Science of the Total Environment 363 (2006) 1 – 10

www.elsevier.com/locate/scitotenv

The role of clinoptilolite in organo-zeolitic-soil systems used for

phytoremediation
Peter J. Leggo a,*, Béatrice Ledésert b, Graham Christie c
a
Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3 EQ, UK
b
Université de Cergy-Pontoise, Département des Sciences de la Terre et de l’Environnement, F-95031 Neuville-sur-Oise, France
c
Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge CB2 1QT, UK
Received 1 July 2005; received in revised form 12 September 2005; accepted 12 September 2005
Available online 17 October 2005

Abstract

The present work is an extension of earlier research [Leggo, P. (2000) An investigation of plant growth in an organo-zeolitic
substrate and its ecological significance. Plant and Soil, 219: 135–146.] in which zeolitic tuffaceous rock containing clinoptilolite, a
commonly occurring natural zeolite mineral, was composted with organic waste to produce a very effective bio-fertilizer thought to
be activated by a very large increase in the population of nitrifiers. This material has now been modified by the addition of extra
untreated zeolitic tuff (i.e. un-ammoniated clinoptilolite). Comparing plant growth in a series of substrates containing increasing
amounts of zeolitic tuff the limit of growth enhancement has been established. Aqueous leachate analysis has demonstrated a
correlation between shoot growth and the mobilization of cations in the soil pore water. Measurement of soil water suction pressure
has shown that soil moisture is directly related to the amount of the zeolitic tuff amendment. It has also been found that the zeolite
component of the soil system supports biofilm formation and this behaviour is thought to account for the additional plant growth in
substrates containing extra untreated tuff. Similar trends are found for plants growing in clean and metal polluted soils. It is now
clear that organo-zeolitic-soil systems offer an opportunity to re-vegetate land made barren by metal pollution and as a
consequence, erosion and dissemination of contaminants are reduced. By harvesting metal tolerant plants it would appear feasible
to recover metals and clean the rhizosphere simultaneously.
D 2005 Elsevier B.V. All rights reserved.

Keywords: Plant growth; Zeolite; Bio-fertilizer; Nitrifiers; Biofilm; Metal contamination; Recovery

1. Introduction ing practices are devoid of the major plant nutrients

(Bloomfield et al., 1982) and soil contaminated by
The disposal of mine and metallurgical waste with- them commonly remain barren or colonized only by
out treatment presents a very difficult problem, which metal tolerant plants (Brooks et al., 1998). Such barren
if ignored leads to a legacy of pollution that covers an sites have no defence against wind erosion, leaching
infinite time scale. The wastes from mining and smelt- by rain and surface run-off. As a consequence toxic
metal elements are free to migrate and invariably
terminate in the food chain. It has long been appreci-
* Corresponding author. Tel.: +44 1223 333 462; fax: +44 1223
ated that phytoremediation provides an answer to this
333 450. problem, but it has always been difficult to sustain
E-mail address: leggo@esc.cam.ac.uk (P.J. Leggo). plant growth on such inhospitable soils (Walker et
0048-9697/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.scitotenv.2005.09.055
2 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10

al., 1996). To help overcome this problem organo- increase in growth rate, related to an apparent large
zeolitic bio-fertilizers, in contrast to chemical fertili- increase in the nitrifying bacterial population. It has
zers, can be used to a great advantage. It has been also been shown that a similar effect occurs when
shown that the unique mineral properties of clinopti- soils, highly polluted with heavy metals, are treated in
lolite, a zeolite occurring naturally as a component of the same way (Leggo and Ledésert, 2001).
altered volcanic rock (zeolitic tuff), when composted Studies of the development of bacterial populations
together with poultry manure becomes ammoniated. in organo-zeolitic amended soils (Andronikashvili et
On introduction into a moist soil the ammonium ions al., 1999) have shown that the quantity of micro-organ-
are back exchanged and oxidized by nitrifying bacte- isms increases by 2.5 times and changes occur in their
ria, significantly enhancing the soil microbial popula- qualitative characteristics. It was also discovered that
tion (Leggo, 2000). The results given in this paper the microbial activity of the amended soil was not
show that additional plant growth occurs on increasing slowed down during winter as much as soil treated
the concentration of un-ammoniated zeolitic tuff in the with chemical fertilizers or left fallow. The amended
substrate. As untreated tuff contains no ammoniated soil was thus found to have more vitality which benefit-
zeolite the additional increase in plant growth must be ed growth the following season. By measuring soil
due to another function rather than a further increase in respiration, to index microbiological activity, it was
nitrifying bacteria due to oxidation of back exchanged found that the incorporation of crushed zeolitic tuff
ammonium ions. To address this question the rate of increased the rate of decomposition of soil organic
loss of zeolite water, during a drying period, has been material, indicating high microbial activity. The micro-
measured as it is known that nitrifying bacteria are bial dynamics revealed rises and falls in the number of
sensitive to soil moisture content (Paul and Clark, micro-organisms and their biomass. The frequency of
1996). Electron microscopy of zeolite crystal surfaces this pulsation was higher than in untreated soil indicat-
was employed to record differences that may exist ing that a clinoptilolite–organic mixture favours inten-
between zeolite from the amended substrates and that sive development of micro-organisms. It was also
from the untreated rock. The hypothesis is proposed observed that mycolitic bacteria, causing lysis in fila-
that an external zeolite crystal surface, in a moist soil, mentous fungi, prevail and the quantity of the fungal
would be covered with water molecules and in such a species decreases by half. An increase in actinomyces
condition would lend itself to the formation of a occurs which favours the sterilisation of undesirable
biofilm that might induce extra nitrifying activity. As micro-flora and Azotobacters, which fix atmospheric
one would expect such activity to increase the ionic nitrogen, also increases by some 30–40%. Amoeboid
mobility of the substrate pore water, electrical conduc- aggregates, 50–60 mm in size, were also found in the
tivity measurements of aqueous leachates were made amended soil. This latter fact indicates that changes in
to reveal any such changes. soil porosity must occur to accommodate such large
microbial bodies. That such porosity changes do occur
2. Background studies was also inferred from the results of leaching experi-
ments of organo-zeolitic amended soils. The leachates,
In recent years phytotechnology has attracted much low in colloidal particles and characterised by high
attention as a means of remediation of contaminated electrical conductivity, indicated that the flocculation
land. Several comprehensive reviews are available (Salt of colloidal clay particles and a loosening of the soil
et al., 1998; Long et al., 2002; McInyre, 2003), and in structure had occurred Leggo, 2000). This effect would
this paper the topic is further addressed by investigating act to increase aeration, favouring the development of
the plant growth enhancement properties of a specific aerobic bacteria.
organo-zeolitic-soil system. Many studies have been It is known that zeolitic tuff can be used to exchange
made (Ming and Allen, 2001) that focus on the ion- NH4+ from municipal and agricultural waste streams
exchange properties of zeolite minerals and how these (Ames, 1967) and was first used in Japan for agronomic
can be used to support plant growth in aqueous culture, and horticultural purposes (Minato, 1968; Mumpton,
but relatively little has been published on plant growth 1999). Composting experiments have shown that by
behaviour in organo-zeolitic-soil substrates of the type incorporating clinoptilolitic tuff with animal waste, am-
described in this paper. monia can be retained by the zeolite (Witter and Lopez-
In a former study (Leggo, 2000) Spring Wheat Real, 1988). It is, however, only in recent years that the
(Triticum aestivum L., cv. Paragon), grown in the relationship between ammoniated zeolite and the bac-
organo-zeolitic-soil substrate, showed a pronounced terial population of organo-zeolitic amended soils has
 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10 3

