there are good publications on sample size estimation and statistical power calculation. In brief, your estimated sample size depends on the magnitude of your expected effect, which is the difference between the mean values of your two compared groups divided by the standard deviation, your chosen significance level alpha, for example 0.05 and your chosen statistical power, e.g. 80%. There are also free softwares available like G*power which you could use. Just have a look at the attached publication and link: