Putting smart molecules to work

A nanotechnology to predict bioremediation potential

For the past six years Alcontrol Laboratories have been working in collaboration with The University of East Anglia and Lancaster University (within the DTI LINK Bioremediation Programme) to establish and validate laboratory methods to assess bioremediation potential of contaminated land. This work has now validated a method that, through the application of ‘smart molecules’, can be used to extract contaminants from soil in a way that reflects their microbial bioaccessibility. In this way it has been possible to successfully pre-determine biodegradation endpoints for key contaminants. The application of this nanotechnology has been successfully applied to spiked soils, reference materials and genuinely impacted soils from a municipal gas plant site. The novel extraction method that has been developed makes use of a group of compounds known as cyclodextrins. These compounds are comprised of glucose monomers that have polymerised to form a ring of glucose units (Figure 1). This molecular structure results in two key attributes: firstly, the –OH groups that project radially from the molecule confer high aqueous solubility upon the molecules, and secondly (and arguably more importantly) at the heart of their molecular structure is a hydrophobic cavity. As a consequence, it is possible for the cavity to act as a site with which hydrophobic organic contaminants (HOCs) can interact. In essence, when aqueous extraction solutions containing the cyclodextrin molecules are used to extract hydrophobic molecules associated with the soil/sediment media they migrate to the hydrophobic cavities of the cyclodextrin molecules, and can then be quantified. The application of this extraction method holds great potential for the screening of soils with a view to assessing, in a meaningful way, how the HOCs are partitioned. With this information a judgement can be made as to the potential for remediation using biological methods. Furthermore, this information, by indicating when remediation is complete, may be valuable in the management of site operations. To appreciate the significance of this novel extraction method it is important to understand firstly how hydrophobic compounds interact with solid media (e.g. soil), and secondly, to appreciate how these interactions influence bioremediation potential.

Soil-HOC interactions

Since the industrial revolution HOCs have been released into the environment from a variety of sources. Owing to their physical and chemical properties it is common for these compounds to accumulate in soils. However, it is far from the case that once these compounds arrive in the soil environment they stop equilibrating with the soil. Indeed, as time progresses the way in which these compounds are partitioned in a soil changes. Soil is, of course, a complex environmental compartment, but for simplicity let us consider it to be comprised of mineral material (sand, silt and clay), some organic mater (from the decomposition of plant residues) and finally (although very importantly) pore-space filled with either water or air. It is with these components that the HOC must equilibrate (Figure 2). The HOC’s vapour pressure dictates the fraction of compound that will be in the gas phase and its aqueous solubility the fraction that will be in the aqueous phase. The remaining fraction then has the potential to interact with the mineral and organic components of the soil. The extent to which a compound partitions onto and into these components is driven by the compound’s hydrophobicity (i.e. how much it ‘hates’ water). The more hydrophobic a compound is, the higher its partition coefficient will be (log Kow values of between 2 and 7 typical for HOCs). Sorption is widely accepted to control these interactions. However, other processes such as entrapment within soil micro-pores or organic matter have also been proposed as mechanisms for compound association with soils. Soil organic matter is of course the place we would intuitively expect organic contaminants to partition, and indeed this is the case.

HOC bioavailability and bioremediation potential

Biodegradation has been identified as a major loss process for organic contaminants in soils and, as a result, microbial strategies have been developed for the remediation of contaminated land. In the bioremediation of contaminated land, prediction of the biodegradable fraction would be important for determining bioremediation endpoints. Importantly, sequestration of HOCs through processes of sorption and diffusion into the soil’s structure renders HOCs inaccessible to the indigenous microorganisms, thus retarding biodegradation and, hence, bioremediation. Ultimately this equilibration process (discussed above) results in a reduction in the availability of the compounds with respect to both extractability and their ability to interact with biological receptors. Thus, on one hand, the equilibration process reduces the risk associated with these residues but on the other provides a bottleneck to bioremediation. Clearly, where bioremediation is being considered as a remediation option, knowledge regarding the potential of its success would be desirable. Bioremediation is dependent upon two prerequisites being met: (1) the compound must be bioaccessible; and (2) the compound must be inherently biodegradable. Of course, site conditions must also be conducive to degradation, for example, sufficient electron acceptors and nutrients. This said, the cornerstone to the potential for bioremediation is the bioaccessibility of the compound for degradation. Organic compound bioaccessibility can be assessed biologically. Quite simply, contaminant concentrations might be established before and after a laboratory scale optimised remediation experiment in which oxygen and nutrients are provided (perhaps with augmentation of catabolically competent micro-organisms). However, these methods are time consuming, require many replicates and as a consequence are encumbered with high costs. An attractive alternative is to establish chemical methods, that are quick and reproducible but that still reflect compound bioaccessibility.

