User manual | ZEA MAYS L GROWN ON TWO CONTRASTING SOILS By

VANADIUM AVAILIBILTY TO MAIZE (ZEA MAYS L.)
GROWN ON TWO CONTRASTING SOILS
By
Jandre McCoy Bekker
Submitted in partial fulfilment of the requirements for the
degree MSc. Agric Soil Science
In the faculty of Natural & Agricultural Sciences
University of Pretoria
Pretoria
Supervisor: Mr. P.C. de Jager
Co-supervisor: Dr. E.H. Tesfamariam
Declaration:
I, Jandre McCoy Bekker declare that the thesis, which I hereby submit for the degree
MSc. Agric Soil Science at the University of Pretoria, is my own work and has not
previously been submitted by me for a degree at this or any other tertiary institution.
SIGNATURE: Jandre McCoy Bekker
DATE: 09/07/2014
i
TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION ..................................................................................... 1
1.1 Background................................................................................................................ 1
1.2 Study aims ................................................................................................................. 3
CHAPTER 2: LITERATURE REVIEW ........................................................................... 5
2.1 Soil acidity ................................................................................................................. 5
2.1.1 Origin of soil acidity ................................................................................................ 5
2.1.2 Effect of soil acidity on nutrient availability, and plant growth .................................. 7
2.2 Liming materials ........................................................................................................ 9
2.3 Vanadium in the soil ................................................................................................ 12
2.3.1 Origin and concentration in soils ............................................................................ 12
2.3.2 Vanadium dynamics in soil .................................................................................... 13
2.3.3 Speciation of vanadium ......................................................................................... 13
2.3.4 Vanadium sorption ................................................................................................ 15
2.3.5 Vanadium mobility ................................................................................................ 16
2.4 Vanadium in the plant .............................................................................................. 16
2.4.1 Concentration and toxicity ..................................................................................... 16
2.4.2 Vanadium uptake and translocation in plants .......................................................... 18
2.4.3 Vanadium interaction with other nutrients .............................................................. 18
2.5 Guidelines for vanadium and the use of slag.............................................................. 19
2.6 Extraction methods and analyses............................................................................... 19
CHAPTER 3: VADADIUM TOXICITY THRESHOLD LEVELS FOR MAIZE: A
POT TRIAL ................................................................................................................. 21
3.1 Introduction ............................................................................................................. 21
3.2 Aims ........................................................................................................................ 22
3.3 Materials and methods .............................................................................................. 22
3.3.1 Soils used in this study .......................................................................................... 22
3.3.2 Vanadium content of mono ammonium phosphate.................................................. 23
3.3.3 Liming, fertilization and vanadium loading ............................................................ 23
3.3.4 Planting and harvesting the maize in the vanadium toxicity trail.............................. 24
3.3.5 Chemical analysis .................................................................................................. 25
3.3.6 Statistical analysis ................................................................................................ 26
ii
3.4 Results and discussion .............................................................................................. 26
3.4.1 Background concentrations of vanadium in soils .................................................... 26
3.4.2 Effect of vanadium loading rates on dry matter yield .............................................. 27
3.4.3 Toxicity at total vanadium concentration in the soil and the plant ............................ 30
3.4.4 Toxicity symptoms ................................................................................................ 31
3.4.5 The extractability of vanadium; comparing Bray 1 and ammonium acetate .............. 32
3.4.6 Root analysis ......................................................................................................... 34
3.4.7 Vanadium interaction with other nutrients .............................................................. 35
3.5 Conclusion ............................................................................................................... 36
CHAPTER 4: PHYTO-AVAILABILITY OF VANADIUM FROM BASIC
OXYGEN FURNACE SLAG USED IN AGRICULTURE ............................................ 39
4.1 Introduction ............................................................................................................. 39
4.2 Aims ........................................................................................................................ 40
4.3 Materials and methods .............................................................................................. 40
4.3.1 Slag trial ............................................................................................................... 40
4.3.2 Field evaluation ..................................................................................................... 45
4.3.3 Soil and plant analysis ........................................................................................... 47
4.3.4 Statistical analysis ................................................................................................. 47
4.4 Results and discussion .............................................................................................. 47
4.4.1 Biomass production of maize as influenced by soil acidity and liming ..................... 47
4.4.2 Estimating vanadium loading rates for the slag pot trail .......................................... 48
4.4.3 Vanadium plant availability and plant uptake for the slag pot trial ........................... 51
4.4.4 Vanadium translocation in the maize plant in the pot trail ....................................... 52
4.4.5 Influence of vanadium loading through slag application on its Bray 1
extractability of vanadium in the pot trail ....................................................................... 52
4.4.6 Influence of vanadium loading through slag application on its ammonium
acetate extractability in the pot trail ............................................................................... 53
4.4.7 Estimating vanadium loading rates for the field evaluation sites .............................. 54
4.4.8 Vanadium plant availability and plant uptake in the field evaluation site ................. 57
4.4.9 Vanadium translocation in the maize plant from the field evaluation sites ............... 58
4.4.10 Influence of vanadium loading through slag application on Bray 1
extractability of vanadium at the field evaluation site ..................................................... 58
iii
4.4.11 Influence of vanadium loading through slag application on acetate
extractability of Vanadium at the field evaluation site .................................................... 62
4.5 Conclusion ............................................................................................................... 62
CHAPTER 5: SUMMARY AND CONCLUSION .......................................................... 64
REFERENCES .............................................................................................................. 67
APPENDIX A: Google maps of the Ogies and Delmas evaluation sites ........................... 78
APPENDIX B: The frequency distribution and cumulative normal distribution
curves for the different soil and soil depths. ................................................................... 80
iv
LIST OF TABLES
TABLE 2.1 Average composition of slags before refining into agricultural lime
(Source: Columbus Steel and Highveld Steel)…………….........................................12
TABLE 3.1. The total concentrations of some of the elements in the two soils
determined by a XRF scan…………………………………………………………...22
TABLE 3.2 A summary of the results from the V toxicity trail (total V concentrations,
Bray 1 and ammonium acetate extractable)……………………………..…………...30
TABLE 4.1. The calcium carbonate equivalent (CCE) values for the three slags in the
two soils………………………………………………………………………………42
TABLE 4.2. The liming rates for the different soils and the different liming
materials……………………………………………………………………………...45
TABLE 4.3 The different V accumulation rates and the predicted years to reach the V
threshold values of the slag used in the Nooit and PPS soil …………………...........50
v
LIST OF FIGURES
FIGURE 3.1 V concentrations in soils unamended with V compared to soils that
received 250 mg kg-1 NH4VO3. .............................................................................. 27
FIGURE 3.2 The effect of increasing V concentrations in on the growth of maize
plants (Nooit soil). ................................................................................................. 28
FIGURE 3.3 Aboveground biomass of maize plants (Dry mass) as affected by
increased V loading rate. ........................................................................................ 29
FIGURE 3.4 A maize plant showing stunted growth at the highest V loading rate (250
mg kg-1) in the Nooit soil after six weeks. ............................................................... 32
FIGURE 3.5 The Bray 1 extractable V and P as influenced by the increasing V
loading rates. ......................................................................................................... 33
FIGURE 3.6 The ammonium acetate extractable K at different V loading rates. ...... 36
FIGURE 4.1. The V concentrations of the three slags as determined by various
methods. ................................................................................................................ 41
FIGURE 4.2. The vanadium, chromium and nickel content of the three slags used as
determined by XRF and HClO4:2HNO3 acid digestion. ........................................... 41
FIGURE 4.3. The pH values of the soil solution after the two soils were limed with
the various liming material. .................................................................................... 43
FIGURE 4.4 The dry mass of the maize plants (above ground biomass) grown in the
PPS soil at the various target pH levels achieved with different liming rates. ........... 48
FIGURE 4.5 The total V concentrations in the Nooit soil after the different liming
treatments. ............................................................................................................. 51
FIGURE 4.6 The total V concentrations in the PPS soil after the different liming
treatments. ............................................................................................................. 51
FIGURE 4.7 The Bray 1 extractable V in the Nooit soil as influenced by the different
lime treatments. ..................................................................................................... 53
FIGURE 4.8 The Bray 1 extractable V in the PPS soil as influenced by the different
lime treatments. ..................................................................................................... 53
FIGURE 4.9 The frequency distribution for the total V concentrations of the 0-30 cm
soil samples at the Delmas site, (control samples omitted)....................................... 54
vi
FIGURE 4.10 The total V cumulative normal distribution curve for the 0-30 cm soil
samples of the Delmas site. The red data points represent the samples collected
outside the field. The orange data point is the average of the outside samples. ......... 55
FIGURE 4.11 The total V concentration frequency distribution of the 0-30 cm soil
samples at the Ogies site, (control samples omitted). ............................................... 56
FIGURE 4.12 The cumulative normal distribution curve for the 0-30 cm soil samples
of the Ogies site. The red data points represent the samples collected outside the field.
The orange data point is the average of the outside samples. ................................... 57
FIGURE 4.13 The V concentrations in the above ground maize plant at the two field
evaluation sites at 120 days of growth. ................................................................... 58
FIGURE 4.14 The Bray 1 extractable V frequency distribution of the 0-30 cm soil
samples at the Delmas site, without the control samples. ......................................... 59
FIGURE 4.15 The cumulative normal distribution curve for the Bray 1 extractable V
of the 0-30 cm soil samples of the Delmas site. ....................................................... 60
FIGURE 4.16 The Bray 1 extractable V frequency distribution of the 0-30 cm soil
samples at the Ogies site, without the control samples. ............................................ 60
FIGURE 4.17 The cumulative normal distribution curve for the Bray 1 extractable V
of the 0-30 cm soil samples of the Ogies site .......................................................... 61
vii
LIST OF ACRONYMS AND ABBREVIATIONS
AEC anion exchange capacity
BOF basic oxygen furnace slag
CCE calcium carbonate equivalent
CEC cation exchange capacity
ICP-AES inductively coupled plasma atomic emission spectrophotometer
viii
ABSTRACT
Soils on the Eastern Highveld acidify naturally due to the high rainfall and the use of
certain nitrogen fertilizers. Liming materials are used to ameliorate soil acidity. A
large fraction of this liming material used is Basic oxygen furnace slag (BOF slag), a
secondary by-product from the Iron and Steel industry, commercially known as
Aglime. These slag contain various concentration of heavy metals, like vanadium.
No vanadium guidelines exist in South Africa. This study was done to determine the
V threshold values where maize experienced reduced growth and to determine the V
loading rate through slag application.
A pot trial was used to determine the concentrations where vanadium reduces plant
growth, and to establish toxicity levels in maize. The pot trail was used to establish V
threshold values with various indicators, like total V concentration in the soil, total V
concentration in the plant, Bray 1 extractable V and ammonium acetate extractable V,
where maize experienced reduced growth.
The threshold value where maize showed reduced growth in sandstone derived soils
was at a total V concentration in the soil of 73.3 mg kg-1. The Bray 1 extractable V at
this threshold was 23.5 mg kg-1 and there was no V in the above ground plant material
in the maize. The ammonium acetate extractability at this level was 1.68 mg kg-1. V
toxicity occurred at a total V concentration of 150 mg kg-1, with Bray 1 extractable V
at 77.6 mg kg-1 and total V in the maize plant 14.8 mg kg-1
For the dolerite derived soil the threshold value was determined to be 235 mg kg-1 for
the total V concentration in the soil. The Total V concentration in the plant was 0.5
mg kg-1 and the Bray 1 extractable V was 30.3 mg kg-1. The ammonium acetate
extractable V was 1.69 mg kg-1.
A pot trail and field evaluation site was used to determine the V loading through slag
application. Three slag where used containing different V concentrations, slag A
containing the highest V (918 mg kg-1) and B (153 mg kg-1) and C (88.6 mg kg-1) had
ix
a lower V concentrations. Theoretical V loading values was determined for three
different slags containing different V concentrations and by using the threshold V
concentration generated in chapter 3, the period to reach the critical V threshold value
for liming with slag A was determined. If all factors (V concentration and
incorporation depth), were to be kept constant, it will take an estimated 186 years of
liming with slag A for the sandstone derived soil to reach the threshold value of 100
mg kg-1 where V negatively affect the growth of maize plants. This period was
calculated to be 472 years for the dolerite derived soil, due to the higher Fe content
and finer textured soil, which increase the V sorption capacity of the soil. The safe
period for the slag B and C in respect with V is much longer than slag A, but other
heavy metal concentrations must be kept in mind for they too can accumulate in the
soil and can influence the growth of maize negatively if certain threshold values are
reached.The V concentration of all the parameters generated in the V toxicity pot trail
was far below the threshold values of the slag pot trail and field evaluation site. This
indicated that the slag use with high V concentration on the short term (10 years) is
unlikely to negatively influence crop production.
x
CHAPTER 1: INTRODUCTION
1.1 Background
South Africa has limited agricultural resources and is classified as an arid country
because the mean annual rainfall is 497 mm which is far less than the world average
of 860 mm. The annual potential evaporation ranges from 1100 mm to 3000 mm
(Huntley et al., 1989). Only 14 % of the land is suitable for dry land production of
which only 3 % is high potential (Scotney et al., 1990). Mining activities on the
Eastern Highveld have increased drastically over the last ten years and the impact on
arable land is devastating, especially in areas where open cast mining is practiced,
resulting in the loss of precious agricultural land. In most cases the most fertile land is
removed for mining and so the land available for sustainable food production on the
Eastern Highveld and has decreased. It is therefore of great importance that the fertile
land that is left for crop production should be cultivated responsibly considering the
fact that there is a world food shortage at hand (FAO, 2011). One of the most
common soil degradation problems encountered in highly weathered soils under
cultivation is soil acidification. Soil acidification significantly affects crop production
and should be prevented to ensure sustainable crop production for the centuries to
come.
Soils on the Eastern Highveld (Area include eastern part of Gauteng, Southern part of
Mpumalanga and Northern part of the Free State in South Africa) acidify naturally
due to the high rainfall (800 mm y-1) and the use of certain nitrogen fertilizers,
especially ammonium based fertilisers (Beukes, 1995). Most of these soils are
predominantly derived from sandstone parent materials, which have a low buffer
capacity and are sensitive to acidification (Fey & Dodds, 1998). Farmers on the
Eastern Highveld use tons of liming material each year to ameliorate soil acidity
aiming to maintain a soil pH (H2O) range of between 6 and 6.5. This is the pH range
where the essential elements for plant growth are readily available and where some
elements like aluminium (Al), iron (Fe) and manganese (Mn) are unlikely to be toxic
to plants. An estimated 4.7 million mega gram of lime are needed for the cultivated
1
land in South Africa to ameliorate soil acidity and about 1 ton of lime is needed
annually to prevent renewed acidification (DME, 2005).
A large fraction of this liming material used is BOF slag, a secondary by-product
from the Iron and Steel industry, commercially known as Aglime. Slag contain
various oxides, carbonates and silicates, and is sieved to micro fine particles (<0.25
mm), and therefore have a relative high acid neutralizing capacity. However, the
disadvantage of slag is that it might contain heavy metals. Some slag used, for
example, contains high levels of vanadium (V).
High levels of V can be toxic to certain plant species, which lead to an under
developed root system and stunted growth. In animals the toxicity of V has been
found to be high when it’s been given parentally, low when it is orally administrated
and moderate in the case of respiratory exposure. In humans both acute and chronic
poisoning have been described in workers engaged in the industrial production and
use of high levels of V. Most of the reported clinical symptoms reflect irritate effects
of V on the respiratory (Vouk, 1979).
Vanadium has not been a specific research focus in South Africa and little work has
been done to understand V dynamics in agriculture. Other heavy metals like lead (Pb),
nickel (Ni), cadmium (Cd) and chromium (Cr) have been the main focus of various
studies. Previous and current studies at the University of Pretoria focussed on slag use
in agriculture, however V dynamics was not the main focus of this investigation (van
der Waals, 2001).
No stipulated guidelines exist in South Africa governing heavy metal additions to
soils through the application of slag. The proposed allowable heavy metal
concentrations for fertilizers (Act No. 36 of 1947) do not include V. This is due to the
fact that little research has been done on V and possibly also because V is not
considered to be as environmentally detrimental as Pb and Cd. In this study the
maximum permissible V concentration in soil and maize plant will be determined.
There is no prescribe method to analyse plant available V, which is a more accurate
2
measurement to test for V toxicity in soils. Bray1 and ammonium acetate extraction
methods will be tested as V extraction methods for soil to predict the level of V
accumulation.
