Capillary movement of liquid in granular beds in microgravity

0273-l 177(95)00882-9

Printedin Great 0273-1177/% $9.50 + 0.00

CAPILLARY MOVEMENT OF LIQUID IN GRANULAR BEDS IN MICROGRAVITY B. S. Yendler,* B. Webbon,** I. Podolski***

and R. J. Bulat

* Bionetics Corporation/NASA Ames Research Center, Moffett Field, CA 94305, U.S.A. ** NASA Ames Research Center, Moffett FieM, CA 94305, U,S.A *** Moscow institute of Biomedical Problems, 123007 Moscow, Russia t Wisconsin Centerfor Space Automation and Robotics, University of Wisconsin, Madison, WI 53706, U.S.A.

ABSTRACT A more complete understanding of the dynamics of capillary flow through an unsaturated porous medium would be useful for the development of an effective water and nutrient delivery system for the growth of plants in space. An experiment was conducted on the Mir Space Station that used an experimental cuvette called “Capillary Test Bed” to compare fluid migration under terrestrial laboratory conditions by positioning the cuvette such that the hydrostatic force is negated and on Mir Differences in fluid migration in the cuvette were observed with under microgravity conditions. migration being slower in microgravity compared with some ground control experiments. INTRODUCTION One of the technologies that will be required for plant growth in microgravity is controlled delivery of nutrients to plant roots. While efforts are underway to develop such systems, some fundamental scientific data must be collected in order to understand how nutrient delivery systems should be developed. Of particular interest is an understanding of the capillary migration of liquids in microgravity. Experimental predictions there exists propagated movement. which was experiments performed microgravity packed bed 1993

data /l/ have shown that capillary movement in granular beds follows theoretical in the case of small particles. However, in the data for granular beds of large particles, significant deviation from theory 111. Yendler and Webbon /l/.observed that water horizontally through a granular bed overcoming the effect of gravity on the water Under such conditions, the height of the saturation bed was equal to the capillary rise, only on the order of several particle diameters for particles of 1 mm. Therefore, evaluating capillary movement in granular beds with large particles can not be accurately on the ground. Such experiments could be most effectively conducted only under conditions. In this paper, we report on an experiment on water propagation through a of near monodisperse glass spheres conducted on board Mir Space Station in January,

EXPERIMENTAL Spherical glass beads of 1.5 mm (Cataphote, Inc., Jackson,MS) were used in the study. The particles were class II beads of density 2.58 gfcm3. Deionized water (pure and colored by FD&C Red dye No. 40) was used in the ground studies. The surface tension of water with and without dye was compared, and the dye was found to have no effect. Water for the experiment conducted on board Mir was obtained from Mir’s potable water system. (4/5)233

(419234

0. S. Yendler et al.

The Capillary Test Bed (CTB), an experimental cuvette, was used in this experiment. The cuvette has been designed to study steady state, one-dimensional water distribution in granular substrates under microgravity conditions /2/ The CTB is a rectangular box with three transparent walls, one of them (a wide side) has a grid. The CTB has an inlet tube for water injection and air venting. A detailed description of the experimental cuvette is given in Podolsky and Mashinsky 121. A study of capillary movement in granular beds requires that water is been supplied to the granular bed without pressure. The capillary force, which moves the water through the granular bed, is inversely proportional to the diameter of the particle. In the case of particles with diameter of 1.5 mm, which have been used in this experiment, the capillary pressure is equivalent to about 4 mm of water. As a result, even a slight positive or negative pressure in the water supply reservoir would have a major effect on water movement and change the experimental results.

