Polycyclic Aromatic Hydrocarbons (PAHs) have been detected in rubber and plastic components of a number of consumer products such as toys, tools for domestic use, sports equipment, and footwear, with carbon black and extender oils having been identified as principal sources.
Journal of Chromatography A, 1566 (2018) 13–22 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma A fast and selective method for the determination of carcinogenic polycyclic aromatic hydrocarbons in rubber and plastic materials Otmar Geiss ∗ , Chiara Senaldi, Ivana Bianchi, Ana Lucena, Salvatore Tirendi, Josefa Barrero-Moreno European Commission, Joint Research Centre, Directorate F Health, Consumers and Reference Materials, 21027, Ispra, VA, Italy a r t i c l e i n f o Article history: Received 22 March 2018 Received in revised form 15 June 2018 Accepted 19 June 2018 Available online 21 June 2018 Keywords: Polycyclic aromatic hydrocarbons Rubber and plastic materials Randall hot extraction Molecularly imprinted polymers a b s t r a c t Polycyclic Aromatic Hydrocarbons (PAHs) have been detected in rubber and plastic components of a number of consumer products such as toys, tools for domestic use, sports equipment, and footwear, with carbon black and extender oils having been identified as principal sources In response to these findings, the European Union Regulation (EU) No 1272/2013 was adopted in December 2013, amending entry 50 in Annex XVII to the Registration, Evaluation, Authorisation, and Restriction of Chemicals (REACH) directive establishing a restriction on the content of eight individual carcinogenic PAHs in plastic and rubber parts of products supplied to the public This work proposes a simple, relatively fast, and cost effective method for determining the concentrations of each of these eight carcinogenic PAHs for compliance testing Existing methodologies were taken as a starting point, improving in particular the extraction and the clean-up procedures Randall hot extraction and ultrasonic extraction were compared with regard to their extraction efficiency Randall hot extraction proved to be more efficient (10–40%, depending on PAH) Sample extract clean-up performance was qualitatively assessed for silica-packed columns and molecularly imprinted polymers (MIPs) solid phase extraction (SPE) cartridges The use of highly selective MIP-SPE cartridges removed most of the undesired contaminants, highlighting their superiority with regard to traditional, silica-based purification methodologies The introduction of Randall-hot extraction for sample extraction and MIP-based solid phase extraction cartridges for selective clean-up represents a novel advance compared with previously reported methods in this field In combination with gas chromatography-mass spectrometry (GC–MS) analyses in selected ion mode, the method was found to be excellent in terms of extraction efficiency, extract purity, and speed © 2018 The Authors Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) Introduction Polycyclic aromatic hydrocarbons (PAHs) are a ubiquitous group of several hundred chemically related, environmentally persistent compounds, many of which have mutagenic, toxic, and/or carcinogenic properties [1,2] Known sources of PAHs in consumer products are extender oils and carbon black which may unintentionally contain various levels of PAHs [3,4] Carbon black is used as reinforcing filler in rubber formulations or as pigment in plastics [5], whereas extender oils are used as plasticiser oils/softeners Examples of consumer products in which PAHs were detected are: household items, tools, clothing, footwear, toys and childcare arti- ∗ Corresponding author at: European Commission, Joint Research Centre, Directorate F – Health, Consumers and Reference, Materials, Via E Fermi, 2749, 21027, Ispra, VA, Italy E-mail address: otmar.geiss@ec.europa.eu (O Geiss) cles [5] as well as floor tiles used in gyms or on synthetic turf fields made of recycled rubber [6] Consumer exposure, in particular of children, to PAHs via oral and dermal contact with these articles has been the subject of considerable public attention over the past decade [7], triggering the establishment of restrictions on PAH content According to paragraphs and of entry 50 in Annex XVII to Regulation (EC) No 1907/2006 of the European Parliament and Council from 18 December 2006 concerning the Registration, Evaluation, Authorisation, and Restriction of Chemicals (REACH) [8], only articles compliant with the conditions laid out therein, including the limits on polycyclic aromatic hydrocarbons (PAHs), can be sold to the general public in the European Union (EU) For eight priority PAHs – all presumed carcinogens (group 1B according to CLP classification) - content limits of rubber and plastic components were set to 0.5 mg kg−1 for toys and childcare articles and to mg kg−1 for all other consumer articles Reliable methods/standards to analytically determine the concentrations of each of these eight carcinogenic PAHs are required for compliance test- https://doi.org/10.1016/j.chroma.2018.06.047 0021-9673/© 2018 The Authors Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) 14 O Geiss et al / J Chromatogr A 1566 (2018) 13–22 ing against the established limit values The European Commission recently (December 2017) launched a standardisation request to the European Committee for Standardisation [7] as no such harmonised European or international standard is available at present A number of methods for the determination of PAHs in a variety of matrices, that are in most cases neither rubber nor plastics, are available From among these methods, that used in the German Product Safety Commission (AfPS) GS 2014:01 PAK specification [9], in which the AfPS sets the requirements of PAH testing in the course of GS mark certification, comes closest to EU requirements This AfPS requirement restricts 18 PAHs in articles and consumer products and proposes a method for the analysis of PAHs in rubber and plastic, based on h ultrasound extraction in toluene at 60 ◦ C, purification of the