| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a The Connecticut Agricultural Experiment Station, P.O. Box 1106, New Haven, CT 06504
b Dept. of Plant Science, U-67, University of Connecticut, Storrs, CT 06269
Corresponding author (c.schulthess{at}uconn.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ATR, attenuated total reflectance • CIR, cylindrical internal reflectance • DR, diffuse reflectance • DRIFT, diffuse reflectance infrared Fourier transformed • EM, electrophoretic mobility • FTIR, Fourier transformed infrared • 
(CO), difference between symmetric and asymmetric C–O stretching vibrations
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Van Geen et al. (1994) postulated inner-sphere surface complexation reactions with both the carbonate anion and the bicarbonate anion as possible surface species. Infrared studies of CO2 adsorption on goethite indicate the presence of inner-sphere surface complexes of carbonate species (Russell et al., 1975). Under evacuated conditions, the bicarbonate species is predominant, while the carbonate species is predominant upon exposure of the goethite films to air. Fourier transformed infrared studies of carbonate adsorption on hydrous ferric oxides indicate monodentate inner-sphere carbonate as the predominant surface species under both evacuated and air-dried conditions (Harrison and Berkheiser, 1982; Nanzyo, 1986). However, spectroscopic techniques that require evacuated or air-dried conditions have the potential to modify surface complexes on surfaces that are normally present in an aqueous environment (Hug, 1997). To identify adsorbed species present at the mineral–water interface, spectroscopy techniques are needed that can be applied in situ in the presence of water (Brown, 1990; Johnston et al., 1995). Attenuated total reflectance (ATR)–Fourier transformed infrared (FTIR) spectroscopy offers the possibility for in situ spectroscopy investigations of adsorbed oxyanions at metal (hydr)oxide–water interfaces (Tejedor-Tejedor and Anderson, 1990; Biber and Stumm, 1994; Su and Suarez, 1995, 1997; Hug, 1997).
Zeltner and Anderson (1988) reported cylindrical internal reflectance (CIR, a type of ATR)–FTIR spectra of goethite with adsorbed carbonate species, but a detailed peak analysis was not performed and no assignment of bands and surface species was made. Su and Suarez (1997) were only able to observe carbonate bands in ATR–FTIR spectra of goethite suspensions with a very high concentration of 1 M NaHCO3. The observed bands were assigned to monodentate carbonate. In the same study, both the carbonate and bicarbonate species were found to be present on amorphous Fe hydroxide. The apparent pK for the protonation reaction of this carbonate surface complex was believed to be between pH 4 and 6.
In general, anion adsorption on metal (hydr)oxide surfaces is accompanied by proton uptake (Stumm and Morgan, 1996). The proton stoichiometry of anion adsorption has been used to infer which surface species and type of surface complex might be present at the mineral surface. Schulthess et al. (1998) determined the proton stoichiometry of the bicarbonate adsorption reaction on Al oxide. This macroscopic proton stoichiometry of bicarbonate adsorption on hydrated 
-Al2O3 has been related to the structure and speciation of adsorbed carbonate (Wijnja and Schulthess, 1999). Attenuated total reflectance–Fourier transformed infrared spectra indicated that only the monodentate inner-sphere complexed CO3 surface species is present at the surface in the pH range of 5.2 to 7.2. Diffuse reflectance infrared Fourier transformed (DRIFT) spectroscopy spectra indicated the existence of extra protonated surface groups associated with the adsorbed CO3. It was proposed that the bicarbonate adsorption reaction was accompanied by a concurrent proton adsorption reaction of a surface group. This was found to be consistent with the proton stoichiometry of approximately 1:1 for the bicarbonate adsorption determined in titration experiments (Schulthess et al., 1998).
