| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Arcadis Geraghty & Miller International Ltd., Newmarket, UK
b Univ. of Cambridge, Department of Engineering, Cambridge, UK
c Kajima Technical Research Inst., Tokyo, Japan
* Corresponding author (ks207{at}cam.ac.uk).
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Received 22 March 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
Abbreviations: AS/SVE, air sparging and soil vapor extraction; CSL, coarse sand lens; CT, column test; FSL, fine sand lens; GS, gas chromatograph; LNAPL, light nonaqueous phase liquid; NAPL, nonaqueous phase liquid; SVE, soil vapor extraction; VOC, volatile organic compounds.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
The air sparging and soil vapor extraction (AS/SVE) technology has become increasingly popular in practice due to its simplicity in implementation and moderate capital costs relative to other conventional approaches (Johnson et al., 1993). However, air sparging involves complex physical and chemical processes. Therefore, engineering design of this technology is mainly based on empirical knowledge and the design engineer's experience.
In air sparging, the nature and extent of the air pathways determine the region of influence of the remedial process. The efficiency depends on microscopic effects such as the mode of air flow (channels or bubbles) and air channel density, as well as on macroscopic effects such as the spatial distribution of air paths. Ji et al. (1993) showed that injected air travels as stable air channels for soils with grain size of 0.75 mm or less, whereas it travels as bubbles for soils with grain size of 4 mm and greater. The transition occurs in soils with grain size of approximately 2 mm, which typically represent medium to coarse sands (Semer et al., 1998). The operational factors such as air flow rate (Leeson et al., 1995; Rutherford and Johnson, 1996), injection pressure (Ji et al., 1993), and spacing of injection and extraction wells (USEPA, 1997; Rogers and Ong, 2000) also influence the density and distribution of air channels in the subsurface. Burns and Zhang (2001) demonstrated that the average bubble size and range of size distribution increased as the injection pressure and size of the injection orifice were increased.
Site geology is considered to have a significant influence on the efficiency of air sparging systems. A symmetric air plume around the injection point has been observed in laboratory experiments with homogeneous saturated soils or glass beads (Ji et al., 1993; Elder and Benson 1999; Peterson et al., 1999; Adams and Reddy 2000), whereas in the field, injected air is more likely to travel in channels with irregular or asymmetric shape because of the heterogeneous nature of the subsurface (Ji et al., 1993; Johnson et al., 1993; Hinchee, 1994; Ahlfeld et al., 1994). Reddy and Adams (2001) investigated the removal of dissolved-phase hydrocarbon under different heterogeneous soil conditions and observed spatial variations in the hydrocarbon removal due to soil heterogeneity. When bypassing of air channels to the contaminated zone occurs, due to encountering low-permeability soils, remediation time is governed by the diffusion process of contaminants toward the air channels. Both laboratory and field-scale studies strongly suggest an advantage to operating air sparging systems in a flow interruption or pulsed mode (Heron et al., 2002; Yang et al., 2005).
Most of the past research on AS/SVE focused on removal of dissolved organic contaminants. Remediation of free-phase NAPLs in the field, compared with the contaminant dissolved in water, is further influenced by its complex spatial distribution of NAPL in multiphase systems. Aquifer heterogeneity has been shown to increase the complexity of NAPL movement and subsequent entrapment, and the spatial distribution is controlled by unstable fingering, preferential channeling, and both micro- (pore) and macroscale (layering and soil texture contrast) heterogeneity of subsurface formations (Schwille, 1988; Illangasekare et al., 1995a, 1995b). Final NAPL distribution in these complex geological environments is manifested by zones of entrapment ranging from low saturation (residual, ganglia, and blobs) to high saturation (lenses, pools, and macroscale entrapment zones resulting from capillary barriers).
