Micro-pollutants: Dioxins and Furans

SUWIC has developed world-leading expertise in dioxin characterisation and control

This page gives highlights of nine research projects:


Project 1 - Characterisation and Control of Dioxin/Furan Emissions from Waste Incinerator Plants (Funded by EPSRC & Industry)


Academic Staff: Dr V.Nasserzadeh Sharifi & Professor J.Swithenbank.
Research Staff/Students: Dr Y.R.Goh (Research Associate), S.Gan (PhD Student), Dr P.J.Clarkson (Research Assistant), A.Parracho (MSc student) ,

Previous research has established that the dominant mechanisms of PCDD/F formation involve heterogeneous synthesis reactions catalysed by flyash, and that the most important factors influencing formation rates and yields in the components of incinerator plant downstream of the main combustion zones are: i) temperatures, ii) Quantities, physical condition and composition of fly-ash and iii) Gas stream composition and combustion efficiency. This EPSRC research project studied the relationships of dioxin formation with design and operational parameters, with a view to understand and predict formation rates and yields, and develop effective methods of control and abatement.

As part of our experimental program, an extensive series of tests was carried out at two large MSW incinerator plants. Ash samples were collected from various locations (ranging from the boiler to the gas clean up system). The results showed concentrations of PCDD/Fs found in the ash samples (in terms of the International Toxic Equivalents, I-TEQs) varied depending on where the samples were collected. Bottom ash samples contained almost negligible PCDD/F. The low PCDD/F concentrations in bottom ash can be explained by the fact that the high temperature exceeding 850°C in the combustion zone does not favour the formation of PCDD/Fs and any PCDD/Fs originally present in the raw waste material is thermally destroyed in the furnace. The results showed that maximum PCDD/F yields and formation rates occured in the temperature range 250- 400°C, although uncertainties remain as to how rates vary both within this range and at its extremities.

In parallel with the analytical tests, attempts were made in order to synthesise the current knowledge based on formation kinetics into a mechanistic model simulating PCDD/F formation in municipal waste incinerators. This model was then used to predict PCDD/F emissions from two operating MSW incinerators. The PCDD/F formation rate varies with temperature (T), oxygen concentration, carbon [C] and Cu2+ ion presence in the fly ash (De-novo mechanism). Using our model, the PCDD/Fs concentrations (CPCCD/F ) formed within a narrow range of oxygen concentration (7 - 12%) in the flue gas can be determined. The comparison between model predictions with measured data from ten other plants indicates that PCDD/Fs emissions are remarkably well characterised.

Back to top


Project 2 - Development of a Novel and Efficient Process for Detoxification and Re-Use of Contaminated Fly-Ash (Funded by EPSRC & Industry)


Academic Staff: Dr V.Nasserzadeh Sharifi, Professor C.W.McLeod & Professor J.Swithenbank
Research Staff/Students: D.Ward (Research Assistant), Dr P.J.Clarkson (Research Assistant), P.H.Lee (PhD Student),

Approximately 100,000 tonnes of toxic fly ash is produced each year by the UK MSW incineration industry alone. At present the ash produced by the waste incineration industry is landfilled but at considerable cost due to toxic heavy metals and PCDD/F content. However, if the ash is converted to a non-toxic product, the ash can become a potential resource to the construction industry or be landfilled at a significantly reduced cost. In response to the demand for cheap, reliable technologies for the detoxification of fly ash, a novel, energy efficient sintering process has been developed.

