No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Photoemission and Related Properties of the Alkali‐Antimonides
1.For the pruposes of this paper, these particular compounds will be termed the “alkali‐antimonides” although compounds are formed of these materials in different ratios.
2.The chemical composition was established independently by N. S. Zaitsev [J. Tech. Phys. (U.S.S.R.) 9, 661 (1939)]
2.and by A. H. Sommer [Nature 148, 468 (1941);
2.A. H. Sommer, Proc. Phys. Soc. (London) 55, 145 (1943)].
3.P. Görlich, Z. Physik 101, 335 (1936);
3.P. Görlich, Phil. Mag.  25, 256 (1938).
4.P. Görlich, Advances in Electronics and Electron Phys. 11, 1 (1959).
5.A. H. Sommer, Rev. Sci. Instr. 26, 725 (1955).
6.A. H. Sommer and W. E. Spicer, Methods of Experimental Physics, edited by K. Lark‐Horovitz and V. A. Johnson (Academic Press, Inc., New York, 1959), Vol. 6B, Chap. 12.4, p. 376.
7.Since the maximum escape depth of photoelectrons is about 250 A, absorption coefficients greater than are necessary to provide the high yields obtained in good emitters. Such high absorption coefficients are associated only with excitation from one fundamental band to another.
8.L. Apker and E. Taft, J. Opt. Soc. Am. 43, 78 (1953).
9.H. Miyazawa [J. Phys. Soc. Japan 8, 169 (1953)]
9.has suggested that Apker and Taft’s data could be attributed to structure in the valence band of If this were so, the structure which appears at low energies for photons of about 4 ev should move to higher energies with increasing photon energy. This is not the case. The mechanism suggested by Apker and Taft has been so well established in a large variety of semiconductors (see, for example, footnotes 10 and 14) that one must assume that it occurs in photoemitters. Taft and Philipp [E. A. Taft and H. R. Philipp, Phys. Rev. 115, 1583 (1959)] have shown that valence band structure can be identified in the velocity distribution of photoelectrons from but this is not the structure referred to by Miyazawa.
10.K. G. McKay, Phys. Rev. 94, 877 (1954);
10.A. G. Chynoweth and K. G. McKay, Phys. Rev. 102, 369 (1956); , Phys. Rev.
10.A. G. Chynoweth and K. G. McKay, 106, 418 (1957); , Phys. Rev.
10.A. G. Chynoweth and K. G. McKay, J. Appl. Phys. 30, 1811 (1959), and the references given in these papers.
11.B. Senitzky, Phys. Rev. 116, 874 (1959);
11.A. G. Chynoweth, J. Appl. Phys. 31, 1161 (1960).
12.P. A. Wolff, Phys. Rev. 95, 1415 (1954).
13.A. G. Chynoweth and K. G. McKay, Phys. Rev. 108, 29 (1957).
14.J. Tau, J. Phys. Chem. Solids 8, 219 (1959);
14.V. Vavilov, J. Phys. Chem. Solids 8, 223 (1959)., J. Phys. Chem.
15.This is supported by the temperature dependence of photoemission for photon energies greater than those at the threshold. See the reference to Miyazawa’s article in footnote 9.
16.J. A. Burton, Phys. Rev. 72, 531(A (1947) and private communication. Some of Burton’s data are given by V. K. Zworykin and E. G. Ramberg in Photoelectricity (John Wiley & Sons, Inc., New York, 1949), p. 59.
17.A. J. Dekker, Solid State Physics, edited by F. Seitz and D. Turnbull (Academic Press, Inc., New York, 1958), Vol. 6, p. 251.
18.Assuming a mean free path between lattice collisions of about 30 A, this is the sort of escape depth which would be expected for energy loss due to phonon production.
19.W. E. Spicer, Phys. Rev. 112, 114 (1958).