Quantifying HOCs: past present and future

The measurement of chemical and biological availability has attracted considerable interest. Many of these studies have observed that as the soil contact time of a chemical increases, there are commensurate reductions in chemical availability (as defined by extractability) and biological availability (as defined by microbial degradation, uptake or toxicity) the formation of recalcitrant (more persistent residues, which are extractable using aggressive techniques) and non-extractable fractions. From these studies, the concept of chemically predicting bioavailability has evolved. Traditionally, exhaustive solvent extraction methods have been used to measure the total organic contaminant concentrations in soil. Commonplace methods include the use of Soxhlet extraction or more recently Accelerated Solvent Extraction (ASE) wherein organic solvents with pronounced solvating powers (for HOCs) are used. Such methods have little relevance to the amount of contaminant that may interact with soil biota or pose an ecological risk. For this reason, non exhaustive extraction techniques that reflect the bioaccessible fraction of contaminants have been sought over the last decade. Indeed it is now widely accepted that assessing compound partitioning is more important than solely estimating the total mass of contaminants in soil. Should you need persuading of the validity of such endeavours, ask yourself this question: ‘if I have a site that has been putatively impacted with heavy metals do I ask my analytical house to a) digest my soil samples with aqua regia (thereby dissolving virtually everything present; including the aluminosilcate minerals), or, b) digest or leach the heavy metals from my samples in a way that will give meaningful concentrations with which to pragmatically assess the risk my site represent and how it should be managed?’ Having given this some thought why might you persist in asking your analytical house to rigorously extract your samples with hot organic solvent in order that you can use this information to pragmatically manage your site?

The development of the cyclodextrin method

As already mentioned cyclodextrins are macrocyclic molecules comprising of a ring of glucose units (Figure 1). Thus, cyclodextrins are very water soluble but have a cavity that is hydrophobic. As a consequence it is possible for the cavity to act as a site with which HOCs can interact. These properties are exploited in an aqueous based extraction method to selectively extract HOCs that are accessible for microbial biodegradation at the exclusion of contaminants that are not accessible for microbial biodegradation. Figure 3 illustrates how the aqueous cyclodextrin extraction method mimics the key contaminant transfer process to soil pore water and subsequent degradation of these bioaccessible residues by soil microorganisms. This process is paralleled where cyclodextrins are used as an aqueous extraction phase wherein the same fraction of accessible compounds move through the aqueous phase and are trapped in the cyclodextrin cavity. Significantly, HOCs of low aqueous solubility are effectively removed by the cyclodextrin molecules as they move to the aqueous phase. As a consequence, aqueous solubility does not limit the extent of extraction as would be the case if a water only extraction was performed. Thus, the entire bioaccessible pool is quantified. Outlined in the following sections are data that support the discussed mechanistic reasoning and the application of an aqueous cyclodextrin extraction method for the robust and reliable prediction of biodegradation potential for a range of HOCs. This validation is presented for simple single compound spiked soils, reference materials, and complex multi-contaminant genuinely contaminated soils from a municipal gas plant.

Validation phase I: spiked soils

Four dissimilar soils with contrasting textures (silty loam, loamy sand, sandy clay and peat) were spiked with phenanthrene (a representative 3-ringed PAH). Samples were analysed at t = 1d, 20d, 5d and 100d. At each sample point total residues present were determined. In addition, residues extractable using an optimised cyclodextrin procedure and using standard aqueous leaching methods with CO 2 equilibrated water were determined. To complement these measurements of chemical extractability biodegradation experiments were undertake to establish the size of the biodegradable fraction of phenanthrene at each of the sampling points. These biodegradation experiments involved the inoculation and incubation (28d) of soil samples with a solution of micro-organisms that had been pre-established to be able to effectively degrade phenanthere. Inoculation was made at a high cell density to ensure rapid degradation of biodegradable residues. Comparison was then drawn between extents of degradation and the size of the extractable phenanthrene fractions established using either aqueous solutions of cyclodextrin and or the water only leachate (Figure 4). Results clearly show that where aqueous solutions of cylcodextrin were used to extract phenanthrene residues from the dissimilar soil the strength of biodegradation prediction was high; Frame A (Figure 4) indicates that across the data set extraction with cyclodextrin accounted for virtually all of the phenanthere degraded (y = 1.1x-6.1; the gradient, approximately equal to one, indicated consistency between predicted and observed biodegradation). Furthermore, the strength of this relationship was statistically high (r 2 = 0.78). In contrast, where a water only leachate was obtained to evaluate the fraction of phenanthere that was potentially bioaccessible for biodegradation, the strength of prediction was low; Frame B (Figure 4) indicates that across the data set extraction with water only accounted for only a fraction of the amount of phenanthere degraded (y=0.11x + 83.8). Furthermore the strength of this relationship was statistically low (r 2 = 0.17). Thus, for these dissimilar spiked soils that had equilibrated for increasing amounts of time, the data supports the application of aqueous cyclodextrin solutions to directly predict biodegradation endpoints.

