US3661724A

US3661724A - Closed loop hygrometry

Info

Publication number: US3661724A
Application number: US25208A
Authority: US
Inventors: Allen Strickler
Original assignee: Beckman Instruments Inc
Current assignee: Beckman Coulter Inc
Priority date: 1970-04-02
Filing date: 1970-04-02
Publication date: 1972-05-09
Anticipated expiration: 1989-05-09

Abstract

Method and apparatus for determining water vapor concentration in a fluid. The apparatus includes a hygrometer having an inlet and outlet and a pump having an inlet and an outlet. A layer of semipermeable barrier material permeable to moisture and having two surfaces, the first surface being exposed to the fluid, is provided. Conduit means connects the pump, hygrometer, and the second surface of the barrier in series so that a carrier gas may be circulated by the pump in a closed loop across the second surface of the barrier and through the hygrometer. Water vapor may pass through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier, whereby the water vapor concentration in the fluid can be determined from the proportional water vapor concentration measured by the hygrometer. The method of determining water vapor concentration in a fluid comprises exposing the first surface of a layer of semipermeable barrier material having two surfaces to the fluid. A carrier gas is circulated in a closed loop across the second surface of the barrier and through a hygrometer so that water vapor passes through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier. The water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer.

Description

United States Patent Strickler 1 May 9, 1972 [5 1 CLOSED LOOP HYGROMETRY 57 ABSTRACT [72] Inventor: Allen Strickler, Fullerton, Calif. Method and apparatus for determining water vapor concentration in a fluid. The apparatus includes a hygrometer having {73] Asslgnee Beckman Instruments an inlet and outlet and a pump having an inlet and an outlet. A 22] Filed: A r. 2, 1970 layer of semipemieable barrier material errneable to l p P Appl. No.: 25,208

Primary Examiner-T. Tung Attorney-William F. McDonald and Robert J. Steinmeyer moisture and having two surfaces, the first surface being exposed to the fluid, is provided. Conduit means connects the pump, hygrometer, and the second surface of the barrier in series so that a carrier gas may be circulated by the pump in a closed loop across the second surface of the barrier and through the hygrometer. Water vapor may pass through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier, whereby the water vapor concentration in the fluid can be determined from the proportional water vapor concentration measured by the hygrometer. The method of determining water vapor concentration in a fluid comprises exposing the first surface of a layer of semipermeable barrier material having two surfaces to the fluid. A carrier gas is circulated in a closed loop across the second surface of the barrier and through a hygrometer so that water vapor passes through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier. The water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer.

25 Claims, 11 Drawing Figures PATENTEDMM 9 I972 3 661 .724

sum 1 OF 4 FIG. I

\ 38 g E 5s ALLEN STRICKLER FIG. 2

ATTORNEY PATENTEUMM 9 I912 8,661,724

SHEET 2 [1F 4 FIG. 4

m -E 5 I o (T) an E Q .1

M E 33% RH (AMB.ATMOS.) l0: 0 25% RH O r l INVENTOR. 5 ALLEN STRICKLER ATTORNEY PATENTEDMAY 9 1972 3,661,724

SHEET 3 BF 4 u .5 x 6 I o E I O O '.1-r:'-r-:':':"':

IO 5o 8O I00 PER CENT RH FIG. 6

AMBiENT ATMOS.

(24.5% RH I o APPROX.) y 25 RH 2 /SAMPLEOFF g I V WWW. A r I I r wwwm I .7 'PPM' I no FEM FIG. 7

INVENTOR. ALLEN STRICKLER BYmwWg2/%% ATTORNEY PATENTEDHAY 91912 3,661,724

' saw Mr 4 v 1 HOUR AMB.ATMOS. 52% RH (74 5F) NO SHIELD CHART DIVISIONS AMB.ATMOS. 52% RH (74.5F)

WITH SHIELD FIG. 8

AMB. ATMOS. 55% RH (75F) NO SHIELD V NO SHIELD WITH SHIELD FIG. 9

CHART DIVISIONS FLOW RATE: FLOW RATE: IOO ML/MIN 5O ML/MIN FIG. IO

25% RH (74 F) CHART DIVISIONS -uo MIN-- CHART DIVISIONS F INVENTOR.

