Tải bản đầy đủ - 0trang
V. Nature of Organic Soil-Binding Suhstances
SOIL A G G K E G x l l O N
original material, dead microbial cells, or secondary decomposition
4. Organic binding substances synthesized by the soil organisms.
Which of these are the most important is a matter of conjecture,
but the last two are probably very important if not the most important.
Geltzer (1937) came to the conclusion that during the decomposition
of organic matter in the soil there is an accumulation of synthetic
microbial substances which bring about the binding of soil particles
into aggregates. Peele (1940) demonstrated that bacterial mucus from
several species produced water-stable aggregates when incorporated
with the soil.
I . Pol ysacchurides
In a study designed to determine the nature of soil-binding substances synthesized by soil organisms, a polysaccharide of the levaii
type produced by Bacillus subtilis was found to be effective binding
material (Martin, 1945). I n a continuation of this study (Martin,
1946), several bacterial polysaccharides were found to be very effective
binding agents (see Table 11). As little as 0.1 g. of one material in 100
Effect of Bacterial Polysaccharides on Aggregation of Declo Loam’
of ( 5 0 - r
Fructosan froiii Amtobarter intlic.tirrr
Dextran froill a soil bacteriuiii No. 1
L)extmn from a soil bacteriutn No. 2
Dextran froiu Leuconostor de.rtranicz~rrr
JAMES P. MARTIN
g. soil bound 23 g. of dispersed silt plus clay into water-stable aggregates larger than 50 p in diameter. Related studies carried out in England by Geoghegan and Brian (1946, 1948) demonstrated the marked
binding action of microbial polysaccharides. In the first report, a relationship between the nitrogen content of the polysaccharide and its
aggregating effect was indicated, namely, the preparations containing
0.2 to 0.3 per cent or more of nitrogen were more effective than those
rontaining less than 0.1 per cent. Later (1948) this was shown to be
due to a degradation of the material during purification.
The investigators who have reported the influence of microbial
polysaccharides on soil binding have emphasized that these compounds
are undoubtedly not the only materials produced through microbial
decomposition processes which aid in soil aggregation, but that they are
probably important because similar type compounds are apparently
present in soil humus. It has been estimated that from approximately
5 to 20 per cent of the soil organic matter consists of polysaccharide substances, primarily of the polyuronide type (Norman and Bartholomew,
1943; Shorey and Martin, 1930). The fact that marked aggregation of
soils results from applications of polysaccharide concentrations which
are less than that apparently found in many soils suggests that these
materials may play an important function in soil granulation.
The polysaccharide fraction of the soil could be derived from plant
polysaccharides, microbial polysaccharides, or both. During the decomposition of plant residues in composts, the cellulose fraction almost
completely disappears, whereas a large amount of other types of polysaccharides, primarily polyuronides, remains in the residue (Waksman,
1938). It was suggested that some of the plant and synthesized microbial
polyuronides are somewhat resistant to decomposition and therefore
persist in the residue. On the basis of decarboxylation rate curves of
soil organic matter and bacterial gums, Fuller (1946, 1947) suggested
a microbial origin for the soil uronides.
Bremmer ( 1950) believes that the estimates of polyuronides in the
soil are impossibly high and is of the opinion that the method for estimating soil uronic carbon, which involves prolonged boiling with 12 per
cent HCl, splits off carbon dioxide from other soil constituents. I n this
connection, Broadbent (1953) points out that the carbon content of the
organic matter of most surface soils is approximately 50 per cent,
whereas that of pure uronide is 40.9 per cent. Any increase in the
polyuronide fraction would lower the carbon content of the organic
fraction. Actually, the carbon content of the soil organic matter decreases with depth, whereas the polyuronide content estimated by the
usual procedures increases; this is indirect evidence that the procedure
has merit. Broadbent further states that the exchange capacity of the
organic matter increases with depth, indicating more acidic groups.
