Tải bản đầy đủ - 0 (trang)
VI. Surface Area and Pore Size

VI. Surface Area and Pore Size

Tải bản đầy đủ - 0trang













0 .2

.4 .6


.8 1

0 . 2 . 4 .6 .8 1




6 810

pore width (nm)


4 6810

pore width (nm)

Figure 7 Nitrogen sorption isotherm and derived cumulative pore size distribution (a) natural

aggregates of a vertisol (total porosity, 0.13 cm3g-I) and (b) Ca-Willalooka illite-compressed cores

(total porosity, 0.23 cm3g-I). (After Murray er al.. 1985.)

aggregates of a Queensland vertisol (No. 10). It can be noted that the plot of

cumulative volume versus pore size indicates modal values of about 4 and 3 nm

respectively. The slope of the desorption isotherm for Willalooka illite indicates a

mixture of slit- and wedge-shaped pores. The isothern for the vertisol shows a

large hysteresis and its shape indicates a predominance of slit-shaped pores.

There is a considerable variation in the isotherms for the other vertisols studied

with some having a sloping desorption isotherm.

Figure 7 also shows that the pore size in these materials is predominantly < 10

nm. For the vertisol, 50% of the pore space is in pores less than < 3 nm and 95%



of the surface area is in these pores. For the Willalooka illite, 85% of the pore

space and virtually all of the surface area available to nitrogen are in voids less

than 5 nm. From these and other observations it can be concluded that these pore

sizes, which have been almost totally neglected, are a common feature of clay

soils. Sills et al. (1973) have studied the movement of the pore peak as illite is

added in increasing proportions to kaolinite. As the percentage of illite increases

the large voids between kaolinite crystals are filled with illite particles and as a

result the pore sizes decrease substantially.

From this information it can be readily understood that pore sizes for finegrained illites and montmorillonites in the dry state are predominately less than 5

nm which thus allows for the effective interplay of interparticle forces. Murray

and Quirk (1994) have discussed the conditions for particle bending on the

drying cycle of a clay when very large forces are operative for close distances of

approach of the surfaces. The release of this mechanical energy would assist the

early stage of the swelling process; however, as noted earlier, for most practical

purposes the surfaces of clay particles within a soil are always separated by at

least two layers of water.


Table IX shows the magnitude of the surface areas of clay particles. The

significant feature is the close agreement between the surface areas obtained from

both arms of the sorption isotherm. The B.E.T. and Kelvin equations involve

entirely different sets of assumptions (Murray et al., 1985; Murray and Quirk,

Table IX

Comparison of the Specific Areas Obtained from Nitrogen

Adsorption Isotherms (B.E.T. Equation) and the Desorption

Isotherms (Kelvin Equation)u


Redhill montmorilloniteb

Willalooka illite ( < O . 1

Rocky Gully kaoliniteb

Queensland vertisol No. 1 I

Urrbrae B aggregatesb

Adsorption isotherm

(m2g- I)

Desorption isotherm












The clay cores were Ca saturated and the vertisols and Urrbrae loam B

horizon were aggregates as sampled from the field.

From Aylmore and Quirk (1967).

From Murray er al. (1 985).



1990b) and hence the agreement between the surface areas derived from both

arms of the sorption isotherm is noteworthy. This must mean that there are no

“ink bottle” pores or restrictions hindering the access of nitrogen to the whole

surface area and its removal on the desorption isotherm. Virtually every pore

within clay domains has access to the sample surface via at least one path which

contains no finer voids or restrictions. That is, domains or regimes of common

particle orientation are bounded by an extensive network of larger voids or

cracks. These flaws or discontinuities must be an intrinsic feature of clay materials (Murray and Quirk, 1990a,b, 1994) and have been described as intrinsic

failure (Quirk, 1978). The intrinsic failure pores are obviously sites of potential

weakness in the structure. Williams et al. (1967) obtained a reduction of the

nitrogen surface area and porosity of Urrbrae loam B horizon to about two-thirds

of the untreated material by the adsorption of 0.035-g polyvinyl alcohol per gram

of soil. The molecular weight of the PVA was 25,000 and the effect was much

less when a PVA with a MW of 70,000 was used. The authors describe the effect

as being due to “peripheral pore occupation” by the polymer, preventing access

of N2 molecules to pores and surfaces within clay domains. In the light of more

recent work it seems a reasonable hypothesis that the polymer modifies access to

the whole surface by affecting access via the set of intrinsic failures.