Recently, it has been demonstrated that the proton stoichiometry of oxyanion adsorption is strongly related to the electrostatic interaction of the oxyanion with the surface (Rietra et al., 1999). Modeling of proton–anion adsorption data shows that the proton–anion stoichiometry is primarily determined by the charge distribution of an oxyanion at the mineral–water interface. This charge distribution can be determined by applying the Pauling charge distribution in a structure of an adsorbed complex. In this way, the proton stoichiometry can be related to the structure of the adsorbed oxyanion complex. It was suggested that proton coadsorption occurs on other surface groups that are not directly involved in anion complexation.
The objective of this study was to determine the mechanisms and the proton stoichiometry of the (bi)carbonate adsorption at the goethite–water interface. The speciation and coordination state of adsorbed carbonate surface complexes were identified using ATR–FTIR spectroscopy. The spectra were collected from systems with carbonate concentrations of 
1 mM, which is similar to carbonate concentrations found in natural environments. The effect of bicarbonate adsorption on OH-stretching vibrations of goethite surface groups was determined by using DRIFT spectroscopy. The proton stoichiometry of the adsorption of bicarbonate was determined using a pH-stat technique. The relation between the structure of the adsorbed carbonate and the proton stoichiometry and possible adsorption reactions will be discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
30 g L-1) was stored at room temperature. The BET surface area was 76 ± 3 m2g-1. X-ray diffraction data indicated the presence of goethite (Wijnja, 1999). Fourier transformed infrared and Raman spectra indicated the predominance of goethite (Wijnja, 1999; Wijnja and Schulthess, 2000b).
FTIR Spectroscopy
The ATR–FTIR spectra were collected using a Mattson FTIR, Model GL 6020 equipped with an MCT detector (Mattson Instruments, Madison, WI). The ATR accessory was the Squarecol (Specac, Smyrna, GA) liquid ATR cell.
The collection of ATR–FTIR spectra of goethite followed a procedure similar to that described by Peak et al. (1999). Goethite was deposited on the ZnSe crystal of the Squarecol (Specac) liquid ATR cell. A 25-µL aliquot of a 150 g L-1 goethite suspension was injected into a droplet of 0.01 M NaCl (prepared with CO2-free Milli-Q water) on one side of the crystal (6 by 50 mm) and evenly spread across the surface of the crystal. The deposited suspension was allowed to dry under a pure air (CO2-free) atmosphere and subsequently rinsed to remove any loose particles by allowing a drop of 0.01 M NaCl to move across the deposit. Once the deposit was dry again, it was placed in the cell holder, followed by the spectroscopy measurement described below.
The ATR–FTIR spectra of the adsorbed CO3 species on the goethite were obtained by first collecting a background spectrum of the cell containing the ZnSe crystal with deposited goethite film filled with 3 mL of 0.01 M NaCl solution. Next, 30 µL of 0.1 M NaHCO3 solution was added. The pH was adjusted by adding 0.1 M HCl. The approximate pH of the solution in the cell was checked by putting a droplet of the solution on a pH test paper. The final pH was measured after completion of the measurement by transferring the solution into a plastic vial and measuring the pH using a flat-surface pH electrode (Accumet, Fisher Scientific, Pittsburgh, PA). Additional experiments were performed to confirm the pH effect on the spectra by using a flow-cell setup similar to the setup used by Peak et al. (1999). In this setup, the pH of the solution in the cell can be more accurately controlled and measured. Using this setup, the spectra of carbonate on goethite at pH 6.5 and 4.8 were determined and confirmed the spectra collected under the same conditions using the earlier-described procedure. For the measurements in D2O, all stock solutions were prepared in D2O. After mixing with a pipet and reacting for 5 min, a sample spectrum was collected. The final spectra were the result of 2000 coadded scans collected at 4 cm-1 resolution.
