This study is aimed toward identifying the reasons why large discrepancies exist in the literature relative to the kinetic constants which are supposed to describe the thermal oxidation of single crystal silicon (Si) in pure oxygen false(O2false) . In order to obtain sufficient quality and quantity of silicon dioxide false(SiO2false) film growth thickness‐time data, an automated ellipsometer was used to measure the SiO2 thickness in situ. The resulting data was fitted to the generally accepted linear‐parabolic model by several commonly used methods and the results compared. Careful attention was given to eliminating trace amounts of H2O and Na so that the data are representative of oxidation in pure O2 ; the oxidation data was compiled in the temperature range of 780°– 980°C. Calculated standard deviation values were used to evaluate the quality of the fit of the data to the model. From this analysis an initial regime of rapid oxidation was identified which does not conform to linear‐parabolic kinetics. This regime extends up to about 350Å. A best fit of the data to the model was achieved using data above 350Å. By using either data below 350Å or only data greater than about 1100Å, large curve‐fitting errors (∼50% in terms of rate constants) were obtained. It was concluded that this source of error in combination with impurity effects, insufficient data, and the specific form of the curve‐fitting equation could yield the large reported discrepancies. The activation energy calculated from the linear rate constants of this study (1.5 eV) indicates that 0‒0 bond breaking is important for linear kinetics and the activation energy for the parabolic rate constants (2.3 eV) is too large to be correlated with a reported value (1.2 eV) for the diffusion of O2 through SiO2 .

Effects associated with the incorporation of chlorine during the thermal growth of SiO2 (500-1400A) on Si are investigated as a function of additive species (HC1, C12), additive concentration, and oxidation temperature (900~176 Chlorine profiles are measured mainly by nuclear backscattering spectroscopy, and oxide thicknesses and refractive, index changes by ellipsometry. The highest chlorine concentrations in unannealed oxides always occur at the Si-SiO2 interface. Under otherwise similar conditions, the use of C12 results in higher, more evenly distributed chlorine levels (up to 1.2 • 1021 at era-3). The incorporated chlorine is gradually removed by annealing at temperatures in excess of II00~ as well as under conditions where further oxide growth takes place without the presence, of HCI or Cl2. Additional oxide formation in steam effectively removes all the incorporated chlorine. MOS capacitors made from oxides grown in HCI or Cl2 show fixed charge levels lower than 10 n cm -2 and mobile charge levels less than 1O l~ cm -2. When biased at high fields (>6 MV/cm) signihcant shifts in the flatband potential of Si were observed. Chlorine-grown oxides exhibit positive and negative shifts attributed to a mobile species; the smaller negative shifts in HCI oxides are attributed to interface states. Other electrical data of intentionally Na-contaminated HCI oxides suggest that only a constant fraction of the Na is trapped.

No abstract

The thermally activated growth of oxide on silicon as a function of time obeys a linear‐parabolic relationship, the linear part of which stems from interface limited reactions. In Part I of this paper, it has been reported that this linear part cannot result from a single rate‐limiting reaction step, because the order of the over‐all reaction rate differs for different substrate orientations at a fixed temperature and varies for a given orientation as a function of temperature. A kinetic model for the reaction between silicon and oxygen at the normalSi‐SiO2 interface is now presented to account for the experimental data false(dnormalSiO2 300Aå,Tnormalox=700°–1000°C,normalpO2=0.01–1.0 normalatmfalse) . Two parallel, competing reactions are postulated to occur. In the first of these, molecular oxygen reacts directly with silicon to form silicon dioxide and atomic oxygen; the second reaction involves the dissociation of O2. The atomic oxygen thus formed, may either react with silicon or recombine to molecular oxygen. An analysis of the data shows that a difference in the activation energies (i.e., 1.91 vs. 0.58 eV) associated with these competing reaction steps is responsible for the shift in their relative importance as a function of temperature.

No abstract