extract on silica gel packed columns, and subsequent analysis with GC MS The Fraunhofer Institute for Process Engineering and Packaging (IVV) has recently published a method [10] to determine PAH concentrations in rubber material, based on a more efficient and faster extraction process, that uses accelerated solvent extraction with cyclohexane instead of ultrasound extraction with toluene After extraction, the sample is purified with normal-phase SPE cartridges and analysed with GC–MS ISO/TS 16,190 [11] describes a test method to quantitatively determine polycyclic aromatic hydrocarbons (PAH) in footwear materials by using n-hexane at 60 ◦ C in an ultrasonic bath for h to extract the test material An aliquot is then analysed using either GC–MS or HPLC without prior purification of the sample extract CEN EN 16,143-2013 [12] describes a method for the determination of benzo[a]pyrene and other selected PAHs in petroleum products (e.g extender oils) The product is dissolved in n-pentane and submitted to a double cleaning step using silica-based column chromatography The final extract is then analysed by GC–MS ISO method 21,461 [13] allows for the selective determination of polyaromaticity of oil in vulcanized rubber compounds This method is not specific to individual PAHs and is based on high-cost nuclear magnetic resonance (NMR) spectrometry A study commissioned by the carbon black industry [14] investigated the extraction and migration behaviour of PAHs from rubber formulations containing carbon black They proposed a method in which the sample material is Soxhlet-extracted with toluene, the sample extract then purified with silica gel packed chromatographic columns, and the analysis carried out with GC–MS While some of the above methods entail high equipment costs as a main challenge for implementing them in enforcement scenarios, others lack the specificity and capability of determining individual PAH concentrations in plastic and rubber materials which imposes severe limitations on their applicability for these purposes In addition, some of these methods suffer from outdated protocols, often in the extraction or purification steps which are no longer state of the art A comparison of already published methods/studies in terms of sample extraction and sample purification can be found in Table The purpose of this study was to develop a simple, relatively fast, and cost effective method for the determination of priority PAHs in rubber and plastic materials Existing methodologies served as starting points with the particular aim to improve the extraction and the clean-up steps Randall hot extraction and ultrasonic extraction were compared with regard to their extraction efficiency Sample extract clean-up performance was qualitatively assessed for silica-packed columns and molecularly imprinted polymers (MIPs) solid phase extraction (SPE) cartridges We found that the optimal method in terms of extraction efficiency, extract purity, and time demands, was Randall hot extraction of the rubber or plastic material, followed by sample extract clean-up with PAH specific solid phase extraction cartridges based on molecularly imprinted polymers (MIPs), in combination with GC–MS analysis in selected ion mode The introduction of Randall-hot extraction for sample extraction and MIP-based solid phase extraction cartridges for selective clean-up represents a novel and significant advance compared with previously reported methods in this field Materials and methods 2.1 Selection of test materials and sample preparation We chose three test materials for this study: carbon black N772 (2.5%) containing soft polyvinylchloride (PVC) and two natural/butadiene rubber (NR/BR) blends; one containing carbon black N375 (24.1%) and distillate aromatic extract (DAE, 2.7%), and the other containing carbon black N375 (24.1%) and treated distillate aromatic extract (TDAE, 2.7%) These materials were chosen for two reasons: (i) they should be representative of the two polymer groups: plastics and rubber; and (ii) they should be able to account for the impact different PAH concentrations in the same material could have on factors such as the recovery rates Soft PVC was chosen as the representative of plastic materials as its high phthalate content poses a challenge in the clean-up step Since singularly available, pure carbon blacks N375 and N772, as well as pure treated and untreated distillate aromatic extracts (the same used during the manufacturing process of the NR/BR blends and soft PVC), were separately analysed as well Although not described in this study, the present protocol was also applied for the analysis of low density polyethylene, polystyrene, ethylene propylene diene monomer and silicone rubber and proved to be suitable All test materials were custom made at external laboratories It should be noted that the rubber blends and the PVC used in this study are not necessarily representative of common commercial consumer products 2.2 Sample preparation – size of sample material Using scissors, the sample materials were cut into pieces with an edge length of < mm 2.3 Extraction While the main extraction method used in this study is Randall hot extraction, we also employed ultrasonic extraction for purposes of comparison The weighed-in quantities of sample material were adjusted so their PAH content would fall within the linear calibration range of the GC detector The weighings given below are therefore only valid for the specific materials investigated in this study 2.3.1 Randall hot extraction Around 40 mg of the DAE containing NR/BR, or 100 mg of either the TDAE containing NR/BR or the PVC were weighed exactly into cellulose extraction thimbles (Whatman, Maidstone, UK, single thickness, 33 mm x 80 mm, Product Code 2,800,338), which were transferred to the respective extraction cups, adding 20 L of isotope labelled internal standard (2500 ng mL−1 in toluene) to the interior base of the extraction thimbles, next to the sample material Then, 95 mL of toluene were added to the extraction cups The extraction was done with a