Microporous and Mesoporous Materials 61 (2003) 127–136

www.elsevier.com/locate/micromeso

Sorption characteristics of zinc and iron

by natural zeolite and bentonite
A.S. Sheta *, A.M. Falatah, M.S. Al-Sewailem, E.M. Khaled, A.S.H. Sallam
Soil Science Department, College of Agriculture, King Saud University, P.O. Box 2460, Riyadh 11451, Saudi Arabia
Received 22 July 2002; received in revised form 15 October 2002; accepted 16 October 2002

Abstract

Understanding the sorption process in natural zeolites is necessary for effective utilization of these minerals as
nutrient adsorbents and consequently as controlled releases of plant nutrients. This research was undertaken to
characterize the ability of five natural zeolites and bentonite minerals to adsorb and release zinc and iron. The
potential for sorption of these ions were evaluated by applying the Langmuir and Freundlich equations. Zinc
sorption data followed the Langmuir equation. The ability for Zn sorption were in the following order: chabaz-
ite > analcime > clinoptilolite1 > bentonite > clinoptilolite2 > phillipsite. Diethylene triamine pentaacetic acid (DTPA)
extractable Zn decreased with the increase in successive extractions. All sorbed zinc was extracted by DTPA in
most zeolite species after four successive extractions while only 50% was readily extractable from chabazite. Iron
sorption data followed the S-type isotherm and the results were described by a Freundlich adsorption model. The
iron activity ratio (Feox /Fed ) of sorbed Fe indicated the dominance of amorphous or poorly crystalline phases with
zeolites and crystalline iron oxide phases with bentonite. The results suggest that natural zeolites, particularly
chabazite and bentonite minerals, have a high potential for Zn and Fe sorption with a high capacity for slow release
fertilizers.
Ó 2003 Elsevier Inc. All rights reserved.

Keywords: Zeolite minerals; Bentonite; Zn and Fe sorption; Amorphous iron

1. Introduction the most effective cation exchangers known and

they have two to three times the cation exchange
Zeolites and bentonite are naturally occurring capacity (CEC) than that of most smectites and
structured and phyllosilicate minerals respectively, vermiculite [1]. Agriculture applications related to
with high cation exchange and ion adsorption ca- ion exchange, adsorption and desorption of ions by
pacity. In particular crystalline zeolites are some of zeolites have been reported by many investigators
[2–4]. NHþ 4 -exchanged clinoptilolite added to light
and medium textured soils produced a positive
growth response in radishes when banded on soil
[5]. Natural zeolites indicated a potential use as an
*
Corresponding author. Fax: +966-1-476-8440. NHþ þ
4 adsorbent and a controlled release NH4 fer-
E-mail address: [email protected] (A.S. Sheta). tilizer [3]. Coarse texture arid land soils are low

1387-1811/03/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S1387-1811(03)00360-3
128 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136

