Danoix F, Grancher G, Bostel A, Blavette D
Groupe de Physique Des Matériaux, UMR CNRS 6634, Institut de Matériaux de Rouen, Université de Rouen, Site Universitaire du Madrillet, Avenue de l'Université, 76 801 Saint Etienne du Rouvray, France.
Ultramicroscopy. 2007 Sep;107(9):739-43. doi: 10.1016/j.ultramic.2007.02.005. Epub 2007 Mar 2.
In a companion paper [F. Danoix, G. Grancher, A. Bostel, D. Blavette, Surf. Interface Anal. this issue (previous paper).], the derivation of variances of the estimates of measured composition, and the underlying hypotheses, have been revisited in the the case of conventional one dimensional (1D) atom probes. In this second paper, we will concentrate on the analytical derivation of the variance when the estimate of composition is obtained from a 3D atom probe. As will be discussed, when the position information is available, compositions can be derived either from constant number of atoms, or from constant volume, blocks. The analytical treatment in the first case is identical to the one developed for conventional 1D instruments, and will not be discussed further in this paper. Conversely, in the second case, the analytical treatment is different, as well as the formula of the variance. In particular, it will be shown that the detection efficiency plays an important role in the determination of the variance.
在一篇配套论文[F. 达诺伊克斯、G. 格兰彻、A. 博斯特尔、D. 布拉韦特，《表面与界面分析》本期（上一篇论文）]中，针对传统一维（1D）原子探针的情况，重新审视了测量成分估计值的方差推导以及相关基本假设。在第二篇论文中，我们将专注于从三维（3D）原子探针获得成分估计值时方差的解析推导。如后文将讨论的，当位置信息可用时，成分可以从恒定原子数或恒定体积的块中推导得出。第一种情况下的解析处理与传统一维仪器所采用的相同，本文将不再进一步讨论。相反，在第二种情况下，解析处理不同，方差公式也不同。特别要指出的是，检测效率在方差的确定中起着重要作用。
