MSc Chemistry
Science for Energy and Sustainability
Master Thesis
Measuring chemical additives in plastic beach litter from Mumbai,
India
Non-target screening and quantitative analysis of selected organophosphorus
flame retardants
February 2019 - October 2019, 42 ECTS
Hannah van de Kerkhof
UvA 10549153/ VU 2628495
Supervisors
Prof. Dr. Jacob de Boer
Dr. Sicco Brandsma
Dr. Heather A. Leslie
Dr. Mahua Saha (National Institute of Oceanography, Goa, India)Environment & Health Department
Vrije Universiteit Amsterdam
Acknowledgements
After 8 months of research, I have finalised my master’s degree in chemistry at the Vrije Universiteit Amsterdam. I learnt a lot about chemical research while working at the Environment & Health Department. I would like to thank my supervisors Jacob de Boer, Sicco Brandsma and Heather Leslie for their guidance and feedback during my thesis. Special thanks to Mahua Saha for the collaboration on this project and Skype and Whatsapp contact between Amsterdam and Goa throughout my research. I would also like to thank Martin Velzen, Martijn Schaap and Quinn Groenewoud for all the support and help during my lab work.
I would like to express my gratefulness to the Dopper Foundation for the financial support that enabled me to follow a course in Sweden during my project and present my work at the SETAC conference in Canada. Finally, I would like to thank all other interns for nice time at the department and my friends and family for the support!
Abstract
Waste mismanagement in India leads to high amounts of plastic contaminants ending up in rivers and on beaches, through which they enter the ocean. Besides physical harm of plastic debris, chemical additives and pollutants adsorbed to and leaching from plastics are of concern for both human health and the environment. To investigate the types and amounts of toxic chemicals that can potentially leach from plastic debris, plastic beach litter was collected from five beaches in Mumbai, India (2018). All collected items were counted and categorized by product type, creating a dataset of the size and item composition of the plastic waste stream. Polymer types in each item were identified by Fourier-transform infrared spectroscopy (FT-IR). Among the most commonly found items were packaging material (29%), caps (27%) and straws and lollipop sticks (27%), containing mostly polypropylene (PP) and polyethylene (PE). The plastic beach litter and matched new products purchased from local supermarkets were analysed for chemical additives to examine the hypothesis that chemicals have leached from the plastic during its lifetime and stay in the marine environment. A non-target screening with Direct Probe mass spectrometry (DIP-MS) was used for qualitative screening of chemical additives present in the plastics. Phthalates and nonylphenols were most frequently detected in the beach litter and new products. Quantitative analysis of three organophosphorus flame retardants (OPFRs): triphenyl phosphate (TPHP), tris(2-chloroisopropyl)phosphate (TCPP) and tris(2-chloroethyl) phosphate (TCEP) was subsequently performed on selected beach plastic samples and new products by solvent extraction and analysis by gas chromatography-mass spectrometry (GC-MS). The selected OPFRs are toxic to aquatic wildlife and recent literature indicates endocrine disruption, reproductive and developmental toxicity associated with human exposure. TPHP was detected most frequently in the plastic items, followed by TCPP. Only TPHP was measured above limit of quantification at 4-73 ng/g in beach samples and 4-44 ng/g in the new products. No significant difference of TPHP concentration was found between beach plastics and new products, partly due to high uncertainties in sampling and product collection. The concentrations of TPHP found in this study are relatively low are more likely adsorbed from the environment than being present as additive in the products. These levels were below the limit of detection of the DIP-MS screening method, demonstrating that DIP-MS is a valuable screening method for additives present in plastic but is not suitable for detection of trace amounts of chemical pollutants adsorbed from the environment.
