Researchers are using NMR spectroscopy to understand the factors that keep soils healthy.

For millennia, farmers have handed down lore to their children, offering such life-sustaining knowledge as whether you’re working the right soil, what crops to rotate and how often, and when to plow things under. But only recently have researchers started to wonder what components go into making a fertile and sustainable soil, and these questions couldn’t have arisen at a better time. As corporate farming has geared up and farmers are trying to maximize their return on investment—getting multiple harvests from a single field, relying more heavily on fertilizers, pesticides, and herbicides—we are just now beginning to see the toll these practices are taking on the health of the soil and the environment.

The Dirt on Dirt
What gives the soil its richness? Soil health is largely dictated by the content of its soil organic matter (SOM; see box, “Content in Context”), which performs a variety of functions. From a physical perspective, SOM lends structure to soil, keeping it from being too compact and thereby facilitating water retention, aeration, and stability. This structural effect also inhibits erosion because water is able to percolate through pores in the soil rather than stay near the surface where it washes away valuable resources. Biologically, SOM facilitates plant and microbial growth by providing a source of carbon, nitrogen, and phosphorus. Because of its binding and pH-buffering capacity, SOM also alters the effects of anthropogenic substances such as herbicides and pesticides. In fact, SOM dictates, as much as, if not more than climate, the type and quality of plants growing on a given plot.

But this physical and biological balance is easily tipped by procedures such as field tillage and drainage. When exposed to air, SOM decomposes more quickly than under anaerobic conditions, and the quality of the soil decreases. This problem is aggravated by the excessive use of fertilizers, which add nitrogen to the soil and facilitate decomposition. Ironically, the short-term fertility gain is offset by a long-term fertility loss. Although the process of soil degradation is a long one, it is a difficult process to reverse.

To precisely understand how these various components work and interact requires an analysis of the compounds at the molecular level.

Elemental Analysis
In determining the health of a given soil plot, elemental analysis is necessary to quantify the amount of various soil components such as carbon, nitrogen, phosphorus, and sulfur. These studies also provide valuable information about possible contaminants within the soil. Similarly, elemental analysis can give you a hint as to the type of SOM that you are examining as C:N ratios correlate with the amount of humic and fulvic acids (some of the polymeric organic compounds) in the soil.

“You really need elemental analysis,” says Caroline Preston, a soil researcher with the Pacific Forestry Centre (Victoria, BC). “You want to look at the C:N ratio, how that’s changing, and maybe some other indices of nutrient availability and quality.”

Elemental analysis falls short, however, in the fact that it cannot be used to identify the compounds within the SOM, information that is critical for a complete understanding of how such compounds interact with each other and with anthropogenic compounds.

Pyrolysis-GC-MS
A relatively common method for analyzing SOM is pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS). In this technique, a soil sample is heated on a filament, and the volatile products are separated by gas chromatography. The components are then detected by mass spectrometry, from which the compounds are identified on the basis of their ion peak and fragmentation pattern.

This method has been used extensively by Patrick Hatcher and his colleagues at Ohio State University (Columbus) to study polycyclic aromatic hydrocarbons (PAHs) in aquatic sediments. In a recent study of the San Diego Bay area, Hatcher’s group noted the influence of surrounding SOM, specifically lignins, to the sedimentary organic matter and the possible role that this might play in PAH contamination (1). But for all of its analytical speed, Py-GC-MS has limitations.

“Although pyrolysis-GC-MS tells you something, it doesn’t capture the whole [SOM] profile,” offers Preston. Part of the problem stems from the fact that MS suffers from molecular weight limits. As such, the SOM has to be broken into smaller components through the thermal degradation step, which can result in the loss of some molecular species or the generation of artifacts.

There have been big improvements in the technique, says Preston, “but it tends to not pick up everything. Even in terms of sampling, you only put micrograms on the wire, so I always worry about sampling error.”