Remediation of NAPL continues to produce significant engineering challenges due to the difficulties in locating the entrapped NAPLs, inability to efficiently remove NAPL from the soils, and limitations in the modeling and assessment tools that are designed to predict their fate and transport behavior (Soga et al., 2004). Few studies have been reported that assess the efficiency of air sparging on the removal of NAPLs (Rogers and Ong, 2000; Johnson et al., 1999; Waduge et al., 2003, 2004). Although there is an argument that off-site migration of NAPL can occur under the influence of injected air, Adams and Reddy (2000) state that air sparging can be successfully used to remove NAPLs with the proper designing of an air sparging system. Furthermore, they have hypothesized that the occupation of pore space in the saturated zone by injected air will decrease the relative permeability of NAPLs preventing off-site migration.
Due to the lack of data in the literature regarding the removal of NAPLs by air sparging and the difficulty of modeling both macroscopic and microscopic heterogeneity effects on air–water distribution (Thomson and Johnson, 2000), the aim of this study is to investigate the governing factors that influence the system efficiency (in particular, the amount of ultimate mass removal that can be achieved) by physical modeling. The factors considered are air-flow rate, flow interruption, subsurface heterogeneity (i.e., lens), and NAPL saturation. A series of one- and two-dimensional physical model experiments were conducted for our investigation.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
Model LNAPL
Toluene, a major constituent of gasoline and other petroleum products, was used as the model LNAPL. Toluene has high volatility and no color; its properties are listed in Table 1. Toluene was provided by Fisher Scientific International (Leicestershire, UK). For the experiments, toluene was mixed with nonvolatile, organic-soluble Red Oil O (Sigma-Aldrich Company, Dorset, UK) (0.5 g of dye in 1 L toluene) to achieve visibility during migration through the porous medium. Wilkins et al. (1995) reported that under the given mixture conditions, this dye had negligible effect on the physical and chemical properties of toluene.
|
|
Experimental Procedure
Sixteen tests were performed to investigate the effect of air flow rate, initial NAPL mass, mode of injection, and NAPL entrapment condition on the mass removal behavior (Table 2). All column tests contained a coarse Fraction B sand layer sandwiched between finer Fraction C layers, as shown in Fig. 1. The column was filled with sands under water (wet packing). During sand packing, the water level in the tank was gradually increased so that it was maintained approximately 50 to 60 mm above the sand surface. The wet packing gave a homogeneous sand condition and also provided a fully saturated condition. In dry packing, air is more likely to be trapped among sand particles, which can contribute to the creation of preferential air flows (Ji et al., 1993). Fraction C sand was placed in the column up to 250 mm from the bottom. This height was used because visual observations of trial tests confirmed that the injected air occupied the whole cross-section within 100 to 150 mm. It was therefore possible to achieve a one-dimensional uniform air flow over the cross-section at 250-mm height to flow through the NAPL source zone. A special device with a hopper, valve, and meshed outlet was used for pouring sand into the column. The use of the sand-pouring device minimized the possibility of nonuniform sand packing. The next 100 mm of soil column was filled with coarser Fraction B sand. The rest of the column was then filled with Fraction C sand. Both mass and volume of the sands were measured, and the average porosities of the Fraction B and C sand layers were 0.394 and 0.37, respectively.
|
Air was injected into the soil model through the air sparger placed at the middle of the base of the soil column. A control panel featuring regulator, gauge, and flow meter was used to control and monitor both air injection pressure and flow rate. This helped to avoid soil fracturing. The injected air passed through the soil column and exited from the top outlet. The effluent air was released into a fume hood. Three different air flow rates of 1, 2, and 5 L min–1 were applied.
All the tests except CT1 and CT2 were conducted by a sequence of flow interruption, 10 h of operational period followed by 14 h of shutdown period. Either two or three cycles of flow interruption were performed. In CT1 and CT2, continuous air injection was applied.
During testing, gas samples were collected from the outlet at different time intervals using a gas-tight 100-µL syringe. The gas concentration of toluene was measured by direct injection of gas samples into an Agilent 6850 Series 11 gas chromatograph (GC) (Agilent Technologies, West Lothian, UK). The gas sample volume was 50 µL. The GC was equipped with a flame ionization detector and a capillary column of 0.32-mm i.d, 0.5-µm film, and 30-m length. External standards were prepared for each analysis to generate a calibration curve. The detection limit of the method was approximately 10 ppb toluene.