An ash detoxification rig was developed. Raw ash collected from two large MSW incinerators was fed into the flame zone of a 300 kW natural gas burner, ensuring rapid and effective heating of all particles. Upon leaving the burner chamber the ash is recovered from the hot combustion gases by a cyclone. Cyclone temperatures are maintained between 850-900°C, within this range softening and partial melting of the material occurs and hence the ash is sintered. The regenerative heating system comprises two burners, two regenerators, a flow reversal system and associated controls. The regenerators consist of a pair of pebble beds which pre-heat the combustion air. While burner 1 fires using air fed through regenerator 1, burner 2 acts as an exhaust port, drawing off the hot waste gases and thereby heating regenerator 2. When regenerator 2 is sufficiently hot, the reversal valves operate and the system flow is reversed. The previously cooled regenerator 1 is now re-heated by the reversed waste gas flow and conversely air pre-heat is now supplied via the opposing regenerator 2. Throughout operation, the flow is alternated between the twin halves of the system, recouping and re-using flue heat. The recuperation of energy by the regenerators means that only a small amount of fuel is required to maintain system temperature. In addition to the recovery of the waste heat, the pebble beds are also intended to both prevent the escape of volatile metals and the reformation of PCDD/F. Exhaust gases experience considerable and rapid cooling (approximately 850-150°C) whilst transiting the relevant bed. Such cooling encourages the removal of vapour phase metals by precipitation to pebble surfaces, thereby allowing scope for possible recovery. In addition, cooling the gases rapidly through the PCDD/F formation window also inhibits formation. Further PCDD/F inhibition is also intended by the removal of particulates before gas cooling. In order to assess the effectiveness of the ash detoxification process concerning PCDD/F, ash samples before and after the sintering process were analysed.

The results obtained from the GC-MS/MS have shown that it is possible to efficiently destroy the PCDD/Fs that are present in fly ash using our sintering process. A dramatic reduction in the dioxin/furan concentrationswas observed in the thermally treated ash samples. The percentage destruction of the PCDD/Fs in the ash samples was found to be in the order of above 96%. Hence, it can be concluded that our ash detoxification process is an effective technique for the detoxification of MSW incinerator fly ash. The treated product was also found to contain very low levels of leachable heavy metals.

Back to top


Project 3 - Safety/Risk Assessment for Municipal/Clinical/Toxic Waste Incinerator Plant Personnel (Funded by EPSRC and Industry)


Academic Staff: Dr V.Nasserzadeh Sharifi & Professor J Swithenbank.
Research Staff/Students: I.Delay (MPhil Student),

The overall objective of this project was to provide a methodology for assessing the risks to human health (the work-force inside the incinerator plant) posed by incinerators which can be used directly by the incineration industry in the UK and Europe. Use of the methodology improves the basis for decision making in assessments of the comparative benefits of alternative waste management strategies. The main tasks of the programme were: i) to review risk assessment methodologies with special reference to MSW, clinical and toxic waste incinerators, ii) to establish a Best Practice Risk Assessment Methodology for MSW, clinical and toxic waste incinerator staff, iii) to carry out a full risk assessment for a case study of an integrated incinerator, iv) to produce guidelines for use by the incineration industry.

Although the use of personal protective clothing and modern pollution technology should be able to minimise potential exposure to fly ash and slag and the absorption of toxic chemicals from this source, work practice observations illustrating possible sources of dermal exposure are numerous (e.g., failure to wear protective clothing, direct handling of contaminated tools, failure to wash hands, forearms and face before eating or drinking). Potential toxic substance found in airborne emissions and the solid residues in ash and slag include heavy metals (lead, cadmium, mercury, arsenic), total respirable particulates, dioxins/furans, polycyclic aromatic hydrocarbons and solvents including benzene.

A preliminary environmental monitoring programme was carried out at one of Europe's clinical waste incinerator plants. A number of wipe samples were collected from seven working surfaces in the plant where activities with the highest level of exposure to ash and incinerator waste. These samples were analysed for metals, dioxins and organics content. The adhesive tape method was used for particulate removal on working surfaces because of its superiority in recovery efficiency and suitability for application to irregular, curved and vertical surfaces. The results of the analysis by ICP showed significant amount of metals in the samples, in particular lead. The detected lead levels ranged from 11.92 μg/100cm2 in the control room to 11015 μg /100cm2 in the boiler area. Small quantities of dioxins/furans were also found in the samples taken from the boiler area . Our preliminary results show that the workers who clean boiler tubes and furnaces can be classified as having the highest levels of exposure to metals and dioxins/furans in the incinerator plants. Continuing risk assessment of dioxins/furans exposure level at other plants is underway.