BY ALLEN STRICKLER FIG. 11

ATTORNEY CLOSED LOOP HYGROMETRY BACKGROUND OF THE INVENTION The instant invention relates to the field of hygrometry. l-Iygrometers, such as electrolytic hygrometers of the type commonly called Keidel cells, are widely used for determining the moisture content of water vapor concentration in fluid streams in industrial processes in which the presence of even minute percentages of moisture has great significance. This type hygrometer has numerous advantages over other moisture determining devices, particularly when used to monitor the water content of continuous process streams. The electrolytic hygrometer is quite selective to water, has a rapid speed of response and is completely quantitative over its operable range of moisture concentration. This eliminates the need for frequent calibration and standard samples. In general, however, the levels of humidity for which this type hygrometer is useful are in the range of to 1,000 parts per million by volume.

The advantages of this instrument, based on an absolute coulometric principle as described in detail in U. S. Pat. No. 2,830,945, have not heretofore been available for measuring ambient atmospheric humidity where the encountered water concentration range is about 0lO0,000 parts per million or the flow rate of water vapor through the hygrometer is more than about 100 micrograms water per minute. At high moisture levels the partially hydrated phosphorous pentoxidephosphoric acid coating in the cell is converted to a relatively thick film of liquid phosphoric acid. This condition leads to a loss of the coating or breakthrough and renders the instrument inoperative. Other hygrometers which are based on other operating principles are suitable for high moisture levels. However, these hygrometers have their own limitations. They tend to be insensitive at low vapor levels. They frequently require large samples. They are inconvenient to use in some environments. They may lack precision and accuracy and may show long response and recovery times and frequently are not self-regenerating.

SUMMARY OF THE INVENTION It is an object of the instant invention to provide a method of, and apparatus for measuring moisture or water vapor concentrations in a fluid varying from 0-l00 percent relative humidity over a wide temperature range. Advantageously, the system according to the instant invention may be used with high accuracy with the Water concentration varying from 0-l 0 parts per million. It is an advantage of the instant invention that water vapor in a fluid which lies over or around materials in a liquid or solid state may be measured, in a range of less than 1 to more than parts per million. This measure can then be correlated with the water content of the liquid or solid sample. It is a further advantage of the instant invention that it is useful over a wide humidity range and at the same time has all the advantages of the conventional electrolytic hygrometer, as described in detail in U. S. Pat. No. 2,830,945 and known as a Keidel cell.

The apparatus for determining water vapor concentration in a fluid, according to the instant invention, includes a hygrometer having an inlet and an outlet and a pump having an inlet and an outlet. The first surface of a layer of semipermeable barrier material, permeable to moisture and having two surfaces, is exposed to the fluid. Conduit means connects the pump, hygrometer, and the second surface of the barrier in series so that a carrier gas may be circulated by the pump in a closed loop across the second surface of the barrier and through the hygrometer. Water vapor may pass through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier. The water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer. In this way, the water vapor concentration to which the hygrometer is exposed is confined to a low range within which the hygrometer performs best with respect to accuracy, linearity, stability and long life.

The method of determining water concentration in the fluid, according to the instant invention, comprises exposing a first surface of a layer of semipermeable barrier material having two surfaces to the fluid. A carrier gas is circulated in a closed loop across the second surface of the barrier and through a hygrometer. Water vapor passes through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier, whereby the water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic representation of a water vapor concentration measuring system according to the instant invention.

FIG. 2 is an elevational view, in cross section, of a sensor head useful in the practice of the instant invention.

FIG. 3 is an elevational view, in cross section, of a shield which may be used in conjunction with the head of FIG. 2.

FIG. 4 is a schematic representation of another system embodying the principles of the instant invention.

FIG. 5 is a graphical representation of various humidity levels as measured in accordance with the instant invention.

FIG. 6 is a graphical representation of humidity and hygrometer signal in accordance with the instant invention.

FIG. 7 is a graphical presentation showing the sensitivity of the instant invention to small changes in water vapor concentration in a fluid.

FIG. 8 is a graphical presentation showing the effects of a shield around the sensor head.

FIG. 9 is similar to FIG. 9 showing the use of a different shield.

