Tải bản đầy đủ - 0 (trang)
IV. Identification and Quantitative Estimation

IV. Identification and Quantitative Estimation

Tải bản đầy đủ - 0trang


23 1

and it is not specific for allophane. Brydon and Day (1970) have shown

that soil materials having reactive A1(OH) groups such as the Bf horizon

of Podzols, give a positive test.

A combined use of dissolution treatment with an analysis of kinetics

(Langston and Jenne, 1964; Follett et al., 1965a,b; Segalen, 1968) or with

difference infrared spectroscopy (Wada and Greenland, 1970; Wada and

Tokashiki, 1972) have been used for characterization and estimation of

amorphous aluminosilicates. In the former procedure, the steady state portion of the dissolution-time curve was taken to indicate a limited attack on

crystalline materials. The difference in the rate constants for various reagents was interpreted in terms of soil clays, being a continuum from completely disordered, through poorly ordered, to well crystallized material

(Follett et al., 1965a). In the latter procedure, the difference spectra representing the infrared absorption of the materials removed by the dissolution treatments were obtained. These spectra and the spectra of the residue

were used for identification and characterization of the major constituents

and for allocation of the weight loss data.

Tweneboah et al. (1967) proposed a 12-hour extraction using 0.5 M

CaCl, at pH 1.5 for “active” aluminum oxides. This proposal is based

on the observed kinetics for the removal of iron, aluminum, and silicon

from soils and clays, and on the resulting reduction of the positive charges

developed at low pH.

As a measure of the amorphous portion of the iron oxides, a modified

Tamm’s oxalate method, involving a 2-hour extraction with 0.2 M ammonium oxalate-oxalic acid at pH 3.5, has been used widely (Schwertmann,

1964; Blume and Schwertmann, 1969). The validity of the method is based

on (1) a significant correlation between goethite content (X-ray, DTA)

and the residue, (2) the absence of a correlation between dithionite-soluble

and oxalate-soluble (Fe) in numerous soil samples, ( 3 ) nature of the solution-time curve, and (4) the observation that the hematite-goethite ratio

in mixtures of crystalline and amorphous oxides does not change consistently upon oxalate treatment.

However, the oxalate solution also extracts aluminum from amorphous

aluminum oxides and hydroxides (McKeague, 1967), and iron and aluminum from organic matter complexes (McKeague, 1967; McKeague

and Day, 1966; Blume and Schwertmann, 1969) as well as from allophane

(Miyazawa, 1966; Hattori and Morita, 1966; Dudas and Harward, 1971;

Higashi and Ikeda, 1974). It was shown that 0.1 M sodium pyrophosphate

more specifically extracts iron and aluminum in sesquioxide-organic complexes which have accumulated in soils such as Spodosols. Both the dithionite-citrate (Mehra and Jackson, 1960) and the pyrophosphate-dithionite

(Franzmeier et al., 1965) procedures remove some iron from crystal-



line oxides as well as from amorphous and crystalline clays (McKeague, 1967; BIume and Schwertmann, 1969; Dudas and Harward,


The association of amorphous silicates with “free” iron oxides was inferred from dissolution of silicon and aluminum in addition to iron during

treatment for removal of iron oxides from soil clays (Mitchell and MacKenzie, 1954; Mehra and Jackson, 1960; Follett et al., 1965b; Yoshinaga,

1966; Weaver et al., 1968). For volcanic ash soil clays in which allophane

predominates, the difference spectra have indicated that allophane-like aluminosilicates, with Si02/A1,03 ratios ranging from 0.4 to 0.8, are dissolved

by the dithionite-citrate treatment (Wada and Greenland, 1970; Wada and

Tokashiki, 1972; Tokashiki and Wada, 1972a). The spectra are characterized by the “Si-0” stretching absorption maximum in a range from 940

to 980 cm-’, and by the fairly strong absorption in the region from 600

to 800 cm-l. Relatively small amounts of unidentified constituents which

show weak absorption as well as absorption maxima different from those

described above, were dissolved from soil clays in which gibbsite and/or

crystalline layer silicates were the principal constituents. The spectra of

the constituents dissolved in 2% Na,C03 solution were also related to the

major mineral composition of the clays. The allophanelike constituents dissolved give the “Si-0’ stretching absorption maximum in a very narrow

range from 950 to 960 cm-I, and absorb relatively less in the region from

600 to 800 cm-’ compared with the dithionite-citrate soluble fraction.