Back to top


Project 4 - The Use of Derivatised Silica for the Clean Up of Organic Extracts for the Analysis of Polychlorinated Dibenzo-p-Dioxins and Furans


Academic Staff: Professor C.W. McLeod, Dr .V. Nasserzadeh Sharifi & Professor J.Swithenbank.
Research Staff/Students: Dr P.J.Clarkson (Research Assistant), V.Browne (MEng student),

Derivatised silica materials have been investigated for use in the clean up of organic extracts for the analysis of PCDD/F's. The current accepted methodologies involving large packed bed silica columns require large quantities of silica and solvent. With the development of solid phase extraction (SPE) technologies it has been possible reduce the size of many clean up procedures thus saving resources, time and money. In SPE, compounds are separated according to their relative affinities for the stationary phase and the mobile phase in much the same was as in reversed phase chromatography. For the clean up of PCDD/F's octadecyl bonded silica (C18-Si), cyano-propyl bonded silica (CN-Si) and phenyl bonded silica (Ph-Si) have been investigated. The derivatisation process replaces the active adsorptive hydroxyl sites on the surface of the silica with a different functional group thus changing the properties of the material.

Three types of modified silica were prepared using one of the three trichlorosilane derivatives octadecyltrichlorosilane, phenyltrichlorosilane or 4-(trichlorosilyl)-butyronitrile. Anhydrous silica (7g) and the trichlorosilane derivative (5mL) are refluxed for 2 hours in pyridine (40mL). The modified silica is then filtered and washed with toluene (200mL), methanol (200mL), methanol with 10% NaOH(200mL), distilled water(200mL). The modified silica is then dried in an oven at approximately 100?C overnight.

A solution of 1% diesel, 1ppm acenaphthylene, fluorene, anthracene, phenanthrene chrysene, and 1ppm 2,3,7,8-TCDD in hexane was prepared as a test mix. This mixture was to represent several major classes of organic compounds: aliphatic, aromatic and PCDD/F. 2g of derivatised silica was lightly packed into a glass column and pre-eluted with 10mL n-hexane. To the column 1mL of the test mixture was added. To obtain an elution profile of each of the silica's the analytes were eluted with 5mL n-hexane and then 5mL toluene, collected in 1mL aliquots. The individual aliquots were analysed using GC-MS for the aliphatics and aromatics and GC-MS-MS for the TCDD.

The Varian 3800 gas chromatograph with Varian Saturn 2000 ion trap mass spectrometer was used with a 30m x 0.25mm ID x 0.25m m film DBX 5 capillary column. For the analysis of PCDD/F, the GC oven temperature was programmed from 155°C (3.5 mins) to 235°C at 25°C min-1, followed by 10 minutes at 235°C, then to 275°C at 5°C min-1, then to 320°C at 10°C min-1; the final temperature being held for 3 minutes. A temperature programme of 50°C (3 mins) to 320°C (5 mins) at 15°C min-1 was used for the analysis of the aliphatic and aromatic compounds. The MS/MS parameters for the analysis of PCDD/F were optimised to give maximum sensitivity and selectivity. Quantitation was achieved by calculating the peak area for both parent and daughter ions of the congeners under investigation. For the purposes of this investigation, a single internal standard compound was used, and all congeners were quantified relative to this.

From the data obtained, elution profiles were obtained. The results showed that the phenyl bonded silica could seperate the aliphatic compounds from the aromatic and PCDD/Fs. Whereas the C18-Si can separate the aromatics from the aliphatics and TCDD. And finally the CN-Si material can seperate all three investigated classes. Hence it has been shown that different derivatised silica materials exhibit different elution profiles for the test mix used. Therefore it is possible to selectively elute the compounds of interest, in this case TCDD. By using a combination of two different phases, such as C18-Si and then CN-Si it is possible to remove all major interferences from the sample before analysis. By using SPE in such away it is possible to reduce the amount of material used in the analysis of PCDD/F, as well as reducing the volume of solvent used. It has been shown that this technique is a viable alternative to other clean up methodologies and it will be investigated further using isotope dilution to validate the method further.

