An investigation into Forensic Soil Analysis

An investigation into Forensic Soil Analysis

Abstract

In this review soil examination, preparation and comparison is being discussed. Procedures used by many authors on soil examination are the topic being reviewed here. The review begins with Transport of soil, preparation of soil, Attenuated Total Reflectance – Fourier-transform infrared spectroscopy (ATR-FTIR), Multivariate analysis, Microscopy examination, X-ray Absorption Spectroscopy, pH analysis and Laser Induced Breakdown Spectroscopy (AIBS). Soils may establish evidence that connects a person or object to a particular location.

Introduction

Soils are a very complex material; it can consist of both organic and inorganic components. The soils stem can transfer components onto an object or persons. Due to its complex material through examination of its components it can provide information about its geological origin, dominant vegetation, management and environment. Soil consists of high volume of information that has a great impact on Forensic analysis. Individual analytical techniques have different scales of resolution and relevance depending on the nature of the criminal case and context, each method has its strengths and weaknesses. The soil that transfers from the stem onto an object or person can be recovered and used as evidence in a forensic investigation. Soil identification of source, linking to a crime is dependent of the characteristics of the organic and inorganic samples which can be compared with other samples. For example, the referenced soil sample can be compared with the suspect’s soil sample to aid in placing someone at the scene of the crime. An overview of soil characterisation methods including chemical analysis, mineralogy, infrared spectroscopy and microscopic analysis are presented here. Using soil as evidence has a century of history, dating all the way back to the writings of Sir Arthur Conan Doyle between 1887 and 1893 through his ‘Sherlock Holmes’ book series. Although soil evidence is a historical one, there are a lot of new developments that are barely known by analysts. It has a lot to do with equipment given in the laboratory.

1. Transport of soil to the Laboratory

When a soil sample is collected for several days before arrival to the laboratory, field-moist soil undergoes significant biological changes at room or elevated temperatures when placed into air-tight containers. Organic matter can release various element ions such as Cadmium Phosphate PO₄^3-, Sulphate SO₄^2-, Borate BO₃^3- and Ammonium NH₄⁺ into the soil. In anaerobic conditions organic matter decomposition can result in loss of Nitrogen (N) from the soil. When transporting soil for a long period of time the sample should be kept in a cool environment – 5 to 10⁰C and the excess water should be removed by drying partially, this keeps the soil moist. A soil sample can be frozen but similar to the effect high temperature has on the soil; it alters the physiochemical properties within the soil. (Benton Jones, 2001)

2. Preparation of the Laboratory Samples.

There are two main ways to prepare a sample before analysing it, but some of the methods have a different way of preparing and this is stated in the methods section. First sample preparation is drying; this is to air-dry the soil samples at laboratory temperature 21 to 27⁰C before crushing and sieving. The drying procedure should be done in a timely manner to avoid mineralization. The texture, organic matter and moisture will determine how long it will take for the soil to get to the air-dried condition required. Soils that are high in clay and/or organic matter constant need a long air-drying time compared to that of the sandy-textured soil. When drying the samples the temperature shouldn’t exceed 38⁰C, because this can cause changes in the physiochemical properties within the soil. Drying of the sample can affect the determination of micronutrients Copper (Cu), Iron (Fe), Manganese (Mn), and Zinc (Zn), since drying caused significant changes in the soil there was a time in the laboratory which they would analysis the soil taken from the field as received using the slurry method. This slurry proved to be a lot more time-consuming for processing larger number of samples. The moisture of an air-dried sample is determined from the physiochemical properties of the soil and the humidity of air surrounding the sample. This variability has little effect on the soil analysis; the only effect is that the soil is measured in volume instead of weight. (Benton Jones, 2001)

