Soil Science Society of America Journal 63:1703-1708 (1999)
© 1999 Soil Science Society of America
DIVISION S-3-SOIL BIOLOGY & BIOCHEMISTRY
Pseudomonas cepacia–Mediated Rock Phosphate Solubilization in Kaolinite and Montmorillonite Suspensions
B. Bar-Yosefa,
R.D. Rogersb,
J.H. Wolframb and
E. Richmanb
a Agricultural Research Organization, Inst. of Soil, Water, and Environmental Sciences, Bet Dagan 50250, Israel
b Idaho National Engineering and Environmental Lab., Biotechnology Unit, Idaho Falls, ID 83415 USA
vwbysf{at}agri.gov.il
 |
ABSTRACT
|
|---|
Pseudomonas cepacia is known as a rock phosphate (RP) solubilizer in bioreactors and in soils. The objectives of this study were to determine the production rates of gluconic acid (GA, pKd 3.41) and 2-ketogluconic acid (KGA, pKd 2.66) by the bacteria in the presence of clay minerals which prevail in soils, and the resulting rate and extent of orthophosphate (OP) release into the suspension solutions. Suspensions (1:40) of RP, RP + Ca–kaolinite (CaKL), RP + Ca–montmorillonite (CaMT), and RP + K–montmorillonite (KMT) were inoculated with P. cepacia E37. The electrical conductivity (EC) and pH, and the OP, glucose, GA, KGA, Ca, and Al concentrations were determined in the suspension solutions as functions of time. In a given clay system, the rate-limiting step in RP dissolution was the rate of GA release by the E37. This acid lowered the pH of all the clay suspensions to 2.7 to 2.8, which resulted in a pronounced increase in the OP concentration in solution, Cp. As glucose was depleted from the system, the KGA concentration increased with a concomitant lowering in pH to 
2.5. At this pH, a sharp decline in Cp occurred, which was attributed to Al release by the alumosilicates, and formation of a new, stable Al–P or Fe–P solid phase. The E37 glucose oxidation efficiency (GOE) was considerably inhibited in CaKL as compared with CaMT or KMT. The GA and KGA adsorption by the clays or their Ca complexation did not play a role in the E37-mediated RP solubilization.
Abbreviations: CaKL, Ca–kaolinite • CaMT, Ca–montmorillonite • EC, electrical conductivity • GA, gluconic acid • GOE, glucose oxidation efficiency • IAP, ion activity product • KGA, 2-ketogluconic acid • KKL, K–kaolinite • KMT, K–montmorillonite • OP, orthophosphate • RP, rock phosphate
|
INTRODUCTION
|
|---|
THE EFFICACY OF VARIOUS Pseudomonas species in dissolving RP from suspension, agar, and soil has received considerable attention during the last two decades (Azcon et al., 1976; Ralston and McBride, 1976; Gaur et al., 1980; Kundu and Gaur, 1980, 1984; Rogers et al., 1986; Illmer and Schinner, 1992). The dissolution of RP involves two steps: (i) production of the monocarboxylic GA and KGA by the bacteria and (ii) dissociation of these acids (pKd 3.41 and 2.66 at ionic strength I = 0.2 M, for GA and KGA, respectively; Moghimi and Tate, 1978) and subsequent dissolution of the RP by the resulting protons. The acids are produced by the oxidation of glucose via the extracellular membrane-bound glucose dehydrogenase (Babu-Khan et al., 1995). Pseudomonas cepacia E37 was reported to produce 2 mmol GA d-1 under optimal growing conditions in a continuous bioreactor and 0.4 mmol KGA d-1 in a suspension volume of 125 mL (109 cells mL-1) (Rogers, 1989). The rate of RP dissolution is usually controlled by the diffusion of the dissolution products, Ca and P, into the surrounding medium (Kirk and Nye, 1986). The diffusion rate is determined by, among other factors, the buffering capacity of the system for Ca, OP, and pH, which explains differences in RP dissolution rates observed in soils differing in chemical characteristics (Robinson et al., 1992). Rock phosphate dissolution in a batch reactor in the presence of 0.15 M H2SO4 had a half life of 1 to 3 min, depending on the RP source (Venturino et al., 1990). In some British soils, 50% of the RP that dissolved in 60 d was found in the soil solution after 2 to 3 d (Robinson et al., 1992).
