Diffusional isotope fractionation has been widely used to explain lithium (Li) isotope variations in minerals and rocks. Isotopic mass dependence of Li diffusion can be empirically expressed as $\frac{D_{7_{\mathit{Li}}}}{D_{6_{\mathit{Li}}}}={\left(\frac{6}{7}\right)}^{\beta }$, where $D$ is the diffusivity of a Li isotope. The knowledge about temperature and compositional dependence of the $\beta$ factor which is essential for understanding diffusion profiles and mechanisms remains unclear. Based on the potential energy and interatomic forces generated by deep neural networks trained with ab initio data, we performed deep potential molecular dynamics (DPMD) simulations of several Li pseudo-isotopes (with mass = 2, 7, 21, 42 g/mol) in albite, hydrous albite, and model basalt melts to evaluate the $\beta$ factor. Our calculated diffusivities for ⁷Li in albite and model basalt melts at 1800 K compare well with experimental results. We found that $\beta$ in albite melt decreases from $0.267\pm 0.006$ at 4000 K to $0.225\pm 0.004$ at 1800 K. The presence of water appears to slightly weaken the temperature dependence of $\beta$, with $\beta$ decreasing from $0.250\pm 0.012$ to $0.228\pm 0.031$ in hydrous albite melt. The calculated $\beta$ in model basalt melt takes much smaller values, decreasing from $0.215\pm 0.006$ at 4000 K to $0.132\pm 0.015$ at 1800 K. Our prediction of $\beta$ in albite and hydrous albite melts is in good agreement with experimental data. More importantly, our results suggest that Li isotope diffusion in silicate melts is strongly dependent on melt composition. The temperature and compositional effects on $\beta$ can be qualitatively explained in terms of ionic porosity and the coupled relationship between Li diffusion and the mobility of the silicate melt network. Two types of diffusion experiments are suggested to test our predicted temperature and compositional dependence of $\beta$. This study shows that DPMD is a promising tool to simulate the diffusion of elements and isotopes in silicate melts.

扩散同位素分馏已广泛用于解释矿物和岩石中锂（Li）同位素的变化。Li扩散的同位素质量依赖性可以根据经验表示为$\frac{d_{7_{\mathit{里}}}}{d_{6_{\mathit{里}}}}={\left(\frac{6}{7}\right)}^{\beta }$， 在哪里 $d$是Li同位素的扩散率。关于温度和成分依赖性的知识$\beta$对于理解扩散分布和机制至关重要的因素尚不清楚。基于由头算数据训练的深度神经网络生成的势能和原子间力，我们对几种Li伪同位素（质量= 2、7、21、42 g / mol）进行了深势分子动力学（DPMD）模拟。在钠长石，含水钠长石和玄武岩模型中熔融以评估$\beta$因素。我们计算出的⁷ Li在1800 K的钠长石和玄武岩熔体中的扩散率与实验结果相比较。我们发现$\beta$ 钠长石中的熔体从 $0.267\pm 0.006$ 在4000 K至 $0.225\pm 0.004$ 在1800 K时。水的存在似乎稍微减弱了温度对温度的依赖性。 $\beta$， 和 $\beta$ 从减少 $0.250\pm 0.012$ 至 $0.228\pm 0.031$在含水钠长石熔体中。经计算$\beta$ 在模型中，玄武岩熔体的取值要小得多，从 $0.215\pm 0.006$ 在4000 K至 $0.132\pm 0.015$ 在1800K。 $\beta$钠长石和含水钠长石熔体的熔体与实验数据非常吻合。更重要的是，我们的结果表明，锂同位素在硅酸盐熔体中的扩散强烈依赖于熔体成分。温度和成分对$\beta$可以用离子孔隙率和锂扩散与硅酸盐熔体网络迁移率之间的耦合关系定性地解释。建议使用两种类型的扩散实验来测试我们预测的温度和成分依赖性。$\beta$。这项研究表明，DPMD是模拟硅酸盐熔体中元素和同位素扩散的有前途的工具。