Fig. 1. Reservoir Assembly. 1 - flexible reservoir, 2 - outlet with valve, 3 - plastic case, 4 - strings A collapsible reservoir assembly was essentially no pressure. A sketch of reservoir with a very thin flexible wall, two strings that both keep the reservoir water is leaving the reservoir.

designed and the assembly an outlet with from moving

fabricated that provided water to the CTB with is shown in Fig. 1 The assembly consists of a a valve, a plastic case to protect the reservoir, and towards the outlet and blocking the outlet while

Prior to launch toward Mir, the CTB was loaded with beads and the reservoir was primed with dry dye (FD&C Red dye No 40). The reservoir was not filled with water because Mir Space Station regulations require that the reservoir be launched dry. The CTB and reservoir assembly were launched disconnected. On board Mir, the reservoir was filled with 50 mL of water from the Mir potable water system, colored by the powdered dye as water was added. To initiate the experiment, a cosmonaut connected the reservoir assembly to the CTB and slightly squeezed the reservoir by blowing air through the open end of the plastic case (3). This forced the water to prime the first several layers of beads. The position of the water front, relative to the grid as a function of time was recorded manually by the cosmonaut. Results of the experiment are presented in Fig. 2. 16 14 12 10 6 6 4

0

2

4

6

6

10

12

14

16

16

20

Time (min)

+

-ground,

ground

level dlff.

mm l -

Jan., 93, MIR Space Station

- -&--

ground, data from /I/

Fig. 2 Distance of water propagation versus time in a bed of glass beads of 1.5 mm. (Filled squires and empty circles with solid lines - experimental results of this study)

Capillary Movement

of Liquid in Granular Beds in Micrograwty

(‘t/5 )2?5

Ground experiments were likewise conducted. The CTB and the reservoir assembly were placed horizontally. The water in the reservoir and water in the CTB were at the same level at the beginning of an experimental run, in order to minimize any hydraulic head. The effect of the hydraulic head on capillary movement will be discussed in the following section. Results of multiple runs conducted on ground are shown in Fig. 2. Ground experimental data from mutilple runs /l/ are also shown in Fig. 2. Scattering of ground experimental data in Fig. 2 can the result of small variations of the water level in the supply reservoir. HYDRAULIC

HEAD

In order to estimate the effect of the hydraulic head on water propagation, conducted when water level in the reservoir was at different levels.

Fig. 3 Schematic

diagram of the experiment

conducted

an experiment

was

on the ground.

In the situation depicted in Fig. 3, when the water level in a supply reservoir is at the same level as in a packed bed, a height of a saturated granular bed is equal to a capillary rise. Water column in the saturated bed exerts pressure in tangential to gravity direction. This pressure helps to move water through the bed. Therefore, this hydraulic head must be taken into account, when a water movement in the bed is calculated. In order to qualitatively estimate the impact of this hydraulic head on water propagation, a one dimensional model /3,4/ will be used with the assumption that the hydraulic head equals half of maximum capillary pressure. In the equations /3,4/, it is assumed that the liquid flow is laminar and inertia effects can be neglected. The rate of propagation will then be calculated from: 1 -_= L

zf+;

(1)

:-

where 1 is the distance penetrated by liquid at time t, L and T are length and time scales, and H is the hydraulic head. Expressions for L and T are given in /l/. The length L gives a vertical capillary rise in a packed bed under gravity. Results of calculation (1) for different H/L are shown in Fig. 4. Approximately, the upper curve corresponds to the case with a positive hydraulic head when water level in reservoir is at the ton level of granular bed. The middle curve corresponds microgravity conditions when there is no hydraulic heyad. A case of water level at the level of the bottom of granular bed is depicted by lower curve. Experimental data for similar case is shown in Fig. 2 ( “ground, level diff.“).

(I

2

4

6

Dimensionless

Fig. 4 Dimensionless different

dimensionless

= 0 (microgravity),

hydraulic

distance

of propagation

10

l/L versus

dimensionless

time

tfT

H/L (1 - upper curve H/L = 0.5 , 2- middle curve H/L = -0.5 )

heads

3 - lower curve

8 Time

for H/L

(4/5)236

B. S. Yendler ctal.

Expression (1) allows us to estimate the effect of variations of the water level in the supply reservoir on water propagation in our experiments with the packed bed of 1.5 mm glass beads. An average capillary rise was approximately 6.2 mm in our experiments. Therefore, a variation of 1 mm of the water level will result in 18% change in the distance and the speed of water propagation in our case. Comparison of data in Fig. 2 and 4 shows that the differences between ground and microgravity experiments can not be explained entirely by the effect of a hydraulic head on water propagation. As data in Fig. 4 indicate, a hydraulic head shifts the curve up or down, but does not change the shape of the curve, as is evident in Fig. 2.