The possible role of Ca complexation by GA and KGA in dissolving RP is negligible, because of the low stability constants of these complexes (K11 = 16 and 0.7, respectively; Moghimi and Tate, 1978).
The overall objective of this work was to study the effect of P. cepacia E37 on RP dissolution in kaolinite and montmorillonite suspensions, as part of an effort to evaluate the suitability of E37 as a RP solubilizer in plant rhizospheres. Our specific objective was to quantify the interrelationships among glucose oxidation by E37 in the presence and absence of two clay minerals prevalent in soils, the resulting GA and KGA concentrations, and temporal pH and OP concentration in the suspension solutions.
|
Materials and methods
|
|---|
Preparing the Pseudomonas cepacia E37 Stock
The bacteria were propagated in a stirred tank bioreactor under optimal growth conditions to give an average bacterial population of 109 cells mL-1. Details of the propagation method, optimal growth conditions, the isolation of E37, and some relevant genetic background can be found in Rogers (1992).
Clay Preparation and Titration
The clays were Georgia kaolinite and Wyoming montmorillonite. Each clay was saturated with either Ca or K by three successive equilibrations with 1 M CaCl2 or KCl (15 L kg-1 raw clay). After equilibration, the clays were washed with distilled water to EC < 0.1 dS m-1. In the course of the last three washes, all particles which settled down to a depth of 25 cm in <30 min were discarded. The suspensions were concentrated by centrifugation and dried by lyophilization. Clay charging curves were obtained by titrating 300 mL of 0.01 M KCl and 7.5 g of clay with 0.1 M HCl and KOH or Ca(OH)2 (depending on system); the time between successive acid or base additions was 7 ± 2 min. Consumption of H or OH was defined as the amounts added minus the observed increase in H or OH in the solution phase (Fig. 1)
. The zero points of titration pH of the CaMT, KMT, CaKL, and K–kaolinite (KKL) suspensions were 8.4, 9.4, 5.8, and 5.8, respectively (Fig. 1).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1 Protons and hydroxyls consumption (application minus addition in solution) by Ca–montmorillonite (CaMT) and K–montmorillonite (KMT) and Ca–kaolinite (CaKL) and K–kaolinite (KKL) (1:40 clay/electrolyte [0.01 M Cl-] suspensions) as a function of suspension pH
|
|
Experimental Procedures
The treatments, run in duplicates, are summarized in Table 1
. Unless otherwise stated, replicates were within ± 10% of each other, and only means are presented below. Clays, RP, and electrolyte solutions were presterilized three times at 121°C for 20 to 30 min in all treatments. Suspensions were prepared by adding 7.5 g of clay, 290 mL of bacteria minimum salt medium (with or without inoculum), 5 mL of pH-adjusting solution [HCl, KOH, or Ca(OH)2], and 5 mL of the appropriate OP solution to 500-mL Erlenmeyer flasks. The RP (Western Phosphate rock, National Institute of Standard and Technology Reference Material 694) included 138 g P, 52 g Si, 9.5 g Al, and 310 g Ca kg-1. In all cases, the rock was ground to pass a 0.005-cm sieve (250 mesh). The E37 inoculum was added to the flasks after a preequilibration period as specified in Table 1. Flask openings were plugged with sponges and covered with aluminum foil, and the flasks were vigorously shaken horizontally. At predetermined times, 20-mL suspension aliquots were drawn under sterile conditions and immediately centrifuged. Supernatants were sampled for prompt pH and EC determination, and the rest was acidified by 0.001 M HCl for storage. At the end of the experiment, the clay in the centrifuge tubes was resuspended and equilibrated for 24 h with 0.5 M NaHCO3 at a 20:1 solution/clay ratio to determine OP, GA, and KGA desorption. The minimum salt medium composition was: 1 g (NH4)2SO4 L-1; 0.3 g MgSO4 L-1; 1 mg MnCl2 · H2O L-1; 2.6 mg Na2MoO4 L-1; 0.06 g FeSO4 · 7H2O L-1, and 10 g dextrose L-1.
Chemical analysis of the solutions involved measurement of Ca by atomic absorption, of K by emission, of Al by inductively coupled plasma analysis, and of OP colorimetrically by the molybdenum blue procedure (Olsen and Sommer, 1982). Gluconic and 2-ketogluconic acids were determined by a composite high performance liquid chromatography system (linear UVIS 200 UV detector, Polypore H 220 by 4.6 mm column, 0.01 M H2SO4 eluent), and glucose by glucose oxidase analyzer YSI 2700 (YSI, Yellow Springs, OH).