Recently, Forsyth ( 1950) isolated two polysaccharides containing
uronide constituents from soil organic matter. Stevenson et al. (1952)
found polysaccharide components including galacturonic acid in the
hydrolyzate of organic colloid from soil. These findings further support the belief that the polysaccharides constitute an important fraction
of the soil organic matter and could, therefore, contribute to soil aggregation under field conditions. Additional evidence of the possible importance of the soil polysaccharides was obtained by Swaby (1950),
who noted that the binding action of humus extracts was destroyed by
acid or alkaline hydrolysis; this suggests the importance of proteins,
polysaccharides, or both.
Most plant and microbial polysaccharides are subject to rapid attack
by soil organisms (Norman and Bartholomew, 1940; Martin, 1946).
Their persistence in the soil has been attributed to a possible combination with other soil constituents including clays which render them
more resistant to microbial attack (Norman and Bartholomew, 1943;
Martin, 1946; Fuller, 1947).
2. Other Organic Substances
I n addition to microbial polysaccharides, some plant polysaccharides,
certain modified lignins, proteins, oils, fats, and waxes, which are rela ted in chemical composition to microbial decomposition products, or
synthesized compounds have been found to increase the stability of soil
aggregates. Alginates have been tested by Hedrick and Mowry (1952),
Quastel ( 1952), and others. Geoghegan (1950) found pectin and alginic
acid to be effective bindings agents in an acid soil but not in soil saturated with sodium or calcium. McCalla (1950) reported that egg
albumin, casein, and certain oils, fats, waxes, and resins increased
structural stability but other proteins and various carbohydrates did
not. In a study by Martin (1946), certain microbial polysaccharides
were found to be better aggregate stabilizers than were white pine
lignin o r casein, although the lignin and casein produced a rather
marked binding action. In this study, it was necessary to get the l i p i n
in solution before it would bind the soil particles. Oxidation of humus
extracts (Swaby, 1950) with hypoiodite, which is supposed to destroy
lignin-like compounds but not polysaccharides, reduced the binding
action of the extracts. This provides indirect evidence that lignh-like
colloidal materials contribute to the binding action of soil humus.
More work is needed to evaluate the contribution of a variety of
natural organic substances to soil aggregate stability and to determine
the nature of the clay-organic matter complex.
VI. SYNTHETICSOIL CONDITIONERS
1 . Nature of Materials Used
The discovery that soil microorganisms synthesized polysaccharides
and other compounds which enhanced soil granulation stimulated the
search for synthetic compounds which would act in a similar manner
to the natural products but would persist for longer periods of time in
the soil. The alginates, which are similar to some bacterial polyuronides,
were used first for structural improvement (Quastel and Webley, 1947;
Hedrick and Mowry, 1952; Quastel, 1952, 1953) but with only fair
success. Several tons of material per acre were required, and it was so
readily decomposed that it lasted for only a short time in the soil; in
addition it reduced the available nitrogen content to deficiency levels.
The silicates of potassium and sodium have been tested by Dutt
(1947), Laws and Page (1946), Raney (1953), and others; waterproofing chemicals such as stearic and abietic acid, by McCalla (1946b)
and Winterkorn et al. (1945) ; and volatile and water-soluble silicones,
by Van Bavel (1950). All have increased soil granulation, but they are
currently not being used in the field except possibly for certain engineering purposes because usually high rates of application are needed,
with resulting high alkalinity, waterproofing effects, and toxicity to
soil microorganisms. In addition the silicones are applied with difficulty
and are costly. Satisfactory effects of these compounds on crop growth
have not been established.
