| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Rua dos Funcionários, 1540, Juvevê, Departamento de Solos, Universidade Federal do Paraná, Curitiba (PR), 80035-050, Brazil
b Dep. of Agricultural Chemistry and Soil Science, The Univ. of Sydney, N.S.W. 2006, Australia
c Departamento de Solos, Universidade Federal de Viçosa, Viçosa (MG), 36571-000, Brazil
* Corresponding author (b.singh{at}acss.usyd.edu.au)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
, and dehydroxylation temperature between 489 to 518°C. Kaolinite in the clay fraction has relatively poor crystal order with a mean crystallinity index value(CI) of 12.7. The dominant forms of the clay-fraction kaolinite were elongated and rounded, with relatively lower proportions of hexagonal particles. The silt-fraction kaolinite showed a tendency to form subspherical large aggregates with high stability. The average Fe2O3 level in the kaolinite of the clay fraction (19.1 g kg-1) was higher than that obtained for the silt fraction (6.6 g kg-1). The smaller kaolinite particles of the clay fraction showed a lower degree of crystal order, higher K and Mg levels, and lower dehydroxylation temperatures. From the strong relationship between the asymmetry index (AI) of the (001) diffraction line and the level of K in kaolinite from the younger soils, we believe that both K and Mg in kaolinite are part of residual micaceous layers interleaved in kaolinite crystals.
Abbreviations: AEM, analytical electron microscopy • AI, asymmetry index • ALN, average layer number • CI, Hughes and Brown crystallinity index • EDS, energy dispersive spectrometry • DT, dehydroxylation temperature • ICP-AES, inductively coupled plasma atomic emission spectrometer • MCD, mean crystal dimension • SEM, scanning electron microscope • SSA, specific surface area • TEM, transmission electron microscope • WHH, width at half height • XRD, x-ray diffraction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Interstratification and impurities of other minerals with kaolinite can influence levels of exchangeable and nonexchangeable K and Mg in soils. The K and Mg content of well-crystallized kaolinites are normally very low. From the AEM analysis of single crystals, Singh and Gilkes (1992) showed that kaolinite in southwestern Australian soils have K2O levels ranging between 1.0 to 2.9 g kg-1. The authors postulated that the K is associated with the presence of 1 to 2.9% micaceous interlayers. The mica layers were not detected by XRD. Georgia kaolinite used in the same study revealed 2.1 g kg-1 of K2O and distinct mica peaks, which were observed by XRD. The presence of such mica zones may be frequent in soil kaolinites because pseudomorphic alteration of mica to kaolinite is commonly found in saprolites (Pinto et al., 1972; Rebertus et al., 1986; Graham et al., 1989).
The specific objective of this study was to test the hypothesis that K and Mg are present as micaceous inclusions in kaolinite in highly weathered soils from Brazil.
|
MATERIAL AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
X-Ray Diffraction
Random and parallel-oriented samples were analyzed by XRD using Cu K
radiation from a Philips PW1050 vertical goniometer (Philips Analytical, Cambridge, UK) equipped with 1° divergence and receiving slits and a graphite monochromator. Minerals in the clay fractions were identified from the oriented and random powder diffraction patterns following the procedures given by Brown and Brindley (1980).
About 25 mg of octacosane [CH3(CH2)26CH3], a long chain alkane, were melted into oriented samples to serve as an internal standard for XRD analysis of kaolinite basal spacings (Brindley and Wan, 1974). The diffraction patterns were obtained in a horizontal scale ranging from 3 to 30 °2 at an angular speed of 0.1 °2 min-1 and a step size of 0.02 °2. Initially, the d-values for octacosane peaks were calibrated using quartz as an internal standard. Separate oriented specimens were prepared to determine the AI of the 001 and 002 diffraction peaks of kaolinite (Singh and Gilkes, 1992). The CI of kaolinite was calculated from the random powder diffraction patterns following the procedure given by Hughes and Brown (1979). The CI = A/B, where A is the intensity above background of the reflection at 4.46 Å and B is the intensity above background at about 2.43 Å.