Velp SER 158 solvent autoextractor (Velp Scientifica, Usmate, Italy) set to the highest heating level, using 120 for immersion, 20 for removal, 30 to wash, for recovery, and 15 for cooling, which resulted in a total extraction time of just over h and a final sample extract volume of approximately 20 mL Gaskets made of Vaflon were used between the connection funnel solvent and the extraction cup 2.3.2 Ultrasonic extraction Around 40 mg of the DAE containing NR/BR or around 100 mg of either the TDAE containing NR/BR or the PVC were weighed exactly Table Comparison of methods/studies for the extraction and determination of PAHs from rubber and/or plastics with regard to sample extraction and sample purification Sample Extraction Sample extract purification Remarks Technique Timed Coste Extraction efficiency Technique Time consuming? Coste Clean-up efficiency Study commissioned 2009 by carbon black industry [14] Toluene Soxhlet 16 h (320 cycles) 3-5 kD (6 pos.) Complete extractiondc Silica gel packed column Yes Packing and long conditioning procedure Variable Unselective Only polar compounds are retained ISO/TS 16,190 [11] 2013 n-Hexane Ultrasounds extractions 1h 1-2 kD Incomplete extractionb No clean-up n/a n/a No clean-up CEN EN 16,143 [12] 2013 n-Pentan n/aa n/aa n/aa n/aa Silica gel packed column Yes Packing and long conditioning procedure Variable Unselective Only polar compounds are retained AFPS GS2014:01 PAK [9] 2014 Toluene Ultrasounds extractions 1h Incomplete extractionb Silica gel packed column Yes Packing and long conditioning procedure Variable Unselective Only polar compounds are retained Fraunhofer IVV [10] 2017 Cyclo-hexane Accelerated solvent extraction (ASE) 45 45-50 kD Complete extractionc Normal phase SPE No Commercially available D /cartr Unselective Only polar compounds are retained Toluene Randall-hot extraction 3h 10-14 kD (6 pos.) Complete extractionc MIP-SPE No Commercially available and easy/fast protocol D /cartr Highly selective Method proposed in this study a b c d e Year of publication Extraction of PAHs from cured rubber formulations containing carbon black Not tested for plastic materials Method for determination of PAHs in footwear materials Not specific for rubber and plastic materials Method for determination of benzo[a]pyrene and other PAHs in petroleum products Used in the course of German GS mark certification Method which comes closest to the method described in this study Method for extraction of PAHs from recycled rubber Not tested for plastic materials O Geiss et al / J Chromatogr A 1566 (2018) 13–22 Extraction Solvent Method/Study Petroleum product directly solved in n-pentane No extraction required The current study suggests that ultrasound extraction may be less efficient compared to Randall hot extraction Extraction efficiency verified in referenced study Total extraction time Indicative acquisition cost (Italy) 15 16 O Geiss et al / J Chromatogr A 1566 (2018) 13–22 into 100 mL Erlenmeyer flasks, to which 95 mL toluene and 20 L of isotope labelled internal standard (2500 ng mL−1 in toluene) were added The capped flasks were then placed into an ultrasonic bath (800 W, 59 KHz, bath area 900 cm2 ) for h at 60 ◦ C 2.4 Clean-Up/Purification Toluene sample extracts were evaporated to dryness with a rotary evaporator, in a 60 ◦ C water bath and adjusting the vacuum to 90 mbar The clean-up procedure was done with solid phase extraction cartridges (SPE) filled with a sorbent based on highly PAH-selective molecularly imprinted polymers (MIPs) Results from this approach were compared against results obtained applying the German AfPS method [9], which is based on adsorption chromatography on silica gel TM 2.4.1 Solid phase extraction with SupelMIP cartridges The dry extract was reconstituted in mL hexane and cleanedup using solid phase extraction (SPE) cartridges filled with MIPs TM (Supelco, SupelMIP PAHs, 50 mg/3 mL, product code 52773-U) The following procedure was used: conditioning mL cyclohexane, loading of sample, washing with mL cyclohexane, elution with x mL ethylacetate, evaporating the ethylacetate extract to dryness with a nitrogen evaporator (heating block set to 40 ◦ C), and reconstituting in mL toluene for GC–MS analysis 2.4.2 Clean-up with silica gel packed columns A glass chromatography column (20 cm × cm) was packed with g of previously deactivated silica (Supelco, Washed silica, product code 21342-U), employing the wet packing method, to which cm of anhydrous sodium sulphate was added Deactivation was achieved by adding 10% in weight of ultrapure water to the silica and subsequent homogenisation for h The column was conditioned with 10 mL petroleum ether The dry extract was then reconstituted in mL toluene and loaded onto the column Elution of PAH was achieved with 50 mL petroleum ether In the next step, the petroleum ether extract was evaporated to dryness with a nitrogen evaporator (heating block set to 40 ◦ C) and reconstituted in mL toluene for GC–MS analysis TM 2.4.3 Recovery determination using SupelMIP SPE cartridges To determine the recovery rates, 100 mg of distillate aromatic extract, for which the PAH content had previously been determined, were initially dissolved in 100 mL hexane Then mL of this extract was loaded onto the SPE columns and treated as described above (Section 2.4.1) A further mL of hexane extract was evaporated to dryness with a nitrogen evaporator and reconstituted in mL toluene Both solutions were analysed with GC–MS to determine the ratio of the respective peak areas Absolute masses loaded onto the cartridges were 18 ng of benzo[a]anthracene (BaA), 90 ng of chrysene (Chr), 40 ng of benzo[b]fluoranthene (BbF), ng of benzo[k]fluoranthen (BkF), ng of benzo[j]fluoranthene (BjF), 100 ng of benzo[e]pyrene (BeP), 25 ng of benzo[a]pyrene (BaP) and ng of dibenzo[a,h]anthracene (DbahA) Recovery determinations were made in quintuplicate 2.5 Analysis 2.5.1 Preparation of isotope labelled internal and native PAH standard solutions Native and isotope labelled internal PAH standard calibration kits were purchased from Lab