to very low in extractable micronutrients and many 2. Experimental

are regarded as potentially deficient. However,
some micronutrient compounds have been added Five naturally occurring zeolite minerals and
to the soils either directly or as an incidental com- one bentonite were used in this study namely,
ponent of other fertilizers. Considerable research clinoptilolite1 and clinoptilolite2 supplied by Ab-
has been conducted concerning the factors which erhill
s Holdings Inc. 1 Abbotsford, BC, Canada
influence the adsorption and desorption of Zn and and Axis Trade Corp., 1 Arizona, USA, respec-
Fe in soils. Zinc adsorption was found to be pH tively. Minerals Research, 1 Clarkson, NY, USA,
dependent [6], related to soil CEC [6,7]. Several supplied phillipsite, chabazite and analcime.
investigators have reported that the adsorption Baroid Saudi Arabia Ltd., 1 Dammam, supplied
mechanism is a major contributing factor to low the bentonite mineral. The locations, mineralogi-
concentrations of Zn in soil solution [8,9]. Zinc cal properties, CEC (cmol kg1 ), surface area
sorption on soils is affected by the soil pH, clay (m2 g1 ), Si/Al ratio and total Fe2 O3 and ZnO
minerals, and Fe-, Al-, and Mn-hydrous oxides, (g kg1 ) are given in Table 1. Representative zeo-
organic material, and carbonates [10]. Adsorption lites and bentonite samples were used without any
mechanisms suggested include the ion exchange [6] chemical pretreatment. Impurities such as quartz,
or physisorption and chemisorption [11]. Soil pH mica, feldspars were identified using X-ray dif-
affects Zn adsorption, either by changing the num- fraction. In preparation for sorption and extrac-
ber of adsorption sites or by changing the con- tion experiments, the samples were grinded in an
centration of the Zn species which is preferentially agate mortar and passed through a 1.0 mm sieve.
adsorbed [12]. Desorption of native and added zinc The zinc sorption experiment were carried out
from a range of New Zealand soils in relation to using the batch equilibrium by weighing triplicate
soil properties has been studied by Singh et al. [13]. 1.0 g samples in 50 ml centrifuge tubes. The initial
They reported that CEC and organic carbon were Zn concentrations added to the samples are 0.0,
the dominant soil variables contributing to sorp- 5.0, 25.0, 50.0, 100.0, 250.0 and 500.0 mg l1 pre-
tion or desorption of zinc. Seasonal precipitated pared from ZnSO4 . Twenty ml of the initial con-
iron oxides in vertisol of Southeast Texas were centrations were added to each sample, then
studied by Golden et al. [14]. They reported that suspensions were shaken for 2 h at constant room
iron oxides precipitated on exposed surfaces on ped temperature (20 °C) followed by centrifugation for
surfaces and within soil pores were relatively 15 min. The zinc concentrations in equilibrium
poorly crystallized while those precipitated on rice solutions were measured by atomic absorption
root surfaces were well crystallized. They refer the spectrophotometry (AAS) according to [16]. The
presence of poorly crystallized oxides to the pres- amount of Zn sorbed (mmol kg1 ) was calculated
ence of soluble Si and P during flooding. The by the difference between Zn added and that re-
presence of Mn lowers the crystalinity of geothite mained in the equilibrium solutions.
formed from slow oxidation of FeCl2 solution at The Langmuir equation was used to describe the
pH 7 [15]. Sorption and desorption characteris- Zn sorption. The model used is
tics of micronutrients (Fe, Zn, Mn, Cu and B) C=x=m ¼ 1=kb þ C=b ð1Þ
by natural zeolite minerals is almost scarce in
literature, despite the relative importance of zeo- where C and x=m are the equilibrium Zn concen-
lites as natural carrier for nutrients. The objec- tration (mmol l1 ) and the amount of Zn sorbed
tives of this research included (1) the evaluation (mmol kg1 ), respectively. The empirical constants
of Zn and Fe sorption characteristics by five nat- b and k are related to the adsorption maximum
ural zeolites and bentonite minerals, and (2) the
quantification of retained Zn and Fe forms through
the extraction using diethylene triamine pen- 1
Trade names and company names are included for the
taacetic acid (DTPA) and selective dissolution benefit of the reader and do not imply endorsement or
methods. preferential treatment of the product.
 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136 129

Table 1
Selected chemical analyses and textural properties of zeolites and bentonite minerals used in the study
Location Impuritiesa Surface areab Si/Al CECc Total (g kg1 )
(m2 g1 ) ratio (cmol kg1 ) Fe2 O3 ZnO
Clinoptilolite1 Southwestern Quartz, feld- 407.9 5.50 98.4 13.4 0.08
USA spar, mica
Clinoptilolite2 Cucurpe, Quartz, 428.3 5.82 75.6 13.3 0.07
Sonora Mexico mordenite
Phillipsite Pine Valley, Mica 961.7 3.87 223.0 19.3 0.09
Nevada
Chabazite Christmas, Mica, quartz 1100.0 4.54 195.0 18.1 0.05
Arizona
Analcime Barstow, Quartz 313.0 3.41 39.0 15.7 0.09
California
Bentonite Dammam, Quartz, mica 1009.0 2.32 88.0 44.9 0.12
Saudi Arabia
a
X-ray diffraction.
b
Surface area by ethylene glycol monoethyl ether method.
c
CEC––ammonium acetate method.

