properties of organo-mineral complexes ...clays.org/journal/archive/volume 37/37-3-235.pdfvol. 37,...

Clays and Clay Minerals, Vol. 37, No. 3, 235-242, 1989.

PROPERTIES OF ORGANO-MINERAL COMPLEXES FORMED BY DIFFERENT ADDITION SEQUENCES OF HYDROXY-A1,

MONTMORILLONITE, AND TANNIC ACID

A. BUONDONNO, D. FELLECA, AND A. VIOLANTE

Istituto di Chimica Agraria, Universit~ di Napoli, via Universit~ 100, 1-80055 Portici, Italy

Abstract--A study was carried out on the influence of the sequence of addition of montmorillonite, hydroxy-A1 ions, and tannic acid ([A1] = 0.015 M; 6 mmole A1/g of clay; tannic acid/A1 molar ratio = 0.1) on the nature of organo-clay complexes formed at pH 4.5. A negligible amount of tannate was held on the clay surfaces in the absence of A1, whereas in the presence of A1, hydroxy-Al-tannate species were easily adsorbed on clay surfaces. Their distribution on the external surface and in the interlayer space of montmorillonite, however, was a consequence of how the components reacted with each other. The complexes showed broad X-ray powder diffraction peaks at 15.6 to 19.2 A at room temperature. They also showed different behavior to preheating, ethylene glycol solvation, and chemical treatments. Whereas tannic acid showed two prominent exothermic peaks at 390 ~ and 510~ all hydroxy-Al-tannate-mont- morinonite complexes showed a broad exotherm at about 400~ Some complexes showed, in addition, a small inflection between 350 ~ and 420~ The complexes also showed distinct differences in cation-exchange capacity, carbon content, extractable AI, titratable acidity, and mode of aggregation after drying. Key Words--Differential thermal analysis, Hydroxy-Al, Interlayer space, Montmorillonite, Organo clay, Tannic acid, X-ray powder diffraction.

INTRODUCTION

Hydroxy-Al-clay complexes, resembling hydroxy- interlayered smectite and vermiculite in soils, can be synthesized by neutralizing A1 ions in clay suspensions or by mixing hydroxy-A1 polymers or oxides with clay (Rich, 1968; Barnhisel, 1977). The formation, stability, and nature of hydroxy-Al-clay complexes are affected by many factors, such as the amount of A1 saturating the clay surfaces, the pH, and the type of expandable phyllosilicates (Hsu and Bates, 1964; Hsu, 1968; Bry- don and Kodama, 1966; Barnhisel, 1977; Sawhney, 1968). Few studies have been carried out on the influence of organic acids on the Al-interlayering in montmorillonite.

The presence of chelating organic acids significantly retards the crystallization process of AI(OH)3 polymorphs (Kwong and Huang, 1975; Violante and Vio- lante, 1980; Violante and Huang, 1985) and stabilize the chlorite-like complexes in montmorillonite, even in the alkaline pH range (Violante and Violante, 1978b; Violante and Jackson, 1979, 1981; Huang and Vio- lante, 1986). Recent studies have indicated a structural perturbation of interlayer A1 hydroxide sheets in montmorillonite by the presence of organic acids, such as citric, tannic, or humic acid, in the system (Goh and Huang, 1984, 1986; Singer and Huang, 1986). Multi- functional organic ligands of trivalent metal cations (AI and Fe) may act as stabilizing agents in the formation of aggregates. The interaction probably in- volves bridges of hydroxy polymers between clay sur-

Copyright �9 1989, The Clay Minerals Society

faces and the ligand groups of organic molecules (Tisdall and Oades, 1982).

Laboratory research on organic ligands-Al-clay complexes are also of great importance in order to under- stand the properties oforgano-mineral complexes present in soils (McKeague et al., 1986). Various monomers, polymers, and/or solid phases form when AI ions are partially neutralized with a base (Hsu, 1977; Huang and Violante, 1986). Small monomers or large polymers of A1 are differently distributed and held on clay surfaces (Hsu and Bates, 1964; Hsu, 1968; Barnhisel, 1977). Furthermore, different hydrolytic products of A1 can react with an organic ligand, forming species having different shape, size, and charge. Consequently, the nature of ligand-hydroxy-A1 ions and their interaction with clay surfaces are probably influenced not only by the OH/A1 molar ratio and the kind of organic anion (Goh and Huang, 1984, 1986), but also by how the components react with each other.

The aim of the present work was to determine the influence of the sequence of addition of montmorillonite, hydroxy-A1 ions, and tannic acid on the chemical and mineralogical properties of organo-mineral complexes formed at initial pH 4.5.