20.As Dekker has shown [A. J. Dekker, Solid State Physics (Prentice‐Hall, Inc., Englewood Cliffs, New Jersey, 1957), p. 432], Eq. (4) is given by diffusion with loss. Hebb [M. H. Hebb, Phys. Rev. 81, 702 (1951)] has pointed out that a more exact approach to the actual physical situation is given by the “age theory,” developed for use in neutron diffusion. Using this, Hebb derived a more exact expression for the escape probability. However, he found that the experimental photoemission data he was treating could be fitted equally well by Eq. (3). A similar conclusion was reached by Lye and Dekker
20.[R.G. Lye and A. J. Dekker, Phys. Rev. 107, 977 (1957)] in considering the secondary emisson form semiconductors.
21.For a prior discussion of these classifications, see E. A. Taft, H. R. Philipp, and L. Apker, Phys. Rev. 110, 876 (1958).
22.The principal dependence of B(hν) on hν would be expected to arise from the fact that only that part of the velocity which is directed perpendicular to the surface is effective in overcoming the surface barrier.
23.L. Apker, E. A. Taft, and J. Dickev, Phys. Rev. 74, 1462 (1948).
24.E. A. Taft, H. R. Philipp, and L. Apker, Phys. Rev. 110, 876 (1958).
25.To be more exact, β will be a function of the initial energy of the excited electron and the integration in Eq. (1) should be carried out for each value of initial energy.
26.N. D. Morgulis, P. G. Borzyak, and B. I. Djatlowizkaja, Izvest. Akad. Nauk S.S.S.R., Ser Fiz. 12, 126 (1948).
27.P. G. Borzyak, J. Tech. Phys. (U.S.S.R.) 20, 923 (1950).
28.G. Wallis, Ann. Physik  17, 401 (1956).
29.Since data indicate appreciable absorption as well as photoconductivity below 1.6 ev [see footnotes 19, 27, 28, and S. Imamura, J. Phys. Soc. Japan 14, 1497 (1959)],
29.there may be an indirect transition a few tenths of an electron volt below 1.6 ev; however, since there are alternative explanations for this absorption and photoconductivity, additional work must be done before a definitive statement can be made. In any case, it appears that the suggestion of Miyazawa (see article cited in footnote 9), that there is an indirect band to band transition at about 0.5 ev, can be ruled out. The study of the indirect optical transition in silicon and SiC [see W. C. Dash and R. Newman, Phys. Rev. 99, 1151 (1955);
29.H. R. Philipp, Phys. Rev. 111, 440 (1958)] suggests strongly that such a transition should have been seen in the photoconductivity measurements if it lay so far below the direct gand gap., J. Phys. Soc. Jpn.
30.Imamura (see reference in footnote 29) obtained values of between 1.6 and 2.2 ev for the “absorption edge” in these materials. This edge may be analogous to that found at 2.1 ev in germanium [see H. R. Philipp and E. A. Taft, Phys. Rev. 113, 1002 (1959)
30.and J. C. Phillips, J. Phys. Chem. Solids 12, 208 (1960)].However, such an absorption edge should not be confused with the band gap. The band gap values obtained by Imamura from his photoconductivity measurements are in fairly good agreement with those of Spicer. Imamura derives a complicated model for the band structure of these materials. As in the case of Miyazawa, the validity of this is very doubtful, since he has to assume that an indirect optical transition is very improbable compared to a direct transition. In the case of Imamura, it is necessary to assume the indirect transition to be about less probable than the direct transition when both are energetically possible.
31.This should just be considered to give, to a good approximation, the general envelope of the vs hν curve in the threshold region, and not as evidence that structure does not exist in the actual vs hν curve. It is worth noting the analogy with the optical absorption data on Wallis (footnote 28) found that the room temperature optical absorption coefficient could be well fitted by a curve of the form but Taft and Philipp (see work cited in footnote 9) found that the structure appeared when the was cooled to about 90 °K. Similarly, Eq. (5) should just be expected to hold in the threshold region; otherwise, the absorption coefficient would increase without limit as hν increases. Fortunately, the yield is not strongly dependent on where is comparable to i.e., away from the threshold.