Dissolution of allophane, imogolite, and gibbsite by the 0.5 N NaOH

treatment were shown on the difference spectra (Wada and Greenland,

1970; Wada and Tokashiki, 1972). No crystalline layer silicate except for

halloysite was dissolved in any significant amount unless the clay was

nearly free from allophane and/or allophanelike constituents. The allophanes dissolved in 0.5 N NaOH gave the “Si-0” stretching absorption

maximum in the widest frequency range from 940 to 1050 cm-I. This variation correlates with the largest variation in the Si0,/A1,03 ratio of the

soluble fraction (from 1.1 to 2.5). The smaller the SiO,/Al,O, ratio, the

lower the absorption frequency (Tokashiki and Wada, 1971, 1972a).

Tokashiki and Wada (1972a) compared the weight loss of the clays

and the sum of extracted Si02, A1203,and Fe,03 by the successive dithionite-citrate, 2% Na,C03, and 0.5 N NaOH treatments, They found that the

combined weight loss for the former two treatments and the weight loss

for the last treatment can approximately be equated with the amounts of

the constituents dissolved by these reagents. Then, the weight losses and

the weight of the residue remaining after the 0.5 N NaOH treatment can

be used to estimate the amounts of amorphous and crystalline constituents

by following the indications from infrared and other instrumental analyses.



- O.S.


A': Humus

AI(Fe) : Humus

AI(Fe) : Humus


- Mt

- Vt










Mt : Ch

-V t : C h


- Gb



X 1000 years





FIG. 4. Formation and transformation of clay minerals important in volcanic

ash soils developed in humid, temperate zones (Wada and Aomine, 1973; modified

on the basis of more recent data). Abbreviations: A, allophane; A', allophane-like;

AI(Fe), sesequioxides; Ch, chlorite; Gb, gibbsite; Ht, halloysite; Im, imogolite; Mt,

montmorillonite; Mt :Ch, montmorillonite-chlorite intergrades; O.S., opaline silica;

Vt, vermiculite; Vt :Ch, vermiculite-chlorite intergrades. Horizontal bars indicate

approximate duration of the respective constituents.


Formation a n d Transformation

Formation and transformation of amorphous clay materials in soils proceeds along with those of crystalline layer silicates and accumulation of

soil organic matter, and are in response to environmental changes during

soil development. This recognition is important in understanding the nature

of the processes in which amorphous clay materials are involved. Figure

4 illustrates the processes for soils developed from volcanic ash in a humid,

temperate climatic zone. Similar situations, though different in the trend

and magnitude, would also occur in any other soils.


The occurrence of the laminar opaline silica was studied in detail by

Shoji and Masui (1971). Opaline silica particles are abundant in recent

volcanic ash soils, particularly in Hokkaido soils less than 500 years old.

Although the contents are low, soils of this area which are more than 7000

years old still contain some opaline silica particles, whereas they are rare

in Kanto soils more than 6000 years old and in Kyushu soils more than

4000 years old. These differences were attributed to differences in climatic

conditions which result in slower weathering of opaline silica in the Hokkaido soils. The relative abundance of opaline silica in the A-horizon,

compared with that in the B- or C-horizons, was interpreted as formation




by supersaturation of silica and by concentration of bases due to surface

evaporation of soil water. The possibility of biogenesis should also be


There is an inverse relationship between opaline silica and allophane

in their occurrence in volcanic ash soils (Fig. 4). Shoji and Masui (1972)

noted the presence of opaline silica in the fine clay fractions of A horizons,

but it was absent in B and C horizons. The reverse was true for allophane.