DRIFT Spectroscopy
The DRIFT spectra were collected with a Perkin-Elmer FTIR 1600 spectrophotometer (Perkin-Elmer, Norwalk, CT) and a Perkin Elmer diffuse reflectance (DR) sampling cell following the procedure described by Wijnja and Schulthess (1999). After centrifuging the goethite suspensions, an aliquot of the oxide paste was spread on a paper filter (Whatman no. 1) and allowed to dry 30 to 45 min in a desiccator that was purged with CO2-free air. A 0.010-g sample of this dried oxide was mixed gently with 0.30 g of KBr using a pestle and mortar and analyzed in the DR cell. The KBr powder was used as a background. In addition to each suspension sample containing adsorbed carbonate, a reference suspension without carbonates but with the same pH and similar ionic strength was prepared. The spectrum of the reference Al-oxide was subtracted from the spectrum of the goethite with adsorbed carbonate. The spectra were the result of 64 scans.
pH-stat Experiments
The coadsorption of protons with CO3 adsorption on goethite was measured at pH 6 using a pH-stat technique. Titration experiments were performed using a Titrino 716 autotitrator (Metrohm, Herisau, Switzerland). Titration samples were prepared in a 60-mL (nominal) Nalgene polyethylene bottle, consisting of a goethite suspension (0.304 g per sample) with a background salt concentration of 0.011 M NaCl and an initial total volume of 33.72 mL. The same systems but without NaCl addition were also analyzed. Atmospheric exposure was minimized by tightly covering with parafilm after the pH electrode (Accumet) and the titrator buret had been inserted in the container. This suspension was continuously stirred using a magnetic stirrer bar and titrated to pH 6 with standardized 0.1 M HCl. Subsequently, a 0.35-mL aliquot of a 0.1 M NaHCO3 was injected into the suspension. The titrations showed an initial fast reaction and stabilization after 5 to 8 min. The titration was stopped after 20 min and the suspension was transferred into a centrifuge tube and subsequently centrifuged for 15 min at 20000 x g. The supernatants were analyzed for the remaining concentration of CO3 using a TOC/TIC analyzer (TOC-5000, Shimazdu, Braintree, MA) and CO3 adsorption was determined according to the procedure outlined by Schulthess et al. (1998). Blank experiments were performed in 0.011 M NaCl solutions.
The proton coadsorption was determined based on the known volumes and concentrations of added solutions, the measured pH, and the measured concentration of the remaining CO3 in solution. The amount of adsorbed protons is found from the net difference between the amount of protons added and the change in the amount of protons remaining in solution after equilibration (Hsol) (Hiemstra and Van Riemsdijk, 1996). Since carbonate is added in the form of a NaHCO3 solution and the proton stoichiometry of the adsorption of bicarbonate is to be determined, the HCO3 species has to be considered as the reference species. The amount of protons added is then the amount present in the HCl titrant. The change in Hsol can be calculated from the speciation of the remaining CO3 in solution according to:
 | (1) |
V the total volume of the added titrant and NaHCO3 solution, and CO2(g) the amount of CO2 degassed. The V(H+ - OH-) term is negligible in our systems. The concentration of the carbonate solution species at pH 6 was determined using the pKa value for H2CO*3 of 6.35 (Stumm and Morgan, 1996). The amount of CO2 degassing can be determined by mass balance if the amount of CO3 adsorbed is known. The amount adsorbed was determined according to the procedure outlined by Schulthess and McCarthy (1990). In this procedure, the amount of CO3 adsorbed is determined based on the remaining CO3 concentrations in the suspension and corresponding blank solution (0.011 M NaCl). Each degassed CO2 molecule is associated with the uptake of one proton by a HCO3 anion:
 | (2) |
 | (3) |
This also must be taken into account when estimating the change in the amount of protons remaining in solution. The overall proton stoichiometry was determined by the ratio of the amount of protons adsorbed and the amount of CO3 adsorbed from the goethite suspension. The reported values are the result of duplicate measurements.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
1 mM NaHCO3 in H2O and D2O. The spectra collected in H2O (Fig. 1A and B) show a doublet of bands at 1315 and 1510 cm-1 and a band at 1068 cm-1. No significant changes in the band positions are observed with a decrease in pH from 6.5 to 4.8, indicating that the same surface species was present under these conditions. The intensity of the 1510 cm-1 band appears to increase relative to the 1315 cm-1 band, which may be attributed to changes in H-bonding at the goethite–water interface. Complementary data for the spectral region of 1600 to 1700 cm-1, which is obscured by the strong H2O bending vibration, are provided by the spectra collected in D2O (Fig. 1C and D). These spectra show bands at approximately the same wavenumbers as the spectra collected in H2O. The band at the 1310 to 1320 cm-1 region has lower intensity compared with the spectra collected in H2O. The 1510 cm-1 band is broadened by the presence of a band at 1462 cm-1. No significant bands are observed in the 1600 to 1700 cm-1 region in the spectra collected in D2O. The spectral region around 1200 cm-1 is obscured by the strong D2O bending vibration.