Although such a system of discontinuities does not contribute significantly to

either the porosity or the surface area of the clay (Murray and Quirk, 1990a), it

imposes a limit on the strength of the clay matrix. The intrinsic failure pores are

probably the precursors of cracking in general and the network of flaws probably

influences the slaking and dispersion of clays or collections of domains (microggregates) (see model in Williams et al., 1967). It also seems reasonable to

conclude that intrinsic failure pores are involved as a source of weakness leading

to failure of sodic soils at low electrolyte concentrations. The actual sites of

weakness may be the connections, between the walls of these pores, which must

exist at periodic intervals. Thus the intrinsic failure pores can make a contribution to physical behavior which is quite disproportionate to their actual volume.

The presence of intrinsic failures within a clay mass which swells isotropically

is preordained by the presence and nature of clay domains since the clay domains

themselves provide a discontinuity. The swelling of the domains, although unidimensional for domain, create a competition for space since each domain swells

in different directions because of their random array. This contributes to the

swelling process by increasing the volume of the intrinsic failure pore space.




The interleaving of lamellar particles produces slits and wedge-shaped pores

with maximum widths which reflect the thickness of the particles. Additionally



there are surfaces which are essentially in adhesive contact but which separate on

hydration; most of the total surface area of smectites is disposed in such regions

of close approach. Figure 8 shows a two-dimensional structure generated by

stacking lamellar particles with random lengths of 0-1000 units and random, but

discrete, thicknesses of 1-10 units. The porosity (0.36 cm3 ~ m - ~surface



(73 m2g-l), and pore size distribution derived from this artificial assembly are

very similar to the experimentally determined values obtained for a relatively

coarse illite (Quirk and Murray, 1991).

The relative abundance of slits and wedge-shaped pores, and the acuity (i.e.,

the dihedral angle) of the wedge-shaped ones, must depend on a number of

factors, including the aspect ratio and pliability of the particles, the forces between the particles, and the history of the material (Murray and Quirk, 1994).

Obviously the aspect ratio of particles is important. Thin particles lead to an

abundance of slits while “blocky” particles produce inferior orientation. True

slits can only arise in two situations. The first of these occurs when two particles

are propped apart at two or more points by particles of equal thickness; this is

relatively improbable. The second occurs when two particles are propped apart

by a single, flat, third particle; this is a more probable event. However, the

pliability of particles acts to reduce the abundance of true slits, especially of

long, narrow ones. These expectations of clay microstructure are supported by

experimental measurements of pore size distribution.




F’igure 8 A randomly packed but oriented assembly of lamella particles with random lateral

extents (
area” (73 m2g-l) and porosity (0.36 cm3 c m 3 ) are quite consistent with values measured for illite

samples. (After Quirk and Murray, 1991.)






Baver (1940) remarked, “The colloidal material is responsible for the cementation of primary particles into stable aggregates.” He considered that the clay

particles themselves, oxides of iron, aluminium and silicon, and the organic

colloids were significant with respect to cementation.

To obtain a more complete appreciation of the relative importance of different

colloids a variety of methodologies have been used to assess stability. These

include the examination of a large number of soils with a view to establishing

correlations between water stability and a particular soil component; the major

investigation by Kemper and Koch (1966) is an example of this methodology.

