray Diffraction, Dover Publications Incorporated, New York, 1990. w28x B. Jonsson, S. Hogmark, Thin Solid Films 114 1984. 257 269. w29x O. Kubaschewski, C.B. Alcock, Metallurgical Thermochemistry-Fifth Edition, Pergamon Press, 1979, p. 379. w30x M. Venkatraman, J.P. Newmann, Binary Alloy Phase Diagrams 2 1990. 1293 1297. w31x P.M. Hansen, Constitution of Binary Alloys, McGraw Hill, 1958, pp. 539 541. w32x R.P. Elliott, Const. Binary Alloys-First Supplement, McGraw Hill, 1965, pp. 351 352. w33x F.A. Shunk, Const. Binary Alloys-Sec Supplement, McGraw Hill, 1969, p. 273.

4. With increasing nitrogen contents in the sputtering gas, the microhardness of the coatings increase. This is due to the composite effect of CrN in Cr. 5. The adhesion property of the coatings increases with increasing nitrogen content in the sputtering gas. 6. Thermodynamic analysis indicates that Cr2 N is not stable at the deposition conditions of this study. Experimental results also verify the non-existence of Cr2 N at 200C. The region below 500C of the Cr-N binary phase diagram is proposed with a eutectoid reaction occurring at approximately 430C.

Acknowledgements The authors would like to thank the EPA, NASA, the WSU-PCC program and the Air Force research laboratory, for their support of this project. We would also like to thank the Porter-Guertin company for providing electroplated Cr on steel samples. References
w1x R.Y. Lin, Recent coating technology development, in: K.S. Shin, J.K. Yoon, and S.J. Kim Eds... Proceedings of the Sec Pacic Rim International Conference on Advanced Materials and Processing, The Korean Institute of Metals and Materials, 1995, pp. 571 578 w2x M.S. Hall, J.D. Dietz, C.D. Cooper, R.L. Wayson, D. Bauman, Plat. Surf. Fin. 79 11. 1992. 18 22. w3x A. Aubert, R. Gillet, A. Gauvher, J. Terrat, Thin Solid Films 108 1983. 165 172. w4x D. Dubiel, in: W.D. Munz, E. Broszeit, H. Oechsner, K.T. Rie, G.K. Wolf, Eds... Proceedings of the International Conference on Plasma Surface Engineering, 1989, 853 860 w5x T. Hurkmans, D.B. Lewis, J.S. Brooks, W.D. Munz, Chromium nitride coatings grown by unbalanced magnetron UBM. and combined arcrunbalanced magnetron ABSTM. deposition techniques, Surf. Coat. Technol. 86 87 1996. 192 199. w6x M. Pakala, R.Y. Lin, Surf. Coat. Technol. 81 1996. 233 239. w7x M. Pakala, R.Y. Lin, Sputter deposition of hard chromium coatings on steel, in: T.S. Sudarshan et al. Ed.., Surface Modication Technologies-Ninth Edition, TMS Publications, Warrendale, PA, 1996, pp. 467 475. w8x M. Pakala, Z. Fan, R.Y. Lin, Sputter deposition of chromium nitride coatings on steel, in: W.D. Cho, H.Y. Sohn Eds.., Value-Addition Metallurgy, TMS Publications, Warrendale, PA, 1998, pp. 237 248. w9x R.R. Aharonov, B.F. Coll, R.P. Fontana, Surf. Coat. Technol. 61 1993. 223 226. w10x S.J. Bull, D.S. Rickerby, Surf. Coat. Technol. 43 44 1990. 732 744.