SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Agricola
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Related Collections
Right arrow Zeoponics
Right arrow Kinetics
Right arrow Soil Chemistry
Published in Soil Sci. Soc. Am. J. 69:362-370 (2005).
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

Division S-2—Soil Chemistry

Dissolution Kinetics of Iron-, Manganese-, and Copper-Containing Synthetic Hydroxyapatites

B. Suttera,c,*, L. R. Hossnera and D. W. Mingb

a Dep. of Soil and Crop Sciences, Texas A&M Univ., College Station, TX 77843
b NASA Johnson Space Center, Houston, TX 77058
c SETI Institute, NASA Ames Research Center, Moffett Field, CA 94035

* Corresponding author (bsutter{at}mail.arc.nasa.gov)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Micronutrient-substituted synthetic hydroxyapatite (SHA) is being evaluated by the National Aeronautics and Space Administration's (NASA) Advanced Life Support (ALS) Program for crop production on long-duration human missions to the International Space Station or for future Lunar or Martian outposts. The stirred-flow technique was utilized to characterize Ca, P, Fe, Mn, and Cu release characteristics from Fe-, Mn-, and Cu-containing SHA in deionized (DI) water, citric acid, and diethylene-triamine-pentaacetic acid (DTPA). Initially, Ca and P release rates decreased rapidly with time and were controlled by a non-SHA calcium phosphate phase(s) with low Ca/P solution molar ratios (0.91–1.51) relative to solid SHA ratios (1.56–1.64). At later times, Ca/P solution molar ratios (1.47–1.79) were near solid SHA ratios and release rates decreased slowly indicating that SHA controlled Ca and P release. Substituted SHA materials had faster dissolution rates relative to unsubstituted SHA. The initial metal release rate order was Mn >> Cu > Fe which followed metal-oxide/phosphate solubility suggesting that poorly crystalline metal-oxides/phosphates were dominating metal release. Similar metal release rates for all substituted SHA (approximately 0.01 cmol kg–1 min–1) at the end of the DTPA experiment indicated that SHA dissolution was supplying the metals into solution and that poorly crystalline metal-oxide/phosphates were not controlling metal release. Results indicate that non-SHA Ca-phosphate phases and poorly crystalline metal-oxide/phosphates will contribute Ca, P, and metals. After these phases have dissolved, substituted SHA will be the source of Ca, P, and metals for plants.

Abbreviations: ALS, advanced life support • DI, deionized • DTPA, diethylene-triamine pentaacetic acid • EPR, electron paramagnetic resonance • SHA, synthetic hydroxyapatite


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE NASA'S ALS program is developing slow-release fertilizers for long-duration human missions to the International Space Station and future outposts on the Moon and Mars. Crops will supply food and recycle air and water for the astronauts (Averner, 1989; Allen et al., 1995). The slow-release nutrient medium under development by NASA is termed zeoponics, and is composed of NH4+– and K+–exchanged zeolite (clinoptilolite) and a nutrient (Fe, Mn, Cu, Zn, Mo, B, S, Cl, Mg) containing SHA [Ca10(PO4)6(OH)2] (Ming et al., 1995; Golden and Ming, 1999; Steinberg et al., 2000). Nuclear magnetic resonance spectroscopy, electron paramagnetic resonance (EPR) spectroscopy, infrared spectroscopy, x-ray diffraction analysis, and Rietveld structure refinement have collectively determined that Fe, Mn, and Cu have substituted for Ca in the SHA materials used in this research (Sutter et al., 2002a, 2002b, 2003). Electron paramagnetic resonance spectroscopy also indicated that poorly crystalline Fe-, Mn-, and Cu-oxide or phosphate phases are associated with Fe-, Mn-, and Cu-SHA materials (Sutter et al., 2002b). For the remainder of this paper these phases will be referred to as metal phases. Nutrients substituted into the sparingly soluble SHA structure would be expected to be slowly released and at rates suitable for plant growth; however, the metal phases occurring with SHA may play a significant role in controlling the release rates of the metals. NASA's interest in this slow nutrient release system is two fold—(1) slow release of nutrients over long periods of time will minimize resupply needs; and (2) a self-sustaining nutrient medium will simplify the plant growth requirements during long-duration missions, for example, compared with a hydroponic system that requires sophisticated pumps and monitoring systems (Ming et al., 1995; Henderson et al., 2000). Nutrient substituted hydroxyapatites may be an ideal source of nutrients for plants used in life support systems on space missions expected to last over a year (Ming et al., 1995; Henderson et al., 2000; Steinberg et al., 2000).