“Siliceous allophane,” though it does not seem to be well defined, was

claimed to be present in both the horizons, The presence of both opaline

silica and allophanelike constituents, but not of allophane, at relatively

early stage of weathering of volcanic ash was also indicated by selective

dissolution and infrared spectroscopy for some surface soils (Tokashiki

and Wada, 1972b). In these soils, organic matter accumulated in amounts

equivalent to as much as 15-17% carbon. These observations suggest that

aluminum released by weathering of the parent material is retained by

organic matter which is stabilized against biotic degradation and leaching.

The resulting suppression of aluminum activities in soil solution would

then favor the formation of opaline silica and prevent the formation of


The occurrence of precious opal in Australia was reviewed by Darragh

et al. (1966) and has implications to the genesis of opal in soils. They

estimated that opal has been deposited under a cover of 5-40 m of rock

during the Tertiary period with climatic conditions not greatly different

from those at present. Field evidence suggested that opal was formed by

slow evaporation of very limited and highly localized supplies of groundwater, a process similar in some respects to the formation of surface duricrust but occurring under a protective cover of rock sufficient to isolate

the system from seasonal drying and flooding. Opal host cavities have characteristically been those situated in positions permitting steady-state concentration of groundwater by evaporation. Soluble salts have been removed

by diffusion to zones above the cavities, where these salts crystallized as

capillary flow ceased.





Reports of the occurrence of discrete, amorphous aluminum oxides and

hydroxides in soils are still rare, though some of them have been referred

to in Section 111, B. This may be due in part to the difficulties in establishing their presence, and in part to the reactivity of these compounds

with silica in solution or with organic matter in soils. The interlayer space

of expandable 2: 1 layer silicates, however, provides sites for precipitation



of aluminum hydrous oxides in isolation from interfering agents. Rich

( 1968) considered that the most favorable soil conditions for the hydroxyinterlayer formation is moderate pH (4.6-5.8), frequent wetting and drying

cycles, and low organic matter content. In volcanic ash soils, Uchiyama

et al. (1968b,c) found that the aluminum interlayering in the expandable

2: 1 layer silicates takes place in soils more than 500 years old, possibly

paralleling the formation of allophane (Fig. 4).

It has been recognized that there are large amounts of nonsilicate forms

of aluminum together with iron in the illuvial horizons of Podzols. Recognition that the aluminum and iron is primarily in complexes with organic

matter has been obtained by extracting the soils with either pyrophosphate

solution (Franzmeier et af., 1965; McKeague, 1967; Bascomb, 1968)

or acid ammonium oxalate solution (McKeague, 1967; Blume and

Schwertmann, 1969). That the crystallization of amorphous ferric hydroxide is significantly retarded or even inhibited by organic compounds

was confirmed either by boiling amorphous ferric hydroxide from a bog

soil in a strong alkaline solution with and without pretreatment with H,O,

(Schwertmann, 1966) or in vitro using citrate as a model organic compound (Schwertmann et al., 1968; Schwertmann, 1970).

The presence of sesquioxides in combination with organic matter in volcanic ash soils has also been indicated by extracting the soils with dithionite-citrate solution (Oba and Okano, 1967; Oba and Hayashi, 1971).

The content of extractable aluminum was higher in the surface compared

with the subsurface soils and the SiO,/Al,O, ratio of the extracted material

was generally lower than 0.5. Kato (1970a) also emphasized the importance of dithionite-citrate soluble sesquioxides, particularly alumina, in the

accumulation of humus in these soils. By applying selective dissolution and

difference infrared spectroscopy Tokashiki and Wada ( 1972b) found allophane to be essentially absent in a deeply buried volcanic ash soils which

was about 10,000 years old and contained organic matter equivalent to

14.4% carbon.





1. Soils Developed on Volcanic Ash

It has been established that the formation of allophane from volcanic

ash constitutes the central feature in the development of soils called Andosols, Andepts, or Humic Allophane soils. These soils occur throughout a

wide range of climatic conditions from the cold subhumid regions to the

humid equatorial tropics except for those in desert and semiarid regions

(Wright, 1964; Flach, 1964). Under subhumid to humid, temperate cli-



mate, nearly all kinds of volcanic ash, either basaltic (e.g., Kanno, 1961;

Siefferman and Millot, 1969: Hamblin and Greenland, 1972), andesitic

(e.g., Fieldes, 1955; Kanno, 1961), dacitic (e.g., Chichester et al., 1969;

Wada and Tokashiki, 1972) or rhyolitic ash (e.g., Fieldes, 1955) produce

allophane and allophanelike constituents, though they may be different in

nature, stability, and amount. Typical Andosols are not usually formed

from consolidated tuffs (Wright, 1964), but there are reports that a considerable amount of allophane forms from these parent materials and

affects soil properties (Saly and Mihalik, 1970; Sakr and Meyer, 1970).