FIG. 10. is a graphical presentation showing the effect of the flow rate on the signal to noise ratio in an electrolytic hygrometer used in accordance with the instant invention.

FIG. 11 is a graphical presentation of water concentration varying in accordance with a biological experiment.

In all FIGS, like reference numerals have been used for corresponding parts.

DETAILED DESCRIPTION Referring to FIG. 1, the apparatus includes a hygrometer 10, for example of the Keidel type, having an inlet 12 and an outlet 14. A first conduit section 16 connects the inlet end 12 of hygrometer 10 with a sensor head 18. A second conduit section 20 connects the outlet 14 of hygrometer 10 with the inlet 22 of a pump 24 having an inlet 22 and an outlet 26. A third conduit section 28 connects the outlet 26 of pump 24 with sensor head 18. The three conduit sections, 16, 20, 28, constitute a conduit means.

In FIG. 2, it may be seen that sensor head 18 includes a base member 30 having

passages

32 and 34 therein.

Passages

32 and 34 are in communication with one another by means of cylindrical bore 36 in base 30. A set 38 is provided in cylindrical bore 36. When sensor head 18 is assembled, a layer of semi-permeable barrier material 40 having a first surface 42 and a second surface 44 rests against seat 38. Diaphragm 46 having an aperture 48 supports barrier material 40. The barrier material 40 and diaphragm 46 are sealed in position against seat 38 by washer 50 and locknut 52 which is adapted to be threadably engaged in cylindrical bore 38. As shown, second surface 44 of barrier 40 is connected by means of first conduit section 16 with the inlet l2'of hygrometer 10 and also by means of third conduit section 28 with the outlet 26 of pump 24. Thus, the conduit means, conduit sections, 16, 20, and 28, connect second surface 44, hygrometer 10, and pump 24 in series. At the same time, the first surface 42 of barrier 40 is exposed through aperture 48 in diaphragm 46 to the fluid in which the water concentration is to be determined. The sensor head 18 may be used as shown when the water concentration in a fluid is to be measured directly. The head 18 is simply exposed to the mass of the fluid. If desired, the water concentration in a discrete sample may be measured. For this purpose a cup 54 is provided which may be threaded over sensor 18. If the fluid is so turbulent that the turbulence is affecting the measurements, a shield 56, as shown in FIG. 3, may be utilized to protect the first surface 42 of barrier 40. Shield 56 is shown as a cylindrical sleeve which may be threadably engaged with sensor head 30 in the'same manner as cup 54.

As shown in FIG. 4, if desired, a dryer 58 may be inserted in third conduit section 28 between pump 24 and sensor 18 to dry the circulating carrier gas in the system.

A carrier gas is circulated by pump 24 in the closed loop fonned by the conduit sections, 16, 20, and 28, and pump 24, sensor 18, and hygrometer 10, passing across the second surface 44 of the barrier 40 and through hygrometer 10. Water vapor may pass through barrier 40 from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across barrier 40. The water vapor concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer. The carrier gas may, and usually will equilibrate in respect to total pressure across barrier 40 to attain the same pressure as that in the fluid space in which the water vapor is being measured. Thus there will be pressure equilibrium across the barrier, except with respect to the water vapor pressure, the water vapor in the loop being continuously consumed by the hygrometer.

In a first series of tests, diffusion barrier 40 was a disc of microporous polyvinylidene fluoride, 0.006 inch thick. The exposed area of the disc was controlled by using a small aperture 48 in diaphragm 46 which limited the amount of vapor entering the carrier gas stream. In this way the operating limit of the Keidel hygrometer of approximately 1,000 parts per million of water assuming a typical flow rate of 100 ml per minute, or 100 micrograms water per minute, was not exceeded and the water concentration in the circulating gas could be limited to no more than about half this value. This is desirable because unlike most electronic instruments, the electrolytic cell shows an increase in relative noise level with increased signal. By operating a cell at a maximum of 500 to 750 parts per million water level the signal is quieter and the cell life is prolonged. A series of constant humidity salt solu tions were prepared, as shown in Table I. Saturated salt solutions have the advantage that a wide range of humidity levels may be accurately set and maintained. They are quite accurate so long as the ambient temperature is maintained constant.