Dissolution studies of micronutrient substituted SHA are a necessary step in assessing the suitability of SHA to supply nutrients to plants. Dissolution studies of micronutrient substituted SHA are rare. Okazaki et al. (1985)(1986) performed batch dissolution studies of Fe-SHA and determined that precipitation of Fe compounds on the Fe-SHA surface inhibited Fe-SHA dissolution. Golden and Ming (1999) exposed micronutrient substituted SHA to DI water for 60 d and found that solution Ca and P were supersaturated with respect to unsubstituted SHA. When micronutrient substituted SHA was extracted with 2% citric acid, 24% less P was available compared with non-substituted SHA (Golden and Ming, 1999). Golden and Ming (1999) also determined that the amount of DTPA-extractable micronutrients from Fe-, Mn-, Cu-, and Zn-containing SHA was directly related to the level of micronutrient substitution into the SHA structure. Wheat (Triticum aestivum) growth studies have demonstrated that sufficient levels of Ca, P, Fe, Mn, and Cu can be supplied by micronutrient substituted SHA (Ming et al., 1995; Henderson et al., 2000; Steinberg et al., 2000; Gruener et al., 2003).

No studies have investigated the release kinetics of Ca, P, Fe, Mn, and Cu from Fe-, Mn-, and Cu-substituted SHA. The objective of this research was to use the stirred-flow technique (Carski and Sparks, 1985) to characterize the dissolution kinetics of Fe-, Mn-, and Cu-substituted SHA in DI water and organic acid solutions. This work will provide a better understanding of the nature of Ca, P, Fe, Mn, and Cu release from substituted SHA material, and lead to techniques or modifications that may improve the use of SHA materials as plant nutrient sources.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Synthesis of Synthetic Hydroxyapatite Materials
The Fe-, Mn-, and Cu-containing SHA and metal free SHA (pure-SHA) were produced by a procedure similar to the method of Golden and Ming (1999). Pure-SHA was synthesized by dissolving Ca(NO3)2 · H2O (235 g) in 20% (v/v) NH4OH (420 mL) while (NH4)2HPO4 (72.2 g) was dissolved in 380 mL of DI water. Thirty milliliters of 20% (v/v) NH4OH was added to the solution containing the dissolved (NH4)2HPO4. The P solution was then added to the Ca solution and mixed by a propeller stirrer for 24 h. After mixing, the precipitate was allowed to age for 48 h. The SHA precipitate was washed four times in 2.5 L of DI water to remove excess NH4OH and NO3. After washing, the SHA precipitate was filtered on a Whatman #41 (Whatman, Clifton, NJ) filter paper to remove excess water. The SHA precipitate was then heated in an oven maintained at 400°C for 24 h. The cooled SHA was ground in an agate mortar and pestle to produce a powder that passed through a 47-µm (325 mesh) sieve.

The Fe-, Mn-, and Cu-containing SHA were synthesized by dissolving the amounts of the transition metal reagents listed in Table 1 in 100 mL of DI water. The transition metal solution was added to the P solution and mixed for 5 min. The metal + P solution was added to the Ca solution and the remaining procedure was followed as described above. The numeral associated with each transition metal SHA is the approximate metal concentrations (g kg–1) in SHA.


View this table:
[in this window]
[in a new window]
  		Table 1. Total amounts of starting materials used as transition metal sources in hydroxyapatite synthesis and total elemental concentration of the synthetic hydroxyapatite (SHA) materials.

 
Chemical and Infrared Analyses
Total Ca, P, Fe, Mn, and Cu analyses were determined with a Cameca SX-50 electron microprobe (Cameca, Trumbull, CT). The SHA materials were pressed into pellets (103.5 MPa) and analyzed at 15 kV and 10 nA with a beam diameter of 20 µm. The electron microprobe sample stage was moved back and forth at 20-µm steps over a length of 200 µm to obtain an average chemical analysis. Smithsonian and C.M. Taylor (Gold Beach, OR) polished standards were used to calibrate the electron microprobe. Results of the electron microprobe analyses are provided in Table 1. All SHA samples (1.0 wt%) were ground with KBr and analyzed by a PerkinElmer 2000 Fourier transform infrared (FT-IR) (PerkinElmer, Norwalk, CT) spectrometer in diffuse reflectance mode with a resolution of 4 cm–1 and a scan range of 4000 to 450 cm–1.

