In gamma-ray spectrometry for volume samples, the self-attenuation effect should be considered in the case of differences in chemical composition and density between the efficiency calibration source for quantitative analysis of sample and the sample actually measured. In particular, the lower the gamma-ray energy, the greater the gamma-ray attenuation due to the self-attenuation effect of the sample. So, the attenuation effect of low-energy gamma-rays in the sample should be corrected to avoid over- or under-estimation of its radioactivity. One of the most important factors in correcting the self-attenuation effect of the sample is the linear attenuation coefficient for the sample, which can be directly calculated using a collimator. The larger the size of the collimator, the more advantageous it is to calculate the linear attenuation coefficient of the sample, but excessive size may limit the use of the collimator in a typical environmental laboratory due to its heavy weight. Therefore, it is necessary to optimize the collimator size and structure according to the measurement environment and purpose. This study is to optimize a collimator that can determine the effective linear attenuation coefficient of low-energy gamma-rays, and verify its applicability. The overall structure of the designed collimator was optimized for gamma-ray energy of less than 100 keV and cylindrical plastic bottle with diameter of 60 mm and a height of 40 mm. The materials of optimized collimator consisted of tungsten. Acryl and acetal were used to form the housing of the collimator, which fixes the central axis of the bottle, collimator and point-like source. In addition, using the housing, the height of the tungsten is adjusted according to the height of the sample. For applicability evaluation of the optimized collimator, IAEA reference material in solid form were used. The sample was filled in the bottle with heights of 1, 2, 3 and 4 cm respectively. Using the collimator and point-like source of 210Pb (46.5 keV), 241Am (59.5 keV), and 57Co (121.1 keV), the linear attenuation coefficient and the radioactivity for the samples were calculated. As a result, to calculate the linear attenuation coefficient using the optimized collimator, a relatively high sample height is required. However, the optimized collimator can be used to determine the linear attenuation coefficients of low-energy gamma-rays for the self-attenuation correction regardless of the sample height. It is concluded that the optimized collimator can be useful to correct the sample selfattenuation effect.