Speciation
A program was written to calculate all relevant P, GA, and KGA species activities and concentrations from measured total concentrations of P, Ca, GA, KGA, EC, and pH. Thermodynamic constants were taken from Lindsay (1979) and Moghimi and Tate (1978). Ionic strength, I (M), was approximated according to
(Griffin and Jurinak, 1973). Ion activity coefficients were calculated with the Debye–Huckel equation. The calculation involved four nonlinear equations with four unknowns solved by the Newton iterative method.
|
Results
|
|---|
Rock Phosphate Dissolution in Calcium–Kaolinite Suspensions
At the end of the preinoculation equilibration (te, Table 1), Cp in the RP–no-clay–pH0 4 suspension was 86 ± 2 µM P (Fig. 2a)
, that is, 96% of the initial RP. Here pH0 is the suspension pH at time zero. In the RP + CaKL–pH0 4 suspension, Cp at te was 54 ± 2 µM (Fig. 3)
. Assuming a complete RP dissolution in the presence of CaKL, as in its absence, the reduced Cp is explained in terms of P adsorption by the kaolinite (1.45 mmol P kg-1). With pH0 8, Cp at te was zero both in the presence and absence of CaKL (Fig. 2a and 3a). One day after inoculation
, Cp in the RP–no-clay–pH0 8 suspension increased to 90 µM P, concomitantly with a steep decrease in suspension pH from 8 to 2.7 (Fig. 2). The E37-mediated RP dissolution rate was quite similar to the RP dissolution rate in the abiotic system at pH0 4 [77 mg RP/(40 mL suspension x 24 h)]. No RP dissolution was observed in the abiotic systems with pH0 8 (Fig. 2a). The presence of clay in the biotic RP + CaKL–pH0 8 suspension slowed down the increase in Cp relative to the no-clay suspension (compare Fig. 3a and 2a), and the maximum Cp amounted to only 52 µM P. The inhibition at t = 48 h stemmed from P adsorption by kaolinite and from the pH buffering capacity of CaKL; in the presence of CaKL the pH was 5, whereas in its absence the pH was 2.7 (Fig. 3b vs. Fig. 2b). Both in the absence and in the presence of CaKL, an abrupt decline in Cp was observed 24 h after it had peaked. A similar, yet smaller decline in Cp in the presence of E37 was observed for pH0 4 (Fig. 2a and 3a).
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 2 (a) Orthophosphate concentration (Cp) and (b) pH in solution of rock phosphate (no-clay) suspensions as a function of equilibration time. Inoculation took place at t = 24 h. An asterisk (*) represents a significant difference (t test comparison, P < 0.05) between biotic (E37) and abiotic (no E37) treatments at given time and initial pH treatment
|
|
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 3 (a) Orthophosphate concentration (Cp) and (b) pH in solution of Ca–kaolinite + rock-phosphate suspensions (1:40) as a function of time. Inoculation took place at t = 24 h. An asterisk (*) represents a significant difference (t test comparison, P < 0.05) between biotic (E37) and abiotic (no E37) treatments at given time and initial pH treatment
|
|
The similarity between the rate of decrease in pH and the rate of increase in Cp during the first 24 h after inoculation in the no-clay–pH0 8 system (Fig. 2) is in accord with the presumption that the rate-limiting step in RP solubilization was acid release by the bacteria. This presumption is based on published data showing that the rate of diffusion-controlled RP dissolution (Venturino et al., 1990) is considerably faster than that of proton contribution by GA + KGA produced by E37 in a bioreactor (Rogers, 1989). Because of the small pH buffering capacity of the no-clay system, pH0 in the biotic suspensions had no effect on the pH at t = 48 h (Fig. 2). In the presence of CaKL, an additional 24 h were required to obtain the same suspension pH under biotic conditions at pH0 8 and 4 (Fig. 3b).
Decreasing the CaKL suspension pH from 3 to 2.5 was associated with a considerable decrease in Cp (Fig. 3a and 3b): at pH 2.8 to 3.0, Cp in the pH0 8 system continued to increase between 72 and 124 h, while at a suspension pH of 2.5 (t = 168 h) Cp dropped to nearly zero. A similar effect of suspension pH decrease from 2.7 to 2.8 to 2.5 on Cp was observed for pH0 4 (Fig. 3a and 3b).