Stirred-Flow Dissolution
Stirred-flow dissolution experiments were performed in duplicate using an Advantec/MFS, Inc. model UHP-43 stir-cell (Advantec, Dublin, CA) utilizing 0.2-µm cellulose nitrate filter paper. Fifty milligrams of pure-, Fe12-, Mn11-, and Cu12-SHA were separately added to 50 mL of solution in the stir-cell. A peristaltic pump maintained a constant flow rate of 2.0 mL min–1 (±0.05 mL min–1). The stir-cell rested on a magnetic stir plate that turned an internal propeller in the stir-cell. The stir speed setting was constant for all experiments. Three dissolution treatments consisted of DI water, 0.1 mM citric acid, and DTPA. Dissolution treatments were buffered at pH 7.0 with 0.001 M [N-Morpholino]propane-sulfonic acid and 0.1 N ionic strength with NaNO3. Effluent collection periods were for 10 min beginning with six consecutive samples collected for the first 60 min followed by seven more samples that were collected beginning on the 55th minute of each hour (115–125, 175–185, 235–245, 295–305, 355–365, 415–425, and 475–485 min). Time zero was defined as the point at which the first drop of effluent entered the first collection vial.

Effluent Ca concentrations were determined by flame atomic absorption spectroscopy with a PerkinElmer 3100 atomic absorption spectrometer (PerkinElmer, Norwalk, CT). Phosphorus concentrations were determined by the vanadomolybdophosphoric acid method (Kuo, 1996) where P absorbance was measured with a Beckman DU 640B spectrophotometer (Beckman Coulter, Fullerton, CA). Effluent Fe, Mn, and Cu concentrations were determined by flame (PerkinElmer 3100, Norwalk, CT) and graphite furnace (PerkinElmer 3100; HGA-600, Norwalk, CT) atomic absorption spectroscopy.

Dissolution Kinetics
The rates of Ca, P, Fe, Mn, and Cu release were calculated by the following procedure. First, we used a nonlinear least squares fitting routine to determine the integrated kinetic equation that best fit the release concentrations as a function of time. Next, we used the best-fit integrated kinetic equation to determine the cumulative amounts of elements released as a function of time and, lastly, we determined the kinetics of cumulative Ca, P, Fe, Mn, and Cu release. For the sake of brevity, only the kinetics of cumulative release are presented here.

Nonlinear least squares analyses (Jandel, 1994) were utilized to determine the rate equation that had the best fit to the cumulative release [Q(t)] versus time data. Cumulative amounts of Ca and P release were described by the power function equation [Q(t) = atb] (R2 = 0.99), while Fe, Mn, and Cu were described by the zero-order [Q(t) = a + kt] and modified-parabolic [Q(t) = a + ktb] equations (R2 = 0.99). The presence of multiple rate constants in the power function and modified-parabolic equations did not allow for easy comparisons of the rate constants. Rate comparisons were simplified by plotting the derivative of the rate equations versus time


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Stirred-Flow Dissolution and Organic Acids
The stirred-flow technique (Carski and Sparks, 1985) was used because solutes are continuously expelled from the solid–solution system, which is similar to real soils where solutes are removed by leaching and plant uptake. The stirred-flow system also provides a well mixed solution that eliminates bulk solution diffusion and minimizes film diffusion around the SHA particles (Sparks, 1989; Unwin and Macpherson, 1995). However, intraparticle diffusion is not eliminated in the stir-flow system causing kinetics to be transport or diffusion-controlled where only apparent rate parameters can be calculated.

Citric acid was utilized in this research because it occurs naturally in soils (Sposito, 1989; Basu et al., 1994; Zhang and Bloom, 1999) and is known to increase the solubility of apatite (Arbel et al., 1991; Kamh et al., 1999). Although not found naturally in soils, DTPA has been used in standard soil tests to determine plant available Fe, Mn, Cu, and Zn (Lindsay and Norvell, 1978).