Because the gas outlet tube had a 4-mm diam., it was assumed that the gas sample represented the average condition of whole volume of air at the corresponding time. The collected gas samples were immediately analyzed (the accuracy of gas sample could vary by ± 2 mg L–1) to monitor the gas concentration profile with time and hence to evaluate the mass removal from the system.
Two-Dimensional Tank Experiments
Apparatus
A laboratory soil tank with internal dimensional of 1200 mm (length), 800 mm (height) and 150 mm (width) was used to carry out the two-dimensional tank experiments. A schematic diagram of the tank is shown in Fig. 2 (upper). The front surface was made of toughened glass to achieve subsurface visibility during testing. The remaining sides were built with aluminum. As shown in Fig. 2a, there were two wells at the lateral sides of the tank; one was used as an extraction well, and the other was open to the atmosphere. The wells were made by combining a screen and a porous material, which prevented sand from washing into wells but still allowed water to flow. The bottom of the tank was connected to a water supply through a movable reservoir, by which it was possible to control the height of the water table inside the tank. Filter papers were placed at the bottom of the tank to prevent the fines from washing out from the tank.
|
As shown in Fig. 2, the SVE system had a pump, vacuum gauge, regulator, and flow meter, while the air sparging system consisted of a compressor, regulator, pressure gauge, flow meter, and sparger. A plastic silencer of 38.5-mm diam. and 123.5-mm effective length was used as a sparger to inject air into the soil model. The sparger was fixed at location G6 of the tank shown in Fig. 2b.
Experimental Procedure
In this study, two scenarios, (i) a coarse sand lens (CSL) in a fine sand matrix and (ii) a fine sand lens (FSL) in a coarse sand matrix, were considered at low water table (LW), middle water table (MW), and high water table (HW). Three tests (CSL-LW, CSL-MW, and CSL-HW) were performed for the CSL scenario, and two tests (FSL-HW and FSL-LW) were conducted for the FSL scenario. Table 3 summarizes the details of the experiments.
|
Upon filling to the required height before placing a sand lens, two aluminum plates, slightly narrower than the tank, were used to create a sand lens in the middle of the Fraction C sand matrix. The plates were pushed into the soil with the same spacing as the length of the lens. Care was taken to ensure accurate placement of sand on both sides without disturbing the stability of the plates. When the required height of the lens was achieved, the plates were removed and Fraction C sand was then filled to the height of the soil model. The mass of the dry sand and the gross volume of the sand pack were measured to calculate the porosity of each sand. The average porosities of Fraction C sand matrix and Fraction B and D sand lenses were 0.40, 0.44, and 0.38, respectively. The schematic diagrams of the soil models constructed for the experiments are shown in Fig. 3 and 4.
|
|
NAPL Entrapment
Coarse Sand Lens Cases
In the experiments with the coarse sand lens in a fine sand matrix (CSL-LW, CSL-MW, and CSL-HW), toluene was spilled using a line source. The line source (145-mm length, 25-mm diam., with a 2-mm split at the bottom over the complete length) was used to achieve a spill over the entire tank width to create a two-dimensional NAPL plume in the tank. The line source was buried approximately 60 to 80 mm below the soil surface. Before spilling, a rubber membrane was placed over the entire soil surface to ensure that all contaminated vapor would go into the extraction well. The system was then sealed by placing a lid on top of the tank. A total of 8 x 10–4 m3 (687 g) toluene was spilled at an average spilling rate of 8000 mm3 s–1. Once spilling of toluene was complete, the systems were left for redistribution until the movement of LNAPL had ceased.
Figure 3 illustrates the final distribution of toluene in the soil models. By changing the location of the water table in the tank, three different entrapment conditions were achieved, simulating possible conditions encountered in the field near the water table. In CSL-LW, the water table was placed at the bottom of the tank, as shown in Fig. 3a, and the majority of spilled toluene migrated through the coarse sand lens, finally resting on the fine sand capillary zone with some lateral spreading. In contrast, when the water table was placed at the midheight of the coarse sand lens (CSL-MW), most of the spilled toluene was trapped in the upper region of the coarse sand lens at high NAPL saturation as shown in Fig. 3b. This was because the lower part of the lens was fully saturated with water. At the beginning of air sparging, the visual inspection indicated that the NAPL pooled on top of the capillary fringe had a saturation greater than the residual.