CTB MODIFICATION Results from these experiments indicate that some modifications of the CTB are desirable. One such change involves the introduction of the water as a uniform front into a granular bed. If the introduction of water into the bed is not uniform, as in the current CTB, the water front becomes unstable and can break into narrow wetting columns or fingers /5/. The position of the front inside of the bed is not visible, as the bed is not transparent . Hence, the front position in bulk can be only assumed based on a position of the front periphery, which we can observe through the transparent walls of the CTB. A uniform frontal movement of water into the bed results in significant reduction in any experimental error. Another modification of the design is to use a cylinder container for a granular bed . It is recognized that a solid wall affects the structure of a packed bed in equivalent of 10 times the diameter of the particles; that is, the porosity of the packed bed next to the wall tends to be higher than in the bulk. Obviously, this effect becomes even more pronounced in the corners of a rectangular container. For this reason, a cylindrical container should be at least 40 times the particle diameter. An existing design (see Fig. 1) relies too much on the cosmonaut’s ability to introduce the water in a way that only the first layers of particles are wetted and not to flood the CTB. Any new design of the experimental cuvette should provide means for the controlled wetting of the first layers of particles at the entrance of the particle bed when the experiment is initiated.

Ualue

1

Bead Column

1 Uent

Reseruoir

Compression

Plunger

Fig. 5 Conceptual experiments.

design of a capillary

-I

‘Support

test bed (modified)

Screens’

assembly

proposed

for follow on

A conceptual design of a CTB-M that does not have mentioned above disadvantages is shown in Fig. 5. A special valve has a flow cross section which is equal to the bead column internal diameter. This allows water to be introduced into the packed bed uniformly. The reservoir assembly and the bead column have vents to equalize pressure inside of the CTB-M. Support screens hold the particles within the bed. Next to the valve is a hydrophilic screen, so water can flow through this screen. The other screen is made from hydrophobic material in order to allow passage of air and retain water inside of the CTB-M.

(415)237

Capillary Movement of Liquid in Granular Beds in Microgravity

To initiate the experiment, a compression plunger is released so that this squeezes the flexible reservoir, creating a pressure inside of the reservoir. The valve is open then. The pressure pushes out a predetermined amount of water which is enough to fill the void space in the valve and to wet the A capillary force that is exerted in the packed bed pulls water from the first layers of particles. reservoir during the experiment. The experiment

is scheduled

to be conducted

on STS-63, February,

1995.

CONCLUSIONS Experimental results obtained on the Mir Space Station reveal a difference in capillary flow in a granular bed of large particles in microgravity or in 1 g. This difference can be attributed to several conditions including malfunctioning of the CTB. However, the most likely reason for the difference is the effect of gravity on water propagation on the ground even under conditions when the hydraulic head is negligible. It is important to stress, however, that a single experiment can not be used as a basis for drawing any statistically justified conclusion. More flight experiments are planned using the modified CTB to obtain reliable information on water propagation in packed beds of large particles in microgravity. ACKNOWLEDGMENT This work was done partially while B. Yendler was holding a National Research Council-NASA Ames Research Associateship. We would like to express our thanks to J. Carbo of NASA Ames for help in preparation of the experiments. REFERENCES 1. Yendler, No. 932164.

B., and B. Webbon, 23rd International

Capillary Movement of Liquid in Granular Beds. Conference on Environmental Systems, (1993).

SAE Tech. Paper

2. Podolsky, I., and Mashinsky, A., Peculiarities of Moisture Transfer in Capillary-Porous During Space Flight, Advances in Space Research, @( 11); 39-46, (1994) 3. Batten, G.L. “Liquid Imbibition 513,

in Capillaries

Sci. ,102,

(1984)

4. Levin,S., and Neale, G.H., ” The Theory Trans. 2.71, 12, (1975)

5.

and Packed Beds.” , J. Colloid Interface

Substitutes

Raats, P.A.C.

Proc.. 37, 681-685,

Unstable Wetting (1973)

of the Rate of Wetting

Fronts in Uniform

of a Porous Medium”,

and Nonuniform

Faraday

Soils. Soil Sci. Sot. Amer.