The screen cup is made from 200-mesh stainless steel screen with an opening width 

 of 0.0074 cm. (74 microns). Each screen cup is made by cutting pieces of screen 1.8 

 cm. 2 and then rolling the screen over a short piece of 0.475 cm. diameter Teflon rod. 

 The screen overlaps slightly and is then soldered with a high grade silver solder along 

 the seam. The outside diameter of the screen cup is 0.581 cm., and the inside diameter 

 is 0.475 cm., which facilitates a snug fit over the Teflon insert. A circular disk of 

 0.475 cm. diameter screen is s i lver- soldered on one end of the screen to close the 

 tube. After soldering, the screen is boiled in distilled water, followed by a rinse in 

 acetone, and rinsed again in distilled water to remove all traces of solder flux. The 

 screen is then slipped all the way up on the Teflon insert and permanently fixed in 

 place with epoxy resin. The inside volume within which the thermocouple is housed is 

 0.188 cm. 3 . 



Calibration and Measurement Procedures 



Calibration of these instruments is achieved by mounting the complete psychrometer 

 in a small test tube lined with filter paper moistened with various KCC solutions (see 

 Appendix 3 for water potentials of various molal solutions). The filter paper lining 

 should include a disk at the bottom of the test tube (1.1 cm. by 9.9 cm. inside dimen- 

 sions) together with a strip of filter paper around the inside circumference of the 

 tube; the filter paper should extend up the side walls about 1 inch from the bottom 

 (Whatman #1 chromatography paper). About 6 to 8 drops of solution are sufficient to 

 fully moisten the filter paper. The psychrometer is lowered into the test tube until 

 the screen cup containing the thermocouple is below the filter paper lining, and is 

 then sealed with a rubber stopper through which the copper lead wires extend. The 

 stopper and lead wires can be coated with vacuum grease or RTV to insure a tight seal. 



The test tubes containing the psychrometers are then completely immersed into a 

 constant temperature bath at 25°C. Usually temperature and vapor equilibrium can be 

 achieved within 30 minutes (figure 6), but at least 2 hours should be allowed to elim- 

 inate all temperature gradients. Because the thermocouple psychrometers are very tem- 

 perature sensitive, calibration is repeated at several temperatures between 10°C. and 

 35°C. However, it has been found (Wiebe and others 1970) that chromel-constantan psy- 

 chrometers follow a rather simple relation between e.m.f. output vs. temperature. This 

 relation is expressed by: 



0.027T + 0.325 K J 



where CF is a correction factor to be multiplied by the output in /jv at the known 

 psychrometer temperature, T(°C). Equation (5) permits calibrations to be made at 25°C 

 from which water potentials at any other temperature can be determined. In some cases, 

 however, individual thermocouple differences may cause the temperature response to 

 deviate slightly from the theoretical value expressed by equation (5) (Wiebe 1970). 

 Therefore, it is recommended that thermocouple psychrometers be calibrated at several 

 temperatures as described above. 



To obtain actual psychrometer readings, two basic instruments are used: (1) a 

 sensitive microvoltmeter ; and (2) a special circuit used for cooling the thermocouple 

 wet junction. The most commonly used voltmeters include a Hewlett-Packard Model 419A, 

 or the Keithley Model 155 (figure 7) , both of which are portable and permit measure- 

 ments to the nearest 0.1 uv. In addition to these, even more sensitive recording 

 instruments are often used, such as strip-chart recorders with an amplifier and more 

 sensitive microvoltmeters (Appendix 4). A diagram of the cooling circuit ("switchbox") 

 is shown in Appendix 6, together with a list of the component parts. The switchbox 

 (figure 7) is mounted in an insulated aluminum box to prevent rapid temperature fluctua 

 tions. The switchbox permits a direct connection of the thermocouple to either the 

 voltmeter or the cooling circuit. 



13 