Allophane does not appear at a very early stage of weathering of volcanic ash (see Section 2V, A and Fig. 4). Also, not all soils with well

developed Andosol characteristics contain allophane (Tokashiki and

Wada, 1972b). An analysis by selective dissolution-difference infrared

spectroscopy of a series of deposits in the same profile in the Kuju district

has shown that allophane is absent or nearly absent, both in the present

day and in the oldest buried, surface soils. All other buried surface soils

and subsoils contain allophane. Allophanelike constituents were present

in all the soils examined except for the oldest, buried surface soil. A balance between the release rate of aluminum from volcanic ash by weathering

and the supply rate of organic matter seems to control the formation of

allophane in these soils.

Hisingerite is normally found as a product of hydrothermal alteration

of pyroxene and olivine (Whelan and Goldich, 1961). Its occurrence in

association with ironstone (a,y-Fe,O,.H,O) in a deep soil overlying limestone in Victoria was reported by Ingles and Willoughby (1967). Silica

adsorption by the hydrated surface of the ironstone and replacement of

aluminum in allophane adjacent to the ironstone by Fe was suggested as

the probable genetic processes.

Imogolite as well as allophane imparts typical Andosol characteristics

to the soils. The number of reports on occurrence of imogolite has steadily

been increasing by use of high-resolution electron microscopy. Imogolite

has been found in weathered volcanic ash in Japan (Yoshinaga and

Aomine, 1962b; Aomine and Miyauchi, 1965; Kawasaki and Aomine,

1966; Kanno et al., 1968; Wada and Tokashiki, 1972; Shoji and Masui,

1972; Aomine and Mizota, 1973), Chile (Besoain, 1968/1969; Aomine

et al., 1972), Oregon, U S A . (Dingus et al., 1973), Papua (Greenland

et al., 1969; Parfitt, 1972), and Africa (Sieffermann and Millot, 1969) and

in pumice tuff soils in West Germany (Jaritz, 1967). Occurrence of relatively pure imogolite as gels in several pumice beds in Japan (Miyauchi and

Aomine, 1966b; Yoshinaga, 1968; Wada and Matsubara, 1968; Yoshinaga

and Yamaguchi, 1970b, Tazaki, 1971) has also been reported. The youngest soil containing imogolite has been found in volcanic ash with the 14C



age of 1610 years (Shoji and Masui, 1972) and the oldest one in the

pumice bed with the 14C age of 30,200 years (Tazaki, 1972).

Association of allophane and imogolite is usually found in volcanic ash

soils. A gradation from allophane to imogolite was observed in a series

of soils developed from the glassy volcanic ash of Kyushu “Imogo”

(Aomine and Miyauchi, 1965). In the pumice beds, allophane with the

higher Si02/Al,0, ratio was found within the pumice grains, whereas imogolite occurred exclusively as macroscopic gel films (Wada and Matsubara, 1968; Aomine and Mizota, 1973). Formation of imogolite in a

particular ash or pumice bed in the northern Kanto district usually occurs

with a relatively thin depositional overburden which serves as a silica source

(Aomine and Mizota, 1973). The association of irnogolite and gibbsite, but

not of imogolite and halloysite, has also been noted (Wada and Matsubara,

1968; Yoshinaga and Yamaguchi, 1970b; Tazaki, 1971; Aomine and

Mizota, 1973). Development of structural order in imogolite has often

been taken as indicating that it constitutes an intermediate phase from

allophane to crystalline layer silicates, particularly to halloysite, but its

SiQ2/Al,0s ratio close to 1.0 and unique tubular structure unit do not

support this view. It is more likely that imogolite represents an intermediate

phase in transformation from allophane to gibbsite in a desilication process

(Fig. 4).

