The release of deuterium from sputter magnetron produced tungsten co-deposit layers is studied by thermal desorption mass spectrometry and modelled with the diffusion reaction codes TESSIM and FACE. Layers up to ∼ 2 µm thick, produced at substrate deposition temperatures up to 513 K are modeled. TESSIM simulations are found to require activity from at least nine traps in the range $1.0-2.5$ eV in order to reproduce the experimentally observed desorption. FACE simulations, which utilize a quasi-continuous distribution of traps, suggest some discreteness in trapping energy above ∼ 1.5 eV, but smoothness in the distribution below. Both codes indicate a quasi-exponential decrease in trap concentration with trap energy. When examined for predictive capability, the developed tungsten co-deposit models accurately reproduce experimental changes in the desorption heating rate from 0.3 to 3 Ks${}^{-1},$ and give reasonable agreement with experimentally different layer thicknesses and deposition temperatures in the parameter ranges explored. Measured D/W ratios in the co-deposits are also found to be in good agreement with literature based predictive scalings.

通过热脱附质谱研究了溅射磁控管产生的钨共沉积层中的氘释放，并用扩散反应代码TESSIM和FACE进行了建模。对在高达513 K的基底沉积温度下产生的约2 µm厚的层进行了建模。发现TESSIM模拟要求该范围内至少9个陷阱的活动$1.0-2.5$eV以重现实验观察到的解吸。利用陷阱的准连续分布的FACE模拟表明，在约1.5 eV以上的陷阱能量中存在一些离散，但在下面的分布中却具有平滑性。这两个代码都表明阱浓度随阱能量的下降呈指数下降。当检查其预测能力时，已开发的钨共沉积模型可以准确地再现从0.3到3 Ks的解吸加热速率的实验变化${}^{-\mathrm{1个}}，$并在探索的参数范围内与实验上不同的层厚度和沉积温度达成合理的一致。还发现在共沉积物中测得的D / W比值与基于文献的预测标度有很好的一致性。