Back to top


Project 5 - Development of a Method for the Extraction of Polychlorinated Dibenzo-p-Dioxins and Furans from Milk and Serum using Accelerated Solvent Extraction


Academic Staff: Professor C.W.McLeod, Dr V.Nasserzadeh Sharifi & Professor J.Swithenbank.
Research Staff/Students: V.Browne (MEng student), Dr P.J.Clarkson (Research Assistant),

Over a quarter of the estimated average daily intake of dioxins per person is thought to be due to the consumption of milk and other milk related products. High dioxin levels in cows milk and human milk have been reported, so there is great scientific value in the analysis of such samples. Extraction of milk is a difficult procedure due to the complex nature of milk, with its composition of sugars, vitamins, proteins and minerals being colloidally dispersed. The fat content is present as globules, which are emulsified and suspended in the aqueous phase of the milk. Due to the lipophilic nature of dioxins and furans it is necessary to rupture these fat globules for efficient extraction from the sample matrix. Other biological samples have been studied and a comparison showed that dioxin levels were highest in blood, followed by adipose tissue, and human milk. The same theory is applied for the extraction of these samples due to their similar compositions. Digestion of the fat contents in these samples has been achieved in many ways; the use of concentrated sulphuric acid, potassium hydroxide, formic acid and homogenisation being just some.

Accelerated Solvent Extraction is a technique that combines elevated temperatures and pressures with liquid solvents. The solvents used are the same as used in standard liquid extraction. ASE gives recoveries similar to those obtained with Soxhlet extraction and other techniques but requires much shorter extraction times and uses less solvent. In addition, the samples are extracted at temperatures above the boiling point of the solvent used and this improves the kinetics of mass transfer. By immobilising the liquid, such as milk or serum on a solid support matrix, it is possible to extract the compounds of interest using ASE. This work investigated the use of Hydromatrix, a diatomeous earth as a solid support matrix.

A 7mL aliquot of milk or serum was placed in an 11mL stainless steel extraction cell with 3.5g of Hydromatrix. The sample was spiked with 5 µL of a 10ppm standard solution of 2,3,7,8 tetrachloro dibenzo-p-dioxin. The sample was then extracted using an ASE. Extractions were performed using accelerated solvent extraction with hexane:2-propanol in a 2:1 ratio as the extraction solvent. The extraction conditions used were as follows: Temperature 120oC, pressure 1500psi, static time 1 minute with 3 static cycles. The extracts were cleaned up using standard procedure and analysed using the Varian 3800 gas chromatograph with Varian Saturn 2000 ion trap mass spectrometer.

The GC-MS results showed that the 2,3,7,8-TCDD spike was present in the extracted samples. Using the internal standard it was possible to calculate recoveries of between 85% and 95%. As the dioxins are present in the lipid layer, to confirm that the extraction was successful, separate HPLC analysis of the extract show high levels of fatty acid. This data therefore confirms the success of this extraction of 2,3,7,8-TCDD and other PCDD/Fs from biological matrices.

As Hydromatrix remains as a powder after being saturated in the milk, the milk sample can be considered as a solid sample and therefore ideal for ASE. These initial results show that by using an immobilising support such as Hydromatrix it is possible to extract PCDD/F from liquids using ASE. This method has many potential advantages over the current methods, such as speed; the fact that no chemical pre-treatment is necessary; that the contact between analyst and sample is reduced and that the volume of sample is low. Combined with the reliability and reproducibility of the ASE equipment these initial results show that there is potential for the further development of this methodology.

Back to top


Project 6 - Tree Bark as a Passive Sampler for Polychlorinated Dibenzo-p-Dioxins and Furans


Academic Staff: Professor C.W.McLeod, Dr V.Nasserzadeh Sharifi & Professor J.Swithenbank.
Research Staff/Students: Dr P.J.Clarkson (Research Assistant),

Tree bark is an effective substrate for collection of airborne-derived environmental contaminants such a heavy metals, PCBs and PAHs. Given that PCDD/F are likely to be associated with fine airborne particulate matter which, through wet and dry deposition processes, will be retained and accumulated by bark over a relatively long time period, tree bark represents a potentially important new sampling strategy for high sensitivity PCDD/F measurement.