MATERIALS A N D METHODS

Starting montmorillonite

The < 2-#m fraction ofmontmorillonite from Uri, Sardinia, was separated by sedimentation after dispersion in water. The fraction was washed with 0.1 M HC1 and then saturated with Na by washing with 0.1 M NaCI solution. The Na-exchanged

235

236 Buondonno, Felleca, and Violante Clays and Clay Minerals

montmorillonite was then washed with water, dialyzed until it was chloride-free, and then dispersed in distilled water (16.0 g/liter).

Preparation of complexes The following components were used to prepare the com-

plexes of montmorillonite, hydroxy-A1, and tannic acid: 5.0 g of Uri montmorillonite (312.5 ml of the clay suspension); 30 mmole of A1 (300 ml of 0.1 M AIC13); 3 mmole of tannic acid (M.W. = 1701). Montmorillonite, aluminum, and tannic acid are hereafter abbreviated in the complexes as M, A, and T, respectively.

A complex, indicated as AMT, was obtained by simulta- neously mixing montmorillonite, tannic acid, and aluminum, and slowly adding 0.1 M NaOH (10 ml/min) under vigorous stirring until a pH of 4.5 was reached.

Complexes, designated as AM+T, MT+A, A T + M , were prepared by initially mixing the first named components, namely aluminum and montmorillonite, montmorillonite and tannic acid, or aluminum and tannic acid, respectively, and adding 0.1 M NaOH to pH 4.5. After 2 hr, the last component, namely tannic acid, aluminum, or montmorillonite, was added to the respective suspension previously prepared, and the pH was brought again to 4.5. The complex A + M T was prepared by adding 0.1 M NaOH to the aluminum solution until a pH of 4.5 was reached. After 2 hr, a suspension obtained by mixing montmorillonite and tannic acid was added to the hydroxy-A1 solution previously obtained, and the pH was brought again to pH 4.5. Other samples, prepared as stan- dards, were obtained by mixing (1) montmorillonite and tannic acid, (2) aluminum and montmorillonite, and (3) aluminum and tannic acid and adding 0.1 M NaOH until a pH of 4.5 was reached. These samples are designated as MT, AM, and AT, respectively. Finally, a suspension of montmorillonite (M) and solutions of tannic acid (T) or aluminum chloride (A) were also brought to pH 4.5 by adding 0.1 M HC1 or 0.1 M NaOH.

The volume of all samples was adjusted to two liters to produce aluminum and tannic acid concentrations, respectively, of 0.015 M and 0.0015 M (tannic acid/A1 molar ratio = 0.1; 6 mmole of Al/g clay). A few drops of toluene were added to the suspensions to prohibit microbial growth. The samples were aged for two months at room temperature in polypropylene containers and filtered through Millipore membrane filters (pore size = 0.45 gm). The residues were immediately washed with deionized water, resuspended in water, sonified, and freeze-dried. Only the sample containing tannic acid (T) was freeze dried without filtration through Millipore membrane filter. No precipitate was collected from the sample containing hydroxy-Al ions (A). After drying, the samples were very lightly ground to pass a 0.5-mm sieve before use.

Examination of the complexes Al in the filtrates was determined by atomic absorption

speetrophotometry (Perldn Elmer 3030 B). The amount of A1 depleted from solution was determined by the difference between the initial Al concentration and the concentration in the filtrate. Exchangeable A1 was determined by washing 1013 mg of each sample twice with 1 M KC1 (20 ml) and determining A1 in the combined supernatants after centrifuging for 15 min at 15,000 r.p.m. (27,000 g) and filtering through a membrane filter (<0.01 t~m). A1 was also extracted from 100 mg of each sample by 20 ml of 0.2 M acid ammonium oxalate at pH 3.0 or by 20 ml of 0.5 M NaOH (McKeague, 1978). A1 in the supernatants was determined by atomic absorption, as described above. The cation-exchange capacity (CEC) of the complexes was determined by the BaC12-TEA method (Peech et aL, 1962). The carbon content was measured by the

method of Walkley and Black (1934). All analyses were replicated three times.

Oriented aggregate specimens for X-ray powder diffraction (XRD) were obtained by drying aliquots of samples (previously sonified) on glass slides. XRD patterns were recorded for K-saturated complexes at room temperature (R.T.) and after heat treatments (110 ~ 300* or 550~ and for Mg-saturated specimens, which were solvated with ethylene glycol (EG), with a Philips diffractometer using Fe-filtered CoKa (or Ni-filtered CuKa) radiation.

For differential thermal analyses (DTA), 50 mg of the sam- pies was heated under static air in a Netzsch Thermal Analyzer from 25 ~ to 1000~ at a rate of 10*C/min, using alumina as the reference material.

