R
European Journal of Soil Science, March 1999, 50, 117±125
In¯uence of the nature of clay minerals on the ®xation of
radiocaesium traces in an acid brown earth±podzol
weathering sequence
E. MAESa, A. ISERENTANTa, J. HERBAUTSb & B. DELVAUXa
a
Universite Catholique de Louvain (UCL), unite Sciences du Sol, Place Croix du Sud 2/10, 1348 Louvain-La-Neuve, Belgium, and
Universite Libre de Bruxelles (ULB), Laboratoire de GeÂneÂtique et d'Ecologie veÂgeÂtales, ChausseÂe de Wavre 1850, 1160 Bruxelles,
Belgium
b
Summary
The magnitude of radiocaesium ®xation by micaceous clay minerals is affected by their transformation,
which depends on weathering in soil. The net retention of radiocaesium traces was quanti®ed by
sorption±desorption experiments in the various horizons of four sandy soils forming an acid brown earth±
podzol weathering sequence derived from sandy sediments and characterized by marked changes in
mineral composition. The features of the 2:1 minerals of the four soils, resulting from an aluminization
process in depth and a desaluminization process towards the surface, had a strong in¯uence on Cs+
®xation. Beneath the desaluminization front, which deepens from the acid brown earth to the podzol,
hydroxy interlayered vermiculite was dominant and the 137Cs+ ®xation was the weakest. At the
desaluminization front depth, vermiculite was responsible for the strongest 137Cs+ ®xation. In the upper
layers, smectite appeared in the podzolized soils and the 137Cs+ ®xation decreased. The magnitude in Cs+
®xation therefore appeared as a tracer of the transformation process affecting the 2:1 clay minerals in the
acid brown earth±podzol weathering sequence. This magnitude was positively correlated with the
vermiculite content of the studied soil materials estimated by the rubidium saturation method.
Introduction
Soils retain radiocaesium because they contain a few highly
selective sites associated with the presence of micaceous
minerals (Sawhney, 1972; Eberl, 1980; Cremers et al.,
1988). Micas are common in some soils derived from
sediments and weathered rocks, particularly of the sialic
type. In soils where the transformation of micas is active
these minerals undergo changes in their structure and
properties. The K+ depletion, gain in hydrated exchangeable
cations and oxidation of structural Fe2+ cause the
transformation of biotite into vermiculite and smectite
(Fanning et al., 1989) in which Al interlayering may occur
(Barnhisel & Bertsch, 1989). The retention of 137Cs+ traces
is strongly in¯uenced by these changes, as reported by
Maes et al. (1999) in a laboratory weathering sequence
biotite ® vermiculite ® oxidized vermiculite ® hydroxy
interlayered vermiculite (HIV). These authors found that the
interlayer occupancy by potassium in biotite and hydroxyAl groups in HIV strongly limited Cs+ ®xation. Vermiculite
®xed large amounts of radiocaesium, but the oxidized
Correspondence: B. Delvaux. E-mail: [email protected]
Received 20 January 1998; revised version accepted 27 October 1998
#
1999 Blackwell Science Ltd
vermiculite retained much more radiocaesium in acid
conditions, because of its dioctahedral character or a
greater resistance to acid weathering or both. It was
concluded that Cs+ ®xation in the studied weathering
sequence occurred on vermiculitic sites associated with
micaceous wedge zones.
Such observations suggest that weathering in soil strongly
in¯uences the ®xation of Cs+. Large differences may be
expected not only in the different pro®les of a soil weathering
sequence but also in the various horizons at each site. In acid
forest soils strong mineralogical variations are characteristic.
They are related both to horizon differentiation and to the
distribution of plant roots, which are dynamic weathering
agents in soils (Courchesne & Gobran, 1997).
This paper describes the 137Cs+ retention properties of the A
and B horizons in a soil weathering sequence acid brown
earth±podzol developed in sandy sediments. The soil sequence
is represented by four distinct pro®les in which vermiculite,
smectite and hydroxy interlayered 2:1 clay minerals are
derived from the weathering of mica and chlorite. The
137
Cs+ retention properties are assessed in relation to the
mineralogical variations and the quanti®cation of vermiculitic
minerals.
117
L
118 E. Maes et al.
Figure 1 Schematic diagram of the
evolution of the clay mineralogy in the acid
brown earth±podzol weathering sequence
(adapted from Herbauts, 1982).
