ENERGY SPECTRUM OF EJECTED ATOMS DURING HIGH ENERGY SPUTTERING OF GOLD
ENERGY SPECTRUM OF EJECTED ATOMS DURING HIGH ENERGY SPUTTERING OF GOLD
复制标题
DOI：
10.1080/14786436808227358
复制
发表时间：
1968-01-01
影响因子：
1.6
通讯作者：
THOMPSON, MW
中科院分区：
文献类型：
--
作者：
THOMPSON, MW
使用第一部分中描述的技术，已经确定了几种金样品（单晶和多晶聚集体）的溅射原子能谱。实验选择了⟨110⟩、⟨110⟩和⟨121⟩喷射方向。使用 43 kev A+ 离子、43 kev Xe+ 离子或 66 kev Xe+ 离子进行轰击。喷射原子的能量范围从约 10−2ev 到约 104ev。通常在 1 至 10 ev 之间的能谱中观察到峰值。在该峰的高能侧，光谱的表现大致类似于 E−2，尽管与此存在显着偏差，特别是对于 ⟨100⟩ 喷射。开发了一个理论模型，其中喷射主要是由轰击离子产生的原子碰撞级联产生的。级联中的能量由两体碰撞共享并且平均无碰撞路径与能量无关的假设导致 E−2 谱，并且该预测应大致保持在 10 到 103ev 之间。结果表明，具有结合能 Eb 的原子表面结合的结果是原子轨迹的折射和光谱中靠近 E=Eb 的峰值。这种随机级联模型无法解释光谱的所有特征，特别是当它们被绘制为飞行时间光谱时出现的尖峰。但更先进的模型允许通过级联生成集中的碰撞序列，非常适合数据。这允许两种类型的序列（辅助序列和简单序列），并真实地代表它们各自的特征。与高级模型预测的偏差仅限于低于 1 ev 的能量（人们认为热尖峰会对此产生影响）和高于 103ev 的能量。其中无碰撞路径可能通过通道效应和横截面随能量的正常减小而增强。在室温下，差异仅占总溅射产率的 10% 左右。随机级联和聚焦碰撞序列的贡献大致相等。通过比较预测和观察到的光谱，可以推断，首先，对于沿 ⟨110⟩ 行行进的简单聚焦碰撞序列，传播的最大能量为 167±25 ev，其最大范围至少为 21 次碰撞；其次，对于⟨110⟩辅助序列，能量限制为500±100 ev，最大范围约为23次碰撞：第三，这些实验中与金表面的有效结合能从2·5到4·1 ev变化。推导出Au的原子间势，与观察到的⟨110⟩聚焦能和体积弹性模量一致。在玻恩-迈耶形式.Aexp(—r/b)中，常数为A= 200±60 kev 和b= 2 ·88/(14 ·3 ·0 ·4) Å，有效范围为1 ·4 到2 ·88 Å。
Using the technique described in part I, the energy spectrum of sputtered atoms has been determined for several gold specimens, both single crystals and polycrystalline aggregates. The ⟨110⟩, ⟨110⟩ and ⟨121⟩ directions of ejection were chosen for the experiments. Bombardment was with either 43 kev A+ions, 43 kev Xe+ions or 66 kev Xe+ions.The energies of the ejected atoms ranged from some 10−2ev up to about 104ev. A peak was generally observed in the energy spectrum between 1 and 10 ev. On the high energy side of this peak the spectrum behaved roughly likeE−2, though there were significant deviations from this, particularly for ⟨100⟩ ejection.A theoretical model is developed in which ejection results principally from the generation of atomic collision cascades by the bombarding ion. The assumptions that energy in the cascades is shared by two-body collisions and that the mean collision free path is independent of energy lead to anE−2spectrum and this prediction should hold roughly from 10 to 103ev. It is shown that the consequence of surface binding of atoms, with a binding energyEb, is a refraction of atomic trajectories and a peak in the spectrum nearE=Eb.This random cascade model is unable to explain all features of the spectra, particularly the sharp peaks that appear when they are plotted as time-of-flight spectra. But a more advanced model, which allows for the generation of focused collision sequences by the cascade, fits the data extremely well. This allows for both types of sequence, assisted and simple, and represents their separate characteristics realistically. Deviations from the predictions of the advanced model are confined to energies below 1 ev, where thermal spikes are thought to contribute, and above 103ev. where the collision free path is probably enhanced by channelling and the normal decrease in cross section with energy. At room temperature the discrepancy only amounts to about 10% of the total sputtering yield. The random cascade and focused collision sequences contribute roughly equal amounts.From a comparison of the predicted and observed spectra it can be deduced, firstly, that for simple focused collision sequences travelling along the ⟨110⟩ rows the maximum energy of propagation is 167±25 ev and their maximum range is at least 21 collisions; secondly, that for ⟨110⟩ assisted sequences the energy limit is 500±100 ev and the maximum range is about 23 collisions: thirdly, that the effective binding energy to the gold surface in these experiments varied from 2·5 to 4·1 ev.An interatomic potential for Au is deduced, consistent with the observed ⟨110⟩ focusing energy and the bulk modulus of elasticity. In the Born–Mayer form.Aexp(—r/b), the constants areA= 200±60 kev andb= 2 ·88/(14 ·3 ·0 ·4) Å, valid from 1 ·4 to 2 ·88 Å.