Diffusion and sputtering techniques are described for applying low resistance, highly transparent, antireflecting, electrical contacts to CdS. The diffusion process entails heating CdS for several minutes in the presence of indium vapor. The resulting crystal surface resistance has been adjusted to range as low as 12 ohms/square without any observable crystal discoloration. The sputtering process employs an 82–18% indium-tin alloy that is placed onto the crystal surface using oxygen as the sputtering atmosphere. The resulting film is Sn-doped In2O3, measuring 100–150 ohms/square, and showing 80–100% transmittance from 4600 to 20 000 A.

No abstract

Graphic means for fabricating active circuit elements have been examined in an exploratory program. Compatible and complementary semiconductor‐,dielectric‐, and metal‐inks have been developed and tested. Their use in various juxtaposed and overlayed configurations has produced operational insulated‐gate field‐effect transistors. The ultimate goal is to establish processes compatible with existing processes for silk‐screened resistors and capacitors and produce high‐volume, low‐cost active circuits. Four types of graphically produced transistors typically exhibit two characteristic modes of behavior; high transconductance false(≤2000µnormalmhosfalse) coupled with poor frequency response false(<100 normalcpsfalse) ; and low transconductance false(≤20µnormalmhosfalse) coupled with better frequency response false(>100 normalcpsfalse) . These behaviors correlate with gate‐dielectric parameters. Various organic and inorganic gate‐dielectric materials have been tested; best results to date have been obtained with nitrocellulose, silicate cement, glyceryl monostearate, or barium titanate in association with normalCdS:normalCdSe . Sintering of the semiconductor layer and postprocessing of the inks after each printing deposition have been found necessary to enhance device performance. The times involved with these operations are comparable to conventional screened conductor, resistor and capacitor processing periods.

No abstract

values of d/ a, diverging when the range of validity of their method is exceeded. VII. SUMMARY 1. The basic theory of approximate calculation of capacitance by the method of subareas is recast 15 -17 for the problem of a rectangular parallel-plate capacitor and illustrated by application to the technically important case of the square-plate capacitor.2. As manifest by Fig. 3, subarea calculation of the square-plate capacitor, using 36 subareas per conductor, yields values of capacitances very close to the exact values. Additionally, this computation affords the distribution of charge over the surface of the conducting plate, as illustrated in Fig. 4 for d/ a= 1.3. The universal curve of Fig. 5 enables immediate graphical determination of the capacitance of a square parallel-plate capacitor of a wide range of specified dimensions.4. Table III shows a comparison of subarea values with values calculated from the commonly used approximate expression Cn=A~o/4?rd and from the more exact edge-corrected equation advanced by Scott and Curtis and affords knowledge of the relative accuracy of these two equations as a function of the geometrical parameters of the capacitor.A self-refining method for growing three different forms of CdS single crystals is described together with fabrication details of the 13oo-watt three-temperature zone furnace employed. The growth chamber measuring 11 in. in diamX21 in. long is charged with CdS powder and either H 2 S or argon at a pressure of 1 atmosphere or less, and subjected from 4 to 7 days to different temperatures (1250±2°C maximum) and gradients depending on the type of crystal desired. Single crystals vary from thin plates, rods, and whiskers weighing fractions of a gram to lO-g polygonal chunks. The method is believed to be suitable for growing single crystals of many other elemental and compound semiconductors.