become recognised (Leggo, 2000; McGilloway et al., Ltd, Bracknell, England has shown the lead, copper
2003). In the latter paper it was found that high nitrite and zinc concentrations to be 49.3, 18.5 and 142 mg
and nitrate levels in zeoponic substrates (artificially ion- kg 1, respectively.
exchanged zeolitic tuff in sterile aqueous solution) The reference soil came from the University of
could only be produced by inoculation with nitrifying Cambridge Botanic Garden and is described in Leggo
bacteria, which is a strong argument for the importance (2000) and details of the contaminated soil which came
of bacterial activity in organo-zeolitic-soil systems. from the Metaleurope Nord site at Evin-Malmaison,
The classical approach to the prevention of erosion near Lille, are described in Leggo and Ledésert (2001).
and transportation of toxic mine waste, by the agents of
wind and rain, has been to stabilize the surface by re- 3.2. Analytical methods
vegetation (Bradshaw and Chadwick, 1980). This has
met with some success, but in cases in which soils A control experiment was conducted with Spring
contain more than 0.10 wt.% zinc and comparable Wheat grown in uncontaminated Botanic Garden soil.
levels of other metals such as cadmium, lead and Spring Wheat was chosen as individual plants can be
copper, it can be difficult to restore permanent vegeta- replicated easily facilitating statistical analysis of plant
tion. This is particularly true for light textured soils with growth. The Spring Wheat (T. aestivum L., cv. Red
low clay content made acidic by the oxidation of metal Fife) was potted in 2 kg substrates, replicated four
sulphide waste (Li and Chaney, 1998). Studies made on times, with initially five wheat caryopses per pot.
the use of sewage sludge, which commonly contains After germination the seedlings were thinned to two
appreciable quantities of metal elements, have clearly plants per pot and grown in a greenhouse. Over the
revealed that phytotoxic effects can be considerable winter months artificial lighting (high pressure Sodium
(Alloway and Jackson, 1991), although much work SON/T rated at 400 W) was used to simulate a 16 h day.
still needs to be done to establish the toxicity levels Plant growth in the control substrate, C.Gp 1, was
of individual metals. Again, the stress placed on soil compared to that in a series of amended substrates.
micro-organisms, at trace metal concentrations, can Substrate C.Gp 2 contained 16.7 vol.% bio-fertilizer
lead to an inadequate supply of plant nutrients in which the ammoniated zeolitic tuff content is ap-
(McGrath, 2003) to the detriment of plant growth. proximately 3.45 wt.% (69 g/2 kg substrate). Other

3. Materials and analytical methods Table 1

Plant dry weights
3.1. Materials Sample Shoot (g) F S.E. (n = 4)
C.Gp 1 Unamended substrate 1.42 F 0.37
The zeolitic tuff used came from the Bulgarian C.Gp 2 Substrate + bio-fertilizer 7.67 F 2.27
deposit at Beli-Plast, S.E. Bulgaria. This rock contains (34.5 g excess zeolitic tuff kg 1)
80–90 (vol.%) clinoptilolite with calcium and potassi- C.Gp 3 Ditto + 157 wt.% 11.41 F 2.25
um exchangeable cations in similar concentrations, with (88.3 g excess zeolitic tuff kg 1)
sodium and magnesium having much lower values. An C.Gp 4 Ditto + 297 wt.% 11.32 F 1.63
(136.8 g excess zeolitic tuff kg 1)
average formula for the unit cell of the Beli-Plast C.Gp 5 Ditto + 423 Wt% 10.83 F 3.06
clinoptilolite, based on 72 oxygen atoms, is: (180.4 g excess zeolitic tuff kg 1)
C.Gp 6 Ditto + 536 Wt% 12.50 F 0.28
(K1.32 Ca2.03 Na0.4 Mg0.32 Fe0.05) [Al6.6 Si29.4 O72] (219.5 g excess zeolitic tuff kg 1)
.n H2O R.H.Gp 1 Unamended substrate 6.76
R.H.Gp 2 Substrate + bio-fertilizer 25.83
(34.5 g zeolitic tuff kg 1)
where n is in the order of 22 (Burlet, 1997; Passaglia R.H.Gp 3 Ditto + 157 wt.% 26.49
and Sheppard, 2001). (excess zeolitic tuff kg 1 as above)
The bio-fertilizer is made by mixing zeolitic tuff, R.H.Gp 4 Ditto + 297 wt.% 25.46
grain size 0.5–2.0 mm, with poultry manure, by vol- (excess zeolitic tuff kg 1 as above)
R.H.Gp 5 Ditto + 423 wt.% 25.10
ume, in a ratio of one part crushed rock to two parts (excess zeolitic tuff kg 1 as above)
manure. Details of its preparation are described in R.H.Gp 6 Ditto + 536 wt.% 24.51
European Patent No. EP 1208922 B1 (Leggo, 2004). (excess zeolitic tuff kg 1 as above)
Analysis of the heavy metal content of a recent batch of C.Gp refers to the Spring Wheat control set, R.H.Gp refers to the high
the bio-fertilizer by Natural Resources Management density ryegrass set.
4 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10