Overall Trends of Calcium and Phosphorus Release from Synthetic Hydroxyapatites
The initial high release rates of Ca and P rapidly decreased for all SHA materials in DI-water, citric acid, and DTPA between 0 and 60 min (Fig. 1 and 2). The reduction in Ca and P concentrations continued more slowly beyond 60 min. These overall trends indicated that at least two phases were responsible for Ca and P release. Phase A was represented by the initial high Ca and P release rates that decreased rapidly before 60 min while Phase B was represented by lower Ca and P release rates that decreased slowly after 60 min. Phase A is attributed to the dissolution of ultrafine particles created during grinding (Holdren and Berner, 1979) and/or a highly reactive calcium phosphate phase separate from SHA (see discussion below). Phase B is attributed to SHA, which slowly releases Ca and P into solution. Calcium release rates for Cu12-SHA in DTPA gradually increased with time, which was different from Ca release from the other materials. This anomalous behavior will be discussed below.



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 1. Calcium release rates versus time for pure-SHA, Fe12-SHA, Mn11-SHA, and Cu12-SHA treated with deionized water, citric acid, and DTPA.

 


View larger version (35K):
[in this window]
[in a new window]
  		Fig. 2. Phosphorus release rates versus time for pure-SHA, Fe12-SHA, Mn11-SHA, and Cu12-SHA treated with deionized water, citric acid, and DTPA.

 
Calcium/Phosphorus Solution Molar Ratios
All solution Ca/P molar ratios for Mn11-SHA were lower than the solid Mn11-SHA Ca/P molar ratio during the first 120 min of dissolution (Fig. 3; Table 2). After 180 min, the Mn11-SHA solution Ca/P molar ratios approached that of the solid Mn11-SHA Ca/P ratio (Fig. 3). The Ca/P molar ratio release trend observed for Mn11-SHA was similar to the Ca/P molar ratio release trends of the pure-, Fe12-, and Cu12-SHA materials (data not shown). Initial (0–180 min) Ca/P molar ratios for pure-, Fe12-, and Cu12-SHA materials were all lower than their starting solid phase Ca/P ratios (Table 2). By the end of the dissolution experiments, the pure-, Fe12-, and Cu12-SHA solution Ca/P molar ratios were near their corresponding solid phase Ca/P SHA ratios (Table 2).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3. Calcium/P molar ratio release versus time of Mn11-synthetic hydroxyapatite (SHA) treated with deionized (DI) water, DTPA, and citric acid. Line represents solid Ca/P SHA molar ratio as calculated from Table 2. Error bars are standard deviation of the mean.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Calcium/P bulk molar ratios and solution molar ratios measured from the 0- to 10- and 475- to 485-min solution collections for the pure-, Fe12-, Mn11-, and Cu12-synthetic hydroxyapatite (SHA) materials. Standard deviations of the mean are in parentheses.

 
The lower solution Ca/P molar ratios relative to solid SHA Ca/P ratios suggested that a phase (Phase A) other than SHA was controlling Ca and P release in the initial stages of dissolution. A calcium-phosphate phase such as dicalcium phosphate (CaHPO4) with a Ca/P molar ratio lower than SHA could have coated or partially coated the SHA surface and may have been responsible for controlling initial Ca and P release (Kaufman and Kleinberg, 1979; Anderson and Elliott, 1992). Temporary accumulation of Ca at the surface (Thomann et al., 1990) or excess surface phosphate groups (Isowa, 1996) have also been proposed to contribute to the initial release of Ca and P during SHA dissolution. Since this phenomenon is observed with pure-SHA material, the metal-phases detected by the EPR analyses (Sutter et al., 2002b) are not likely to be the main cause of the initially low Ca/P release ratios.

Ultrafine SHA particles created during grinding are not the reason for the initially high rate of Ca and P release because the solution Ca/P molar ratios were not similar to the solid SHA Ca/P ratios. The initial Ca/P molar ratios suggests that a highly reactive, non-SHA Ca-phosphate phase(s) dominated the initial release of Ca and P The dissolution of SHA probably occurred throughout the entire experiment but the Ca/P molar ratios of SHA were not detected in initial solutions because of the overwhelming influence of the other Ca-phosphate phase(s). Once the non-SHA solid phase was dissolved and flushed from the stir-cell, the solution Ca/P molar ratios approached the Ca/P ratios in SHA.