Another involves the effect of the addition of various substances such as organic

materials, lime, and silicates to assess the role that such substances may have in

conferring stability. One issue of “Soil Science” (73,No. 6, 1952) was devoted

to the role of synthetic polymers in stabilizing soil aggregates. Yet another

approach involves subtractive techniques whereby a chemical reagent specific for

a particular soil constituent is used to remove that constituent to determine its

role in stabilizing aggregates. There is an exceedingly large body of literature

concerned with these topics; however, in this review attention is given to a few

pieces of work which illustrate the importance of microstructure and the disposition of cements or stabilizing materials.

Greenland er al. (1962) examined the permeability of prewet aggregates of a

red-brown earth (alfisol) after treatment with alkaline sodium periodate. The

reduction in permeability was small for samples from plots which had been under

pasture for many years, but large for samples which had been under pasture for 4

years or less. They concluded that aggregate breakdown as a result of the periodate treatment is probably due to the disruption of polysaccharide and polyuronide moieties in the soil and that stability under cropping and young pastures

is due to relatively ephemeral materials. This is consistent with the correlation

established by Chesters et al. (1957) between stability and polysaccharide content. The nature of the organic matter in virgin soil samples or accumulating over

time under pasture remains largely unresolved but several possibilities exist such

as a greater aromatic content or the slow accumulation of clay-polyvalent metalorganic matter complexes as suggested by Edwards and Bremner (1967) with

respect to the formation of aggregates less than 250 pm.

Desphande et al. (1968) investigated a number of red soils for which iron

oxides were thought to exert a favorable effect on soil physical properties; these

included krasnozem (oxisols), terra rossa (rhodic calcixerolls) and related rendzina (lithic calcixerolls), and a lateritic red earth (Ultisols) soils. The removal of



iron oxides by sodium dithionite treatment did not result in any changes in the

permeability of a bed of aggregates as compared with the control treatment,

sodium sulfate. The changes in specific area following the removal of iron oxides

indicated that, in all but one instance, the iron oxides are principally present as

discrete particles with relatively large surface areas. They qualified their conclusion by suggesting that it was possible that minor amounts of iron oxide may be

present as active-binding agents. However, Desphande et al. (1968) also reported that acid treatment (0.1 M HCI), which removed silicon, aluminium, and

minor amounts of iron, as compared with the 2-15% by weight removed by

dithonite treatment from the soils studied, mostly produced larger changes in

physical properties.

Because of the effects of the removal of relatively small amounts of aluminium

on the results for permeability, wet sieving, mechanical analysis, and swelling

determinations, Desphande et al. (1968) considered that reactive aluminium in

the form of interlayers or islands between contiguous crystals or clay domains

could confer stability. The authors observed that ferric oxides are not known to

form chloritic type interlayers, although chlorites containing high proportions of

Fez+ in the brucite layer are well known. The small quantity of silicon and

aluminium may arise from very small particles with a composition similar to the

clay minerals.

These considerations suggest that the various empirical approaches to assessing aggregate stability are limited and are not capable of unequivocal interpretation with respect to the basis of stability. Virtually no attention has been given to

the disposition of stabilizing agents in the porous matrix of the soil or indeed to

the nature of the porous matrix itself.

The actual forces involved in cementation need definition. It should be emphasized again that a soil is a condensed particle system to which the excellent

studies of heterocoagulation between clays and oxides (Tama and El-Swaify,

1978) may have limited applicability in considering the role of oxides or their

precursors in stabilizing soil aggregates. The basic questions which need attention are: What is the nature of the forces between differently charged particles at

close distances of approach? What is the relative role of van der Waals forces in

such an interaction? It seems timely to address this difficult problem since considerable information on the surface chemistry of oxides has accumulated in

recent decades (Bowden et al., 1980; Goldberg, 1992).






The outstanding feature of soil structural behavior is the profound influence

that organic matter, or rather some moiety of it, can have on the water stability of

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

VI. Surface Area and Pore Size

Tải bản đầy đủ ngay(0 tr)