Comparison of simulated void structure with NMR and thin section data

by G.Peter Matthews, Deborah Holtham, and Christophe Canonville
School of Earth, Ocean and Environmental Sciences, University of Plymouth

Although NMR can give information abut pore size and degree of saturation in soil, it cannot give information about connectivity, which is an essential factor in soil hydraulic properties. Also, NMR investigations of pore size require calibration, which often relies on the capillary bundle (bundle-of-tubes) approximation, and can produce widely different calibration factors for each sample, so reducing the predictive capability of the technique. This presentation addresses these problems for sandstones, rather than soils, and suggests suitable approaches for the investigation of soil structure.
Mercury intrusion curves for five widely different sandstones were simulated by the Pore-Cor network model, to produce structures. These structures were then mathematically microtomed into 100 horizontal slices. The resulting fragment areas agreed much more closely to experiment than areas calculated using the capillary bundle approximation.
The relative signal strengths of the NMR T2 relaxation times depend on the volume of water generating each signal. So we converted this to a relative number of features, assuming all features are isotropic with diameter or side length d, by dividing each signal amplitude by its respective d3. This gave rise to number densities of features that were skewed very markedly to smaller sizes. We assumed that the NMR method was insensitive to features of diameters below 0.1um, so only compared simulated features above this size. Although the shapes of the simulated and experimental NMR distributions differed markedly, the mean sizes of the ranges trended well against each other, whereas corresponding results from the capillary bundle approximation gave a weak and incorrect trend against experiment.
The results suggest that calibration of NMR T2 relaxation times using the capillary bundle approximation is dangerous. The results also lead to suggestions about improvements in interpreting the NMR data and making further developments to the model, which are applicable to soil as well as sandstones.
 

Matthews, G.P., Canonville, C.F., and Moss A.K. Use of a void network model to correlate porosity, mercury porosimetry, thin section, absolute permeability and NMR relaxation time data for sandstone rocks, accepted for publication, Sep 2005.
 

J.C. Price* and G.P. Matthews
Environmental and Fluid Modelling Group, University of Plymouth, Plymouth, PL4 8AA.
* Presenting Author

See abstract.

D. A. L. Holtham 1,3, D. Scholefield 1, A. Cresswell 2 and G. P. Matthews 3

1 Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon, UK.
2 Institute of Grassland and Environmental Research, Plas Gogerddan, Aberystwyth, UK.
3 Department of Environmental Sciences, University of Plymouth, Plymouth, UK.

Introduction
Achieving an effective balance between environmental protection and agricultural production requires understanding and management of soil processes. Good soil structure is the basis of good agricultural production and for agriculture to be sustainable it is important that the soil resource is not degraded. Sustainable agriculture must also be economically viable and unfortunately the economic targets of farmers cannot easily be reconciled with reduced use of fertilizers in conventional grassland systems. This has prompted a marked swing to ‘organic’ production, which is based principally on the acquisition of nitrogen (N) through biological fixation in the root nodules of legumes. Previous work at IGER has revealed that structural differentiation under white clover is phenomenally rapid and enhanced when compared with ryegrass (Mytton et al., 1993). However, the mechanisms of aggregation are poorly understood. Soil structural differentiation determines both the proportion of accumulated nutrient that actually leaches and the concentration at which it enters water bodies (Scholefield et al., 1996). Legume-based systems are not environmentally benign: similar amounts of N and phosphorus (P) are leached from beneath grass-clover swards as those leached form beneath fertilised grass operating at the same level of production (Tyson et al., 1997; Cuttle et al., 1998). In some circumstances, clover rich swards can give rise to very high levels of nitrate leaching (Loiseau et al., 2001).

Aims
This study aims to investigate the mechanisms of accelerated and enhanced structuring in soils under white clover and to assess the concomitant changes in nutrient leaching. Thus achieving a balanced insight into the sustainability and environmental consequences of manipulating soil structure in agricultural systems.

Experimental
A sandy loam of the Crediton Series (FAO - Dystric Eutrocrept) was dried, sieved and re-packed to field bulk density in plastic cylinders (110 mm diameter x 170 mm in height). The seedbed was inoculated with Rhizobium and sown with perennial ryegrass, white clover, or a mixture of the two species.

Results
After 14 weeks of growth in a glasshouse a greater degree of structural differentiation was seen in surface layers of soil beneath clover compared with both N fertilised grass and unplanted controls. Preliminary results indicate increased permeability of air and water in soils under white clover and enhanced structural stability to water when compared to ryegrass. There was a marked difference in nitrate breakthrough curves for clover and grass treatments.