The mean crystal dimension (MCD), the thickness normal to the diffracting plane, was calculated from WHH using the Scherrer's equation (Klug and Alexander, 1954), and the average layer numbers (ALN) was calculated by dividing MCD by the d(001) values.
Thermal Analysis
Simultaneous differential thermal analysis, differential thermogravimetry, and thermogravimetric analyses of the deferrated samples were done using a Stanton Redcroft STA-780 Series instrument (Fire Testing Technology, West Sussex, UK) by heating a 20-mg sample from ambient temperature to 1000°C at 10°C min-1 under an N2 atmosphere.
Specific Surface Area
For the measurement of external surface area, the kaolinite samples were degassed overnight at 373 K. The measurements were made using a Micromeretics Gemini III 2375 surface area meter (Micromeretics Instruments, Norcross, GA) by N2 adsorption (relative pressure ranged between 0.05–0.30) at 77 K and using the BET equation.
Chemical Analysis
The deferrated clay samples were treated with 0.5 M NaOH to remove amorphous alumino-silicates and gibbsite (Jackson, 1979). Kaolinite was dissolved by boiling in 5 M NaOH for 60 min (Norrish and Taylor, 1961). The residual sodalite formed during NaOH treatment was removed by two successive extractions with 0.5 M HCl (Singh and Gilkes, 1991). The extracts obtained from the 5 M NaOH and 0.5 M HCl treatments were analyzed for Al, Si, Fe, Ti, K, and Mg with a Perkin Elmer Optima 3000 (Perkin Elmer Analytical Instruments, Norwalk, CT) inductively coupled plasma atomic emission spectrometer (ICP-AES). The quantitative estimation of kaolinite and mica in the clay fraction of soils was done from the chemical analyses of extracts after 5 M NaOH and NaHSO4, respectively (Norrish and Taylor, 1961; Jackson et al., 1986; Melo, 1998).
The kaolinite in the silt fraction was removed from the untreated natural sample by heating at 550°C for 4 h followed by boiling in 0.5 M NaOH for 2.5 min. (Jackson, 1979). The extract was analyzed for Si, Al, Fe, Ti, K, and Mg contents by ICP-AES.
Electron Microscopy
Kaolinite in the clay fraction was examined with a Philips EM 400T transmission electron microscope (Philips Analytical, Cambridge, UK) operated at 200 kV and coupled with an energy dispersive spectrometry (EDS) system. The silt-fraction kaolinite was examined with a JEOL JXA-840 scanning electron microscope (JEOL USA, Inc., Peabody, MA) operated at 20 kV and equipped with an EDS system.
The proportion of euhedral faces was calculated by dividing the sum of euhedral faces for a sample by the maximum possible number of euhedral faces, six per crystal.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The high kaolinite content is consistent with the intense weathering conditions experienced by these soils in the humid tropics (Juo, 1980; Dixon, 1989). Under high temperatures, humidity, and leaching conditions, most common primary minerals such as micas and feldspars weather directly to kaolinite (Grant, 1964; Anand et al., 1985; Rebertus et al., 1986; Nwadialo and Lietzke, 1989).
XRD–Based Measurements
The d(001) values for kaolinite in the clay fraction varied from 0.713 to 0.728 nm with a median value of 0.719 nm (Table 2). The range for the 002 d-values was narrow (0.357–0.359 nm) compared with 001 d-values. The wider range for the 001 d-values may be partly because of inherent greater inaccuracy in measuring d-values at smaller 2 angles.
|
The WHH, which is inversely related to the size of coherently diffracting domains, ranged between 0.30 and 0.97 °2 with a median value of 0.41 °2. The median ALN for kaolinite in the studied samples was 27. Kaolinite in Samples 13, 16, and 17 have the highest WHH values (Table 2) and contain very high amounts of K and Mg in the clay fraction (Table 3). Kaolinite crystals in the younger soils (Samples 16 to 21) are smaller than the older soils and contain a higher nonexchangeable K and Mg. A negative correlation between MCD (001) and K (r = -0.710***; significant at the 0.001 level) and Mg (r = -0.920***) contents was observed for all samples (Table 4). The kaolinite in highly weathered soils, developed on Tertiary sediments (Barreiras Group), has the lowest WHH and the highest MCD values.