Effect of Clay Type on Rock Phosphate Solubilization and Gluconic Acid + 2-Ketogluconic Acid Release by E37
At any t > 50 h, Cp in CaKL suspension exceeded Cp in CaMT and KMT suspensions (in that order) (Fig. 4a)
. The higher Cp in the CaKL suspension could be explained in terms of a slightly higher pH, in the critical range of 3 to 2.5, than with the other clays. At identical pH and ionic strength, P adsorption by kaolinite surpasses that by montmorillonite (Bar-Yosef et al., 1988a). The CaKL and KKL had almost identical charging curves (Fig. 1) and similar temporal Cp values; therefore, the Cp and pH data for KKL are not presented. According to the charging curves (Fig. 1), P adsorption by CaMT should exceed that by KMT at a similar pH; this has been reported in the literature (Bar-Yosef et al., 1988a) and was found in our study in treatments in which OP was added to the suspensions (data not presented). However, this result was not obtained when P was added as RP in the presence of E37 (similar Cp in CaMT and KMT solutions, Fig. 4). The reason for this discrepancy was the lower GA + KGA concentration in CaMT than in KMT suspension (Fig. 5)
. The lower GA + KGA concentration in CaKL than in CaMT (Fig. 5) also explains the difference in Cp between the two suspensions discussed above (Fig. 4).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 4 (a) Orthophosphate concentration (Cp) and (b) pH in solution of Ca–kaolinite (CaKL), Ca–montmorillonite (CaMT) and K–montmorillonite (KMT) suspensions (1:40) as a function of equilibration time. Rock phosphate was added to all suspensions at t = 0. Inoculation with E37 took place at t = 24 h (CaKL) or 6 h (CaMT, KMT)
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5 Gluconic acid (GA) + 2-ketogluconic acid (KGA) concentration in solution of Ca–kaolinite (CaKL), Ca–montmorillonite (CaMT), and K–montmorillonite (KMT) suspensions (1:40) as a function of equilibration time. Phosphorus was added as rock phosphate at t = 0. Inoculation with E37 took place at t = 24 h (CaKL) or 6 h (CaMT, KMT)
|
|
A distinct difference between KL and MT suspensions was that in CaKL Cp continued to increase between 48 and 96 h after inoculation, while in CaMT it decreased (Fig. 4a). The different trends can be explained by the different pHs in the kaolinite and montmorillonite suspensions (Fig. 4b) induced by differences in GA + KGA concentrations (Fig. 5). In all clays, the decline in Cp to 0 coincided with suspension pH values between 2.5 and 2.7 (Fig. 4). The fact that in CaKL, Cp at pH 2.9 was still 53 µM P leads to the conclusion that a change in P phase occurred as the pH dropped below 2.9, and that this new phase binds P much stronger than the clay surface.
When P was added to biotic CaKL as OP rather than RP, the GA + KGA concentration and the suspension pH values were very similar to those found in the RP system (data for OP not presented). This proves that E37 activity was unaffected by the high initial Cp (46 vs. 1 µM P). Unlike the increase in Cp with time characteristic of systems with RP, Cp declined monotonically with time when P was added as OP. The reason for the monotonic decline in Cp was the E37-mediated gradual decrease in pH and increased P adsorption (data not presented). The relatively slow P adsorption was controlled by the rate of GA + KGA release by the E37.
The incremental increase in EC [
EC = (EC at t) - (EC at inoculation time)] varied among the suspensions between 0 (abiotic) and 1.6 dS m-1 (detailed data not presented). This EC was generated by E37, but it was small relative to the increase in GA + KGA concentration, which varied between 0 (abiotic) and 30 mM. Had the GA and KGA been fully dissociated, the EC value corresponding with the 30 mM concentration could have been 3 dS m-1. A plot of EC vs. GA + KGA for all the data points at t = 24 and 144 h after inoculation (18 observations for each t) gave two straight lines with identical slopes of 0.03 dS m-1 mM-1. At 24 h the intercept was 0, and at 144 h it was 0.5 dS m-1. The slope implies that across the experimental GA + KGA concentration range, only 30% of the organic acids in solution were dissociated. This dissociation percentage is expected at pHs below the pKdissociation of GA and KGA (3.4 and 2.7, respectively), but at higher pHs this could also have been caused by Ca chelation.