For scanning electron microscopic (SEM) studies, the samples were mounted on aluminum stubs and sputter-coated with gold. A Cambridge $4 Stereoscan instrument was used.

Potentiometric titrations Undisturbed suspensions of the samples (10-20 ml) or dia-

lyzed and freeze-dried complexes (50 mg) were potentiomet- rically titrated to pH 8.0 in H20 or in 1.0 M KC1 (final volume 25 ml) in an atmosphere of N2 gas by adding CO2-free stan- dard 0.05 M NaOH at a feed-rate of 5 gl/min. An automatic titrator Metrohm Herisau E 536, in conjunction with 1-ml automatic syringe buret 655 Dosimat was used. All the titrations were replicated three times.

R E S U L T S AND D I S C U S S I O N

Chemical analyses of the complexes

T h e da t a in Tab le 1 show tha t in the presence o f m o n t m o r i l l o n i t e (complex A M ) 30.7% o f the A1 in i - t ial ly a d d e d r e m a i n e d in so lu t ion , whereas in the presence o f t a n n i c ac id (complex AT) on ly 11.4% o f the A1 was f o u n d in the supe rna t an t . In the h y d r o x y - A l - t a n - n a t e - m o n t m o r i l l o n i t e c o m p l e x e s ( A M T , A M + T , A T + M , M T + A , or A + M T ) , however , > 9 0 % o f the A1 was dep le t ed f r o m solu t ions , a d s o r b e d by m o n t - mor i l lon i t e , or p rec ip i t a t ed as separa te sol id phases . O n the cont ra ry , in the absence o f b o t h m o n t m o r i l - lon i te a n d t a n n i c ac id all A1 was f o u n d in the so lu t ion (no t r epo r t ed in the Tab le 1). Di f ferent quan t i t i e s o f c a r b o n a n d A1 ex t rac tab le by 0.5 M N a O H or by ac id a m m o n i u m oxala te at p H 3.0 were also de tec ted in the complexes .

M u c h m o r e Al was r e m o v e d f r o m the samples by 0.5 M N a O H t h a n by ac id a m m o n i u m oxalate. F r o m the A T c o m p l e x 1630 a n d 1140 m e q o f A1/100 g were ext rac ted , respect ively , by N a O H or a m m o n i u m oxalate. T h i s c o m p l e x was no t comple t e ly so lubi l ized by N a O H af ter one t r e a t m e n t , a n d a ve ry smal l sol id phase r e m a i n e d o n the filter, e v e n af te r two t r e a t m e n t s (1760 m e q o f AI solubi l ized) , i nd ica t ing a ve ry h igh s tabi l i ty to chemica l t r e a t m e n t s o f O H - A l - t a n n a t e species. M o n t m o r i l l o n i t e a n d the M T complex were only slightly so lub i l i zed by c h e m i c a l t r e a t m e n t s (Tab le 1).

T h e a m o u n t s o f A1 ex t rac ted f r o m h y d r o x y - A l - t a n - n a t e - m o n t m o r i l l o n i t e complexes by N a O H ranged f rom 800 to 890 m e q / 1 0 0 g, whereas the a m o u n t s o f A1 ex t rac ted by oxala te r anged f r o m 715 to 765 m e q / 1 0 0 g. T h e c h e m i c a l t r e a t m e n t s d id n o t revea l s ignif icant

Vol. 37, No. 3, 1989 Properties of synthetic organo-mineral complexes 237

Table 1. Total A1 remained in solution, NaOH-extractable, oxalate-extractable, and KCl-exchangeable A1, carbon content and cation-exchange capacity (CEC) of the samples.

Extractable AI (me, q/100 g)

Sample L AI (%)2 NaOH Oxalate KCI Carbon (g/100 g) CEC (meq/100 g)

M -- 70 35 -- -- -- 75.0 AM 30.7 810 640 7 -- -- 27.4 MT -- 60 25 -- 1.03 (2. l ) 4 60.0 AT 11.4 1630 1140 -- 37.8 (40.3) -- AMT 7.2 810 715 11 22.1 (22.7) 111.2 AM+T 6.1 845 765 16 23.4 (24.8) 126,0 AT+M 6.7 805 720 11 17.9 (22.6) 101.7 MT+A 7.8 890 765 14 23.0 (23.3) 135.3 A+MT 6.7 800 755 52 19.5 (23.4) 140.0 Average error (%) 1.4 4.2 3.7 2.4 3.0 2.6 2.9

Symbols of the samples explained in Materials and Methods section. z Percentage of A1 initially added (in the supematant). 3 After Walkley and Black (1934) 30-min oxidation). 4 Determined by modified Walkley and Black (1934) method (150-min oxidation).

differences among the complexes. Small amounts o f A1 were exchanged from the complexes by 1 M KC1 (1.4- 6.5% of A1 extractable by NaOH).

