There seems to be no literature that has reported plasma-nitriding of protrusions using the RF magnetron sputter apparatus. Thus, the experiments were carried out changing the nitrogen gas pressure, a mixing rate of argon gas, sputter power and sputter time as are classified in Table 1. In group 1, the plasma-nitriding is carried out only by using nitrogen gas (purity: 99.999%) of 130 Pa, which is the lowest pressure level for DC plasma-nitriding but will be very high for RF-magnetron nitriding, at a sputter power of 50 or 100 W for 7.2 ks or 14 ks. The specimens nitrided with these conditions are called, e.g. N2130 Pa–50 W–7.2 ks specimen, N2130 Pa–100 W–7.2 ks specimen, or N2130 Pa–(50, 100)W–7.2 ks specimens together. The reason for much shorter nitriding time than the conventional DC plasma-nitriding time is that the nitride layers should be limited only on protrusion surface even if they Marimastat are formed. However, from the experimental results as is described later, the indentation resistance of protrusions was only 1.3 times as large as blood group or type of the as-sputter-etched protrusions even for the nitriding time of 14 ks. So that the second process classified as group 2 was tried by decreasing the nitrogen pressure to 1.2 Pa. Although the indentation resistance largely increased in the N21.2 Pa-(100, 200)W-7.2 ks specimens, the nitriding power of 100 and 200 W eventually resulted in a loss of sharpness and formation of brittle nitrides on the protrusions. Reduction of the nitriding power to 50 W suppressed the nitride formation but the indentation resistance was still about 1.4 times as large as that of the as-sputter-etched specimen. Finally, the third nitriding processes of group 3 were tried at a lower nitrogen gas pressure of 0.53 Pa mixed with argon gas of 0.67 Pa; N20.53 Pa–Ar0.67 Pa–50 W–1.8 ks specimen, etc.