Clay–Phosphorus Desorption by NaHCO3
Bicarbonate at 0.5 M (1:20 clay/extracting solution ratio) has been shown to be an effective extractor of OP from kaolinite and montmorillonite adsorption sites (Kafkafi et al., 1988). In our study, the percentages of sorbed P recovered by NaHCO3 in the abiotic CaKL system were 60 and 72% when the equilibrium pHs were 4 and 8.4, respectively. In the biotic CaKL only 2.5% of the sorbed P was recovered by NaHCO3, regardless of the initial pH of the system (Table 2)
. The striking difference in P desorption between biotic and abiotic systems supports the suggestion that at pH 2.5, caused by the GA and KGA, P retention shifted from adsorption, which controlled Cp between pH 8 and 2.7, to a new and less soluble P solid phase. Similar NaHCO3 recovery results were obtained in the other clay suspensions studied (data not presented).
View this table:
[in this window]
[in a new window]
|
Table 2 Orthophosphate desorption from Ca–kaolinite (CaKL) suspensions by 0.5 M NaHCO3 in the presence and absence of E37 at the end of experiment (t = 168 h)
|
|
No GA or KGA were detected in any of the clay NaHCO3 extracts (data not presented), which indicates that these acids were not adsorbed by kaolinite and montmorillonite.
Glucose Oxidation by E37
Glucose oxidation efficiency is defined as the incremental increase in GA + KGA concentration (M) divided by the incremental decrease in glucose concentration (M) within a given time interval
. At t = 24 h after inoculation, GOE in all RP + CaKL and RP–no-clay suspensions was in the range 0 .67 to 0.74 (Table 3)
; the only exception was the RP + CaKL–pH0 8] suspension in which GOE was 0.40, and in which the highest Cp was obtained (Fig. 4). Forty-eight hours after inoculation, the GOE decreased to 0.3 to 0.4 in suspensions with pH0 = 8, but was unchanged relative to that after 24 h in suspensions with pH0 = 4 (Table 2). In the RP + CaKL–pH0 8 suspension GOE increased to 0.58. At t > 48 h, the glucose concentration in all suspensions sharply decreased (data not presented), the GA and KGA were consumed by the bacteria for maintenance, and the GOE could no longer be determined. In the RP + CaKL–pH0 8 suspension the GOE at t = 72 h was 0.75. We do not have a satisfactory explanation for why bacteria in this suspension were slower in attaining the maximum GOE than those in the other suspensions, but this appears to have been the reason for the difference among the Cp(t) functions in the various suspensions investigated. Measurements of initial glucose concentrations and the initial numbers of cells in suspensions exclude the possibility that those factors affected the above-mentioned results.
View this table:
[in this window]
[in a new window]
|
Table 3 Glucose oxidation efficiency (GOE , M/M) by E37 in the time interval 0 to 24 and 24 to 48 h after inoculation in Ca–kaolinite (CaKL) and no-clay rock phosphate (RP) suspensions
|
|
In all the suspensions, the predominant glucose oxidation product during the first 24 h after inoculation was GA. After 48 h, the concentration of KGA increased with elapsed time, while that of GA declined and reached zero at t = 168 h (data not presented).
|
Discussion
|
|---|
Between inoculation and 100 h later, E37 produced on the average 20, 12, 8, and 6 µmol GA + KGA h-1 in 40 mL of suspension in RP without clay, RP + KMT, RP + CaMT, and RP + CaKL suspensions, respectively (Fig. 5). The experimental error was ± 15% of the mean. Data from previous continuous bioreactor studies with E37 at constant pH of 3.5 and glucose concentration of 10 g L-1 (optimal growth conditions), showed a production rate of 30 µmol h-1 in 40 mL (Rogers, 1989). The differences among the above-mentioned suspensions cannot be attributed to experimental variability in initial glucose concentration or bacteria number, nor to GA and KGA adsorption by the clays, as adsorption was estimated to be negligible. It is speculated that the decline in glucose oxidation stemmed from bacterial cell coating by the clay (Bashan and Levanony, 1988; Fendorf et al., 1997). The coating could form a direct contact between the extracellular glucose oxidation enzyme and the clay surface, resulting in reduced enzymatic activity. The order of decline in GA + KGA production is consistent with the order of increase in P adsorption by the three clays, indicating that adsorption was involved in the enzyme deactivation.
