Maria Luisa Polignano

Comparison among lifetime techniques for the detection of transition metal contamination

Abstract A systematic comparison among the most common methods (surface photovoltage (SPV), Elymat, and microwave-detected photoconductive decay (μ-PCD)) for lifetime measurements is presented. Though these techniques are very different from each other, we show that, where bulk-diffused impurities are concerned, they agree very well with each other, provided they are properly used. In order to validate these techniques for the quantitative evaluation of bulk-diffused contaminants, iron and chromium implantations were carried out. An excellent correspondence was found between Elymat and μ-PCD data. In addition, the lifetime dependence on iron dose was studied and the expected behaviour was verified over two orders of magnitude. Elymat and μ-PCD measurements have also been applied to the study of contaminants segregated at wafer surface, such as nickel and copper. Both these techniques are very sensitive to surface-segregated metals, though under these conditions the correlation between Elymat and μ-PCD data is somewhat different with respect to samples with metals dissolved in the bulk. A measurement procedure is proposed in order to discriminate bulk and surface recombination.

Crystal defects and junction properties in the evolution of device fabrication technology

In this paper, the correlation between dislocation density and transistor leakage current is demonstrated. The stress evolution and the generation of defects are studied as a function of the process step, and experimental evidence is given of the role of structure geometry in determining the stress level and hence defect formation. Finally, the role of high-dose implantations and the related silicon amorphization and recrystallization is investigated.

Quantitative evaluation of bulk-diffused metal contamination by lifetime techniques

Abstract In this paper we present a systematic comparison among the most common methods (surface photovoltage, Elymat and microwave-detected photoconductive decay) for lifetime measurements. The possibility to identify contaminants and to quantitatively evaluate their concentration by lifetime techniques is investigated. Though these techniques are very different from each other, we show that in relation to bulk-diffused impurities, they agree very well with each other, provided that the dependence of carrier lifetime on the injection level is taken into account when comparing them. Iron and cromium implanted wafers are used in order to validate these techniques for the quantitative evaluation of bulk-diffused contaminants. Iron concentrations obtained from lifetime data are compared to measurements of chemical concentration of iron at wafer surface. A very good correlation is obtained between estimates of iron concentration from lifetime data and O 2 -leak SIMS data in samples contaminated by implantation of dopant ions.

Metal contamination monitoring and gettering

Abstract Some commonly used techniques for metal contamination monitoring by lifetime measurements (surface photovoltage, Elymat and microwave-detected photoconductive decay) are discussed and compared. In order to validate these techniques for the detection and the quantitative evaluation of bulk-diffused contamination, iron and chromium implanted samples have been studied. Though these techniques are very different from each other, we show that for what concerns bulk-diffused impurities they agree very well with each other, provided in the comparison the dependence of carrier lifetime on the injection level is taken into account. In addition, the possibility to extend these techniques to surface characterization (e.g. for the detection of surface-precipitated metals) is studied. Nickel and copper implantated wafers were used to this purpose. Both Elymat and μ-PCD are found to be very sensitive to surface-segregated metals, though under these conditions the correlation between Elymat and μ-PCD data is somewhat different with respect to samples with metals dissolved in the bulk. Finally, some gettering techniques are reviewed and compared for what concerns gettering efficiency. It is pointed out that gettering efficiency is sensitively reduced in high temperature rapid thermal treatments. This fact can be explained by efficiency loss of the segregation mechanism.

Molibdenum contamination in silicon 1. Molibdenum detection by lifetime techniques

Abstract In this work the use of lifetime techniques in the presence of a non-uniform profile of recombination centers is discussed. It is shown that measurements with different probing depths can be used in order to obtain information about the in-depth distribution of recombination centers. Molibdenum was chosen for this study, because it is probably the most common slowly diffusing contaminant. In this study, it is also shown that molibdenum inactivation does take place by segregation at the wafer surface. The evidence from lifetime techniques is confirmed by TEM analyses, revealing tiny molibdenum clusters at the wafer surface.