substrates, Gp 3, 4, 5, and 6 contained the same initial Measurements of the variation of soil water suction
proportion of bio-fertilizer to which various excess pressure, as an indication of soil moisture content (Hil-
amounts of untreated zeolitic tuff were added which, lel, 1982), were made over various time periods and the
on re-potting to 2 kg, represented additions of 157 wt.% drying rates calculated. The measurements were made,
(177 g), 297 wt.% (274 g), 423 wt.% (361 g), and 536 under greenhouse conditions, using a jet filled tensiom-
wt.% (439 g), respectively. The plants were harvested eter model 2725, manufactured by the Soilmoisture
after sixteen weeks; shoots were cut at the base of the Equipment Corp., Santa Barbara, USA. This instrument
roots and dried at 70 8C for 48 h before weighing. is calibrated to measure soil suction pressure in centi-
Ryegrass (Lolium perenne L.) was used to investi- bars (ikPa) over a range 0 (water saturated soil) to
gate perennial plant growth behaviour in the metal 100 (air dry soil). No attempt was made to calibrate
polluted soil as grasses are to be used in future field matrix suction pressure against absolute water content
trials. As with the Spring Wheat control experiment, the as relative differences in drying rate were considered
plants were kept watered to field capacity with de- sufficient to show the effect of zeolitic tuff on the water
ionized water without further addition of any nutrients. holding capacity of the substrates. As only one instru-
The ryegrass was sown in high density (ca. 600 plants/ ment was available measurements were made on each
pot) in 2 kg pots. A low density set (ca. 25 plants/pot) substrate sequentially. This procedure took several
was grown to show morphological differences but not weeks during which time a continuous record of air
used for analytical purposes. In the case of the high temperature was made. Each substrate was, in turn,
density set it was not possible to separate and measure watered to field capacity and allowed to dry at ambient
individual shoot dry mass and the data shown (Table 1) air temperature. The elapsed time of a measured change
represents bulk shoot dry weight only. This approach in soil suction pressure was recorded and the rate of
did not lend itself to replication and statistical analysis loss of apparent soil moisture was calculated. These
was therefore not made. Shoots were harvested by rates were corrected for air temperature variation, which
cutting 3 cm above the substrate, dried to constant affected the evaporation rate, by averaging the maxi-
weight and analysed for zinc, lead and copper; the mum daily air temperature, over the duration of the
roots were discarded. individual drying periods, and normalizing against the
Total nitrogen measurements on the Botanic Garden maximum average daily air temperature.
soil were made on samples air-dried at a temperature A Philips XL 30 FEG scanning electron microscope,
not greater than 30 8C, sieved to pass a 0.5 mm screen with an environmental chamber operating under cryo-
and analyses using the Dumas technique. Procedure genic conditions, was used to image zeolitite grains
used is cited in the AOCA official methods of from both an untreated rock hand specimen and an
analysis (1990), method 949.12. The instrument was amended organo-zeolitic-soil substrate. Zeolitic tuff
calibrated using standard reference materials GBW- grains were recovered from a C.Gp 3 substrate and
7401,7402,7403. Analyses of zinc, lead and copper washed with chilled de-ionized water. The sample was
trace concentrations were made using standard atomic then carefully dried by bblottingQ with a Whatman filter
absorption techniques in the Department of Earth paper before quenching in a solid–liquid nitrogen slush.
Sciences, Cambridge. The instrument was calibrated The quench cooled sample was then placed in the
using solutions diluted from 1000 ppm Merck single environmental chamber, maintained at a temperature
element standard solutions. Dry shoot material was of 157 8C. Magnification of 20,000 times was
ground to a fine powder in a coffee grinder and ashed obtained using an accelerating voltage of 5 kV and
in a muffle furnace overnight at 450 8C and extracted spot size of 2 Am.
with HCl. Blanks were run for all batches and analytical Preliminary microbiological analyses conducted on
precision was better than F5%. the control, amended soils and extracted zeolitic tuff
Leachate samples, collected as in the earlier work grains comprised plate count experiments and 16 s
(Leggo, 2000) and replicated three times, were taken sequence analysis of the predominant colony types.
from the substrates at the end of the growing season. Total viable counts were determined by diluting 1 g
Electrical conductivity was measured with an Electron- (wet weight) aliquots of freshly sampled soil in 10 ml
ic Instruments Ltd., conductivity meter Model MC-1, sterile distilled water. Samples were vortex mixed for 1
mark V calibrated with a 0.1 N solution of KCl at an min, then treated for 15 s in a sonicating water bath,
ambient air temperature of 20 8C and pH recorded with before serially diluting with sterile water and plating 100
a Cranwell CR 95 digital pH/mV meter using an Ama- Al aliquots on dilute nutrient broth (0.13 g/l) (Oxoid)
gru S5 electrode. solidified with 1.5% agar (w/v). Plates were incubated
 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10 5