List of abbreviations
Ace Acetone
APCI Atmospheric pressure chemical ionization ASE Accelerated solvent extractor
ATR Attenuated total reflectance BPA Bisphenol A
BS Beach sample DCM Dichloromethane DIP Direct Probe
EC European Commission ECHA European Chemical Agency EtAc Ethyl Acetate
EU European Union FR Flame retardant
FT-IR Fourier-transform infrared spectroscopy GC Gas Chromatography
HDPE High density polyethylene LDPE Low density polyethylene LOD Limit of detection LOQ Limit of quantification MeOH Methanol
MS Mass spectrometry NP Nonylphenol
OPFR Organophosphorus flame retardants PAH Polycyclic aromatic hydrocarbons PC Polycarbonate
PCB Polychlorinated biphenyls PE Polyethylene
PFR Phosphorus flame retardant POP Persistent Organic Pollutant PP Polypropylene
PS Polystyrene PVC Polyvinyl chloride
REACH Registration, Evaluation, Authorisation and Restriction of Chemical Substances RSD Relative Standard Deviation
SD Standard deviation SIM Selected ion monitoring
SUP Single use plastic
TCEP Tris(2-chloroethyl) phosphate TCPP Tris(2-chloroisopropyl)phosphate TPHP Triphenyl phosphate
Table of Contents
ACKNOWLEDGEMENTS ... 2
ABSTRACT ... 3
LIST OF ABBREVIATIONS ... 4
1 INTRODUCTION ... 7
2 CHEMICAL ADDITIVES AND TOXICITY ... 9
2.1FLAME RETARDANTS ... 9
2.1.1 Organophosphorus flame retardants ... 9
2.2PHTHALATES ... 10
2.3NONYLPHENOLS ... 10
2.4BISPHENOL A ... 10
2.6LEACHING OF CHEMICALS FROM PLASTIC DEBRIS ... 11
2.6ADSORPTION OF POLLUTANTS ... 12
3 MATERIALS AND METHODS ... 13
3.1COLLECTION OF SAMPLES ... 13
3.2SAMPLE SELECTION FOR ANALYSIS ... 14
3.3ANALYTICAL PROCEDURE ... 15
3.3.1 Chemicals ... 16
3.3.2 Identification polymers ... 16
3.3.3 Screening with DPI-MS ... 16
3.3.4 Solvent extraction ... 18
3.3.5 Clean-up by SPE ... 18
3.3.6 GC-MS analysis ... 19
3.3.7 Method validation ... 20
3.3.8 Contamination and background measurement ... 21
4 RESULTS ... 22
4.1IDENTIFICATION PRODUCT TYPES ... 22
4.2IDENTIFICATION POLYMERS ... 23
4.2DIRECT PROBE MASS SPECTROMETRY ... 25
4.3EXTRACTION AND GC-MEASUREMENT SELECTED OPFR’S ... 26
5 DISCUSSION ... 29
5.1METHOD DEVELOPMENT ... 29
5.2PRODUCT TYPE ... 29
5.3POLYMER TYPE ... 29
5.4DIP-MS RESULTS ... 30
5.5EXTRACTION AND GC-MS RESULTS ... 32
6 CONCLUSION ... 35
6.1OUTCOME RESEARCH ... 35
6.2RECOMMENDATIONS FOR FUTURE RESEARCH ... 35
7 REFERENCES ... 36
9 APPENDIX ... 43
I PICTURES AND POLYMER TYPES ANALYSED BEACH SAMPLES ... 43
II PICTURES AND POLYMER TYPES NEW PRODUCTS ... 51
III MATCHING BEACH SAMPLES AND NEW PRODUCTS ANALYSED WITH GC-MS ... 54
IV LIST OF COMPOUNDS USED IN DIP-MS SCREENING WITH MASS OF IONIZED MOLECULES ... 56
VCALIBRATION CURVES TPHP,TCPP AND TCEP USED FOR GC-MS DATA ANALYSIS ... 58
1 Introduction
Plastics are synthetic organic polymers that are used for a large variety of products and are found everywhere in our daily lives. This human-made substance holds great features for usage since it is durable, cheap, light weight and easy mouldable. However, the increase in usage has also led to an increase of plastic pollution, which is now recognised as one of the world’s major environmental concerns1. Waste mismanagement leads to plastic waste ending in the environment and the ocean2. If
the current trend will continue, the amount of plastic in the ocean will double from 2010 to 20253.
Currently, plastic accounts for 60-95% of marine debris4. The limited options for removal, together with
the persistence and hazards of the material makes plastic pollution in marine environment of high environmental concern. The plastic debris can cause physical stress to marine wildlife by entanglement or ingestion1. Microplastics have gained focus in research, since these are increasingly found in nature
and easily consumed by (small) organisms5. Microplastics, commonly defined as plastic fragments
smaller between 0.1 µm and 5 mm, are used as raw material in manufacturing of plastic products but are also added to personal care products as exfoliating agents (primary microplastics)6–8. However, the
main source of microplastics are secondary microplastics, formed by fragmentation and weathering of larger plastic pieces9. Even though health effects of microplastic uptake for humans are yet unknown,
several studies show negative effects on aquatic organisms10,11.