Materials and methods
The acid brown earth±podzol weathering sequence
Four soil pro®les studied earlier by Herbauts (1982) were
sampled on the Lower Lias outcrop in southeast Belgium. The
altitude ranges between 320 and 335 m, the annual rainfall
amounts to 1100 mm, and the mean annual temperature is
7.7°C. All the soils were sampled in deciduous forest with
Fagus sylvatica, Quercus petraea and Quercus robur as
dominant species.
The bedrock (calcareous sandstones of Lower Lias age) is
covered by a two-layered sheet: an autochthonous sandy layer
formed by the dissolution of the calcareous bedrock is overlain
by a mixture of this sandy material with loessic silt-sized
particles. The soil sequence developed in the upper material is
as follows: acid brown earth ® ochreous brown earth ®
brown podzolic soil ® podzol, i.e. Dystric Cambisol ® Haplic
Podzol according to the World Reference Base for Soil
Resources (WRB) classi®cation (FAO, 1998). Following a
gradient of increasing podzolization, this sequence of soil
coincides with the evolution of the humus from mull to
moder-mor. This evolution coincides with the morphological
development of the pro®le from Ah-Bw to Ah-Bh and
eventually Ah-E-Bh. The increasing podzolization is well
illustrated by (i) selective chemical extraction of Fe and Al and
(ii) the mineralogical composition of the clay fraction (< 2 mm)
(Herbauts, 1982).
The general evolution of the clay minerals in the sequence is
illustrated in Figure 1 (Herbauts, 1982). In the moderately acid
B horizons, Al interlayering affects the 2:1 clay minerals
produced by the weathering of mica and chlorite: 2:1±2:2
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
hydroxy-Al intergrades are the dominant clay minerals. In the
more acidic surface horizons, aluminium is complexed by
organic compounds and removal of Al interlayers occurs. The
latter process has been called `dechloritization' by Frink
(1969), `de-aluminization' by Vicente et al. (1977) and
`desaluminization' by Herbauts (1982). In the surface
horizons, desaluminization occurs with increasing intensity
towards the more weathered soils, resulting in the dominance
of smectite-like minerals. These smectite-like minerals present
a 1.7±1.8-nm re¯ection in their X-ray diffraction after
saturation with Mg2+ and ethylene-glycol solvation, but
collapse to 1.0 nm when heated (200°C) after saturation with
K+. These minerals are identi®ed as degradation smectites
derived from micaceous minerals. From the ochreous brown
earth to the podzol, the deepening of the desaluminization
front is therefore associated with the advent of smectitic clays.
Whatever the soil weathering stage, the sand and silt fractions
have homogeneous mineralogical compositions dominated by
quartz (Herbauts, 1982).
Soil characterization
The following horizons were sampled in the four soil pro®les
of the sequence: the Ah and B horizons of the acid brown
earth, the Ah, AB and Bw1 horizons of the ochreous
brown earth, the Ah, Bh, Bw1 and Bw2 horizons of the
brown podzolic soil, and the Ah, E, Bhs, Bs and C horizons of
the podzol. The samples from mineral horizons were air-dried
and sieved to < 2 mm. The samples from organic-rich Ah
horizons were sieved to < 2 mm without drying and kept in
closed bags at 4°C to avoid their becoming hydrophobic. The
R
137
Cs+ ®xation in an acid brown earth±podzol weathering sequence 119
methanol. The sample was heated at 110°C overnight, cooled
in a desiccator and then washed four times with 10 ml of 0.5 M
NH4Cl. Supernatants were discarded. The sample was then
digested in HF±HNO3±HClO4 (Lim & Jackson, 1982). The Rb
content was determined by atomic absorption spectrophotometry. The vermiculite content in g kg±1 was evaluated by Rb
content (cmolc kg±1)/0.154: the evaluation is based on a CEC
of the vermiculite of 154 cmolc kg±1 (Ross et al., 1989).
particle size analysis was done after dispersion using an
ultrasonic probe (50 W, 15 min) and Na+ resins as a dispersing
agent (Rouiller et al., 1972). Total carbon (Walkley-Black),
exchangeable bases and cation exchange capacity (CEC) (1 M
NH4OAc at pH 7), and KCl-extractable Al and H were
determined using the procedures outlined by Page et al.
