Ultratrace analysis is a term which still awaits a clear definition but it is gaining rising importance especially for hightech materials and processes. Relevant ultratrace levels of importance depend largely on the investigated matrix but generally range below 1 ppm. Some factors leading to fiction in ultratrace analysis are discussed such as: The yet undefined terms typical analysis, actual analysis and guaranteed analysis Misleading certificates of analysis for ultrapure materials The listing of total purities according to the number of nines concept with a lot of exceptions Purity data based exclusively on integral physical methods Extremely small analytical volumes versus the degree of heterogeneous distribution of a trace component In extreme cases, heterogeneous ultratrace distribution might limit the applicability of a respective material The enormous economic pressure exerted on analysts in industry and non ideal conditions in relevant analytical laboratories. On the other hand, some of the most fascinating aspects of ultratrace analysis are elucidated: Spectroscopic and other methods of end determination are approaching single atom detection Topochemical materials characterization is today feasable with atomic resolution. However the high costs of relevant instrumentation and of the necessary highly qualified personnel hinders a faster progress and the exorbitant prices of ultrapure materials often hinder their use and less pure material is used instead. A good example of what is possible today in ultratrace characterization of complex matrices is given by an overview on the results of a European working group "Ultrapure refractory metals and their silicides for microelectronic applications" of the COST 503/II-Action (Fig. 1). Within this working group the following methods were applied: Mass spectrometric methods: SSMS, GDMS, SIMS, IDMS, ICP-MS, classical spectrometric methods: ICP-OES, GFAAS after TMS. Trace-matrix separation methods were developed for the matrices Mo and W (some also for Ti, Nb and Ta) in the batch mode and by use of column chromatography (HPLC, IC). Activation analytical methods (mainly by NAA) were also developed. All these methods are critically compared. Essential progress in the production of ultrapure refractory metal and silicide sputter targets as well as their safe characterization was possible by this joint effort. Many erroneous results of the past could be explained and can now be avoided.