(mmol kg1 ) and bonding strength (l mmol1 ), re- the CBD (sodium citrate–dithionite–bicarbonate)
spectively. This equation will be used to compare method [19] for the free Fe oxide form (Fed ) de-
the adsorption maximum (b) between zeolites and termination. Amorphous and poorly crystalline Fe
bentonite since these adsorbents and the experi- oxides form (Feox ) were extracted from another
mental conditions are identical as proposed by duplicate sample using NH4 -oxalate in darkness at
Veith and Sposito [17]. The sorbed Zn was pH 3 [20,21]. Soluble Fe in all extracts was mea-
sequentially extracted four times with DTPA, ac- sured by AAS [22].
cording to the method described by Lindsay and
Norvell [18]. Zinc in all extracts was measured by
AAS. 3. Results and discussion
Sorption of Fe was carried out on the five
mentioned zeolite species and bentonite samples Zinc sorption isotherms of Zn sorbed x=m
by the same method as already described for Zn. (mmol kg1 ) and the equilibrium Zn concentra-
The initial concentrations are 0.0, 5.0, 25.0, 50.0, tions C mmol l1 are presented in Fig. 1. The
250.0 and 500.0 mg l1 prepared from FeCl2 Æ pattern of the isotherms are quite similar with a
3H2 O salt. Six replicates were prepared in the ad- slight difference between zeolites and bentonite. At
sorption experiment in order to characterize the lower initial concentrations, the isotherms have a
sorbed Fe. The Freundlich model was used to relatively high slope whereas at higher concentra-
describe Fe sorption by zeolites and bentonite tions the slope was relatively low with a defined
minerals. The linear Freundlich model is plateau for the adsorption maximum in most of
the samples studied. Sorption isotherms follow the
log x=m ¼ log k þ 1=n log C ð2Þ
L-shaped type similar to that described by Sposito
where x=m is the amount of Fe sorbed [23]. Such adsorption behavior could be explained
(mmol kg1 ), k and n are constants, and C is the by the high affinity of zeolites and bentonite for Zn
equilibrium Fe concentrations (mmol l1 ). The at low concentrations, which decrease as Zn con-
sorbed Fe was characterized as follows: the avail- centration increases. Langmuir constants for Zn
able Fe was extracted from the duplicate sample sorption were calculated from the best fitting
sequentially (four times) using the DTPA extract straight line between C=x=m and C (Table 2). Data
[18]. Other duplicate samples were extracted using of maximum adsorption (b mmol kg1 ) have been
130 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136

Fig. 1. Sorption isotherms of Zn for zeolite species and bentonite.

in the following decreasing order: chabazite > l mmol1 ) was in the order: chabazite > anal-
analcime > clinoptilolite1 > bentonite > clinoptilo- cime > phillipsite > bentonite > clinoptilolite2 >
lite2 > phillipsite. The binding strength values (k in clinoptilolite1. Results also show that chabazite

Table 2
Langmuir equations and constants (b and k) for Zn sorption by zeolites and bentonite
Sample Langmuir equations r2 b (mmol kg1 ) k (l mmol1 )
Clinoptilolite1 Y ¼ 0:0207x þ 0:0109 0.9572 48.31 1.90
Clinoptilolite2 Y ¼ 0:0328x þ 0:0136 0.9925 30.49 2.41
Phillipsite Y ¼ 0:0385x þ 0:0129 0.9978 25.97 2.99
Chabazite Y ¼ 0:0082x þ 0:0006 0.9949 122.00 13.67
Analcime Y ¼ 0:0178x þ 0:0031 0.9831 56.18 5.74
Bentonite Y ¼ 0:0212x þ 0:0087 0.9621 47.17 2.44
 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136 131

