This topic causes more confusion than most, with a bewildering array of methods, and difficulties in comparing data from different laboratories. Many clients do not understand exactly what is being measured, or what they should be asking their laboratory to provide with respect to any specific site.
By definition, petroleum compounds are derived from crude oil, but crude oil itself is a highly variable material consisting of thousands of compounds, depending on where it was formed. All crude oil is formed from buried organic material subjected to intense heat and pressure, but the chemical composition can vary from light (volatile) condensates to solid, tarry bitumens. Chemically, they are all composed principally of carbon and hydrogen, but contain varying proportions of other elements: nitrogen, sulphur and oxygen being the most common (NSO). For example, a North Sea crude contains < 0.5% sulphur, whereas a crude from the Kirkuk field (Middle East) can contain 3 - 4% sulphur.
“North Sea crude contains < 0.5% sulphur, a crude from the Kirkuk field (Middle East) can contain 3 - 4%”
The non-hydrocarbons (bottom half of Fig. 1) are sometimes referred to as heterohydrocarbons, as they contain the additional elements, N, S and O (nitrogen, sulphur and oxygen) as well as carbon and hydrogen. However, in laboratories, most of the analytical efforts relate to the top half of Fig. 1, the true hydrocarbons, and these consist of two main groups, the aliphatics and the aromatics. The major difference between them is that aromatics are all based on the benzene ring and contain resonating double bonds, and this structural difference has a major effect on their toxicity.
Alternative names for aliphatic species: alkanes, paraffins, saturated compounds, mineral oil, naphthenes, non-polar compounds.
Aromatics are either monoaromatic (one benzene ring, e.g. BTEX benzene, toluene, ethyl benzene, xylenes) or polyaromatic (PAH’s – multiple benzene rings, e.g. naphthalene, benzo-a-pyrene), and aromatics include the more carcinogenic compounds, and have much lower action values in soil.
Common Petroleum Products
Once the oil reaches the refinery, it will undergo a number of processes to split the complex mixture in more useful refinery ‘cuts’. These are commonly temperature dependent, and will have varying physical properties associated with them.
A general rule is that as refinery fractions increase in boiling point range and density, they will decrease in volatility and solubility in water. The increasing molecular size of these compounds is often referred to as the carbon number. For example, pentane has five carbon atoms, whereas longer chain aliphatics commonly contain 30 or 40 carbons.
Chromatography is a process whereby complex mixtures are separated using columns (the stationary phase) and a moving liquid or gas (mobile phase), which passes through the column. Compounds within the mixture will have differing affinities for the column or the mobile phase, and thus move at differing speeds through the column, thereby allowing separation to take place. This is referred to as partitioning, and is dependent upon the volatility, molecular size, and most importantly, the polarity (which reflects the reactivity). Non-polar compounds, such as aliphatics will pass through a column on the solvent front, whereas larger or more polar compounds (e.g. polyaromatic hydrocarbons), will pass through more slowly. The flow will then pass into a detector, which will enable identification and quantification of the separated compounds. Most methods for TPH will involve some kind of gas chromatography, depending upon the compounds of interest.
The capillary column can be up to 30 m in length, with an internal diameter of <0.5 mm, and is made of glass, and coiled around the central spindle. It is housed within an oven, as the temperature is ramped up at a controlled rate to assist with the separation of individual compounds. For TPH, there are two main systems, one for the more volatile compounds (VPH), and one for the heavier, extractable compounds (EPH).
Volatile Petroleum Hydrocarbons (VPH)
These are compounds covering a carbon range of C5 – 12, and are sometimes referred to as GRO (gasoline range organics) or PRO (petrol range organics), and laboratories commonly use a headspace GCFID method for analysis, which enables the total VPH to be determined, plus individual BTEX compounds can also be identified and quantified. A soil or water sample is placed in an autosampler vial, but only half filled. Standards are added and the vial is sealed. It is then loaded onto an autosampler, where each vial is picked up in turn, and inserted into a small unit where it is heated and agitated. All volatile hydrocarbons will therefore pass into the headspace in the vial, and when the autosampler needle pierces the septum of the vial, it will take a sample of the headspace, not the soil or water, and then inject this onto the column.
Benzene, toluene, ethylbenzene and xylene will always elute in the same order, due to their polarities, and once the retention time windows are determined by running a standard, then peaks identified in these time windows in the samples can be assumed to be the respective compound, and can be quantified by determining peak height/area with that of an internal standard. Using GC and a FID (flame ionisation detector), identification is therefore performed by retention time only, unlike with GCMS (mass spectroscopy) where the mass of fragment ions is used. MTBE (methyl tertiary butyl ether) is the anti-knock additive in petrol, and is commonly requested for analysis. Due to its extreme volatility and appearance at the beginning of the chromatogram, it is recommended that this compound is always analysed using a GCMS system, which provides a more reliable result than a GCFID. A real sample will often contain up to 400 compounds, so the accompanying chromatogram will appear more complex.