As Fieldes (1955) and a number of subsequent investigators have observed, transformation of allophane to halloysite and sometimes to metahalloysite takes place in old and buried volcanic ash soils (Fig. 4). Depositional overburden generally favors this transformation (Mejia et al., 1968;

Aomine and Mizota, 1973) through effects of either the weathering time

since deposition and/or the silica supply for resilication. The effects of

overburden may be indirect. There are observations that stagnant moisture

regime favors formation of halloysite (Kanno, 1959; Aomine and Wada,

1962; Dudas, 1972). Authigenic halloysite occurs just above the paleosol

contact in well drained sites of Mazama ash (Dudas and Harward, 1974).

This localized formation was related to increased moisture regime above

the less permeable paleosol. Halloysite has formed from ashes and pumices

of various compositions, from basalt (Kurabayashi and Tsuchiya, 1960;

Sieffermann and Millot, 1969) through andesite (Birrell et al., 1955;

Aomine and Wada, 1962; Calhoun et al., 1972) to dacite (Dingus et al.,

1973; Dudas and Harward, 1974). It usually appeared as unique spherules, but not tubes. These spherules have diameters of 0.1-0.5 pm and

consist of curled short flakes (Birrell et al., 1955; Kurabayashi and

Tsuchiya, 1960; Aomine and Wada, 1962; Sieffermann and Millot, 1969).

Development of crystallinity is seen in the lattice images which appear in

these flakes (Fig. 3 ) . Substantial dissolution of these spherules takes place



in hot 0.5 N NaOH, but allophane is absent in them as was shown by

difference infrared spectroscopy (Wada and Tokashiki, 1972). The age of

the ash deposits containing halloysite has been found to range from 8650

(Aomine and Miyauchi, 1963) to 10,000 years old or more in the temperate zone, while it was found in a 4000-year-old ash soil in tropical

St. Vincent (Hay, 1960).

Transformation of allophane and imogolite to gibbsite takes place in

the final stages of weathering of volcanic ash in an environment which

favors dedication (Fig. 4 ) . In an old soil (<33,000 years old) derived

from dacitic ash, no allophane was detected and gibbsite was found to

amount to 30-45% of the clay fraction (Wada and Aomine, 1966; Wada

and Greenland, 1970). The association of allophane and gibbsite as well

as imogolite and gibbsite was found in many volcanic ash soils and pumice

beds. There are, however, some differences in the occurrence of gibbsite;

it occurs often as white concretions in the pumice, scoria, and volcanic

sand and gravel beds, but not in the fine-textured volcanic ash soils. These

associations do not mean that gibbsite always forms by transformation

either from allophone or from imogolite. Field and microscopic observations made on pumice and scoria beds (Wada and Matsubara,

1968; Aomine and Mizota, 1973) indicate that gibbsite forms separately,

possibly because of the heterogeneity of parent materials or changes in

environment from time to time.

The transformation of allophane to expandable 2: 1 layer silicates was

suggested by Masui et al. (1966) as an important process in weathering

of volcanic ash. This inference was made on the basis of their finding

that the contents of the 2:l layer silicates increase and those

of material soluble in dithionite-citrate and 0.5 N NaOH decrease with

weathering, where the degree of weathering was measured by the content

of clay in the soils. They overlooked, however, the effect of different kinds

of volcanic ash on weathering. A comparison of the mineralogical analyses

between the soils derived from andesitic and dacitic ashes in Kyushu

(Tokashiki and Wada, 1972b; Wada and Aomine, 1973) indicated that

the 2: 1 and 2 : 1: 1 layer silicate contents are much higher in soils derived

from dacitic ashes than in those from andesitic ashes, where the amount

of quartz in the clay fraction served in distinguishing the two ashes. The

infrared spectra and petrographic data obtained by Masui et al. (1966)

also support this interpretation and show that the contents of the 2: 1 layer

silicates in the soil is high when the quartz content in the clay fraction

is high and the anorthite content in feldspar is low. Higher contents of mica

and hornblende in dacitic ash may suggest that these 2: 1 layer silicates

are inherited, but their origin is beyond the scope of the present article.