mineral has the highest ability for Zn sorption un- tive available Zn to plants in zeolite species or
der all the studied concentrations, while bentonite bentonite mineral was low and almost negligible.
has an intermediate ability. The amount of Zn Moreover, mixing such minerals with soils low in
sorbed by chabazite ranges between 92.8 and 75.8% available Zn cannot support the plant
s need for Zn.
from the added Zn while it was only 74.4–20.2% for On the other hand, data in Table 3 indicated that
phillipsite. These data reflect large differences be- more Zn was extracted after the fourth extraction
tween zeolite minerals for Zn sorption, particularly by DTPA from chabazite treated samples com-
in the case of chabazite mineral. Differences in the pared with other samples. For example, at higher
mineralogical structure of zeolite species and sur- initial Zn added, chabazite releases 97 mg kg1 in
face characteristics could play an important role in the fourth extraction while clinoptilolite1, clinop-
the sorption behavior of Zn ions besides the pres- tilolite2, phillipsite, analcime and bentonite releases
ence of impurities in these natural samples. Data of 45, 27.8, 80.5, 67, and 64.4 mg kg1 , respectively.
the native Zn extracted by successive DTPA ex- Also, the percentage of cumulative extracted Zn (%
tractions (Table 3) decreased drastically after the from sorbed) varied considerably with the sorbed
first extraction and were almost negligible after the amounts and with zeolite species or bentonite.
third extraction in all mineral species. The cumu- Chabazite showed the lowest desorbed percentage
lative amounts of native Zn ranges from 0.91 particularly at high levels of sorbed Zn. Only 53% is
mg kg1 for bentonite to 0.18 mg kg1 for either readily extractable by DTPA after four successive
clinoptilolite1 or clinoptilolite2. Therefore, the na- extractions leaving 47% Zn retained by the mineral

Table 3
Amounts of sorbed Zn and DTPA extractable Zn (four successive extractions) from Zn treated and untreated zeolites and bentonite
P P
Zn added Sorbed Zn DTPA Extractable/ Sorbed Zn DTPA Extractable/
1 1
(mg kg ) (mg kg ) extractable Zna sorbed (%) (mg kg1 ) extractable Zna sorbed (%)
(mg kg1 ) (mg kg1 )
Clinoptilolite1 Clinoptilolite2
0 0 0.18 – 0 0.18 –
100 88.2 43 49 84.2 73 87
500 403 286 71 360.6 303 84
1000 600 547 91 550 445 81
2000 1020 994 97 960 812 85
5000 2610 2272 87 1690 1357 80
10000 2860 2743 96 1867 1558 83
Phillipsite Chabazite
0 0 0.27 – 0 0.31 –
100 74.4 73 98 89.6 72 81
500 379.8 325 86 463.6 302 65
1000 529.2 475 90 819.4 583 71
2000 1016 973 96 1733.4 893 52
5000 1530 1362 89 4410 2801 64
10000 1600 1547 97 7580 3993 53
Analcime Bentonite
0 0 0.35 – 0 0.91 –
100 89 65 73 88.8 46 52
500 457.8 301 66 452.6 257 57
1000 800 683 85 604 454 75
2000 1520 1318 87 1072 899 84
5000 2650 1884 71 2610 2024 78
10000 3660 2560 70 2860 2368 83
a
Summation of DTPA extractable Zn (four successive extractions).
132 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136

even though the mineral showed the highest amount Iron sorption isotherms for zeolite species and
of Zn extracted in the fourth extraction. This find- bentonite are presented in Fig. 2. All the isotherms
ing may reflect clearly the positive role on the pos- could be described by the S-type isotherm [23]. The
sibility of using chabazite as a slow release for Zn. slope of the isotherm initially increased with the
Analcime showed a relatively similar trend since it increase in Fe concentration, but eventually
retained about 30% of the sorbed Zn after four showed a low slope at relatively high equilibrium
extraction treatments by DTPA. These results Fe concentrations. This type of isotherm indicates
agrees well with bonding energy data calculated that at lower concentrations the surface has a low
from the Langmuir model (Table 2). Other minerals affinity which increases at higher concentration
show a relatively low ability to retain Zn against the [23]. Nevertheless, the S-type isotherm has rarely
extraction by DTPA, for example, phillipsite re- been observed for the adsorption of heavy metals
leases up to 97% and clinoptilolite1 about 96% in soil [24]. There were some differences between
at the highest sorbed amounts. the studied samples, particularly at high concen-

Fig. 2. Iron sorption isotherms for zeolite species and bentonite.

