CHAPTER 11

GEOTECHNICAL PROPERTIES OF SELECTED MATERIALS FROM THE SLUMGULLION LANDSLIDE

by Alan F. Chleborad, Sharon F. Diehl, and Susan H. Cannon

Introduction

Two fine-grained, clay-rich materials (altered volcanic materials), which are commonly exposed on the surface of the active part of the Slumgullion landslide and easily distinguished in the field on the basis of gross color differences (one mostly various hues of yellow and the other mostly reddish brown), were block sampled to obtain material for laboratory strength and physical-property testing. The two materials appear to form much of the matrix of the landslide debris and also occur as local masses within the deposit. Consequently, we consider them important to the characterization of the geotechnical properties of the landslide deposit. "Popcorn weathering," characteristic of many soils with a high smectite content, was observed at both sampling locations (fig. 1) and at many other locations on the slide surface where similar materials were exposed. Slender, needle-like gypsum (selenite) crystals, as much as 10 cm long, are common in and on surface exposures of the yellow material (AT-1). These crystals were rarely observed in or on exposures of the reddish-brown material (AT-3).

Composition and Geotechnical Properties

Semiquantitative, bulk compositional data for AT-1 and AT-3 were collected on a Cambridge Mark 2 scanning electron microscope (SEM) with an attached energy dispersive X-ray (EDX). The compositional analyses show that AT-1 is relatively rich in sulfur and calcium, whereas sample AT-3 has a higher iron content (fig. 2A). The higher sulfur content for sample AT-1 relative to AT-3 is consistent with the higher sulfate mineral (gypsum) content of the yellow material noted in the field.

Prior to X-ray analyses, the samples were crushed, dispersed in distilled water, and centrifuged to achieve a less than or equal to 2-µm clay suspension. X-ray analyses were performed on oriented clay mounts of the less than or equal to 2-µm clay fraction. The X-ray diffraction analyses indicate that smectite is the only clay present in detectable amounts in the sample of yellow material, but both discrete smectite and kaolinite are present in the reddish-brown material. A scanning electron photomicrograph of sample AT-3 (fig. 2B) shows a webby, highly crenulated morphology that is characteristic of smectite. The complex microstructure of the clays suggest that their engineering behavior is complex; however, no general quantitative relationship has been established between microstructure of clays and engineering properties.

Grain size curves for the two materials indicate a clay-size (<0.002 mm) fraction of 58 percent for the yellow material and 36 percent for the reddish-brown sample (fig. 3). Preliminary examination of the >200-sieve-size (silt and fine to medium sand sizes >0.075 mm) fraction of the reddish-brown sample using a binocular microscope revealed that that fraction is composed almost entirely of angular, reddish-brown volcanic rock fragments. The >200-sieve-size fraction of the yellow material consists almost entirely of white to pale yellow, subangular to subrounded, volcanic rock fragments.

Atterberg limits were determined according to ASTM standards (ASTM, 1994). The natural moisture content of the samples was maintained prior to testing and the samples were oven-dried to equilibrium at a temperature of 60^oC (4 days drying time) to reduce the degree of dehydration of gypsum that could occur at the standard 110^oC drying temperature (ASTM, 1994). The samples were then dried to equilibrium an additional 24 hr at 110^oC to determine the additional moisture loss and possible effect of the higher drying temperature on the limits determination. Although the effect of the higher drying temperature was greatest for the yellow sample (table 1), possibly reflecting a higher gypsum and clay content, both changes are minor and suggest that additional dehydration of gypsum or other hydrous compounds in the samples at the higher temperature does not significantly affect the results. Comparison of plots for the two materials on Casagrande's plasticity chart (fig. 4A) with those for the common clay minerals (fig. 4B) shows that sample AT-3 plots near the zone typical for montmorillonites (smectites), but sample AT-1 (the yellow material) exhibits apparently anomalous plasticity characteristics by plotting below the "A-line". In contrast, both materials exhibit a high dry strength that is characteristic of clays with high plasticity. It has been reported that the composition and concentration of soluble salts in a soil can affect liquid and plastic limits. For example, a reduction in the liquid limit of sodium montmorillonite may result from increased salt concentration due to a corresponding decrease in interparticle repulsion (Warkentin, 1961). Compositional analyses and field observations indicate that sulfates are the most abundant soluble salts in samples AT-1 and AT-3; however, the effect of the sulfates on salt concentration and the limits is undetermined.