NMR
One analytical technique that has really started to make inroads in SOM analysis is nuclear magnetic resonance (NMR). “The big thing about NMR,” offers Preston, “is that it shows you all of the carbon, although not necessarily in a quantitative manner. If you’re going to do really expensive and time-consuming things like wet chemical analysis, NMR can really give you a guide of what to look for.”

One of the more common NMR techniques used to study SOM is cross-polarization with magic-angle spinning (CP-MAS) NMR, a technique used by Preston in her studies of forest SOM. Cross-polarization refers to the transfer of magnetic polarization from one nuclear type to another. This is typically performed to increase the signal from low-frequency, low-abundance nuclei (e.g., ¹³C) using the polarization of high-frequency, high-abundance nuclei (e.g., ¹H). MAS is a process in which a sample is spun rapidly at an angle of 54.7° relative to the static field. This magic angle corresponds to the diagonal of a cube, and the spinning averages the signal over all three dimensions.

In 1999, Daniel Kile and colleagues at the U.S. Geological Survey (Boulder and Denver, CO) used CP-MAS NMR to examine the correlation between the sorption of nonionic compounds such as PAHs and specific SOM constituents (2). Previous studies indicated that the ability of a soil to sorb PAHs (its partition coefficient or K_oc) correlated with the aromaticity of the soil humic fractions. Kile’s group, however, found a poor correlation between the level of aromatic carbon and the K_oc of soil with carbon tetrachloride, their model nonionic compound. Instead, they found that the combination of aliphatic and aromatic carbon content determined how well the compound partitioned to the soil, suggesting that the polar quality of the carbon was a larger factor.

Two years later, however, studying the binding of the nonionic pesticides carbaryl and phosalone to various soils from Pakistan and Australia, Riaz Ahmad and colleagues at the University of Adelaide and CSIRO Land and Water (Glen Osmond, Australia) found a striking correlation between aromatic carbon and the K_oc of the pesticides (3). But this finding was tempered by differences between the sorption properties of many soil samples. The finding prompted the warning that “the use of an average soil K_oc value of a pesticide is insufficient to assess a soil’s ability to sorb the pesticide. Another key to the interpretation of these sorption results may be found in the conformation of the molecular components of the SOM.”

Of course, soil analysis isn’t strictly limited to solid-state NMR. There are numerous examples in which solid- and solution-state experiments were combined. Recently, Nathalie Mahieu and her colleagues at the University of London and the International Rice Research Institute (Manila) used ¹³C- and ¹⁵N-CP-MAS NMR in conjunction with ³¹P solution NMR to determine the effects of repeated annual cropping and flooding of rice fields on the humic acid fractions (4). As rice farming has intensified in Asia over the past 30 years, farmers are trying to get two or three crops from a field in a single season. This has meant that fields remain submerged for longer periods of time and represents a potential drain on nutrient resources.

Mahieu’s group found that regular soil aeration—accomplished through repeated field drainage—facilitated the processing or humification of the SOM. This process was counteracted, however, by repeated cropping because there was a distinct buildup of less-decomposed matter with a coincident depletion of the more labile compounds.

A Hybrid Technique
In many SOM studies using NMR, organic samples are extensively extracted, separating the organic components from the clay to which they are normally attached. Some researchers are concerned that they might be looking at artificial reactivities produced by the reaction procedure. Solid-state techniques such as CP-MAS NMR address some of these concerns, according to André Simpson and his colleagues at Mississippi State University (Starkville) and Bruker Analytik (Rheinstetten, Germany), but the method does not “distinguish structures at the solvent–surface interface, and unlike solution-state experiments, cannot distinguish signals due to interactions arising through space (nuclear Overhauser effects, NOE) from couplings which occur through bonds.”

To get around these problems, Simpson’s group used an NMR method that is something of a hybrid between solid- and solution-state experiments: high resolution- (HR-) MAS (5). This is a solid-state method that uses a solution-state instrument but allows the analysis of materials that swell, become partially soluble, or dissolve in a solvent even when solid components remain. Components in contact with the solvent become NMR visible, and their dipole interactions decrease.