 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136 133

trations, that could be explained by the unique and maghemite (c-Fe2 O3 ) crystalline minerals in
framework structure of zeolite species and the FeCl2 –NH4 OH systems at pHs of 6.0 and 8.0, re-
phyllosilicate structure of bentonite. For example, spectively. Therefore, it was likely that the appli-
at the highest Fe applications clinoptilolite1 seems cation of Fe to different zeolites or bentonite
to retain the highest amount, followed by ben- minerals may lead to oxidation and precipitation
tonite, while analcime was the lowest one. Sorp- of different forms of Fe. However, no attempt was
tion data followed the Freundlich adsorption made to protect the systems from possible oxida-
equation (Table 4) and the fit was better for all the tion by air or even changing the suspension pH
studied mineral species in comparison with the fit during the experiment. The nature of precipitates
for the Langmuir equation. The results show that was attributed to the influence of different surface
most of the obtained plots were quite similar in properties present in the samples including the
shape and varied by the Freundlich sorption con- accessory minerals, oxides, hydroxides and the
stants, i.e., log k and n (Table 4). The calculated n amorphous aluminosilicates. Data of Fed (Table 6)
value was qualitatively related to the distribution indicated that the untreated zeolite species and
of site-bonding energies [25]. The n values for all bentonite have considerable amounts of native
the studied mineral samples were lower than 1, Fed and the highest amount was extracted from
which may indicate that the distribution of bond- phillipsite (6760 mg kg1 ) and bentonite (3720
ing energies will vary with adsorption density [26]. mg kg1 ). Other zeolite minerals have relatively
The small differences encountered in Freundlich low contents of native Fed , which range from 1000
sorption constants could be related to the hetero- mg kg1 for chabazite to 798 mg kg1 for clinop-
geneity of sorption sites in zeolites and bentonite. tilolite1. Data also indicated that the amounts of
The close similarity of these constants for ben- Fed increased with the increase of sorbed Fe and
tonite and zeolites may indicate that the unique this was clear from the differences between Fe in
structure of different zeolites have no clear role in treated and untreated samples. Bentonite showed
the sorption behavior of Fe added under the ex- the highest Fed from the sorbed Fe compared with
perimental conditions in comparison with the zeolite minerals which may indicate the possible
bentonite structure. The percentage Fe sorbed/ role of bentonite surfaces for precipitation of ad-
added data (Table 5) were increased with the in- ded Fe to form free iron oxides compared with
crease in initial Fe added up to 250 mg l1 . It then zeolite surfaces. In that respect Golden et al. [14]
decreased for all samples. It appears that most of found that iron oxides precipitated on exposed
the added Fe at initial concentrations <250 mg l1 surfaces of flooded montmorillonitic soils within
could be retained on the exchange sites or precip- soil pores were relatively poorly crystallized while
itated as insoluble Fe compounds. The extent of those precipitated on rice-root surfaces were well
such processes was related to the characteristics of crystallized. On the other hand, data of amor-
zeolites and bentonite minerals as well as to the phous Fe extracted by ammonium oxalate (Feox )
chemical composition of the equilibrium system. from untreated samples indicated that phillipsite
In that respect Krishnamurtti and Huang [27] re- contains relatively high amounts (3249 mg kg1 )
ported the formation of lepidocrocite (c-FeOOH) followed by bentonite (1155 mg kg1 ). Also,

Table 4
Freundlich sorption equations and constants (log k and n) for Fe sorption on zeolites and bentonite
Mineral Equations r2 log k (mmol kg1 ) n (kg l1 )
Clinoptilolite1 Y ¼ 1:562x þ 0:7961 0.8576 0.796 0.64
Clinoptilolite2 Y ¼ 1:1507x þ 1:4333 0.7581 1.433 0.87
Phillipsite Y ¼ 1:5332x þ 0:607 0.7449 0.607 0.65
Chabazite Y ¼ 1:1909x þ 1:210 0.8248 1.210 0.84
Analcime Y ¼ 1:0864x þ 1:2766 0.8991 1.277 0.92
Bentonite Y ¼ 1:3166x þ 1:2055 0.8313 1.210 0.76
134 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136