The risk posed by V as a result of agricultural slag use, is a function of 1) The V
concentration in the slag; 2) The amount of slag added per unit area to ameliorate soil
acidity; 3) Soil properties, (for example, clay mineralogy, clay content and
exchangeable acidity) of the soil receiving the slag; 4) the incorporation depth of slag
in the soil, 5) the type of crop to be planted on that soil, as well as the number of years
that the slag was used.
Therefore, it is of vital importance to understand the dynamics of V in the soil system,
to ensure the long-term sustainability of using slag in South African agricultural lands
for food production.
1.2 Study aims
The overarching aim of this study was to gain insight into the dynamics of V in the
plant- soil environment, focus on maize. The objectives were to:
a) Determine the threshold levels in two contrasting soils at which the growth of
maize are influenced and / or toxic symptoms appear. It is hypothesised that the
concentration where the growth of maize is negatively influenced will differ between
soils with different V sorption capacities.
b) Evaluate the use of two extraction methods, to create threshold concentrations were
V becomes toxic. It is hypothesised that Bray 1 would be a better extraction method
compared to ammonium acetate because of the suggested analogy between phosphate
and vanadate, and that vanadate is the dominate V state in oxidised conditions.
c) Determine the V loading rate through slag application and estimate the duration a
specific slag can be used before plant growth is expected to be negatively impacted,
and investigate the translocation of V in the maize plant. It is hypothesised that V
loading over the short term (10 years) would not have a negative effect on the growth
3
of maize plants, but might become a problem when V accumulation takes place over
100 years and longer.
4
CHAPTER 2: LITERATURE REVIEW
2.1 Soil acidity
Soil acidification is a natural process, which can be accelerated through the activity of
plants, animals and humans or can be managed through good farming practises. Van
Breemen (1991) describes soil acidification in several ways; an increase in soil acidity
and / or a decrease in soil pH, a decrease in base saturation, an unbalanced availability
of elements in the root environment, or a decrease of the acid neutralising capacity of
the soil. An acid soil is a soil with a pH (H 2O) below 7 which might contain
phytotoxic Al and Mn compounds at pH below 4 (Venter, 2004).
2.1.1 Origin of soil acidity
Acidification is the result of the addition, or generation of H+ ions. The most common
contributor of H+ ions is when carbon dioxide dissolves in water to form carbonic
acid. Carbonic acid dissociate releasing H + ions (Equation 2.1), but the contribution of
H+ ions is negligible when the pH is below 5.0. Another contributor is organic
material that was subjected to microbial mediated oxidation. Humic substances
formed, contain, for example, carboxylic functional groups (Equation 2.2) that can
protonate and deprotonate depending on the pH. Decreasing the pH below the pKa
value these functional groups will result in their protonation and increase in
exchangeable or reserve acidity of the soil. An increase in pH will result in the release
of the protons in the soil solution, increasing soluble acidity and buffering pH change.
Dissolved organic substances can also form soluble complexes with basic cations
(Ca2+ and Mg2+) and facilitate the loss of these cations through leaching.
CO2 + H2O → H2CO3  HCO3- + H+ pKa = 6.35
(2.1)
[RCH2OH…] + O2 + H2O  RCOOH ↔ RCOO- + H+ pKa = 3 to 5
(2.2)
5
H+ ions are generally produced during oxidation. During nitrification, ammonium
(NH4+) ions from organic matter or fertilizers are subjected to microbial oxidation and
release two H+ ions for each NH4+ ion oxidized ( Equation 2.3).
NH4+ + 2O2  H2O + 2H+ + NO3-
(2.3)
Organic –SH groups and sulphur containing minerals like pyrite (FeS2) can also be
oxidised to form sulphuric acid (Equation 2.4). Pyrite is a mineral commonly
associated with coal and is the main cause of acid mine drainage on the Eastern
Highveld. When these acids dissociate in water, H + ions are produced. This is
typically how acid rain is formed. Sulphur and nitrogen gasses from the combustion
of fossil fuels, forest fires and lightning dissociate completely in rain drops because
they are strong acids with low pKa values (Equation 2.5 & 2.6). Plants also contribute
to H+ ions to the soil. Plants must balance their cations and anions uptake. When a
cation is taken up, the plant can take up an anion, or exude H + ions to maintain
balance between positive and negative charges in the plant (Brady & Weil, 2003).
FeS2 + 3½O2 + H2O  FeSO4 + 2H+ + SO42-
(2.4)
H2SO4  SO42- + 2H+
(2.5)
HNO3  NO3- + H+
(2.6)
H+ ions formation as a result of dissociation reactions (Brady & Weil, 2003).
Soils in high rainfall areas acidify naturally because of the leaching of basic cations
(calsium (Ca), magnesium (Mg) and potassium (K)) in well drained soils. This
happens in the case of N and S mineralization and oxidation of organic N and S when
H+ ions are produced. Protons will replace above mentioned basic cations and these
cations in turn will leach with NO 3- and SO42- as charge balancing cation. The H+
ions, however, remain behind in the topsoil, resulting in the acidification of the
topsoil. This will lower soil pH in this layer and the pH buffer capacity (Bolan et al.,
2003). The intensity of acidification is dependent on the geology and topography of a
specific area. This is a slow natural process (Helyar & Porter, 1989). Soils in the
Highveld of the Mpumalanga Province are very sensitive to acidification because of
6
high rainfall (800 mm year-1) and the majority of these soils are sandier soils with
poor buffering capacities (Fey & Dodds, 1998). This natural process can be
accelerated by human activity such as inappropriate fertiliser use. Apart from acid
rain, ammonium based nitrogen fertilisers and urea has been the main anthropogenic
contributors to acidification of agricultural soils. With the addition of N fertilizers,
nitrification produces H+ and NO3-. These products from nitrification can leach below
the root zone that can lead to subsoil acidity (Rowell & Wild, 1985; Bouman et al.,
1995; Juo et al., 1995).
However, acidity is not only generated by oxidation reactions. Soluble mono calcium
phosphate is the principal P component in superphosphate fertilizers. In the immediate
vicinity of a superphosphate fertilizer granule, brushite (CaHPO4.2H2O) can
precipitate. A mole of burshite precipitated will generate one mole of protons
(Equation 2.7). With the dissolution of mono calcium phosphate, the formation of
dicalsium phosphate accompanied with phosphoric acid which dissociates into
phosphate and H+ ions.
Ca2+ + H2PO4- + 2H2O  CaHPO4.2H2O (brushite) + H+ (2.7)
Phosphate ions are strongly adsorbed to most soils, where H 2PO4-_induced leaching of
basic cations is unlikely to occur (Bolan et al., 2003).
2.1.2 Effect of soil acidity on nutrient availability, and plant growth
Acidification influences the physical, chemical and biological characteristics of soils
and therefore affects the transformation and biogeochemical cycling of both nutrients
and heavy metals in soils. pH affect the surface charge and subsequent adsorption of
solutes by variable charge soil components, such as layer silicate clays, organic
matter, and oxides of Fe and Al (Adriano, 2001). pH affects the following: sorption of
metal cations and anions in soil, metal speciation, complexation of metals with
organic matter, precipitation/dissolution reactions, redox reactions, mobility and
leaching, dispersion of colloids, and the bioavailability of nutrients and trace metals
(Bolan et al., 2003).
7
Surface charge controls a number of physical and chemical properties of soil. Soil
solution pH is one of the major factors controlling surface properties of variable
charge components (Sposito, 1984; Barrow, 1985; Sparks, 1986). pH affects the
surface charge through the supply of H+ for adsorption onto metal oxides and the
dissociation of the functional groups in the soil organic matter. An increase in pH
increases the net negative charge or cation exchange capacity, and a decrease in pH
increases the net positive charge or anion exchange capacity (Singh & Uehara, 1986).
The change in surface charge is the main reason for the effect of pH on anion and
cation adsorption. Lower pH values increase the Al concentration in the soil solution,
occupying a larger fraction of the cation exchange sites, and reducing base saturation
(Ritchie, 1989).
Soil pH is one of the important factors that influence the availability of nutrients
(Fölcher, 1975; Sumner, 1975; Sumner et al., 1991). Acidification also leads to
manganese (Mn) and iron (Fe) toxicities as well as deficiencies of phosphorous (P),
calcium (Ca), magnesium (Mg), potassium (K), sulphur (S), and molybdenum (Mo)
(Ceballos et al., 1995; Zeigler et al., 1995).
Acidity influences the nitrification of N, which is reduced at pH of 6 and undetectable
at pH lower than 4 (Alexander, 1977). Acidity also affect the solubility of soil P. Fe
and Al concentrations are increased in the soil solution with a decrease in pH,
increasing the adsorption/ precipitation of P. In variable charged soils, increasing
acidity decreases the cation exchange capacity reducing the soil’s ability to retain K,
Ca and Mg, resulting in more K, Ca and Mg in the soil solution which is prone to
leaching (Blue, 1986; Alibrahim, 1988).
Acidification affects the transformation of heavy metal ions through the altering of the
surface charge in variable charged soils, altering the speciation of metals and
influencing the reduction and oxidation reactions of metals. Thus, pH influences the
pollution hazard of several heavy metals (Løbersli et al., 1991). Metals are more
available in acid soil with the exception of a few metalloids like arsenate,
8
molybdenate, selenate, vanadate and some valence state of chromium that is more
available under alkaline conditions like chromate (Adriano, 1986).
Root growth restriction can occur in deep soil profiles due to subsoil acidity. Retarded
root growth is the primarily effect of aluminium (Al) toxicity (Foy, 1974; Pinkerton &
Simpson, 1986) and this affect the water uptake capability of roots which impact
negatively on yield due to limited water uptake by the underdeveloped root system
(Coventry et al., 1997).
2.2 Liming materials
The commonly used liming materials are dolomitic lime (≤ 15 mg kg-1 CaCO3 and ≥
70 mg kg-1 MgCO3), calsitic lime (≥ 70 mg kg-1 CaCO3 and ≤ 15 mg kg-1 MgCO3),
magnesite (containing ≥ 975 mg kg-1 MgCO3 and ≤ 25 mg kg-1 CaCO3) and BOF
slags from the steel industry containing Ca and Mg alumino-silicates (Barber, 1967).
The acid neutralizing value of liming materials is expressed in terms of calcium
carbonate equivalent (CCE), which is the acid neutralizing capacity of a liming
material expressed as a weight percentage of pure CaCO 3. The quantity of liming
material required to ameliorate soil acidity depend on the reactivity which determine
the neutralizing value of the liming material and the pH buffer capacity of the soil.
Venter (2004) mentions four approaches to ameliorate soil acidity which is used as
lime recommendation guidelines in South Africa. The pH-texture method uses the
clay content of soils and a specific target pH to determine the lime requirement. The
SMP-buffer method implies the measurement of the change in pH of a buffer solution
to which the test soil was added. Soils can be incubated with CaCO 3 to obtain a
calibration curve. Lime requirement can be determined in terms of CaCO 3. The
Eksteen method was developed by Eksteen (1969) for the winter rainfall region of
South Africa, and is expressed as follows:
X = [RH – (Ca + Mg)] / (R + 1)
9
X = lime requirement in t ha-1 for a 0.15 m soil layer, R = the ratio of exchangeable
Ca and Mg to exchangeable acidity to be achieved, H = exchangeable acidity
measured in the soil, and (Ca + Mg) = HCl extractable Ca and Mg. The acid
saturation method was developed for KwaZulu-Natal and soils are limed to an
acceptable level of exchangeable acidity.
Bornman (1985) developed the Resin Suspension Method (RSM) to predict the
reactivity of liming materials, compared to analytical grade CaCO 3. The SMP-buffer
method in combination with the RSM method can be used to calculate the quantity of
slag required to obtain the desired pH level in the soil.
Liming enhances the physical, chemical and biological characteristics of soil through
the direct amelioration of soil acidity and through indirect effects on the mobilization
of plant nutrients, immobilization of toxic heavy metals and the improvement of soil
structure (Haynes & Naidu, 1998).
The calcium in liming materials helps with the formation of soil aggregates,
improving soil structure. The lime induced improvement in aggregate stability is
manifested through the effect of liming on dispersion and flocculation of soil particles
(Chan & Heenan., 1998).
Acid soils are limed to overcome the chemical problems associated with acid soils
like the high concentration of acid ions (H+ and Al3+) and toxic elements (Mn2+), and
low concentrations of basic cations (Ca and Mg) and other nutrients like P and Mo.
The hydrolysis of the basic cations in lime produce OH- ions which neutralize the H+
ions, decreasing the activity and bioavailability of Al and Mn. Liming increase the
solubility of P and Mo, increasing their availability. Lime also provides basic nutrient
cations (Ca and Mg) and can be used to correct nutrient imbalances. The elevation in
pH due to the addition of lime results in the precipitation of exchangeable Al and
increase the negative charge or CEC (Bolan et al., 2003).
Other liming materials include different slags derived from the processing of steel.
10
The slag’s composition differs due to different processes used in the processing of
steel, therefore the neutralising reaction is not precisely known. It is not a pure
carbonate but also contains Ca-silicate and could react (together with other
constituents) in the following manner:
SiO5Ca3 + 6H+  3Ca2+ + SiO2 + 3H2O
(2.8)
SiO4Ca2 + 4H+  2Ca2+ + SiO2 + 2H2O
(2.9)
2CaO.Fe2O3 + 16H+  2Ca2+ + 4Fe3+ + 8H2O
(2.10)
FeO.MnO + 4H+  Fe2+ + Mn2+ + 2H2O
(2.11)
CaO + H2O  Ca(OH)2
(2.12)
Ca(OH)2 + 2H+  Ca2+ + 2H2O
(2.13)
MgO + 2H+  Mg2+ + H2O
(2.14)
(After Vanacker, 1999).
Several types of slags are used as agricultural liming materials, such as blast furnace
slag, open-hearth slag, and basic slag. These are all the by-products of steel
processing plants and their composition depend on the raw material (ore), the specific
process, and the product of the plant. Due to this, Ca and Mg aluminosilicates, Fe,
Mn, phosphates, and other elements such as Cr, Ni, Co, Cd, Cu, Zn, Mo, and Ba may
occur in differing quantities. Some of the elements can reach concentrations of several
parts per thousand, which are then added to the soil.
Two main forms of slags are produced in and distributed from the heart of the Loskop
Dam catchment area near Witbank and Middelburg (part of the Eastern Highveld).
This is also the area in which soil acidification is reaching critical proportions in parts,
and in which many soils are very sensitive to acidification (Fey & Dodds, 1998). The
waste product, if finely ground, has a high Calcium Carbonate Equivalent (CCE) and
is relatively cheap in comparison to other mined liming materials. Due to high
transport costs and the proximity of the slags, large quantities are used in the area. As
mentioned earlier, a source of concern is the fact that slags contain varying amounts
of trace elements and heavy metals and that although most of the elements are not
plant available at the pH levels in the soil after liming, the build-up of these elements
11
could be detrimental in the long-term. Slag use as a liming material serves as a waste
management strategy for the Iron and Steel industry. In 2004 the total amount of lime
used in RSA was 1.39 Mt tons, and 50.5 % was slag from the Iron and Steel
industires. Table 2.1 gives an indication of the average composition of the slags.
TABLE 2.1. Average composition of slags before refining into agricultural lime
(Source: Columbus Steel and Highveld Steel).
Chemical
Columbus Steel
Highveld Steel
Compound
%
%
Al2O3
2.3
1.3
CaO
49.5
55.4
Cr2O3
3.0
-
FeO
0.6
16.7
MgO
11.4
6.5
MnO
0.9
1.1
P
-
0.4
S
-
0.4
SiO2
30.6
17.9
TiO2
0.7
-
V2O5
-
1.6
After Van Der Waals, 2001.
2.3 Vanadium in the soil
2.3.1 Origin and concentration in soils
According to Nriagu (1998), the average concentration of V in the earth’s crust is
100-150 mg kg-1, which is almost the same as Ni, Zn and Pb. V is more abundant in
mafic rocks (high in Fe and Mg) than in silicic rocks. Concentrations in gabbro and
norite range between 200-300 mg kg-1. The V concentration in silicic rocks like
granite is about 80 mg kg-1. In sandstone the V concentration is very low and has an
average concentration of 12 mg kg-1. V concentration in iron ores is in the range of
600 - 4100 mg kg-1 and in rock phosphates 10 - 1000 mg kg-1. The production of
12
super phosphate from V containing rock phosphate can further concentrate V to 50 2000 mg kg-1 (Evans & Landergen (1978) and ATSDR (1992).
