BACKGROUND: Marine sedimentary carbonate rock is an important carrier for recording seawater information. The Sr isotope composition (87Sr/86Sr) of carbonate rocks can reflect the relative contribution of the continental crust and mantle to the Sr isotope composition of seawater. The long-term variation trend of Sr isotope composition in geological history can be used to interpret global tectonic events, weathering rate changes, biogeochemical cycles, and determine the age of marine sedimentary strata. However, the carbonate rocks likely contain non-carbonate fractions to varying degrees, which lead to the whole rock Sr isotope composition being unequal to that of the primary carbonate fraction. In order to obtain the primary carbonate fraction that reflects the primitive seawater, an effective leaching method is required. OBJECTIVES: To identify experimental procedures and target leaching steps that can effectively extract representative primary carbonate fractions in carbonate rock samples of varying purity and variety. METHODS: Reference materials of dolostone and limestone (GBW03105a and ECRM-782-1) were selected to represent carbonate rock samples with high purity, and natural samples of limestone (C-3, purity: 85%) and dolostone (E-3, purity: 65%) were selected to represent samples with low purity. The leaching solution of all steps was measured for Ca and Mg contents by inductively coupled plasma-optical emission spectrometry (ICP-OES), for Sr, Mn and Al contents by inductively coupled plasma-mass spectrometry (ICP-MS). The Sr isotope was measured by multi-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS) after purification of the leaching solution. Through the utilization of various indicators, such as Sr/Ca and Mn/Sr, the targeted leaching steps were ascertained. RESULTS: (1) Factors affecting the Sr isotope composition of the primary carbonate fraction. The detection of Sr isotope composition of primary carbonate fraction is affected by the soluble and exchangeable Sr, resulting in higher 87Sr/86Sr values. Thus, the pre-leaching step is essential for the multiple-step leaching method. It is shown that an excess of 5% acetic acid can cause leaching of non-carbonate fraction of limestone and affect the Sr isotope composition of the primary carbonate fraction. (2) Comparison and selection of multiple-step extraction methods. (a) It is recommended to use the method proposed by Li, et al[29] for limestone samples. It was found that the carbonate of GBW03105a and C-3 dissolved in the target steps (A10-A11) proposed by Liu, et al[13] accounted for 46.51% and 39.49% of the total carbonate fraction. A large number of target carbonate was dissolved rapidly in these two steps without considerable differentiation. In the method proposed by Li, et al[29], GBW03105a and C-3 target steps (B7-B9) dissolved carbonate accounted for 26.41% and 30.31% of the total carbonate fraction. The target steps are close to the step of non-carbonate fractions dissolved in strong acid. For natural samples with complex mineral compositions, the method proposed by Li, et al[29] may have leached non-carbonate fractions, which resulted in a higher 87Sr/86Sr value for testing. Therefore, the more conservative method proposed by Li, et al[29] is recommended for leaching unknown limestone samples, and B7-B9 are the target steps for leaching representative primary carbonate fraction. (b) The method proposed by Liu, et al[13] is recommended for dolostone samples. Liu, et al[13] selected acetic acid with different concentration ranging from 0.25%-10% for leaching. The carbonate fraction is almost completely dissolved, and the lowest Sr isotope value measured is lower than the method proposed by Li, et al[29] . Hence, to choose the concentration of acetic acid from low to high for leaching is helpful to separate the target fractions of dolostone samples. The same concentration of acetic acid is chosen by the method of Li, et al[29], and the dissolution rate of samples with different purity began to slow down at A9, leaving about 30% of the carbonate fraction not leached. Thus, it is difficult to judge whether the target carbonate fraction using this method has been leached completely. For the multiple-step leaching of unknown dolostone samples, the method proposed by Liu, et al[13] is recommended here, and A14-A15 are selected as the target steps for leaching representative primary carbonate fraction. (3) Optimization of a multiple step leaching method. The insoluble powder in the leaching solution will be digested by the subsequent addition of nitric acid, which can affect the Sr isotope value of the target fraction. This study has found that the Al/Ca ratio in the acetic acid leaching part is lower than that in Li, et al[13] (Fig.3b), indicating that the filter can reduce pollution caused by the dissolution of non-target fractions. In addition, the preleaching steps can also be optimized. The experimental data showed that 5mL of 1mol/L ammonium acetate could be selected for pre-leaching to simplify the experimental procedure and shorten the processing time. CONCLUSIONS: Based on prior research, a focused multiple-step leaching method for carbonate rocks is proposed. Limestone samples (purity≥85%) are suitable for 9-step leaching with 1% acetic acid, and the target steps are L7-L9; dolomite samples (purity≥65%) are suitable for the 14-step leaching method with 0.25%-10% acetic acid, and the target steps are D13-D14. The Sr isotope value of the primary carbonate fraction of the European Committee for Steel Standardization (ECISS) dolostone reference material ECRM-782-1 has been reported for the first time, which is 0.707868±0.000034 (n=12, 2SD). In the future, the experimental methods will be further improved to encompass samples from diverse sources and varying purity, thereby ensuring the reliability and universality of the experimental approach. [ABSTRACT FROM AUTHOR]