In the hydroxy-Al-tannate-montmorillonite complexes the C content ranged from 17.9 ( A T + M ) to 23.4% (AM+T), i f oxidation by dichromate was carried out for 30 min, according to Walkley and Black method (1934). In contrast, if oxidation was carried out for an additional 120 rain at 80~ a higher C content was found for all the complexes (the data are reported in parentheses in Table 1). The increase of carbon content, however, was negligible for the complexes AMT (from 22.1 to 22.7%) and M T + A (from 23.0 to 23.3%). Greater increases were found for the complexes A T + M (from 17.9 to 22.6%) and A + M T (from 19.5 to 23.4%). The higher stability to oxidative treatment of the latter complexes can be attributed, at least in part, to the stronger aggregation o f the particles in these samples after drying (as is discussed below; Table 2 and Figure 1).

Only 1.0% (or 2.1% after a 150-min oxidative treatment) carbon was found in the complex MT, indicating that in the absence orAl a negligible amount of tannate was held on the clay surfaces. In contrast, for the complex AT, considerable tannate (37.8% C) was coprecipitated with AI.

The CEC of the Uri montmoril lonite was 75.0 meq/ 100 g, but in the presence o f A1 (complex AM) the CEC was reduced to 27.4 meq/100 g, and in the presence o f tannic acid (complex MT) to 60 meq/100 g. The reduction of the CEC of montmoril lonite is thought to be due to interlayering ofhydroxy-A1 (Barnhisel, 1977).

The hyd roxy -Al - t anna t e -mon tmor i l l on i t e complexes showed very high CECs, ranging from 101.7 to 140 meq/100 g. Evidently, the presence of negative sites due to dissociated carboxylic or phenolic groups

of the tannate anions increased the CEC. In fact, the complexes A M T and A T + M , which contained, after a 150-min oxidative treatment, the lowest C contents (22.7 and 22.6% of C), showed the lowest CECs (111.2 and 101.7 meq/100 g, respectively). The complexes M T + A and A + M T , which contained similar C contents (23.3 and 23.4% C, respectively), also showed similar CECs (135.3 and 140 meq/100 g, respectively); however, the CEC of the complex AM + T, which contained the highest C content (24.8%), was relatively low (126 meq/100 g), probably because of a better interlayering of the OH-Al-tannate ions in this sample (vide infra).

Titratable acidity

The acidity values determined to pH 8.0 in HzO or 1 M KCI of the undisturbed suspensions differed mark-

Table 2. Titralable acidity of the undisturbed suspensions and of dialyzed and freeze-dried samples titrated to pH 8.0 by NaOH 0.05 M in HzO or 1 M KC1.

Suspensions titrated in Solid phases titrated in (meq/liter) (meq/100 g)

Sample ~ H20 KCI H20 KCI

A 16.6 15.5 -- - -

T 4.4 4.9 - - - -

A T 6.9 9.8 5.5 25.5 AM 11.5 11.8 14.6 13.8 AMT 4.7 8.1 13.0 75.4 AM+T 5.7 8.5 20,0 92.5 AT+M 5,0 8.6 1,5 7.0 MT+A 5.5 9.2 9.5 70.5 A+MT 6.5 10.3 3.5 16.5 Average error (%) 2.2 1.9 2.9 2.0

l Symbols of the samples explained in Materials and Meth- ods section.


edly (Table 2). About 16.6 and 4.4 meq/l i ter of base was required to titrate in water samples A and T, respectively. In contrast, the total titratable acidity of the complex AT was particularly low (6.9 meq/liter), indicating that the chelation of AI by tannate ligands neutralized many acidic groups.

The A M complex showed a lower ti tratable acidity (11.5 meq/liter) than the AI solution (16.6 meq/liter), probably because large amounts of hydroxy-A1 ions were interstratified (vide infra). Indeed, Violante and Violante (1978a) demonstrated that if OH-A1 ions are fixed between the structural layers of montmori l lonite , a complete t i tration of Al-ions is extremely difficult. The amounts of base consumed in titrating A, T, or A M were almost the same in H20 and in KC1. In contrast, more base was added to neutralize the AT complex in KC1 (9.8 meq/liter) than in HzO (6.9 meq/ liter).