The natural concentration of total V concentration in soil varies between 10 mg kg -1
and 220 mg kg-1 (Kabata-Pendias & Pendias, 1993). Panichev et al. (2006) however
reported concentrations as high as 400 mg kg-1. This variation is, however, a function
of the parent material from which the soils are derived and other possible sources of
contamination, including liming with materials having high levels of V.
The largest contributors of anthropogenic V are emission from the combustion of
fossil fuels by power plants (USEPA, 1987). The average concentration of V in hard
coals is 19 mg kg-1 (Yudovich, 1972). In 1972 oil combustion contributed 94.1% to
the emissions of V into the atmosphere, coal contributed 4.5 % and metallurgical and
other processes 1.2% (Van Zinderen Bakker & Jaworski, 1980). Other contributors
are the emissions from petroleum refineries and metal plants (Kabata-Pendias, 1993)
and in this study slag application to ameliorate soil acidity.
2.3.2 Vanadium dynamics in soil
There are about 80 different minerals containing V (Fleischer, 1987; Clark, 1990).
These minerals are placed in four groups according to their crystal chemical structure.
The groups are; sulfides, sulfosalts (derived from oxidized sulphide ores) of Pb,
copper (Cu), Zn and Mn, silicates and oxides. There exist a group of minerals which
contains pentavalent V in their structure occurring as isolated VO 4 tetrahedra which is
the result of the oxidation of base metal sulfides (Evans & Landergen, 1978). A
number of minerals in this group are isostructural with phosphates and vanadates like
vanadinite (Pb5(VO4)3Cl).
2.3.3 Speciation of vanadium
The speciation of an element is an important aspect to understand and determine the
toxicity, bio-availability and its mobility in the environment (Bendito & Rubio, 1999;
Ebdon et al., 2001; Sturgeon, 2000). The speciation of elements is influenced by pH
13
and its oxidation state. V exist in the +3, +4 and +5 oxidations states in nature and
below pH 8, the +4 and +5 states dominate under oxidising conditions (Peterson and
Girling, 1981).
Mononuclear vanadate oxyanions (H2VO4- and HVO42-) with a structural analogy like
phosphate, are the dominant species of V(V) in dilute solutions (Wehrli & Stumm,
1989; Wanty & Goldhaber, 1992). In more concentrated solution (> 100 µmol L-1) V
might form polynuclear species (Cruywagen & Heyns, 1991) like decavanadate
(HxV10O28x-6) and metavanadate ((VO3)xx-).
Pyrovanadates (V2O74-) replace
mononuclear oxyanions as the primary V(V) species at V concentrations higher than
0.1 mol L-1 (Baes & Mesmer, 1976). V (+4) and V (+5) are bound to oxygen to form
oxyanions which tend to complex with ligands like phosphorous (P) and sulphur (S)
(WHO, 1988).
The metavanadate (VO3-) anion can be reduced to the vanadyl (VO2+) cation, which
may be an important form of V in soil according to Berrow et al. (1978). VO2+ is
immobilized as humic acid complexes while the anions like orthovanadate (VO 43-)
and metavanadate. (VO3-) are mobile in soils and potentially more toxic to soil micro
biota (Goodman & Cheshire, 1975; Bloomfield, 1981). Little work has been done to
establish methods to quantitatively separate the different species, which have different
nutritional and toxicity properties (Mandiwana et al. 2005).
Poledniok and Buhl (2002) used the Tessier et al. (1979) extraction analysis
procedure for heavy metals (Cd, Co, Cu, Ni, Pb, Zn, Fe and Mn) to determine five
fractions of V in two soils. The first is the exchangeable fraction, which contains
metals adsorbed on the solid surface. The second fraction is metals bounded or
precipitated by carbonates. The third fraction is metals adsorbed on surfaces of
precipitating Fe, Al and Mn oxides and this is under oxidised conditions. The fourth
fraction is metals adsorbed to the surface or bonded to organic matter. The fifth
fraction comprises of metals built into the crystal lattice of primary and secondary
minerals, which are not plant available. These authors reported that the exchangeable
V was the same for both soils but not the total V concentration, which indicate that
14
plant available V does not depend on total V concentration in the soil. The different
concentrations of Fe, Mn and organic material determined the fraction of V held in
that specific fraction. In both soils no V and carbonate compounds were found and
this may indicate that V does not form compounds with carbonates.
2.3.4 Vanadium sorption
The aqueous chemistry of V is controlled by its redox state and sorption onto iron
oxide and clay minerals (Trefry & Metz, 1989). Sorption onto minerals like goethite
and hematite controls V concentration in groundwater (Breit & Goldhaber, 1989;
Wanty & Goldhaber, 1992).
Keeney-Kennicutt & Morse (1984) suggested that VO 2+ formed inner-sphere surface
complexation at low pH. At higher pH, inner-sphere surface complexation was
suggested through the adsorption of HVO42-. The vanadate ion acts like phosphate and
can adsorb through ligand exchange (Sigg & Stumm, 1980).
The V sorption capacity of soils directly influences the mobility and plant availability
of V. If the structural analogy between vanadate (H 2VO4-) and phosphate (H2PO4-)
ions exist then it can be hypothesised that vanadate reacts like phosphate in the soil
and the soil properties that influence P fixation will influence the V sorption and
availability (Rehder, 1999; Crans, 1994). Common factors influencing phosphate
availability are pH, AEC, Fe, Al, and Mn (oxy) hydroxides, the percentage carbon (C)
and texture (Brady & Weil, 2002).
Wang and Lui (1999) determined the sorption capacity of two soils by shaking the
soils with different concentrations of V. Different concentrations of NH 4VO3 was
added to two soils. The soil with the highest concentration of total Mn and Fe sorbed
more V. pH also influenced the sorption of V. As the pH increased, the sorption
capacity decreased and the plant availability of V increased.
15
2.3.5 Vanadium mobility
Vanadium (+5) is more soluble than V (+4) rendering it potentially more mobile and
plant available and V (3+) is not found in oxidized conditions like oxidised soils (Van
Zinderen Bakker & Jaworski, 1980). In neutral or alkaline soils V is more mobile
relative to other heavy metals (Brooks, 1972). Martin and Kaplan (1996) found that V
is not very mobile in the soil that they tested. In this study a V salt (VOSO 4·H2O) was
incorporated into the top 7.5 cm of the soil and samples were taken at various depths
for 30 months. There was little increase in V concentration at a soil depth below 15
cm due to the presence of gibbsite and Fe oxide. One should mention that the V
loading rate (5.6 kg ha-1) was very low. It is not clear from the literature whether V, in
the presence of Fe oxides, is removed from the aqueous phase as a precipitate, such as
Fe(VO3)2, or sorbed onto the surfaces of Fe oxides (Rai & Zachara, 1984).
2.4 Vanadium in the plant
2.4.1 Concentration and toxicity
The question whether V is an essential nutrient for higher plants have not been
answered (Morrell et al., 1986; Kabata-Pendias & Pendias, 1992). According to
Kabata-Pendias & Pendias (1993), V positively influence chlorophyll synthesis,
potassium (K) consumption and nitrogen assimilation at concentrations lower than 2
ppm, while higher concentrations of V can be toxic and may cause sclerosis and limit
plant growth. The concentration at which an element becomes toxic depends on the
soil type and plant species. Plant species have different toxicity levels because some
plants have developed mechanism to tolerate high heavy metals concentrations. An
example of a V accumulator is Astralgus confertifloris with V concentration up to 144
mg kg-1, growing in vanadium rich, sandstone derived soils in Utah, USA (Cannon,
1963).
Various symptoms were observed where plants were exposed to high concentrations
of V. Somasundaram et al. (1994) found that V caused inhibition of chlorophyll
biosynthesis, soluble protein and net photosynthesis of seedlings of rice (Oryza sativa
L.). V can cause a decrease in Ca absorption by sorghum root tips (Wilkinson &
16
Duncan, 1993), decrease phosphate uptake by maize roots (Sklenar et al., 1994) and
inhibit the growth of onion roots (Hidalgo et al., 1988).
Nutrient solution studies are used to determine the essentiality or the toxicity of
certain elements. It is also used to determine the factors that influence uptake like pH
and other cations or anions in solution. Trails were conducted to determine the
toxicity levels of V in a solution. V concentrations of 10 to 20 mg/l in nutrient
solution were reported to be harmful to plants (Arnon & Wessel, 1953; Cannon,
1963). Welch (1973) found that CaSO4 in solution enhanced the V uptake by roots
and speculated that the V taken up by the plant cells accumulates within the interior of
the cell and Ca is required for the retention of the absorbed V by the cells.
Wang and Liu (1999) conducted a greenhouse pot trial, to determine the V toxicity of
soya bean seedlings in two soils. They applied NH 4VO3 to the soil through seven V
loading rates (0, 5, 10, 15 30, 50 and 75 mg kg-1). They found that the toxicity levels
in soil differ as a function of texture and Fe and Mn content. Soya bean seedlings in a
sandy loam (20 % clay) soil with Fe concentrations of 22 000 mg kg-1 showed
reduced growth at V loading rate of 30 mg kg-1 soil. In a sandy clay loam (32 % clay)
soil with a Fe content of 46 000 mg kg-1 a V loading of 75 mg kg-1 did not affect the
growth of soya bean plants negatively. This was due to a higher V fixation in the
sandy clay loam. Neither plant tissue concentrations nor the plant available fraction of
V were mentioned where toxicity occurred.
Kaplan et al. (1990) used VOSO4 to determine V toxicity in collards (Brassica
oleracea) in a sandy (6 % clay) and in a loamy sand (14 % clay) soil. The V loading
rates were 0, 20, 40, 60, 80, and 100 mg kg-1. Collards in the sandy soil with a V
loading rate of 80 mg kg-1 showed a significant reduction in above ground biomass.
This was at a higher V loading rate than the soya seedlings. This shows that different
plant species have different threshold concentrations for V and that the mineralogy of
the soil also influences V dynamics in the soil. The total V concentration in the plant
at this loading rate was 10.3 mg kg-1. Collards growing in the loamy sand with a
higher Fe concentration, organic material and cation exchange capacity did not show
17
a reduction in above ground biomass even at a V loading rate of 100 mg kg -1. The
total V concentration in these collard plants was 7.34 mg kg-1.
2.4.2 Vanadium uptake and translocation in plants
Higher plants do not accumulate V and concentration is the highest in the roots where
it is thought to precipitate out as calcium vanadate (Cannon, 1963; Hempill, 1972;
Lepp, 1977; Wallace et al., 1977. According to Peterson and Girling (1981) this might
be a mechanism that the plant developed to tolerate high V concentrations. Kaplan et
al., (1989) confirmed that V accumulates in the roots which place V in a group of
metals that is immobile in plants. When vanadate is taken up by the plant and
translocated to the cells, it is easily reduced to vanadyl (VO 2+) (Crans & Tracey,
1998).
Martin and Kaplan (1996) planted bush beans (Phaseolus vulgaris L.) to determine
the plant availability of V. Vanadyl sulfate (VOSO 4) was used as the V source and the
loading rate was 5.6 kg ha-1, which is a fairly low loading rate (± 2.15 mg kg-1) if the
bulk density is taken as 1300 kg m3 and the soil depth of 20 cm. The roots contained
the highest concentration of V, which correlate with other literature. While the
concentration in the above ground parts were appreciably low. V plant availability
was drastically reduced as time progressed and the concentration was negligible after
18 months.
Welch (1973) found that V is passively absorbed by barley roots. The uptake was
highly pH dependant and had a linear relationship with the concentration of V. The
highest uptake was at pH 4, constant between pH 5-8, with VO3- the dominant specie,
and decreased to very low at pH 10.
2.4.3 Vanadium interaction with other nutrients
According to Rehder (1999) and Crans (1994), there is a structural analogy between
vanadate (H2VO4-) and phosphate (H2PO4-) ions. The accumulation of V by plants
may reduce the uptake of P, which plays an important physiological role. Welch
18
(1973) found that the anion that produced the greatest inhibition of V uptake was
H2PO42-, but this inhibition was only 27% less than the control. Little inhibition of V
uptake by roots occurred with anions like MoO 43-, BO3-, Cl-, SeO42-, CrO42- and NO3-.
This shows that V uptake is not greatly affected by other anions.
There is also a possibility that vanadyl, being a cation, can interact with calcium
(Ca2+) and magnesium (Mg2+) (Olness et al. 2001). This might indicate that before
there is V toxicity, vanadium might influence plant nutrition to some extent.
2.5 Guidelines for vanadium and the use of slag
The Russians stipulate that 150 mg kg-1 is the maximum total V concentration
allowed in soil (Ghost, 1985), but they do not mention the method of analysis.
Canadian soil quality guideline for the protection of environmental and human health
stipulates an upper limit of a total V concentration of 130 mg kg-1 in soil used for
agriculture (Canadian Council. 1997). In South Africa no guidelines exist considering
V concentration in ameliorants or in the environment.
Slag falls in the same category as fertilizers under Act No 36 of 1947. Guidelines
exist for maximum allowable concentrations for some heavy metals but not V.
Criteria used to evaluate slag are the fineness of the slag, which determines the
reactivity and the Mg and Ca content, which will class it into calcitic or dolemitic
“lime”.
One of the remedies for heavy metal pollution is to apply slag to increase the pH and
this will reduce the availability of most heavy metals (Logan, 1992; Hooda et al,
1997; Chaney et al, 2001). The problem with V is that it becomes more mobile and
plant available as the pH increases (Adriano, 1986).
2.6 Extraction methods and analyses
Mandiwana et al. (2005) stated that there is little work done to quantitatively separate
between different species of V so they experimented with a few methods. Mandiwana
& Panichev (2004) determined 5 fractions of V using a sequential fractionation
19
procedure (and V determination with atomic absorption spectrometer). Firstly the
water soluble fraction of V (V) was determined by shaking 0.25 g of soil with 25 ml
de-ionised water for 24 h. The second was the determination of V (V) extracted with
carbonated water. The soil was also shaken with deionised water before bubbling CO2
through the suspension. The third method was to determine the V (V) extracted with
0.1 M sodium bicarbonate (Na2CO3). The precipitates of this method where ashed and
then digested with HF and H2SO4 to determine V (IV). The total amount of V was
determined by digesting 0.25 g of soil with HF and HClO 4. Mandiwana & Panichev
(2004) found V (V) is easily leached from the soil by carbonated water. They also
found that 50.8% of V present in polluted soils near a V mine, containing 7160 mg V
kg-1 soil, was V (V). In another speciation analysis Mandiwana et al. (2005) used
PO43- instead of Na2CO3 to determine the V (V) fraction in the soil. They concluded
that (NH4)2HPO4 and Na3PO4 are excellent leaching agents, like Na2CO3, and that the
use of P fertilizers will potentially increase the mobility of V (V) in the soil.
20
CHAPTER 3: VADADIUM TOXICITY THRESHOLD
LEVELS FOR MAIZE: A POT TRIAL
3.1 Introduction
Greenhouse pot trials are commonly employed to study the uptake of elements and
their effects on plant growth at various concentrations. The plant availability of V
from slag was an important aspect that needed assessment. While field trials and field
monitoring can give an indication of actual phyto-availability, too many variables
exist that can complicate results and conclusion obtained. At field scale, for example,
it is difficult to apply slag uniformly over large areas. Factors that can further
exacerbate spatial variability are application history of fertilizer or liming material
that could have contained V and natural variability of soil properties that can
influence V plant availability (mineralogy, pH, texture etc.). Furthermore, especially
under dryland conditions, any stress the crops experience can potentially alter the
uptake of V and the uptake of essential elements. At greenhouse level many of these
variables can be controlled and V uptake by crops can be investigated under more
controlled conditions.
Subjecting crops to incrementally higher application levels of V, toxicity threshold
levels for both the plant and soil type can be established. By doing this, lower limits
for reduced growth and upper toxicity limits can be established. This, in turn, can be
compared to typical V loading rates when BOF slag is used and thus enabling some
predictive capability on the long term use of V containing slag in agriculture based on
current application rates. At high loading rates the toxic concentration levels, as well
as the visual symptoms of toxicity can be established. An analogue exists between the
chemistry of orthovanadate and orthophosphate (Rheder, 1999; Crans, 1994). It is
therefore hypothesised that extractants used to assess plant available P can also be
used to determine plant available V.
21
3.2 Aims
The aims of this chapter were to:
1) Establish V threshold concentrations where the growth of maize (Zea mays L.) is
visibly influence;
2) Evaluate the ability of two commonly used soil extractants in agriculture (Bray 1
and Ammonium acetate) to predict plant available V.