Next to physical harm, research is recently focussing on toxic and health effects concerned with chemical additives that can potentially leach from plastic products12. Chemicals such as plasticizers,
flame retardants, modifiers and colouring agents are often added during manufacturing to improve the quality and performance of the plastic material. Since these compounds are not chemically bound to the polymer matrix, they can leach out of the plastics during usage and after disposal. This is of high concern since many of these additives have been listed as hazardous13. For example bisphenol-A,
nonylphenol and triphenyl phosphate have shown endocrine disruptive effects to organisms exposed to these additives13,14. In addition to human and environmental health effects, these compounds
complicate recycling of plastic waste2,15. Next to the release of additives, marine plastic debris has
shown to be a sink for persistent organic pollutants (POP’s) present in the environment4,12. By
adsorption and release of hydrophobic organic contaminants, such as polychlorinated biphenyls (PCBs) and polycyclic aromatic hydrocarbons (PAHs)16, plastic debris forms a transfer mechanism of POP’s
from the littoral to marine environments affecting aquatic wildlife17–19. It is important to monitor the
presence of these chemicals in plastic products and marine plastic debris, in order to investigate possible health effects concerned with usage of plastic products and presence in the environment20.
India has a long coastline and is dealing with large quantities of plastic pollution along this coast21. As
the second largest populated country, India is ranked 12th on the list of countries mostly contributing
to plastic pollution worldwide3. The rapid growth of the Indian economy lead to a rapid increase in the
plastic consumption and waste generation22, accumulating on beaches due to poor waste
management21. The amount of non-recycled plastic waste in India is estimated to reach 13 million
tonnes in 203022. Big quantities of plastic along the coastline do not only scare tourists but also harm
the (marine) environment23. Few studies have been focusing on the presence and distribution of plastic
debris along the coast of India 8,21, but there is still a lack of data about the composition and
environmental effects of this plastic pollution21. Next to addressing toxic effects of this plastic pollution,
the identification of the plastic debris is important to investigate the source of the plastic. Tackling the problem of plastic pollution at the source is important since beach clean-ups seem effective, but disrupt the beach ecosystems and do not remove small particles (<20 mm)21,24. With almost 20 million
citizens25, Mumbai is one of the biggest cities in the world. It deals with plastic waste from both the
dense population as well as from industry located in the metropolitan city26.
Measuring additives in plastic can be challenging since additives can be present in low concentrations and the mixture of multiple chemicals and polymer matrix can lead to interferences27. The lack of data
about the identity and amount of chemicals used as additives in plastic products, demands a broad and fast analytical screening method28. In this study, the novel Direct Probe (DIP) - atmospheric pressure
chemical ionization (APCI)28,29 mass spectrometric (MS) technique will be used to conduct a qualitative
screening of presence of toxic additives in plastic samples. DIP-APCI/MS provides a quick and broad screening without sample preparation29. Screening results will be used select target chemicals for a
quantitative gas-chromatography (GC)-based analysis. The quantitative analysis will provide the possibility to compare concentrations of beach samples and their new products to examine leaching effects. The aim of this research is to (ii) identify beach litter from three beaches, (ii) screen the beach litter for presence of a wide variety of chemical additives and (iii) check hypothesis of leaching by comparing concentrations of selected additives in beach plastics and their new products. In addition, methods for qualitative screening and quantitative analysis will be optimised and evaluated.
Identification
Polymer type (FTIR)
Screening
Qualitative (DIP-MS)Target analysis
Quantitative (GC-MS)
Selection target compound
Information matrix for method development
2 Chemical additives and toxicity
Multiple types of chemicals are mixed with polymers during manufacturing of plastic products to modify their properties13. Depending on the polymer type and product specifications, a mixture of
additives can be used. The next section describes the chemicals, with associated hazards, commonly found in plastic products13,30.