In CSL-HW, the water table was placed below the lens during the spill stage, as shown in Fig. 3c, and the toluene initially pooled on the bottom interface of the two sands. The water table was then raised above the lens, and the toluene spread throughout the coarse sand lens due to buoyancy. The majority of the toluene was entrapped in the lens due to the capillary barrier effect at the upper sand interface with a NAPL saturation greater than the residual value.
Fine Sand Lens Cases
In the experiments with a fine sand lens in a coarse sand matrix (FSL-HW and FSL-LW), toluene was injected directly into the unsaturated zone of the finer lens through three syringes with 100-mm-long needles using a pump. This was done to entrap the toluene in the fine lens. A total of 600 g toluene was injected (through the sampling ports C5, C6, and C7) at a rate of 155 mm3 s–1. The water table was then raised. When the water table was raised, the toluene redistributed by buoyancy, and some came out of the fine sand lens, as shown in Fig. 4a for CSL-HW. In CSL-LW, the water table was lowered again before the operation of AS/SVE, and the toluene that came out of the fine sand lens was re-entrapped. The final distribution of toluene is shown in Fig. 4b.
Air Sparging Operation
In the AS/SVE stage, air was injected at a specified flow rate of 0.005 m3 min–1 from the sparger and extracted at a flow rate of 0.012 m3 min–1 from the extraction well. As the effective injected airflow was 0.005 m3 min–1 at pressure gauge reading of 0.04 bar, the flow through the well open to the atmosphere was 0.00696 m3 min–1. Ten cycles of 10 h AS/SVE operational periods followed by 14 h of shutdown were conducted. Toluene concentrations in the effluent gas from the extraction well were measured to evaluate the total mass removed from the system.
|
Results and Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
To show the air flow rate effect on mass removal of free-phase hydrocarbons, Fig. 5a through 5c plot the variation of fractional mass remaining with different flow rates (1, 2, and 5 L min–1) for 10, 50, and 100 mL of toluene, respectively. Remaining mass fraction was defined as the ratio between the remaining toluene mass within the system to the initial injected toluene mass into the system. Injected air volume was normalized by the volume of injected air needed as a theoretical minimum for equilibrium volatilization the toluene mass within the system.
|
At a later stage of air sparging, the removal curves deviated from one another where the mass removal was limited by mass transfer from NAPL phase to gaseous phase. The mass removal for a given volume of injected air, particularly at the later stage, was greater with a low flow rate, as shown in Fig. 5, emphasizing the fact that mass transfer from NAPL to gas phase was rate limited. Wilkins et al. (1995) and Yoon et al. (2002) suggested that NAPL–gas phase mass transfer is highly dependent on gas velocity (i.e., rate limited). Consequently, it can be argued that the resident time or contact time between NAPL and air also plays a major role in mass partitioning from NAPL to gas phase, in addition to the limited contact area issue.
For a given set of air flow rate and initial NAPL mass, the mass removal rate when the NAPL was initially entrapped above the water table was similar to that when it was below the water table. This indicates that the air sparging created a similar pattern of microscopic air channels in both scenarios.
Effect of Air Injection Flow Rate on Final Mass Removal
Figure 5 shows that at the end of the continuous air injection (CT1 and CT2) and of the first air injection cycle (CT3–CT16), the remaining mass fraction reached asymptotic values, which were less dependent on injected air flow rate. These values were approximately 0.44, 0.42, and 0.28 for 10, 50, and 100 mL of toluene, respectively. Several previous laboratory studies (Adams and Reddy, 1999, 2000; Reddy et al., 1999) with dissolved-phase contaminants reported an improved mass removal with a higher air injection flow rate up to a threshold value. In contrast, the current data showed a negligible effect of air injection flow rate on mass removal from free-phase NAPL. Similar observations were made by Johnson et al. (1999), Rogers and Ong (2000), and Braida and Ong (2000). An increase in injection flow rate increases air channel density inside the aqueous phase and enlarges the air–water interfacial contact area. For contaminants dissolved in the aqueous phase, the removal rate therefore increases. On the other hand, the spatial distribution of free-phase NAPL in the pores is not uniform due to local pore-scale heterogeneities, and there is limited NAPL surface available for the air to get in contact with. Hence, it is considered that there was not much change in the air–NAPL interfacial contact area within the investigated range of air flow rate.
