CCA agent consisting of Cu (II), Cr (VI), and As (V) were previously used for preservation of wood products such as ground plates, telegraph poles, and timber. CCA products are not now manufactured in Japan, but about 20,000m(3) of CCA-preserved wood waste is predicted to be produced annually in the coming decades. Under the Construction Materials Recycling Law of 2000, this CCA-preserved wood waste is required to be recycled appropriately. Pulp alternative is one possible destination of recycling that utilizes the properties of wood. For this purpose, CCA removal must be compatible with the conservation of wood components. In this study, an agent for CCA removal was chosen on the basis of CCA distribution in the CCA-preserved wood waste, and then removal of CCA was carried out in conjunction with investigation of the loss of wood components. Three CCA-preserved wood wastes (samples A, B, and C) were obtained from the CCA-waste storage facility in Chugoku region, Japan. After chipping the wood waste to less than 5 mm, each sample was subjected to an electron probe micro analysis (EPMA) to determine the distribution of CCA. EPMA showed that CCA mainly existed in intra-cellular sections. Since intra-cellular sections contain lignin, which consists of phenolic structures, an oxidizing agent, hydrogen peroxide was chosen for removal of CCA. For optimized removal using hydrogen peroxide, we investigated effects of 4 operation parameters, including treatment time, temperature, hydrogen peroxide concentration, and solid/liquid ratio (hydrogen peroxide/CCA-preserved wood waste), on CCA removal and loss of wood components. Finally, removal of CCA was carried out at 90 C for 3 h with 4% hydrogen peroxide in a ratio of 15ml.g(-1). The treatment using hydrogen peroxide attained almost 100% removal efficiencies for Cr and As in samples B & C, whereas the values for sample A were 68% and 66%, respectively. A removal efficiency of more than 95% was observed for Cu in all samples. These results confirmed the feasibility of removing CCA by using hydrogen peroxide, although the removal efficiencies might vary with characteristics of the targeted wood waste. As for wood components, the total weight loss after the treatment was 23%, 22%, and 30% for the samples A, B, and C, respectively, while loss of holocellulose was 23%, 26%, and 30%. Despite these losses, the ratio of holocellulose to total weight in the treated sample was not changed, thereby indicating that the treated wood waste has potential for use as a pulp alternative.