excited ground electronic state of Br2 was made. NeC1, remains as the only example of a van der Waals molecule whose metastable states are observed to live more than Acknowledgment.The reaction rate constants for OH + Cl,, Br2, and BrCl are determined to be 6.70 X cm3 molecule-' s-', respectively. Rate measurements are carried out at 298 K in He by using a discharge flow apparatus.Two OH sources are used: H + NO2, and F + H,O. OH decays are monitored by resonance fluorescence at 309 nm; C1and Br are measured simultaneously with OH by resolved resonance fluorescence at 134.7 and 157.7 nm, respectively.Measurement of OH + BrCl requires corrections for C1, and Br, which exist in equilibrium with BrC1. Atom yields from the title reactions are compared to the reference reactions H + C1, and Br, and analyzed by a computer model of the reaction kinetics. The products in each case are found to be the halogen atoms (Cl and Br), and not the halogen oxide radicals (C10 and BrO). Using the observed, forward reaction rate constants and the equilibrium constants derived from thermodyanmic data, we calculate the rate constants of the reverse reactions, C1 + HOC1, Br + HOBr, and C1 + HOBr, to be 2.5 X 2.4 X cm3 molecule-' s-I, respectively. The reactivity of OH toward the three halogen molecules is related to the ionization potentials of the halogens, lower ionization potentials promoting faster reactions. In addition, a comparison of the results obtained here with the rate constant measurements of other radical-halogen reactions indicates that a radical's reactivity correlates with its electron affinity. On the basis of these trends, the activation energies of the title reactions are estimated to be 2.1,0.0, and 3.4 kcal mol-', respectively. The dependence of the rate constant on the ionization potential and electron affinity of the reactants may serve as a guide for estimating the rates of similar radical-molecule reactions. 5.28 X lo-", and 1.49 X and 3.7 X IntroductionThe chemistry of the hydroxyl radical is fairly well characterized in many systems. The O H radical is of such paramount importance in the atmosphere and in flames that many of its reactions with molecules such as CO and CHI have received considerably attention. The reactions of OH with halogens and interhalogens have been far less completely determined; only

No abstract

This work examines the pH and time dependence of metal deposition onto hydrophilic silicon surfaces. We find that the surface concentration of metal species is proportional to the pH. Time dependence is seen for only a few metals, notably Cr and Fe in this study, especially at high pH. Si surfaces with wet chemical oxides were exposed to a dilute solution of ten metals: Al, Ba, Ca, Cr, Cu, Fe, K, Ni, Sr, and Zn, in dilute nitric acid solution, from pH 3 to 5.6, for different periods of time, from about 1 to 1000 s. At equilibrium the surface concentration of metals is approximately proportional to the ratio of ionic charge to ionic radius of the metal ions for many of the metal ions which exist as hydrated cations at that particular pH. We observe this proportionality over the pH range under study, although the surface concentration varies over several orders of magnitude. Cations which form amphoteric species, such as A13+ which forms A1,0, and Al(0H); at high pH, and several other cations that do not exist as simple cations at some pHs (e.g., Fe3', Zn2+, and Ni2' at high pH), do not show this dependence on ionic charge and radius.

The kinetic behavior of OH with I2 and IC1 is reported within the context of a series of radical-halogen reactions to investigate the mechanism of such reactions through studies of reactivity trends. We report 298 K rates and product channels for OH + I2 -I + HOI (kla) and OH + I2 -IO + HI (klb, AH = 25 kcal mol-') of kl = kla + k1b = 1.6 X lo-'' cm3 molecule-' kcal mol-'), OH + IC1 -C10 + HI (kzc, A H = 16.9 kcal mol-'), and OH + IC1 -IO + HC1 (k2d, A H = 7.5 kcal mol-') of k2 = 2.0 X lo-'' cm3 molecule-' s-', where k2a + k2c + k2d < 0.01k2b. Atomic I and c1 products of the title reactions are measured by resolved resonance fluorescence vapor lamps. IO and C10 are detected by chemically converting them, using NO, to I and C1, respectively. A factor of 2 uncertainty must be associated with these values because of the inherent difficulty of, and uncertainty arising from, the separation of the homogeneous and heterogeneous reaction components. In addition, the rate constant of the reaction H + IC1 has been measured to be (5 f 2) X lo-" cm3 molecule-' s-I at 298 K; its only product channel is I + HC1. Observed products of the OH-halogen reactions suggest that -15 < AHf(HOI) < -1 kcal mol-'. Magnitudes of the rate constants of these OH reactions substantiate the reactivity trends in the OH-halogen system based on an electron-transfer mechanism from the highest occupied molecular orbital of the halogen to the lowest unoccupied orbital of the hydroxyl. The rapidity of the OH + I2 reaction makes it a possible source of HOI for photochemical studies.S-', with klb/kla < 0.01, and for OH + IC1 -CI + HOI (kza, AH unknown), OH + IC1 -I + HOC1 (k2b, A H = -6.1

The contamination of cleaned silicon wafer surfaces by metal ions is a significant problem. The RCA cleaning process developed by Kern and Puotinen 1 has been very successful at reducing metal contamination to low levels; however, we now are attempting to produce surfaces cleaner than anything dreamed of 30 years ago. Various cleaning recipes have been established to produce very clean surfaces. These are generally made up from a few treatments that include ammonia and hydrogen peroxide, hydrochloric acid (usually with hydrogen peroxide), and hydrofluoric acid. No matter which treatments are included in a cleaning process sequence, the final step is normally a rinse in deionized (DI) water, although several alternatives have been described. Miyamoto and co-workers have described using low concentrations of HF in the water. 2 Shiramizu and co-workers have described the use of electrolysis-ionized water containing NH 4 Cl to keep the pH low and remove and keep off metallic impurities. 3 Lampert and Fabry have suggested the addition of CO 2 to water. 4 Loewenstein and Mertens have described the use of HNO 3 in preventing metal deposition. 5 These approaches arise out of a growing understanding that DI water can itself contribute to contamination of the silicon wafer, and that low pH in the rinse water can significantly reduce the metallic contamination on the surfaces of treated wafers.The nature of the adsorbing species has also been an area of study in the literature. Ohmi and co-workers used the segregation of metallic impurities across the surface of a silicon wafer to make inferences regarding adsorbing species. 6 The approach of drying a puddle of water containing dissolved metal ions is difficult to interpret, however, as metal ion concentrations, wafer and solution temperatures, and possibly even the nature of the surface are changing over the 3 h spent drying the puddle. Mouche and co-workers have investigated the mechanism of metal ion adsorption onto silicon. 7 Their work concentration on adsorption from industrially important solutions (SC1, SC2, Caro's acid, dilute HF, and H 2 O) for cleaning silicon wafers. As such, their work lacked a systematic control of pH. Several investigators have used thermodynamic analyses. Helms and co-workers consider the formation of metal silicates, 8 and the phase diagrams of the ternary metal-oxide-silicon systems thus involved. While informative, metal silicates often need temperatures of 800ЊC or higher to form, 9 and so are not accessible at the low temperatures involved in aqueous processing. Building on the work of Pourbaix, 10 Mori and co-workers have concentrated on the changes in the metal species present in solution as the pH changes, combining this with calculations of the free energy change of adsorption for different species. 11 Mori concluded that neutral metal hydroxides are the main adsorption species in SC1 solutions. Loewenstein and Mertens, on the other hand, found that cations are the adsorbing species in acidic solution. 12 All of these studies lack a...