Effect of Flow Interruption
An increase in mass removal was observed in most experiments by flow interruption or pulsing as indicated by the arrows in Fig. 5. The exception was at the beginning of the second cycle with 10 mL of toluene, in which a very limited increase of mass removal was observed (Fig. 5a). During flow interruption (or cyclic air injection), it is possible for NAPL to mobilize and redistribute due to the movement of water. As reported by Ji et al. (1993), any microscale variations in pore structure alter the air flow pattern in the porous medium. Therefore, changes to the system due to flow interruption include NAPL redistribution inside the soil model and the formation of new air channels, resulting in new contacts between free- or dissolved-phase NAPL and air and enhancing mass partitioning.
Flow interruption also enhances the contact of toluene NAPL with decontaminated water, improving dissolution by introducing a higher concentration gradient than that with partially contaminated water. Furthermore, each shutdown period allows for diffusion and reestablishment of system equilibrium. As a result, the gas concentration increased when the system was resumed after the shutdown period.
Threshold NAPL Saturation
The very small mass removal rate in the second cycle in Fig. 5a and the third cycle in Fig. 5b and Fig. 5c suggests that the mass of toluene left in the system at the beginning of these cycles might have been close to the threshold value, beyond which there would be no further increase in mass removal by the interruption of air injection. This behavior leads to the question, "Is there a mass removal limit at which NAPL cannot be removed even after many flow interruption cycles?"
The toluene saturation was estimated using the height of toluene distribution (which was measured based on the visual observation against the column wall) in the Fraction B sand layer and the amount of toluene in the layer. Following the redistribution of toluene during the shutdown period, it is assumed to be appropriate to treat each cycle as its own column test. The fractional mass removal (removed mass/initial mass) was calculated at different cycles. The initial mass for each cycle was determined based on the mass left from the previous cycle. This allowed combining the data obtained from the tests with different initial toluene volumes and different cycle stages. Each air injection cycle was stopped after a 10-h air injection period in which mass removal flattened, indicating insignificant variation of mass removal with time.
Figure 6 shows the variation of fractional mass removal with the percentage of NAPL saturation left in the soil. A higher mass removal is possible at higher NAPL saturations. However, the fractional mass removal declines with the decrease of the NAPL saturation. The fraction mass removal becomes zero below a NAPL saturation of 4%, suggesting that it is not possible to remove the NAPL through air sparging beyond this value even after many flow interruption cycles.
|
Tank Experiments
Air Flow Patterns
As air was injected from the bottom center of the tank (G6 in Fig. 2b), it traveled through the porous medium in pore-scale air channels in all the tests. The air flow pattern was obtained from visual observations made against the front face of the tank.
The air emanating from the point of injection expanded laterally in the homogeneous uniform Fraction C sand matrix while moving upward toward the water table. The zone of influence in the fine sand matrix increased gradually with height until it reached the coarse sand lens in the CSL experiments. Once injected air entered into the more permeable sand lens, air channels passed vertically through the medium, showing no lateral expansion. The observed shape of the zone of influence was roughly symmetrical around the injection point, similar to the observations made by Ji et al. (1993), Adams and Reddy (2000), and Heron et al. (2002).
In the FSL experiments, the injected air moved upward with lateral spreading in the coarse sand matrix until it encountered the bottom interface of the fine sand lens. The air plume that impinged on the lower surface of the low permeability sand created a zone of high air saturation below the fine sand lens and circumvented the lens, similar to the observation made by Ji et al. (1993), McCray and Falta (1997), and Reddy and Adams (2001). However, the air plume was still roughly symmetric around the point of air injection.