(1982, pp. 570, 160, 163, respectively); pH was measured in
both H2O and 1 M KCl (10 g:25 ml). Elemental analyses were
made in HF±H2SO4 digests of calcined < 2 mm soil samples
(Page et al., 1982, p. 7) by atomic absorption spectrophotometry.
Results and discussion
Soil characteristics
Radiocaesium sorption±desorption experiments
Table 1 presents major characteristics of each horizon of the
four soils. All the horizons are acidic and depleted of
exchangeable bases, with base saturation less than 5% except
in the case of the Ah horizon in the acid brown earth (16%).
Aluminium extractable in KCl represents 30±60% of the
effective cation exchange capacity (ECEC) in the organic-rich
horizons and 65±94% in the mineral horizons.
The depth distribution of organic carbon reveals a regular
decrease in the pro®le of the acid brown soil (59), the ochreous
brown soil (57) and the brown podzolic soil (29). In the last
pro®le (79), the organic carbon decreases sharply in the E
horizon and increases in the Bhs, which is typical of a welldeveloped podzol.
Sand is the dominant (73±96%) particle size fraction in all
the horizons. This re¯ects the siliceous character of the soil
parent material. The clay content is very small (0.7±9.2%) and
decreases from the acid brown soil to the podzol.
The total reserve in bases (TRB) sums to the total
content of basic cations and is an estimator of the content
of weatherable minerals (Herbillon, 1988). The total
reserves are invariably small in all the soil horizons
(< 23 cmolc kg±1). They markedly decrease, however, from
the acid brown earth to the podzol. Potassium is the
dominant cation in the TRB (37±53%) because of the
presence of K-bearing primary minerals such feldspars and
micas (Herbauts, 1982). The increase in the proportion of
K in the TRB from the acid brown soil (37±40%) to the
podzol (48±53%) illustrates the greater resistance of the Kbearing minerals to acid weathering. The distribution of
both clay content and TRB suggests increased weathering
of primary silicate minerals from the acid brown earth to
the podzol.
Soil samples (air-dried in the case of mineral horizons, sieved
to pass 2 mm) were equilibrated with a mixed KCl±CaCl2
solution with a total chloride concentration of 10±3 M, a
potassium adsorption ratio (PAR) value of 0.057 mM1/2 (PAR
is de®ned as (K)/Ö(Ca) where () refers to ion concentration in
(mM) and a K:Ca molar ratio of 0.082. Dialysis bags
containing 1 g of soil and 5 ml of the mixed KCl±CaCl2
solution were placed in 95 ml of 137CsCl-labelled KCl±CaCl2
solution (carrier free 137Cs, 137Cs+ concentration 1 3 10±10 M
representing a g activity of 44 kBq dm±3). The extent of the
137
Cs+ sorption, which continued for 7 days, was measured by
g counting of the solution phase using an NaI scintillator g
counter (Auto-gamma 5000 series, Canberra) for 5 min. At the
end of the sorption the relative quantity of sorbed radiocaesium, the distribution coef®cient Kd, characterizes the
sorption properties of the soil. The distribution coef®cient is
de®ned by the following equation: Kd = XCs/[Cs], where XCs is
the 137Cs g activity on the adsorbed phase in Bq kg±1 and [Cs]
is the 137Cs g activity in the liquid phase in Bq dm±3.
Desorption was done using a method adapted from Wauters
et al. (1994) in which 100 ml of the mixed KCl±CaCl2 solution
with 5.5 g of K+±Ca2+-saturated resin and 1 g of soil was
shaken end-over-end. Both sorption and desorption were
carried out in an identical K+±Ca2+ electrolyte background
because both the soil materials and the desorbing resin had
previously been equilibrated with the same mixed KCl±CaCl2
solution. The desorption was monitored for 25 days: both the
solution and the resin were replaced after 1, 2, 4, 7, 11 and
18 days, and counted for their g activity. The 137Cs+ net
retention was evaluated by the relative quantity of radiocaesium retained at the end of the desorption process. This
relative quantity is expressed as a percentage of the initial
137
Cs+ input.
Cs sorption±desorption data
The 137Cs+ sorption±desorption data for each soil horizon
are presented in Figure 2. Except for the Ah and the E
horizon in the podzol, the sorption was fast and large in all
cases, exceeding 95% of the initial 137Cs+ input after
7 days. The expression of the sorption by the Kd gives
more pronounced differences between the various horizons,
Quanti®cation of the vermiculitic sites
A rubidium ®xation method adapted from Ross et al. (1989)
was used to quantify the vermiculitic sites of the soil materials.