for 10 days at 25 8C prior to colony counting. All plate the shoot dry weight in substrates containing more than
count experiments were conducted in triplicate. 157 wt.% untreated zeolitic tuff. This decrease corre-
lates with an increase in the shoot zinc concentration
4. Results and discussion and it is thought that the zinc uptake is producing a
phytotoxic effect. This is not surprising as it is well
4.1. Plant growth behaviour known that zinc tolerance is not high in most plants.
For example, the critical toxicity level in the leaves of
Both the Spring Wheat control and ryegrass data sets crop plants is given as 100–300 mg Zn kg 1 dry weight
show very large increases in the shoot dry weight of (Ruano et al., 1988). With regard to this apparent zinc
plants grown in the amended substrates (Table 1). This phytotoxicity the question arises whether it is the result
is thought to be due to a greater degree of mineraliza- of elevated zinc concentrations or the result of substrate
tion of the soil pore water which increases the avail- dilution of available nitrogen. To help clarify this ques-
ability of plant nutrients. In the case of the Spring tion four samples of the Botanic Garden soil were
Wheat control set (Fig. 1), the data shows that the analysed for total nitrogen and compared with thirteen
shoot dry weight, due to the application of the bio- analyses of soil from the contaminated site. The garden
fertilizer, increases by a factor of more than 5 and on soil value (2.38 F 0.05 g kg 1) was found to be within
addition of 157 wt.% (88.3 g kg 1) excess untreated the range of values from the contaminated site (1.32–
zeolitic tuff, the shoot dry weight increases further by a 2.48 g kg 1). Although the contaminated unamended
factor of nearly 1.5. The dramatic initial growth behav- substrates contain on average a little less total nitrogen
iour is thought to be due to the release of ammonium it is thought that the dramatic growth enhancement of
ions from the bio-fertilizer promoting a very large plants growing in the amended substrates demonstrates
population of nitrifying bacteria in the rhizosphere an abundance of available nitrogen. It would, therefore,
(Leggo, 2000). However, the second effect cannot be seem unlikely that the trend in the ryegrass growth can
caused in this way as this material contains no ex- be attributed to a dilution of available nitrogen.
changeable ammonium ions. It was found that above The fact that the increase in zinc uptake correlates
an addition of 157 wt.% untreated zeolitic tuff the with an increase in the zeolitic tuff concentration of the
growth effect reaches a plateau. However, at an excess substrate introduces an interesting question. In the case
greater than 400 wt.% the growth rate, within experi- of clinoptilolite several studies have shown that zinc has
mental error, appears to increase again. a low ion-exchange selectivity (Semmens and Seyfarth,
A similar behaviour in plant growth rate is exhibited 1978; Tsitsishvili et al., 1992; Zamzow et al., 1990) and
by the ryegrass growing in the contaminated substrates therefore, would not be as susceptible to ion-exchange
(Fig. 1). However, an apparent slight decrease occurs in as most other base metals. This fact might lead to the
assumption that Zn would be bioavailable and concen-
trate in plants. However, it has been shown that Zn
30 uptake decreases with increasing zeolite concentration
RH Gp.2 RH Gp.3 RH Gp.4 RH Gp.5
(Castaldi et al., 2005) in zeolite-soil substrates. In our
RH Gp.6
25 107 138 case Zn uptake is seen to increase with increase in the
141 163
170
concentration of the zeolitic component of the substrate.
Shoot Dry Weight (g)

20
It would appear that this apparent contradiction can be
15
explained by the fact that the amendment used in our
work contains an exogenous organic substance (poultry
C Gp.6
10 C Gp.3 C Gp.4 C Gp.5 manure) which, together with the soil and zeolitic com-
RH Gp.1
74 C Gp.2 ponents, causes a high degree of mineralization; shown
5 by the high electrical conductivity of the leachates from
C Gp.1 the amended substrates (Table 3). This assumption is
0
0 50 100 150 200 250 supported by a work that has shown that the introduction
Zeolitic Tuff (g. kg -1substrate) of exogenic humic substances enhances the heavy metal
availability to plants due to the formation of metal–
Fig. 1. Effect of zeolitic tuff on shoot growth, groups 3–6 contain organic complexes, rather than insoluble inorganic
excess untreated tuff. Upper curve: Ryegrass (high density set) grown
in contaminated substrate [numbers refer to respective Zn concentra-
salts (Halim et al., 2003).
tions (mg kg 1) in ryegrass shoots]. Lower curve: Spring Wheat, Although it was found that a lower up-take of copper
control set, grown in uncontaminated soil. occurs in the ryegrass shoots, concentrations above 20–
6 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10