|
|
The thickness of kaolinite particles is correlated with CI (r = 0.730***) and the correlation between crystal volume [volume = a x b x MCD(001); a = longest width of crystal, b = shortest width of crystal] and CI is even higher (0.930***). The results show that well-ordered kaolinite crystals have a larger volume than the poorly ordered kaolinites. There is a weak negative correlation between crystal volume of kaolinite with K and Mg contents (Table 4). These results suggest that better crystalline kaolinites have greater crystal volume and fewer impurities (such as K and Mg).
Specific Surface Area
The N2–BET surface area for the kaolinite-rich soil clay fraction (median = 44.5 m2 g-1) was nearly twice as large as reported values for soil kaolinite (24 m2 g-1; Singh and Gilkes, 1992). This result is consistent with the broad basal reflections observed in the random powder diffraction patterns and the negative relationship between the specific surface area (SSA) and MCD (r = -0.740***). The SSA values of B horizon samples are considerably larger than values for the C horizon samples (Table 2). The lower SSA at greater depth could be either because of greater weathering of kaolinite in surface soils or authigenic formation of large kaolinite in the C horizons due perhaps to less inhibition at crystal growth by organic matter (or both). A significant positive correlation was observed between SSA and K and Mg contents in kaolinite (Table 4). These results are in accordance to the XRD observations and suggest that smaller kaolinite crystals contain higher amounts of K and Mg.
Dehydroxylation Temperature
The dehydroxylation temperature (DT) for kaolinite in the clay fraction ranged from 489 to 518°C (Table 2) with a median value of 514°C. The lower DT for soil kaolinite compared with specimen kaolinite (>540°C) may be because of reduced crystal size and poor crystallinity (Smykatz-Kloss, 1975; Singh and Gilkes, 1992). Dehydroxylation temperature has a significant positive relationship with CI (r = 0.820***) and MCD(001) (r = 0.740***) and a negative relationship with SSA (-0.60**; significant at the 0.01 level). The median value of DT was lower for the silt-fraction kaolinite (500°C) than the clay-fraction kaolinite, possibly indicating a higher degree of structural disorder in the silt-fraction kaolinite. There is significant negative relationship between DT for kaolinite in the clay fraction and K and Mg contents (Table 4), which further suggests that K and Mg are concentrated mainly in smaller kaolinite crystals.
Chemical Composition of Kaolinite
The chemical composition of the kaolinite for the silt and clay fractions is presented in Table 4. For most samples, the SiO2 and Al2O3 contents in kaolinite are within the range of the published values (Newman and Brown, 1987). In general SiO2 and Al2O3 contents were lower and Fe2O3 content was higher in the clay-fraction kaolinite compared with the silt-fraction kaolinite. Small amounts of Ti, K, and Mg were also present in both silt and clay fractions of all samples. Some of these elements may be present in the form of impurities admixed with the kaolinite. It is well recognized that the selective dissolution method has limitations, and some impurities (crystalline and amorphous) may have dissolved with kaolinite during NaOH extraction treatments of silt and clay fractions.
The most striking feature in the chemical composition of soil kaolinite compared with standard kaolinite is the presence of high amount of Fe. The Fe2O3 content in the clay-fraction kaolinite ranged from 10.9 to 28.9 g kg-1. The association of Fe with kaolinite has been observed in several previous studies (Jepson and Rowse, 1975; Singh and Gilkes, 1992). It has been shown using a number of independent techniques that Fe is present in the Fe3+ form and substitutes for Al in the octahedral sheet of kaolinite (Meads and Malden, 1975; Herbillon et al., 1976; Singh and Gilkes, 1992). We did not observe correlation between the concentration of the Fe in the parent rock and the substitution of Fe in kaolinite. For example, the Fe2O3 values of 19.1 and 20.4 g kg-1 in kaolinite from Samples 14 and 15, respectively (soils originated from basalt), were close to the mean (19.1 g kg-1) and median (18.8 g kg-1) values of all samples.