Certain cellulose esters have been used successfully for structural
improvement of the soil. Among these are cellulose acetate, cellulose
methyl ether, methyl cellulose, carboxymethyl hydroxyethyl cellulose, and the variously substituted carboxymethyl celluloses (Felber
and Gardner, 1944; Quastel and Webley, 1947; Hedrick and Mowry,
1952; Martin and Kleinkauf, 1951; Raney, 1953). These compounds
act very much like the natural polysaccharides and are capable of
bonding directly to the silts and clays to effect an immediate improvement in soil granulation. Aggregate stability is largely a function of
the degree of substitution on the cellulose molecule. The materials are,
however, subject to decomposition by the soil microflora, so that induced changes tend to be of short duration. Unpublished work with
carboxymethyl cellulose at Ohio State College by P. E. Baldridge,
J. J. Doyle, and G. S. Taylor has shown that increases in the lower
plastic limit and intrinsic permeability of Hoytville clay and Miami
silt loam occur at concentrations of from 0.025 to 0.50 per cent by
weight, and that substitution of more than one carboxymethyl group per
ailhydroglucose unit is necessary to produce aggregates which do not
deteriorate after one month of alternate wetting and drying of the soil.
Certain water-soluble, polymeric electrolytes of high molecular
weight which are markedly resistant to microbial decomposition have
since 1952 been commercially available for use in the amelioration of
poor soil structure. Some 61 polymers with soil aggregate stabilizing
properties are described by Mowry and Hedrick (1953a, b) in the
Monsanto Chemical Company patents. These compounds are characterized as follows:
“The various polyelectrolytes . . . are ethylenic polymers having
numerous side chains distributed along a . . . linear carbon atom
molecule. The side chains may be hydrocarbon groups . . . , sulfonic acid groups . . . , phosphoric acid . . . , heterocyclic nitrogen
groups, aminoalkyl groups, alkoxy radicals (or derivatives thereof), the
number of which groups and the relative proportions of hydrophilic
and hydrophobic groups being such as to provide a water-soluble
polymeric compound having substantially a large number of ionizable
For best results it was noted that molecular weights in excess of
10,000 were desirable and that with some polymers best effects were
reached at 30,000 to 100,000. Cross-linked polymers were not as effective as linear polymers.
Three polymers have mostly been used: (1) hydrolyzed polyacrylonitrile (HPAN) supplied largely as a sodium polyacrylate; (2) a mixture of calcium hydroxide and a copolymer of vinyl acetate and the
partial methyl ester of maleic acid (VAMA); and (3) a copolymer of
isobutylene and the half ammonium salt-half amide of maleic acid
(IBMA). Under a variety of trade names, these compounds are on the
market in powder, flake, or liquid forms. They are usually formulated
with inactive diluents for the reduction of hygroscopicity and for easier
and more accurate application, thus reducing the amount of active material present.
2 . Factors Influencing Polymer Effectiveness
Studies by Pearson and Jamison (1953) and others (Martin, 1953;
Sherwood and Engibous, 1953) have shown that to be effective the
polymers are best mixed with the soil at rates varying from 0.02 to 0.2
per cent. The soil should contain enough moisture for good workability.
Gumming occurs if the soil is too wet. The soil should be remixed after
JAMES P. MARTIN
irrigation or precipitation if the soils are too dry. Liquid preparations
are used on prepared seedbeds to ameliorate crusting or to control
Allison (1952), Demortier and Droeven (1953), Fuller et al.
(1953), Hedrick and Mowry (1952), and numerous other investigators
have demonstrated that the synthetic polyelectrolytes are effective in
changing the structural properties of soils. The water stability of soil
aggregates is greatly increased. As a consequence of greater aggregation,
changes in porosity, water permeability, apparent density, the lower
plastic limit, and workability ensue. All the polymers behave similarly.
In general, greater aggregation has been obtained in the fine-textured
The Effect of Various Fertilizers on the Aggregate Stability of Padding Clay When
Treated with Soil Conditioners’
% aggregation ( < 0 . 2 5 mm.)
0.1% VAMA None
results obtained by M. B. Jones, Ohio State University.