In combination with ion trap GC-MS the possibility therefore exists to develop a low cost high throughput screening capability for detection of airborne-derived PCDD/F. By comparing background levels from soil with data from seasonal sampling of vegetation such as leaves and grasses it is possible to obtain temporal data for the release of PCDD/F into the environment. However, to study the long-term accumulation of organic pollutants in the environment it is necessary to use a species that is not subject to seasonal change. As trees are ubiquitous throughout both the urban and rural environment, it was decided to investigate the use of tree bark as a matrix for the trapping of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans. This work utilises accelerated solvent extraction/column clean-up and GC-MS-MS and targets sampling sites in and around a major UK city.

Between 30g to 100g of bark were removed from the surface of each tree, 1-2 m above the ground. The nature of the bark determined the sampling procedure: Peeling bark was manually collected by removing the bark bits from the trunk. Other types of bark were removed by scraping the surface and collecting the material. As the organic compounds of interest are deposited on the surface of the tree bark, only the external outer bark, no deeper than 2 mm, was required. Samples were frozen in liquid nitrogen prior to crushing in a teemer mill. The powdered tree bark was stored at room temperature in aluminium foil. Extractions were performed using accelerated solvent extraction with toluene as the extraction solvent. The extracts were cleaned up using our standard procedure and analysed using the Varian 3800 gas chromatograph with Varian Saturn 2000 ion trap mass spectrometer. The validity of the method is shown in Fig. 8, with the separation of 13 TCDD congeners.

It was observed that tree bark samples taken from near to different emission sources showed different patterns of congeners. Sources of dioxin in the samples can be tentatively identified based on the congener patterns obtained. Bark sampled from near a solid waste incinerator showed distinctive patterns of HpCDD and OCDD congeners. Whereas bark samples from near a decommissioned chemical waste incinerator, show distinctive TCDD and PCDD patterns. Distinct differences were observed in the PCDF data.

The results from both the hospital and chemical incinerators show that the persistence of dioxin in the environment is a long-term problem, with the tree bark still showing appreciable levels of dioxin several years after the removal of the emission source. At present the processes by which the majority of the dioxins are deposited on the surface of the bark is unclear, yet the obvious health implications associated with inhalation of fine airborne particulate matter needs to be assessed. Using a simple quantitative approach it is possible to determine that PCDD were present in the range of 10-190 ng kg-1 and PCDF in the range 10-100 ng kg-1.

Tree bark has been shown to act as a passive sampling media for dioxins, which in combination with GC-MS-MS provide a new measurement strategy for long term assessment of environmental contamination. By combining the data obtained from the analysis of dioxins in tree bark with that from seasonally dependent herbage such as grasses it would be possible to better characterise the occurrence, spatial distribution and sources of dioxins in the environment. There is scope for further development and refinement of the methodology and future papers will deal with fully quantitative analysis using EPA 1613 and direct comparison with high resolution GC/MS. It is proposed to perform a detailed survey of the urban and rural environment (tree bark, soils, herbage) in the near future.

Back to top


Project 7 - Application of the Immunoassay Analysis Technique for the Measurement of PCDD/F Concentrations in Fly Ash


Academic Staff: Professor C.W.McLeod, Dr V. Nasserzadeh Sharifi & Professor J.Swithenbank.
Research Staff/Students: D.Ward (Research Assistant), Dr P.J. Clarkson (Research Assistant).