Service Analytica Srl (Anzola Dell’Emilia, Italy) The native standard contained benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[j]fluoranthene, benzo[e]pyrene, benzo[a]pyrene, and dibenzo[a,h]anthracene, each at a concentration of 25 g mL−1 in toluene The isotope labelled standard contained benzo[a]anthracene-d12, chrysene-d12, benzo[b]fluoranthene-d12, benzo[k]fluoranthened12, benzo[e]pyrene-d12, benzo[a]pyrene-d12 and dibenzo[a,h]anthracene-d14, each at a concentration of 25 g mL−1 in toluene Spiking solutions of the isotope labelled internal standards were prepared in toluene at a concentration of 2500 ng mL−1 2.5.2 Analytical determination of PAHs with GC–MS The concentrations of each of the eight PAHs were determined through gas chromatography coupled to a mass spectrometer (Agilent 7890 A/5975C, Santa Clara CA, USA) in selected ion mode (SIM) against internal standards Five calibration levels were prepared in the range 0–200 ng mL−1 (5, 10, 50, 100, and 200 ng mL−1 ; internal standard 50 ng mL−1 ) The separation column used was a Rxi-PAH column (Restek, Bellefonte PA, USA, 30 m, 0.25 mm ID, 0.10 m df, product code 49,318) with an injection volume of L in splitless mode (1 valve time) The injector and transfer line temperatures were set at 300 ◦ C and 320 ◦ C respectively, the carrier gas was helium at a constant flow of 1.75 mL min−1 Additional details can be found in the supplementary data (SD1) 2.6 Quality control and quality assurance To test for possible contaminations during work-off (extraction, clean-up and analysis), blank (95 mL toluene) solutions spiked with 20 L internal standard solution (2500 ng mL−1 ) were analysed for each batch of analyses Pure toluene was injected prior to each sample sequence to exclude solvent contamination Solvents were always found free of PAH contamination A reference chromatogram with all native and deuterated standards can be found in the supplementary data (SD2) 2.7 Method performance parameters 2.7.1 Recovery, precision, analytical range, linearity and control sample The recovery was determined for concentrations at the lower and upper end of the calibration curve Elftex® TP carbon black containing polyvinylchloride (PVC), which was previously shown not to contain any PAHs, was used as an interference-free blank matrix Approximately 50 mg of the blank matrix was spiked with 10 L and 80 L of native standard solution (2500 ng mL−1 ) corresponding to 25 ng and 200 ng of each of the PAHs respectively Sample extraction, sample concentration and clean-up were carried out as described earlier The dry extract obtained after the evaporation step (2.4.1) was reconstituted with 50 ng mL−1 internal standard containing mL toluene for GC–MS analysis Recovery was calculated by dividing the analytically determined concentration with the theoretical spiked concentration Five replicates and one non-spiked blank sample were analysed for each spiking level For the determination of the precision, samples were analysed in quintuplicate Native and internal standards were used to calibrate the GC–MS instrument over a concentration range of 0–200 ng mL−1 Each batch of analyses included a control sample, consisting of 95 mL of toluene spiked with 20 L internal standard and 40 L native standard solutions (both at 2500 ng mL−1 ) All subsequent steps (sample concentration and clean-up) were carried out as described above for the sample extracts 2.7.2 Trueness In the absence of a reference material, trueness - defined as the closeness of agreement between a test result and a reference value could not be directly determined However, the content of distilled aromatic extracts (2.7%) and carbon black (24.1%) were known for O Geiss et al / J Chromatogr A 1566 (2018) 13–22 the NR/BR blends and each of these two ingredients were separately available and could be analysed with regard to their PAH content Assuming that carbon black and the distilled aromatic extracts were the only sources of PAHs, the theoretical final mass-concentration for the test-material could be determined and compared against the effectively measured concentrations 2.7.3 Limit of Quantification (LoQ) The limit of quantification was determined following the approach specified in the European Pharmacopoeia [15] The peakto-peak noise around the analytes retention times was measured, and subsequently, the concentration of the respective analytes that would yield a signal-to-noise ratio equal to 10:1 was estimated Results and Discussion 3.1 Extraction 3.1.1 Considerations concerning the selection of an appropriate extraction solvent Selecting a suitable solvent is crucial for the extraction process as its effectiveness depends on the solvent’s polarity and boiling point The non-specific interaction between carbon surfaces with organic molecules is dominated by dispersion forces [16] Due to the large surface area of carbon black, it has a relatively strong adsorption affinity for PAHs, with extractability becoming optimal at high temperatures The two most commonly reported [9–11,14] solvents for the extraction of PAHs in various matrices are toluene (bp 101.6 ◦ C) and hexane/cyclohexane (bp 68.7/80.7 ◦ C) Hexane has a lower polarity index (close to the polarity index of the highly lipophilic PAHs) However, its boiling point is around 40 ◦ C lower than for toluene For this reason, toluene was given preference over hexane as extraction solvent 3.1.2 Randall hot extraction compared to ultrasound extraction Randall hot extraction and ultrasonic extraction were compared with regard to their extraction efficiency for the materials under investigation and the conditions used With the exception of the extraction technique, all other steps were identical (clean-up with SPE and analysis) Mass concentrations achieved when extracting with Randall hot extraction were always higher compared to the concentrations obtained with ultrasound extraction (Fig 1) This difference appears to be independent of the PAH content (Fig 1A and B) and the type of material The contents achieved with ultrasound extraction were between 