An average “starter solution” fertilizer formulation is 13-46-13 used a t the rate of 4 or 5 pounds/lO gallons
water. One-half pint of this is used per transplant. I n these experiments, 70 mi. of solution of approximately the
above concentration in terms of N, PiOs a n d KzO were applied t o PO0 g. of Padding clay. This amount of
solution will bring the air-dry soil up to approximately field capacity. The soil conditioners together with the
fertilizer salts in solution were sprayed on the soil during mixing for uniform wetting. The soil was then placed
in tumblers and allowed to stand overnight. It was then air dried and subjected to wet sieve aggregate analysis.
rather than in the coarse-textured soils (Martin et al., 1952; Sherwood
and Engibous, 1953).
The results of unpublished fertilizer compatibility studies with the
synthetic polyelectrolytes made by M. B. Jones at Ohio State University
are illustrated in Table 111. These data and others (Martin, 1953) indicate that HPAN is influenced more by fertilizer than either VAMA or
IBMA but that the magnitude of the effect is generally small. Field
trials in which HPAN and VAMA were used with ammonium nitrate,
ammonium sulfate, superphosphate, and potassium chloride showed no
measurable differences in levels of aggregation attained.
3 . Persistence in Soil
FieId and laboratory tests indicate that the synthetic polymers are
markedly resistant to microbial decomposition, although aggregate deterioration from cultivation, freezing and thawing, and other natural
causes is indicated (Hedrick and Mowry, 1952; Martin, 1953). Tests
with carbonl4-1abeled HPAN and VAMA substantiated these observations as to the durability of the polymers (Martin, 1953). Brookston
silty clay loam and Hoytville clay were incubated 130 days at 2 7 O C .
with continuous aeration and at field moisture except for freezing and
drying cycles. Radioactive carbon dioxide equivalent to 2.7 per cent of
the added HPAN and 0.2 per cent of the added VAMA was produced
as a result of microbial decomposition. The addition of 1 per cent ryegrass increased the decomposition of the VAMA to 0.3 per cent. Comparable conclusions were reached from studies in Arizona (Fuller and
4 . Comparison with Natural Organic Binding Substances
As noted above, the chief differences between the natural and the
synthetic polymers (VAMA, HPAN, and possibly IBMA) is that the
aggregating substances produced through microbial activity are much
more subject to decomposition. It may take up to 2 to 5 per cent organic
residue to produce the same level of aggregation obtained with 0.05 per
cent IBMA or VAMA, and the aggregating effect of the former will
deteriorate much more quickly. One should not overlook the fact, however, that the modification of soil structure affected by organic residues
is not their only function, and the synthetic conditioners are not likely
to replace the organics in management procedures but may be used in
addition to or with them.
Recent tests were made by Martin and Aldrich (1955) at the University of California Citrus Experiment Station, Riverside, to obtain
some direct comparisons of the binding action of certain natural and
JAMES P. MARTIN
synthetic materials. Briefly, VAMA, two dextrans from soil bacteria,
IBMA, fructosan from Bacillus subtilis, and mesquite gum exerted the
greatest initial binding effect; carboxymethyl cellulose and pectin
exerted an intermediate effect; and ammonium alginate, arabogalactan (larchwood gum), and ammonium lignin sulfonate produced very
little binding action. On the basis of the pipet method used for estimating soil aggregation in these studies, it appears that some of the
microbial substances are initially just as effective as the better synthetic
Slater and Rodriguez (1954) found that the stability of aggregates
determined by wet sieving did not account for differences in structural
quality between naturally stable and conditioner-stabilized Christiana
silt loam. The two soils were equally resistant to slaking, but penetration and seed germination were better in the treated soil. It was suggested that the consistency of water-stable aggregates may be more
important than their size.