Kinetics of Calcium and Phosphorus Release
Comparisons of Solution Treatments
Citric acid possesses four functional groups and was expected to enhance the release of Ca and P relative to the DI water treatment. However, the Ca and P release rates in citric acid were lower than in DI water (Fig. 1 and 2) (Table 3). The lower release rates may be due to the precipitation of Ca-citrate onto the SHA surface, which inhibits Ca and P release (Christoffersen et al., 1983; Misra, 1996). Calcium-citrate was detected by infrared analyses in all the citric acid treated SHA materials (Fig. 4). Christoffersen et al. (1983) demonstrated that citric acid concentrations <0.03 mM inhibited SHA dissolution. Low citrate concentrations were proposed to complex very little Ca2+; and instead, adsorb to the SHA surface inhibiting dissolution (Christoffersen et al., 1983). Citrate concentrations >0.03 mM allow for more Ca2+ to be complexed by citrate, which counteracts the adsorption effects enhancing SHA dissolution. The experiments in this work were conducted with 0.1 mM citric acid, which suggests that adsorption effects should of not of been a factor according to Christoffersen et al. (1983). Differences in experimental conditions and SHA materials between our work and Christoffersen et al. (1983) may have allowed for calcium-citrate precipitation in our experiments despite the use of higher citric acid concentrations. Calcium-citrate precipitation may not be a factor in the ALS rhizosphere because citric acid, if present, will likely occur at elevated concentrations. Growing area restrictions will lead to high planting/rooting densities where organic acid concentrations are expected to be higher than in terrestrial cropping systems (Eick et al., 1996). Future stirred-flow studies should examine the effects of higher citric acid concentrations (>0.1 mM) on SHA dissolution to verify what concentrations of citric acid are required to prevent calcium-citrate precipitation.


View this table:
[in this window]
[in a new window]
  		Table 3. Calculated rates of Ca and P release at 485 min from synthetic hydroxyapatite (SHA) in deionized (DI) water, citric acid, and DTPA.

 


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Infrared spectra of pure-, Fe12-, Mn11-, and Cu12-synthetic hydroxyapatite (SHA) materials exposed to deionized water and 0.1 mM citric acid. Arrows indicate the vibrational modes of calcium citrate that occurred between 1573 and 1579 cm–1.

 
The DTPA treatments resulted in the greatest release rates of Ca and P relative to citric acid (Fig. 1 and 2) (Table 3) because DTPA has a higher Ca-ligand stability constant than citric acid (Table 4). DTPA has a higher Ca-ligand stabilty constant than citric acid because DTPA has eight functional groups while citric acid has only four in which to bind Ca.


View this table:
[in this window]
[in a new window]
  		Table 4. Stability constants (Martell and Smith, 1982) for citric acid and DTPA and morpholine-propane-sulfonic acid (MOPS) (Anwar and Azab, 1999) when complexed with, Fe3+, Mn2+, and Cu2+. The reaction that is described by these log K values is M + L {leftrightarrow} ML where M is the metal and L is the ligand.

 
Comparisons Between Synthetic Hydroxyapatite Materials
Release rates of Ca and P from Fe12- and Mn11-SHA were in some cases significantly greater than the release rates of Ca and P from pure-SHA in the DI water and DTPA solutions (Table 3) (Fig. 1 and 2). The higher release rates of Ca and P from Fe12- and Mn11-SHA were attributed to greater structural strain and/or smaller crystallite size of the metal-SHA materials relative to pure-SHA (Baig et al., 1999) as calculated by Rietveld refinement (Sutter et al., 2003). The cases where Ca and P release rates are not significantly greater than pure-SHA suggest that the level of strain or the amount of relatively small crystals caused by substitution was not enough to significantly increase the dissolution rates. However, while the rates are not significantly higher, they are all consistently higher indicating that dissolution rates increased as a result of metal substitution.

Rietveld refinement of XRD spectra determined that the metals occurred in the six-fold coordinated Ca(2) site of SHA (Sutter et al., 2003). Accordingly, the effective ionic radii for the metals are 0.065, 0.083, and 0.077 nm for Fe3+, Mn2+, and Cu2+, respectively (Shannon, 1976). Structural strain in SHA was created when the smaller Fe3+ (0.065 nm) and Mn2+ (0.083 nm) ions substituted for the larger Ca2+ ion (0.1 nm) (Sutter et al., 2003). The Ca, P, and O atoms positions in the metal substituted SHA structure are modified relative to unsubsituted SHA to accommodate the smaller metal ions; thereby increasing the structural strain of the substituted SHA materials. The presence of metals during SHA synthesis also apparently inhibited SHA crystal growth leading to the formation of smaller crystals (Sutter et al., 2003). Strain parameters of Rietveld analysis have been shown to correlate more to synthetic carbonated apatite solubility than crystallite size (Baig et al., 1999). Sufficient data were not collected to determine which factor; smaller crystallite size or greater strain, contributed the most to higher Ca and P release from the substituted SHA materials.