10–40 % lower, depending on the specific PAH, suggesting that ultrasound extraction under these conditions may be less efficient compared to Randall hot extraction Our extraction procedure deviated from the AfPS protocol [9] in that the AfPS protocol extracts higher weightings of sample material in lower amounts of solvent This deviation was necessary to ensure identical extraction conditions in terms of loading factor for the Randall and ultrasound extractions and should have no effect on extraction efficiency The Randall hot extraction process represents an improvement over the classical Soxhlet extraction technique in that it considerably shortens the extraction time Compared to the classical Soxhlet method where the condensed solvent is at a temperature below the boiling point, the Randall method has the sample material completely immersed in boiling solvent which provides great time savings as analytes are more soluble in boiling solvent Other benefits of the hot extraction process include short process paths, low solvent requirements, and a process that is gentler on the extract (due to the shorter extraction period) A general drawback of ultrasound extraction is linked to the underlying principle behind the effects of ultrasound sonication in liquids: cavitation Cavitation refers to the formation, growth, and collapse 17 of vapour or gas bubbles due to ultrasound [17] Depending on the frequency and intensity of the ultrasound waves, the cavitation bubbles produce very high temperatures and pressures, which can cause degradation and transformation of organic molecules [18] This can result in lower recoveries of the desired analytes and/or increase the amount of undesired compounds extracted from the matrix 3.1.3 Completeness of extraction The Randall hot extraction process can be split into steps: (i)immersion: the thimble is lowered into the boiling solvent, (ii)rinsing/washing: the thimble is raised above the boiling solvent for a period of time until residual extract is removed from the solid material by the condensed solvent, and (iii) recovery: part of the solvent is removed from the extraction cup, concentrating the analytes for further processing Within the margin of error provided by the standard deviations, no difference in mass concentrations (content) was observed between immersion times of h and h (Fig 1D) which indicates that the extraction was already complete after immersing for h Hamm and co-authors [14] investigated the extraction time/cycles necessary to obtain complete extraction of PAHs from carbon black with traditional Soxhlet extraction [19,20] and concluded that 16 h (320 cycles) with toluene were required Randall-extraction is known to shorten extraction times by a factor of to when compared to classical Soxhlet extraction This is equivalent to a total extraction time of 3–4 hours (including all three steps) and confirms the quantitative extraction of all PAHs in our study 3.2 Purification Rubber and plastic materials are complex matrices and during the extraction process a number of undesired substances such as monomers, aliphatic hydrocarbons, and aromatic hydrocarbons (other than PAHs), as well as plasticizer-additives are extracted together with the desired analytes These contaminations lead to higher detection/quantification limits, accelerated dirtying of the mass-spectrometer, and - in the presence of phthalates - may render determination of the correct analyte outright impossible For this reason, sample extracts undergo a clean-up before being injected into the GC MS system The selection of appropriate and efficient clean-up procedures in the case of complex matrices can be challenging [21,22] Some of the existing methods listed in Table either forego this clean-up process entirely [11] or rely on a normal phase silica clean-up using SPE-cartridges [10] or silica-gel packed glass columns [9,12,14], which are non-selective and coextract compounds having similar physicochemical characteristics [23] In this work, we tested a new SPE-phase, based on molecularly imprinted polymer (MIP) technology, on extracts obtained through Randall hot extraction on both the NR/BR blend and the phthalate containing soft-PVC MIPs are highly cross-linked polymer phases that have pre-determined selectivity for a single analyte or a group of structurally related analytes To the best of our knowledge, these SPE cartridges have so far only been used for the extraction and analysis of PAHs in olive oil [24], environmental water samples [25], and from tea leaves [26] Having extracted the same amount (35 mg) of rubber for all three samples (no clean-up, clean-up with MIPs SPE and clean-up with silica-gel packed column) allows for direct comparisons of the areas below the different peaks in the obtained chromatograms (Fig 2) Clearly, the silica-gel packed column did not remove most of the undesired contaminants Baselines from silica gel purified and unpurified extracts show similar total ion currents (in Scan and Selected Ion modes) Silica-gel primarily retains polar compounds which appear to be absent in the sample extract In particular, the aliphatic hydrocarbons eluted in the first part of the chromatogram were not removed from the sam- 18 O Geiss et al / J Chromatogr A 1566 (2018) 13–22 Fig Comparison of extraction efficiencies Comparison of the extraction efficiencies for Randall hot extraction and ultrasound extraction for (A) NR/BR with carbon black N375 and TDEA (100 mg weigh-in), (B) NR/BR with carbon black N375 and DEA (40 mg weigh-in), (C) soft-PVC (100 mg weigh-in) (D) Comparing extraction efficiencies for 2-hour and 4-hour immersion times ple extract The extract purified with the MIP-SPE columns shows a much cleaner chromatogram, indicated by the lower total ion current baseline A dominant peak elutes around 21 deriving from the release of an additive from the frits used in the SPE-columns However, the retention time of this peak did not interfere with the retention times of the analytes under the chromatographic conditions described in this work A clean-up step before