The h y d r o x y - A l - t a n n a t e - m o n t m o r i l l o n i t e complexes showed acidity values which differed strongly from sample to sample. I f the ti trations were carried out in H20, 4.7 meq of base per liter of suspension was consumed to neutralize the complex A M T and 6.5 meq/l i ter to neutralize the complex A + MT. Titrat ions carried out in 1 M KC1 resulted in an increase of the ti tratable acidity values from 8.1 (AMT) to 10.3 (A + MT) meq/liter. Probably the addit ion of a strong electrolyte, such as KCI, weakened some bonds between hydroxy-Al- tannate-montmori l loni te panicles, thereby rendering many acidic sites more accessible and, hence, facilitating the neutralization reactions (Violante and Violante, 1978a).

The h y d r o x y - A l - t a n n a t e - m o n t m o r i l l o n i t e complexes, which were dialyzed until chloride-free and then freeze-dried under controlled conditions, showed acidity values which differed significantly from sample to sample (Table 2). The total NaOH needed to titrate the A T + M and A + M T complexes was only 1.5 and 3.5 meq/100 g, if the neutralization took place in H20, and 7.0 and 16.5 meq/100 g, i f it took place in KC1. The M T + A , AMT, and A M + T complexes required, respectively, 9.5, 13.0, and 20.0 meq/100 g in H20 and 70.5, 75.4, and 92.5 meq/100 g in KC1.

The titratable acidity values in 1 M KCI of the freeze- dried complexes were 4.6 to 7.4 times higher than those determined in 1-120, whereas the acidity values in KC1 of the undisturbed suspensions of the same complexes were only 1.5 to 1.7 t imes higher than those determined in H20 (Table 2). Probably, the addit ion of KC1 weakened many new addit ive bonds, which formed, after drying, between organo-mineral particles.

Whereas the titratable acidity values in 1 M KC1 of the undisturbed suspensions o f the complexes were in the order:

AMT < A M + T < A T + M < M T + A < A + M T ,

those of the same complexes after drying were in the order:

A T + M < A + M T << M T + A < A M T < A M + T .

These findings require an explanation. Evidently, the total acidity of each complex was influenced by the nature o f the hydroxy-Al-tannate ions and by their distribution on the external surfaces and interlayer space of montmori l loni te and, after drying, by the formation, stability, and strength of aggregates formed between the particles.

Indeed, the hydroxy-Al-tannate ions were found to be relatively more uniformly distributed in the interlamellar spaces of montmori l loni te in the AMT and A M + T complexes than in the A + M T , M T + A or AT + M complexes (vide infra). Consequently, the lower ti tratable acidity values ascertained before drying in the A M T and A M + T complexes (Table 2) can be attributed, at least in part, to an incomplete neutralization of better interlayered OH-Al- tannate ions (Vio- lante and Violante, 1978a).

The different nature and distr ibution o f the OH-A1- tannate ions promoted different aggregation of the par- t ides after drying. The stability and strength of the aggregates drastically inhibited a complete neutralization reaction of the acidic sites. In other words, the stronger the aggregation o f the particles, the lower the titratable acidity value. Indeed, electron optical obser- vations showed that all the freeze-dried complexes were characterized by different morphology and state o f aggregation (Figure 1). The A T + M and A + M T complexes, which showed after drying the lowest t i tratable acidity values (Table 2), appeared to be more aggre- gated than the complexes AMT and M T + A (Figure 1), which showed higher ti tratable acidity values (Table 2).

Thermal analyses

Figures 2A and 2B show D T A curves of tannic acid (T), montmori l loni te (M), and the AT, MT, AM, A T + M , and A M T complexes. The thermal decom- posit ion of tannic acid (Figure 2A) was nearly complete by 650~ The two shoulder-like exotherms between 200* and 350~ and the prominent exotherms at 390 ~ and 510~ were probably due to decarboxylation and decomposi t ion of the tannic acid (Schnitzer and Ko- dama, 1977). Two addit ional very small exotherms were observed at about 630 ~ and 870"C. The D T A curve of AT complex (Figure 2A) shows a single broad exotherm between 200 ~ and 620"C and centered at about 550~ indicating that the decomposi t ion of tannic acid coprecipitated with A1 took place over a wide temperature range. A small exotherm at 860~ was also present on the D T A curve o f this complex.

Schnitzer and Kodama (1972) demonstrated that complexes o f fulvic acid containing 14 different mono-


Figure 1. Scanning electron micrographs of the complexes AMT, AT+M, A+MT, and MT+A. Symbols of the complexes explained in Materials and Methods section. Bar indicates 10 •m.

