X-Işını floresans tekniklerinde matris çoğaltması ile matris soğutmasının dengelenmesi

Abstract

ÖZET Uygun enerjideki bir fotonun, herhangi bir elementin atom yörüngesinden bir elektron koparması sonucunda, boşalan yere farklı bağ enerjisine sahip bir elektron transfer olur. Bu enerji farklılığından dolayı bir miktar enerji hedef elementin floresan X-ışını olarak açığa çıkmaktadır. Hedef atomun etrafında bulunan diğer atomlar ya üretilen floresans X~ışınım algılayıcıya ulaşmadan soğururlar (MATRÎS SOĞURMA ETKÎSÎ) veya da hedef atomu uyarımda kullanılan birincil X-ışınları uyarılarak, kendi floresans X-ışınlarını üreterek hedef atom için ek bir ışınlama kaynağı haline gelirler (MATRÎS ÇOĞALTMA ETKÎSÎ). Uygulamada karşılaşılan bu problemler X- ışını floresans tekniklerinin sistematik hata kaynakları olarak bilinmektedir. Bu çalışmada ilk defa her iki problemin yani, hem matris çoğaltmanın ve hem de matris soğurmanın aynı anda ele alınarak tek bir malzeme içindeki davranışlarının gözlemlenmesi amaçlanmıştır. Ancak bu gözlem sonucunda iki etkinin birbirlerini dengeleyip dengelemeyecekleri anlaşılabilir. Bu amaçla; değişik oranlarda bakır, demir ve germanyum elementleri içeren farklı miktarlarda ve farklı fiziksel fazlarda bulunan 320 örnek hazırlanmış ve bakırın analizi amaçlanmıştır. Bu örneklerde demirin matris soğurma ve germanyumun da matris çoğaltma sağlamak amacı ile kullanıldıkları enerjilerinin karşılaştırılmasından da anlaşılmaktadır (Cu ka =8.047 keV kab = 8.980, Fe ka = 6.403 keV kab = 7.111 keV, Ge ka = 9.885 keV kab = 11.103 kev). Elde edilen bakır, demir ve germanyum sayımlarının yanı sıra bakırın demir veya germanyum sayımları üzerinden normalize edilmiş sayımlar ile birlikte 200 'ü aşkın değerlendirme grafikleri elde edilmiştir. Bu sonuçlara dayanarak matris çoğaltma etkisinin izlenen şiddetinin örnek kütle yoğunluluğuna ve dolayısı ile öz soğurma faktörüne bağımlı olduğu sonucu ilk defa ortaya konulmuştur. Ayrıca ilk defa matris çoğaltma etkisi ile matris soğurma etkilerinin arasında, öz soğurma etkisinin faktörü olarak bir korelasyon kurulması ile bu etkilerin giderilmesi başarılmıştır. iv