In a 15-ml centrifuge tube, 1 g of soil (< 2 mm) was washed
three times with 10 ml of 0.5 M RbCl and once with 80%
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
0±5
5±15
15±30
0±4
4±12
12±24
24±40
59 Ah
59 B
57 Ah
57 AB
57 Bw1
Ah
Bh
Bw1
Bw2
Ah
E
Bhs
Bs
C
29
29
29
29
79
79
79
79
79
a
0±10
10±20
Horizon
4.9
5.1
4.3
4.7
4.9
4.8
4.3
4.9
4.9
4.0
4.3
4.9
4.2
5.0
H2O
pH
3.3
3.9
3.6
4.2
4.7
3.4
3.3
4.3
4.6
3.3
3.5
4.6
3.7
4.3
KCl
8.3
1.4
5.7
4.6
1.6
61.4
21.3
11.9
4.7
33.7
15.0
7.6
25.7
4.9
Organic C
/g kg±1
Total exchangeable bases. b Total reserve in bases.
0±10
10±30
30±45
45±60
60±80
Depth
/cm
0.29
0.04
0.12
0.09
0.02
0.95
0.10
0.08
0.02
0.29
0.13
0.02
2.35
0.24
7.84
1.18
5.00
3.90
1.58
23.70
8.39
5.65
3.20
11.10
5.27
3.54
14.87
4.78
Exchange complex
TEBa
CEC
________
/cmolc kg±1 ________
0.59
0.24
2.17
1.66
0.38
1.43
1.99
1.35
0.73
1.48
1.89
0.87
1.33
1.29
0.35
0.09
0.57
0.34
0.08
0.55
0.98
0.59
0.03
0.63
0.55
0.03
0.64
0.16
KCl-extractable
Al
H
________
/cmolc kg±1 ________
6.50
4.86
8.87
9.55
10.44
12.19
11.73
12.97
11.36
22.25
17.54
22.33
19.25
20.14
TRBb
/cmolc kg±1
3.18
2.55
4.25
5.09
5.52
4.88
4.46
4.88
4.88
11.04
6.79
8.28
7.22
8.07
Total K
/cmolc kg±1
95.7
93.4
92.2
93.4
92.7
88.1
84.9
83.7
83.5
76.9
76.0
73.1
82.0
80.7
3.1
5.9
4.3
4.1
2.6
7.9
10.4
11.2
11.4
17.2
17.6
20.9
9.5
10.1
1.2
0.7
3.5
2.5
4.7
4.0
4.7
5.1
5.1
5.9
6.4
6.0
8.5
9.2
Particle size analysis
Sand
Silt
Clay
_____________________
/% ____________________
Table 1 Major properties of the four soils forming an acid brown earth±podzol weathering sequence developed in sandy material (59, acid brown earth; 57, ochreous brown earth; 29,
brown podzolic soil; 79, podzol)
L
120 E. Maes et al.
R
137
Cs+ ®xation in an acid brown earth±podzol weathering sequence 121
Figure 2 137Cs+ sorption±desorption isotherms of each horizon (size fraction
< 2 mm) of the four soils forming the acid
brown earth±podzol weathering sequence. S,
sorption; D, desorption by K+±Ca2+-saturated resin.
with Kd ranging between 252 and 26 216 dm3 kg±1. These
values, when plotted on a log scale (Figure 3), are
correlated with the 137Cs+ net retention value, which is
the relative quantity of radiocaesium retained at the end of
the desorption process (r = 0.70).
The 137Cs+ net retention values are plotted in relation to
soil horizons in Figure 4. These values range from 14.3 to
96.9% of the initial 137Cs+ input. The net 137Cs+ retention
value in the Ah horizons decreased gradually from the acid
brown soil (80%) to the ochreous brown earth and the
brown podzolic soil (73% and 60%, respectively) and then
strongly to the podzol (14%). In each pro®le of the
sequence, the maximum net 137Cs+ retention value is
observed at successively greater depth amounting to 80,
97, 91 and 93% in, respectively, the Ah horizon of the acid
brown soil, the AB horizon of the ochreous brown soil, the
Bh horizon of the brown podzolic soil and the Bhs horizon
of the podzol.