, di-, and trivalent ions had different thermal stabilities, depending on the nature of the cation. They ascertained that the principal exotherm of the Al-fulvic acid complex (pH = 5.0) was at higher temperature than that of fulvic acid. The present results seem to support the findings of Schnitzer and Kodama (1972). In all the hydroxy-Al-tannate-montmorillonite complexes only a broad exotherm was noted, but at about 400~176 The DTA patterns of the A T + M , A M T (Figures 2A and 2B), and M T + A (not shown in Figure 2) complexes contained, in addition, small inflections at 350 ~ 420"1:2 and a small exotherm at 9300--940~ The D T A patterns of the A M + T and A + M T complexes (not reported in Figure 2) had a small exotherm at about 880~ similar to that found on DTA curve of tannic acid (Figure 2A).

These data seem to confirm the findings of Satoh (1976) on natural organo-mineral complexes in volcanic ash soils. Satoh (1976) found that for a clay complex fraction (7.8% C content) containing montmori l- loni te and noncrys ta l l ine mater ia l s the pr inc ipa l exothermic peak was at 295~ with a shoulder at about 3900C, but that the humic acid fraction extracted from this complex with 0.1 M NaOH gave the exothermic

440 420

55O

930

860 AT

100

20 260' 6~ " 9.~0 Temperature (~

410 B 370~

94 AMT

AM

M

130

b 200 ' 600 '950 Temperature (~

Figure 2. Differential thermal analysis curves of." (A) tannic acid (T), AT and AT + M complexes; (B) montmorillonite (M), MT, AM, and AMT complexes. Symbols of the complexes explained in Materials and Methods section.


R.T.

MT 19.2

A M T

AM*T (~AT*M) ,

A § S ~ , I I I 12 10 8 6 4

M(~MT) E.G.

16.~

A M T

AT+M

AM+T

A + M T

i i i

12 10 8

" 17.1/~ R.T. 3 0 0 ~ 5 5 0 ~ E.G.

12.4 ,~, 11.9 15.7

A + M T 11.6

!4.7 16.3 1.1 A M T J A 12 .~ ~

10.7

A M +T,..,,.,~ . . . . . . . . 10 8 6 4 10 8 6 4 10 8 6 4 10 8 6 4

i

6 4

" 9 . 9 / ~

550~ ~ 300~ 14.8

1-~ MT) AM

AMT �9 '

AM +T

AT+M ~ ; I AT+M

A+MT L~_L__~ L----J A+MT L.-- t t ~ 12 10 8 6 4 12 10 .8 6 4

020 CoKa Radiation

Figure 3. X-ray powder diffraction (XRD) patterns at room temperature (R.T.) and after heating at 300* or 550~ of K-saturated or Mg-saturated and ethylene glycol-solvated (EG) samples. XRD patterns of the complex MT+ A are similar to those of the complex AT + M and are not listed here for clarity. Symbols of the complexes explained in Materials and Meth- ods section.

peaks at higher temperature (320 ~ and 4800C). Prob- ably, montmoril lonite surfaces catalyzed the pyrolysis reaction of the organic ligands (Schnitzer and Kodama, 1977).

The DTA curve of the MT complex (Figure 2B)

020 CuKa, R a d i a t i o n

Figure 4. X-ray powder (XRD) diffraction patterns at room temperature (R.T.) and after heating at 300 ~ or 550*(2 of the K-saturated or Mg-saturated and ethylene glycol-solvated (EG) complexes A+MT, AMT, and AM+T after acid ammonium oxalate treatment. Symbols of the complexes explained in Materials and Methods section.

confirms that only a few tannate ions were adsorbed in the absence of Al on the clay surfaces (Table 1). In fact, two barely visible exotherms at about 370 ~ and 500~ were noted on the DTA curve. The endotherm at about 690~ present in montmoril lonite and in the MT complex, was extremely weak in the curves of the AM and the hydroxy-Al - t anna te -montmor i l lon i t e complexes, indicating interlayering (Brydon and Ko- dama, 1966). Further evidences for interlamellar ad- sorptions ofhydroxy-Al-tannate ions were obtained by XRD analyses (vide infra).

X- ray powder diffraction analyses

The XRD patterns of montmoril lonite and the complexes at room temperature, after EG-solvation and heating (300 ~ and 550~ are shown in Figure 3. At room temperature montmoril lonite had a basal spacing of 11.9 ~, which gradually collapsed to 9.6 ,~ on heating to 550~ and expanded with EG to 17.1 ~,. The XRD pattern of the MT complex showed a d-value of 13.2 /k at room temperature and an intense broad peak at about 11.3 ~ after the sample had been heated at 110~ (10.3 ~ for montmorillonite; not shown), suggesting that some tannic acid was retained. Samples of montmorillonite and the MT complex heated at 300 ~ and 550~ or saturated with Mg and solvated with EG gave XRD patterns that were practically the same, evidently because negligible tannate was interlayered (Table 1) or still present after heating (Figure 2).