Table I shows the salt solutions used and the corresponding relative humidities.

TABLE I FIG. 5 is a graphical presentation of the recording of salt solution humidities ranging from 25 to 76 percent relative humidity. Distilled water was used for the 100 percent relative humidity level and the ambient atmosphere for the 33 percent relative humidity level. These tests were run by inserting the salt solution in cup 54 of the apparatus which was used in the closed loop as shown in FIG. 4 with a second electrolytic cell as dryer 58. It will be noted that in each case, reading from right to left, the hygrometer rapidly reached a signal level which could be correlated with the relative humidity of the solution in cup 54. FIG. 6 is a plot of the relative humidity vs.

scale amplitude based on FIG. 5. The linearity achieved is striking, with only a slight departure from linearity being seen at the highest or percent relative humidity level. The carrier gas used was dry air. The area of the barrier 40 exposed through the aperture 48 in diaphragm 46 was 0.040 inch.

The system was arranged so that total air pressure on the two sides of the barrier was equalized. There remained, however, a partial pressure difference of water vapor across barrier 40. The partial pressure of water vapor was essentially zero in the carrier gas stream as it entered space 30 and proceeded across the second surface 44 of barrier 40. To ensure that the dry side of the gas stream stayed dry, a drying element 58 was inserted in series with pump 24 and the inlet to sensor base member 30. Both chemical drying tubes and a second electrolytic cell were used as the drying element and both were successful.

In another series of tests, a different material, woven stainless steel screening of 200 by 1,400 mesh was utilized. The screen had a nominal pore size of 5 microns. The rate of water vapor passing therethrough was controlled by reducing the size of the aperture 48 in diaphragm 46. The moisture arising from a variety of tobaccos and foodstuffs in cup 54 was detected and measured in terms of equilibrium relative humidity. It was possible to determine the difference in moisture content of cigarettes from a fresh package and those from the same package which had been opened for several hours. Differences in moisture in several brands of pipe tobacco were also detected.

It was also determined in these tests that the system according to the instant invention functions independently of flow rate so long as the flow rate is stabilized and held constant after either increasing or decreasing. Otherwise, a varying flow rate produces pressure changes which generate a pumping action across the barrier which in turn alters the mechanism from diffusion to mass transfer of water vapor.

In one test a cup 54 was in position containing a 25 percent relative humidity sample. The cup 54 was then removed and the system immediately responded to the change in humidity and recorded the relative humidity of the ambient atmosphere. The results may be seen in FIG. 7 where the curve, reading from right to left, moved up as the instrument sensed the 25 percent humidity sample. The sample was then removed and the curve dropped down to indicate the 24.5 percent relative humidity of the atmosphere. It is readily apparent that the trace in this area is one-half a division below the 25 percent relative humidity trace. This illustrates the instant invention is capable of sensing and measuring slight changes in relative humidity or water concentration.

Similarly, a beaker of water placed about 2% inches below barrier 40 was sensed in about 15 seconds; The signal stabilized at about 50 percent relative humidity in about 60 seconds. The beaker was removed and the ambient humidity again correctly registered in about 1% minutes. When the palm of the human hand was lightly pressed against the opening of cylindrical bore 36 a moisture concentration was registered of more than twice that over water at 25 C. This may not only have been due to the moisture evaporatingfrom the hand but also to the higher temperature of the human body. This may alter the operating temperature of the barrier, and since the diffusion constant of water vapor in air is temperature dependent, this can affect the vapor transfer rate at a given vapor pressure differential. Thus, if there is a change in temperature at the barrier, the hygrometer reading has to be corrected for the temperature at which the measurement is made. Throughout the immediately preceding series of tests, the diffusion screen was of the stainless steel mesh type described previously. It was found that with this type of barrier 40, a seven thirty-second inch (0.219 inch) diameter aperture 48 in diaphragm 46 set the 100 percent relative humidity level at 750 parts per million across the hygrometer.