At the beginning it should be mentioned that, the topic of this work was appeared as the question of ` What is the probability of setting up a balance between the matrix absorption and the matrix enhancement effects in X-ray florescence techniques`? However, before going further, one should have a quite clear scope of the main principles and the conformity of the application limits due to these principles in the x-ray fluorescence techniques. X-ray fluorescence spectrometry is known as a very effective application of nuclear techniques in material evaluation, serving both the scientific researches and industry. The basic principle that this technique has been built on is, the interaction of a photon 'holding a sufficient energy with an electron of the stable atom which is due to be detected. The evaluation of the results of such a detection enables to analyse the elemental composition of the examined material, both quantitatively and qualitatively. From the nuclear physics point of view, if photon being produced from a source is aimed towards an atom and is carrying an energy which could break the bounding energy of an electron in the target, it can remove this electron from its base orbital and up grade it to an upper orbital or throw it to the inter atomic space. This of course causes an instability in the target atom. In such a condition the excited atom starts rearranging its electron shells vand one of the electrons from an upper orbital transfers to the empty place of the lost electron. This procedure that brings the atom to its stable position happens in less than lO`lsSec. Mean while the difference between bounding energy of the transferred electron and its destination is emitted from the atom as the characteristic X-ray of the target element, The emitted characteristic X-ray which is namely called fluorescence X-ray of the element is monochromic, and its energy defines the target element's characteristics. As it's being seen the theory is quite obvious and straight away, but the same thing can not be referred to the practical applications. Problems related with the practical works and studies are generally originated from the impossibility of working with a single atom of any element. In this mean it is unavoidable to have a mass of atoms which provides matrix effects on the fluorescence of the target element. These effects are called systematic error sources for X-ray fluorescence techniques and are classified in two different aspects, one is the matrix absorption effect and the other is the matrix enhancement effect. Matrix absorption effect is referred to the absorption of the fluorescence photon produced by the target element's atom on the way of detection by any other element's atom. This of course causes a reduction in the detected fluorescence X-ray. The matrix enhancement effect is caused by the atoms of any other element which may act as a secondary source by emitting its own florescence with the energy of higher than absorption edge of the target element. This provides an increase in the detected fluorescence of the target element. Many studies have been done to correct this systematic errors in the X-ray fluorescence techniques. Most of these studies are dealing with the matrix absorption effect, but none was found to be. referred to the both effects acting on one element at the same VItime. Many of these studies have ended with a correction technique for the matrix effects in X-ray- fluorescence, but most of them has neglected the enhancement effect. It has also been observed that some authors have tried to fix a certain value for the enhancement and generalise it to all applications. As the conditions of each application differs in accordingly with the elemental configuration and physical phase of the material under exam, the generalisation of one correction method could end the application results with unavoidable errors, In this work, it was planed to study the multi element samples case, generating both the matrix absorption and the matrix enhancement in the same sample. In this purpose more than 300 samples containing variable percents of Copper, Iron and Germanium were prepared. These samples can be categorized by the mean of their total mass in four different groups (lgr., 50rag., 7*100 jjI. and 12*100il.), so this could give the opportunitie of evaluating the dependent of the matrix effects in multi-element samples to the thickness. Such a composition can also give the ability of comparing the self absorption oscillation in the variable thickness for the examined samples. All the prepared samples were counted in the same condition such as geometry of the counting system. A well equipped X-ray florescence research laboratory under the authority of the Turkish Atomic Energy Organisation was permitted to be used in this work. The experiment system was constructed of a plutonium 238 annular source, a sample holder, a HpGe detector, a high voltage supplier, a preamplifier, an amplification device, an oscilloscope, a multichannel buffer, a mini exchanger, a multichannel analyser and its computer and plotter. All the elements of each sample were counted by the multichannel analyser and the results were normalised for 100 Sec. In according to the matrix elements count rates one could judge the efficiency of the back scattering correction method for the multi-element material. Almost 200 graphs were plotted for different sample groups. VI iIt was observed that, as the efficiency of the matris absorption reduces by the thickness reduction in the samples containing germanium, the enhancement effect increases. Also, the expected enhancement for the copper in the germanium matrix was unexpectedly appear as a matrix absorption effect in the samples thicker than 50 mg mass concentration. The neutralisation of the germanium matrix effects for the samples around 50 mg mass concentration, indicates the balance realisation between the self absorption and the enhancement effect of the germanium. As the self absorption is minimised in the thin film samples, an intensive enhancement effect appears for all the samples containing germanium. This in deed confirms the efficiency of the thin film correction method for the self absorption and the matrix absorption effects, but it also indicates the unconformity and the limitation of this technique for the material evaluation of 3uch configuration. Also a correlation was successfully established between the matrix enhancement and self absorption in the samples with the medium thickness. In the thick samples this correlation was found to be applicable only for a certain mass concentrations of the germanium, iron and copper. The normalisation of the copper count rate by the absorber matrix count rate with the purpose of matrix effect correction was found to be useful in some cases but not very effective. Normalisation of the copper count rate by the total counts of the germanium and iron for all the samples was found to be more effective. The results obtained in this work is indicating the importance of consideration of both the matrix absorption and matrix enhancement effects as well as the self absorption in the multi-element materials. It should be also noticed that the dependent of the both the matrix absorption and the matrix enhancement effect to the self absorption effect of the sample should be consider as the matter of the sample preparation. The techniques used for sample preparation in X-ray vmfluorescence techniques, should mach the elemental composition and the object of the operation. As the use of the X-ray fluorescence techniques are getting more common in the industry, in according to obtain and supply more accuracy in the consequences of the applications it is suggested to take more care of the composition of the material under the exam. In this mean the correction and sample preparation techniques should be selected more attentively. This gains more importance in working with mobile and single channel analysers. For such a condition it is suggested to try to compare a thin film sample ' s count result with a thick sample result by the mean of a standard sample count results. The comparison of the mathematical corrections for matrix absorption with the technique introduced in this work can be suggested. It has to be noticed that the most of these corrections are ampiric and do not imply the matrix enhancement effect in multi-element samples. Such a correction method could be a subject to a detailed research, but the results might not be very reliable because of the random nature of the X-rays. IX