3.3 Materials and methods
3.3.1 Soils used in this study
Two contrasting soils where chosen according to specific soil characteristics. The
Nooit (Nooit) soil was a Clovelly soil from the Nooit experimental farm near Ermelo,
with a textural class of a loamy sand (12% clay). Quartz (57%) dominated the clay
fraction of the Nooit soil and also contained of 20% Mica, 20% kaolinite, and 5%
Goethite. The main soil selection criteria was P adsorption capacity, would be a
function of above mentioned mineralogy, and the abundance of these minerals would
also be reflected by the total Al, Fe and Mn content of the soil. The Nooit soil has
lower concentrations of total Al, Fe and Mn content (table 3.1) compared to the PPS
soil, which lead to a low P adsorption capacity and because of the expected analogy
between P and V, a low V adsorption capacity.
TABLE 3.1. The total concentrations of some of the elements in the two soils
determined by a XRF scan.
Al
mg kg
Si
-1
mg kg
Fe
-1
mg kg
Mn
-1
mg kg-1
Nooit
14000
22670
16590
140
PPS
29100
17060
40980
270
The PPS soil was a topsoil sample collected from a Hutton soil form. This soil was a
sandy clay loam (25% clay). The clay fraction of PPS is dominated by kaolinite
(65%) and also contains 25% mica, 6% hematite and 4% smectite. PPS was collected
22
at the Hatfield experimental farm. The PPS soil also had a higher total Fe and Mn
concentrations compared to the Nooit soil and expected higher V sorption capacity.
3.3.2 Vanadium content of mono ammonium phosphate
In this study mono-ammonium phosphate (MAP) was used as the P fertiliser. A XRF
scan revealed that MAP contained 0.005 % V (50 mg kg-1) (Fig 3.1). This is a relative
low value and the V loading through the use of MAP as P fertilizer was expected to
be negligible.
3.3.3 Liming, fertilization and vanadium loading
The soils used in this study were limed to a target pH(H 2O) of 6.5. The amount of
lime required to reach this pH was estimated by establishing liming requirements for
each soil.
Air-dried Nooit and PPS soil was passed through a 5 mm sieve and was weighed of in
4 kg pots. Analytical grade CaCO3 was used as the liming material in the V toxicity
trial. The Nooit soil received 4.00 g CaCO3 per pot which was equivalent to 2.60 tons
ha-1 and the PPS soil 6.40 g CaCO 3 per pot which was equivalent to 4.14 tons ha-1.
The bulk density of the soil was assumed to be 1300 kg m-3 and the incorporation
depth taken as 0.2 m.
The soils were placed in a concrete mixer and the CaCO 3 was added and mixed for 5
minutes. The soils where subjected to wetting and drying and after six weeks the
target pH was reached. After the incubation to correct the pH the soils were air dried.
V was added to two soils, Nooit and PPS, to determine the V concentration in the soil
where plant growth is affected negatively. The soils were treated with various levels
of V. Ammonium metavanadate (NH4VO3), a V salt, was used for the different V
loading rates. The different V loading rates were 0, 15, 30, 60, 100 and 250 mg kg-1
V. The respective treatments were separately mixed for the second time with a
concrete mixer and transferred back to the pots. The loading range proved to be
critical V loading rates that affect the growth of certain plant species as found in the
literature (Wang & Liu, 1999; Kaplan et al., 1990).
23
The soils were also fertilized with mono ammonium phosphate (MAP), ammonium
sulphate (NH4)2SO4 and potassium chloride (KCl). The MAP was mixed into the soil
together with the NH4VO3. The (NH4)2SO4 and KCl were dissolved in water and then
added to the pots after planting. The lower V loading rates received less N because
NH4VO3 was the V source, (NH4)2SO4 was used to ensure all the treatments received
the same amount of N. Each pot received the equivalent of; 180 kg ha -1 N, 70 kg ha 1
P and 140 kg ha
-1
K at planting. Each treatment was replicated four times, which
gave a total of 48 pots for the V toxicity trial.
3.3.4 Planting and harvesting the maize in the vanadium toxicity trail
Maize was used as the test crop. Five maize plants were planted in each pot and were
reduced to two plants per pot after emergence. After 6 weeks of growth the maize
plants were pot bound and whole maize plants (above ground matter) were harvested.
After harvest the plants were washed with deionised water, air dried at 60 oC for two
days and weighed. The maize plants where milled and passed through a sieve.
Representative air-dried soil samples were taken from each pot and passed through a
2 mm sieve.
Therefore to obtain a better picture of V uptake by maize, root analyses were
performed. However, root analysis is often fraud with difficulties. Soil is often
embedded in the roots and it is difficult to remove all the soil and completely
eliminate soil contamination. Furthermore, precipitates of calcium, ferric and
aluminium vanadate are expected to be sparingly soluble in water. Any precipitates on
outer surfaces of roots would be difficult to rinse off with deionised water.
The ideal would have been to sample the roots of all the treatments, this was a
shortcoming of this study. Only the roots of the control treatment and the 250 V
loading rate of both soils were sampled and analysed. The highest application level
for the Nooit soil resulted in underdeveloped roots and in order to obtain enough
material to perform analysis, a composite sample were prepared from the different
replicates. This means that no statistical analyses were possible for the root data.
24
3.3.5 Chemical analysis
The soil pH was determined by weighing 20 g of soil in glass beakers and 50 ml
deionised water was added (1:2.5 basis). The suspension was stirred and left for 1
hour and the pH reading was taken with a glass electrode pH meter. The soil pH was
taken to ensure that all the pots where at the same pH level.
Bray 1 (0.03 M NH4F + 0.025 M HCl) was used as an extractant for extractable PO4-3
and presumably also VO3-1 and/or VO43-. Bray extractable phosphate has shown to be
a good extractant to assess plant available P. It was therefore used as an extractant for
V because of the analogy between orthovanadate and orthoposphate chemistry
(Rheder, 1999; Crans, 1994). Eight grams of soil was weighed into Schott bottles and
60 ml of Bray 1 was added. The suspension was shaken for 60 seconds by hand and a
drop of super flock was added. The suspension was filtrated through a 2 µm Whatman
filter paper and analysed using an axially viewed Inductively Coupled Plasma Atomic
Emission Spectrometer (ICP – AES).
Ammonium acetate was used to extract K +, Ca2+, Mg2+ Na+ and presumably VO2+.
Ammonium acetate extractable cations represent the soluble and exchangeable
fractions of the elements in the soil and are generally considered to represent the plant
available forms. Five grams of soil was weighed off in Schott bottles and 50 ml of
ammonium acetate was added. The suspension was shaken for 30 min on a reciprocal
shaker. The suspension was filtrated through a 2 µm Whatman filter paper and was
analysed using an axially viewed ICP – AES.
The soil and plant samples were digested with a HClO 4:2HNO3 digestion (ALASA,
1998) to determine the total concentrations of certain elements. For the
HClO4:2HNO3 0.5 g of the plant and soil samples was weighed of in 100 ml glass
tubes and then 5 ml of the HClO 4:2HNO3 solution was added. The samples were
digested at 230 oC on the digestion block and then diluted to 100 ml with deionised
water. Afterwards, the samples were analysed for V using an axially viewed ICPAES.
25
The V content of the soils was determined by means of three methods. The first was a
XRF scan which is a qualitative- semi-quantitative powder scan. The second was a
HNO3 digestion (EPA 3050). The V in solution was determined by means of
inductively coupled plasma mass spectrometry (ICP-MS). The third method was a
HClO4:2HNO3 digestion (ALASA, 1998). The V concentration was determined by
means of ICP-MS.
3.3.6 Statistical analysis
The statistical program SAS was used for statistical analyses of the data generated.
The Tukey test was used to determine treatment effects at a probability level of ά =
0.05.
3.4 Results and discussion
After six weeks of growth the maize plants were all pot bounded. It was therefore
reasonable to expect that good exploitation of the soil volume in the pots occurred. It
is important to evaluate V uptake at various V loading rates. This will enable the
establishment of lower limits where reduced growth will occur and upper levels where
toxicity occur. With this established, predictions can be made on cumulative long
term loading of V through slag application and management strategies can be
developed to ensure long term slag use in agriculture.
3.4.1 Background concentrations of vanadium in soils
The background V concentration in soils over the world varies greatly, from 0-400 mg
kg-1. The V concentration in the control treatments of the two soils used in the pot
trials was 26.4 (+/- 1.07) mg kg-1 for the sandstone derived Nooit soil and 56.0 (+/10.5) mg kg-1 for the dolerite derived PPS soil, determined by HClO4:2HNO3
digestion (ALASA, 1998). These background values were relatively low compared to
concentrations found in the literature (Kabata-Pendias & Pendias, 1993; Panichev et
al. 2006). The PPS 250 and Nooit 250 are the V concentrations in the soils after the V
loading of 250 mg kg-1 in the toxicity trial. The XRF scan showed the highest V
26
concentration of the three methods for total V concentration, and the HClO 4:2HNO3
the lowest (Fig 3.1).
350
300
V mg kg-1
250
200
XRF
HNO3
150
HCl04:2HNO3
100
50
0
Nooit
PPS
NOOIT 250
PPS 250
MAP
FIGURE 3.1 V concentrations in soils unamended with V compared to soils that
received 250 mg kg-1 NH4VO3.
3.4.2 Effect of vanadium loading rates on dry matter yield
Fig 3.2 and Fig 3.3 illustrate the effect of high V concentrations on plant growth. In
terms of loading rates, the Nooit soil showed reduced growth at the 100 mg kg-1 V
(100 level) and stunted growth at the 250 mg kg-1 V loading rate (250 level) (Fig 3.2).
According to Tukey’s test, dry matter yield obtained from the 250 level was
significantly lower than the other treatments. The 100 level was also significantly
lower than the rest of the treatments except for the 60 mg kg-1 V loading rate. For the
Nooit soil V loading rates greater than 60 mg kg-1 reduced biomass production and
loading rates greater than 100 mg kg-1 was seemingly detrimental for plant growth.
According to Kaplan et al. (1990) collards (Brassica oleracea) showed reduced
biomass production at a V loading rate of 80 mg kg-1 in a 6 % clay soil. However
Wang & Lui (1999) found that soya beans (Glysine max) showed reduced growth in a
20 % clay soil at a V loading rate of 30 mg kg-1. This might be due to the fact that
soya beans are more sensitive to V compared to other plants.
27
250
100
(mg kg-1)
V loading rates
60
30
15
0
FIGURE 3.2 The effect of increasing V concentrations in on the growth of maize
plants (Nooit soil).
28
FIGURE 3.3 Aboveground biomass of maize plants (Dry mass) as affected by
increased V loading rate.
Dry mass production did decrease for the PPS soil from a mean of 19.5 (+/- 3.49)
grams per pot for the 100 level to about 15.4 (+/- 1.82) grams per pot for the 250 level
but according to Tukey’s test, this was not statistically significant at α = 0.05.
There are a few reasons that can explain why the maize plants growing in the PPS soil
did not show reduced growth or toxicity symptoms. Maize dry matter where not
affected by the increasing V rates. The PPS was a dolerite derived soil with higher
clay content than the Nooit soil. This suggests that the PPS most likely had a greater
surface area available where V sorption could be facilitated compared to sandstone
derived Nooit soil. The PPS soil also have a higher total concentration of Fe (40 980
mg kg-1) compared to the Nooit soil (16 590 mg kg-1). Ferric sesqui-oxides, for
example, the ferric (oxy) hydroxides, goethite (α –FeOOH) and the ferric oxide
hematite (α –Fe2O3) are known to almost irreversibly sorb phosphate and according to
Sigg & Stumm (1980), vanadate sorption characterstics resembles that of phosphate.
The PPS soil had 6% hematite while the Nooit soil only had 2% goethite. Vanadate is
specifically sorbed and forms bi-nuclear bidentate innersphere complexes on goethite
(Peacock & Sherman, 2006). It was therefore expected that the PPS soil would have a
greater ability to attenuate V.
29
3.4.3 Toxicity at total vanadium concentration in the soil and the plant
Table 3.2 summarise the values obtained through various V extractants (total V
concentration and extractable V concentration) compared to V content in the plant
material as well as dry mass production.
The average total V concentration at the 250 V loading rate in the Nooit soil was 150
(+/- 18.2) mg kg-1 and 14.8 (+/-1.24) mg kg-1 in the above ground plant tissue (dry
mass) which is significantly different from the lower values. The V concentration in
the lower loading rates is close to the detection limit and very low. This supports the
fact that V is not tanslocated to the above ground parts of maize plants.
TABLE 3.2 A summary of the results from the V toxicity trail (total V concentrations,
Bray 1 and ammonium acetate extractable).
Soil
Nooit
PPS





V
loading
Rate
0
15
30
60
100
250
0
15
30
60
100
250
Total V soil
26.4 (c)
29.7 (c)
37.4 (bc)
48.0 (bc)
73.3 (b)
150 (a)
56.0 (c)
62.1 (c)
64.8 (c)
79.4 (c)
112 (b)
235 (a)
Total V Plant
material (above
ground mass)
V (Bray 1)
(mg kg-1)
0.18 (b)
0.12 (e)
0.14 (b)
2.26 (de)
0.00 (b)
5.83 (d)
0.00 (b)
12.4 (c)
0.00 (b)
23.5 (b)
14.8 (c)
77.6 (a)
0.03 (b)
0.05 (ab)
0.07 (ab)
0.14 (ab)
0.10 (ab)
0.50 (a)
0.07 (c)
0.43 (c)
1.79 (c)
2.45 (c)
6.36 (b)
30.2 (a)
V (OAc)
0.00 (d)
0.05 (d)
0.17 (d)
0.64 (c)
1.68 (b)
10.2 (a)
0.01 (c)
0.00 (c)
0.02 (c)
0.07 (c)
0.21 (b)
1.69 (a)
Total V soil – Pseudo total V concentration in the soil determined by the
HClO4:2HNO3 digestion.
Total V Plant (above ground) – V content in the plant determined by the
HClO4:2HNO3 digestion.
V (Bray 1) – Bray 1 extractable V determined by the Bray 1 extraction method
V (AA) – Ammonium acetate extractable V determined by the ammonium
acetate extraction method.
(a) – Tukey’s test for significant difference
The above ground plant tissue can be used to test for V toxicity. 14.8 mg kg-1 seems
to be a threshold value for V in the above ground plant material of maize (Table 3.2).
30
14.8 mg kg-1 might be a threshold value for maize but will differ for different plant
species. According to Kaplan et al. (1990) collards (Brassica oleracea) showed
reduced biomass production at a V loading rate of 80 mg kg-1 and had a total V
concentration in the plant of 10.32 mg kg-1 and 11.80 mg kg-1 at a V loading rate of
100 mg kg-1. Currently no basic guidelines on critical threshold levels exist in South
Africa. This value can be used as the upper limit for V in above ground material for
maize.
The pseudo total V concentration at the 250 V loading rate for the PPS soil was 235
(+/- 22.0) mg kg-1 and 0.50 (+/- 0.30) mg kg-1 in the above ground plant tissue (dry
mass). The V levels in the above ground material for the 250 level in the PPS soil was
far less than the V levels in the Nooit soil. This again indicates the low plant
availability of V in the PPS soil which has a significant V fixing ability.
3.4.4 Toxicity symptoms
The V toxicity symptoms were stunted growth with sclerosis on the older leaves (Fig
3.4) and a small, underdeveloped root system. This correlates with the findings of
Wang and Lui (1999), who described V toxicity in soya beans with stunted growth,
yellow and withered foliage and small roots showing sign of senesces.
31
FIGURE 3.4 A maize plant showing stunted growth at the highest V loading rate (250
mg kg-1) in the Nooit soil after six weeks.
3.4.5 The extractability of vanadium; comparing Bray 1 and ammonium acetate
The Bray 1 extractable V for the Nooit soil at the 0 level was 0.12 (+/- 0.02) and
increased to 23.5 (+/- 0.98) mg kg-1 at the 100 level and 77.6 (+/- 4.00) mg kg-1 at the
250 level. The Bray 1 extractable V for the PPS soil at the 0 V loading rate was 0.07
(+/- 0.04) mg/kg and increased to 30.2 (+/- 2.77) mg kg-1 at the 250 V loading rate.
For the highest V application level, the Bray 1 extractable fraction represented 12.1 %
of the total V. This shows that the Bray 1 extractable V increased as the V
concentration in the soil increased (Fig 3.5).
32
FIGURE 3.5 The Bray 1 extractable V and P as influenced by the increasing V
loading rates.