As can be seen from the previous examples, important factors in the successful practice of both the method and apparatus of the instant invention are the porosity and dimensions of the barrier material 40 separating the sample vapor from the near zero humidity of the circulating gas stream. By choosing an appropriate material and dimensions, the rate of diffusion may, within limits, be made to assume almost any desired value. However, the permitted diffusion rate is not the only important property of the barrier. Water adsorption or absorption, chemical stability and dimensional stability of the diffusion element all must be considered in selecting a barrier material. Since all the desired properties are not available in any one material, a compromise product must be utilized.

in investigations, true organic membranes of cellophane and dialysis film, and silicon rubber all passed water vapor and effectively protected the electrolytic cell from particulate matter. These materials, however, showed relatively long response times due to water absorption and hold-up in the body of the membrane. Sintered polyethylene was tested and found to be generally satisfactory but over a period of time tended to change shape around the seal with diaphragm 46. The 200 X 1,400 mesh stainless steel screen with a nominal. pore size of 5 microns was utilized and found quite satisfactory. However, the porosity was so high that the area of admittance in diaphragm 46, i.e., the size of aperture 48, was necessarily reduced to a very tiny hole, 0.008 inch diameter, in order to limit the flow rate of water vapor across barrier 40 to not more than about 100 micrograms water per minute into the carrier gas stream. Very small apertures 48 can result in variations in response and reproducibility.

in the next series of tests. an organic microporous filtering membrane of pure polyvinylidine fluoride, 0.006 inches thick was again used. The average pore size of the material was only 0.45 microns with a maximum of 1.5 microns. The material is hydrophobic, having a moisture pickup of 2 micron grams per liter square inch, is chemical resistant, and very stable up to a temperature of about 300 F. The material is fragile but was adequately protected by diaphragm 46 having an aperture, 48, diameter of 0.400 inch. During these tests a chemical drying agent was used in dryer 58 rather than the electrolytic cell employed in previous tests. The drying agent was packed magnesium perchlorate. This drying agent was as satisfactory as the electrolytic cell, and the signal-to-noise ratio in hygrometer improved since one source of noise, the drying electrolytic cell, was eliminated. 56

in this series of ambient atmosphere tests it was found that unless the ambient atmosphere was very quiet, the hygrometer readout showed a noisy signal, in marked contrast with the low noise signal of discrete samples positioned in cup 54. Accordingly, a shield of 56 in the form of a short collar, as shown in FIG. 3, was attached to sensor base member 30. This did suppress the atmospheric noise. However, the signal amplitude fell 10 percent from its average value. A sample cup 54 placed next to shield 6 showed an unchanged diffusion rate for the system. Substituting an open mesh screen shield for the collar type shield 56 made no difference in the results. For example, as shown in FIGS. 8 and 9, when there was a no-shield signal having a 46 to 52 chart division amplitude, the system with a collar type shield stabilized at around a 36 amplitude. Similarly, a 54-58 signal stabilized at around 53 with a screen type shield.

The reason for this is not fully understood. Two hypotheses have been advanced. First it is hypothesized that the noise represents variable depletion of moisture at the outer boundary layer of the barrier in the turbulent air and that very quiet air might depress the signal. Secondly, a so-called microphonic effect might occur. The diffusion membrane rises and falls with changing pressures resulting from intense acoustic sources, or from turbulent air, generating a pumping action. The latter hypothesis was investigated. A barrier 40 of microporous vinyl backed with a Teflon diaphragm 46 having only a 0.040 inch aperture 48 was used. The curious effect vanished. While this result does not exclude a contribution of the other effect, it does provide some support for the microphone hypothesis for this phenomenon.

The examples show that the instant invention is independent of and functions well regardless of the flow rate of the carrier gas stream within wide limits. The signal amplitude is constant from 20 to 400 milliliters per minute. However, it is desirable to avoid very slow or very fast flow rates. As may be seen in FIG. 10, even at low moisture levels, there is increased noise pickup at flow rates below 50 milliliters per minute. As shown in FIG. 10, when the flow rate is doubled to about milliliters per minute, the signal-to-noise ratio improves. These effects are even more pronounced at high vapor levels such as 100 percent relative humidity. However, if the flow rate is too fast, some electrolytic cells show breakthrough, i.e., the cell fails to electrolyze all the moisture which is passing through it. Some accuracy of the instrument is therefore lost although it may still be precise in the sense that it gives repeatable and relatively correct readings for sample changes. Another reason for avoiding very high flow rates became evident when a cell was subjected to a flow rate of 600 milliliters per minute. The cell lost almost all sensitivity and thereafter showed excessive breakthrough. Apparently the fast flow of the circulating gas blew the coating right out of the cell. Experimentation with various flow rates indicates a desirable flow rate is 100 :25 milliliters per minute. This rate satisfactorily smooths the signal and at the same time preserves the coating in the hygrometer cell.

