Tải bản đầy đủ - 0trang
II. Problems of Conventional Soil Testing Practice
EUF DETERMINATION OF SOIL NUTRIENTS
red relict soil ( 2 6 . 2 mg exchangeable K )
A gray brown luvisol ( 2 3 . 8 rng exchangeable K )
FIG. 2. Differences in the decrease in K conconcentration in the soil solution and in the yield
depressions of ryegrass in two different soils.
problematic. If strong solutions are used for extraction, both the effectively as
well as the potentially available fractions will be obtained together (for example,
20 mg of K per 100 g of soil). It is possible that up to 80% of this amount is
effectively available. But even if the effectively available portion is only 20%,
this level is not without importance for the K uptake rate of the plant.
Therefore, it is reasonable to expect from soil analysis that the ionic concentration in the soil solution should be measured, as the soil solution is the nutrient
solution of the plant under natural conditions. There are several methods of
obtaining the soil solution, as described by Briggs and McCall (1904), Van Zyl
( 1916), Lipmann ( 1918), Northrup ( 1918), Parker ( 1921) , von Wrangell ( 1930),
Magistad et al. (1945), and Adam and Winsor (1973). Being very timeconsuming, these methods are not suitable for routine investigations. Moreover,
the ionic concentration in the soil solution is only a “momentary value.” It remains unknown how this value changes in the course of the vegetation period.
Yet this is a highly important factor for the nutrient uptake of the plant.
An example is given in Fig. 2. This figure shows the differences in the
decrease in the K concentration in the soil solution and in the depressions in yield
of ryegrass in two different soils during the vegetation period. Although the
contents of exchangeable K are almost equal, the K concentration in the soil
solution decreases more rapidly in the red relict soil than in the gray-brown
luvisol. This can be ascribed to the different mineral composition of the clay
fractions. The clay minerals of the red relict soil belong to the kaolinitic group
and have a low capacity to release K into the soil solution, whereas the illitic clay
minerals of the gray-brown luvisol are capable of keeping the K concentration in
the soil solution relatively stable up to a certain level. For this soil the yield
K . NhMETH
depression rate is consequently very slow. The contents of exchangeable K
alone, therefore, cannot indicate the expected change in the K concentration in
the soil solution during the vegetation period.
It is even more difficult to determine the amount of K fertilizer that has to be
applied in order to raise the K concentration in the soil solution by a determined
value (for example, from 0.2 to 1.O me/l). Figure 3 shows clearly the considerable differences in the increase in the K concentration in the soil solution of the
various soils at equal K fertilizer levels, although the contents of exchangeable K
were practically the same (9-10.5 mg per 100 g). The contents of exchangeable
K are consequently not sufficient information for calculating the K fertilizer
Figure 3 furthermore shows that the K fertilizer requirements can be very
different even at comparable clay contents (1 9-2 1%) and almost equal K concentrations (0.2-0.3 me/l) in the soil solution, if the mineral composition of the clay
fraction is different.
In agricultural practice, soil analysis should therefore provide the following
(1) What is the nutrient concentration in the soil solution, or what is the
effectively available amount of nutrients in the soil?
(2) Which of the changes of these effectively available amounts in the course
of the vegetation period are due to nutrient removal, leaching, weathering, etc.?
(3) What is the amount of nutrients to be applied to the soil in order to raise
the effectively available amount to the required value (fertilizer requirement)?
It will be the objective of this paper to explain how the requirements specified
above can be achieved.
A. COMPONENT PROCESSES OF ELECTRO-ULTRAFILTRATION
1. Dialysis, Ultrafiltration, and Electrodialysis
To provide information on the processes taking place in electro-ultrafiltration,
the component processes, which have been actually known for a long time, will
first be described.
A century ago, dialysis had already been used for the separation of ions from
colloids (for example, clay minerals). This method is based on the principle of
diffusion along a concentration gradient through a semipermeable membrane.
EUF DETERMINATION OF SOIL NUTRIENTS
red relict soil ( 19 % clay)
gray brown luvisol(21% cloy)
gray brown luvisol (3174 clay)
alluvial soil (38 7" clay)
K fertilizatian (kglha)
FIG. 3. Relationship between the K Concentration in rhe soil solution and the amounts of K
fertilizer of soils with equal contents of exchangeable K (9-10 mg per 100 g of soil)
The process is very slow, but it can be accelerated by increasing the surface of
the membrane, the concentration gradient, and the temperature.
