Drying of a model soil

Drying experiments have been carried out with model soils made of different pastes filling granular packings. A detailed information concerning the time evolution of the water saturation distribution inside the sample was obtained from magnetic resonance imaging measurements. This study makes it possible to understand the physical origin of the drying characteristics of these materials. The drying curves exhibit a constant-rate period (CRP) and a falling-rate period (FRP) but the relative durations of these periods depend on the paste structure. With a kaolin suspension the CRP lasts down to very low water densities and is associated with a homogeneous drying of the paste throughout the sample. With a bentonite suspension the CRP is shorter and the drying in the FRP results from a complex process involving fractures progressing downward through the pasty matrix. With a gel the CRP period is even shorter and the drying in the FRP results from the progression of a dry front through the packing as a result of the shrinkage of the gel matrix. This provides an overview of the main possible processes at work when drying a soil as a function of its components along with some practical means for slowing down drying from soils.

Figure 1

Apparent water density or saturation profiles in the bentonite sample in Fig. 11a at the initial time (filled squares) and after 21 days of drying (open squares): (a) apparent density as measured by NMR; (b) apparent saturation after correction for the sample or magnetic field heterogeneities; and (c) after additional correction for relaxation effects.

Figure 2

Ratio of the effective water saturation (obtained from sample weighting) to the NMR signal saturation as a function of the NMR signal saturation for the kaolin (squares) and bentonite samples in Fig. 11a (open circles) and Fig. 11b (filled circles). The inclined dotted line corresponds to a power-law model with a power of $-2/3$.

Figure 3

Theoretical saturation profiles in the sample for case (a) (left) and case (b) (right) as described in the text.

Figure 4

Drying curves of analogous samples (height: 57 mm; average bead diameter: $300\mu \text{m}$) with different interstitial pastes: kaolin suspension (squares); bentonite suspension (circles); and gel (stars).

Figure 5

Drying curves of analogous samples (height: 57 mm) with either one-particle size $\left(300\pm 50\mu \text{m}\right)$ with different interstitial pastes—kaolin suspension (squares), bentonite suspension (circles), and gel (stars)—or with half the packing height (bottom) made of large particles (1 mm) and the other part (top) made of smaller particles $\left(300\pm 50\mu \text{m}\right)$—kaolin suspension (two-colored squares) and bentonite suspension (two-colored circles). The dotted line is the linear model fitted to kaolin data in Fig. 2. The inset compares the drying curves of different granular packings (with a constant volume fraction of beads: 62%) with a height of 1.5 cm: pure $\text{water}+\text{beads}$ (continuous line); kaolin paste $\left(25\mathrm{\%}\right)+\text{small}$ beads $\left(300\pm 50\mu \text{m}\right)$ (squares); kaolin paste $\left(25\mathrm{\%}\right)+\text{intermediate}$ beads $\left(600\pm 50\mu \text{m}\right)$ (circles); kaolin paste $\left(25\mathrm{\%}\right)+\text{large}$ beads (1 mm) (crosses); and more concentrated kaolin paste $\left(31\mathrm{\%}\right)+\text{small}$ beads $\left(300\pm 50\mu \text{m}\right)$ (cross squares).

Figure 6

Bead to clay volume fraction as a function of the height within kaolin (squares) and bentonite samples (circles) (times ten in that case) of two different heights: 69 mm (filled symbols) and 89 mm (open symbols).

Figure 7

Drying curves for kaolin samples of different heights as a function of the time scaled by the sample height: 20 mm (crosses), 67 mm (squares), 89 mm (stars), and 145 mm (circles). The straight line is a guide for the eyes.

Figure 8

Density profiles as measured by NMR and then corrected (see text) for a kaolin sample (sample height: 57 mm) at different times during drying: (from top to bottom) 0 (filled squares), 2.5, 5, 8.5, 11, 15, 18, and 20 days. MRI pictures (b) of a longitudinal cross section (see text) along the central axis of the same sample at different times: 0, 2.5, 5, 8.5, and 11 days (sample $\text{diameter}=\text{white}$ or gray area of 34 mm diameter).

Figure 9

Water saturation profiles in time as measured by NMR for a kaolin sample with thickness of 17 mm (increasing time from top to bottom). The first seven profiles were measured every 42 min after the beginning of the test; the others were measured for each 120 min.

Figure 10

Drying curves for bentonite samples of different heights as a function of the time scaled by the square sample height: 15 mm (crosses), 34 mm (squares), 67 mm (stars), 89 mm (circles), and 145 mm (triangles).

Figure 11

Density profiles as measured by NMR for bentonite samples [(a) $L_0=57\text{mm}$ and (b) $L_0=72\text{mm}$] at different times during drying: (from top to bottom) 0 (filled squares), 3, 4, 6, 10, 18, 21, 24, 28, 31, 34, and 38 days (empty squares). MRI pictures (c) of a longitudinal cross section (see text) along the central axis of sample (b) at different times: 0, 3, 6, 10, 18, and 21 days (sample $\text{diameter}=\text{white}$ or gray area of 34 mm diameter).

Figure 12

Water saturation profiles within bentonite samples of different heights (the distance from bottom has been scaled by the sample height) at different times scaled by the square height (successively cross, two-colored, empty, and filled symbols): 11.5 cm (squares) at 109, 261, 806, and $1100\text{min}{\text{cm}}^{-2}$; 7.2 cm (triangles) at 111, -, 778, and $1056\text{min}{\text{cm}}^{-2}$; 5.9 cm (circles) at 124, 248, 745, and $1158\text{min}{\text{cm}}^{-2}$; 4.7 cm (stars) at -, 261, 847, and $1369\text{min}{\text{cm}}^{-2}$; and 3.4 cm (losanges) at -, -, 747, and $\text{−}\text{min}{\text{cm}}^{-2}$.

Figure 13

Density profiles as measured by NMR for a two-layered (diameter: $300\mu \text{m}$ above, 1 mm below) bentonite sample $\left(L_0=57\text{mm}\right)$ at different times during drying: (from top to bottom) 0 (filled squares), 3, 6, 10, 18, 21, 24, 28, 31, 34, and 38 days (empty squares).

Figure 14

Drying curve for the gel sample in terms of the evaporated water fraction: deduced from sample mass measurements (open squares) or computed from MRI profiles without signal correction (filled squares). The dotted line has a slope of $\frac{1}{2}$.

Figure 15

Density profiles as measured by NMR for the gel samples at different times during drying: (from top to bottom) 0 (filled squares), 3, 6, 10, 18, 21, 24, 28, 31, 34, 38, and 66 days. MRI pictures (b) of a longitudinal cross section (see text) along the central axis of the same sample at different times: 3, 6, 13, 18, and 21 days (sample $\text{diameter}=\text{white}$ or gray area of 34 mm diameter).

Figure 16

(Color online) Schemes of drying processes of pastes in bead packing. (a) Kaolin suspension and (b) Carbopol gel saturating bead packing evolve in two different ways. For the kaolin air penetrates by capillary effects inside the porous medium; (c) the solid kaolin particles get closer to each other and the paste concentration increases homogeneously throughout the sample. (d) For the gel only the elements close to the free surface are squeezed and finally dried, air penetrates in the form of a front progressing through the sample.