As shown in the above examples, the diffusion rate is conveniently controlled by the size of the aperture 48 in the diaphragm 46 pressing against the barrier material 40. An aperture of 0.100 inches in diameter in conjunction with the polyvinylidene fluoride barrier material described, allows the transfer of about 300 parts per million water at 100 percent relative humidity at a rate of not more than about 100 micrograms water per minute. The aperture 48 may be enlarged to increase the span of low vapor samples and reduced to measure high vapor samples. In general, the 0.100 inch opening was satisfactory to cover a range of 0 to 33,000 parts per million water in the fluid.

In the tests with the constant humidity salt solutions it was found that the instant invention is capable of reacting rapidly to both small and large step changes in vapor pressure. Sixtythree percent of the signal equilibration level for large step changes was consistently reached in 45 seconds. Small changes, illustrating sensitivity, were consistently observed by the system almost immediately, as shown in FIG. 7.

In another test a severed fresh green leaf was taped to the open side of sensor head 18. As shown in FIG. 11, the system immediately responded to the presence of the green leaf and its transpiration by a rapidly rising trace. The trace then dropped back to normal as the rate of transpiration of the leaf decreased to zero with time. It was further found the moisture content of tobaccos and dehydrated foods can be inferred from the water vapor arising from such materials.

It is to be understood that various changes and variations can be made in the foregoing method and apparatus without departing from the spirit and scope of the instant invention. Accordingly, the instant invention is not to be limited thereby but only by the claims wherein what is claimed is:

1. Apparatus for determining water vapor concentration in a fluid which comprises:

a. a hygrometer having an inlet and an outlet;

b. a layer of semipenneable barrier material permeable to moisture and having two surfaces, the first surface being exposed to the fluid;

c. a pump having an inlet and an outlet;

d. a conduit means connecting the pump, hygrometer, and the second surface of the barrier in series and in a closed loop that a carrier gas may be circulated by the pump in the closed loop across the second surface of the barrier and through the hygrometer, the total pressure of the carrier gas reaching equilibrium with the total pressure of the fluid and water vapor may pass through the barrier from the fluid into the carrier gas at a transfer rate proportional 4. The apparatus of claim 1 wherein the barrier material has an average pore size of about 5 microns. 5. The apparatus of claim 1 wherein the barrier material has an average pore size of about 0.45 microns.

6. The apparatus of claim 5 wherein the maximum pore size is about 1.5 microns.

7. The apparatus of claim 1 wherein the barrier material is a woven metal screen.

8. The apparatus of claim 1 wherein the barrier material is an organic membrane.

9. The apparatus of claim 8 wherein the membrane is a material comprising polyvinylidene fluoride.

10. The apparatus of claim 1 including means for shielding the first surface of the barrier material from turbulence in the fluid to which it is exposed.

11. The apparatus of claim 1 including an apertured diaphragm supporting the barrier material.

12. The apparatus of claim 11 wherein the size of the aperture is such that the flow rate of water across the barrier is not more thanabout 100 micrograms water per minute.

13. The apparatus of claim 11 wherein the size of the aperture is such that the water vapor concentration in the carrier gas will be not greater than about 750 parts per million.

14. The apparatus of claim 11 wherein the size of the aperture is such that the water vapor concentration in the carrier gas will be not greater than about 300 parts per million.

15. The apparatus of claim '11 wherein the diameter of the aperture is from about 0.008 inch to about 0.219 inch.

16. The apparatus of claim 15 wherein the diameter of the aperture is about 0.100'inch.

17. The method of determining water vapor concentration in a fluid which comprises:

a. exposing a first surface of a layer of semipermeable barrier material having two surfaces to the fluid;

b. circulating a carrier gas in a closed loop across the second surface of the barrier and through a hygrometer so that the total pressureof the carrier gas reaches equilibrium with the total pressure of the fluid and water vapor passes through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier, whereby the water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer.