30 mg kg 1 dry weight are known to be toxic to many Table 3

1
plants (Robson and Reuter, 1981) and it is most likely Leachate electrical conductivity, units (AS cm ) and pH in ryegrass
substrates
that this element also contributes to the phytotoxic effect.
Sample EC F S.E. (n = 3) pH
The same is not true of Pb, as concentrations in the shoots
are very low considering the high Pb concentration of the C.Gp 1 Unamended substrate 333 F 29 n.d.
C.Gp 2 Substrate + bio-fertilizer 2150 F 132 n.d.
contaminated soil. However, it is well known that Pb is
C.Gp 3 Ditto + 157 wt.% 4467 F 797 n.d.
not easily translocated in plant tissue and this fact is excess untreated zeolitic tuff
supported by measurements of the concentration factor C.Gp 4 Ditto + 297 wt.% 3860 F 251 n.d.
for this element (Chamberlain, 1983). Such values (ratio excess untreated zeolitic tuff
of metal concentration in plant to metal concentration in C.Gp 5 Ditto + 423 wt.% 3050 F 50 n.d.
excess untreated zeolitic tuff
substrate) have been calculated for other elements in a
C.Gp 6 Ditto + 536 wt.% 4100 F 350 n.d.
range of plants and soils, and it has been found that Zn excess untreated zeolitic tuff
and Cu have a much higher plant uptake than Pb which is R.H. Gp 1 Unamended substrate 125 F 11 8.05
attributed to higher mobility and bioavailability of these R.H. Gp 2 Substrate + bio-fertilizer 375 F 6 7.63
elements in the substrate (Ross, 1994). This characteris- R.H. Gp 3 Ditto + 157 wt.% 185 F 5 7.83
excess untreated zeolitic tuff
tic is also shown in our work by the concentration factors
R.H. Gp 4 Ditto + 297 wt.% 196 F 3 7.86
for ryegrass shoots, growing in the maximum amended excess untreated zeolitic tuff
contaminated soil (i.e. Group 6 plants) which are respec- R.H. Gp 5 Ditto + 423 wt.% 136 F 38 8.03
tively Zn = 0.19, Cu = 0.82 and Pb = 0.01. excess untreated zeolitic tuff
R.H Gp 6 Ditto + 536 wt.% 527 F 12 7.93
excess untreated zeolitic tuff
4.2. Changes in substrate texture, water-filled pore
space and moisture retention C.Gp refers to the Spring Wheat control set grown in uncontaminated
soil, R.H.Gp refers to the high density ryegrass set grown in contam-
inated soil.
As soil water suction pressure varies as a linear
function of soil moisture content (Hillel, 1982) the rate
of change of soil water suction pressure is an indication processes, as a function of water-filled pore space,
of the rate of change of soil moisture. It can be seen applies to the nitrification/denitrification cycle (Linn
(Table 2) that the unamended substrate C.Gp 1 has the and Doran, 1984). It is, therefore, thought that the
highest drying rate and by increasing the concentration bzeolitic waterQ plays a prominent role in supporting
of zeolitic tuff the rate is substantially lowered. This the nitrifying bacterial population which, in turn, bene-
moisture retention property is thought to be due to the fits the plant’s nutrient supply.
desorption of loosely bound water present on, or close
to, the external zeolite surface, as the heat required to 4.3. Effect of excess zeolitic tuff on cation mobilization
remove more tightly bound water, present in the zeolite in soil pore water
channels (Bish and Carey, 2001), would be far greater
than that available in an air-dried soil. As soil bacterial Although no clear differences can be seen in the pH
populations are dependant on a moist environment to be values (Table 3) the changes in electrical conductivity
fully functional (Atlas and Bartha, 1997) it is logical to of the leachates behave in an interesting fashion. The
expect that nitrifiers would remain active longer in the values obtained from the amended uncontaminated sub-
substrate with the slowest drying rate. This is supported strates are an order of magnitude greater than those
by the knowledge that the relative rate of microbial from the contaminated substrates. An identical behav-
iour was seen in the earlier work (Leggo, 2000) which
was correlated to nitrate and major cation concentra-
Table 2
Soil water suction pressure measurements
tions and explained as being due to a surge in the
nitrifying bacterial population. In the present study
Sample Period (h) Change in suction Rate of change
pressure (kPa) (kPa h 1)
the uncontaminated soil substrates show a large initial
change from 333 F 29 AS cm 1 in the unamended
C.Gp 1 120 66.2 0.60
C.Gp 2 216 72.5 0.40
substrate to 2150 F 132 AS cm 1 in the amended sub-
C.Gp 3 264 78.5 0.34 strate (Table 3). This is followed by a further increase
C.Gp 4 206 67.0 0.32 until a plateau of between 3000 and 4000 AS cm 1 is
C.Gp 5 239 69.0 0.25 reached in substrates containing excess untreated zeo-
C.Gp 6 336 55.0 0.11 litic tuff. The contaminated substrates react quite dif-
 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10 7

4
A similar rise in electrical conductivity occurs but in
3.5 C.Gp.3 C Gp.6 this case it is thought that an increase in pore water
C Gp.4 C Gp.5
C Gp.2
3 mobility causes the base metals to become more bio-
log10 E C(µS.cm -1)