The release rates of Ca and P from the Mn11- and Cu12-SHA materials were similar to Ca and P release rates from pure-SHA in the citric acid treatment (Table 3). The presence of the calcium-citrate precipitates on SHA surfaces (see above) apparently negated the effect that Mn and Cu substitution had on Ca and P release. The incorporation of Fe into SHA, however, was able to counteract the adsorption behavior of Ca-citrate precipitates; thereby, enhance Ca and P release relative to pure-SHA. The ability of the Fe12-SHA to negate the effects of the Ca-citrate precipitation may be due to the higher structural strain and smaller crystallite size of Fe12-SHA relative to the other SHA materials (Sutter et al., 2003).

The substitution of Cu [Cu2+ (0.077 nm)] into SHA was expected to enhance Ca and P release similar to the effects that Fe and Mn had on Ca and P release. However, the Ca and P release rates from Cu12-SHA in the DI water treatment were similar and lower than the rates of Ca and P release from pure-SHA (Table 3; Fig. 1 and 2). The source of Cu for SHA synthesis was nitrate based while the Mn and Fe starting materials were sulfate based (Table 1). Low N levels occur for all SHA materials (Table 1) indicating that very little nitrate was substituted into the SHA structure while elevated S levels were present in the Mn and Fe-SHA materials due to the metal-sulfate starting materials. The incorporation of sulfate into the Fe- and Mn-SHA structure would be expected to substitute for phosphate (Golden and Ming, 1999), and like metal substitutions for Ca, may have contributed to enhanced Fe- and Mn-SHA dissolution relative to the non-sulfate containing Cu-SHA.

While the incorporation of sulfate into the Fe- and Mn-SHA structure may explain the enhanced dissolution of the Fe- and Mn-SHA materials relative to Cu-SHA, it does not explain why Cu-SHA has lower Ca and initially (0–60 min) lower P release rates relative to pure-SHA (Fig. 1 and 2). A possibility exists that copper-phase(s) detected in EPR analyses of Cu-SHA materials (Sutter et al., 2002b) may have occurred on the SHA surface, which could have inhibited Ca and P release. Kohen et al. (1984) and Okazaki et al. (1985) have reported that insoluble Fe precipitates on the Fe-SHA surface inhibited Fe-SHA dissolution. Perhaps, Cu precipitates on Cu-SHA retarded Ca and P release from Cu-SHA.

The release rate of Ca from Cu12-SHA in all treatments showed a minimal change with time relative to the other materials (Fig. 1). A copper-phosphate phase may be present on the Cu12-SHA surface. The copper-phosphate phase could have inhibited Ca release but was able to supply P albeit at an initially lower rate than pure-SHA. The Ca and P release rate from Cu12-SHA in DTPA was initially (0–60 min) lower than pure-SHA but at later times became greater than pure-SHA. This suggested that with time, DTPA was able to extract enough of the copper-phosphate phase, which allowed Ca and P to be released from Cu12-SHA at rates higher than pure-SHA.

Iron and Mn phase(s) were also observed by EPR (Sutter et al., 2002b); however, Ca and P release from Fe12- and Mn11-SHA were not inhibited relative to pure-SHA in the DI water treatment. Manganese release rates were greater than Cu release rates in DI water (see below) (Fig. 5). Surface Mn phase(s) were apparently extracted from SHA more readily than the surface Cu phase(s); therefore, the Mn phase may not have inhibited Ca and P release as much as the Cu phase(s). Alternatively, the concentration of the Cu phase(s) coating the SHA surface may have been greater than the Mn phase(s) as well as Fe-phase(s) resulting in lower levels of Ca and P release from Cu12-SHA relative to Fe12- and Mn11-SHA. Backscattered electron (BSE) and transmission electron microscopy (TEM) detected individual Fe phase(s) separate from the Fe-SHA materials (Sutter et al., 2003). The occurrence of Fe phases may have been divided between Fe phases that occur on the SHA surface and Fe phases that occur separate from SHA. With lower levels of Fe-phases on the surface than Cu-phases, Ca and P release rates from Fe12-SHA occurred at higher rates than from Cu12-SHA.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 5. Manganese, Cu, and Fe release rates from Fe12-, Mn11-, and Cu12-synthetic hydroxyapatite (SHA) in deionized (DI) water and citric acid.