The main objective of this project was to compare the use of the Enzyme Immunoassay technique (EIA) for the analysis of PCDD/F with the conventional GC-MS method. The CAPE Technology's High Performance Dioxin/Furan Immunoassay technique is a newly developed enzyme immunoassay approach for analysis of PCDD/Fs . The basis of all immunoassays is the specificity of an antibody molecule for a particular chemical structure. The noncovalent binding of the target analyte to the antibody's binding site occurs because of the lock and key fit between these complementary structures. Molecules which differ slightly from the target structure may also bind to the antibody, though less tightly than the target analyte. Molecules which differ greatly from the target structure will not bind to the antibody. Cross-reactivity, the ability of an other molecules to bind to the antibody, decreases as the structural similarity to the target analyte decreases. The dioxin/furan congener recognition profiles correlates to the toxic equivalency factors (TEFs) of the individual PCDD/F congeners. Hence because of this specificity, results obtained by this technique thereby correlate to ITEQs.

Fly ash samples collected from the bag filter plant of a MSW incinerator were tested in this project. Half of the samples were analysed in accordance with the analysis methodology described earlier. The other half of the samples were prepared for EIA analysis in two consecutive batches, each consisting of three sub-samples and a blank. Samples of treated ash were prepared first, so as to avoid cross contamination. Thirteen EIA tubes were prepared for the final analysis; 8 for prepared samples and 5 for calibration control standards (including a negative control). 10 μl of each prepared sample was added to individually allotted tubes and likewise for each control standard. Tubes were allowed to incubate at room temperature for approximately 17 hours. After this period the tube contents were dumped and 500 μl of Competitor-HRP Conjugate Solution added. After a further 15 minutes the contents were again dumped and 500 μl of HRP-Substrate Solution added. As per manufacturers instructions 30 minute was allowed to enable blue colouring of tube contents to develop, as was observed. After the allotted time, 500 μl of Stop Solution was added to each tube. Immediately the tube contents were seen to turn yellow, as expected. Using the dedicated differential photometer, the optical density (OD) of each tube was measured in comparison to a blank tube containing de-ionised water. This then allowed the PCDD/F concentration in the samples to be calculated.

In comparison to the GC-MS technique, the immunoassay technique was found to be simple to perform and a much lesser degree of technical expertise is required. Despite the 12-24 hour incubation period, the actual number of man hours devoted to EIA analysis was relatively short (about 30-45 minutes for 13 tubes). Hence the amount of actual 'hands on' technician time required to produce a result can be approximated at a few minutes per sample. As with GC-MS analysis, most of the time was spent in sample preparation, taking about 2 complete working days per batch. Apart from the addition of spiking materials and a final re-elution to methanol, virtually the same laborious extraction and clean-up technique is used as for GC-MS analysis. Parallel GC-MS/MS analysis of prepared samples revealed numerous PCDD/F compounds in samples, as indicated by the EIA technique. The EIA technique however, offers no accountability for PCDD/F quantities lost during extraction and clean-up and hence original sample concentration results are speculative. Triplicate analysis of the ash yield varying results and this can be attributed to EIA inconsistencies, the heterogeneous nature of the sample or variations in preparation loss. The detection limit of the EIA technique on the basis of a sample matrix load of 1000mg per tube is 4 pg/g ITEQ. This is relatively higher than many GC-MS instruments but may still be sufficient for some applications. The technique is thus valuable for extensive scoping studies complemented with GC/MS analysis of selected samples.

Back to top


Project 8 - Polycyclic Aromatic Hydrocarbon Emissions from Waste Incineration


Academic Staff: Professor P.T.Williams, (Leeds University), Dr V.Nasserzadeh Sharifi & Professor J.Swithenbank, (Sheffield University)
Research Staff/Students: C. Salt (PhD Student)

Polycyclic aromatic hydrocarbons (PAH) are one group of chemicals which are at present unregulated but are of great concern since amongst the environmental chemical groups, PAH are known to be the most carcinogenic. There are very little sampling and monitoring data concerning the emissions of PAH from waste incineration and also their formation under furnace and flue gas conditions are not known. Whilst PAH are as yet unregulated, the incineration industry, environmental groups and legislators would derive great benefit from an understanding of the levels of PAH emitted to the environment. Additionally, an understanding of their reaction mechanisms under a variety of process conditions would consequently lead to the development of control strategies.