C Gp.1
2.5
RH Gp.6
available. These observations infer subtle relationships
RH Gp.2
2 RH Gp.1
RH Gp.3 RH Gp.4
RH Gp.5
between metal phytotoxicity, soil microbial activity and
1.5 plant growth.
1
It should be noted that by increasing the zeolitic tuff
concentration the resulting dilution of the organic ma-
0.5
terial would be expected to impose a growth restriction
0
0 50 100 150 200 250 on the heterotrophic bacteria. In the case of a confined
Zeolitic tuff (g.kg -1substrate) microbial ecology such as a biofilm the chemolitho-
trophic nitrifiers, that would be in competition for
Fig. 2. Influence of zeolitic tuff on leachate electrical conductivity. oxygen and space, might be promoted.
Upper curve: Spring Wheat control, amended uncontaminated soil
substrate. Lower curve: Ryegrass, amended contaminated soil substrate.
4.4. Biofilm formation and preliminary microbiological
investigation
ferently (Fig. 2) as the effect of the addition of extra,
untreated, zeolitic tuff acts to initially depress the elec- Since the early work on microbial slimes (Atkin-
trical conductivity to a minimum limit of 136 F 38 AS son, 1964) it is known that wet surfaces commonly
cm 1, whereas at the highest tuff concentration the support micro-organisms trapped in gelatinous matri-
electrical conductivity of the leachate responds posi- ces of extracellular bio-polymers, mostly polysaccar-
tively by increasing to 527 F 12 AS cm 1.
The behaviour of the leachate electrical conductivity
from the uncontaminated soil substrates mirrors the
enhanced plant growth effect. This correspondence is
thought to be due, in general, to an increase in miner-
alization having a positive effect on the nutrient supply.
In the case of the contaminated substrates increasing
ionic mobility also increases the concentration of bio-
available metal cations. As already shown the phyto-
toxic limits, 100–300 mg Zn kg 1 dry shoot weight and
20–30 mg Cu kg 1 dry shoot weight are quickly
reached in the amended contaminated substrates, and
appear to be detrimental to plant growth.
The erratic behaviour in electrical conductivity of
the leachates from the contaminated substrates, contain-
ing in excess of 200 g kg 1 zeolitic tuff, indicates that
changes are occurring in the ionic mobility of the pore
water that influence plant growth behaviour. It would
be expected that an increase in metal ion concentration
would adversely affect the microbial community and
ultimately result in a phytotoxic effect, as postulated.
The abrupt rise in the conductivity of the RH Gp.6
leachate sample implies that relief of microbial stress
might be occurring, at this level of tuff amendment of
contaminated soil. In the case of Spring Wheat, grow-
ing in uncontaminated soils, the abrupt increase in dry
weight correlates with an increase in soil ionic mobility
as would be expected, from former observations. A
similar growth enhancement is not seen in the plants Fig. 3. Clinoptilolite crystal surfaces. Top: Clean unmodified surfaces
from the contaminated soils, as a slight reduction in found in dry, untreated, zeolitic tuff. Bottom: Biofilm covered crystal
plant growth occurs above a concentration of 157 wt.%. surfaces, from a moist amended substrate.
8 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10

ides and water. It is now generally accepted that ables and at present it is not possible to explain the
bacteria, in most ecosystems, are attached to surfaces interaction in detail. Much more work will be required
by bbiofilmsQ, which now have become recognised in to determine the interactive forces between zeolite sur-
many different environments. A comparison has been faces, microbial cells, their enclosing biofilm and the
made (Fig. 3) between, essentially dry, clinoptilolite resulting mechanism of enhanced nitrification.
crystals taken from a hand specimen of zeolitic tuff In an initial attempt to study the microbiology of the
and those of the same material from one of the organo-zeolitic system viable counts obtained from
amended uncontaminated soil substrate, in a moist amended soil samples ranged from 1.16108 to
condition (C.Gp 3). The sample from the soil sub- 1.57108 CFU/g. This was of the same order of mag-
strate shows the clinoptilolite surfaces to be covered nitude as the control sample (1.20108 CFU/g). Colo-
in a biofilm which encapsulate clusters of densely ny morphologies between the control, amended soils,
packed bacteria together with other granular particles, and bacteria isolated from zeolite tuff grains, were also
whereas those of the dry sample show no trace of a comparable in appearance and constitution. Microscop-
biofilm. Although the zeolite biofilm has a typical ic investigation reveals an abundant diversity of un-
morphology the cell size of the encapsulated bacteria specified small rod-shaped bacteria in all the samples
is much smaller than normal. This is though to be due including the biofilm. Isolates of the predominant col-
to shrinkage that has occurred during cryofixation of ony type were identified by partial 16s rRNA sequenc-
the sample. Apart from the uniform coating of bacte- ing, essentially as described by Janssen et al. (2002),
rial cells on the crystal surfaces irregular shaped using the oligonucleotide primers 27f and 1492r, and
clusters, in some cases connected by minute strands sequencing primer 519r. Sequence comparison with
of extra cellular film, are seen to be part of the those in the GenBank database using BLAST software
complex. That clinoptilolite in a damp organo-zeolit- (Altschul et al., 1990) identified several isolates as
ic-soil substrate is susceptible to biofilm formation is belonging to the genus Ralstonia. However, at this
not surprising as the strong bonding of microbes in stage it is unclear whether these bacteria, or indeed
clinoptilolite-coagulant complexes was found in water any of the other isolates, are responsible for any of
treatment plants that use zeolitic tuff in water purifi- the perceived benefits associated with the organo-zeo-
cation processes (Kalló, 2001). This property is used litic enriched soil. Although the presence of chemo-
successfully in filter beds to remove bacteria such as lithotrophic nitrifying species remains, at this time,
total coliforms, fecal coliforms, fecal streptococci and undetected it is realised that in situ methods of molec-
dissolved organic matter (Garcia et al., 1992a). In this ular biology e.g., fluorescent in situ hybridization per-
study water was percolated through a volcanic tuff formed with 16S rRNA-targeted oligonucleotide
comprising 60 wt.% phillipsite, 32 wt.% unaltered probes, are required to identify ammonia-oxidizing
glass, 7 wt.% plagioclase feldspars and 1 wt.% cal- and nitrite-oxidizing bacteria and such studies are
cite. After treatment it was observed that an amor- planned for the future.
phous biofilm containing an agglomeration of micro-
organisms was found to cover most of the material. 4.5. Sustainability of plant growth
It is pertinent to remark, here, that enhancement of
nitrification can be produced artificially by the co-im- Although metal ions are far more mobile in the
mobilization of clinoptilolite and nitrifying bacteria by amended substrates their bioavailability does not, in
sodium alginate (Yang, 1997). It is thought that the the substrates studied, cause a degree of metal toxicity,
zeolite surface property acts to provide a substrate to in either Spring Wheat (Leggo and Ledésert, 2001) or
the bacteria by the transfer of ammonium cations. In ryegrass, which is fatal to the plant. On the contrary, a
this respect the observations reported in the present very large increase in root and shoot growth occurs in
work appear little different although, in our case, the all the amended groups. It is only on further inspection
effect appears to be a natural phenomenon. of the data that it becomes apparent, in the case of
The present study, although far from definitive, sug- ryegrass grown in the contaminated substrates, that a
gests that water present in the zeolite pore space and limit does exist above which increasing the amount of
lost by de-sorption to the external crystal surfaces un-ammoniated zeolitic tuff begins to affect plant
provides a film of water molecules which will accom- growth adversely. However, the improved nutritional
modate the adsorption of bacteria and formation of a quality of the contaminated substrate allows ryegrass to
biofilm. However, the attachment of micro-organisms develop a vastly improved root structure (Fig. 4). At the
to surfaces is a complex process involving many vari- end of the experiments described in this paper the
 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10 9