5. Effect on Plant Growth
An advantage of the microbially resistant synthetic polymers over
the natural soil conditioner substances is that they can be used as research tools to elucidate the importance of soil structure in plant response studies without the introduction of fertility factors, and at levels
of application such that they make up a minute or insignificant part of
the soil mass. It is now well established that the use of the synthetics on
some soils for structural improvement has effected significant improvement in plant growth and yield, whereas in other soils yield increases
have not occurred even though striking differences in aggregate stability
have been brought about (Allison, 1952; Fuller et al., 1952; Martin,
1953; Martin et al., 1953; Pearson and Jamison, 1953). Root crops
often improve in quality and come out of the ground cleaner following
conditioner treatment of the soil.
Decreased plant growth in a soil containing appreciable exchangeable sodium could be caused by poor soil structure, the sodium ion, or
both. Use of a soil conditioner that aggregates the soil in the presence
of exchangeable sodium offers a means of better evaluating the two
effects. A study of this type (Martin and Jones, 1954) indicated that
up to 75 per cent of reduced plant growth at certain soil sodium percentages could be ascribed to the dispersing action of the sodium ion.
Very striking yield diff ereiices have occurred where early spring
drainage has been improved or surface crusts ameliorated by conditioner
treatment (see Table IV). Stands of direct seeded tomatoes were increased from 2150 to 3640 plants per acre from treatment of Paulding
S OI L AGGREGATION
The Effect ot 0.05% Surface Applications of Soil-Aggregating Chemicals on the
Rapidity of Corn Emergence and Early Growth of Corn.
Miami Silt Loam. Columbus. 1953'
L.S.D. (0.05 level)
1 Unpublished data by De Ment. hpplication rates calculated for a ?/i-inch soil depth. The experimental
area was disked and cultimulclied. Ten corn seed were planted per pot followed by liquid applications of soilaggregating chemirals, and tlirn hand irrigated with R water applicator which produced large droplets. An
intense rain fell 0 days later. Corn planted on August 26. Plant heiglits and weigllts were taken 43 days later.
2 Carboxymethyl cellulose lE0 H.
clay with 0.15 per cent VAMA to a depth of 4 to 6 inches. Seed germinated but seedlings did not survive the waterlogged condition of untreated soils during periods of spring rains. Yields were increased from
9 to 22 tons per acre (Martin, 1953). It was noted in this experiment
and others (Martin et al., 1952; Swanson, 1953) that crops often not
only get off to a better start in polymer-treated soils but tend to mature
more quickly. The latter observation merits further research endeavor.
One way that has been found to reduce rates of application of the
synthetic polymers to economical levels is to treat only the top half inch
of soil for the reduction of surface crusting (Sherwood and Engibous,
1953). IBMA, HPAN, and carboxymethyl cellulose in solution at 0.2
to 0.4 per cent concentration, sprayed or sprinkled in a band treatment
or put out as a jet stream over the row, have proved successful in this
respect. The application rates required for improved emergence under
these conditions range from about 30 pounds per acre for complete
surface coverage, 2 to 5 pounds for band treatments, to a approximately
1 pound for the jet stream procedure. These rates are low enough to
be of practical significance.
A further implication of crust reduction from polymer treatment
is that more water infiltrates to the rhizosphere to meet crop moisture
requirements. Growth differences in crust control experiments can in
part be attributed to such moisture differences. Increase in rhizosphere
moisture infiltration is undoubtedly more important to crop production
than changes in the moisture capacities or in available moisture, which
are only slightly influenced by polymer treatment (Hedrick and
Mowry, 1952; Peters et al., 1953; Sherwood and Engibous, 1953).