Second preparation procedure is Crushing/Grinding/Sieving, following drying the soil is crushed either by hand or by using a mechanical device and is then passed through a 10-mesh (2-mm). the grinding process can have an effect on the extracting of Copper (Cu), Iron (Fe), Manganese (Mn), and Zinc (Zn), Potassium (K) and Phosphorus (P). Removal of stones and any other substances are done when the soil is sieved through the 10-mesh, this leaves a sample that can be easily handled and stored in the laboratory. When preparing a sample for testing in the laboratory, the sample can be easily contaminated by coming in contact with surfaces or with dust residue from a previous sample. The crushing and sieving devices used the preparation should be free of elements that can be determined in the analysis. “For example, brass sieves should not be used if Cu and Zn are elements to be determined” (Benton Jones, 2001). For crushing and sieving sample reduction may be required and this takes precise care as the sample needs to be thoroughly mixed before it can be divided. Reduction of particle size can have an effect on determination of a number of elements such as Cu, Fe and Zn. (Kahn, 1979) and also for equilibrium extraction reagent procedures (Houba, 1993). Both authors agree that the reduction of particle size effects certain element determination but understands that it must be done to make the sample easier to handle.

Once the soil has been dried, crushed and screened it can be then stored in a dry environment indefinitely without any changes occurring within the soil test values.

3. ATR – FTIR Method

For the FTIR method, three samples were taken from the Canberra area and all of the samples were analysed individually. The soil was oven dried for 24-48 hours @47⁰C, a mortar and pestle were then used to crush the soil lightly. The dried soil was then sieved to get it to its finest particles. Each soil specimens were analysed “using a Smiths Detection Identify IR portable FTIR spectrometer which uses a diamond ATR sampling interface”. Each soil specimen was analysed in triplicate. The diamond cell detection window has a 1.3mm diameter; the soil specimen must cover this window. The sieved soil specimen was pressed down on the ATR crystal. The range of the spectra was 4000-650cm^-1, the soil was scanned 64 times to get a clear reading. (Woods et al. 2014)

The results from the spectra were placed into Minitab16 (version 2.2) this is a statistical software, this constructing a plot of the first two components in the soil specimen. (Woods et al. 2014)

The three soil samples appeared to be the same as the soil in the Canberra area were rudosols and chromosols from volcanic mountains, alluvial fans, granitic material and metasediments. (Woods et al. 2014)

Variations present at peak 3700/3620cm-1 (Fig.1) and at 796/779/694cm-1 (Fig.2). The variations at 3700/3620cm-1 present can be a results of Kaolin Minerals and the variations at 796/779/694cm-1 present could be a results of quartz. (Woods et al, 2014)

4. Multivariate analysis

Multivariate analysis can only be performed when the Qualitative examination has been completed. In the Qualitative analysis, the comparison of the soil spectra has to be done to calculate the discrimination power. This formula was first described by Smalldon and Moffat;

Formula 1: The discrimination power formula. (Smalldon and Moffat 1973)

In the field of forensic science, the statistical data is generally interpreted on either raw or log-transformed data. To individualize and separate the elements from multi-elements evidence/exhibit the Pattern recognition statistical method is used. By using the standardization method the data has to be first normalized to either mean 0 or 100 due to the complex nature of spectroscopic and chromatographic technique. The pre-treatment given to the raw data is an essential part of the effectiveness of the analysis. All statistical analysis in this article is carried out by using SPSS 20 (IBM).