2.1 Flame retardants
Flame retardants (FRs) are a group of chemicals that are used to reduce the flammability of products. Since these chemicals are designed to be stable to ensure fire persistence of the product during its lifetime, these compounds are persistent and bio-accumulate in the environment13. Brominated FRs
are a group of FRs that show to be toxic and are of concern for human health13. The usage of a number
of brominated FRs is therefore restricted in the European Union13. As an alternative, phosphorus
compounds and other new FR are increasingly used28.
2.1.1 Organophosphorus flame retardants
Organophosphorus compounds are widely used as flame retardants, plasticizers and anti-foaming additives in production of plastics, textiles and electronics31. Organophosphorus flame retardants
(OPFR’s) are widely found in our environment and presence has been detected in (indoor)air, dust, water and soil31,32. Halogenated OPFRs, are mainly used as FR whereas non-halogenated OPFR’s as
triphenyl phosphate (TPHP), are also used as plasticizers31,32. In this study, quantitative analysis was
executed on TPHP, chloroisopropyl)phosphate (TCPP, also known as TCIPP) and tris(2-chloroethyl) phosphate (TCEP). TPHP is frequently detected in commercial plastic products29. Recent
studies provide evidence for (neuro) toxic effects of TPHP on both humans and aquatic organisms32–34,
raising concern about this compound. TPHP is also classified as endocrine disrupting compound by the United Nations Environment Program35 and included in the list of most hazardous substances
associated with plastic packaging, developed by Groh et al. (2019), based on harmonized hazard classifications36. The halogenated OPFRs TCEP and TCPP are indicated as reproductive toxins as well as
toxic for aquatic organisms32. TCEP is indicated as neurotoxic and carcinogenic 32,37 and TCPP is
Figure 2: Chemical structures of selected OPFR's for target analysis.
2.2 Phthalates
Phthalates are often used as plasticisers in a wide variety of plastic products20.Plasticizers are added to
the polymers to make the material flexible, soft and mouldable. Plasticizers can make up to 50 % of the total weight of the final product20. Phthalates can be hormone disruptive, even at low concentrations,
and are suspected to be carcinogenic 13,20. Usage of several phthalates in children’s toys and other
products are restricted in the European Union (EU)40. Among the different hormone disruptive
chemicals such as pesticides and bisphenol A, phthalates are most frequently found in the environment and domestic waste water 20.
2.3 Nonylphenols
Nonylphenols (NPs) are degradation products of nonylphenol ethoxylates, used as surfactant and antioxidant in the production of many products such as paints and pesticides13,41. NPs are also used as
plasticizer and antioxidant in the production of plastics13,20. Leaching of NP from plastic packaging has
been demonstrated to contribute to human exposure to NP, together with intake via contaminated food and inhalation41. NPs are considered endocrine disruptive and usage is prohibited in the EU 13,20.
2.4 Bisphenol A
Bisphenol A (BPA) is used as monomer in the production of polycarbonate (PC) plastics. In other polymer types, such as polypropylene (PP) and polyethylene (PE), BPA can be used as a plasticizer or antioxidant13,20. BPA has been identified as endocrine disruptive and can influence reproduction13,41.
Leaching from plastic food and drink packaging has been identified as a cause of human exposure to BPA, similar to intake of food from cans with epoxy coatings13,41. BPA is listed as substance of very
high concern in the Registration, Evaluation, Authorization and Restriction of Chemicals (REACH) of
the European Union. Usage of BPA in materials that come in contact with food and children’s toys are restricted in the EU.
Figure 3: Chemical structure Bisphenol-A
2.6 Leaching of chemicals from plastic debris
Chemicals added during the production of the plastic are often not chemically bound to the plastic and can leach out during all stages of the life cycle of the product such as production, usage and littering13,30.
Depending on the medium of contact, emissions of additives can end up in air, water, soil or organisms. The leaching mechanism is difficult to predict and depends on many factors including chemical properties of the substance, physical stress on the plastic, morphology of the polymer, temperature and particle surface 42.
Affinity and solubility of the chemical additives is different for solid, aqueous and lipid phases, depending on the chemical properties of the compound. The partition between these phases is described by the partition coefficient, representing the equilibrium ratio of the concentrations of the substance in the different phases. Equation 1 show the formula of the often-reported octanol-water partition coefficient (Kow), in which Cwater is the molar concentration of the substance in the water phase
and Coctano the molar concentration of the substance in the octanol phase, when the system is at
equilibrium. The Kow represents the hydrophilicity of a substance and a few are listed in Table 1.