using the column could however be added to remove the impurities in case of interference Also in selected ion monitoring mode (lower part of Fig 2), the MIP-SPE baseline is lower compared to the other two Large amounts of phthalates were extracted together with the PAHs, which, even in selected ion monitoring mode, resulted in a relatively large increase in the baseline signal which in turn needed to be removed before injection (Fig 3) Both the MIP-SPE and the silica-gel packed columns successfully removed, almost quantitatively, the more polar phthalates from the sample extract By overlaying a standard chromatogram (panel in Fig 3) onto the sample extract chromatogram (panel in Fig 3), we demonstrate that the eight priority PAHs and their respective deuterated forms not co-elute with impurities released from the frits of the SPE columns attributed to the design and intrinsic functioning of MIPs, which are a class of highly cross linked polymer-based molecular recognition elements engineered to bind one specific target compound or a class of structurally related compounds The MIP material is designed with cavities that are sterically and chemically complementary to the target analytes [27] Compared to most other PAHs investigated in this study, benzo[a]anthracene has a relatively small molecular structure and might therefore be retained less efficiently Also benzo[j]fluoranthene exhibited a relatively low recovery rate which can possibly be attributed both to the relatively low absolute amount loaded onto the SPE cartridge and non-baseline separation from benzo[k]fluoranthene which added uncertainty to the integration procedure (cf., RSD column in Table 2) This explanation is only partly satisfactory as it should equally apply to benzo[k]fluoranthene, which exhibited a higher recovery rate All other recovery rates were found to be above 84% The simple usage-protocol (fast), low amount of required solvents, commercial availability (no need to pack the column), and, above all, high selectivity resulting in lower baselines (i.e., lower detection limits), highlight the superiority of the MIP-based solid phase extraction procedure with regard to traditional, silica-based purification methodologies TM 3.2.1 Recovery experiments of SupelMIP SPE cartridges Depending on the type of compound, PAH recoveries ranged from 51 to 95 % (Table 2) The lowest recoveries were found for benzo[a]anthracene and benzo[j]fluoranthene with recoveries of 59% and 51%, respectively A similarly low recovery rate for benzo[a]anthracene has been reported in an application note for the extraction and analysis of TM PAHs in olive oils using SupelMIP SPE [24] and can likely be 3.3 Method performance parameters 3.3.1 Recovery, precision, analytical range and linearity Recoveries and relative standard deviations for each of the eight PAHs are summarised in Table Values are comparable for both spiking levels and range from 55 to 85% O Geiss et al / J Chromatogr A 1566 (2018) 13–22 19 Fig Comparison clean-up efficiency NR/BR extract Overlaid chromatograms of non-purified NR/BR extract, with MIP-SPE columns cleaned NR/BR extracts and with silica-gel packed columns purified NR/BR extracts in Scan mode (upper) and selected ion monitoring mode (lower) Sample weigh-in around 40 mg Table TM Recovery rates for priority PAHs purified on SupelMIP SPE cartridges Compound Absolute mass loaded on column [ng] Average recovery [%] RSD [%] n = Recovery of spiked olive oila [%] Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[j]fluoranthene Benzo[e]pyrene Benzo[a]pyrene Dibenzo[a,h]anthracene 18 90 40 100 25 59 84 94 89 51 95 93 91 4.3 3.5 2.6 10.8 9.3 1.4 3.5 3.2 65 70 82 84 n/ab n/ab 87 82 a b From Supelco application note 192 [24] Recovery rates not reported in Supelco application note 192 Since the recovery values were determined by spiking blank PVC matrix, values may vary for other plastic or rubber materials Due to its high content of phthalates, the clean-up step for this material can be considered as being particularly challenging Therefore it can reasonably be assumed that the recovery rates will be similar or better for other rubber or plastic materials For the determination of the precision, samples were analysed in quintuplicate and the relative standard deviation was generally below 5%, except for benzo[k]- and benzo[j]fluoranthene, where it reached 10% due to non-baseline peak separation The recovery of the control sample ranged from 79 to 99 % Native and internal standards were used to calibrate the GC–MS instrument, yielding an R-squared >0.99 for all compounds The detector proved to have a linear response over the range 0–200 ng mL−1 and all sample extract concentrations were within this range 3.3.2 Indirect determination of trueness In the absence of a reference material, trueness - defined as the closeness of agreement between a test result and a reference value - could not be directly determined However, the content of treated and untreated distilled aromatic extract (2.7%) and carbon black (24.1%) were known for the NR/BR blends and each of these 20 O Geiss et al / J Chromatogr A 1566 (2018) 13–22 Fig Comparison clean-up efficiency PVC extract Soft-PVC extracts (100 mg sample weigh-in): (1) Standard chromatogram (scan mode) on same retention time scale as overlaid sample extract chromatograms in panel (2); (2) overlaid chromatograms of MIP-SPE column cleaned extract and silica-gel packed column purified extract; (3) non purified sample extract Table Recovery and RSD [%] values for low and high spiking levels Spiking Level Compound 25 ng Recovery ± RSD [%] n = 200 ng Recovery ± RSD [%] n = Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[j]fluoranthene Benzo[e]pyrene Benzo[a]pyrene Dibenzo[a,h]anthracene 67.0 ± 5.0 84.0 ± 5.8 72.4 ± 2.5 84.6 ± 4.4 65.6 ± 5.2 72.2 ± 3.1 72.3 ± 3.8 54.7 ± 7.4 62.2 ± 8.0 75.4 ± 5.7 74.2 ± 3.0 79.5 ± 8.2 62.6 ± 3.8 76.1 ± 3.4 65.2 ± 6.1 62.2 ± 4.3 two ingredients were separately available and could be analysed with regard to their PAH content Assuming that carbon black and the