5. Microscopy Examination and Colour Analysis

A microscope can be useful in the soil analysis in determining particle sizes. The microscope analysis is often performed after the colour of the soil has been determined also after any recordings of fibres, metals, paints, glass and plastic have been made. To make the identification of minerals much easier the soil sample is soaked in water this removes the organic debris within the soil. In some occasions this can also identify grains of starch, ceramics and some abrasives. In a lot of Forensic cases the Electron Microscopes have been used to provide a higher resolution than that of a lower-power optical microscope. (Tibbett and O Carter, 2008). In the Wells and Chapman criminal murder case in 2002, soil analysis was used to place the accused Huntley at the scene of the crime. The microscope analysis was used to examine the sample from the crime scene vs. the suspects sample found in the car, which could distinguish that both samples matched up. Every soil has a unique colour, all soil samples are analysed under the microscope to determine the size and shape of each particle but it is also used to determine the colour of the soil. The soil colour is compared to the Munsell colour chart to determine exactly what colour the soil is. The Munsell colour system is a colour space specifying colours based on three colour properties: hue, value (lightness) and chroma (colour purity). Analysing the colour of soil can have a huge impact on a criminal case because soils from different areas have different colours and it is rare for soil from two different areas to have the same colour dry or wet. For example, in an isolated area, a woman who had run on a road in Central Park (CP) was found near death. She was hit with a large rock in the head and dragged several hundred feet into a ravine where she was brutalized. It is thought that she managed to travel a short distance after the attack and collapsed from her injuries. She was bleeding until the morning when some passers- by discovered her. In a short time, many suspects were arrested and questioned. After questioning some were arrested. Their clothes were taken as proof and packed by the crime scene unit. . The clothing was taken to the laboratory for analysis.

Table 1: Results of the soil comparisons conducted in this case (Petraco & Kubic, 2008)

Table 2: Associative data compiled on the soil data sheet (Petraco & Kubic, 2008)

From table 1, the questioned soil aggregates S3, S4 and S5 were removed from the clothing of the suspect. Two aggregate soil specimens S2 and S2A were removed from the clothing of the victim, which the investigating officer sent to the laboratory. On the crime scene, several known soil specimens S1 and S7 have been collected. The information displayed in table 1 and 2 shows that there are many similarities between the known and the questioned specimens. Although some colour and vegetation differences between the specimens are shown in Table 2, it is important to remember that this was a very dynamic event. The victim was dragged several hundred feet; many times she was thrown to the ground; after the attack she walked or crawled a distance; the suspects travelled and hid in the park. Such differences in colour and the quantity of plant debris between the specimens in question and the specimens in question and between the specimens in question could therefore easily be accounted for. Table 2 shows an interesting side note concerning the importance of the colour of a soil in forensic soil comparisons. Although the specimen bogs area (S7B) and the specimen high point area (S7A) were only 15 inches. Apart from that, both had the same composition of minerals, while their tangible colours and the calculated Munsell1 colours were very different. Examiners who use soil colour exclusively as the primary indicator of whether a questioned soil specimen originated from a given source might wish to reconsider this reasoning. The numerous consistencies documented on the soil data sheet and the gradient have enabled one of the authors to link the specimens S3, S4 and S5 to the known sources represented by S1, S2, S2A, S7A and S7B. The author testified in two separate disputes about the origin of the questioned specimens found on the three suspects. One author testified that, in his opinion, the three suspects got the soil from Central Park on their clothes. (Petraco & Kubic, 2008)

6. X-ray Absorption Spectroscopy

X-ray Absorption Spectroscopy (XAS) has been used on a number of soil science, mineralogy and geochemistry. A high – quality spectra of XAS can be obtained from a small sample of heterogeneous mixtures of gases, liquids and/or solids, which makes this technique ideal for soil analysis. The XAS spectrum identifies the chemical speciation of an element, including mineral, non-crystalline solid or adsorbed phases. To obtain the nature of the complex sample the application of several different experimental techniques with each individual measurement providing both unique and complementary information is required. X-ray diffraction (XRD) is the most commonly used technique for identifying the minerals in soil. The XRD relies on long range ordering of atomic planes to probe crystalline structure at a length scale of approximately 50 Å or more. XAS and XRD can be used to determine distances between atoms. The XAS technique can analysis dry and moist soil, the x-ray beam probes a millimetre of the soil sample. To get the sample to a millimetre, the sample can be sieved and diluted to homogenize samples.