|
The spectra of the adsorbed carbonate species were different from the spectra of carbonate anion solution species as found by Wijnja and Schulthess (1999). The 3(CO) stretch vibration band of noncoordinated carbonate anion appears to be split into two bands at 1315 and 1510 cm-1, and the infrared inactive mode of the free carbonate anion at 1068 cm-1 (CO-stretch) also becomes active in the spectra of adsorbed carbonate. This splitting of the 3(CO) stretch band into two bands and the appearance of the 1068 cm-1 band indicate a lowering of the symmetry of the carbonate ion (Fujita et al., 1962; Gatehouse et al., 1958) (see also Table 1). This indicates that adsorbed carbonate is directly coordinated to the surface of goethite through formation of an inner-sphere surface complex. The coordination state of the adsorbed carbonate species can be determined based on the degree of splitting of the 3(CO) stretch vibration [(CO)] described by Wijnja and Schulthess (1999). This will be discussed in detail later. The vibrational characteristics of the free carbonate anion and complexed carbonate are summarized in Table 1. The band at 1315 cm-1 can be assigned to the symmetric OIICOII stretching vibration, the band at 1510 cm-1 to asymmetric OIICOII stretching vibration, and the band at 1068 cm-1 to the C–OI stretch of adsorbed carbonate anion. In D2O, the asymmetric OIICOII stretching band is partially shifted to 1462 cm-1, while the symmetric OIICOII stretching vibration at 1319 is reduced in intensity compared with the spectra collected in H2O.
|
1510 cm-1). Lercher et al. (1984) also observed a shift of the symmetric C–O stretching vibration and an unchanged asymmetric C–O stretching vibration of carbonate on Al–Mg mixed oxides. The shift to lower wavenumbers of the symmetric C–O stretching vibration of adsorbed carbonate on goethite results in a greater split between the symmetric and asymmetric C–O stretching vibrations [(CO)]. As mentioned earlier, the magnitude of (CO) is used to determine the coordination state of adsorbed carbonate. As pointed out by Wijnja and Schulthess (1999), the polarizing power of the metal ion should be taken into account in using the (CO) criterion for distinguishing between monodentate and bidentate coordination. Based on the correlation reported by Jolivet et al. (1982), it can be determined that carbonate complexes with Fe3+ would have a somewhat larger (CO) than complexes with Al3+ and Co3+ due to the slightly higher polarizing power (9.2 for Fe3+ vs. 8 for Al3+ and Co3+). Based on this and the criteria for assigning the coordination state outlined by Wijnja and Schulthess (1999) (see also Table 1), the observed difference between the two bands [(CO)] of 140 to 195 cm-1 can be interpreted as indicative of a monodentate coordination state of carbonate on the goethite surface. This conclusion agrees with the results of Russell et al. (1975), who observed bands at similar wavenumbers in spectra of CO2 adsorbed on goethite that was exposed to air. Fourier transformed infrared studies of carbonates on hydrous ferric oxides also showed bands indicative of monodentate coordinated carbonate (Harrison and Berkheiser, 1982; Nanzyo, 1986) (Table 1). Monodentate coordinated carbonate surface species were also found to be present in our earlier study with aged -Al2O3 (Wijnja and Schulthess, 1999) and other IR studies on the interaction of bicarbonate with Al hydroxides (Serna et al., 1977; Su and Suarez, 1997). The effect of adsorbed carbonate on the surface OH groups of goethite was evaluated by using DRIFT spectroscopy to obtain spectral information in the OH-stretching region. Tejedor-Tejedor and Anderson (1986) observed the surface hydroxyl bands of goethite in DRIFT, but these bands did not show up in ATR–FTIR spectra. Infrared studies have indicated the