A second method of separating plant nutrients from soil is based on the principle
of ultrafzltration. This is a filtration process by means of which the soil colloids
are collected on a filter and the sorbed ions are removed by leaching. This method
is also time-consuming, because the filtration rate decreases considerably with
increasing dispersity of the soil.
After the development of electrodialysis by Morse and Pierce (cited by
Bechold, 1925) in 1903, ion diffusion through membranes could be accelerated
considerably. This method could also be used for soil tests (Konig et al.,
1913; Mattson, 1926; Norman et ul., 1927; Rost, 1928; Bradfield, 1928). It has
been found that the various soils release very different amounts of ions. Investigations with synthetic permutite and feldspar powder showed that mainly exchangeable ions were detected by electrodialysis. The feldspar powder released
very small amounts of cations (Bradfield, 1928). Mattson (1926) as well as
Bradfield (1928) demonstrated that the exchangeable cations can be determined
quantitatively. The time of dialysis, however, amounted to 20-50 hours.
2. Electro-ultrafiltrationat Constant Voltage and
Temperature during the Extraction Process
In addition to requiring a considerable expenditure of time, there were undesirable effects involved in electrodialysis;of these, the decrease in pH during the
extraction proved to be most important. To overcome this undesirable decrease in
pH, Bechold ( 1925) suggested combining ultrafiltration with electrodialysis. He
called the new method electro-ultrafiltration. It differs basically from electrodidysis in that the secondary products of the electrodialysis (hydroxides and
acids) are removed by suction and cannot enter the middle cell, so that the change
in pH is reduced to the minimum.
The electro-ultrafiltration equipment used by Bechold and Konig was improved by Kottgen in the thirties and forties and used for soil tests (Kottgen and
Diehl, 1929; Kottgen, 1937, 1940; Kottgen and Jung, 1941; Jung and Nemeth,
1966, 1969). More recently, Grimme (1978, 1979) has worked with constant
voltage with the objective of evaluating the kinetics of nutrient desorption.
B . ELECTRO-ULTRAFILTRATIONWITH VARYING
VOLTAGE AND TEMPERATURE DURING THE EXTRACTION
Once the significance of the soil solution in nutrient transport (diffusion and
mass flow) to the plant root had been recognized, means of measurement were
sought that would allow the nutrient concentration in the soil solution and its
buffering to be measured in routine analysis. A further development of the EUF
procedure appeared to be suitable for this purpose.
1 . Principle of the Method
When an electrical potential is applied to a soil suspension, the following
reactions can occur at the cathode:
2Na+ + 2e 4 2Na0
2Na0 2Hz0 -+ 2NaOH
+ H2 + 67.4 kcal
Similar reactions are possible for K, Mg, Ca, etc.
At the anode, where oxidative processes take place, the following reactions
NO5 - e = [NO,]
[NO,]. + HzO + NO5 a [OH]
2[OH]* = HZOz + HZO - HOp
However, the processes taking place at the electrodes at high voltages (200 and
400 V) and different temperatures need further investigation. More details are
given by Nemeth (1972, 1976).
2 . EUF Apparatus and Extraction .
A three-cell apparatus is used in electro-ultrafiltration. The middle cell containing the soil suspension (soi1:water = 1:lO) has a stirrer and a water inflow
EUF DETERMINATION OF SOIL NUTRIENTS
FIG. 4. Part of the automatic EUF equipment.
(Fig. 4). Each side of this middle cell is provided with a micropore filter attached
to the platinum electrodes that separate the middle cell from the two outside
compartments (Fig. 5). These two other cells have vacuum connections. Therefore, the hydroxides [Na(OH), K(OH), Ca(OH),, NH,(OH), etc.] accumulating
at the cathode and the acids (HN03, HzS04etc.) accumulating at the anode are
washed away by the continuous stream of water to the collecting tanks. The
FIG.5. Application of the micropore filters to the platinum electrodes.
procedure is regulated in such a way as to obtain fractions in 5 min or other
intervals. In contrast to the situation that prevails with electrodialysis (Konig et
al., 1913; Mattson, 1926; Norman et al., 1927; Bradfield, 1928), the pH
of the soil suspension remains constant during EUF extraction. This is very
important, because the pH exerts a decisive influence on the desorption and
The negatively charged clay minerals of the soil migrate to the anode and are
deposited on the anode filter (Fig. 6). The water permeability of the anode filter
is reduced by the accumulation of clay. The amount of water flowing through this
filter is therefore inversely proportional to the clay content, so that the latter can
be determined indirectly, as indicated in Fig. 7 (Nemeth, 1976). The clay content
determined by the conventional method is plotted on the ordinate. The abscissa
indicates the flow of water through the anode during the period from 10 to 35
min. As can be seen, the correlation is very close, and thus a rapid determination
of the clay content fof;practical purposes is made possible, which is necessary for
the assessment of fertilizer requirements.