5. pH Analysis

Soil pH is the measure of Hydrogen ion (H⁺) activity in the soil solution. pH is expressed as the following equation: pH = -log10(H+)

A soil is either acidic or basic, they are acidic because they have ionized H⁺ ions and they are basic because they have ionized OH^– ions. pH is therefore measured on soil acidity or basicity on a scale from 0-14, pH of 7.0 is the neutral point meaning it is neither acidic nor basic. The pH of soil can be determined with the use of chemical dyes (Jackson, 1958; Woodruff, 1961; Hesse, 1971); all these three authors agree to analysis the pH with the use of the dyes. However, the normal procedure is to use a glass electrode and a calomel reference cell. These electrodes can be separate or combined into one electrode. History has identified that two separate electrodes is best as the combined electrode is more prone to problems (Peck, 1983). The characteristics and the techniques used to measure the pH of the soil is a huge contribution to finding the correct measurement. (Benton Jones, 2001)

The determination of the soil of pH is a three-step procedure:

(1) Prepare the soil-water:

When using a pH meter, the pH is measured in a soil/water slurry ratio; the ratio is normally 1:1 or 1:2. The sample can be measured in a soil slurry of either 0.01M CaCl₂.2H₂O or 1N KCl. When the pH is measured in the salt solutions the pH value is different. The pH determined in 0.01M CaCl₂ is from 0.3-0.5 units less than that determined in water and 0.7-1.0 units less when determined in 1N KCl. Therefore the determination of the pH must be identified based on the solution.

Standardized water pH Determination –

• Weigh 10g air dried <10-mesh-sieved (2mm) soil into a beaker.
• Pipette 10ml of water into the beaker and mix for 5 seconds with a glass rod.
• Let the soil-water sit for 30 minutes.
• Preform pH measurements at 20 to 25⁰C.
• Before analysing stir the soil-water with a glass rod.
• Place the electrode into the soil-water and place on mechanical stirrer and stir gently.
• Read the pH after 30-60 seconds

(2) Calibration of the pH meter:

To calibrate the pH meter it requires the use of two buffer standards of known pH. For acidic soils, a buffer of pH 7.0 and another one of pH 4.0 or 5.0 will be used and for basic soils, a buffer of pH 10.0 and pH 6.0 will be used. The meter should be adjusted to read the high pH buffer first followed by the lower pH buffer. Rinse the lower surfaces with water before placing the electrode into the buffer solutions. Swirl the solution once the electrode has enter the buffer solution to ensure the electrode has complete contact. The electrode surfaces are cleaned with water between solutions. If the pH meter does not calibrate after using the buffers, then the glass electrode may need replacing. If the meter is slow to reach its indicated reading, this may be a result in the calomel reference being clogged. In this case, gently polish the tip of the electrode with emery paper.

(3) Place the electrode into the prepared slurry and read the pH

Table 3: pH values of the Acidic and Basic categories (Benton Jones, 2001)

Image 1: pH values of chemical elements (Benton Jones, 2001)