distilled aromatic extracts were the only sources of PAHs, the theoretical final mass-concentration for the test-material could be determined from c(PAH, mgkg−1 ) = c(PAHCarbon Black , mgkg−1 )xContent in rubber [%] + c(PAHDAE/TDAE , mgkg−1 )xContent in rubber [%], (1) where c(PAHCarbonBlack ) is the individual PAH concentration in the carbon black (N375) and c(PAHDAE/TDAE ) the individual PAH concentration in the treated/untreated distilled aromatic extract, respectively These concentrations are multiplied with their respective relative contents Results are generally in good agreement (Table 4) with only benzo[a]pyrene showing a higher discrepancy between the theoretical and the measured values for both NR/BR blends Unfortunately, these results cannot be used for a fully quantitative trueness evaluation of the method as an insufficient number of replicate measurements were done for the mass-concentration determination of carbon black N375, TDAE, DAE and NR/BR (N375/TDAE) Calculation of the uncertainties under these conditions was not possible Moreover, the values provided by the manufacturer for the content of carbon black and distilled aromatic extract can only be considered semi-quantitative estimates 3.3.3 Limit of Quantification (LoQ) The limit of quantification is the lowest concentration of an analyte that can be determined with acceptable precision and accuracy Different guidelines for the determination of the LoQ are available By using the signal-to-noise method, the peak-to-peak noise around the analyte retention time is measured, and then, the concentration of the analyte that yields a signal-to-noise ratio of 10:1 is estimated [15] The absolute and relative limits of quantification are reported in Table If required, the sensitivity could be improved by reducing the volume of toluene used to reconstitute the dry-extract after nitrogen evaporation (Section 2.4.1) Conclusions This work proposes a simple, relatively fast, and cost effective method for the determination of eight priority polycyclic aromatic hydrocarbons in rubber and plastic materials Compared with other published methods/studies, the method proposed in this study has two relevant advantages: In terms of extraction efficiency, the Randall hot extraction process proved to be more efficient compared to ultrasound extraction, and represents an improvement also over the classical Soxhlet extraction technique in that it con- O Geiss et al / J Chromatogr A 1566 (2018) 13–22 21 Table Theoretical and measured mass-concentrations of PAHs in NR/BR blends Mass Concentration [mg kg−1 ] Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[j]fluoranthene Benzo[e]pyrene Benzo[a]pyrene Dibenzo[a,h]anthracene a Carbon black N375 Treated Distillate Aromatic Extract (TDAE) Distillate Aromatic Extract (DAE)a 0.8 1.3 4.9 1.7 2.0 14.7 20.2 1.4 0.16 0.57 0.28 0.1 0.04 1.16 0.21 18.3 88.0 40.1 6.2 8.4 101.3 25.4 2.2 NR/BR, N375, TDAE NR/BR, N375, DAE Theoretical Measured Theoretical Measured 0.2 0.3 1.2 0.4 0.5 3.6 4.9 0.3 0.2 0.3 0.9 0.3 0.3 3.2 3.7 0.4 0.7 2.7 2.3 0.6 0.7 6.3 5.5 0.4 0.5 2.1 2.1 0.5 0.5 6.6 4.7 0.5 Not available on the market Used only for the purpose of this research project Table Absolute and relative limits of quantification Compound Absolute LoQ [ng mL−1 ] Relative LoQ (500 mg of sample material) [mg kg−1 ] Relative LoQ (100 mg of sample material) [mg kg−1 ] Relative LoQ (20 mg of sample material) [mg kg−1 ] Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[f]fluoranthene Benzo[e]pyrene Benzo[a]pyrene Dibenzo[a,h]athrancene 0.8 0.8 3.6 0.9 0.8 0.8 1.1 0.6 0.002 0.002 0.007 0.002 0.002 0.002 0.002 0.001 0.008 0.008 0.036 0.009 0.008 0.008 0.011 0.006 0.04 0.04 0.18 0.05 0.04 0.04 0.06 0.03 siderably shortens the extraction time Other benefits of the hot extraction process include short process paths, low solvent requirements, and a process that is gentler on the extract (due to the shorter extraction period) In terms of sample extract purification efficiency, the simple usage-protocol (fast), low amount of required solvents, commercial availability (no need to pack the column), and, above all, high selectivity resulting in lower baselines (i.e., better detection limits), highlight the superiority of the MIP-based solid phase extraction procedure with regard to traditional, silica-based purification methodologies The introduction of Randall-hot extraction for sample extraction and MIP-based solid phase extraction cartridges for selective clean-up represent a novel and significant improvement compared with previously reported methods in this field The findings of this work may be beneficial to the work conducted by the European Committee for Standardisation (CEN) which has recently (December 2017) been mandated by the European Commission to develop a European standard for the analytical determination of the individual concentrations of the carcinogenic PAHs, listed in entry 50 of Annex XVII to the REACH Regulation, in the plastic and rubber components of articles, in support to the enforcement of the provisions in paragraphs and of the restriction Disclaimers The information and views set out in this study are those of the author(s) and not necessarily reflect the official opinion of the European Commission The European Commission does not guarantee the accuracy of the data included in this study Neither the European Commission nor any person acting on the European Commission’s behalf may be held responsible for the use which may be made of the information contained therein Appendix A Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.chroma.2018 06.047 References [1] H.I Abdel-Shafy, M.S.M Mansour, A review on polycyclic aromatic hydrocarbons: source, environmental impact, effect on human health and remediation, Egypt J Petroleum 25 (2016) 107–123 [2] N.