In the AM complex hydroxy-A1 ions increased the basal spacing of the montmoril lonite from 11.9 to 15.6 /~. This complex expanded only slightly on EG solvation (16.0/~) and partially collapsed after heat treat-


ments (14.1 /k). The strong and relatively symmetric peaks of this sample at 300* and 550~ indicate a suf- ficiently uniform orientation of OH-AI ions in the interlayer space of montmoril lonite (Goh and Huang, 1984, 1986).

The hydroxy-Al - t anna te -mon tmor i l l on i t e complexes typically gave broad and asymmetric XRD peaks from 15.6 (AMT) to 19.2 /~ (A+MT) at room temperature, from 15.0 to 19.2 Zk at 110~ (not shown) and from 16.3 to 19.0 ~ after EG solvation (Figure 3). Whereas the AMT and A M + T complexes after being heated at 300* and 550~ gave small, diffuse but evi- dent peaks (14.5-14.7 and 14.1 /~, respectively), similar to those found for the AM complex at the same temperature, the other complexes (AT+M, A + M T , and M T + A ) gave only humps between 5.0* and 9.0~

The absence of distinct XRD peaks for some hydroxy-Al-tannate-montmorillonite complexes heated at 300 ~ and/or 550~ was probably due to the contin- uous removal of tannic acid on heating which left hydroxy-Al pillars randomly and differently distributed in the interlayers ofmontmorillonite (Singer and Huang, 1986). The broadening of the peaks with increasing preheating temperature was probably due to a random orientation of the polymers in the interlamellar spaces (Goh and Huang, 1986). Evidently, tannate ions inhibited the formation of an uniform distribution of aluminous species in the interlayer space of montmorillonite.

The present data, however, suggest that if AI and montmoril lonite were initially mixed together and the pH adjusted to 4.5 (as for the A M + T and AMT complexes), the aluminous species were more uniformly distributed in the inteflamellar space of clay. In contrast, if large OH-Al-tannate or OH-AI ions, already formed, were added, respectively, to montmoril lonite or montmoril lonite + tannic acid (as for the A T + M and A + M T complexes), the interlayers of montmorillonite were nonuniformly filled.

In fact, the complexes AMT and A M + T not only showed distinct peaks after preheating at 300 ~ or 550~ and the lowest values oftitratable acidity in KCI before drying (Table 2), but also a higher stability to the treatment by ammonium oxalate (Figure 4). Much more interlayer material remained after the chemical treatment in the AMT (d(001) = 14.2 A at room temperature and 11.6 .~ at 550~ and A M + T (d(001) = 14.7 /~ at room temperature and 10.7/~ at 550~ complexes than in the A + M T (d(001) = 12.4/k at room temperature and 10.1 A at 550~ A T + M , and M T + A (not reported in Figure 4) complexes. In the A + M T complex, which gave the greatest d(001) value before oxalate treatment (19.2 A, Figure 3) and the greatest CEC and titratable acidity before drying (Table 2), the OH-Al-tannate ions were probably much more con- centrated and massed at the edges Of montmoril lonite particles and thus more easily extractable by acid am-

monium oxalate solution at pH 3.0 (Figure 4) and more easily neutralized by a base (Table 2).

SUMMARY

In summary, the organo-mineral complexes, which were formed by differently mixing hydroxy-Al, montmorillonite, and tannic acid, showed different chemical and mineralogical properties. The characteristic and peculiar properties of each complex were probably due to (1) the amount and nature of the hydroxy-Al-tannate ions and to their distribution on the external surface and in the interlayer space of montmorillonite, and (2) the formation, mainly after drying, of aggregates of different stability and strength of the bonds among the particles.

REFERENCES

Barnhisel, R. I. (1977) Chlorites and hydroxy interlayered vermiculite and smectite: in Minerals in SoilEnvironments, J. B. Dixon and S. B. Weed, eds., Soil Sci. Soc. Amer., Madison, Wisconsin, 331-356.

Brydon, J. E. and Kodama, H. (1966) The nature of aluminum hydroxide-montmorillonite complexes: Amer. Mineral. 51, 875-889.

Goh, T. B. and Huang, P.M. (1984) Formation ofhydroxy- Al-montmorillonite complexes as influenced by citric acid: Can. J. Soil Sci. 64, 411-421.

Goh, T. B. and Huang, P. M. (1986) Comparison of the influence of citric and tannic acid on hydroxy-Al interlayering of montmorillonite: Clays & Clay Minerals 34, 37- 44.

Hsu, P. H. (1968) Heterogeneity ofmontmorillonite surface and its effect on the nature of hydroxy-aluminum interlayers: Clays & Clay Minerals 16, 303-311.