Conclusions
This study has confirmed previous work in demonstrating enhanced soil structuring under white clover compared to ryegrass. This enhanced structuring gave rise to greater permeability of air and water, and these soil qualities are important attributes of sustainable systems. However, there may be potentially negative impacts associated with clover induced soil aggregation that need further investigation.

References
Cuttle, S. P., Scurlock, R. V. and Davies B. M. S. 1998. A 6-year comparison of nitrate leaching from clover/grass and N-fertilised grass grazed by sheep. Journal of Agricultural Science 131: 39-50.
Loiseau, P., Carrere, P., Lafarge, M., Delpy, R. and Dublanchet, J. 2001. Effect of soil-N and urine-N on nitrate leaching under pure grass, pure clover and mixed grass/clover swards. European Journal of Agronomy 14: 113-121.
Mytton, L. R., Cresswell, A. and Colbourn, P. 1993. Improvement in soil structure associated with white clover. Grass and Forage Science 48: 84-90.
Scholefield, D., Lord, E. I., Rodda H. J. E. and Webb, B. W. 1996. Estimating peak nitrate concentrations from annual nitrate loads. Journal of Hydrology 186: 355-373.
Tyson, K. C., Scholefield, D., Jarvis, S. C. and Stone A. C. 1997. A comparison of animal output and nitrogen leaching losses recorded from drained fertilised grass and grass/clover pasture. Journal of Agricultural Science 129: 315-323.

P. A. BODURTHA1*, G. P. MATTHEWS1, J. P. KETTLE2, S. LOHMANDER3 AND P. W. JAMES4

1Environmental and Fluid Modelling Group, University of Plymouth, Plymouth PL4 8AA (UK)
2SCA Graphic Research, SE-850 03 Sundsvall (Sweden)
3Swedish Pulp and Paper Research Institute (STFi), S-114-86 Stockholm (Sweden)
4Department of Mathematics and Statistics, University of Plymouth, Plymouth PL4 8AA (UK)

In this study, we have investigated the effects of structural anisotropy of porous media on the permeation of fluids. The motivation for the work was an increased understanding of the permeation of inks into paper coatings, which often contain platey or needle-like particles, which have been aligned during the coating process. Mineral pigments, comprising mainly of calcium carbonate or clay, are often are applied to the surface of paper to improve optical and printing properties. For a high quality image to be achieved, the coating should have sufficient capillarity to allow the ink film to set within the time-scale of a modern printing press. The pore structure and surface chemistry at the paper/ink interface determine the ink uptake of a coating.

The way the fluid permeates through the structure is determined by the size and geometry of the voids, the way they are connected into a network, the surface energy between the fluid and the solid phase, and the density, viscosity and applied pressure of the fluid. If a porous solid is made up from a homogeneous packing of unsorted spherical particles bound together, the resulting void structure will normally be structurally isotropic - i.e. it will have the same structural characteristics in any direction, and hence the same permeation characteristics. However, a sample made up from non-isometric particles such as clay platelets, packed in an aligned manner, would be structurally anisotropic. Consequently, the permeation characteristics of such a sample would also be anisotropic.

The permeation of fluids into four different coatings formulations has been studied in two categories: Speswhite and Amazon90 SD which belong to the Kaolin (clay) mineral group, and OpacarbA40 and Albaglos which belong to the Precipitated Calcium Carbonate (PCC) mineral group. The permeation was measured by two methods. The first involved use of an Ink Surface Interaction Tester, which transferred oil from a gravure roller to a sample at a known contact pressure and speed. The second involved measure of colour density, after the wiping off of excess Croda Red ink at set times after transfer.

The results were modelled using a modified version of the software package 'Pore-Cor', which simulated both permeability and capillary absorption of a wetting liquid into porous media containing anisotropic voids, and allowed the effects of anisotropy to be isolated from other closely related pore properties. The model generated a simplified three-dimensional void network having pores with a rectangular cross-section and throats with an elliptic cross-section. The model was also modified to demonstrate the effect of anisotropy caused by differences in positions of isotropic voids, rather than in the anisotropy of individual voids, specifically the effect of horizontally layering or laminating the structure.

The application of an anisotropic term to the modelled fluid behaviour in describing permeability and capillary absorption gave trends, which correlate better with the experimental data than either a traditional straight aligned-tubes model or an isotropic three-dimensional network. The insights gained from this study have allowed conclusions to be drawn about the nature of fluid permeation; they have therefore opened the way to more sophisticated modelling and the engineering of high performance coating structures.