is thought to be due to a subtle alteration of the

substrate which enhances microbial activity to the ben-
efit of the plant. Changes in moisture content together
with the susceptibility of zeolite crystal surfaces to
sustain biofilm formation are put forward to explain
this phenomenon.
It is demonstrated that loosely bound bzeoliticQ
water plays a role in maintaining a water potential
at which nitrifiers would survive droughty soil con-
ditions and remain active for longer periods that
those in unamended substrates. If the loss of soil
moisture can be held below the bacteriological stress
level by the desorption of bzeoliticQ water then the
relationship between the amount of zeolitic tuff pres-
ent in the substrate and the increase in plant growth
would be expected to be positively correlated. The
leachate conductivities of the uncontaminated control
set are seen to support the concept of enhanced plant
growth being due to increased mineralisation. How-
ever, the data from the leachate analysis of the
contaminated substrates is much more difficult to
interpret and it will require further study to reach a
positive conclusion.
It is encouraging that the work with perennial rye-
grass supports the earlier conclusion (Leggo and Ledé-
sert, 2001) which suggested that plants can be
sustained on barren, metal contaminated sites by the
incorporation of organo-zeolitic bio-fertilizer. Field
scale trials, using an identical organo-zeolitic amend-
Fig. 4. Morphological characteristics of the ryegrass root system. Top: ment, at Lynn Lake, Manitoba, Canada together with
25 plants grown as a set in the unamended contaminated substrate; laboratory experiments with a metal tolerant plant (un-
estimated 10% loss of fine roots during washing. Bottom: 25 plants published data, Ledésert et al., in preparation) show
grown as a set in the amended contaminated substrate; the dense root that plant growth can be successfully sustained in sub-
structure made separation impracticable.
strates containing extremely high metal values and in
the Canadian case over years in freezing sub-arctic
ryegrass grown in the contaminated substrates had winter conditions.
shown little overall change for twelve months with
the morphological and growth differences being sus- Acknowledgements
tained without additional treatment.
The authors would like to thank members of both the
5. Conclusions Department of Earth Sciences and the Botanic Garden,
University of Cambridge who gave their support to the
The present work demonstrates that plant growth project. Financial support has been provided by the
rate responds very positively to the influence of zeolite Concerted Research Program (PRC) of the French
bio-fertilizer. This is thought to be due to active min- Nord-Pas de Calais Council and the CNRS (UMR
eralization caused by a large increase in the population PBDS), University of Lille I.
of chemolithotrophic nitrifying bacteria, sponsored ini-
tially by NH4+ ions diffusing from the ammoniated References
zeolite in the bio-fertilizer. It has now been found
Alloway B, Jackson A. The behaviour of heavy metals in sewage
that when extra untreated zeolitic tuff is added to sludge-amended soils. Sci Total Environ 1991;100:151 – 76.
substrates, already amended with the bio-fertilizer, a Altschul S, Gish W, Miller W, Myers E, Lipman D. Basic local
substantial further increase in plant growth occurs. This alignment search tool. J Mol Biol 1990;215:403 – 10.
10 P.J. Leggo et al. / Science of the Total Environment 363 (2006) 1–10