It is desirable to inject a word of caution at this point. Before interpreting a particular clay mineral as a weathering product of volcanic ash, it



is necessary to establish that it was not of detrital origin and not an initial

constituent of the parent material. These are often assumed and not verified. In cases where quartz was not present as phenocrysts in the ash, it

may be used as an indicator of detrital components, including clay minerals

(Dudas and Harward, 1974). Other primary minerals may be used in

a similar manner to check the validity of interpretations of clay mineral


In summary, it seems evident that the formation and transformation of

allophane and allophanelike constituents in weathering of volcanic ash are

primarily controlled by availability of silica and aluminum in the soil solution. At an early stage of soil formation, the organic matter added to

the soil would retain aluminum released. Formation of allophane would

thereby be inhibited, formation of opaline silica favored, and the organic

matter itself stabilized against biotic degradation and leaching. The presence of allophane in buried surface soils and that of allophane and/or

imogolite in subsoils would take place as a result of desilication process

under limited supply of organic matter. Transformation of allophane to

halloysite would then take place in relatively silica rich environments resulting from either limited or stagnant moisture regime or by coupling with

the desilication process in the overburden deposit. On the other hand,

transformation of allophane and imogolite to gibbsite would take place

as a result of desilication in an open and strongly leaching environment.

Although the forms of aluminum retained by the organic matter need

further clarification, it is likely that they are subject to transformation in

weathering just as allophane and allophanelike constituents are, as

illustrated in Fig. 4.

2 . Soils Developed on Parent Materials Other Than

Volcanic Ash

The presence of allophane in clay fractions of soils derived from parent

material other than volcanic ash has been suggested by a number of investigators. DeMumbrum and Chesters (1964) reported that allophane was

present in several Wisconsin soils including Podzols. The X-ray patterns of

the acid-dispersed clay fractions apparently gave no indication of crystalline layer silicates, but in contradiction to their interpretation, the infrared

spectra indicated predominating presence of crystalline layer silicates rather

than allophane. Fieldes (1966) took dissolution of soil clay by hot 0.5

N NaOH as evidence of the presence of allophanic material in New ZeaLand soils derived from sedimentary rock as well as volcanic ash. Raman

and Mortland (1969/1970) used extraction at pH 3.7 to separate part

of the clay fraction from Onaway sandy loam, a Spodosol, and thereby

concentrated amorphous material which was soluble in 0.5 N NaOH. The

fractions separated from Ap, B2ir, B,t, and C horizons contained 50, 64,



58, and 54% amorphous material, respectively. Their SiO,/Al,O, ratio

varied from 1.25 to 1.71, and their OH water content varied from 15 to

22%. Coen and Arnold ( 1972) also found 0.5 N NaOH soluble amorphous

silicates in some New York Spodosols, and suggested that chlorite in the

albic horizon is weathered to amorphous aluminosilicates and translocated

to the spodic horizon. On the other hand, De Kimpe (1970) concluded

from his study of a Quebec Podzol that the content of amorphous material

was less than 2% of the clay fraction on a quartz-free basis, and that this

material consisted almost exclusively of Fe,O,.

The presence of poorly ordered aluminosilicates in a Podzol with thin

iron pan and noncalcareous humic gley in considerable amounts was indicated by successive extraction of soil clays with Na,C03 solution (Follett

et al., 1965a,b). The materials differ from normal finely particulate allophane and are associated closely with the crystalline layer silicates as a

film or coating. The small granules were soluble in dithionite and were

suggested to be ferruginous complexes containing considerable quantities

of silica and alumina. Disordered silica and aluminosilicates as crust on

clay particles were also observed in a Red-Brown Earth by Greenland and

Wilkinson ( 1969) by electron microscopy of carbon replicas and selective


The presence of allophane in some red, basaltic soils in Australia, referred to as Krasnozems or Red Loams, was suggested from an X-ray study

of their clay fractions. Briner and Jackson (1969, 1970) reported that the

clay fractions of Victorian soils contain 1 5 2 0 % allophane having

SiO,/A1,0, ratios from 2 to 4, and which dissolve in hot 0.5 N KOH.