Table 5
Amounts of Fe sorbed, and DTPA extractable Fe (four successive extractions) for zeolites and bentonite minerals
P P
Initial x=m (mg kg1 ) Sorbed/added DTPA x=m (mg kg1 ) Sorbed/added DTPA
a
concentration (%) extractable Fe (%) extractable Fea
(mg l1 ) (mg kg1 ) (mg kg1 )
Clinoptilolite1 Clinoptilolite2
0.00 – – 35.6 – – 20.6
5.0 23.13 23.1 51.2 36 36.0 36.4
25.0 335.2 67.0 128.0 336.4 67.3 166.4
50.0 797.4 79.7 406.4 796 79.6 389.4
250.0 4261.2 85.2 1361.0 4510 90.2 1390.0
500.0 7600 76.0 1663.6 5894 58.9 1542.2
Phillipsite Chabazite
0 – – 10.0 – – 19.4
5.0 10.8 10.8 17.20 37.4 37.4 35.8
25.0 346.7 69.3 235.2 295.6 59.1 181.0
50.0 600 60.0 539.0 595 59.5 532.2
250.0 4391.3 87.8 1700.7 4393.8 87.9 1605.2
500.0 6352 63.5 1610.2 5520 55.2 2813.6
Analcime Bentonite
0 – – 26.8 – – 13.8
5.0 48.4 48.4 34.2 29.4 29.4 31.4
25.0 296.6 59.3 141.8 363.8 72.8 259.0
50.0 407.8 40.8 376.8 836 83.6 405.0
250.0 3930 78.6 1018.6 4139.3 82.8 1629.0
500.0 5300 53.0 807.0 – – 2320.0
a
Summation of DTPA extractable Fe (four successive extractions).

amounts Feox increased with the increase in first extraction followed by clinoptilolite1 and
amounts of sorbed Fe and the increase was more bentonite. In the fourth extraction clinoptilolite1
pronounced in all zeolite minerals compared with showed the highest extractable Fe followed by
bentonite, particularly at low concentrations of analcime and chabazite. The cumulative concen-
applied Fe. Data of the iron activity ratio (Feox / trations of native Fe extracted from zeolite min-
Fed ) were <1 in all the untreated samples reflecting erals ranged from 35.6 mg kg1 for clinoptilolite1
the dominance of well crystalline Fe oxides. to 10.0 mg kg1 for phillipsite and for bentonite
Treated zeolite samples have iron activity ratios >1 (13.8 mg kg1 ). Amounts of extractable Fe from
in most cases which may reflect the dominance of treated samples increased in the first extraction in
amorphous or poorly crystalline Fe oxides phases all samples with the increase in initial Fe applica-
as a result of the Fe application to zeolite species. tion and with the increase in the sorbed Fe. There
Bentonite shows iron activity ratios <1, indicating was a general decrease in the amount of extract-
high amounts of crystalline Fe oxides (Fed ) and able Fe with successive extraction, but all samples
less amounts of poorly crystalline and amorphous indicated relatively high amounts even after the
oxides (Feox ). Therefore, these data clearly suggest fourth extraction. For example, at the highest ap-
the existence of a variable role of zeolite and plication treatment (500 mg l1 ) about 160 mg kg1
bentonite mineral surfaces on the formation of was extracted from treated chabazite followed
different Fe forms as a result of the Fe applica- by bentonite (153 mg kg1 ) while other zeolites
tion. Data of the native Fe extracted by successive showed low extractable Fe that range from 54
DTPA extractions (Table 3) decreased with suc- mg kg1 for clinoptilolite2 to 68 mg kg1 for anal-
cessive extractions in most of the cases. Analcime cime. The data also indicated that the cumulative
showed the highest extractable native Fe in the amounts of extractable Fe largely increased with
 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136 135