In contrast to Ca, K, Na, etc., the heavy metals (Cu,Cd, Fe, Mn, Ni, Zn, etc.)
are collected by the cathode filter as hydroxides and hydrated oxides (Fig. 6).
Thus, a seperation can be made between the hydroxides of Ca, K, Na, etc., and
the hydrated oxides of heavy metals. (Mg has an intermediate position; see
Section III,C,3 .) Heavy metals determined by EUF represent their effectively
EUF DETERMINATION OF SOIL NUTRIENTS
FIG. 6. Clay minerals accumulated at the anode filter and heavy metals collected at the cathode
plant-available fractions, since only soluble or desorbed ions can move in an
For practical purposes, an extraction duration of 35 min is adequate (seven
fractions of 5 min each). As a result of intensive research and comparison with
conventional soil testing methods, the voltage is changed as follows: 0-5 min,
50 V ( I 1.1 V/cm); 5-30 min, 200 V (44.4V/cm); 30-35 min, 400 V (88.8 V/cm).
In routine investigations three fractions are sufficient: EUF, collected within
71.9 - 0 . 5 5 ~
amount of water Iml)
FIG. 7. Correlation between the clay content and the amount of water that flows through the
anode filter during 10-35 min.
0-10 min, EUFIIcollected within 10-30 min, EUFII, collected within 30-35 min.
The voltage is varied, as stated above. On the other hand, for research purposes it
is more appropriate to obtain the fractions at 5-min intervals in order to get more
information on the rate of supply of nutrients from the soil.
3 . Determination of Nutrients in the EUF Fractions
The determination of nutrients in the EUF fractions can be carried out by
conventional methods. For serial tests, a combination with the AutoAnalyzer has
proved useful. The determination of K, Na, and Ca can be carried out with an
emission spectrophotometer, and that of Mg, Mn, Zn, Fe, etc., by means of an
atomic absorption spectrophotometer, whereas B, PO4, SO4, NO3, NH4 can be
determined colonmetrically. NH4 and NO3 can also be determined by the
4 . Presentation of Results
The quantities determined by EUF in 5-min intervals can be plotted against
time (duration of extraction), so that the amount desorbed (milligramsof nutrient
per 100 g of soil per 5 min) appears on the ordinate, and the duration of desorption and the voltages applied can be read off from the abscissa.
Figure 8 shows two possibile ways of presenting the results. Example
(a) shows the values in the form of a curve, and example (b) shows these values
in the form of columns. In routine tests (three or two fractions), the values are
given in the form of columns or are listed in a table, which is even simpler. From
desorplion lime in minules f I I
desorplion lime in minutes f / I
FIG. 8. Representation of the EUF values at varied voltages (using the nutrient K as an example).
EUF DETERMINATION O F SOIL NUTRIENTS
( a ) 10ppm
_ _ _ 200v
time in minutes ( t I
time in minutes ( 1 )
FIG. 9. Quantities of phosphorus transported by means of EUF from a solution at different
voltages and different P concentrations.
the shape of the curves, however, information on the desorption and solubility
rates can also be obtained.
5 . Significance of Altering the Voltage
The speed of ion migration in electro-ultrafiltration (in the middle cell) is
proportional to the field strength and inversely proportional to the frictional
forces. In turn, the field strength depends on the voltage and the distance between
the electrodes. Since the latter is kept constant, the field strength is raised with an
increase in voltage, and the ions are transported more rapidly to the electrodes.
This is illustrated by Fig. 9, which shows the quantities of phosphorus removed
from a solution by means of EUF at different voltages (200 and 400 V) and
various concentrations. The transported amounts, in parts per million per 5 min,
are plotted on the ordinate, and the duration of extraction can be read off from the
abscissa. It is shown clearly that the phosphate ions are removed more rapidly
from the middle cell of the EUF apparatus with increasing voltage. The difference in the transport rate due to raised voltage is greater, the lower the ion
concentration in the middle cell.
It can further be seen from Fig. 9 that the transport of ions at a given voltage
is more rapid, the higher their concentration. At a P concentration of 10 ppm,
larger quantities of P, compared with the total amount, are transported to the
electrodes in the time unit (for example, 5 min) than are transported at a P concentration of 2 ppm.