It was suspected that the increased V concentration would have an effect on the Bray
1 extractable P. However, the Tukey test only showed a significant increase in the P
extractability for the Nooit soil as the V loading rate increased. One of the reasons for
the increase in V extractability might be due to exchange reaction between the two
anions: orthovanadate (H2VO4-) and orthophosphate (H2PO4-) in the Nooit soil. The
Bray 1 extractable P in the Nooit soil at the 0 V level was 5.93 (+/- 3.30) mg kg-1 and
increased to 16.8 (+/- 3.33) mg kg-1 for the 250 V level, almost three times higher
than the control. The Bray 1 extractable P for the PPS soil at the 0 V loading rate was
13.1 (+/- 1.26) mg kg-1 increased slightly to 17.3 (+/- 4.98) mg kg-1 at the 250 V level.
This represented an increase of 32% but was not statistically significant. The lesser
effect on Bray 1 extractability P in the case of the PPS could have been the result of
less competition and enough surface area and sorption sites for both vanadate and
phosphate.
The sandier Nooit soil probably had less sorption sites for vanadate and phosphate.
This could have resulted in more competitive sorption between V and P. The
ammonium acetate extractable V in the Nooit soil at the 0 level was 0 (+/- 0.02) mg
kg-1 and 10.2 (+/- 0.41) mg kg-1 at the 250 level.
33
The ammonium acetate extractable V at the 0 V loading rate for the PPS soil was 0.01
(+/- 0.01) mg kg-1 which is almost zero and the highest V concentration extracted
with ammonium acetate for the PPS soil was 1.69 (+/- 0.12) mg kg-1 at the 250 V
loading rate.
Ammonium acetate was in general a poorer extractant for V compared to Bray 1.
NH4OAc is a weaker extractent and is buffered at pH = 7. The acetate did not displace
the vanadate. In these oxic soils it was suspected that the applied V (V) was not
reduced to vanadyl. Bray 1 is a stronger extractant (pH = 3) and the F- could easily
displace the sorbed V. Because of the suspected structural analogy between vanadate
(H2VO4-) and phosphate (H2PO4-) ions (Rheder, 1999; Crans, 1994), Bray 1 might be
more appropriate to use when testing for the extractability of V compared with
ammonium acetate, which represents the cationic form of V (VO2+). A good
extraction method to predict plant available V should not only correlate well with the
plant availability of an element but must also be in range with the actual quantity
taken up by the plant.
As discussed previously growth reduction occurred at the highest V application levels
(100 and 250 mg kg-1). The corresponding Bray 1 extractable V levels for the 100 and
the 250 levels were 23.5 and 77.6 mg kg-1, respectively. The data therefore suggested
a Bray threshold level of 23.5 (< 24 mg kg-1 if rounded up) for sandstone derived soils
which is arguable for the most common soils used for maize production on the
Eastern Highveld. The results of the PPS soils, suggested that dolerite derived soils
would sorb V, lowering vanadium’s plant availability even at high V application
levels.
3.4.6 Root analysis
The values obtained through the Bray 1 extraction method did not correlate well with
the total V taken up by the plant. A possible explanation for this was that much of the
V taken up by the plants was immobilised in the roots through the precipitate of, for
example, calcium vanadate (Cannon, 1963; Hempill, 1972; Lepp, 1977; Wallace et
al., 1977).
34
The V content of the roots (dry mass) at the 0 V level for the Nooit soil was 4.51 mg
kg-1 and the Bray 1 extractable V was 0.12 (+/- 0.09) mg kg-1. The V concentration in
the roots increased to 158 mg kg-1 for the 250 V level. The corresponding Bray 1
extractable V at this level was 77.6 (+/- 3.97) mg kg-1 which was about half of the V
taken up by the maize roots. The total V concentration in the above ground plant
material at the 250 level was 14.8 mg kg-1.
In the PPS soil the V content in the roots at the 0 V level was 3.04 mg kg -1 which was
similar to the V concentration in the roots in the Nooit soil. The Bray 1 extractability
was 0.07 (+/- 0.04) mg kg-1. The total V content of the roots in the PPS soil, at the 250
level, was 25.2 mg kg-1 which correlate better with the Bray 1 extractable V of 30.2
(+/- 2.77) mg kg-1 for this loading rate. The total V concentration in the above ground
plant material was 0.5 mg kg-1. This also showed that for the PPS soil, substantial V
immobilisation occurred in the soil itself and the low V levels in above ground
material was not only the result of poor translocation of V in the plant
3.4.7 Vanadium interaction with other nutrients
There are a few researchers that suggested that V might have an influence on the
uptake of other nutrients. Vanadate might influence phosphate uptake, and vanadyl
might influence cations like Ca and Mg (Rehder, 1999; Crans, 1994 & Welch, 1973).
An increase in K extractability was observed (Fig. 3.6), as the V concentration
increased in the soil. The ammonium acetate extractable K in the Nooit soil, at the 0 V
level, was 26.1 (+/- 4.12) mg kg-1 and increased to 44.0 (+/- 4.29) mg kg-1 for the 250
V loading rate. This represented an increase by 69%, which was statistically
significant according to the Tukey’s test (α = 0.05).
The ammonium acetate extractable K in the PPS soil at the 0 V level was 23.2 (+/4.34) mg kg-1 and increased to 29.5 (+/- 6.42) mg kg-1 for the 250 V loading rate,
which is a statistically insignificant increase of 27%.
35
The highest V treatment for the Nooit soil showed an ammonium acetate extractable
V level of 10.2 mg kg-1.
FIGURE 3.6 The ammonium acetate extractable K at different V loading rates.
The plant analysis did not show any increase in P and K levels for the highest V
treatments. This was most likely because sufficient amounts of K and P were supplied
to the plants. The plants would only have responded if they were stressed with respect
to K and P. Under field conditions this might not manifest as theorised. The reason for
this is that at these high levels of V, the maize plant experienced reduced growth and
poorly developed roots, therefore, on a whole, the plant might not benefit from the
greater availability of these nutrients. The poor root development will reduce the yield
of maize plants.
3.5 Conclusion
The threshold value where maize showed reduced growth in the Nooit soil was at a
total V concentration in the soil of 73.3 mg kg-1. The Bray 1 extractable V
concentration at this threshold was 23.5 mg kg-1 and there was no V in the above
ground plant material in the maize. The ammonium acetate extractability at this level
was 1.68 mg kg-1.
36
For the PPS soil the threshold value was observed at a total soil V concentration of
235 mg kg-1. The Total V concentration in the plant was 0.5 mg kg-1 and the Bray 1
extractable V was 30.3 mg kg-1. The ammonium acetate extractable V was 1.69 mg
kg-1. The V concentration in the roots for the 250 level was 158 mg kg-1 for the Nooit
soil and 25.2 mg kg-1 for the PPS soil.
V availability is highly dependent on soil characteristics and differed between the two
contrasting soils used in this study. The results indicated that the factors affecting P
plant availability can possibly be extrapolated to V in typical oxic soil environments.
It is reasonable to expect that soil properties known to influence P sorption, for
example texture, clay mineralogy and the abundance of Fe (oxy) hydroxides, can be
used to predict V plant availability and mobility. Soils with finer texture means that a
larger surface area is available for V sorption that could facilitate V sorption
compared to coarse textured soils. It is suspected that red soils derived from mafic
igneous rock like dolerite, diabase, norite and basalt, with a relative high kaolinite
clay fraction would all behave like the PPS soil and would have higher V sorption
capacities. Fine textured soils with red apedal B horizons like Hutton and Shortland
soil forms will have a higher V tolerance.
No significant reduction in growth or any V toxicity was induced in the high P fixing
PPS soil. This dolerite derived soil exhibited relatively low V plant availability
compared to the Nooit soil even at the highest V loading. Clear threshold levels were
established for the sandstone derived Nooit soil. The maize grown in Nooit soil
showed reduced growth at a V loading rate of 100 mg kg-1 and showed toxicity
symptoms (stunted growth and sclerosis on older leaves) at a V loading rate of 250
mg kg-1. Based on these results sandstone (coarse textured) derived soil will exhibit a
higher V plant availability than a dolerite (fine textured) derived soil with higher
concentrations of Fe.
In terms of V content of above ground material, levels greater than 10 mg kg-1
severely influenced the growth of maize plants. The pot trial also indicated that Bray
1 levels of higher than 23.5 mg kg-1 negatively impacted plant growth. Therefore,
37
Bray 1 can be used as an extractant to predict V plant availability and possible V
toxicity.
38
CHAPTER 4: PHYTO-AVAILABILITY OF VANADIUM
FROM BASIC OXYGEN FURNACE SLAG USED IN
AGRICULTURE
4.1 Introduction
Soils acidify on the Eastern Highveld. BOF slag is used to rectify acidification. The
reason is that most of the BOF slag is generated at steel plants much closer than
natural lime sources, rendering it more economically viable as transport costs keep on
escalating. Although these BOF slag has the same neutralizing capabilities compared
to natural lime source, it contains more impurities like heavy metals, depending on the
steel plant proses. Some of these slags contained various concentrations of different
heavy metals. This study focused on slag containing different concentrations of V.
It seems that higher plants like maize are not likely not accumulate V under natural
conditions (low V concentration) in the above ground parts as seen in the toxicity pot
trial (chapter 3). Usually the V concentration is the highest in the roots where it is
thought to precipitate out possible as calcium vanadate (Cannon, (1963), Hempill,
(1972), Lepp, (1977), Wallace et al., (1977)) Kaplan et el., (1989) place V in a group
of metals that is immobile in plants, V is taken up and accumulates in the roots, but
minimum translocation to other plant parts take place. Higher V concentrations
reduced plant growth and had a negative effect on the dry matter of maize. It is
therefore important to determine the V loading rate through different liming rates.
The field evaluation was done to investigate actual V loading under normal
agricultural practises and to determine the distribution and variation of V under field
conditions and to see if there was any translocation to the different above ground plant
parts. The fields chosen had a six year liming history with a V containing slag. One
39
field was located near Delmas (S 26 09.612 E 28 44.903) (Appendix A, Fig A1) and
the other near Ogies (S26 03.332 E 29 06.934) (Appendix A, Fig A2), both located on
the Eastern Highveld.
As already mentioned slag is used as an alternative liming material, because it is the
closest liming material source for farmers on the Highveld, making it cheaper due to
lower transport cost. The toxicity of V is however a reality and if these concentration
of V are reached in soil, as determined in chapter 3, the consequences are fatal. It is
therefore important to study the V loading through liming with a V containing slag.
4.2 Aims
The aim of the slag pot trial was to determine the plant availability of V from slag
under realistic agronomical slag application rates. The aim for the field monitoring
was to determine the plant availability of V under field conditions and to determine
the translocation of V in mature maize plants.
4.3 Materials and methods
4.3.1 Slag trial
Three slag where used in the slag pot trial. The four liming material was analytical
grade CaCO3. Slag A was a V rich basic oxygen furnace (BOF) slag and has the
highest V concentration and is commercially, slag B was a stainless steel plant slag,
with an intermediate V concentration. Slag C was a steel plant slag with a low V
concentration. The V content of the three slags used in this study was also determined
with the same three methods used for the soils. Slag A had the highest V
concentration followed by slag B and then C (Fig 4.1). The HNO3 acid digestion and
XRF scan methods showed higher V concentration than the HClO 4:HNO3 acid
digestion.
40
FIGURE 4.1. The V concentrations of the three slags as determined by various
methods.
Fig 4.2. shows the concentrations of V, relative to Cr and Ni (as determined by a XRF
scan and a HClO4:2HNO3 acid digestion). Slag A contained the most V and low
levels of Cr. Slag C, on the other hand, contained the most Cr .
12000
10000
V (mg kg -1 )
V Acid
8000
V XRF
Cr Acid
6000
Cr XRF
4000
Ni Acid
Ni XRF
2000
0
Slag A
Slag B
Slag C
FIGURE 4.2. The vanadium, chromium and nickel content of the three slags used as
determined by XRF and HClO4:2HNO3 acid digestion.
The Calcium carbonate equivalent (CCE) values were determined for the slags using
the Soil Suspension Method (Bornman 1985). This was to predict the reactivity of the
41
slag and to apply the correct quantity of the slag to reach the specific target pH values.
50 mg of liming material was added to 50 g of soil. 125 ml of deionised water was
added and the soil suspension was stirred and left to equilibrate over 24 hours. The
soil suspension was stirred thoroughly before the pH reading was taken. The CCE-RS
value relative to analytical grade CaCO3 was determined as follows:
Table 4.1 summarise the CCE-RS values for the three slags in the Nooit and PPS
soils. Slag C had the highest CCE-RS values so less quantities of this slag is
necessary to obtain a certain pH compared to the other slag. Slag A had the lowest
CCE-RS and has the lowest acid neutralising ability, thus a larger quantity of slag A is
necessary to obtain the desired pH.
TABLE 4.1. The calcium carbonate equivalent (CCE) values for the three slags in the
two soils
Slag A
Slag B
Slag C
CCE values
%
Nooit
PPS
44
54
59
60
68
75
Fig 4.3 shows the pH values obtained after liming in the Nooit and PPS soils with the
different liming materials in the slag pot trial (Chapter 4). The different target pH
values was chosen to accomplish 4 different application rates (Target pH 4, 6, 6.5 &
7).The Nooit soil target pH of 4 could not be reached because the Nooit soils initial
pH was 4.7. The pH of the Nooit soil limed with analytical grade CaCO 3 was about
0.5 a unit below the target pH value and about 1 unit below the target pH in the PPS
soil. This shows a slight error in the buffer curve measurement where Ca(OH)2 was
used and which is a stronger base compared to CaCO 3. The pH values obtained with
Slag A was near the target pH value in the Nooit soil but not in the PPS soil. This can
also be attributed to the error with the buffer curve. The pH of the soils limed with
42
slag B and C correlated well with the target pH values. Taking into account the error
with the buffer curve, then the actual CCE values were slightly higher for slag B and
C than initially estimated.
Target pH
pH Nooit
pH PPS
8.0
7.0
6.0
pH
5.0
4.0
3.0
2.0
1.0
0.0
CaCO3
Slag A
Slag B
Slag C
FIGURE 4.3. The pH values of the soil solution after the two soils were limed with
the various liming material.
A greenhouse pot trial and field monitoring were used to gain better understanding of
V release and V plant availability in soils amended with V containing slag. It was also
done to determine the short term risk posed by the use of slag under natural conditions
and to determine the actual V field loading rate through slag application. The V
loading rates used in the pot trial and at the field evaluation site are typical of a six
year liming period on the Eastern Highveld, which will represent short term reactions
in the soil and relatively low cumulative V loading rates. With the V loading rate
calculated, one can make predictions on the sustainability on the use of V containing
slag as a liming material.
The V status of the soil at the field monitoring sites and the slag pot trial will be
investigated on hand with parameters, generated from the V toxicity pot trial. These
parameters are; total V concentration in the soil and in the maize plant and Bray 1 and
ammonium acetate extractable V.
43
The two soils for the slag trial were the same than that used in Chapter 3 (toxicity pot
trial), (Nooit and PPS). The soils chosen, represented typical soil conditions that
requires liming. The soil properties are reported in Chapter 3. The slag trial had 4 pH
target levels; 4.0, 6.0, 6.5 and 7.0. These different target pH levels resulted in four
liming rates simulating a low to high liming rate for one season. Four different liming
materials were used according to the difference in V concentrations; slag A, B and C
and CaCO3. These slags are commonly used as liming materials by farmers on the
Eastern Highveld. Analytical grade CaCO3 was the fourth liming material and was
used for the control, containing no V.
Buffer curves for both soils were determined, using the addition of saturated Ca(OH) 2
solution to 50 g of soil with a solution ratio of 1:1. The solution was left to equilibrate
for 24 hours, after which the pH was determined. The concentration of the saturated
Ca(OH)2 solution was determined by titrating with a 0.01M HCl solution. The moles
of Ca(OH)2 needed per mass of soil to reach this target pH were then converted to an
equivalent amount of pure CaCO3, by multiplying it to the molar mass of CaCO3,
taking into consideration account the difference in molecular weight between the
hydroxyl and carbonate groups.
The Nooit and PPS soils were passed through a 5 mm sieve and placed in 4kg pots.