Cs ®xation and soil characteristics
The decrease of the Cs+ net retention in the Ah horizons
from the brown soil to the podzol could be associated with
a decrease in clay content (Table 1). This relationship
could not be con®rmed, however, when all the horizons
were taken into account (r = 0.34). Similarly, the net 137Cs+
retention value seemed unrelated to the organic carbon
content (r = 0.25). Soil organic matter plays no part in
speci®c Cs+ retention (Cremers et al., 1988). Even in highly
organic soils, the mineral fraction, however small it may
be, is responsible for the Cs+ ®xation in soils (Shand et al.,
1994; Hird et al., 1995).
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
L
122 E. Maes et al.
Figure 3 Relation between log Kd and the 137Cs+ net retention for
all the horizons of the acid brown earth±podzol weathering
sequence (r = 0.70).
Comparison of the depth distribution of the Cs+ net retention
value (Figure 4) with the particle size data (Table 1) also
illustrates that clay content alone provides insuf®cient
information for predicting the adsorption by the soil, as
previously reported by Livens & Loveland (1988). On the
other hand, a comparison of Figures 1 and 4 suggests that
the variation of the Cs+ net retention might be associated with
the nature of the clay minerals present in each soil horizon and
therefore with the weathering stage characterizing each soil in
the sequence.
Cs ®xation and soil clay mineralogy
Fixation of Cs by clay minerals has been widely studied. This
®xation occurs on a few highly selective sites present at the
edge or in the interlayer of micaceous minerals of illitic type
and accounting for about 1±2% of their cation exchange
capacity (Sawhney, 1972; Francis & Brinkley, 1976; Cremers
et al., 1988). In this study we found no signi®cant correlation
between total K content and Cs+ ®xation (r = 0.28). This poor
relation could be because (i) total K does not strictly re¯ect
illite content (mica and K feldspar are present) and (ii) Cs+
®xation is not solely associated with illitic sites.
Various clay minerals were identi®ed in the soil clay
fraction of the sequence: kaolinite, illite, hydroxy interlayered
vermiculite (HIV), vermiculite and degradation smectite. Their
occurrence varies with soil type and depth (Figure 1). Beneath
the desaluminization front, which deepens from the acid brown
earth to the podzol, HIV minerals are present. Just above this
line in each pro®le vermiculite is present. In the ochreous
brown earth and the podzolic soils of the sequence vermiculite
progressively gives way to smectite towards the soil surface.
When the depth distributions of the net 137Cs+ retention
value (Figure 4) are compared with the mineralogical
evolution in each pro®le (Figure 1) it is evident that maximum
137
Cs+ net retention coincides with the depth of the
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
Figure 4 137Cs+ net retention on the fraction < 2 mm of each
horizon of the four soils.
desaluminization front and in the horizon in which the clay
fraction contains vermiculite without Al interlayering and is
devoid of smectitic clay minerals. Weaker Cs+ retention is
associated with the presence of HIV minerals in deeper
horizons and with degradation smectites in the surface
horizons.
Maes et al. (1998) have already observed weak Cs+ retention
by soil dominated by HIV in an acid brown forest soil derived
from siliceous schist. They attributed this poor retention to the
blocking effect of the hydroxy-Al interlayers which impede the
collapse of vermiculitic layers and therefore Cs+ ®xation in the
interlayer position.