𝐾"# =[&-.)(/-0[&'()*+]] (1)
Table 1: Log Kow: n-octanol/water partition coefficient and solubility in water of several chemical additives. Data
compiled from aWei et al. (2015)31, bLatorre et al. (2003)43 .
Compound Log KOW Solubility in water at 25◦C [mg/L]
TPHP 4.59a 1.9a
p-Nonylphenol 4.5b 5b
Next to migration of chemicals at the interface between the plastic and surrounding medium, chemicals can also slowly migrate within the plastic to the surface30,44. Degradation of the plastic into smaller
particles will increase the particle surface and therefore increase the phase interface and lead to higher migration ratios of the substance to the surrounding medium30. The interface between the polymer
and surrounding medium is however dependant on the temperature, morphology of the polymer and properties of the compound and surrounding medium44.
2.6 Adsorption of pollutants
Plastics have shown to be a sink for chemical pollutants present in the environment45,46. Chemicals such
as biocides, surfactants and polycyclic aromatic hydrocarbons (PAHs) are found be adsorbed by marine plastics47, transporting pollutants to new locations and to aquatic organisms through ingestion46.
Sorption affinity is dependent on the chemical structure of the pollutant and polymer type. Due to the larger surface area, PE shows to have a larger affinity, for a wider range of compounds, compared to PP and polyvinyl chloride (PVC)46. For this reason, PE is also used as passive sampler for measurement
3 Materials and methods
3.1 Collection of samples
Collection and counting of product types and samples were executed by Dr. Mahua Saha, Chemical Oceanography Division, National Institute of Oceanography, Dona Paola, Panjim, Goa 403004, India. In April 2018 plastic debris was collected from five locations in Mumbai, India (Figure 4). Main activity and pollution source of each beach are listed in Table 2. At each location, all macroplastics (>2 mm) were collected from three 10x10 m quadrants with 40-50 m gaps in between. The sampling quadrants were located above the high tide zone and close to vegetation on the beach. The collected macro plastic debris (>5 mm) was collected and counted per product category (caps, spoons, straws, etc.). Unidentifiable samples were counted as ‘fragments’. In order to estimate to what extent chemicals have leached out of the plastic debris, new plastic products were purchased at local supermarkets for analysis. It was tried to buy plastic products similar to the selected debris samples, purchased in Mumbai and Goa, India. Samples were stored at 4◦C until analysis.
Table 2: Beaches of sample collection in Mumbai, India with main activity
Beach location Activity/close to
Dadar Beach Sewage
Girgaon Beach Fishing
Juhu Beach Recreation, tourist
Mahim Beach Sewage and recreation
14
Figure 4: Map of India with detailed map of Mumbai and sampling beaches (retrieved from Google maps)
3.2 Sample selection for analysis
For analysis, samples were selected from three beaches from Mumbai: Juhu, Mahim and Versova. These beaches were chosen because of diverse activities, their geographical location and amount of pollution. Two beach samples (BS) (if available) from each product category were selected from two quadrants per beach. A total of 68 samples were selected for screening with direct probe. Pictures of the selection of samples with corresponding names and category can be found in appendix i, a few are shown in Figure 5. In total, 27 new products purchased in Goa and Mumbai were used for analysis (screening). Pictures of the new products with corresponding names and category can be found in appendix ii. Resembling beach and new products were selected for solvent extraction and qualitative analysis of concentrations of selected OPFRs. Matching products are listed in appendix iii, a few are shown in Figure 6.
19/09/2019, 12(31 Google Maps
Kaartgegevens ©2019 Google,ORION-ME,Mapa GISrael 500 km
17/09/2019, 10)45 Google Maps Page 1 of 2 https://www.google.com/maps/place/Mahim+Beach+ ! ,+Mahim,+India/@19.0368902,72.830108,16z/data=!4m5!3m4!1s0x3be702a812c29f55)0x89b4b51e92b6ce16!8m2!3d19.6460556!4d72.7129822 Versova Beach Juhu Beach Mahim Beach Dadar Beach Girgaon Beach
Figure 5: Pictures of a few selected beach samples
Figure 6: Pictures of a few beach samples and resembling new products
3.3 Analytical procedure
First, polymer types of all samples were identified with use of Fourier-Transform infrared spectroscopy (FT-IR). After, all beach samples were screened with use of DIP-APCI time of flight mass spectrometry (TOF)-MS). This method provides a fast screening over a broad range of chemical additives and was used to determine target compounds for further analysis28,29. The selected OPFR target chemicals were
analysed with GC-MS after solvent extraction and silica-based clean-up. The method for clean-up was optimised for removal of polymers and impurities, to improve GC-MS measurements.