Table 1. -- Atterberg limits and natural moisture contents of selected materials from the active part of the Slumgullion landslide. [LL, liquid limit; PL, plastic limit; PI, plasticity index]
Sample	Atterberg limits						Natural moisture content (%)
	(Drying temp. = 60o C)			(Drying temp. = 110o C)
	LL	PL	PI	LL	PL	PI
AT-1 (yellow)	92	49	43	95	45	50	56
AT-3 (reddish-brown)	62	23	39	65	26	39	38

Likewise, the data are insufficient to account for the apparent discrepancy in the activities of the samples. It has been shown that a good correlation exists between the activity of a clay (defined as the ratio of plasticity index to percent by weight of the clay fraction) and clay type (Holtz and Kovacs, 1981). As shown on figure 5A, neither of the materials displays high activities characteristic of montmorillonites.

The swelling potentials of the the two materials were estimated by plotting the activity versus clay fraction on the swelling potential classification chart developed by Seed and others (1962) (fig. 5B). As shown on figure 5B, swelling potential of both materials falls in the high category. The indicated swelling potentials suggest that variations in moisture content of the materials in the field may result in measurable seasonal movement of the ground surface. Such movements may be significant in deformation studies to detect movement of the ground surface associated with landsliding. Mielenz and King (1955) report that gypsum in sodium bentonite (1) will effect partial cation exchange upon wetting of the material and (2) impedes swelling.

Undrained shear strength of the two materials was estimated in the field using an in-situ hand vane tester. The instrument consists of a torque head, including a direct-reading scale, which is turned by hand. Subsequent to excavation of a 3-ft deep pit, a 0.75 inch-diameter vane was inserted into the soil to an additional depth of 6 inches. The vane was rotated at a constant shearing rate of 1 revolution per minute. to obtain estimates of both maximum and minimum shear strength. A set of ten tests on both materials yielded average maximum undrained shear strength values of 575 lb/ft² (28 kPa) for the yellow soil (AT-1) and 785 lb/ft² (38 kPa) for the reddish-brown soil (AT-3). Average minimum shear strength values of 90 lb/ft² (4 kPa) for the yellow material and 110 lb/ft² (5 kPa) for the reddish-brown material were obtained by the continued rotation of the vane after maximum strength was reached. Subsequent to testing, the sheared material was examined for evidence of rock fragments or other hard materials that might have affected the results of the test. In the few instances where this was found to be the case, the data were discarded and a new test location in an undisturbed area of the pit was selected. Natural moisture contents of 56 percent for the yellow material and 38 percent for the reddish-brown material (table 1) were determined in the laboratory subsequent to field testing. Consolidated-undrained triaxial testing of the two materials to determine strength and stress-strain relationships is in progress.

Summary

Preliminary testing and examination of two fine-grained materials (highly altered volcanic material), commonly exposed on the surface of the active part of the Slumgullion landslide, indicate that both are low-strength, clay-rich materials with medium to high plasticity and high swelling potential. Anomalously low activities of the smectite-bearing materials cannot be explained with the available data.

Further testing of materials from the active and inactive parts of the Slumgullion landslide is needed to characterize the geotechnical properties of the landslide materials. To address unanswered questions regarding the type and relative abundance of various minerals and the effect of soluble salts on material properties, additional chemical and semi-quantitative analyses are needed. Planned physical-property testing of surface samples that have been collected at regular intervals on traverses across the lower, middle, and upper parts of the active part of the landslide will further characterize the geotechnical properties of the slide material.

References Cited

American Society for Testing Materials (ASTM), 1994, Soil and Rock: 1994 Annual Book of ASTM standards, sec. 4, v. 04.08, 975 p.

Holtz, R.D., and Kovacs, W.D., 1981, An introduction to geotechnical engineering: Englewood Cliffs, N.J., Prentice-Hall, 733 p.

Mielenz, R.C., and King, M.E., 1955, Physical-chemical properties and engineering performance of clays, in Pask, J.A., and Turner, M.D., eds., Clays and Clay Technology: California Division of Mines Bulletin 169, p. 196-254.

Seed B.H., Woodward R.J., and Lundgren R., 1962, Prediction of swelling potential for compacted clays: Journal of the Soil Mechanics and Foundation Division of American Society of Civil Engineers, v. 88, no. SM-3, p. 53-87.

Skempton, A.W., 1953, The colloidal activity of clays: Proceedings of the 3rd International Conference on Soil Mechanics and Foundation Engineering, v. 1, p. 57.

Varnes, D.J., Smith, W.K., Savage, W.Z., and Varnes, K.L., 1993, Control and deformation surveys at the Slumgullion Slide, Hinsdale County, Colorado--A progress report:U.S. Geological Survey Open-File Report 93-577, 15 p.

Warkentin, B.P., 1961, Interpretation of the upper plastic limit of clays: Nature, v. 190, p. 287-288.