Anthracene and pyrene were selected as representative Polycyclic Aromatic Hydrocarbons (PAHs). Anthracene was selected because of its distinct spectral structure. The PAHs used in this study were purchased from Sigma Chemical Co. Quartz (SiO₂) and aluminum oxide (Al₂O₃) were selected as main soil constituents to represent the different soil surface area types and mineralogies, i.e., soil media type. The quartz was a coarse soil type with a yellowish color and a low surface area compared to the aluminum oxide, which represented the fine soil type, with a white color and a high surface area of about 900 m². To prepare the PAH contaminated soil media, fifty grams of soil mineral was mixed with 10 mL of spiking solution where the PAH was dissolved in dichloromethane. To achieve a homogeneous condition, spiking procedures were thoroughly carried out in several steps. The contaminated soil mineral was placed in the fume hood for three days to evaporate the dichloromethane. At 150 mg PAH/kg dry soil, fluorescence was observed.

Cyclohexane and dimethyl sulfoxide (DMSO) were selected to investigate the interaction arising from the differences in polarity between the contaminants and their surrounding environments. These solvents have low and high dielectric constants, and were chosen to represent the polar and nonpolar environments, respectively. The effect of water was also checked because of its presence in soil pores. Soil media with 0, 2.5, 5, 10, 15, and 20% solvent content were also prepared to examine the effects of the solvent content on the LIF spectral signature. HPLC-grade reagents were obtained from Fisher Scientific Co.

A biosurfactant was used to monitor its effect on removing the organic contaminants adsorbed on soil media using the LIF. Rhamnolipid, the biosurfactant used in this study, is comprised of extra-cellular natural substances produced by Pseudomonas aeruginosa and has low CMC and nontoxic effects to soil microorganisms. Biosurfactant was introduced to enhance the solubility of hydrophobic organic compounds like PAHs. Rhamnolipid solution was purchased from Jeneil Biosurfactant.

LIF system used in this study is dealt with. The principal components are the pulsed UV laser as the excitation source, a fiber-optic probe, a detection unit for time and spectrally resolved detection of the fluorescence light, and the control and data acquisition unit. For the excitation of fluorescence, a diode laser pumped neodymium-doped yttrium aluminum garnet (Nd:YAG) laser with a 7 ns pulse duration and the fourth (266 nm) harmonics of the laser was used for excitation of the fluorescence. The detection unit is basically an optical mulit-channel analyzer with a time resolution of 5 ns, which consist of a spectrograph, a gateable image intensifier and a CCD camera (Figure 1).

The light of a pulsed UV-laser is guided through an optical fiber into the observed soil and excites existing pollutants to fluorescence. The fluorescence light is guided back though additional fibers to the detection unit.

The fiber-optic probe was dipped into the soil sample, rinsed with ethanol and distilled water, and cleaned with a fine laboratory wiping tissue after each measurement. Typically, the fiber-optic probe carrying UV-laser illuminated a surface area of 20 mm². The fluorescence signal was collected from approximately that spot size. The original spectra from LIF system are time resolved spectra in the total wavelength between 0 and 50 ns and in the wavelength range between 258 and 607 nm. For the entire soil sample, at least five measurements were performed and averaged to avoid any inconsistency due to the heterogeneity of the soil matrix.

The LIF provides the full Wavelength-Time-Intensity (WTI), which is the time resolved spectra between 0 and 50 ns over the wavelength range 258 to 607 nm. The wavelength coverage of the fluorescent light emitted from the PAHs, and the specific shape of fluorescence spectra were dependent on the PAH itself.

In this study, the total fluorescence intensity and spectral signature were generally obtained by integrating the spectra over the total time interval. Here, the additional time information was only used for the interpretation of surfactant effect on PAH contaminated soil where surfactant and PAH showed their overlapping fluorescence spectra. Time resolved detection of the fluorescent light yields fluorescence decay times. This information makes it possible to simultaneously discriminate between PAHs and surfactant due to their characteristic decay time. The fluorescence intensity, and its relation to the specific wavelength coverage at each respective PAH concentration, were recorded. The total fluorescence intensity was obtained by integrating these spectra for the wavelength coverage of each PAH. Additionally, fluorescence spectral signatures were monitored to assess the different types of PAH binding on the surface of the soil media.

The different soil media types were tested to investigate the matrix properties on the fluorescence spectra. At a given concentration of 150 mg PAH/kg dry soil, fluorescence was observed. According to the results, the anthracene on different soil minerals showed different spectral structures, implying different types of binding or interactions between the contaminants and soil mineral (Figures 2 and 3). Likewise, the total fluorescence intensity extracted from the fluorescence curve for the wavelength coverage was affected as well.

For anthracene on quartz, the first peak (bending mode) of fluorescence was deteriorated (Figure 3a); this can be explained as the suppression of bending of molecules due to the strong binding of PAH on quartz compared to the aluminum oxide. On the other hand, aluminium oxide had spectra of somewhat released species on the surface, showing more of an activated bending mode (Figure 3b). These findings implied that different interactions between contaminant moleculea and the environment exist.

In order to support this, the fluorescence of extracted anthracene from the contaminated quartz was also observed (Figure 4). Here, quartz of various sizes was used, including quartz Certified Reference Material (CRM), quartz 0.4–0.8 mm and quartz powder. The results showed that the PAH was not destroyed during the measurement when it is compared to PAH on aluminum oxide and in solvent showing distinct spectra structure.