Molybdenum Contamination in Silicon: Detection and Impact on Device Performances

In this paper, a case of molybdenum contamination from wet cleaning is discussed, and various techniques are compared for their ability to detect molybdenum. In addition, the impact of this sort of contamination on the electrical results of a bipolar device is studied.

Investigation of metal contamination by photocurrent measurements: validation and application to ion implantation processes

In this work we present a systematic study about the metal contamination induced by ion implantation, with the aim to identify contamination mechanisms and possible solutions to the problem. Lifetime measurements have been used in order to evaluate the level of contamination in implented wafers. Lifetime values have been extracted from photocurrent measurments (Elymat technique). Implantations of iron and cromium have been used in order to validate the study of lifetime versus injection level as a technique for the identification of contaminants and for the quantitative evaluation of their concentration. The contamination level in ion implanted wafers has been characterized varying main implantation parameters (species of the implanted ion, dose, current, energy, angle) and surface condition (whether bare or oxidized silicon). Ion implantation is responsible for a a heavy lifetime degradation (i.e. metal contamination), which increases in proportion to implantation dose and comes from the side exposed to the ion beam. The distribution of lifetime over wafer surface provides relevant information. Details of the implanter endstation (e.g., the clamping system) usually show up in wafer maps of lifetime. Results coming from different equipments concur to indicate that contaminants come from material sputtered from the loading disk. This conclusion is confirmed by the dependence of lifetime on implantation energy and tilt angle. The chemical nature of the contaminant can in some cases be identified by injection level spectroscopy. Implantation of heavy ions is mainly responsible for iron contamination; some other impurity (maybe cromium) is detected in boron-implanted wafers. From the point of view of device processing, the problem can be circumvented by implantation through a screening exide. Vice versa gettering techniques remove only a limited fraction of the contaminants introduced during the implantation.

Niobium Contamination in Silicon

In this paper niobium is characterized as a silicon contaminant. It is shown that niobium is a relatively slow diffuser, with an intermediate diffusivity between very slow diffusers such as molybdenum and fast diffusers such as iron. Niobium is found to be very effective as a recombination center, and in addition prone to surface segregation. In addition, niobium shows optical activation, but no thermal activation. Three deep levels are revealed in niobium contaminated silicon, plus an additional level observed in high contamination dose samples only. One of these levels is located very close to midgap, and consistently niobium was also found very effective in degrading the generation lifetime.

Highly Sensitive Detection of Inorganic Contamination

As the detection of inorganic contaminants is of steadily increasing importance for the improvement of yields in microelectronic applications, the aim of one of the joint research activity within the European Integrated Activity of Excellence and Networking for Nano- and Micro-Electronics Analysis (ANNA, site: www.ANNA-i3.org) is the development and assessment of new methodolo¬gies and metrologies for the detection of low concentration inorganic contaminants in silicon and in novel materials. A main objective consist in the benchmarking of various analytical techniques avail¬able in the laboratories of the participating ANNA partners, including the improvement of the res¬pective detection limits as well as the quantitation reliablity of selected analytical techniques such as total-reflection x-ray fluorescence (TXRF) analysis.

Study of SiO2Si interfaces by photocurrent measurements

Abstract An experimental study is presented about surface recombination velocity obtained from photocurrent measurements by the Elymat technique. Surface recombination velocity has been studied as a function of process conditions (oxidation cycle, hydrogen annealing, cleaning procedures) with the aim to investigate how this parameter can be used for determining some properties of oxide-silicon interfaces. The data of surface recombination obtained from photocurrent measurements have the expected dependence on the oxidation cycle and on a H2/N2 post-oxidation annealing, so we concluded that these data actually reflect the status of the oxide-silicon interface. In addition, these data are shown to be affected by the cleanliness of the oxidation equipment, and they are unaffected by bulk properties of the oxide. By using these measurements, we show that N2O treated interfaces have a better stability under baking treatments than N2 annealed interfaces. The dependence of surface recombination velocity on injection level can be modeled by analogy with bulk recombination by Shockley—Read—Hall recombination statistics.