1. Introduction Silver is one of the most useful materials which are used as the part of electronics equipments because of its high electric and thermal conductivity, while sulfur compounds in the air easily corrode silver in the form of silver sulfide causing a defective continuity. Thus, reliability of electronics equipments consisting of silver should be secured by enough corrosion resistance tests. Corrosion resistance test time are usually decided using field condition and test condition corrosion rate data, however, reported corrosion rate data are limited1. Thus, in order to decide corrosion resistance test time, year-long time tests have been applied under conditions at which the corrosion rate of silver is unknown. The purpose of this study is to get corrosion rate data quickly by an estimation model for the corrosion rate of silver. The factors which are known to be related to silver corrosion are temperature, humidity and corrosive gas concentration. In this study, relations among these three factors and the corrosion rate were defined by making a corrosion rate estimation model which was consist of these factors. 2. Experimental 2.1. Samples Pure silver plates (99.99%) which used in this study were about 30 x 30 x 0.3mm in size. The weights of the plates were precisely measured before the tests. 2.2. Setting of factors Hydrogen Sulfide (H2S) gas was selected as the corrosive gas because of its high reactivity toward silver. Considered diversity field which electronics are installed, the ranges of each factors were decided in the following. H2S concentration : 0.05 to 1.05ppm. Temperature : 25 to 45℃. Humidity : 60 to 90%RH. In this range, three levels (max., med., and min.) were selected for the each factor, and totally 27 (3 x 3 x 3) condition tests were settled for the tests. 2.3. Exposure experiments Silver plates were exposed by H2S in gas concentration, temperature and humidity controlled. After exposure, weights of silver plates were measured. The exposure tests were performed 30 days in maximum. 2.4. Method of analysis The measured 27 corrosion rate data by the exposure tests were analyzed, and the regression equation, that is, the corrosion rate estimation model were made with analysis software (Design Director Version 1.0, NHK Spring Co., LTD.2). This estimation model was composed of three factors, temperature, humidity, and corrosive gas concentration. 3. Results and Discussions 3.1. Development of estimation model The weights of silver plates showed liner increase with exposure time for all the 27 tests, and these increasing patterns followed the famous classic research3. Thus, it was defined that the weight increase of the silver plates shows the amounts of corrosion, and that the liner slopes show the corrosion rates V. From the 27 tests, relations among the factors and the corrosion rates V were analyzed and significant factors were selected. As a result, we got estimation model shown in the following. Corrosion rate estimation model; V=36.0-1.5T+0.021T2-0.12[H2O]-489.2[H2S]+396.6[H2S]2+28.2T[H2S]-0.39T2[H2S]-25.1T[H2S]2+0.36T2[H2S]2+0.65[H2O][H2S] V: corrosion rate (g/m2・h), T: temperature (℃), [H2O]: humidity (%RH), [H2S]: hydrogen sulfide concentration (ppm) From the amount of coefficient, it was clarified that the three factors, especially H2S concentration, have influence on the corrosion rate of silver, and there are interactions among these factors. 3.2. Check of validity of estimation model Another exposure test under a different condition from the 27 sets was performed in order to check the validity of the estimation model. The Vest. (given by the estimation model) to Vexp.(given by the test) ratio was 1.21, it was said that developed estimation model shows high estimation quality. 4. Conclusion In this study, estimation model for the corrosion rate of silver in H2S gas was developed. This model enabled us to get corrosion rate data quickly without having long time test. Reference 1. R. Minamitani, Zairyo-to-kankyo, 64, 14 (2015). 2. NHK Spring Co., LTD., “Features of Design Director”, http://www.nhkspg.co.jp/eng/products/industry/dd/aboutdd.html. 3. D. W. Rice, P. Peterson, E. B. Rigby, P. B. P. Phipps, R.J. Cappell, and R. Tremoureux, J. Electrochem. Soc., 128, 275 (1981).