The rate of ion migration in electro-ultrafiltration is therefore not constant at
a given voltage and temperature. In the course of electro-ultrafiltrationit decreases
to such an extent that the concentration in the middle cell is lowered. At low
concentrations, the rate of transfer attains an almost constant state in which the
field strength and the frictional forces maintain an equilibrium (Fig. 9b, 200 V).
In order to accelerate the transport of ions, the voltage must therefore be raised.
Relatively high voltages must be applied in order to raise the field strength
(volts per centimeter) significantly, since the distance between the electrodes must
be 4.5 cm in order to have enough space for the soil sample in the middle cell.
The attainment of an equilibrium between field strength and frictional forces is
more complicated in the soil than in a solution. The attainment of constant rates
of extraction is counteracted by the release of ions from the soil reserves. The
faster the release from the reserve and the higher the voltage, the higher are the
rates of extraction (Fig. 10).
Figure 10 shows the rates of P extraction in milligrams per 100 g per 5 min for
two soils-(a) eroded gray-brown luvisol and (b) pararendzina-with equal P
concentrations in the soil solution (in the saturation extract). At 50 V, these
EUF-P values are low, but comparable, for the two soils. The P availability is
accordingly well-characterized at 50 V, since the P concentration in the soil
solution of the two soils is likewise equal. It could be concluded from this result
that the soils should be investigated at 50 V, since the concentration is characterized well and their buffering could also be registered after a long time of
extraction. Unfortunately, the rates of extraction of most nutrients are very low
(hardly measurable) at 50 V, so that a long extraction would be required to record
the buffering as well.
sail lbl pH 7.3
soil lo1 pH 6.6
soil lo1 and ( b ) by three
extraction time in minuter (11
extroclion time in minutes f t )
erlraction time in minufes I t )
FIG. 10. Amounts of P extracted by means of EUF from two soils with equal P concentrations
(1.5 ppm) in the soil solution at constant and varied voltages.
EUF DETERMINATION OF SOIL NUTRIENTS
At 200 V the rates of P extraction are considerably higher, but the P curves of
the two soils are still almost the same. However, a more constant supply becomes
noticeable in soil (b), whereas a slight falling off is noticed in soil (a).
The amounts of P extracted from the two soils are entirely different at 400 V.
The rates of desorption from soil (b) are almost twice as high as those of soil (a),
although the P concentrations in the soil solution of the two soils were comparable. Hence, the P concentration in the soil solution is not correctly assessed at
400 V. On the other hand, the P reserves are not registered at 50 V within 35
min. The difference between the two soils at 50 V is evident only after an
extraction for several hours.
By raising the voltage stepwise (50 V , 200 V , and 400 V), both the P concentration in the soil solution and its buffering can be characterized in one operation
in which the duration of extraction required for practical purposes is only 35 min.
This is the most important advantage of varying the voltage.
6. Significance of Altering the Temperature
The speed of ion migration in the electrical field is raised with an increase in
temperature, so that more nutrients can be extracted per unit of time. Care has to
be taken not to confound this increase in the speed of ion migration through the
increase in temperature with high amounts of effectively available nutrients in the
analyzed soil. It is, therefore, advisable to keep the temperature below 25°C
during extraction by EUF. Higher temperatures during extraction can, on the
other hand, be suitable for the determination of larger quantities from the nutrient
reserves (from the potentially available reserves), thus giving a more precise
insight into the nutrient dynamics of the soil.
Therefore, the EUF procedure was modified by varying the temperature during
the desorption processes. The modified procedure is as follows:
A 20- to 30-min extraction at 200 V and 20°C is suggested for the determination of the effectively available nutrient fractions. Subsequently the temperature
is raised from 20°C to 80"C, the voltage is increased from 200 V to 400 V, and
the extraction is maintained for another 10-15 min. The effectively available
amount of nutrients and the nutrient reserves can be characterized more precisely
when the fractions at 20°C and 80°C are collected at 5-min intervals. Reliable
information will thus be obtained on the rate of supply from the reserves.
Table I shows the K and P desorption of five soils at different K fertilizer
levels (K, and K4)as influenced by voltage and temperature. From this table, it
can be gathered that considerable amounts of K and P are extracted when the
temperature is raised from 20°C to 80°C and at the same time the voltage is
increased from 200 V to 400 V. The K amounts in the second fraction (80°C) are
for many soils higher than the effectively available K contents in the first fraction
(20°C). These amounts are higher, the higher the contents of K-selective clay
minerals in the soil.