Sargeant and Skene (1970) had thought that the higher CEC of clay fractions from the soils derived from the newer basalts compared with the older

basalts might be due to higher contents of allophane, but their determination showed no consistent difference in the allophane content between the

two groups of soils. Prior to these studies, Simonett and Bauleke (1963)

also determined the amounts of allophane by boiling in NaOH. Dissolution

amounted to as much as 30% of the clay, but the SiO,/Al,O, ratios of

the dissolved material were found to be very close to that of kaolinite.

They interpreted the results in terms of the dissolution of somewhat poorly

crystalline kaolinite and halloysite of small particle size. A selective dissolution-infrared spectroscopic analysis of a Glencoe soil, a Krasnozem from

Pleistocene basalt ( Wada and Greenland, 1970) supported this interpretation. Dissolution in 0.5 N NaOH after the dithionite-citrate and

2 % Na,C03 treatments amounted to 13.5% of the clay fraction, but patterns of poorly crystalline layer silicates predominated in the difference

infrared spectrum.

In Hawaii, the occurrence of allophane in weathering of basaltic and


24 1

andesitic rocks has been reported. Bates (1962) studied alteration of

Honolua andesite and found that an amorphous transition state, probably

ranging in composition from allophane to aluminum hydroxide gels, exists

as part of the change from halloysite to gibbsite. Evidence was obtained

by electron microscopy and diffraction work on pseudomorphs after halloysite tubes which were found in certain samples that had been studied in

detail with the light microscope. Patterson (1964) found fairly pure allophane and alumina-silica gel in basalt saprolite at one locality on Maui.

As illustrated in Fig. 1 (top), the main constituent of the gel was later

identified to be imogolite (Wada et al., 1972). Imogolite occurs as veins

and coatings on vesicles and cracks in the saprolite at a depth of 2-6 feet

under tropical rainforest. The allophane is white, and most of it is covered

by the clear or light-colored gel. Both the allophane and dried imogolite

gel contain about 50% alumina, 22 to 26% silica and 25% hydroxyl water.

A minor amount of gibbsite is present in the imogolite gel, but no more

than a trace occurs in the allophane.

Allophane with an approximate composition S O , . 2A1,0, 4.5H20 was

found in granodiorite in California. Its occurrence as cores to plagioclase

crystals suggest a secondary origin, probably representing the first stage of

weathering of the rock (Snetsinger, 1967). Formation of optically isotropic

material in plagioclase at early stages of granite and gneiss weathering was

also observed by Grant (1963, 1964). The appearance of an amorphous

alkali-soluble component, judged to be allophane, was also reported in

weathered andesite saprolite from the Cascade Range of Northern California (Hendricks et al., 1967). The allophane comprises 30-50% of the

clay fractions, the amounts decreasing with weathering. Halloysite, found

to be present in all saprolites, is highest in concentration in the more

strongly weathered members. The SiOr/Al,O, ratios of allophane and halloysite dissolved by 0.5 N NaOH treatment after heating at llO°C and

then after heating at 55OoC, respectively, are, however, not much different,

and both are in the range from 1.3 to 1.9. Yuan (1968) analyzed the

composition of the 0.5 N NaOH and dithionite-citrate soluble fraction in

Regosols in Florida. This fraction comprises 20-60% of the clay fraction

of these very sandy soils and has the Si0,/A1,03 ratios less than 1.7.

The presence of allophane in tropical Latosols has also been inferred

from selective dissolution analysis. Townsend and Reed (1971) found considerable amounts of alkali-soluble silica and alumina in a Panamanian

Latosol which they thought to be possible components of allophane. They

attributed the measured high CEC values of 11 me per 100 g of silt and

25 me per 100 g of fine clay to this amorphous constituent rather than

kaolin minerals present. Le Roux (1973) reported that an alkali-extractable fraction constitutes up to one-third of the total clay fraction in some

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

IV. Identification and Quantitative Estimation

Tải bản đầy đủ ngay(0 tr)