It also cannot be explained by self absorption due to the high concentration or inner filter effect because of its spectral signature shown in 250–370 nm. This observation implies that the fluorescence spectra signature according to the soil mineral type should be carefully considered when it is used as the contaminant indicator with the expression of relative quantities of PAH present in the soil sites of interest. Additionally, this fluorescence spectra signature of anthracene can be a representative probe, which can detect the variation of the state of contaminants in soil. For example, it is possible to deduce the adsorbed or dissolved state of contaminants in the soil environment by comparing the fluorescence spectra of the contaminant in the soil and the solvent.

The polarity of solvent and the local environment have profound effects on the emission spectra of polar fluorophores such as PAHs. Generally, solvent effect was evaluated to determine the polarity of the probe-binding site on the macromolecule. This was accomplished by comparing the emission spectra and /or quantum yields of the fluorophore between when it is bound to the macromolecule and when it is dissolved in the solvents of different polarity. This serves as an explanation in that the different soil media types and surrounding local environment of contaminant show different spectral signatures. It means that the estimation of PAH in the soil media could be performed by comparing the fluorescence spectra of PAH on quartz and in solvent. In order to investigate the possible interaction between PAH and soil media, the fluorescence spectral signatures with the addition of solvent on soil media were observed.

Solvent and water addition showed different responses in the fluorescence spectra, indicating their effects on the interaction between the contaminant and the soil media. The added solvent increased fluorescence intensity in which the contaminant began to dissolve in the additional solvent (Figures 5 and 6). Here, DMSO (C₂H₆O₅) and cyclohexane (C₆H₁₂) were used to describe the different environment around PAH. These solvents with different dielectric constant could affect the solubility of PAHs on soil media, resulting in the change in their fluorescence intensity and spectral signatures.

This figure showed increased PAH fluorescence intensity after addition of both solvents. This implies that as the surface loading of PAH on soil media is increased, the PAH fraction accessible at the window of sensor increased. The fluorescence of aluminum oxide was a sharper structure similar to that in pure solvent. Based on the convoluted structure of aluminium oxide, we can deduce the state or phase of contaminants in soil media; adsorbed species were on the surface or soaked species were in the inner pores. There was no change in the fluorescence spectral signature following solvent addition, as shown in Figure 5. Based on these observations, the contaminants were expected to exist as dissolved species in the solvent or as free molecules and weakly binding species on the surface if adsorbed. In the case of the soil media type such as aluminium oxide, it was deduced to have caused the soaking species to move into the outer space rather than changing the binding on the surface.

As shown in Figure 6, the anthracene on quartz under dry conditions showed typical spectra of adsorbed species on surfaces, implying strong binding on the surface, thus suppressing the bending mode of the molecules of the contaminants. However, as the solvent was added on the soil media, it showed the developed spectral signature. It is possible that as the solvent was added, adsorbed contaminants on quartz surface started to dissolve into the solvent. In other words, the phase of the contaminants in the soil media changed from an adsorbed phase on the surface to a dissolved phase in the solvent. This resulted in the free bending mode of contaminant molecules.

Water had little effect on the state of the contaminants, due to their low solubility in that liquid. Considering the natural water content in soil media, the water effect on fluorescence intensity as well as the spectral signature of the PAHs was assessed.

Quartz is considered as the first layer surface with low available surface area, and with the continuous addition of water, PAH was displaced from the surface, resulting in a decreased intensity. However, the effects of water result in no change to the spectral signature with increasing water content. The state of the contaminant can be deduced as the adsorbed species on surface in spite of water addition. However, for PAH on aluminium oxide, high water content induced a geometric change of the aluminium oxide, rather than a change in PAH solubility. This yielded a different spectral structure (Figure 7), meaning that the available surface area for contaminant, surrounding environment of contaminant, was changed and that interactions between the contaminant and the soil media was affected.

From these results, we can deduce the role of fluorescence spectra, implying the state of contaminant and its physicochemical interactions with the soil media.

The surfactant effect was investigated to monitor the presence of the contaminants, and to predict the possible state of the contaminant in the soil media during the remediation process monitoring. For organic compounds such as PAHs, the remediation process using surfactant is also commonly performed. Therefore, an accurate appreciation of the surfactant effect on the fluorescence of PAHs was required. Based on the appreciation of the phases or states of contaminants in soil media, the LIF test was performed. Time resolution was applied to distinguish the spectra of the PAH and the surfactant. The fluorescence of PAH in solvent in which PAH existed as a free molecule or dissolved states (Figure 8) and that of PAH from contaminated soil media was compared with the fluorescence of PAH with surfactant addition (Figure 9).

In terms of the peak wavelength and spectral signature of pyrene, the LIF results were interpreted. As the surfactant was added on the soil media, the developing spectral signature of pyrene on the soil media and the subsequent increase in its fluorescence intensity were observed. This supports the fact that as the surfactant is added, adsorbed contaminants on soil media surface starts to dissolve in the surfactant, resulting in the free bending mode of the contaminant molecules. These observations make it possible to monitor the remediation process using surfactants for removing organic contaminant.