Hsu, P. H. (1977) Aluminum hydroxides and oxyhydroxides: in Minerals in Soil Environments, J. B. Dixon and S. B. Weed, eds., Soil Sci. Soc. Amer., Madison, Wisconsin, 331-356.

Hsu, P. H. and Bates, T. F. (1964) Fixation of hydroxy- aluminum polymers by vermiculite: Soil Sci. Soc. Amer. Proc. 28, 763-769.

Huang, P. M. and Violante, A. (1986) Influence of organic acids on crystallization and surface properties of precipi- tation products of aluminum: in Interaction of SoilMinerals with Natural Organics and Microbes, P. M. Huang and M. Schnitzer, eds., Soil Sci. Soc. Amer. Spec. Publ. 17, 159- 221.

Kwong, K. F. and Huang, P. M. (1975) Influence of citric acid on the crystallization of aluminum hydroxides: Clays & Clay Minerals 23, 164-165.

McKeague, J. A. (1978) Manual on Soil Sampling and Methods of Analysis, 2nd ed., J. A. McKeague, ed., Cana- dian Soil Survey Committee, 212 pp.

McKeague, J. A., Cheshire, M. V., Andreux, F., and Berthelin, J. (1986) Organo-mineral complexes in relation to pedo- genesis: in Interaction of Soil Minerals with Natural Or- ganics and Microbes, P. M. Huang and M. Schnitzel eds., Soil Sci. Soc. Amer. Spec. Publ. 17, 549-592.

Peech, M., Cowan, R. L., and Baker, J. H. (1962) A critical study of the BaCl2-triethanolamine and ammonium acetate methods for determining the exchangeable hydrogen content of soils: Soil Sci. Soc. Amer. Proc. 26, 37-40.

Rich, C. I. (1968) Hydroxy interlayers in expansible phyllosilicates: Clays & Clay Minerals 16, 15-30.

Satoh, T. (1976) Isolation and characterization of naturally


occurring organo-mineral complexes in some volcanic ash soils: Soil Sci. Plant Nutr. 22, 125-136.

Sawhney, B. L. (1968) Aluminum interlayers in layer sili- cates. Effect of OH/A1 ratio of A1 solution, t ime of reaction, and type of structure: Clays & Clay Minerals 16, 157-163.

Schnitzer, M. and Kodama, H. (1972) Differential thermal analysis of metal-fulvic acid salts and complexes: Geoder- ma 7, 93-103.

Schnitzer, M. and Kodama, H. (1977) Reactions of minerals with soil humic substances: in Minerals in Soil Environ- ments, J. B. Dixon and S. B. Weed, eds., Soil Sci. Soc. Amer., Madison, Wisconsin, 741-770.

Singer, A. and Huang, P .M. (1986) Effect o fhumic acid on Al-interlayering in montmorillonite: Proc. 13th Cong. Int. Soc. Soil Sci. 4, 1478-1479.

Tisdall, J. M. and Oades, J. M. (1982) Organic matter and water-stable aggregates in soils: J. Soil Sci. 33, 141-163.

Violante, A. and Huang, P.M. (1985) Influence of inorganic and organic ligands on the formation of aluminum hydroxides and oxyhydroxides: Clays & Clay Minerals 33, 181-192.

Violante, A. and Jackson, M. L. (1979) Crystallization of nordstrandite in citrate systems and in the presence of montmorillonite: in Proc. Int. Clay Conf., Oxford, 1978,

M. M. Mortland and V. C. Farmer, eds., Elsevier, Am- sterdam, 517-525.

Violante, A. and Jackson, M. L. (1981) Clay influence on the crystallization of AI(OH)3 polymorphs in the presence of citrate, sulfate, or chloride: Geoderma 25, 199-214.

Violante, A. and Violante, P. (1978a) Potentiometric titration of aluminum bentonite in the presence of aluminum hydrous oxide gel: Soil Sci. 126, 136-144.

Violante, A.andViolante, P. (1978b) Influenceofcarboxylic acids on the stability of chlorite-like complexes and on the crystallization ofAl(OH)3 polymorphs: Agrochimica 22, 335- 343.

Violante, A. and Violante, P. (1980) Influence ofpH, concentration, and chelating power of organic anions on the synthesis of aluminum hydroxides and oxyhydroxides: Clays & Clay Minerals 28, 425-433.

Walkley, A. and Black, I. A. (1934) An examination of the Degtjareffmethod for determining soil organic matter, and a proposed modification of the chromic acid titration method: Soil Sci. 37, 29-38.

(Received 11 March 1988; accepted 12 October 1988; Ms. 1771)

properties of organo-mineral complexes ...clays.org/journal/archive/volume 37/37-3-235.pdfvol. 37,...

Documents