This on-going collaborative research programme between the Department of Fuel and Energy, Leeds University and SUWIC strives to investigate the formation of PAH from the incineration of single component waste, simple mixtures and simulated municipal solid waste, through laboratory and full scale incinerator plants experimental programmes. Analyses of flue gas particulate and vapour phase PAH were carried out by GC and GC/MS methods to allow the wide variety of PAH species, including other PAH derivaties such as nitrogen - oxygenated - and sulphur -PAH, to be identified and quantified. The analyses also provides the vital data needed for the extension of SUWIC's Fluid Dynamics Incinerator Combustion (FLIC) code to include the formation of PAH during incinerator bed combustion

Back to top


Project 9 - Investigation into the development of more robust dioxin sample preparation, clean up and analysis protocols (i.e. ion trap MS-MS, High resolution ion trap)


Academic Staff: Dr V Nasserzadeh Sharifi, Professor J Swithenbank, Professor C McLeod
Research Staff/Students: Dr Paul Clarkson (Research Assistant).

Literally hundreds of potentially toxic organic substances occur in a wide variety of environmental matrices at the ultra-trace level. As a result, extensive sample pre-treatment/work-up, under ultra-clean conditions, is required before final measurements are made by gas chromatography-mass spectrometry (GC/MS). Currently data on Dioxin/Furan emissions from different industrial plants is of doubtful value because different sampling and analytical methodologies have been used. Data quality is also under question as a result of so many different and evolving international standards. Thus the emissions rates for these compounds are generally given as a range of values and this range for many sectors is particularly wide, reflecting the uncertainty in the present data.

The MS/MS technique has three basic steps. These are : Step 1) The gate electrode admits electrons from the filament through holes in the end cap. Analyte molecules entering the cavity undergo electron impact ionisation and become fragmented. Step 2) The largest fragments, the parent ions, are captured in a magnetic field by the application of a radio-frequency oscillating voltage to the central electrode. This causes them to circulate in stable trajectories around the trap while all other ion fragments are expelled. Hence the parent ions are selectively isolated. This step virtually eliminates the chemical background interference, assuring selectivity. Step 3) In order to confirm analyte identity, the isolated parent ions undergo a second fragmentation. This is achieved by collision induced dissociation (CID). A CID voltage is applied to the endcap electrodes, further exciting the trapped ions. The energised parents fragment as a result of collisions with the helium carrier gas atoms. An ion detector collects the expelled product daughter ions. Confirmation of the analyte identity is achieved by examining the characteristic mass spectrum of the daughters.

The technique is termed "MS/MS" due to the employment of the duel fragmentation of the analyte. MS/MS operation requires the pre-determination of essential parameters particular to the desired analyte. To perform MS/MS analysis the following information are required: GC column retention time, Storage voltage and CID voltage of each analyte. A series of preliminary tests was therefore carried out using PAH and PCB stock solutions in order to determine the above data.

Determining PCDD/F Retention Times: MS/MS operates by individually searching for each analyte as it leaves the GC column. This is achieved by the setting of time windows or segments in which the MS is instructed to target a particular molecular mass. To be effective the segment must correspond to the analyte GC column retention time. Hence the first step was to determine the retention times of each of the seventeen 2,3,7,8-substitute PCDD/F congeners. This was achieved by the analysis of a PAR (Precision and Recovery) standard solution containing the seventeen 2,3,7,8-PCDD/F substitute congeners. MS parameters for an electron ionisation (EI) analysis method were then established. EI is the simplest mode of Saturn 2000 operation. Compounds are fragmented by a stream of electrons and the resulting ions accelerated towards the detector to produce a mass spectrum. In order to achieve adequate separation of the individual congeners by GC, it was necessary to set an appropriate column oven program. Using EI analysis and the oven temperature sequence, 1m l of PAR solution was injected into the GC-MS. The Saturn 2000 handling software is equiped with a fairly comprehensive mass spectrum library data base. Using the search facility, the system was able to identify the homologue of each of the seventeen PCDD/F compounds. Examination of the PAR chromatogram allowed the determination of GC column retention times for each of the 17 compounds.