The different liming materials were weighed, together with the P fertilizer in the form
of MAP. A summary of the quantities of the lime added to each pot is given in Table
4.2. The four replications were mixed together in a concrete mixer for 5 min to ensure
that the soil and the liming materials were mixed thoroughly. The pots were filled
with 4 kg of soil. There were three wet and dry cycles over a six week period to
ensure enough reaction time for the liming materials and to ensure that the target pH
values were reached. Two maize plants was plant in every pot. The maize plants
received the same fertiliser and were sampled like the maize in the toxicity pot trail
(Chapter 3). The maize plants were harvested after six weeks of growth, and the
above ground plant material was dried and sampled according the different
treatments.
44
TABLE 4.2. The liming rates for the different soils and the different liming
materials
Soil
Nooit
Liming
mat. (CCE)
CaCO3
Slag A
(0.44)
Slag B
(0.59)
Slag C
(0.68)
PPS
CaCO3
Slag A
(0.54)
Slag B
(0.6)
Slag C
(0.75)
Target pH
g/pot
ton/ha
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
4.00
6.00
6.50
7.00
0.00
2.60
4.00
4.60
0.00
5.91
9.09
10.45
0.00
4.41
6.78
7.80
0.00
3.82
5.88
6.76
0.00
4.40
6.40
8.00
0.00
8.15
11.85
14.81
0.00
7.33
10.67
13.33
0.00
5.87
8.53
10.67
0.00
1.69
2.60
2.99
0.00
3.84
5.91
6.80
0.00
2.86
4.41
5.07
0.00
2.49
3.82
4.40
0.00
2.86
4.16
5.20
0.00
5.30
7.70
9.63
0.00
4.77
6.93
8.67
0.00
3.81
5.55
6.93
pH after
liming
4.8
5.4
6.1
6.6
4.6
5.3
6.5
6.9
4.8
6.1
6.8
7.0
4.7
6.2
7.0
7.4
4.0
5.5
5.5
6.1
4.0
5.0
5.5
6.1
4.1
5.5
6.3
6.8
4.1
6.0
6.9
7.1
4.3.2 Field evaluation
Two sites were identified; a 20 ha field under irrigation near Delmas (“Delmas site”)
and 40 ha under dry land production near Ogies (“Ogies site”). These sites were
chosen to investigate the mobility of V, under irrigation compared to dry land maize
production. The soils at the Delmas site consisted of deep (> 1.2 m deep) Hutton and
45
Clovelly soil forms. The Ogies soils consisted mainly of deep Clovelly soil forms.
The soil texture at these sites range from sandy to sandy loam soils (9 - 18% clay).
The texture of the soils from these sites was similar than that of the Nooit soil used in
the pot trials.
Samples were taken on a 100 m grid, which means one sample per hectare. These
sites were chosen because they were limed with Slag A which had the highest total V
concentration (918 mg kg -1). According to the agronomists that advised the farmers
these two sites had a six year liming history, and each received about 6 tons of slag A
over this period, 3 tons/ha every third year.
Soil and plant samples were taken six weeks (42 days) after emergence and 17 weeks
(120 days) after emergence. The samples were taken on a 100 m grid which was
logged as waypoints using the Global Positioning System (GPS) so that the second
sampling could be done near the point from the first sampling. The samples taken
after six weeks were to compare the field evaluation site’s result with the pot trial’s
results. The second sampling was done to determine the V translocation in the mature
maize plant.
With the 120 days sampling, the different plant parts, the cob, upper leaves and lower
leaves, were cut with a pruning shear and sampled separately. The soil samples were
collected at two depth intervals, 0–30cm and 30–60 cm with a Johnson soil auger, to
determine the concentration of V in different horizons and to investigate if there was
any accumulation or removal of V at these two soil depths. Soil samples were also
taken from natural soils nearest to the fields. These natural soils did not receive any
slag at all, and served as the control samples. Samples marked K1 – K4 is the control
waypoints of the Ogies site and samples K5 – K7 was the control samples for the
Delmas site, as seen on the two Google Earth Maps (Appendix A1 & A2).
The plant samples were washed and then oven dried at 60 ºC for two days and sieved.
The soils samples were air dried and passed through a 2 mm sieve.
46
4.3.3 Soil and plant analysis
The soil and plant analysis performed for the slag pot trial and field evaluation were
the same as in chapter 3.
4.3.4 Statistical analysis
The statistical program SAS was used for statistical analyses of the data generated
from the aglime pot trial. The Tukey test was used to determine significant difference
at a significance level of α = 0.05 between the different treatments. Means with the
same letter are not significantly different. Population statistics was done on the field
samples using Microsoft Excel. The frequency distribution and cumulative normal
distribution curves were drawn for the different data sets. A one-tailed T test with
unequal variance was used to determine whether the control samples outside the field
were significantly lower than the soil samples inside the field.
4.4 Results and discussion
4.4.1 Biomass production of maize as influenced by soil acidity and liming
The importance of liming was illustrated in the liming pot trail. Both the controls of
the Nooit and PPS soil received no lime and had very low pH values. The Nooit soil
control had a pH(H2O) of 4.7, however, it showed no reduced growth compared to the
other pots, which received lime and slag (data not shown). Tukey’s test also showed
no significant difference in the above ground dry mass of maize plants growing in
soils with different pH values in the Nooit soil, which had low exchangeable acidity.
In the case of the PPS soil, a Tukey’s test showed a significant difference between the
dry mass of maize plants that was limed and maize plants that received no liming
material (Fig. 4.4). The maize plants growing in the PPS soil which received no
liming material, showed reduced and stunted growth as a result of an under developed
root system, although it received the same amount of fertilizer and water as the other
treatments. This is probably due to the fact that the pH (H2O) of 4 is very low and the
exchangeable acidity was higher compared to the exchangeable acidity of the Nooit
47
soil. At pH 4(H2O) Al is more soluble and can lead to Al toxicities or deficiencies of
P, Ca, Mg, K, S, and Mo (Ceballos et al., 1995; Zeigler et al., 1995). As the pH
Plant dry mass (g pot -1 )
increased, the maize growth was normal.
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
CaCO3
Slag A
Slag B
Slag C
4.0
6.0
6.5
7.0
pH (H2O)
FIGURE 4.4 The dry mass of the maize plants (above ground biomass) grown in the
PPS soil at the various target pH levels achieved with different liming rates.
4.4.2 Estimating vanadium loading rates for the slag pot trail
There are various factors that influence the V loading rates of soils through liming.
The factors include; V content of slag, slag reactivity, which in turn will influence
quantity of slag added to ameliorate soil acidity, and the incorporation depth in the
soil. It is also difficult to apply liming material homogeneously over a field, as wind
direction and particle size of the liming material play a huge role. A soil depth of 20
cm for slag incorporation was used in this study for all the calculations. This is a more
realistic incorporation depth, because slag are more commonly incorporated and
mixed in the soil with a disc harrow which have an average incorporation depth of 20
cm.
Van der Waals & Claassens (unpublished report) made an estimate of the V loading
through slag application based on a V content of 1500 mg kg-1 slag. According to
these authors, V concentration in the soil after 100 years would be 50 mg kg-1 if V
48
background levels were negligible and the slag applied at 1.5 tons/ha/year. Other
assumptions were an incorporation depth of 30 cm and bulk density of 1500 kg m3.
Slag A originated from the same steel plant than that used by Van der Waals &
Claassens, however, this time around the V content was lower. The total V
concentration for slag A, B and C was 918 (+/- 116), 153 (+/- 50.56) and 88.6 (+/9.10) mg kg-1 respectively. Based on this V content the V loading rates for slag A, B
and were calculated at 0.35, 0.06 and 0.03 mg kg-1 year-1 respectively (Table 4.2).
This was based on an incorporation depth of 0.2 m and a bulk density of 1300 kg m-3
and assuming no V losses occur through erosion or leaching. Table 4.2 summarises
the theoretical long term accumulation of V if this yearly loading rate is consistently
applied and the V content of the slags stays constant. A slag incorporation depth of 20
and 30 cm were taken to establish an upper and lower limited to allow for the
variability that can occur because of different means of incorporation. As mentioned
previously 20 cm represent the average depth for disc harrow while 30 cm can be
viewed as a limit for conventional (mouldboard) ploughing. In chapter 3 the threshold
V concentration where maize exhibited reduced growth was determined at 100 mg kg1
for the sandier Nooit soil. This threshold value was used as a reference to estimate
the years it will take to reach critical V level. The Nooit soil is a sandstone derived
soil, commonly cultivated on the Highveld. While the dolerite derived soils like the
PPS soil occurs as dyke intrusions and is less common on the Eastern Highveld.
Depending on the incorporation depth it was estimated that it would take between 189
– 283 years to reach the V threshold value of 100 mg kg-1 where the growth of maize
is influenced negatively. It would take an estimated 470 -700 years to reach the 250
mg kg-1 level for the PPS soil, if all of the parameters that influence V accumulation
are kept constant as assessed in this study. The Russian guideline stipulate a total V
concentration of 150 mg kg-1 and the Canadian threshold V concentration is 130 mg
kg-1 but the type of soil or crop are not mentioned in these guidelines. It would take
343 years to reach the Canadian V threshold concentration and 428 years for the
suggested Russian V concentration. As time progress, it is possible that V could
transform to more stable forms on mineral surfaces, rendering it unavailable for
plants. This could increase the years that V containing liming material could be used
49
before the critical V levels are reached. Further research is necessary to study the long
term availability of V in the soil applied through slag application.
Slag B and C contained lower V compared to slag A, and had lower V accumulation
rates (Table 4.3.) The safe period for these slags in respect with V is much longer but
other heavy metal concentrations must also be kept in mind. Slag containing lower V
concentrations must be used on V sensitive soils like the Nooit soil as an alternative
liming material.
TABLE 4.3 The predicted V accumulation rates and the years to reach the V
threshold level for the sandstone derived Nooit soil (100 mg kg-1) and Canadian
regulatory level (130 mg kg-1) for various incorporation depths and at agronomic slag
application rates
Slag
V
content
of slag
mg kg-1
V loading per
year (mg kg-1 a-1)#
Years to reach:
100 mg kg-1
Incorporation depth
20 cm
30 cm
189
283
919
20 cm
0.35
30 cm
0.24
Slag B
154
0.06
0.04
1127
Slag C
88.6
0.03
0.02
1956
Slag A
130 mg kg-1
20 cm
246
30 cm
368
1690
1465
2197
2934
2543
3814
-1
# Based on an annual application rate of 2 ton ha
Total V analysis was done on the slag amended soils with the aim to corroborate the
calculated V loading rates. However, the liming rates and subsequent V loading rates
were too low to establish measurable differences between the control and slag
treatments (Figures 4.5 & 4.6). A Tukey’s test showed no significant difference
between the total V concentration (HClO4:2HNO3 digestion) of the CaCO3 and slag
treatments for both soils (ALASA, 1998).
50
FIGURE 4.5 The total V concentrations in the Nooit soil after the different liming
treatments.
FIGURE 4.6 The total V concentrations in the PPS soil after the different liming
treatments.
4.4.3 Vanadium plant availability and plant uptake for the slag pot trial
According to literature and the V toxicity trail, the V concentrations in above ground
plant parts are very low under normal V concentration in soils. Vanadium is
immobilised in the roots, or precipitate on the root surface (Cannon, 1963; Hempill,
1972; Lepp, 1977; Wallace et al., 1977).
51
The threshold V concentration in the above ground dry mass of maize plants was
determined in the V toxicity trial to be 14.8 mg kg-1 (Chapter 3). In the slag pot trial
the highest V concentration in the above ground mass of the maize plants was 0.21
mg kg-1, with all the other treatments lower than 0.21 mg kg-1. This shows that V is
not translocated and does not accumulate in the above ground plant part of the maize
plant under low V loading rates.
4.4.4 Vanadium translocation in the maize plant in the pot trail
Plants did not reach maturity in the pot trial, therefore, to study the translocation of V
in the plant, mature plants were sampled, where different plant parts were sampled
separately under field conditions. A total of eight mature maize plants were sampled
at each site. The different plant parts collected were; leaves and stem underneath the
cob, the cob, and leaves and stem above the cob. The plant analyses from the slag pot
trial and the first field evaluation sampling showed that V isn’t readily taken up by the
maize plant. So only some of the samples were taken for the second field evaluation
to test this hypothesis.
4.4.5 Influence of vanadium loading through slag application on its Bray 1
extractability of vanadium in the pot trail
In the slag pot trial vanadium’s Bray 1 extractability increased as the quantity of Slag
A increased, in both soils (Red line Fig. 4.7 & 4.8). The control and other slag
treatments did not show an increase in Bray 1 extractable V due to an increase in slag
application rate. The Bray 1 extractable V at the 0 treatment in the Nooit soil was 0.28
(+/- 0.07) mg kg-1 and 1.55 (+/- 0.06) mg kg-1 at the pH 7 treatment, a 453 % increase.
In the PPS soil the Bray 1 extractable V at the 0 treatment was 0.13 (+/- 0.04) mg kg-1
and 0.65 (+/- 0.05) at the pH 7 treatment, a 400 % increase. This means that if more
slag A is applied, the Bray 1 extractability of V increases.
52
FIGURE 4.7 The Bray 1 extractable V in the Nooit soil as influenced by the different
lime treatments.
FIGURE 4.8 The Bray 1 extractable V in the PPS soil as influenced by the different
lime treatments.
4.4.6 Influence of vanadium loading through slag application on its ammonium
acetate extractability in the pot trail
The ammonium acetate extractable V in both soils was about 0.18 mg kg-1 for all the
treatment levels, which indicate that the dominant available fraction of V in the slag is
53
vanadate or that the V in slag is oxidised to vanadate in a well-drained soil or that
ammonium acetate is a poor V extractant.
4.4.7 Estimating vanadium loading rates for the field evaluation sites
The background V concentrations of the Ogies and Delmas control samples (8.4 &
19.1 mg kg-1) were lower than the Nooit (26.4 mg kg-1) and PPS (49.9 mg kg-1) soils.
These values were also relatively low compared to the natural V concentrations for
soils found in the literature, which ranges between 10 mg kg-1 and 220 mg kg-1
(Kabata-Pendias & Pendias, 1993; Panichev et al. 2006). The theoretical increase of
V over the six year period with slag A should have been 2.1 mg kg-1. The variation for
the top 30 cm at the Delmas site, as shown in the frequency distribution graph (Fig
4.9), however, was too high and it was impossible to detect any increase in V.
7
6
6
Frequency
5
4
4
3
3
3
2
2
27
30
2
1
1
1
0
0
3
6
0
9
12
15
18
21
24
V content (mg kg-1 ) Value shows upper limit
FIGURE 4.9 The frequency distribution for the total V concentrations of the 0-30 cm
soil samples at the Delmas site, (control samples omitted)
The 95 % confidence interval for a normal distribution population of total V
concentration ranged from 10.1 to 31.4 mg kg-1 for the top 30 cm at the Delmas site.
The total V concentration frequency distribution for the 30-60 cm soil depth samples
was similar to the 0-30 cm soil depth samples (Appendixes B, Fig B1). In the 30-60
cm soil depth the range (with a 95 % confidence interval), for total V concentration
54
was 8.9 – 26.4 mg kg-1. According to a one-tailed T-test with unequal variance there
were no difference between the total V concentration of the 0-30 cm and 30-60 cm
soil samples. This indicated that there was no significant or measureable removal
from the topsoil or significant accumulation of V in the subsoil. A cumulative normal
distribution was drawn for the top 30 cm (Fig 4.10, 0-60 cm soil depth in Appendix B,
Fig B2.) to compare the V content of the control soils (red data points) relative to that
collected at the slag amended Delmas site. The orange data point is the mean of the
outside samples. The control samples were located almost at the median (0.5) of the
population, for these samples to suggest a significant difference, it should have been
located in the lower 5% (<0.05) of the distribution. An one-tailed T-test with unequal
variance on the same samples also showed that the controls did not differ significantly
from samples collected at the slag amended Delmas site (P= 0.054 for 0-30cm and
P=0.34 for 30-60cm ). This corroborated the findings of the cumulative normal
distribution approach (Fig 4.10)
Cumulative normal distribution
1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
5
10
15
20
25
30
35
V content (mg kg-1 )
FIGURE 4.10 The total V cumulative normal distribution curve for the 0-30 cm soil
samples of the Delmas site. The red data points represent the samples collected
outside the field. The orange data point is the average of the outside samples.
For the Ogies site, the 95 % confidence interval of total V concentration for the top 30
cm ranged from 8.4 to 21.1 mg kg-1. The frequency distribution figure for the total V
55
concentration for the top 30 cm at the Ogies was closer to a normal distribution
compared to the Delmas population site (Fig 4.11),. The frequency distribution for the
30-60 cm soil depth samples (Appendixes B, Fig B3) looks similar to the 0-30 cm soil
depth samples. In the 30-60 cm soil depth the total V concentration range was 5.50 –
20.0 mg kg-1 for a 95 % confidence interval.