Effect of Source Zone Entrapment Conditions
The mass removal from the physical models was computed from the measured gas concentration–time curves for a given extraction flow rate. The fractional mass removal is defined as the amount of NAPL mass removed compared to the initially spilled mass. The extracted air volume was normalized by the theoretical air volume required to remove the NAPL mass, assuming equilibrium condition. The changes in fractional mass removal with the normalized air volume are shown in Fig. 7.
|
When the CSL and FSL experiments are assessed separately, it can be seen that mass removal efficiency varied by about 10% in each scenario when the location of the water table was varied. The observed variation of mass removal can be described by the effect of moisture content around the entrapped NAPL source as a higher water saturation will limit mass removal by SVE. A lower mass removal was observed in CSL-MW and CSL-HW than in CSL-LW, where most of the toluene was trapped below the saturated zone. A similar phenomenon may explain the lower mass removal of FSL-HW compared with FSL-LW. However, the variation of final mass removal due to different water table locations was secondary compared with that due to soil heterogeneity. The limited influence of water table location was also observed in the column experiments, as described above.
Effect of Flow Interruption
Flow interruption or pulsing was applied in all physical model tests. Similar to the one-dimensional model tests, an increase in mass removal due to the cyclic air injection was observed in all the tank experiments. The spikes in temporal toluene gas concentration at the extraction well, as shown in Fig. 8, represent the enhancement of gas concentrations during the shutdown periods. Within the CSL experiments, CSL-HW gave the smallest rebound since most of the NAPL was entrapped below the water table, whereas CSL-LW had the largest rebound because the majority of the NAPL was left in the unsaturated zone. For the FSL experiments, the rebound of toluene gas concentration rapidly decreased when the water table was higher. The entrapped toluene in the saturated zone of FSL-HW restricted the contact between NAPL and air during the shutdown period. As a result, the amount of mass transferred to the gas phase in FSL-HW was smaller during the shutdown period than FSL-LW in the later stage of air sparging. However, at the beginning of air sparging, there was a considerable concentration increase in FSL-HW after the shutdown period. This may have been due to the presence of toluene close to the upper boundary of the capillary fringe.
|
To illustrate the usefulness of flow interruption data to estimate the ultimate mass that could be removed, the rebounds in gas concentration at the extraction well before and after the shutdown period are plotted against the amount of toluene left inside the soil model for all tank experiments in Fig. 9. The data show that the amount of concentration rebound after a 14-h shutdown period decreased linearly with the amount of toluene left in the system. It appears that there is a threshold toluene mass in which the rebound of gas concentration after flow interruption becomes very small and mass removal enhancement by flow interruption is ineffective. A plot such as Fig. 9 can be useful in the field to assess the effectiveness of the air sparging operation.
|
|
Conclusions |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
Results of the column experiments suggest that the initial mass removal rate from NAPL is directly proportional to the injection air flow rate, suggesting near-equilibrium conditions, but is limited to the volatilization of NAPL that is in direct contact with microscopic air channels. At the later stage of air sparging, the mass removal rate became rate limited so that a higher mass removal, for a given air volume, could be expected with a lower air injection rate.
Flow interruption was used to enhance the mass removal of LNAPLs entrapped around the water table by AS/SVE. However, the fractional mass removal rate became zero at a threshold NAPL saturation beyond which NAPL could not be removed by air sparging even after many interruption cycles. The column experiments show that the threshold saturation of toluene was about 4%. Similar observations were made in the tank experiments. The rebound concentration when the system was resumed following the shutdown period can be used as a parameter to illustrate the potential degree of enhancement due to the flow interruption for field applications.
The data presented in this paper are potentially useful for validating numerical models of air sparging processes. Some attempts were made by Waduge (2003) as a separate study. Results indicate the necessity of incorporating a rate limited mass transfer model into the NAPL–gas mass transfer model. The model should incorporate of NAPL saturation with a threshold value in which NAPL cannot be removed, even after many interruption cycles.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and MethodsResults and DiscussionConclusionsREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