Determining Conditions for Selective Ion Capture - In order to selectively capture the analyte parent ions , appropriate excitation storage levels must be set. The storage level relates to the strength of the magnetic trapping field, and needs to be low enough to capture the target ions, but high enough to enable strong entrapment. The manufacturers operating instructions recommend the mass to charge ratio of 0.4. Hence the excitation storage levels for each of the 17 PCDD/Fs were set.

Determining Conditions for Collision Induced Disassociation (CID) - The capture of an ion is insufficient to prove identity. Selective capture targets analytes by molecular mass and therefore other similarly massive compounds can also be trapped. Positive identification is achieved through collision induced dissociation, (CID). CID is achieved by the electromagnetic excitation and consequent helium collision of the captured parent ions. Essential to the process is the CID voltage or excitation amplitude of the electromagnetic field. Too little can fail to induce fragmentation, too much can result in an over fragmentation, or annihilation of the ion. An appropriate CID voltage should produce a mass spectrum containing discernible daughter ion(s). Determination of the appropriate CID voltage was achieved by automatic method development (AMD). Saturn 2000 AMD operation applies a range of CID voltage levels in rapid and repeating succession. Examination of the mass spectra produced by each CID voltage allowed the most appropriate level to be selected for MS/MS operation. Hence, suitable CID voltages for each of the 17 PCDD/Fs were obtained through AMD analyses of the PAR standard solution.

Testing MS/MS Operating Parameters - Using the above information, the MS/MS operating program was composed for the qualitative analysis of PCDD/Fs. Comparison with the chromatogram obtained through EI analysis, shows MS/MS to eliminate much of the background noise, producing a more flat base line and more clearly defined peaks.

PCDD/F Identification in Real Samples - PCDD/F identification in real samples were determined by comparing chromatograms and mass spectra to those obtained through analysis of the PAR solution. Providing column and oven conditions are unchanged, compound retention times remain identical for successive analyses. Chromatogram peaks generated using EI analysis occur with identical retention times to those generated under MS/MS analysis conditions. Hence by examining the PCDD/F homologue retention times produced by EI analysis of the PAR solution, MS/MS identification in real samples can be achieved. CID of the PCDD/F parent ion results in the loss of 63m/z, possibly a fragment constituting a chloride, a carbon and an oxygen atom. Such behaviour is characteristic of PCDD/F and hence detection and homologue identification in real samples can be achieved through the presence of the appropriate daughter ions in MS/MS mass spectra. Real samples can not be expected to contain only the seventeen 2,3,7,8-substitute congeners; as in a PAR standard solution, any of the 210 congeners could be present. Non-2,3,7,8-substitute congeners will also produce the same daughter ions. Hence detection and identification of the target 17 compounds is achieved through an examination of both chromatogram peak retention time and mass spectra daughter ions.


Back to top

Ash detoxification rig.

The regenerative heating system comprises two burners, two regenerators, a flow reversal system and associated controls.

The regenerators consist of a pair of pebble beds which pre-heat the combustion air. While burner 1 fires using air fed through regenerator 1, burner 2 acts as an exhaust port, drawing off the hot waste gases and thereby heating regenerator 2. When regenerator 2 is sufficiently hot, the reversal valves operate and the system flow is reversed. The previously cooled regenerator 1 is now re-heated by the reversed waste gas flow and conversely air pre-heat is now supplied via the opposing regenerator 2.

Throughout operation, the flow is alternated between the twin halves of the system, recouping and re-using flue heat. The recuperation of energy by the regenerators means that only a small amount of fuel is required to maintain system temperature.

In addition to the recovery of the waste heat, the pebble beds are also intended to both prevent the escape of volatile metals and the reformation of PCDD/F. Exhaust gases experience considerable and rapid cooling (approximately 850-150°C) whilst transiting the relevant bed. Such cooling encourages the removal of vapour phase metals by precipitation to pebble surfaces, thereby allowing scope for possible recovery. In addition, cooling the gases rapidly through the PCDD/F formation window also inhibits formation.