The GA + KGA production rate declined when glucose concentration in suspensions approached zero (96 h, depending on clay system; data not presented). The solution KGA/GA molar ratio increased from 10% at t = 48 h to 100% at t = 168 h (data not presented). The stronger KGA caused an additional decrease in suspension pH from 3 to 2.5 to 2.7. At pH < 2.7, a sharp decline in Cp was observed in all suspensions (with and without clay). To determine the new P solid phase that was formed at this pH in montmorillonite and RP systems, the Al3+ and PO3-4 ion activity product (IAP) was compared with the solubility product constant (Ksp) of variscite (AlPO4 · 2H2O). In kaolinite suspensions the H22PO-4 activity was compared with the theoretical H2PO-4 activity in a variscite–kaolinite system (Table 4)
. Data show that all decreases in pH between the time of maximum Cp and the time of minimum Cp were associated with increases in Al+3 activity and decreases in H2PO-4 activity in the suspension solution (Table 4). In the clay suspensions, the enhanced Al concentration stemmed from montmorillonite and kaolinite dissolution, while in RP it probably emanated from alumosilicates present as contaminants. In CaMT, KMT, and RP without clay, the IAP under conditions of maximum Cp were close to the Ksp of variscite. As the pH dropped below the threshold value of 2.7, Cp became undersaturated with respect to variscite (Table 4). In CaKL, the H2PO-4 activity at maximum Cp was supersaturated respective to the theoretical H2PO-4 activity in a variscite–kaolinite system. At the minimum Cp, the H2PO-4 activity was close to the theoretical activity in the variscite–kaolinite system (Table 4). At the maximum Cp (pH 2.8), the solid phase that controlled H2PO-4 in CaKL suspension could be taranakite, which under the experimental conditions is more soluble than variscite (Lindsay, 1979). The P compound that controlled Cp in montmorillonite and RP at pH < 2.7 could not be determined in this study because Fe, a potential precipitator of
; Lindsay, 1979) was not analyzed.
View this table:
[in this window]
[in a new window]
|
Table 4 Experimental (exptl) pH, H2PO-4, and Al3+ activities coinciding with maximum and minimum OP concentrations in biotic suspension solutions (Cp), and corresponding ion activity products (IAP) respective to Ksp of variscite or variscite in equilibrium with kaolinite (varis-kaol).
|
|
Speciation calculation showed that [CaGA]+ (stability constant K11 = 16), which could stimulate RP dissolution, did not surpass 13% of total Ca in solution. This maximum value was calculated for a RP + CaMT suspension at pH 3.8 and total GA concentration of 20 mM. When the suspension pH was higher or lower than 3.8, the [CaGA]+ complex comprised 2 to 3% of the total Ca concentration. When repeating the computation for
and assuming a KGA concentration of 20 mM, the calculated molar fraction of the [CaKGA]+ complex slightly differed from the molar fraction of [CaGA]+ but constituted <3.8% of total Ca under all circumstances. The speciation results agree with the conclusion of Moghimi and Tate (1978) that formation of [CaKGA]+ should be disregarded as a significant mechanism in RP dissolution.
|
Conclusions
|
|---|
The choice of E37 as an RP biosolubilizer in industry (Rogers et al., 1986) was appropriate, as the bacteria very efficiently produce GA and KGA that can replace the inorganic acids currently used in superphosphate production. In this process, the RP slurry is maintained at pH 3.5 by pH-stat. However, in the case of RP solubilization by E37 in plant rhizospheres, the dissolution reaction should be carefully evaluated, as the KGA might be too strong for soils containing alumosilicate minerals, and may cause local Al–P and probably Fe–P precipitates, which would reduce the advantage of RP dissolution by the weaker GA. Moreover, the low local pH and high Al concentrations may be toxic to plant roots, and may prevent root proliferation in soil subvolumes enriched with RP and E37. The GA and KGA anions seem not to be adsorbed by montmorillonite and kaolinite and presumably cannot compete and desorb P.
Extrapolation of the current results to intact soils should be carefully done because of the vast difference in solution/solid ratio and possible effects of additional soil minerals and metal oxides on RP dissolution and secondary reactions determining the temporal Cp in the solution phase.
|
NOTES
|
|---|
Contribution from the Agricultural Research Organization series 626/98.
Received for publication September 28, 1998.
|
REFERENCES
|
|---|
- Azcon R., Barea J.M., Hayman D.S. Utilization of rock phosphate in alkaline soils by plants inoculated with mycorhizal fungi and phosphorus solubilizing bacteria. Soil Biol. Biochem. 1976;8:135-138.