The degradation smectites identi®ed in the Ah horizons
of the podzolic soils likely derive from the desaluminization of HIV (Herbauts, 1982) as reported by others
(Malcolm et al., 1969; Carnicelli et al., 1997). During
formation from the parent mineral HIV, these degradation
smectites have lost layer charge mainly by loss of Al from
the tetrahedral sheet and oxidation of octahedral Fe2+
R
137
Cs+ ®xation in an acid brown earth±podzol weathering sequence 123
Table 2 Fixed Rb, computed vermiculite
contents in the soil (< 2 mm) and [137Cs+
retained/Rb+ ®xed] ratio (59, acid brown
earth; 57, ochreous brown earth; 29, brown
podzolic soil; 79, podzol)
137
Cs+/Rb+
(3 10±7)
Horizon
Depth
/cm
Rb ®xed
/cmolc kg±1
Vermiculite content
/g kg±1
59 Ah
59 B
0±10
10±20
1.04
0.42
6.7
2.7
7.8
13.9
57 Ah
57 AB
57 Bw1
0±5
5±15
15±30
0.67
1.18
0.30
4.4
7.6
1.9
11.0
8.3
11.7
29
29
29
29
Ah
Bh
Bw1
Bw2
0±4
4±12
12±24
24±40
0.58
0.87
0.37
0.33
3.8
5.6
2.4
2.2
10.5
10.7
8.2
9.5
79
79
79
79
79
Ah
E
Bhs
Bs
C
0±10
10±30
30±45
45±60
60±80
0.09
0.07
1.02
0.61
0.08
0.6
0.5
6.6
3.9
0.5
15.9
47.4
9.1
11.2
31.1
vermiculite in the clay fraction ranges between 3 and 19%,
assuming that vermiculite content of coarser fractions is
negligible. The smallest vermiculite/clay ratio occurs in the
horizons where HIV or degradation smectite are dominant.
Figure 5 illustrates the relation between 137Cs+ net retention
and vermiculite content. This strong linear (r = 0.95) relation
suggests that vermiculitic sites dominate radiocaesium retention. In acid soils vermiculites have a dioctahedral structure
(Douglas, 1989), which probably enhances Cs+ ®xation as
reported by Maes et al. (1999). The relation between Cs+
®xation and vermiculite content also con®rms that weaker Cs+
retention is associated with either degradation smectite or HIV
minerals. Moreover, it highlights the fact that the ®xation of
Cs+ traces in soils strongly depends on the nature of the clay
mineral rather than on clay content. These results suggest that
the speci®c sites for Cs+ ®xation in soils are not illitic sensu
stricto: they are vermiculitic sites, most probably associated
with micaceous interlayers.
Table 2 also indicates that the values of the ®xed Cs+ 3 10±7
/Rb+ ratio range between 7.7 and 47.4. Most of the values for
this ratio fall within a rather narrow range (7.7±13.8), which
suggests that the number of highly speci®c Cs+ sorbing sites,
though small, represents a fairly constant proportion of the
vermiculitic sites.
(Robert, 1973; Borchardt, 1989). Layer charge is a major
determinant of Cs+ ®xation by 2:1 minerals (Cornell, 1993),
and so the weaker Cs+ ®xation capacity in horizons
dominated by degradation smectites could arise from the
smaller layer charge. However, degradation smectites are
reported to have substantial capacity to ®x K (Robert,
1973; Badraoui et al., 1987; Borchardt, 1989), particularly
after wetting and drying (Badraoui et al., 1987). This
capacity has been related to collapse of layers on heating
after saturation with K, which has been observed in the Ah
horizons of podzolic soils (Herbauts, 1982). The weaker
Cs+ ®xation in the Ah horizons dominated by degradation
smectites could therefore be related to the competition
between K+ and Cs+ for speci®c sorbing sites or to
turbostratic disorder of the 2:1 layers (Mamy & Gaultier,
1976; Maes et al., 1985).
Quanti®cation of vermiculitic layers
Vermiculite content was quanti®ed using the rubidium
methodology adapted from Ross et al. (1989). Table 2
presents both ®xed Rb+ and computed vermiculite contents
as well as the values of the ratio ®xed Cs+/®xed Rb+. The
Rb and vermiculite contents range from 0.07 to 1.18 cmolc
kg±1 and 0.05 to 0.76%, respectively. For each of the four
soils, the largest vermiculite content is invariably observed
in the horizon below that in which degradation smectite
was identi®ed and above that in which HIV is dominant.
This observation is supported by X-ray data which indicates
that vermiculite is the dominant 2:1 clay mineral in this
type of horizon (Figure 1). Comparing vermiculite and clay
contents (Tables 1 and 2) reveals that the proportion of
The weathering model
The maximum value of the 137Cs+ net retention deepens with
increasing weathering stage. For each pro®le, this value is
associated with the greatest vermiculite content, located at the
desaluminization front (Figure 1). In the soil sequence, the
®xation of radiocaesium traces therefore provides a valuable
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
L
124 E. Maes et al.
Figure 5 Relation between 137Cs+ net retention and the vermiculite
content estimated by the Rb+ ®xation method for all the horizons of
the soil sequence (r = 0.95).
index of the weathering processes affecting the 2:1 clay
minerals. Both Al interlayering and the diminishing of layer
charge decrease Cs+ ®xation in such minerals. The results
presented here and in the paper by Maes et al. (1999) largely
converge in three respects. First, ®xation of 137Cs+ strongly
depends on the nature of clay minerals in a constant electrolyte
background. Second, the ®xation occurs on vermiculitic sites
most probably associated with micaceous interlayers. Third,
dioctahedral vermiculites are likely to act as ef®cient sinks for
radiocaesium in acid forest soils.