A number of studies have suggested that the soil conditioners may,
under some conditions, influence the uptake of nutrients by plants
partly as a result of increased microbiological activity, and possibly
through increased growth response to fertilizers (Fuller et aZ., 1953;
Fuller and Gairaud, 1954; Martin, 1953). MacIntyre et aZ. (1954)
observed that soil treatment with HPAN, which contains sodium,
tended to increase sodium absorption by plants and decrease absorption
of calcium and magnesium. Studies by Martin and Jones (1954) indicated that, although plant growth was markedly improved in some
tests, soil treatment with VAMA did not influence nutrient absorption
by lettuce and radish plants. Sodium absorption by Love11 peach seedlings, however, tended to be reduced. In field tests at Ohio (Martin,
1953), plant tissue analyses in general indicated little influence of the
polymers on nutrient ion uptake, even when they were applied directly
with commercial fertilizers.
6. Effect on Microbial Activity
Microbiological activity appears to be increased in some soils by improvement in aggregation effected by the synthetic polymers. The
“aeration factor” of Quastel and Webley ( 1947), based on the Warburg
technique, which is essentially a measure of the oxygen demand of the
soil population, is increased by polyelectrolyte treatment, as is the evolution of carbon dioxide (Fuller and Gairaud, 1954). As noted earlier,
the increased evolution of carbon dioxide is not caused by the metabolic
degradation of the applied polymers. Fuller and Gairaud also found
that crop residues added to conditioner-treated soils decomposed more
quickly. Nitrification rates (Sherwood and Engibous, 1953) and nodulation of alfalfa (Hely and Bonnier, 1954) have been reported to increase
following conditioner applications to the soil.
The mechanism by which organic substances bind soil particles is
of interest though still in the theoretical stage. Sideri (1936) and
Myers (1937) demonstrated a direct binding action of clay particles by
polar organic compounds. Kroth and Page (1947) concluded that polar
compounds form physicochemical bonds with the surface-active clays
which prevent breakdown of the aggregate on wetting. The superiority
of polar substances produced through the decay of fresh organic materials was stressed.
Geoghegan (1950) observed that the aggregating effect of several
bacterial levans varied with the molecular weight, that is, the larger the
molecule, the greater the binding action. It was suggested that hydrogen
bonding may be the mechanism by which polysaccharides bind soils.
In this type of bonding, the hydroxyl hydrogen would be attracted to
the oxygen atom at the exchangeable base sites on the clay as well as
to the oxygen atom of the hydroxyl group in the polysaccharide. Deamination, esterification, and acetylation studies of humus extracts and
of some proteins and polyuronides (Swaby, 1950) suggested that NH,,
COOH, and OH groups might all be involved in polar linkage between
organic and inorganic colloids. The studies with humus extracts indicated that alcoholic, phenolic, and possibly amino groups were more important than carboxyl groups.
Ruehrwein and Ward (1952) believed that in order to bind soil
particles the organic molecules must he long enough to bridge the gap
between soil particles. It was postulated that the polymer molecules
(VAMA, HPAN, and IBMA) were long enough to bridge this gap and
that they were strongly adsorbed at the “anchor points” on the clay,
probably by anion exchange. The chain length theory is supported by
the findings of Geoghegan (1950), who noted a relationship between
molecular weight and binding action of a series of bacterial levans. The
molecular weight of dextrans from the Leuconostoc group of organisms
is thought to be about 4,000,000, whereas that of arabogalactan from
larchwood has been estimated to be 2200 (Whistler and Smart, 1953).
In recent studies by Martin and Aldrich (1955) the binding action of
Leuconostoc dextran was very marked, whereas arabogalactan exerted
little o r no binding action. This observation is also in harmony with
the chain length theory.
Besides the length of the chain, the shape of the active molecule and
the spacing of active sites could also influence binding action. If the
molecule contains numerous long side chains or is tightly coiled, the
clay particles may be kept from active binding sites. If active sites on
the organic molecule coincide with active sites on the clay particle,
binding may be more stable than if sites do not coincide.
Peterson (1948) proposed a scheme for calcium linkage between
uronide particles and uronide and clay particles. Such a linkage can be
illustrated as follows:
It was suggested that this type of linkage may be active in the formation
of natural soil aggregates by polyuronides in the soil organic matter and