-D Dat, M.B Chang, Review on characteristics of PAHs in atmosphere, anthropogenic sources and control technologies, Sci Total Environ 609 (2017) 682–693 [3] K.O Goyak, M.H Kung, M Chen, K.K Aldous, J.J Freeman, Development of a screening tool to prioritize testing for the carcinogenic hazard of residual aromatic extracts and related petroleum streams, Toxicol Lett 264 (2016) 99–105 [4] P Lassen, L Hoffmann, M Thomasem, PAHs in toys and childcare products Survey of chemical substances in consumer products No 114 2011, in: Danish Ministry of the Environment, Environmental Protection Agency, 2012 [5] BAuA, Annex XV Restriction Report Proposal for a Restriction of PAH, Bundesanstalt Fuer Arbeitschutz Und Arbeitsmedizin (BAuA), 2010 [6] M.K Peterson, J.C Lemay, S Pacheco Shubin, R.L Prueitt, Comprehensive multipathway risk assessment of chemicals associated with recycled (ăcrumb)ă rubber in synthetic turf fields, Environ Res 160 (2018) 256–268 [7] Commission Implementing Decision of 1.12.2017 on a standardisation request to the European Committee for StandardiZation and to the European Committee for Electrotechnical StandardiZation as regards compliance with maximum content criteria of Polycyclic Aromatic Hydrocarbons in rubber and plastic components of articles placed on the market for supply to the general public in support of Regulation (EC) No 1907/2006 of the European Parliament and of the Council, in: M/556 C(2017)7926 final, Brussels, 2017 [8] Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC, in, Official Journal of the European Union L396, 2006 22 O Geiss et al / J Chromatogr A 1566 (2018) 13–22 [9] BAuA, GS-Spezifikation AfPS GS 2014:01 PAK Testing and validation of Polycyclic Aromatic Hydrocarbons (PAH) in the course of GS-Mark Certification Ausschuss fuer Produktsicherheit, Bundesanstalt fuer Arbeitsschutz und Arbeitsmedizin, (2014) [10] L Gruber, M Schlummer, D Fiedler, M Barwitz, S Smilic, R Franz, Untersuchungen zur Migration von Polyzyklischen Aromatischen Kohlenwasserstoffen (PAK) aus altreifenhaltigen Gummiprodukten Report PA/4453/17, in, Fraunhofer-Institut fuer Verfahrenstechnik und Verpackung, Freising/Germany, 2017 [11] International Organization for Standardization, ISO/TS 16190:2013 Footwear – Critical substances potentially present in footwear and footwear components – Test method to quantitatively determine polycyclic aromatic hydrocarbons (PAH) in footwear materials, in, Geneva, 2013 [12] European Committee for Standardization, EN 16143:2013 Petroleum products - Determination of content of Benzo(a)pyrene (BaP) and selected polycyclic aromatic hydrocarbons (PAH) in extender oils - Procedure using double LC cleaning and GC/MS analysis, in: Brussels, 2013 [13] International Organization for Standardization, ISO 21461:2009 Rubber – Determination of the aromaticity of oil in vulcanized rubber compounds, in, Geneva, 2009 [14] S Hamm, T Frey, R Weinand, G Moninot, N Petiniot, Investigations on the extraction and migration behavior of polycyclic aromatic hydrocarbons (PAHS) from cured rubber formulations containing carbon black as reinforcing agent, Rubber Chem Technol 82 (2009) 214–228 [15] A Shrivastava, V Gupta, Methods for the determination of limit of detection and limit of quantitation of the analytical methods, Chronicles Young Scientists (2011) 21–25 [16] A Andreu, H.F Stoeckli, R.H Bradley, Specific and non-specific interactions on non-porous carbon black surfaces, Carbon 45 (2007) 1854–1864 ˛ [17] M Gagol, A Przyjazny, G Boczkaj, Highly effective degradation of selected groups of organic compounds by cavitation based AOPs under basic pH conditions, Ultrason Sonochem 45 (2018) 257–266 [18] P Riesz, D Berdahl, C.L Christman, Free radical generation by ultrasound in aqueous and nonaqueous solutions, Environ Health Perspect 64 (1985) 233–252 [19] N.J Thiex, S Anderson, B Gildemeister, Crude fat, diethyl ether extraction, in feed, cereal grain, and forage (Randall/Soxtec/submersion method): collaborative study, J AOAC Int 86 (2003) 888–898 [20] S Anderson, Soxtec: Its Principles and Applications, AOCS Press, 2004 ´ [21] E Gilgenast, G Boczkaj, A Przyjazny, M Kaminski, Sample preparation procedure for the determination of polycyclic aromatic hydrocarbons in petroleum vacuum residue and bitumen, Anal Bioanal Chem 401 (2011) 1059–1069 [22] M Patrycja, F André, B Grzegorz, Method for the simultaneous determination of monoaromatic and polycyclic aromatic hydrocarbons in industrial effluents using dispersive liquid–liquid microextraction with gas chromatography–mass spectrometry, J Sep Sci (2018) [23] F Ahmadi, H Rezaei, R Tahvilian, Computational-aided design of molecularly imprinted polymer for selective extraction of methadone from plasma and saliva and determination by gas chromatography, J Chromatogr A 1270 (2012) 9–19 [24] Supelco, Extraction & Analysis of PAHs in Olive Oil Using SupelMIPTM SPE – PAHs and GC-MS, Application Note 192 https://www.sigmaaldrich.com/ content/dam/sigma-aldrich/docs/Supelco/Application Notes/1/t309192.pdf [25] M Villar-Navarro, M.J Martín-Valero, R.M Fernández-Torres, M Callejón-Mochón, M.Á Bello-López, Easy, fast and environmental friendly method for the simultaneous extraction of the 16 EPA PAHs using magnetic molecular imprinted polymers (mag-MIPs), J Chromatogr B 1044–1045 (2017) 63–69 [26] L Drabova, J Pulkrabova, K Kalachova, M Tomaniova, V Kocourek, J Hajslova, Rapid determination of polycyclic aromatic hydrocarbons (PAHs) in tea using two-dimensional gas chromatography coupled with time of flight mass spectrometry, Talanta 100 (2012) 207–216 [27] H Gong, S Hajizadeh, L Jiang, H Ma, L Ye, Dynamic assembly of molecularly imprinted polymer nanoparticles, J Colloid Interface Sci 509 (2018) 463–471 ... formulations containing carbon black Not tested for plastic materials Method for determination of PAHs in footwear materials Not specific for rubber and plastic materials Method for determination of. .. for Standardisation [7] as no such harmonised European or international standard is available at present A number of methods for the determination of PAHs in a variety of matrices, that are in most... Recovery determinations were made in quintuplicate 2.5 Analysis 2.5.1 Preparation of isotope labelled internal and native PAH standard solutions Native and isotope labelled internal PAH standard calibration