FIGURE 4.11 The total V concentration frequency distribution of the 0-30 cm soil
samples at the Ogies site, (control samples omitted).
The cumulative normal distribution curve for the 0-30 cm soil depth of the Ogies soil
samples are illustrated in Fig 4.12 and the 30-60 cm in Appendix B, Fig B4. These
data points for the samples collected outside the field were clearly located in the lower
tail of the distribution, showing a clearer separation from the amended soils and
therefore stronger evidence of some vanadium enrichment in the cultivated fields The
amount of samples were small and arguably would have been better described by a T
distribution. An simple one tailed T test with unequal variance seemed to support the
fact that the total V concentration of the control population for the Ogies site differed
significantly from the population of the field samples in the field for the 0-30 cm (P =
0.0002) and 30-60cm (P = 0.0009). Therefore, a measurable effect of the use of V
containing slag was detected for the Ogies site. The two-tailed T-test, however,
showed no difference in the total V concentrations between the 0-30 cm and 30-60 cm
56
soil depth, again showing no measurable removal or accumulation took place from
either soil depth.
FIGURE 4.12 The cumulative normal distribution curve for the 0-30 cm soil samples
of the Ogies site. The red data points represent the samples collected outside the field.
The orange data point is the average of the outside samples.
Although the population statistic showed a difference in V concentration between the
control and the field samples, the variability of V in the fields made it difficult to
determine a loading rate in the soils with a short liming history (six years). Another
problem is that the liming history before the six years of precision farming at these
two sites was unknown.
4.4.8 Vanadium plant availability and plant uptake in the field evaluation site
In the field evaluation site the average total V concentration in the maize plant was
0.13 mg kg-1 with a maximum of 0.18 mg kg-1 for the Delmas site and an average of
0.19 mg kg-1 and maximum of 0.31 mg kg-1 for the Ogies site. These values were
below the estimated total V concentration of 10 mg kg-1 (chapter 3) where reduced
growth can be expected.
57
4.4.9 Vanadium translocation in the maize plant from the field evaluation sites
Kaplan et al., (1989) confirmed that V accumulates in the roots which place V in a
group of metals that is immobile in plants. There is minimum translocation of V in the
maize plant and the concentration decreased as the distance from the roots increase
(Fig. 4.13). For the Delmas site the leaves below the cob had the highest V
concentration (0.45 +/- 0.08 mg kg-1), then the leaves above the cob (0.04 +/- 0.11 mg
kg-1). The cob itself had no V, and this supports Kaplan’s theory that there is a
minimum transport of V in the plant. The total V concentration in the maize plants
growing at the Ogies site was 0.76 +/- 0.08 mg kg-1 for the leaves and stem
underneath the cob, 0.41 +/- 0.15 mg kg-1 for leaves and stem above the cob and 0.03
+/- 0.08 mg kg-1 in the cob.
FIGURE 4.13 The V concentrations in the above ground maize plant at the two field
evaluation sites at 120 days of growth.
4.4.10 Influence of vanadium loading through slag application on Bray 1
extractability of vanadium at the field evaluation site
The 95 % confidence interval for Bray 1 extractable V ranged from 0.11 to 0.28 mg
kg-1 for the top 30 cm of soil of the Delmas field evaluation site (Fig 4.14). The
frequency distribution for the 30-60 cm soil depth was similar to the 0-30 cm soil
58
depth. In the 30-60 cm soil depth (Appendixes B, Fig B5) the range was 0.11 to 0.19
mg kg-1. The Bray 1 extractable V cumulative normal distribution curve for the 0-30
cm soil depth of the Delmas soil samples are illustrated in Fig 4.15 and for the 30-60
cm soil depth in Appendix B, Fig B6. The red data points represent the samples
collected outside the field. The orange data point is the average of the outside
samples. According to a one tailed T test with unequal variance showed that the Bray
1 extractable V control population of the Delmas did not differ significantly from the
population of the samples in the field for the 0-30 and 30-60 cm soil depth (P= 0.4 &
0.07).
FIGURE 4.14 The Bray 1 extractable V frequency distribution of the 0-30 cm soil
samples at the Delmas site, without the control samples.
59
FIGURE 4.15 The cumulative normal distribution curve for the Bray 1 extractable V
of the 0-30 cm soil samples of the Delmas site.
The 95 % confidence interval for Bray 1 extractable V ranged from 0.13 to 0.95 mg
kg-1 for the top 30 cm of soil of the Ogies field evaluation site (Fig 4.16).
FIGURE 4.16 The Bray 1 extractable V frequency distribution of the 0-30 cm soil
samples at the Ogies site, without the control samples.
The frequency distribution for the 30-60 cm soil depth looks similar to the 0-30 cm
soil depth. In the 30-60 cm soil depth (Appendix B, Fig B7) the range was 0.11 to
0.37 mg kg-1. The cumulative normal distribution curve for the 0-30 cm soil depth of
60
the Ogies soil samples are illustrated in Fig 4.17 and for the 30-60 cm soil depth in
Appendix B, Fig B8. The red data points represent the samples collected outside the
field. The orange data point is the average of the outside samples. According to a one
tailed T test with unequal variance showed that the Bray 1 extractable V control
population of the Ogies site did not differ significantly from the population of the
field samples in the field for the 0-30 and 30-60 cm soil depth (P= 0.15 & 0.18). A
one tailed T test with unequal variance showed a significant difference between the
samples in the field at the two soil depths. The 0-30 cm soil samples had a higher
Bray 1 extractable V compared to the 30 – 60cm soil samples.
FIGURE 4.17 The cumulative normal distribution curve for the Bray 1 extractable V
of the 0-30 cm soil samples of the Ogies site
The data from the field evaluation site for Bray 1 extractable V was lower compare to
the data obtained from the slag pot trail. This suggests that the Bray 1 extractability of
V decrease over time. This happens when V is sorbed to Fe mineral surface like
goethite and a further study should be done to investigate the long term reactions of V
in soils. All the field evaluation site soil samples is far below the Bray 1 extractable V
concentration of 23.6 mg kg-1 soil (chapter 3) where V influence the plant growth of
maize negatively.
61
4.4.11 Influence of vanadium loading through slag application on acetate
extractability of Vanadium at the field evaluation site
The ammonium acetate extractable V in the Delmas and Ogies soils showed no
difference between the control samples and the samples in the field. The Delmas soil
ranged from 0.05 to 0.09 mg kg-1 for the 0-30 cm soil depth and 0.05 to 0.1 mg kg-1
for the 30-60 cm soil depth. The Ogies 0-30 cm samples ranged from 0.06 to 0.17 mg
kg-1 and from 0.05 to 0.14 mg kg-1 for the 30-60 cm soil depth.
4.5 Conclusion
The slag pot trial showed the effect of soil acidity and the importance of liming. The
maize growing in the PPS soil showed reduced growth in the untreated pH(H 2O) = 4
soil. This shows the necessity to lime soils that has acidified to a pH below 4, even
with slag that contain various concentrations of heavy metals. If these concentrations
are determined, the loading rates can be used to manage these limes as sustainable
liming materials.
By using the same theoretical approach as Van der Waals & Claassens to calculate V
accumulation and by using the threshold V concentration generated in chapter 3, the
period to reach the critical V threshold value for liming with slag A was determined.
If all factors (V concentration and incorporation depth), were to be kept constant, it
will take an estimated 186 years of liming with slag a in the Nooit soil to reach the
threshold value of 100 mg kg-1 where V will negatively affect the growth of maize
plants. This period was calculated to be 472 years for the PPS soil, due to the higher
Fe content and finer textured soil, which increase the V sorption capacity of the soil.
The safe period for the slag B and C in respect with V is much longer than slag A, but
other heavy metal concentrations must be kept in mind for they too can accumulate in
the soil and can influence the growth of maize negatively if certain threshold values
are reached.
62
The plant availability of V under the low V loading rates was low. All the V
concentration in the above ground plant parts of the maize plants were far below the
threshold value for total V (14.8 mg kg-1) in the plant as determined in chapter 3. As
the application rate in the pot trail with slag A increased, the Bray 1 extractable V
increased. The Bray 1 data from slag A showed that Bray 1 can be used as a method
to predict the V status in a certain soil.
The second sampling of the maize plant parts confirmed that minimum V was
translocated to the above ground plant part, especially to the cob, which V
concentration was close to zero. This shows that it is safe to use the maize after
harvest in respect with V concentration.
To determine the actual total V loading rate proved to be a difficult task, the main
reason being the low loading rate applied to the pots over the simulated 6 year period.
The data from the field evaluation showed an increase in the V content in the field
samples of the Ogies evaluation site compared to the control samples outside the field,
but the loading rates could not be quantified. The reason for this was the variance of
all the soil samples. A long term evaluation site with an accurate slag history and well
defined control sites at the border of the field, where slag A is applied as liming
material, should be identified and monitored.
The total V concentrations in the plants were also close to zero which supports the
fact that V is not translocated in the plant under natural V concentration. All these
results of the V toxicity pot trail and the slag pot trial shows that the use of slag that
contains high concentrations of slag is save over the short period (10 years) and even
longer if the sites are monitored and managed.
The Bray 1 extractability of V increased as the pH of the soil increased and when slag
A was used as a liming material. All the Bray 1 extraction values for V in the pot trial
and at the field evaluation site were well below the threshold value of 30 mg kg -1.
63
CHAPTER 5: SUMMARY AND CONCLUSION
Soil acidification is a wide spread problem especially on the Eastern Highveld. Slag is
used to rectify soil acidification, but can contain various levels of heavy metals like
vanadium. A toxicity pot trial was used to determine the level where V becomes toxic
to maize plants in two soils, Nooit and PPS. The toxicity pot trail was used to
establish V threshold values with various indicators, like total V concentration in the
soil, total V concentration in the plant, Bray 1 extractable V and ammonium acetate
extractable V.
The threshold value where maize showed reduced growth in the Nooit soil was at a
total V concentration in the soil of 73.3 mg kg-1. The Bray 1 extractable V at this
threshold was 23.5 mg kg-1 and there was no V in the above ground plant material in
the maize. The ammonium acetate extractability at this level was 1.68 mg kg-1. V
toxicity occurred (stunted growth and sclerosis on older leaves) at a total V
concentration of 150 mg kg-1, with Bray 1 extractable V at 77.6 mg kg-1 and total V
concentration in the maize plant of 14.8 mg kg-1.
For the PPS soil, a dolerite soil with a larger V sorption capacity compared to the
Nooit soil, the threshold value was determined to be 235 mg kg-1 for the total V
concentration in the soil. The Total V concentration in the plant was 0.5 mg kg -1 and
the Bray 1 extractable V was 30.3 mg kg-1. The ammonium acetate extractable V was
1.69 mg kg-1. The Russian guideline stipulate that 150 mg kg-1 is the maximum total
V concentration allowed in soil (Ghost, 1985) and the Canadian soil quality guideline
130 mg kg-1 in soil used for agriculture (Canadian Council. 1997). Although both of
these guidelines do not distinguish between soil types, however, these values
correspond with the values obtained for the Nooit soil. At the moment there are no V
guidelines in South Africa, therefore from a regulatory point of view it seems that
both the Canadian and Russian guidelines would be reasonable to use..
V availability is highly dependent on soil characteristics as was evident from the slag
en toxicity pot trail. The results indicated that the factors affecting phosphate plant
availability can possibly be extrapolated to vanadate for oxic soil environments
64
(where V is in the +5 oxidation state). It is reasonable to expect that soil properties
known to influence P sorption, for example texture, clay mineralogy and the
abundance of Fe (oxy) hydroxides, can be used to predict V plant availability and
mobility. Soils with finer texture means that a larger surface area is available for V
sorption that could facilitate V sorption compared to coarse textured soils. Red
coloured soils derived from mafic rocks, with a relative high kaolinite clay fraction
and high P fixing capacity (which the PPS soil represented) is expected to also have
higher V sorption capacities. Fine textured soils with red apedal B horizons like
Hutton and Shortland soil forms will have a higher V tolerance.
Theoretical V loading rates were calculated for three different slags containing
different V concentrations and by using the threshold V concentration generated in
chapter 3, the period to reach the critical V threshold value for liming with slag A was
determined. If all factors (V concentration, incorporation depth and no erosion), were
to be kept constant, it will take an estimated 186 years of liming with slag A in the
Nooit soil to reach the threshold value of 100 mg kg-1 where V will negatively affect
the growth of maize. This period was calculated to be 472 years for the PPS soil, due
to the higher Fe content and finer textured soil, which increase the V sorption capacity
of the soil. The safe period for the slag B and C in respect with V is much longer than
slag A, but other heavy metal concentrations can be limiting and influence the growth
of maize negatively if certain threshold values are reached.
The V concentration of the slag pot trail and field evaluation site was far below the
threshold values compared to the V threshold values generated in the V toxicity pot
trail. Minimal V translocation occured to above ground material in the maize under
conditions of low V loading simulated in the pot trail and in the field evaluation site
was not taken up by the plant and that was harvested at the field evaluation site. This
indicated that the slag use with high V concentration in the short term (< 10 years) is
unlikely to negatively influence crop production because of the low V loading rate to
the soil (0.35 mg kg-1) and little transfer of V to above ground parts of maize.
65
More information is needed to predict the long term use of slag containing high V
concentrations. Sandy soils with long term application rates and control sites, which
received no slag application, should be established and then monitored. If other slag is
available, slag A should not be used for long periods on the same soils, especially if it
is a sandy soil. A long term application site with a detailed slag application history
with numerous control sites which received no slag might give a more accurate V
loading rate.
66
REFERENCES
ACT No. 36 of 1947. Fertilizer, Farm feeds, Agricultural remedies and stock remedies
act, 1947.
ADRIANO DC. 1986. Trace elements in the terrestrial environment. Springer-Verlag,
New York.
ADRIANO DC. 2001. Biochemistry, bioavailability and risks of metals. In Trace
elements in the terrestrial environments 2nd Ed. Springer, New York.
ALASA. 1998. Handbook of feeds and plant analysis. Palic, D. (Ed).
ALEXANDER M. 1977. In Introduction to soil microbiology. 2nd Ed. Wiley, New
York.
ALIBRAHIM A, SAYAG DR, ANDRE LE. 1988. Effect of liming on the fixation
and release of ammonium and potassium by a trioctahedral vermiculite.
Agrochemica 32. 229-235.
ARNON DI, WESSEL G. 1953. Vanadium as an essential element for green plants.
Nature 172. 1039-1040.
ATSDR (Agency for Toxic Substances and Disease Registry). 1992. Toxicological
Profile for Vanadium., U.S. Dept. of Health & Human Services,
Washington, D.C.
BAES CF, MESMER RE. 1976. The Hydrolysis of Cations. John Wiley, New York.
BARBER SA. 1976. Liming materials in practise. In: PEARSON RW, ADAMS F.
(Ed) Soil acidity and liming. Madison, Wisconsin, USA. Am. Soc. Agron p.
125-160.
67
BARROW NJ. 1985. Reactions of anions and cations with variable-charge soils. Adv.
Agron. 38. 183-230.
BENDITO
DP,
RUBIO
S.
1999.
Environmental
Analytical
Chemistry
(Comprehensive Analytical Chemistry, vol. XXXII), Elsevier, Amsterdam.
BERROW ML, WILSON MJ, REAVES GA. 1978. Origin of extractable titanium
and vanadium in the horizons of Scottish podzols. Geoderma 21. 89-103.
BEUKES DJ. 1995. Benefits from identifying and correcting soil acidity in
agriculture. Pamphlet published by the Agricultural Research Council.
BLOOMFIELD C. 1981. The translocation of metals in soil. In The chemistry of soil
processes, GREENLAND DJ, HAYES MHB. (Ed). John Wiley and Sons,
New York.
BLUE WG, FERRER MS. 1986. Potassium retention in a sandy Florida spodosol
with variable organic matter concentrations and liming histories. Soil Crop
Sci. Fl. 45. 90-94.
BOLAN NS, ADRIANO DC, CURTIN D. 2003. Soil acidification and liming
interactions
with
nutrient
and
heavy
metal
transformation
and
bioavailability. Adv. Agron. 78. 215-272.
BORNMAN JJ. 1985. n Suspensiemetode vir die evaluering van lanboukalk. MSc.
(Agric.)-verhandeling. Universiteit van Pretoria.