Conclusions
The soil weathering sequence acid brown earth ® ochreous
brown earth ® brown podzolic soil ® podzol developed
from sandy sediments is characterized by a strong
mineralogical heterogeneity arising from the transformation
of vermiculite. The weathering stage strongly affects the
Cs+-®xation properties. The maximum net 137Cs+ retention
value occurs at increasing depth with increasing weathering
stage and is therefore observed in different horizons from
the acid brown soil to the podzol. The magnitude of Cs+
®xation is positively and strongly related to the soil
vermiculite content estimated by the rubidium method.
Fixation of Cs+ decreases in soil horizons with clay
fractions dominated by hydroxy interlayered vermiculite or
smectite. In a constant electrolyte background, Cs+ ®xation
appears as a good tracer of the transformation processes
affecting vermiculite in the soil sequence. These results
suggest that Cs+ is ®xed on vermiculitic sites most
probably associated with micaceous interlayers.
References
Badraoui, M., Bloom, P.R. & Rust, R.H. 1987. Occurrence of highcharge beidellite in a Vertic Haplaquoll of northwestern Minnesota.
Soil Science Society of America Journal, 51, 813±818.
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125
Barnhisel, R.I. & Bertsch, P.M. 1989. Chlorites and hydroxyinterlayered vermiculite and smectite. In: Minerals in Soil
Environments, 2nd edn (eds J.B. Dixon & S.B. Weed), pp. 729±
788. Book Series No 1, Soil Science Society of America, Madison,
WI.
Borchardt, G. 1989. Smectites. In: Minerals in Soil Environments, 2nd
edn (eds J.B. Dixon & S.B. Weed), pp. 675±727. Book Series No 1,
Soil Science Society of America, Madison, WI.
Carnicelli, S., Mirabella, A., Cecchini, G. & Sanesi, G. 1997.
Weathering of chlorite to a low-charge expandable mineral in a
Spodosol on the Apennine mountains, Italy. Clays and Clay
Minerals, 45, 28±41.
Cornell, R.M. 1993. Adsorption of cesium on minerals: a review.
Journal of Radioanalytical and Nuclear Chemistry, 171, 483±500.
Courchesne, F. & Gobran, G.R. 1997. Mineralogical variations of
bulk and rhizosphere soils from a Norway Spruce stand. Soil
Science Society of America Journal, 61, 1245±1249.
Cremers, A., Elsen, A., De Preter, P. & Maes, A. 1988. Quantitative
analysis of radiocaesium retention in soils. Nature, London, 335,
247±249.
Douglas, L.A. 1989. Vermiculites. In: Minerals in Soil Environments,
2nd edn (eds J.B. Dixon & S.B. Weed), pp. 635±674. Book Series
No 1, Soil Science Society of America, Madison, WI.
Eberl, D.D. 1980. Alkali cation selectivity and ®xation by clay
minerals. Clays and Clay Minerals, 28, 161±172.
Fanning, D.S., Keramidas, V.Z. & El-Desoky, M.A. 1989. Micas. In:
Minerals in Soil Environments, 2nd edn (eds J.B. Dixon & S.B.
Weed), pp. 551±634. Book Series No 1, Soil Science Society of
America, Madison, WI.
FAO 1998. World Reference Base for Soil Resources. World Soil
Resources Reports No 84, FAO, Rome.
Francis, C.W. & Brinkley, F.S. 1976. Preferential adsorption of 137Cs
to micaceous minerals in contaminated freshwater sediment.
Nature, London, 260, 511±513.
Frink, C.R. 1969. Chemical and mineralogical characteristics of
eutrophic lake sediments. Soil Science Society of America
Proceedings, 33, 369±372.
Herbauts, J. 1982. Chemical and mineralogical properties of sandy
and loamy-sandy ochreous brown earths in relation to incipient
podzolization in a brown earth±podzol evolutive sequence. Journal
of Soil Science, 33, 743±762.