“using GC and a FID (flame ionisation detector), identification is therefore performed by retention time only”
Extractable Petroleum Hydrocarbons (EPH)
Extractable petroleum hydrocarbons (EPH), in the carbon range C10 – 40, are extracted using an organic solvent, and if a total figure is all that is required, this extract can be evaporated to dryness and simply weighed to provide the EPH result. This will not contain the Volatile Petroleum Hydrocarbons (VPH), but may contain significant levels of indigenous organic material, particularly from soil samples. The solvent should be strong enough (polar) to extract degraded aromatics, and dichloromethane or hexane/acetone would be suitable for this.
This extract can be run by GCFID to provide a fingerprint chromatogram, which may be used to identify the type of product that may be present in the sample.
However, this extract will contain many thousands of compounds and does not provide sufficient information for a risk assessment, so further analysis is usually required. Gas chromatography is the method of choice, but it is preferable to clean up the extract prior to loading onto a GC column, as this will greatly improve the results. To facilitate this, the EPH extract is passed through a silica/alumina based column which removes the most polar NSO compounds (usually the indigenous compounds, such as humic acids). At this point, a two-stage process, using two different polarity solvents, is often implemented to split the extract into the aliphatic and aromatic fractions.
This will enable both fractions to be run individually on the GC systems and produce chromatograms which are much easier to identify and quantify.
This method is the basis of the Speciated TPH – Criteria Working Group (CWG) banding analysis, carbon range C5 – 40, which provides both aliphatic and aromatic fractions split into specific bands and quantified. It includes both VPH and EPH, but excludes the NSO compounds, and provides the best method for risk assessment, giving a detailed banded breakdown of the type of petroleum material, and more importantly, the proportion of carcinogenic PAHs on the site.
However, PAHs are often better analysed by GCMS, as the detection limits are better, and it is not necessary to perform a clean-up, as the MS is run in SIM (selected ion monitoring) and can ignore any other compounds present in the extract.
There are situations where a detailed breakdown of TPH is not required. This can be when delineating the area of a spill on a site, or for monitoring the progress of bioremediation. In this case, a TPH screen will be adequate, and this covers the carbon range C6 – 35, using one solvent extraction of pentane or hexane for both the VPH (only part) and EPH, with analysis by GCFID in one run. It does not usually include a clean up to remove the indigenous compounds, but this can be done if requested.
Ageing and Forensics
Laboratories are often asked if it is possible to determine when a spill happened, or if it is possible to identify the origin of the spill, and a new specialist subject has developed over the twenty years or so, known as Environmental Forensics, in order to assist consultants with these issues.
Once exposed to the atmosphere, the product will begin to change:
• Volatile compounds will evaporate
• More water-soluble compounds will be removed by rainfall/groundwater
• More reactive compounds will adhere/react with soil particles
• Microbial organisms will start to break down the product
The rate at which this happens will depend on several factors:
• Composition of the product
• Composition of the soil – matrix, pH, particle size
• Degree of exposure
• Moisture content
• Microbial content – both concentration and species
So, this is not an easy result to give. From standard chromatograms, VPH and EPH, we can often state what the product is, and if it is fresh, slightly weathered, moderately weathered, or severely weathered. We can possibly say if the spill happened weeks, months or years ago, but that is probably the limit.
“PAHs are often better analysed by GCMS, as the detection limits are better”
An additional parameter we can measure is the nC17/pristane ratio, or the nC18/phytane ratio. The nC numbers are the normal alkanes, containing 17 and 18 carbons, respectively. Pristane and phytane are isoprenoids associated with these peaks on an EPH chromatogram, and they are very resistant to weathering, whereas the nC17 and nC18 weather quite easily. In a fresh diesel, the nC17 peak is about twice the value of the pristine, and this ratio will therefore change throughout the weathering process.
The peaks labelled IP are all isoprenoids. The pristine and phytane appear as doublet peaks with the nC17 and nC18. In the degraded sample, almost all the normal (n) alkanes have gone, just leaving the isoprenoids. Laboratories are often asked for this ratio, and it does not require running the sample again, and is simply a calculation.
A further test to compare oils is a CORAT analysis – this is a Compound Ratio Analysis Technique developed by scientists at Chevron to assist in identifying oil source rocks.
TPH analysis is a complex subject and far more sophisticated analyses can be performed by GCMS to identify many biomarker compounds present in extracts from oil spills and provide a signature fingerprint, including steranes and triterpenes, but this would be an article in itself, so please contact your laboratory for further information.