 
Kinetics of Iron, Manganese, and Copper Release
Deionized Water
Manganese release rates were greater than Cu release rates in DI water (Fig. 5). Copper was barely detectable while no Fe was detected (detection limit 2.0 µg L–1) in DI water experiments. Copper substitution levels in SHA may be lower than Mn (Sutter et al., 2003); however, they are not likely to be so low as to cause the large difference in Mn and Cu release rates. Calcium and P release rates at 485 min (Table 3) indicated that Mn11- and Cu12-SHA dissolution were similar to each other while Fe12-SHA had similar to slightly higher dissolution rates than Mn11- and Cu12-SHA. If SHA dissolution was supplying most of the Fe, Mn, and Cu, then Fe, Mn, and Cu release rates should at least be approximately the same. This was not observed and suggested that something else is dominating Fe, Mn, and Cu release in DI water. At the pH of these experiments (pH 7), the solubility order of Mn-, Cu-, and Fe-oxides is Mn >>> Cu ~ Fe. The greater Mn release relative to Cu and Fe release may be controlled by metal phases instead of metal-SHA. Sutter et al. (2002b) identified metal phases in these samples using EPR techniques.

While morpholine-propane-sulfonic acid (MOPS) has been considered a pH buffer with relatively weak metal-complexing ability (Bering, 1987), MOPS has been shown (Anwar and Azab, 1999) to have some metal-complexing ability for Mn2+ and Cu2+ (Table 4). Thus MOPS likely had a role in complexing Mn2+ and Cu2+ in the DI water treatment. The greater release rates of metals in the citric acid and DTPA treatments indicated that these ligands enhanced metal release relative to the MOPS which is expected since citric acid and DTPA have higher stability constants than MOPS (Table 4).

Citric Acid
The constant release rate for Mn in citric acid suggested that only one type of Mn phase dominated Mn release. Structural Mn (i.e., Mn substituted for Ca in SHA) could have been released from SHA along with Mn from the Mn-phase. However, structural Mn release is not apparent in the citric acid treatment because the amount of Mn released from the Mn-phase was still too high for structural Mn release to be detected. Iron was detected in the citric acid treatment, but the rates of Fe release were much slower than those of Mn and Cu. The Fe and Cu release rates increased with time suggesting that different phases were providing Fe and Cu. Iron- and Cu-phases may occur as layers on the SHA surface, with the outer layers having a lower solubility than the inner layers closer to the SHA surface. Thus, the more resistant Fe- and Cu-phases were released first, which then exposed underlying and more soluble phases that were subsequently released at a faster rate. The increasing Fe and Cu release rates may also reflect structural Fe and Cu release from SHA that occurred when some of the surface metal-phases were removed.

The presence of calcium citrate, as discussed above, would also be expected to inhibit metal release from surface metal phases. However, if the metal phases were heterogeneously distributed across the SHA surface, then calcium citrate could have precipitated in areas where access to Ca was more available and thus would likely have occurred in areas without metal-phases. The lack of calcium citrate precipitation in areas covered by metal-phases would allow remaining solution citrate to chelate and remove metals from the SHA surface.

Diethylene-Triamine-Pentaacetic Acid (DTPA)
The initial metal release rates for DTPA were greater than the citric acid treatments because DTPA can extract metals more readily due to its higher stability constants for the metals than citric acid (Table 4). The initial metal release rate order were similar to the other treatments and followed the order of Mn > Cu > Fe (Fig. 6). By the end of the experiment, the Fe, Mn, and Cu release rates were similar (approximately 0.01 cmol kg–1 min–1). The Ca and P release rates indicated that the initial SHA dissolution rate order was Fe12- > Mn11- > Cu12-SHA (Fig. 1 and 2). If the metals were associated with the non-SHA calcium phosphate phase or SHA, and not the metal phases, then metal release rates should have followed the order of Fe > Mn > Cu. This did not occur which suggests that the metal phases were involved in controlling the metal release at early times. The release rates of Mn and Cu were initially high and then decreased to lower rates at later times (Fig. 6). The initial high rates of Mn and Cu release were attributed to DTPA extracting the Mn- and Cu-phase(s). Manganese and Cu release was then controlled by SHA dissolution when the release rates became similar at 125 min (Fig. 6). The DI water and citric acid experiments did not have a rapid change in metal release rates (unlike the rapid change in the DTPA experiments) because these experimental conditions did not completely dissolve or remove the metal phases. The 125-min point in the DTPA experiment corresponds to the cumulative release of 7 cmol Cu kg–1 and 9 cmol Mn kg–1 which suggests that these amounts of Cu and Mn need to be released (dissolved) before most of the metal phases have been extracted. This is not to say that 7 to 9 cmol metal kg–1 SHA exist in the metal phase(s) and the remainder is substituted into the SHA. This refers to the level of metals that must be released before metal release from SHA can be detected in the stirred-flow experiments. Synthetic hydroxyapatite was likely releasing metals into solution with the metal phases, but at least 2 h were required before the influence of the metal phase(s) were removed from the stirred-flow reactor. Manganese and Cu that were released beyond these levels can then be attributed to release mostly from the SHA structure. The total amounts of Mn and Cu released in the DI water and citric acid experiments were <3.6 cmol metal kg–1 SHA, which suggests that the metal-phases played a large role in the release of Mn and Cu throughout these experiments.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 6. Manganese, Cu, and Fe release rates from Fe12-, Mn11-, and Cu12-synthetic hydroxyapatite (SHA) in DTPA.