Validation phase II: reference materials

Clearly, the data already presented represents the simplest of situations, namely, a single contaminant that had been augmented into pristine soils. To advance the validation work the next step was to assess the applicability of the cyclodextrin method across a range of compounds present in homogenised reference materials. To this end soils representing a range of contaminant sources, namely, diesel, lubricating oil, spent oxide were ground, sieved and blended to produce homogenised reference materials (RM) with different analyte loadings. Polycyclic Aromatic Hydrocarbon (PAH) concentrations in the RMs ranged from 5.5 + 0.5 to 44.4 + 4.5 mg kg -1 . The RMs were then extracted using a rigorous solvent extraction method (to establish total loading), and also extracted using the aqueous cyclodextrin method (to establish microbially bioaccessible PAH fractions). Biodegradation assays were again undertaken, wherein RMs were inoculated and incubated (28d) with a solution of microorganisms that had been pre established to be able to effectively degrade PAHs. Comparison was then drawn between extents of degradation and the size of the extractable phenanthere fractions established using aqueous solutions of cyclodextrin (Figure 5). The results showed that biodegradation within the four RM substrates were not limited by co-contaminant loadings, but rather by their bioaccessibility. Cyclodextrin extraction correlated well with biodegradation regardless of RM complexity. Across the entire dataset the cyclodextin method was observed to be a robust direct predictor of analyte biodegradation endpoints. Direct prediction was evident in the slope of 0.9 while reproducibility and robustness indicated by the r 2 value of 0.87. Looking specifically at the data by PAH ring size, this general relationship was borne out in more detail: for 2/3 ring PAHs a slope of 0.98 and r 2 value of 0.97 were observed; for 4 ring PAHs a slope of 0.98 and r 2 of 0.77 were observed, and; for 5/6 ring PAHs a slope of 0.80 and r 2 0.93 were observed. Collectively this wave of validation indicated the cyclodextrin method to directly predict biodegradation endpoints in these multi-contaminant reference materials that represented a range of contaminant sources (diesel, lubricating oil and spent oxide).

Validation phase III: real soils

This final phase of validation built upon the results for the simple single contaminant spiked soil and the multi-contaminant (but homogenised) reference materials already discussed. Two PAH contaminated soils from a municipal gas plant (MGP) (ΣPAH = 877 + 52 and 2620 + 344 mg kg -1 ) were used in this phase of validation. In contrast to the RMs the PAH loading in the MGP soils were notable higher. Thus, not only was the cyclodextrin method validated for more heterogeneous genuinely contaminated materials but also at exceptionally high PAH loadings. The same experimental methodology was applied to the MGP soils as for the RMs wherein total PAH loading were established using a harsh solvent extraction method, bioaccessible PAH fractions determined using the cyclodextrin aqueous extraction and biodegradation endpoints established following a 28d of incubation of the soil in the presence of active PAH degrading microorganisms. The results again showed a good strength of prediction between cyclodextrin established biodegradation endpoints and biodegradation endpoints observed (slope of 1.24 and r 2 of 0.94)(Figure 6). Collectively, this wave of validation indicated the cyclodextrin method to directly predict biodegradation endpoints in these multi-contaminant genuinely contaminated soils with very high levels of PAHs. These results in combination with the preceding waves of validation provide extensive support for the use of the cyclodextrin method to reliably predict the bioremediation endpoints for a wide variety of HOCs (with very different properties) present in a wide variety of matrices. Much of this work presented here has been peer reviewed and published in mainstream journal literature (see further information section for a selection of recent relevant publications).

Fulfilling your needs

The application of this extraction method holds great potential for the screening of soils with a view to assessing, in a meaningful way, how the HOCs are partitioned. With this information a judgement can be made as to the potential for remediation using biological methods. At the other end of operations, the cyclodextrin method might also be a beneficial (and cost saving) tool as it will provide information regarding the bioaccessibility of residual compound present as the bioremediation programme advances. In essence, the cyclodextrin method will indicate when no further remediation is probable; as all accessible contaminants have been exhausted. This information could be used to save the expense associated with protracting site operations for no benefit in remediation endpoint. Alcontrol Laboratories have recognised the market need for a rapid, economical screening test to indicate the amenability of soils to successful bioremediation. Alcontrol are now offering this test routinely. This protocol can be performed in five days, and will therefore greatly reduce the timeframes currently involved in assessment for bioremediation potential. Please contact our Sales & Marketing department for further information: mkt@alcontrol.co.uk.

Selected publications

Brian J. Reid, Joanna D. Stokes, Kevin C. Jones, Kirk T. Semple. (2000) “Nonexhasutive Cyclodextrin-Based Extraction Technique for the Evaluation of PAH Bioavailiability”. Environmental Science and Technology, 34, 3174-3179. Kieron J. Doick, Nadia M. Dew, and Kirk T. Semple. (2005) “Linking Catabolism to Cyclodextrin Extractabilty: Determination of the Microbial availability of PAHs in soil”. Environmental Science and Technology, 39, 8958-8864. Ian J. Allan, Kirk T. Semple, Rina Hare, Brian J. Reid. (2006) “Prediction of mono- and polycyclic aromatic hydrocarbon degradation in spiked soils using cyclodextrin extraction”. Environmental Pollution, 144, 562-571. Published: 10th Jun 2007 in AWE International