Titanium enrichment was observed in the clay-fraction kaolinite (median = 2.8 g kg-1) compared with the silt-fraction kaolinite (median = 0.053 g kg-1). A similar observation was made by Nagelschmidt et al. (1949) in Georgia kaolinite and they attributed the origin of Ti to the presence of anatase. Anatase was identified by XRD in the clay fraction of all the samples and its content ranged between 0.21 to 4.70% (Melo, 1998). Some of the Ti associated with kaolinite is also believed to be either substituting in the structure of kaolinite (Dolcater et al., 1970; Jepson and Rowse, 1975) or present as a discrete surface-sorbed form (Weaver, 1976).
The lowest levels of K and Mg were associated with kaolinites in highly weathered soils developed from sediments of the Barreiras Group. On the other hand, the kaolinites from younger soils contain higher amounts of nonexchangeable K and Mg (Table 3). Among these soils, C horizon kaolinite showed higher K content. Despite the lower level of K in the parent material (sandstone) of Sample 12, kaolinite from this sample contains relatively high amounts of K and Mg compared with kaolinite in the soils that formed in basalt (Samples 14 and 15). The kaolinites with higher Mg content also had higher K and Mn content (r = 0.700*** and 0.840***, respectively).
Transmission Electron Microscope and Scanning Electron Microscope Analyses of a Single Crystal
The high Fe2O3 levels in the clay-fraction kaolinite, determined by chemical analysis of the 5 M NaOH extracts, were confirmed by single crystal analysis using EDS (Table 5). Some very high values of Fe2O3 observed by EDS analysis of the silt-fraction kaolinites are possibly due to contamination resulting from surface-sorbed Fe hydroxides because the EDS analyses were carried out using samples that had not been treated with dithionite–citrate–bicarbonate. Samples 14 and 15, which showed the highest Fe content, contained higher amounts of hematite than other samples (Melo, 1998).
|
Potassium and Mg values from EDS analysis are generally much higher than the values obtained by the chemical analysis. The contents of K and Mg are low, and because of very low x-ray count rates the values may not be precise.
Morphological Characteristics of Kaolinite Crystals
The size and shape of kaolinite in representative samples were determined by TEM (Fig. 1). We examined between 80 and 120 kaolinite crystals per sample. The percentage of kaolinite crystals with six euhedral faces varied from 1 to 17% (Table 6). The hexagonal form is commonly observed in kaolin deposits but is rarely observed in highly weathered soils (Hughes and Brown, 1979; Singh and Gilkes, 1992). About 50% of the kaolinite crystals are anhedral (no euhedral faces) with elongated, rounded, subrounded or ill-defined morphologies. Kaolinite crystallization in the presence of other minerals, organic ions, and inorganic cations, other than Si and Al, may produce nonhexagonal crystal forms. Also, with increasing weathering, kaolinite crystals tend to lose euhedral faces, especially in near-surface samples. More than 50% of kaolinite in Samples 12, 13, and 15 have euhedral faces and >25% crystals contain five or six euhedral faces (Table 6). In these samples, kaolinite crystal morphology closely resembles the pseudohexagonal shape commonly observed in geological deposits (Fig. 1).
|
|
A mineral with a tubular form, present in the C horizon of the Humitropept developed on granite (Sample 19) (Fig. 1), was found to be halloysite by the formamide test (Churchman et al., 1984). Halloysite commonly occurs in the saprolites developed on granite (Allen and Hajek, 1989) and being less crystalline is more susceptible to weathering.
Particles with a higher number of euhedral faces (five and six) tend to have higher concentrations of K and Mg (Table 3). The correlation coefficients between the proportion of particles with five faces and MCD(001), surface area, and dehydroxylation temperature (r = -0.8* [significant at the 0.1 level], 0.8*, and -0.7*, respectively) indicate that particles with more euhedral faces are less thick, resulting in a greater surface area and lower DT.