Table 6
Amounts of CBD and NH4 -oxalate extractable Fe (mg kg1 ) from Fe treated and untreated zeolites and bentonite minerals
Treatment Untreated (0) Initial Fe concentration (mg Fe l1 )
5.00 25.0 50.0 250.0 500.0
Clinoptilolite1
Fed (mg kg1 ) 798 820 1000 1348 3780 5800
Fed treated–Fed untreated – 22 202 550 2982 5002
Feox (mg kg1 ) 236 293 697 1065 4410 6825
Feox treated–Feox untreated – 57 461 829 4174 6589
Feox /Fed 0.296 2.59 2.28 1.51 1.40 1.32
Clinoptilolite2
Fed (mg kg1 ) 1180 1236 1524 1760 4020 5160
Fed treated–Fed untreated – 56 344 580 2840 3980
Feox (mg kg1 ) 570 630 975 1425 4860 6450
Feox treated–Feox untreated – 60 405 855 4290 5880
Feox /Fed 0.483 1.07 1.18 1.47 1.51 1.48
Phillipsite
Fed (mg kg1 ) 6760 6840 7160 7500 10100 10760
Fed treated–Fed untreated – 80 400 840 3340 4000
Feox (mg kg1 ) 3249 3345 3750 4225 7125 7590
Feox treated–Feox untreated – 96 501 976 3876 4341
Feox /Fed 0.48 1.2 1.25 1.34 1.16 1.09
Chabazite
Fed (mg kg1 ) 1000 1076 1260 1560 4100 7480
Fed treated–Fed untreated – 76 260 560 3100 6480
Feox (mg kg1 ) 143 195 483 882 3923 4725
Feox treated–Feox untreated – 52 340 739 3780 4582
Feox /Fed 0.14 0.68 1.31 1.32 1.22 0.71

Analcime
Fed (mg kg1 ) 840 900 1170 1472 3390 4120
Fed treated–Fed untreated – 60 330 632 2550 3280
Feox (mg kg1 ) 518 576 1002 1500 4215 5295
Feox treated–Feox untreated – 58 484 982 3697 4777
Feox /Fed 0.62 0.97 1.47 1.55 1.45 1.46
Bentonite
Fed (mg kg1 ) 3720 3829 4200 4800 7520 11000
Fed treated–Fed untreated – 109 480 1080 3800 7280
Feox (mg kg1 ) 1155 1252 1402 1920 3765 8400
Feox treated–Feox untreated – 97 247 765 2610 7245
Feox /Fed 0.31 0.89 0.52 0.71 0.69 0.99
Fed ––Free iron oxides (Fe extracted by the CBD method).
Feox ––Amorphous or poorly crystalline iron (Fe extracted by NH4 -oxalate in darkness).

the increase of the sorbed Fe in all zeolite minerals, were calculated between the amounts of Fed or
and the extent of extractable Fe depends on the Feox and the cumulative DTPA extractable Fe
type of zeolite. For bentonite, relatively high cu- from the sorbed Fe. The results indicated that
mulative extractable Fe was obtained from the cumulative DTPA extractable Fe was highly cor-
retained Fe, particularly at high levels. These data related with Fed (r2 ¼ 0:961) and Feox (r2 ¼ 0:88).
reflect the importance of the studied minerals as a Therefore, the results suggest that sorbed or
carrier for Fe, even though it was precipitated or freshly precipitated Fe on zeolites or bentonite
oxidized in the system. Correlation coefficients could be a source for increasing available iron.
136 A.S. Sheta et al. / Microporous and Mesoporous Materials 61 (2003) 127–136