presence of three major types of surface hydroxyl groups, namely singly, doubly, and triply coordinated surface hydroxyls (Russell et al., 1974; Sun and Doner, 1996). Changes in the spectra of the OH-stretching region upon carbonate adsorption can best be revealed by obtaining a difference spectrum (Turek et al., 1992; Wijnja and Schulthess, 1999). Figure 2 shows the difference spectrum of goethite with adsorbed carbonate including the region above 3200 cm-1. The difference spectrum shows the asymmetric CO-stretching vibration at 1510 cm-1. The symmetric CO-stretching vibration is observed at a slightly lower wavenumber of about 1309 cm-1 compared with 1315 cm-1 in the ATR–FTIR spectra in Fig. 1. Although there is a small increase in the difference between the two CO-stretching bands in the DRIFT spectrum, these bands still indicate a monodentate coordinated carbonate anion.
|
The increased H-bonding in the presence of adsorbed carbonate may also indicate that additional protonated surface groups coexist with adsorbed carbonate. The presence of extra protonated surface groups suggests that protons coadsorb on surface groups that are not directly involved in the complexation with carbonate.
Proton Coadsorption with Carbonate
Combined with spectroscopic data, the proton stoichiometry can provide useful additional information about the adsorption mechanism and effects on surface protonation. The proton coadsorption with the adsorption of bicarbonate was determined by using a pH-stat technique. In systems with low ionic strength, the proton stoichiometry of the bicarbonate adsorption appears to be 0.86 (± 0.02):1. While no NaCl background salt was added to these systems, the addition of NaHCO3 and the HCl titrant resulted in a total Cl concentration of 8 x 10-4 M and a total NaHCO3 concentration of 1 mM. In systems with a background salt concentration of 0.011 M NaCl, the proton-to-bicarbonate ratio was found to be 0.54:1. These data confirm that protons in solution are being consumed and agree with the microscopic data of extra protonated surface groups (Fig. 2). The numerical value of the proton stoichiometry appears to be sensitive to the ionic strength. A possible explanation for this phenomenon will be discussed later.
First, the relationship between the structure of carbonate surface complexes and the proton stoichiometry will be discussed. To compare the results from this work with the data from Rietra et al. (1999), the experimental proton stoichiometry was expressed per carbonate species instead of bicarbonate species. The proton-to-carbonate values were 1.86:1 and 1.54:1 for the systems with low ionic strength and 0.011 M NaCl, respectively. The FTIR data in this study indicated monodentate carbonate surface species. In this structural arrangement, one of the three oxygens of the carbonate anion was bonded to the surface. Using the Pauling charge distribution for adsorbed complexes as described by Rietra et al. (1999), the charge attribution to the surface plane in a monodentate carbonate complex would be -0.67. This relative low charge attribution to the surface would relate to a proton-to-carbonate proton stoichiometry value in the range of 0.8:1 to 1:1. The experimental values found in the present study were much higher than the predicted value. This may suggest that more charge was attributed to the surface than based on the structure of adsorbed carbonate found in the present study. The higher proton coadsorption of carbonate compared with other bivalent oxyanions may also indicate that carbonate had a different interaction effect with the goethite surface that results in higher coadsorption of protons. This may have been due to a specific structural arrangement at the goethite surface that facilitates charge redistribution, as suggested by Russell et al. (1975).