The dimensions of the kaolinite crystals vary both in samples and between samples (Table 7). The longest width (henceforth called a dimension) and the shortest width (the b dimension) were measured for about 80 to 120 crystals for each sample. Kaolinite in Sample 13 showed the smallest crystal size and DT and largest SSA. The values obtained for the kaolinite crystals in the study are close to those reported by Singh and Gilkes (1992). The particle size in the a direction is 34 to 44% larger than in the b direction (Table 7). The values greater than 1 for a/MCD and b/MCD indicate preferential growth of kaolinite along the a and b directions in relation to the c axis (thickness).
|
Kaolinite in the silt fraction appears as thick and stable large flakes, which remain intact even after 0.2 M NaOH and ultrasonic dispersion treatments. Kaolinite in the silt fraction of the C horizon of a Cambisol (Sample 19) presents as mica pseudomorphs. The presence of such pseudomorphs suggest that kaolinite may have formed directly from the weathering of biotite (Pinto et al., 1972; Rebertus, et al., 1986; Graham et al., 1989). In kaolinite formed from biotite weathering, K ions may be retained in residual micaceous layers, and possibly due to the presence of such layers, higher amounts of K were found in the silt-fraction kaolinite of Sample 19 (Table 3).
Crystallographic Predictors of Potassium and Magnesium Contents
According to Lee et al. (1975), the presence of K in kaolinite is because of inclusion of micaceous layers. These authors observed occluded and interleaved micas, revealed by the presence of 1.0-nm spacing in kaolinite (0.7-nm spacing) by high-resolution TEM. In the present study, the AI of the (001) and (002) diffraction peaks may indicate the interstratification of 2:1 layers in kaolinite. Asymmetry in the (001) peak may be partly because of an increasing Lorentz-polarization factor with a decreasing 2 angle (Klug and Alexander, 1954). The plot of AI for the (001) peak against K content shows two distinct groups (Fig. 2a). Kaolinite in Group I samples (11, 13, 16, 17, 18, 19, 20, and 21) have higher K contents (K2O contents between 0.71 to 3.66 g kg-1) than the Group II, which consists of the remaining samples with K2O contents <0.55 g kg-1. The linear regression between AI and K content, for samples of Group I, is statistically significant (R2 = 0.980***) (Fig. 2a). The high AI values for Group I are probably because of interstratification of biotite layers, given the high K and Mg contents (Table 3). For Group I, the AI also has a significant positive relationship with SSA (r = 0.8*), indicating more prevalent interstratification of mica in smaller kaolinite particles. For Group II, the relationship between AI and K content is not very strong (R2 = 0.570***), partly because of the lack of precision in the AI measurement (reduced WHH of the peak) and the narrow ranges for the AI values (0.16 to 0.21) and K content (K2O = 0.15–0.55 g kg-1). Similar to K, AI also increased with increasing Mg contents but the relationship was rather poor for both groups (Fig. 2b).
|
There was a highly significant correlation between mica content and K (r = 0.66**) and Mg (0.800***) contents, which suggests that most of these elements are associated with micas. There was a negative relationship between K and Mg contents, and MCD derived from the 001 kaolinite diffraction peak (Fig. 3). The results in Fig. 3 and the positive correlation of SSA with K and Mg contents (Table 4) indicate that smaller kaolinite crystals generally contain greater amounts of K and Mg. The negative relationships of DT with K and Mg contents in kaolinite also support the observation that K and Mg are concentrated in smaller kaolinite crystals. It is possible that because of the presence of occluded and interstratified layers of mica, crystal growth of kaolinite may have been retarded.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Kaolinite crystals are predominantly elongated, rounded and subrounded, with a relatively lower proportion of hexagonal particles. There is no significant correlation between the form of the particles and other kaolinite properties including their K and Mg contents.
The smaller kaolinite crystals are generally poorly crystalline and have higher K and Mg contents and lower DT. There is a close relationship between the AI of the (001) diffraction peak and K content in the kaolinite for the relatively young soils. The K and Mg in these soil kaolinites are possibly in residual micaceous layers interleaved in the kaolinite structure and thus protected from further weathering.
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 8, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIAL AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