32.One might expect B to have at least a photon energy dependence due to the fact that only the velocity directed perpendicular to the surface barrier is available to overcome that barrier (see footnote 22); however, since the electrons which arrive at the surface will be losing energy principally through phonon production, an electron which has greater than the minimum escape energy strikes the surface several times before its energy is reduced below that of the surface barrier.
33.G. Brauer and E. Zintl, Z. physik Chem. 37B, 323 (1937).
34.A. Solomon (private communication).
35.G. Gnutzmann, Ph.D. dissertation, University of Münster (1953).
36.K. H. Jack and M. M. Wachtel, Proc. Roy. Soc. (London) 239A, 46 (1957);
36.J. J. Scheer and P. Zalm, Philips Research Repts. 14, 143 (1959).
37.W. H. McCarroll, J. Phys. Chem. Solids (to be published).
38.S. Imamura, J. Phys. Soc. Japan 14, 1491 (1959).
39.W. H. McCarroll is now studying the crystal structures of samples taken from sensitive cathodes.
40.T. Sakata, J. Phys. Soc. Japan 8, 125, 272 (1953);
40.T. Sakata, 9, 1030, 1031 (1954)., J. Phys. Soc. Jpn.
41.In this paper, the term “defect level” will be used for any energy level introduced into a semiconductor by a lattice imperfection whether by a foreign atom, vacancy, or any other type of defect.
42.By assuming the conductivity to be due to a single level with no compensation, Sakata obtained values for his defect concentrations between and 40 However, the scheme of defect levels may be much more complicated.
43.It should be noted that in the paper of Jack and Wachtel (footnote 36), no experimental evidence is given for an excess of cesium.
44.N. S. Khlebnikov, J. Tech. Phys. (U.S.S.R.) 17, 333 (1947);
44.K. Miyake, J. Appl. Phys. 31, 76 (1960).
45.A. H. Sommer, J. Appl. Phys. 29, 1568 (1958).
46.At present, it is not certain that the photosensitive cathode has this type of structure. See discussion in Sec. V.
47.Such an analogy may be misleading, since the difference in electronegativity of the components is considerably smaller for compounds such as than it is for See, for example, F. A. Kroger and H. J. Vink, in Solid State Physics, edited by F. Seitz and D. Turnbull (Academic Press, Inc., New York, 1956), Vol. 3, p. 366.
48.In agreement with the previous work on [P. G. Borzyak, Izvest. Akad. Nauk S.S.S.R., Ser. Fiz. 9, 173 (1941); Proc. Conf. on Cathode Electronics, Acad. Sci. UkSSR (1952), p. 18;
48.R. Suhrman and C. Kangro, Naturwissenschaften 40, 137 (1953)].
49.There is no evidence for the random distribution of alkali and antimony atoms in the multi‐alkali lattices.
50.For a more complete discussion of the influence of defect levels on photoemission see W. E. Spicer, RCA Rev. 19, 555 (1958).
51.The depth of photoemission is here taken to be the maximum depth from which appreciable emission is obtained for radiation incident from the vacuum. This will be determined by either the thickness of the semiconductor necessary to absorb most of the light, or by the depth from which an excited electron can escape. The depth of escape will be given by the smaller of the above parameters.
52.The values of electron affinity which will be reported here and which have been reported previously should be understood as “apparent electron affinities” because of the difficulty in separating electron affinity effects from that of band bending. However, we will continue to term our values “electron affinities” until the separation is possible. Since there is good agreement on the threshold of photoemissive response in these materials, it is clear that the electron affinity is reproducible.
53.This last proposal has been made previously by Imamura see footnote 38.
Article metrics loading...
Full text loading...
Most read this month
Most cited this month