SOLID SURFACES and the Gas-Solid Interface

---

Κ AT Ζ

I -Propanol Adsorption on Germanium

237

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: June 1, 1961 | doi: 10.1021/ba-1961-0033.ch024

Literature Cited (1)Bardeen, J., Phys. Rev. 71, 717 ( 1947 ). (2) Garret, C. G. B., Brattain, W., Brown, W. L., Montgomery, H., "Semiconductor Surface Physics," R. H. Kingston, ed., p. 111, University of Pennsylvania Press, Phila­ delphia, 1957. (3) Green, M., "Progress in Semiconductors," A. F. Gibson, ed., Vol. 4, pp. 35-62, Heywood & Co., London, 1960. (4) Kipling, J. J., Peakall, D. B., "Chemisorption," W. E. Garner, ed., p. 59, Academic Press, New York, 1957. (5) Law, J. T., Meigs, P. S., "Semiconductor Surface Physics," R. H . Kingston, ed., p. 383, University of Pennsylvania Press, Philadelphia, 1957. (6) Ligenza, J. R., J. Phys. Chem. 64, 1017 ( 1960). (7) Many, Α., Gerlick, D., Phys. Rev. 107, 404 (1957). (8) Morrison, S. R., "Semiconductor Surface Physics," R. H. Kingston, ed., p. 169, Uni­ versity of Pennsylvania Press, Philadelphia, 1957. (9) Schwab, G.-M., Ibid., p. 283. (10) Schwab, G.-M., Block, J., Müller, W., Schultze, D., Naturwissenschaften 22, 582 (1957). (11) Schwab, G.-M., Block, J., Schultze, D., Angew. Chem. 71, 101-4 ( 1959). (12) Schwab, G.-M., Mutzbauer, G., Naturwissenschaften 46, 13 (1959). (13) Statz, H., deMars, G. Α., Davis, L., Adams, Α., Phys. Rev. 101, 1272 (1956). (14) Yu, Y.-F., Chessick, J. J., Zettlemoyer, A. C., J. Phys. Chem. 63, 1626 (1959). RECEIVED July 20, 1961. Work supported by the U. S. Signal Corps Research Laboratory, Fort Monmouth, N. J., Contract DA 36-039 SC-85130.

Editors Note. The results reported by Zettlemoyer and associates are explained by a theory which has been very neatly corroborated by work on similar systems at the U . S. Army Signal Research and Development Laboratory. The following discus­ sion is included with the consent of the symposium chairman.

Discussion of Adsorption Studies on Metals X.

1-Propanol on Germanium Powders

M. J. KATZ U.S. Army Signal Research and Development Laboratory, Fort Monmouth, N. J. Results obtained recently by us are entirely consistent with the model advanced here for adsorption on the semiconductor surface. Starting from the theory of surface states, Zettlemoyer et al. are able to ac­ count for changes in chemisorption with the variation in conductivity of a series of samples. As the authors suggest, a direct confirmation of the relationship would require simultaneous measurement on a single crystal of chemisorption and the related electronic factors. This requirement is partially met in our experi­ ments. Taken with the work described above, they provide a complementary aspect of the proposed model. Before entering into a discussion of our results, it is appropriate to point out the essential differences in approach. The Lehigh group measured chemisorption on powdered samples and deduced the electronic interaction with the surface from a plausible model. W e, on the other hand, measured the conductivity and lifetime T

Copeland et al.; SOLID SURFACES Advances in Chemistry; American Chemical Society: Washington, DC, 1961.

238

ADVANCES IN CHEMISTRY SERIES

of excess current carriers on single-crystal germanium as a function of the ambient. The crystals were η-type, about 30 ohm-cm., 0.5 mm. thick, polished and etched in 2 to 1 H N 0 to H F . A number of adsorbates were scanned through, including (X, H 0 , and some compounds of the series, R—X, R — N H , and R — O H . Our pro­ cedure was to pump the sample to a 10~ mm. vacuum and to observe the change in lifetime (average life of injected current carriers) by measuring the decay in photovoltage with exposures to the adsorbate. In some cases, where the reaction seemed reversible, a new exposure was made after pumping at room temperature. Otherwise the sample was flashed at about 5 0 0 ° C. at 10 mm. before re-exposing it. This treatment will not remove all residual oxygen and/or hydroxyl so that our surfaces are similar in this respect to the powder surfaces. It seemed worthwhile to proceed in this way, although any given reaction must be understood in terms of previous sample history. We cannot detail here the diverse effects of reversibility, rates, and pressure dependence of the reactions. Although there was no means of measuring the adsorption directly, we could control the extent of the reaction by varying the ambient pressure. In the present context, the following points should be made: 3

2

2

6

Downloaded by UNIV LAVAL on July 11, 2016 | http://pubs.acs.org Publication Date: June 1, 1961 | doi: 10.1021/ba-1961-0033.ch024

-6

1. Systematic variations in lifetime and conductivity as a function of the ambient were obtained on η-type germanium. 2. Upon exposure to various alkanols and alkyl amines (not exceeding 4 carbons) carrier lifetimes increased up to about 50%, whereas the conductivities increased by 1%. Total conductivity was measured—i.e., bulk plus surface—and, therefore, the magnitude of the change does not reflect the true sensitivity of surface conductivity to the reaction. 3. The lifetime of a p-type crystal decreased on exposure to ethylamine— i.e., a change inverse to that obtained with an η-type crystal. 4. C H S H and alkyl halides produced no change in lifetime. Zettlemoyer has shown elsewhere that alkyl halides adsorb to a negligible extent, if at all, on oxidized germanium surfaces. In the case of C H S H , we have an interesting contrast to that of C H O H , which, as indicated above, produced a pronounced increase in lifetime. As the C H S H would not be expected to react as a donor, it would not tend to liberate trapped electrons. 5. Prolonged pumping on the adsorbed alcohols and amines at room tempera­ ture tended to decrease the lifetime. Thus, even the reversibly sorbed fraction may affect the trapped electrons—perhaps through a dipole-dipole interaction. 3

3

3

3

Essentially, the Lehigh theory envisages electrons trapped at surface cation sites where they will act to inhibit donor-type chemisorption. From our point of view, the process of donor chemisorption tends to release the trapped electron, thereby increasing the conductivity and increasing the lifetime by reducing the occupancy at recombination centers. As already noted, this process is observed during the adsorption of R — O H . The theory also predicts similar results for the adsorption of R — N H , which were in fact obtained, although detailed chemisorp­ tion data are not available for this case. Here again the amino group would be expected to exhibit donor properties. 2

RECEIVED October 17,

1961.

Copeland et al.; SOLID SURFACES Advances in Chemistry; American Chemical Society: Washington, DC, 1961.