The researchers used either deuterium oxide (D₂O) or deuterated dimethyl sulfoxide (DMSO-d₆) as the solvent. With D₂O, HR-MAS identifies only constituents at the solvent–soil interface, whereas DMSO-d₆ disrupts the soil aggregates and therefore offers information on the hydrophobic surfaces that might otherwise be buried. The two samples showed striking similarities with regard to the presence of strong aliphatic carbon peaks as well as those for sugars and amino acids, but the D₂O-treated sample lacked peaks in the region of aromatic carbons that were present in the DMSO-d₆-treated sample. This finding suggests that the aromatic moieties are buried within the soil aggregates, which may mean that they are less involved in the binding of nonionic soil contaminants as was argued earlier with extracted humic samples.

HR-MAS also allowed the researchers to analyze the interaction of the soil sample with the herbicide trifluralin, extensively used on U.S. cotton crops (Figure 1). These kinds of experiments will help to explain how and why such compounds interact with various soils and may facilitate the development of better xenobiotics.

Tomorrow’s Experiments
The future of NMR in soil science will be interesting, says Preston. There has been a steady move to higher-field magnets in CP-MAS, but this has led to bigger spinning sidebands, spectral peaks that are separated from true peaks by integer multiples of the sample spinning frequency. This effect is due to the nuclei experiencing altered field strengths caused by problems with magnetic inhomogeneity, chemical shift anisotropy, or dipole or quadrupole coupling. To some extent, the use of higher speed spinning probes is eliminating sidebands, but the spinning probes lower the cross-polarization efficiency. New techniques are being developed to get around this problem.

Preston also believes that the use of multidimensional solution-state NMR has not been fully explored, but that this technique will be more useful for the study of dissolved organic matter and pesticide residue work. Along this line, Simpson’s group has started to explore the use of ¹H-¹H total correlation spectroscopy to help identify the major components at the soil–aqueous interface of whole soil. Similarly, Klaus Schmidt-Rohr and colleagues at Iowa State University (Ames) and the University of Massachusetts (Amherst) are using ¹H–¹³C heteronuclear correlation spectroscopy to improve understanding of the structures of specific humic acids (Figure 2, reference 6).

The Challenge
“I think that the biggest limitation right now in applying NMR is the lack of people in the soil field who understand NMR,” says Preston. “Conversely, the people who run the machines don’t understand the SOM. There is a lack of opportunity for genuine interdisciplinary graduate student training, including the crucial hands-on aspect.”

To address this issue, Preston has spent a lot of time trying to make NMR technology more accessible to soil scientists. She published a paper last year that she hopes can be understood by non-NMR specialists (7). “There’s not that much out there in the literature that they can understand,” she laments. “There was one book on soil NMR that was published a long time ago, but right away it jumps into all this math. [The NMR] has to be on a very practical level that the people in soils can understand.”

If Preston is successful, the future of NMR in soil science will be a bright one.

References

1. Deshmukh, A. P.; Chefetz, B.; Hatcher, P. G. Chemosphere 2001, 45, 1007–1022.
2. Kile, D. E.; Wershaw, R. L.; Chiou, C. T. Environ. Sci. Technol. 1999, 33, 2053–2056.
3. Ahmad, R.; Kokana, R. S.; Alson, A. M.; Skjemstad, J. O. Environ. Sci. Technol. 2001, 35, 878–884.
4. Mahieu, N.; Olk, D. C.; Randall, E. W. J. Environ. Qual. 2002, 31, 421–430.
5. Simpson, A. J.; et al. Environ. Sci. Technol. 2001, 35, 3321–3325.
6. Mao, J.-D.; Xing, B.; Schmidt-Rohr, K. Environ. Sci. Technol. 2001, 35, 1928–1934.
7. Preston, C. M. Can. J. Soil Sci. 2001, 81, 255–270.

   ---

Randall C. Willis is a senior associate editor of Today’s Chemist at Work. Send your comments or questions regarding this article to tcaw@acs.org or the Editorial Office, 1155 16th St N.W., Washington, DC 20036.