Electrophoretic mobility (EM) data of goethite with various different adsorbed oxyanions also showed a different effect of carbonate compared with the other bivalent oxyanions SeO4 and SO4 (Wijnja, 1999). At pH 6, the SeO4 adsorption density (
SeO4) of 0.6 µmol m-2 resulted in an EM value of 0.4 cm2 V-1 s-1 whereas a CO3 of 0.9 µmol m-2 resulted in a much higher EM value of 1.8 cm2 V-1 s-1. The higher EM value in the presence of adsorbed carbonate was consistent with the high proton coadsorption found in the presence study. The same trend in EM values was also observed for these anions on Al oxide (Wijnja and Schulthess, 2000a).
It remains to be explained why there was a decrease in proton coadsorption with increase in ionic strength. In view of electrostatic interactions, it may be possible that the increase in ionic strength affects the electrostatic interactions at the mineral–solution interface, which in turn may affect the proton coadsorption. However, Schulthess and Hu (2001) showed that proton coadsorption on Al oxide is affected by the variable adsorption of the background Cl anion. It was shown that this variable chloride adsorption results in lower experimental proton coadsorption values.
Proposed Adsorption Reaction
The formation of a monodentate inner-sphere carbonate surface complex suggests that a ligand exchange reaction with a surface OH group must have taken place. Structural considerations and IR data indicate that the different types of surface hydroxyl groups on goethite have different reativities for anion complexation (Russell et al., 1974; Sun and Doner, 1996; Hiemstra and Van Riemsdijk, 1996). These studies suggested that the singly coordinated surface hydroxyls (Fe–OH0.5-) are the most reactive in the formation of inner-sphere complexes with anions. It is therefore reasonable to assume that carbonate also forms a surface complex with this singly coordinated surface group through a ligand exchange reaction. The FTIR data in the present study cannot confirm the identity of the surface group involved in the ligand exchange reaction because of the concurrent band development caused by increased H-bonding and extra protonated surface groups. The proton coadsorption data suggest proton uptake with the adsorption of bicarbonate from solution. The increased H-bonding indicated by the DRIFT spectra are consistent with the presence of extra protonated surface groups in association with adsorbed carbonate. This suggests a coadsorption reaction of protons with surface OH groups that are not directly involved in the carbonate complexation. In the pH range of 4.8 to 7 used in this study, it can be expected that the protonation reaction only involves the singly coordinated surface groups (Hiemstra et al., 1996). The following concurrent ligand exchange reaction and proton coadsorption reaction equations are proposed:
 | (4) |
 | (5) |
These equations indicate that the adsorbed carbonate surface species is accompanied by protonated surface groups at the goethite surface. The protonated surface groups may be immediately adjacent to the Fe–OCOO- groups, but their relative locations cannot be confirmed here. Also note that the bicarbonate surface species can be formed upon dehydration of the surface, which was observed by Russell et al. (1975) under evacuated conditions.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication February 4, 2000.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-FeOOH) surfaces, and its implications for anion adsorption. J. Chem. Soc. Faraday Trans. I 71:1623–1630.-Al2O3/water interface. Spectrochim. Acta 55:861–872.This article has been cited by other articles:
|
|
 |
|
  S. Goldberg, L. J. Criscenti, D. R. Turner, J. A. Davis, and K. J. Cantrell Adsorption Desorption Processes in Subsurface Reactive Transport Modeling Vadose Zone J., August 1, 2007; 6(3): 407 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Wijnja and C. P. Schulthess Effect of Carbonate on the Adsorption of Selenate and Sulfate on Goethite Soil Sci. Soc. Am. J., July 1, 2002; 66(4): 1190 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.P. Schulthess and Z. Hu Impact of Chloride Anions on Proton and Selenium Adsorption by an Aluminum Oxide Soil Sci. Soc. Am. J., May 1, 2001; 65(3): 710 - 718. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