 
The Fe release rates decreased from 0.04 to 0.01 cmol Fe kg–1 min–1 for Fe12-SHA; this release rate was a much smaller rate change than the changes in Mn and Cu release rates (approximately 0.28–0.01 cmol kg–1 min–1) (Fig. 6). The Mn and Cu phase(s) were extracted by DTPA, which resulted in the drastic change in the Mn and Cu release rate while the smaller rate change for Fe release suggested that the Fe phase(s) were not completely solubilized. This is not unexpected since the kinetics of Fe release in soils has been demonstrated to be much slower than Mn release rates (Pohlman and McColl, 1986). Similar metal release rates by the end of DTPA experiments from metal-SHA materials, which have roughly similar levels of substituted metals, suggested that Fe like Mn and Cu were released from the SHA structure at later times.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Initial Ca and P release data for the pure- and metal-SHA materials indicated that a phase or phases other than SHA were responsible for Ca and P release early in the dissolution experiments. The nature of the phase(s) responsible for initial Ca and P release is not clear and could be attributed to another Ca-P phase, such as dicalcium phosphate, variations in Ca-P chemistry on the SHA surface or a combination of these factors. The phase(s) responsible for initial Ca and P release contributed Ca and P early in the experiments, but SHA ultimately dominated Ca and P release. The DTPA treatment, as expected, caused the highest release of Ca and P. Calcium-citrate precipitation in the citric acid treatment inhibited Ca and P release from SHA as well as the non-SHA calcium phosphate phases. However, if higher concentrations of citric acid occur in the soil solution than in these experiments, then enhanced Ca and P release dissolution will likely result. The inhibiting effects of surface metal phases (e.g., copper phosphate) on SHA dissolution will likely be removed over time by organic acids exuded by plant roots and microbes.

The constant release rates of metals from SHA in DTPA suggested that metal-substituted SHA has the potential to be a suitable reservoir of metal micronutrients for plant growth as long as SHA can endure. The metal phases associated with metal-SHA materials will also initially provide Mn and Cu while the lower solubility of the Fe phase(s) will provide Fe for a longer time than Mn and Cu phases can supply Mn and Cu. Enhanced release of metals in the citric acid and DTPA treatments demonstrate that organic acids exuded by plant roots and microorganisms in soil systems should play a major role in extracting these metals from the metal-substituted SHA and associated metal phases. Low organic acid concentrations (0.1 mM) were used in this work. Future work should investigate the endurance of SHA and its ability to continually supply Ca, P, Fe, Mn, and Cu under higher organic acid concentrations that may occur in the ALS rhizophere. Iron-, Mn-, and Cu-containing SHA materials are potential long-term sources of Ca, P, Fe, Mn, and Cu to plants for NASA's ALS and terrestrial cropping systems.


   ACKNOWLEDGMENTS
 
This research was partially funded by NASA's Graduate Student Researchers Program (NGT 51229) and the NASA Advanced Life Support Program. The authors gratefully acknowledge the two anonymous reviewers whose critical review greatly improved this manuscript.

Received for publication May 17, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Agricola
Right arrow Articles by Sutter, B.
Right arrow Articles by Ming, D. W.
Related Collections
Right arrow Zeoponics
Right arrow Kinetics
Right arrow Soil Chemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Journal of
Environmental Quality