Herbillon, A.J. 1988. Chemical estimation of weatherable minerals
present in the diagnostic horizons of low activity clay soils. In:
Proceedings of the 8th International Soil Classi®cation Workshop:
Classi®cation, Characterization and Utilization of Oxisols (eds
F.H. Beinroth, M.N. Camargo & H. Eswaran), pp. 39±48.
EMBRAPA, Rio de Janeiro.
Hird, A.B., Rimmer, D.L. & Livens, F.R. 1995. Total caesium-®xing
potentials of acid organic soils. Journal of Environmental
Radioactivity, 26, 103±118.
Lim, C.H. & Jackson, M.L. 1982. Dissolution for total elemental
analysis. In: Methods of Soil Analysis: Part 2, Chemical and
Microbiological Properties, 2nd edn (eds A.L. Page, R.H. Miller &
D.R. Keeney), pp. 1±12. American Society of Agronomy and Soil
Science Society of America, Madison, WI.
Livens, F.R. & Loveland, P.J. 1988. The in¯uence of soil properties
on the environmental mobility of caesium in Cumbria. Soil Use and
Management, 4, 69±75.
R
137
Cs+ ®xation in an acid brown earth±podzol weathering sequence 125
Maes, A., Verheyden, D. & Cremers, A. 1985. Formation of highly
selective cesium-exchange sites in montmorillonites. Clays and
Clay Minerals, 33, 251±257.
Maes, E., Delvaux, B. & Thiry, Y. 1998. Fixation of radiocaesium
in an acid brown soil. European Journal of Soil Science, 49,
133±140.
Maes, E., Vielvoye, L., Stone, W. & Delvaux, B. 1999. Fixation of
radiocaesium traces in a weathering sequence mica ® vermiculite
® hydroxy interlayered vermiculite. European Journal of Soil
Science, 50, 107±115.
Malcolm, R.L., Nettleton, W.D. & McCracken, R.J. 1969. Pedogenic
formation of montmorillonite from a 2:1±2:2 intergrade clay
mineral. Clays and Clay Minerals, 16, 405±414.
Mamy, J. & Gaultier, J.P. 1976. Les pheÂnomeÁnes de diffraction des
rayonnements X et eÂlectroniques par des reÂseaux atomiques:
application aÁ l'eÂtude de l'ordre cristallin dans les mineÂraux
argileux. Annales Agronomiques, 27, 1±16.
Page, A.L., Miller, R.H. & Keeney, D.R. 1982. Methods of Soil
Analysis: Part 2, Chemical and Microbiological Properties, 2nd
edn. American Society of Agronomy and Soil Science Society of
America, Madison, WI.
Robert, M. 1973. The experimental transformation of mica toward
smectite. Relative importance of total charge and tetrahedral
substitution. Clays and Clay Minerals, 21, 167±174.
Ross, G.J., Schuppli, P.A. & Wang, C. 1989. Quantitative determination of vermiculite by a rubidium ®xation method. Soil Science
Society of America Journal, 53, 1588±1589.
Rouiller, J., Burtin, G. & Souchier, B. 1972. La dispersion des
sols dans l'analyse granulomeÂtrique. MeÂthode utilisant les
reÂsines eÂchangeuses d'ions. Bulletin de l'Ecole Nationale
SupeÂrieure d'Agronomie et des Industries Alimentaires, Nancy,
14, 193±205.
Sawhney, B.L. 1972. Selective sorption and ®xation of cations by clay
minerals: a review. Clays and Clay Minerals, 20, 93±100.
Shand, C.A., Cheshire, M.V., Smith, S., Vidal, M. & Rauret, G. 1994.
Distribution of radiocaesium in organic soils. Journal of Environmental Radioactivity, 23, 285±302.
Vicente, M.A., Razzaghe, M. & Robert, M. 1977. Formation of
aluminium hydroxy vermiculite (intergrade) and smectite from
mica under acidic conditions. Clay Minerals, 12, 101±112.
Wauters, J., Sweeck, L., Valcke, E., Elsen, A. & Cremers, A. 1994.
Availability of radiocaesium in soils: a new methodology. Science
of the Total Environment, 157, 239±248.
#
1999 Blackwell Science Ltd, European Journal of Soil Science, 50, 117±125