Ames L. Zeolite removal of ammonium ions from agricultural and editors. Metal-contaminated soils: in situ inactivation and phytor-
other waste waters. Proc 13th pacific northwest industrial waste esoration. Springer-Verlag; 1998. p. 197 – 210.
conf. Pullman, Washington7 Washington State Univ; 1967 Linn D, Doran J. Effect of water-filled pore space on carbon dioxide
p. 135 – 152. and nitrous oxide production in tilled and nontilled soils. Soil Sci
Andronikashvili T, Tsitsishvili G, Kardava M, Gamisonia M. The Soc Amer J 1984;48:1267 – 72.
effect of organic-zeolite fertilizers on microbial landscape of soil. Long X, Yang X, Ni W. Current status and perspective on phytor-
Izv Acad Sci Georgia 1999;T 25(3–4):243 – 52. emediation of heavy metal polluted soils. J Appl Ecol 2002;13
Atkinson B. Biochemical reactors. London7 Pion Press; 1964. 757 – 62.
Atlas R, Bartha R, 1997. Microbial ecology: fundamentals and appli- McGilloway R, Weaver R, Ming D, Gruener J. Nitrification in a
cations, 4th ed. Menlow Park California7 Benjamin/Cummings; zeoponic substrate. Plant Soil 2003;256(2):371 – 8.
1997. p. 309 – 13. McGrath S. Effects of heavy metals from sewage sludge on soil
Bish D, Carey J. Thermal behaviour of natural zeolites Natural microbes in agricultural ecosystems. In: Ross S, editor. Toxic
zeolites: occurrence, properties, applications, reviews in mineral- metals in soil-plant systems. UK7 J.Wiley and Sons; 1994
ogy and geochemisrty. Min Soc America 2001;45:403 – 52. p. 247 – 74.
Bloomfield H, Handley J, Bradshaw A. Nutrient deficiencies and the McInyre T. Phytoremediation of heavy metals from soils. Adv Bio-
aftercare of reclaimed derelic land. J Appl Ecol 1982:151 – 8. chem Biotechnol 2003;78:97 – 123.
Bradshaw A, Chadwick M. The restoration of land. Oxford7 Black- Minato H. Characteristics and uses of natural zeolites. Kaotsugasu 5.
well Scientific Publications; 1980. 317 pp. Mumpton FA, editor. Worldwide deposits and utilisation of
Brooks R, Chiarucci A, Jaffré T. Revegetation and stabilization of natural zeolites. Indust Minerals, No.73, 1968, pp. 40 – 5,
mine dumps and other degraded terrain. In: Brooks R, editor. 536 – 47.
Plants that hyperaccumulate heavy metals. Oxford7 CAB Inter- Ming D, Allen E. Use of natural zeolites in agronomy, horticulture
national. p. 227 – 47. and environmental soil remediation Reviews in mineralogy and
Burlet E. Les tufs volcaniques zéolitisés des Rhodopes (Bulgarie) geochemistry. Mineral Soc Am 2001;45:619 – 54.
caractérisation, propriétés physico-chimiques, MSc thesis, Uni- Mumpton FA. La roca magica: uses of natural zeolites in agri-
versité Aix-Marseille 111, France. 1997. 17 pp (in French with culture and industry. Proc Natl Acad Sci U S A 1999;96:
English abstract). 3463 – 70.
Castaldi P, Santona L, Melis P. Heavy metal immobilization by Passaglia E, Sheppard R. The crystal chemistry of zeolites. In:
chemical amendments in a polluted soil and influence on white Bish DW, Ming DW, editors. Natural zeolites: occurrence, prop-
lupin growth. Chemosphere 2005;60:365 – 71. erties, applications, reviews in mineralogy and geochemistry;
Chamberlain A. Fallout of lead and uptake by crops. Atmos Environ 2001. p. 89 – 91.
1983;17:693 – 706. Paul E, Clark F. Soil microbiology and biochemistry. London7 Aca-
Garcia J, Gonzales M, Notario J. Removal of bacterial indicators of demic Press; 1996. p. 194 – 6.
pollution and organic matter by phillipsite-rich tuff columns. Appl Robson A, Reuter D. Diagnosis of copper deficiency and toxicity. In:
Clay Sci 1992;7:323 – 33. Loneragan JF, Robson AD, Graham RD, editors. Copper in soils
Halim M, Conte P, Piccolo A. Potential availability of heavy metals to and plants. London7 Academic Press; 1981. p. 287 – 312.
phytoextraction from contaminated soils induced by exogenous Ross S. Toxic metals: fate and distribution in contaminated ecosys-
humic stubstances. Chemosphere 2003;52:265 – 75. tems. In: Ross SM, editor. Toxic metals in soil-plant systems. UK7
Hillel D. Introduction to soil physics. Academic Press; 1982 J.Wiley and Sons; 1994. p. 189 – 235.
p. 57 – 86. Ruano A, Poschrenrieder Ch, Barcelo J. Growth and biomass
Janssen P, Yates P, Grinton B, Taylor P, Sait M. Improved culturability partitioning in zinc-toxic bush beans. J Plant Nutr 1988;11
of soil bacteria and isolation in pure culture of novel members of 577 – 88.
the divisions Acidobacteria, Actinobacteria, Proteobacteria, and Salt D, Smith R, Raskin I. Phytoremediation. Annu Rev Plant Physiol
Verrucomicrobia. Appl Environ Microbiol 2002:2391 – 6. Plant Mol Biol 1998;49:643 – 68.
Kalló D. Applications of natural zeolites in water and wastewater Semmens M, Seyfarth M. The selectivity of clinoptilolite for certain
treatment Reviews in mineralogy and geochemistry. Min Soc heavy metals. In: Sand LB, Mumpton FA, editors. Natural zeo-
Amer 2001;45:519 – 50. lites: occurrence, properties, use. New York7 Pergammon Press;
Ledésert B, Leggo P, Demulsant X, Gauthier A, Durand, A, Saumi- 1978. p. 517 – 26.
tou-Laprade P. Use of an organo-zeolitic fertilizer to sustain plant Tsitsishvili GV, Andronikashvili TG, Kirov GN, Filizova. Natural
growth in highly contaminated river sediments. Tests with a metal Zeolites. New York7 Ellis Horwood; 1992. p. 163 – 4.
accumulator (Arabidopsis halleri). In preparation. Yang L. Investigation of nitrification by co-immobilized nitrifiying
Leggo P. An investigation of plant growth in an organo-zeolitic bacteria and zeolite in a batchwise fluidized bed. Water Sci
substrate and its ecological significance. Plant Soil 2000;219 Technol 1997;35(8):169 – 75.
135 – 46. Walker C, Hopkin S, Sibley R, Peakall D. Principles of ecotoxicology.
Leggo P. Method of sustaining plant growth in toxic substrates New York7 Taylor and Francis; 1996. p. 53 – 8.
polluted with heavy metal elements. European Patent No. EP 1 Witter E, Lopez-Real J. Nitrogen losses during the composting of
208 922 B1, patent bulletin 04/10 March 2004. sewage sludge and the effectiveness of clay soil, zeolites and
Leggo P, Ledésert B. Use of organo-zeolitic fertilizer to sustain plant compost in absorbing the volatilized ammonia. Biol Wastes
growth and stabilize metallurgical and mine waste sites. Mineral 1988;23:279 – 94.
Mag 2001;65(5):563 – 70. Zamzow M, Eichbaum B, Sandgren K, Shanks D. Removal of heavy
Li Y-M, Chaney R. Phytostabilization of zinc smelter-contaminated metals and other cations from waste water using zeolites. Sep Sci
sites: the Palmerton case. In: Vansgronveld J, Cunningham S, Technol 1990;25:1555 – 69.