18. The method of claim 17 including the additional step of drying the circulating carrier gas.

19. The method of claim 17 wherein the pressure of the circulating gas across the second surface of the barrier is maintained at the same level as the pressure of the fluid on the first surface of the barrier.

20. The method of claim 17 including the additional step of shielding the first surface of the barrier from turbulence in the fluid.

21. The method of claim 17 wherein the flow rate of the circulating gas is maintained from about 10 to about 400 milliliters per minute. i

22. The method of claim 21 wherein the flow rate of the circulating gas is maintained from about 75 to about 125 milliliters per minute.

23. The method of claim 22 wherein the flow rate of the circulating gas is maintained at about milliliters per minute.

24. The method of claim 17 including the additional step of controlling the passage of water vapor through the barrier so that the concentration of water vapor in the carrier gas is not greater than about 300lparts er million.

25. The method of c aim 1 including the additional step of controlling the passage of water vapor through the barrier to a flow rate of not more than about 100 micrograms water per minute.

* l k l l

Claims

2. The apparatus of claim 1 wherein the hygrometer is an electrolytic hygrometer.
3. The apparatus of claim 1 including means for drying the carrier gas connected between the pump outlet and the barrier.
4. The apparatus of claim 1 wherein the barrier material has an average pore size of about 5 microns.
5. The apparatus of claim 1 wherein the barrier material has an average pore size of about 0.45 microns.
6. The apparatus of claim 5 wherein the maximum pore size is about 1.5 microns.
7. The apparatus of claim 1 wherein the barrier mAterial is a woven metal screen.
8. The apparatus of claim 1 wherein the barrier material is an organic membrane.
9. The apparatus of claim 8 wherein the membrane is a material comprising polyvinylidene fluoride.
10. The apparatus of claim 1 including means for shielding the first surface of the barrier material from turbulence in the fluid to which it is exposed.
11. The apparatus of claim 1 including an apertured diaphragm supporting the barrier material.
12. The apparatus of claim 11 wherein the size of the aperture is such that the flow rate of water across the barrier is not more than about 100 micrograms water per minute.
13. The apparatus of claim 11 wherein the size of the aperture is such that the water vapor concentration in the carrier gas will be not greater than about 750 parts per million.
14. The apparatus of claim 11 wherein the size of the aperture is such that the water vapor concentration in the carrier gas will be not greater than about 300 parts per million.
15. The apparatus of claim 11 wherein the diameter of the aperture is from about 0.008 inch to about 0.219 inch.
16. The apparatus of claim 15 wherein the diameter of the aperture is about 0.100 inch.
17. The method of determining water vapor concentration in a fluid which comprises: a. exposing a first surface of a layer of semipermeable barrier material having two surfaces to the fluid; b. circulating a carrier gas in a closed loop across the second surface of the barrier and through a hygrometer so that the total pressure of the carrier gas reaches equilibrium with the total pressure of the fluid and water vapor passes through the barrier from the fluid into the carrier gas at a transfer rate proportional to the difference in partial water vapor pressure across the barrier, whereby the water concentration in the fluid can be determined from the proportional water vapor concentration as measured by the hygrometer.
18. The method of claim 17 including the additional step of drying the circulating carrier gas.
19. The method of claim 17 wherein the pressure of the circulating gas across the second surface of the barrier is maintained at the same level as the pressure of the fluid on the first surface of the barrier.
20. The method of claim 17 including the additional step of shielding the first surface of the barrier from turbulence in the fluid.
21. The method of claim 17 wherein the flow rate of the circulating gas is maintained from about 10 to about 400 milliliters per minute.
22. The method of claim 21 wherein the flow rate of the circulating gas is maintained from about 75 to about 125 milliliters per minute.
23. The method of claim 22 wherein the flow rate of the circulating gas is maintained at about 100 milliliters per minute.
24. The method of claim 17 including the additional step of controlling the passage of water vapor through the barrier so that the concentration of water vapor in the carrier gas is not greater than about 300 parts per million.
25. The method of claim 17 including the additional step of controlling the passage of water vapor through the barrier to a flow rate of not more than about 100 micrograms water per minute.