4. Conclusions [3] M. Kithome, J.W. Paul, L.M. Lavkulich, A.A. Bomke,

Soil Sci. Soc. Am. J. 62 (1998) 622.
The studied zeolite species (clinoptilolite1 and 2, [4] E.R. Allen, L.R. Hossner, D.W. Ming, D.L. Henninger,
Soil Sci. Soc. Am. J. 60 (1996) 1467.
chabazite, phillipisite and analcime) and bentonite [5] M.D. Lewis, F.D. Moore, K.L. Goldsberry, in: W.G.
showed considerable variations in Zn sorption Pond, F.A. Mumpton (Eds.), Zeo-agriculture: Use of
properties and DTPA extractability. Chabazite Natural Zeolites in Agriculture and Aquaculture, Westview
has the highest ability for Zn sorption, while Press, Boulder, CO, 1984, p. 105.
bentonite has an intermediate ability compared [6] L.M. Shuman, Soil Sci. Soc. Am. Proc. 39 (1975) 454.
[7] E.J. Udo, H.L. Bohn, T.C. Tucker, Soil Sci. Soc. Am. Proc.
with other zeolites. Chabazite retains a relatively 34 (1970) 405.
high percentage of the sorbed Zn against the ex- [8] B.G. Ellis, B.D. Knezek, in: J.J. Mortvedt, P.M. Giordano,
traction by DTPA, even it released more Zn after W.L. Lindsay (Eds.), Micronutrients in Agriculture, Soil
the fourth extraction by DTPA. Iron sorption data Science Society of America, Madison, WI, 1971, p. 59.
[9] U.C. Shukla, S.B. Mittal, Soil Sci. Soc. Am. J. 43 (1979) 905.
were described by a Freundlich adsorption model.
[10] M.A. Elrashidi, G.A. O
Connor, Soil Sci. Soc. Am. J. 46
Most of the sorbed Fe by zeolites was present in (1982) 1153.
the poorly crystalline or amorphous form while [11] M.B. McBride, J. Blasiak, Soil Sci. Soc. Am. J. 43 (1979)
that of bentonite was present in the free iron oxide 866.
form. The results suggest that natural zeolite and [12] N.J. Barrow, J. Soil Sci. 37 (1986) 295.
bentonite minerals have a high potential for Zn [13] D. Singh, R.G. McLaren, K.C. Cameron, Aust. J. Soil Res.
35 (1997) 1253.
and Fe retention. The availability of the retained [14] D.C. Golden, F.T. Turner, H.S. Bhatkar, J.B. Dixon, Soil
Zn and Fe was higher for Zn compared with Fe, Sci. Soc. Am. J. 61 (1997) 958.
and chabazite seems to have the highest ability for [15] Z. Karim, Clays Clay Miner. 32 (1984) 334.
Zn sorption and extractability by DTPA. Ben- [16] D.E. Barker, N.H. Suthr, in: A.L. Page (Ed.), Methods of
tonite has intermediate characteristics for Zn and Soil Analysis: Part 2, second ed., American Society of
Agronomy, Madison, WI, 1982, p. 13.
Fe sorption among the studied zeolite mineral [17] J.A. Veith, G. Sposito, Soil Sci. Soc. Am. J. 41 (1977) 697.
species. [18] W.L. Lindsay, M.A. Norvell, Soil Sci. Soc. Am. J. 42
(1978) 421.
[19] O.P. Mehra, M.L. Jackson, Clays Clay Miner. 7 (1960)
317.
Acknowledgement [20] U. Schwertmann, Z. Pflanzenernaehr., D€ unger und Boden
Kunde 105 (1964)194.
This research was supported in part by the [21] M.V. Fey, J. LeRoux, Clays Clay Miner. 25 (1977) 285.
Saudi Basic Industries Corporation (SABIC). [22] R.V. Olson, R. Ellis Jr., Iron, in: A.L. Page (Ed.), Methods
of Soil Analysis: Part 2, second ed., American Society of
Agronomy, Madison, WI, 1982, p. 301.
[23] G. Sposito (Ed.), The Surface Chemistry of Soils, Oxford
References University Press, New York, 1984.
[24] Y. Yin, H.E. Allen, C.P. Huang, P.F. Sanders, Soil Sci. 162
[1] D.W. Ming, F.A. Mumpton, Zeolites in soils, in: J.B. (1997) 35.
Dixon, S.B. Weed (Eds.), Minerals in Soil Environments, [25] G. Sposito, Soil Sci. Soc. Am. J. 44 (1980) 652.
second ed., Soil Science Soceity of America, Madison, WI, [26] W.P. Inskeep, J. Baham, Soil Sci. Soc. Am. J. 47 (1983)
1989, p. 873. 660.
[2] M.A. Weber, K.A. Barbarick, D.G. Westfall, J. Environ. [27] G.S.R. Krishnamurtti, P.M. Huang, Clays Clay Miner. 37
Qual. 12 (1983) 349. (1989) 451.