BOUMAN OT, CURTIN D, CAMBELL CA, BIEDERBECK VO, UKRAINETZ H.
1995. Soil acidification from long term use of anhydrous ammonia and urea.
Soil Sci. Soc. Am. J. 59. 1488-1494.
BRADY NC, WEIL RR. 2002. The nature and properties of soil. 13th Ed. USA:
Prentice Hall.
68
VAN BREEMEN N. 1991. Soil acidification and alkalisation. In Soil Acidity.
Springer, New York. p 1-7.
BREIT GN, GOLDHABER MB. 1989. Hematite-enriched sandstones and chromiumrich clays – clues to the origin of vanadium-uranium deposits in the
Morrison Formation, South-western Colorado and south-eastern Utah.
Vienna Australia, International Atomic Energy Agency, Technical
Document 500. 201-226.
BROOKS RR. 1972. Geobotany and biochemistry of mineral exploration. Harper and
Row Publishers, New York.
CANADIAN COUNCIL OF MINISTERS OF THE ENVIRONMENT. 1999.
Canadian soil quality guidelines for the protection of environmental and
human health: Vanadium. Winnipeg.
CANNON HL. 1963. The biochemistry of vanadium. Soil Sci. 96. 196-204.
CEBALLOS H, PANDEY S, KNAPP EB, DUQUE J. 1995. Progress from selection
for tolerance to soil acidity in five tropical maize populations. In DATE RA,
GRUNDON NJ, RAYMENT GE, PROBRT ME. (Eds.) Plant Soil
Interaction at Low pH: Principles and Management. Kluwer Academic
Publishers, Dortrecht, The Netherlands. pp. 419-424.
CHAN KY, HEENAN DP. 1998. Effect of lime (CaCO3) application on soil
structural stability of a red earth. Aust. J. Soil Res. 36. 73-86.
CHANEY RL, RYAN JA, KUKIER U, BROWN SL, SIEBIELEC G, MALIK M,
ANGLE JS. 2001. Heavy metal aspect of compost use. In STOFELLA PJ,
KAHN BA. (Ed) Compost utilization in horticulture cropping systems. CRC
Press, p 323-359.
69
CLARK AM. (1990). Hey’s mineral index. London, Chapman & Hall.
COVENTRY DR, SLATTERY WJ, BURNETT VF, GANNING GW. 1997. Liming
and re-liming needs of acidic soils used for crop-pasture rotations. Aust. J.
Exp. Agric., 37. 577-581.
CRANS DC, TRACEY AS. 1998. The chemistry of vanadium in aqueous and
nonaqueous solution. Am. Chem. Soc., 711. 2-29.
CRANS DC. 1994. Aqueous chemistry of labile oxovanadates: relevance to biological
studies. Comments Inorg. Chem., 16. 1-13.
CRUYWAGEN JJ, HEYNS JB. 1991. Vanadium (V) equilibria. Spectrophotometric
and enthalpimetric investigation of the dimerization and deprotonation of
HVO4-. Polyhedron. 10. 249 - 253.
DME (Department: Mining and Energy, South Africa). 2005. Dolomite and limestone
in South Africa: Supply and demand. Report R49/2005.
EBDON
L,
PITTS
L,
CORNELIS
R,
CREWS
H,
DONARD
OFX,
QUEVAUVILLER PH. 2001. Trace element speciation for the environment.
Food and Health. The Royal Society of Chemistry. Cambridge.
EKSTEEN LL. 1969. The determination of the lime requirement of soils for various
crops in the winter rainfall region. Fert. Soc. of S. Afr. J. 2. 13-14.
EVANS HT, LANDERGEN S. 1978. Vanadium. In WEDEPOHL KH. (Ed),
Handbook of Geochemistry Springer-Verlag, Berlin, Vol.II/2, Section 23.
FOOD AND AGRICULTURE ORGINAIZATION. 2011. "The State of Food
Insecurity
in
the
World
2011"
http://www.fao.org/docrep/013/i1683e/i1683e.pdf
70
FEY MV, DODDS HA. 1998. Classifying the sensitivity of soils of the South African
highveld to acidification. S. Afr. J. Plant Soil. 15. 99-103.
FLEISCHER M. 1987. Glossary of Mineral Species. Tucson, Arizona, Mineralogical
Record, Inc.
FÖLSCHER WJ. 1975. Die toestand van suur grond. Fert. Soc. of S. Afr. J., 2. 53-55.
FOY CD. 1974. Effects of aluminium on plant growth. In CARSON E. W., (Ed.). The
plant root and its environment. Univ. Press of Virginia, Charlottesville,
Virginia, USA.
GOODMAN BA, CHESIRE MV. 1975. The bonding of vanadium in complexes with
humic acid: an electron paramagnetic study. Geochim. Cosmochim. Acta 39.
1711.
GHOST. 1985. Nature protection. Soils. Methods of sampling and preparation of soil
samples for chemical analyses. Moscow.
HAYNES RJ, NAIDU R. 1998. Influence of lime, fertilizer and manure applications
on soil organic matter content and soil physical conditions: a review. Nutr.
Cycl. Agroecosyst. 51. 123-137.
HELYAR KR, PORTER WM. 1989. Soil acidification, its measurements and the
process involved. In ROBSON AD. (Ed) Soil acidity and plant growth..
Academic Press, Sydney. p 61-100
HEMPHILL DD. 1972, Availability of trace elements to plants with respect to soilplant interaction. Annals New York Acad. Sci. 199. 46-61.
HIDALGO A, NAVAS P, GARCIA-HERDUGO G. 1988. Growth inhibition induced
by vanadate in onion roots. Environ. Exp. Bot. 28. 131-136.
71
HOODA PS, McNULTY D, ALLOWAY BJ, EITKEN MN. 1997. Plant availability
of heavy metals in soils previously amended with heavy applications of
sewage sludge. J. Sci. Food Agric., 73. 446-454.
HUNTLEY B, SIEGFRIED R, SUNTER C. 1989. South Africa environments into the
21th century. Human and Rousseau, Tafelberg.
JUO ASR, DABIRI A, FRANZLUEBBERS K. 1995. Acidification of a kaolonitic
Alfisol under continuous cropping with nitrogen fertilization in West Africa.
Plant Soil 171. 245-253.
KABATA-PENDIAS A, PENDIAS H. 1992. Trace elements in soil and plants. 2nd
Ed. CRC Press, Boca Raton, FL.
KABATA-PENDIAS A, PENDIAS H. 1993. Biochemistry of trace elements. PWN,
Warsaw Poland.
KAPLAN DI, ADRIANO DC, CARLSON CL, SAJWAN KS. 1989. Vanadium:
Toxicity and accumulation by beans. Biochem. Div. Savannah River
Ecology Laboratory, Drawer E, Aiken, SC2980 1, U.S.A.
KAPLAN DI, SAJWAN KS, ADRIANO DC, GETTIER S. 1990. Phyto-availability
and toxicity of Beryllium and Vanadium. Water Air and Soil Pollut.. 53.
203-212.
KEENEY-KENNICUTT WL, MORSE JW. 1984. The interaction of Np(V)O2+ with
common mineral surfaces in dilute aqueous solutions and seawater. Mar.
Chem. 15. 133-150.
LEPP NW. 1977, Z. The effect of vanadium on germination and seedling growth of
lettuce. Pflanzenphysiol. Bd. 83 185-188.
72
LØBERSLI E, GJENGEDAL E, STEINNES E. 1991. Impact of soil acidification on
the mobility of metals in the soil plant system. In VERNET JP. (Ed) Heavy
metals in the environment. Elsivier, p 37-53.
LOGAN TJ. 1992. Reclamation of chemically degraded soils. In Advances of soil
science. Springer-verlag, New York. 17. 13-35.
MANDIWANA KL, PANICHEV N. 2004. Electrothermal atomic absorption
spectrometric determination of vanadium (V) in soil after leaching with
Na2CO3. Anal. Chim. Acta 517. 1247-1250.
MANDIWANA KL, PANICHEV N, MOLATLHEGI R. 2005. The leaching of V(V)
with PO43- in the speciation analysis of soil. Anal. Chim. Acta 545. 239-243.
MARTIN WH, KAPLAN DI. 1996. Temporal changes in Cadmium, Thallium and
Vanadium mobility in soil and phyto-availability under field conditions.
Water, Air, and Soil Pollut. 101. 399-410.
MORREL BG, LEPP NW, PHIPPS DA. 1986. Vanadium uptake by higher plants:
Some recent developments. Environ. Geochem. Health. 8. 14-18.
NRIAGU JO. 1998. Vanadium in the environment, Part1, Chemistry and
Biochemistry, John Wiley and Sons, Inc., New York.
OLNESS A, PALMQUIST D, RINKE J. 2001. Ionic ratios and crop performance. II.
Interactions amongst V, P, Mg and Ca with soybean yield. J. Agron. Crop.
Sci. 187. 47-52.
PANICHEV N, MANDIWANA K, MOEMA D, MOLATHLEGI R, NGOBENI P.
2006. Distribution of vanadium (V) species between soil and plant in the
vicinity of vanadium mine. J. of Hazard. Mat., 137. 649-653.
73
PETERSON PJ, GIRLING CA. 1981. Other trace metals, In LEPP NW. (Ed). Effect
of heavy metal pollution on plants. Vol. 1. Effects of trace metals on plant
function, Applied Science Pub. London, p 213.
PINKERTON A, SIMPSON JR. 1986. Aluminium effects on the growth and mineral
composition of corn plants cultivated in nutrient solution at low aluminium
activity. J. Plant. Nutr. 19. 729-741.
POLEDNIOK J, BUHL F. 2002. Speciation of vanadium in soil. Department of
Analytical Chemistry, University of Silesia.
RAI D, ZACHARA JM. 1984. Chemical attenuation rates, coefficients, and constants
in leachate. Migration. Volume 1: A Critical Review, EA3356. Electric
Power Research Institute, Palo Alto, CA. p. B17.
REHDER D. 1999. The coordination chemistry of vanadium is related to its
biological functions. Coord. Chem. Rev. 182. 297-322.
RITCHIE GS. 1989. The chemical behaviour of aluminium, hydrogen and manganese
in acid soils. In Soil acidity and plant growth. Academic Press, New York. p
1-60.
ROWELL DL, WILD A. 1985. Causes of soil acidification: a summary. Soil Use and
Management 1. 32-33.
SCOTNEY DM, VOLSCHENK JE, VAN HEERDEN PS. 1990. The potential and
utilization of the natural agricultural resources of South Africa. Dept. Agric.
Develop., Pretoria, South Africa.
SIGG L. STUMM W. 1980. The interactions of anions and weak acids with hydrous
goethite (α-FeOOH) surface. Colloids Surf. 2. 101-117.
74
SINGH U, UEHARA G. 1986. Electrochemistry of the double layer: Principles and
applications to soils. In Soil Physical Chemistry. CRC Press, Boca Raton,
FL. p 1-38.
SKLENAR I, FOX GG, LOUGHMAN BC. 1994. Effect of vanadate on the ATPase,
activity and phosphate absorption capacity of maize roots. Plant Soil. 167.
57-62.
SOMASUNDARAM R, MUTHUCHELIAN K, MURUGESAN S. 1994. Inhibition
of chlorophyll, protein, photosynthesis, nitrate reductase and nitrate content
by vanadium in Oryza sativa L. J. Environ. Biol.15. 41-48.
SPARKS DL. 1986. In Soil Physical Chemistry. CRC Press, Boca Raton, FL.
SPOSITO G. 1984. In The surface chemistry of soils. Oxford University Press, New
York.
STURGEON RE. 2000. Current practice and recent developments in analytical
methodology for trace element analysis of soils, plants, and water. Commun.
Soil. Sci. Plant Anal. 31. 1479-1512.
SUMNER ME. 1975. Soil acidity and its agronomic implications. Fert Soc. S. Afr. J.
2. 47-51.
SUMNER ME, FEY MV, NOMBLE AD. 1991. Nutrient Status and Toxicity in Acid
soils. In ULRICH B, SUMNER ME. (Ed) Soil Acidity. Springer-Verlag,
Berlin Heidelberg, p 149-192.
TESSIER A, CAMPBELL PGC, BISON M. 1979. Sequential extraction procedure
for the speciation of particulate trace metals. Anal. Chem. 51. 844-851.
TREFRY JH, METZ S. 1989. Role of hydrothermal precipitates in the geochemical
cycling of vanadium. Nature 342. 531-533.
75
USEPA (United States of America Environmental Protection Agency) 1987. Health
effects assessments for vanadium and compounds. Environmental Criteria
and Assessment Office, Office of Health and Environmental Assessment,
Office of Research and Development, Cincinnati, OH. PB88-176383.
VAN DER WAALS JH. 2001. Slags as agricultural limes: Reactivity and heavy metal
and phosphorous bio-availability. University of Pretoria.
VAN DER WAALS JH, CLAASSENS AS. (UNPUBLISHED), Long-term heavy
metal pollution risk due to the use of slag as agricultural limes on the South
African Highveld. University of Pretoria.
VAN ZINDEREN BAKKER EM, JAWORSKI JF. 1980. Effects of vanadium in the
Canadian environment. National Research Council of Canada, Associate
Committee on Scientific Criteria for Environmental Quality, Subcommittee
on Heavy Metals and Certain Other Compounds. NRCC No. 18132.
VENTER HJ. 2004. Lime quality and soil acidity criteria for dry land maize.
University of the Free State.
VOUK V. 1979. Vanadium In Handbook on the toxicology of metals, Friberg et. al
.Elsivier North Holland. Biomedical press pp. 659 – 674.
WALLACE A, ALEXANDER GV, CHAUDHRY FM. 1977. Phytotoxicity of
cobalt, vanadium, titanium, silver and chromium. Commun. Soil Sci. Plant
Anal. 8. 751-756.
WANG JF, LIU Z. 1999. Effect of vanadium on the growth of soybean seedlings.
Plant & Soil 216. 47-51.
76
WANTY RB, GOLDHABER HB. 1992. Thermodynamics and kinetics of reactions
involving vanadium in natural systems: Accumulation of vanadium in
sedimentary rocks. Geochim. Cosmochim. Acta 56. 1472-1438.
WEHRLI B, STUMM W. 1989. Vanadyl in natural waters: Adsorption and hydrolysis
promote oxygenation. Geochim. Cosmochim. Acta 53. 69-77.
WELCH RM. 1973. Vanadium uptake by plants. Plant Physiol. 52. 183-185.
WHO (World Health Organization) 1988. Vanadium. Environmental Health Criteria
81. Geneva.
WILKISON RE, DUNCAN RR. 1993. Vanadium influence on calcium (45Ca2+)
absorption by sorghum root tips. J. Plant Nutr. 16. 1991-1994.
YUDOVICH YE. 1972. Mean trace element contents in coal. Geochemistry
international. 9. 712-720
ZEIGLER RS, PANDEY S, MILES J, GOURLEY LM, SARKARUNG S. 1995.
Advances in the selection and breeding of acid tolerant plants: Rice, maize,
sorghum and tropical forages. In DATE RA, GRUNDON NJ, RAYMENT
GE, PROBERT ME. (Eds.). Plant Soil Interactions at Low pH: Principles
and Management. Kluwer Academic Publisher, Dordrecht, The Netherlands.
p 391-406.
77
APPENDIX A: Google maps of the Ogies and Delmas
evaluation sites
FIGURE A.1. A Google Map of the Delmas field evaluation site
78
FIGURE A.2. A Google Map of the Ogies field evaluation site
79
APPENDIX B: The frequency distribution and cumulative
normal distribution curves for the different soil and soil
depths.
FIGURE B1: The total V concentration frequency distribution of the 30-60 cm soil
samples at the Delmas site, without the control samples
FIGURE B2. The total V cumulative normal distribution curve for the 30-60 cm soil
samples of the Delmas site.
80
FIGURE B3. The total V concentration frequency distribution of the 30-60 cm soil
samples at the Ogies site, without the control samples
FIGURE B4. The total V cumulative normal distribution curve for the 30-60 cm soil
samples of the Ogies site.
81
FIGURE B5. The Bray 1 extractable V frequency distribution of the 30-60 cm soil
samples at the Delmas site, without the control samples.
FIGURE B6. The Bray 1 extractable V cumulative normal distribution curve for the
30-60 cm soil samples of the Delmas site.
82
FIGURE B7. The Bray 1 extractable V frequency distribution of the 30-60 cm soil
samples at the Ogies site, without the control samples.
FIGURE B8. The Bray 1 extractable V cumulative normal distribution curve for the
30-60 cm soil samples of the Ogies site
83

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project