- Babu-Khan S., Yeo T.C., Martin W.L., Duron M.R., Rogers R.D., Goldstein A.H. Cloning of mineral phosphate-solubilizing gene from Pseudomonas cepasia. Appl. Environ. Microbiol. 1995;61:972-978.[Abstract]
- Bar-Yosef B., Kafkafi U., Rozenberg R., Sposito G. Phosphorus adsorption by kaolinite and montmorillonite: I. Effect of time, ionic strength and pH. Soil Sci. Soc. Am. J. 1988;52:1580-1585 a.[Abstract/Free Full Text]
- Bashan Y., Levanony H. Adsorption of the rhizosphere bacterium Azospirillum brasilense Cd to soil, sand and peat particles. J. Gen. Microbiol. 1988;134:489-493.[Abstract/Free Full Text]
- Fendorf S.E., Li G., Morra M.J., Dandurand L.M. Imaging a pseuodomonad in mineral suspensions with scanning force and electron microscopy. Soil Sci. Soc. Am. J. 1997;61:109-115.[Abstract/Free Full Text]
- Gaur A.C., Mathur R.S., Sadasivam K.V. Effect of organic material and phosphate-dissolving culture on the yield of wheat and greengram. Indian J. Agron. 1980;25:501-503.
- Griffin R.A., Jurinak J.J. Estimation of activity coefficients from the electrical conductivity of natural aquatic systems and soil extracts. Soil Sci. 1973;116:26-30.
- Illmer P., Schinner F. Solubilization of inorganic phosphates by microorganisms isolated from forest soils. Soil Biol. Biochem. 1992;24:389-395.
- Kafkafi U., Bar-Yosef B., Rozenberg R., Sposito G. Phosphorus adsorption by kaolinite and montmorillonite: II. Organic anion competition. Soil Sci. Soc. Am. J. 1988;52:1585-1589.[Abstract/Free Full Text]
- Kirk G.J.D., Nye P.H. A simple model for predicting the rate of dissolution of sparingly soluble calcium phosphate in soil. 1. The basic model. J. Soil Sci. 1986;37:529-540.
- Kundu B.S., Gaur A.C. Establishment of nitrogen-fixing and phosphate-solubilizing bacteria in rhizosphere and their effect on yield and nutrient uptake of wheat. Plant Soil 1980;57:223-230.
- Kundu B.S., Gaur A.C. Rice response to inoculation with N2-fixing and P-solubilizing microorganisms. Plant Soil 1984;79:227-234.
- Lindsay W.L. Chemical equilibria in soils. New York: Wiley-Interscience, 1979.
- Moghimi A., Tate E. Does 2-ketogluconic chelate calcium in the pH range 2.4 to 6.4?. Soil Biol. Biochem. 1978;10:289-292.
- Olsen S.R., Sommer L.E. Phosphorus. In: Page A.L., et al. , ed. Methods of soil analysis. Part 2, 2nd ed Madison, WI: Agron. Monogr. 9. ASA and SSSA, 1982:403-430.
- Ralston D.B., McBride R.P. Interaction of mineral phosphate-dissolving microbes with red pine seedlings. Plant Soil 1976;45:493-507.
- Robinson J.S., Syers J.K., Bolan N.S. Importance of proton supply and calcium-sink size in the dissolution of phosphate rock materials of different reactivity in soil. J. Soil Sci. 1992;43:447-459.
- Rogers R.D. Biological separation of phosphate from ore. Idaho Falls, ID: Third Quarter report (contract DE-AC07-76ID01570) submitted to the Department of Energy, 1989.
- Rogers R.D. Biological separation of phosphate from ore. Idaho Falls, ID: Second Quarter report (contract DE-AC07-76ID01570) submitted to the Department of Energy, 1992.
- Rogers, R.D., J. Acasi, R.C. Cronn, J.K. Trautman, and J.H. Wolfram. 1986. Biosolubilization of phosphate ore. p. 161–168. In Proc. 118th Annual Meeting. Am. Inst. Mining, Denver, CO.
- Venturino G., Cocquio F., Mulato R., Barbera A. Measurement of acid attack on phosphate rocks. Fert. Res. 1990;26:81-87.
TOP
ABSTRACT
INTRODUCTION