7. Laser Induced Breakdown Spectroscopy – Detection of lead in soil:

Laser induced breakdown spectroscopy is a technique for basic analysis of different states of matter. This technique can be used for detecting and quantifying the components in different types of samples such as environmental specimens, liquid samples, aerosols, metallurgical samples, metallic and non-metallic samples, edible samples, gas samples, soil samples and biological samples. The detection of lead has therefore received the greatest attention from the various heavy metals. In particular, Pb (Lead) – contaminated soil is very lethal for both humans and animals, as animals exposed to Pb have been observed to develop severe health conditions, such as central nervous system disorders, blindness, ataxia and convulsions. According to the United States Environmental Protection Agency (USEPA) (Farel et al, 2005), the safe level of Pb in soil is 400 mg L−1 [13] and in Europe, the corresponding value is 32–100 mg L−1 (Dragović et al, 2008). For quantification of Pb in soils, several conventional elemental analysis techniques, such as inductively coupled plasma- mass spectroscopy( ICP- MS), atomic absorption spectrometry( AAS) and Xray fluorescence (XRF) spectroscopy, have been used. However, these conventional techniques require elaborate sample preparation procedures, which often distort a sample’s true chemical composition. On the other hand, the sample can be analysed in its original form without changing its chemical composition in the LIBS technique. In LIBS, however, the precise estimation of metals in soils is obstructed by the complex matrix of the soil. In order to draw reliable calibration curves for quantitative studies, many standard reference samples must therefore be analysed. Workers at oil field sites and nearby are exposed to various health risks associated with polluting sources, in particular lead- containing dirt, soil or plant intake and drinking water contaminated by lead- polluted soils. For the above reasons, the author was motivated to analyse the fuelled soil for drilling in order to estimate the abundance of Pb in the soil at work. In the work carried out by this author, an optimized LIBS system was used to detect and evaluate Pb contamination in soils in oil field drilling areas in Pakistan. The quantitative measurement of Pb content in soil samples was carried out using two LIBS- based mechanisms, i.e. the standard calibration curve method and a procedure based on the integrated intensity of the strongest peak of an interesting species in a given sample.  The assumption of local thermodynamic equilibrium (LTE) and the formation of optically thin plasma (OTP) were verified using key characteristic plasma parameters such as electron temperature (T) and number density (Ne) to ensure the optimal experimental condition for the accurate evaluation of Pb. In addition, Pb quantification using LIBS and ICP/OES has been shown to be highly consistent. (Rehan et al, 2018)

Five samples of soil were collected from five different locations at a distance of 1 km from the east and west sides of the drilling site. Four samples were taken from each location at a depth of 0–11 cm with respect to the ground surface using a hand spade. Each target sample consisted of homogeneous compositions of 4 samples collected in a 1 km circle, dried in a tray under ambient conditions and special codes were assigned, as explained in table 1. The standard preparation procedure 3050B has been adopted for the analysis of Pb with ICP / OES, as recommended by the EPA for the ICP study. For this purpose, the samples were first digested in 5 percent nitric acid with a purity of approximately 99.99 percent (Fisher Scientific, USA) in a graphite block heater at approximately 95 ° C, followed by the addition of hydrogen peroxide (H2O2), which completed the exothermic reaction and was then refluxed at 95 ° C with strong HCl. The resulting solution was then tested for Pb by ICP spectrometer (Optima 2100-DV; Dual View, Perkin Elmer) using three levels of accuracy reference standards. The parameters of ICP-spectrometer used during this study are listed in Table 2. The soil samples were first mixed with the matrix of Potassium Bromide (KBr) for LIBS measurement. 2 grams of a homogeneous mixture of KBr and soil were palletized with a diameter of 2 cm and a thickness of 1 cm by means of a hydraulic press with an applied pressure of~ 8.0 FF106 Pa for 30 minutes. The homogeneity of our samples was verified by various LIBS measurements at various locations on the pallet surface. The LIBS spectrometer was calibrated for Pb before actual measurement of the soil samples. (Rehan et al, 2018)

Image 2: The result from the LIBS scans. (Rehan et al, 2018)

From image 2 , the typical spectrum of DFS emissions from LIBS covering the spectral range of 200–620 nm revealed the presence of neutral as well as single ionized spectral emission lines of elements such as Al, Fe, Ca, Mg, Mn, and S in addition to Pb atomic. A single laser shot was used to record the LIBS spectrum, followed by an average of 20 laser shots under the same experimental conditions. The correlation between the observed spectral line intensity of Pb and its concentration was first studied to quantify the Pb content. The DFS sample was used to obtain samples of soil with different concentrations of lead. Various Pb emission lines have been tested (239.37, 280.19, 405.78 and 438.64 nm) from soil samples of known Pb concentrations. The intensity of the Pb line was observed at 405.78 nm to increase linearly against the Pb concentration with a R2 value of 0.99. The PB emission line at 405,78 nm was therefore used during this study for further analysis. The optimal sensitivity of the LIBS system to detect Pb was determined by investigating the dependence of LIBS signal intensity on different experimental parameters such as laser energy, laser pulse width, time interval between the laser pulse and plasma emission measurement. The sensitivity and laser- induced disintegration emissions can also be affected by the physic – chemical properties of the target material, the ambient conditions and the pressure used to make the pallets. (Rehan et al, 2018)

Conclusion

This paper shares a methodology and rationale information for forensic soil tests and comparisons that have been successfully used in forensic casework for many years. The methodology in this review are quick and easy to learn and use, allows for quick screening and comparison of complex soil specimens, limits the use and requirement of more time- consuming procedures such as density gradients and gives the forensic science community a fast close to a case.

References

• Smalldon, K.W. & Moffat, A.C., 1973. ‘The calculation of discriminating power for a series of correlated attributes’. Journal of the Forensic Science Society, 13 (4), 1973/10/01/, pp. 291-295.
• Woods, B., Lennard, C., Kirkbride, K.P. & Robertson, J., 2014. ‘Soil examination for a forensic trace evidence laboratory—part 1: Spectroscopic techniques’. Forensic Science International, 245, 12/1/December 2014, pp. 187-194.
• Tibbett and O. Carter, 2008. Soil Analysis in Forensic Taphonomy [online]. Place: CRC Press, Taylor& Francis Group. Available from: https://books.google.ie/books?id=aksRkfr1d6kC&pg=PT324&lpg=PT324&dq=R.+C.+Murray,+Geotimes+2000,+45,+14.&source=bl&ots=xbQ4yMRUG8&sig=1gUYR8qWbWua-76qNAqn-Uvkydk&hl=en&sa=X&ved=2ahUKEwjUvI-TmrHeAhWOuIsKHSFaCnsQ6AEwAXoECAgQAQ#v=onepage&q&f=false [Accessed on 31 October 2018].
• Houba, V.J.G., W.J Chardon, and K.Roelse. 1993. Influence of grinding of soil on apparent chemical composition. Commun. Soil Sci. Plant Anal., 24:1591-1602
• Kahn, A. 1979. Distribution of DTPA-extractable Fe, Zn and Cu in soil particle-size fractions. Commun. Soil Sci. Plant Anal., 10:1211-1218
• Drees, L. R. and Ulery, A. L. (2008) Methods of soil analysis. 2nd ed (Soil Science Society of America book series: no. 5). Available at: http://search.ebscohost.com/login.aspx?direct=true&db=cat05654a&AN=gmit.260928&site=eds-live (Accessed: 12 November 2018).
• M.R. Farfel, A.O. Orlova, R.L. Chaney, P.S.J. Lees, C. Rhode, P.J. Ashley, Biosolids compost amendment for reducing soil lead hazards: a pilot study of Orgro® amendment and grass seeding in urban yards, Sci. Total Environ. 340 (2005) 81–95.
• S. Dragović, N. Mihailović, B. Gajić, Heavy metals in soils: distribution, relationship with soil characteristics and radionuclides and multivariate assessment of contamination sources, Chemospere 72 (2008) 491–495.
• Jackson, M.L 1958. Soil Chemical Analysis, Prentice-Hall, Englewood Cliffs, NJ.
• Woodruff, C.M. 1961. Brom cresol purple indicator of soil pH. Soil Sci., 91:272
• Hesse, P.R. 1971. Soil Chemical Analysis, Chemical Publishing Co., New York.
• Petraco, N., Kubic, T. A. and Petraco, N. D. K. (2008) ‘Case studies in forensic soil examinations’, Forensic Science International, 178(2–3), pp. e23–e27. doi: 10.1016/j.forsciint.2008.03.008.
• Rehan, I., Gondal, M. A. and Rehan, K. (2018) ‘Determination of lead content in drilling fueled soil using laser induced spectral analysis and its cross validation using ICP/OES method’, Talanta, 182, pp. 443–449. doi: 10.1016/j.talanta.2018.02.024.