BS 52, packaging BS 61, spoon BS 40, fragment
BS 23, straw
BS 55, Lollipop stick BS 23, cap
BS. 48, cup NP 18, cap BS 28, hair comb NP 17, cap NP 16, packaging BS 37, cap BS 24, cap BS 35, packaging NP 20, cap
3.3.1 Chemicals
Table 3 lists all chemicals with suppliers from which they were purchased. A stock solution of 1 mg/L PFR mix in MeOH was diluted with toluene and used for method validation and calibration. Internal standards for the extraction procedure were prepared with deuterated compounds. Solutions of TPHP-d15 (645 ng/mL in iso-octane) and TCEP-d12 (684 ng/mL in toluene) were stored at 4◦ C until
usage. Silica with 3% H2O was prepared by drying silica overnight at 140◦C and adding 3% H2O w:w (in
small portions, while shaking) after cooling.
Table 3: List of chemicals used
Name Supplier
2-Propanol VWR Chemicals
Acetone Biosolve
Acetonitrile VWR Chemicals
APCI Low Concentration Tuning Mix
Agilent
Dichloromethane Biosolve
Ethyl acetate Biosolve
n-Hexane Biosolve
Isooctane Merck
Toluene VWR Chemicals
3.3.2 Identification polymers
The polymer type composition of the samples was determined with use of FT-IR. The FT-IR measurement was carried out using a Bruker, LUMOS microscope in attenuated total reflectance (ATR) mode with 128 scans per measurement. Absorption spectra were compared with spectra from the Opus Bruker library and literature8,20 to identify polymer types.
3.3.3 Screening with DPI-MS
Based on the procedure proposed by Ballesteros-Gómez et al. (2013), a micro TOF II MS (Bruker Daltonics, Bremen, Germany; mass accuracy <2 ppm and resolution >16,500 FWHM) equipped with an APCI source and DIP assembly was used to conduct a fast screening of additives in selected plastic samples 28,29. The detection limit of the direct probe is approx. >100 µg/g g or 0.025% (w/w),
determined by earlier experiments executed by our group29. No sample preparation was required for
glass probe. The glass probe was placed in the DIP assembly and 5 μL of calibration mix (APCI low concentration tuning mix) was added on the tip of the glass probe. After introducing the probe in the TOF-MS, samples were measured for 3 minutes over a temperature range from 220 - 260 ◦C. All parameters of the DIP-MS method are shown in Table 4. In between all measurements, blanks with unloaded glass probes were executed to check for background contamination.
Table 4: DIP-APCI-TOF parameters
TOF parameter Positive mode Negative mode
Capillary 4000 V 1000 V
Corona 5000 nA 7000 nA
End plate offset -500 V -1000 V
Nebulizer 3 Bar 3 Bar
Dry Gas 2.5 L/min 3 L/min
Vaporizer temperature 220◦C (0-1 min), 240◦C (1-2 min),
260◦C (2-3 min) 220◦C (0-1 min), 240◦C (1-2 min), 260◦C (2-3 min) Capillary exit 90 V -120 V Skimmer 1 30 V -40 V Hexapole 1 23 V -23 V Hexapole RF 100 Vpp 100 Vpp Skimmer 2 23 V -23 V
Data analysis 4.1 from Bruker was used for internal calibration with use of the signal of the calibration mix. The enhanced quadratic mode was used for internal calibration and mass errors <2 ppm was considered acceptable. Samples were screened for the presence of FRs, plasticizers, NPs and other chemical pollutants, listed in appendix iv with corresponding ion masses. Data processing was done with use of the program Task, by comparing measured ion masses with values from earlier studies on toxic additives with DIP-MS29,49. A mass error of +/- 0.004 was considered acceptable. The list of possible
hazardous additives with corresponding ion masses used can be found in appendix iv.
Since many of the selected chemicals are present in air and dust30, minimal peak intensities and
signal-to-noise ratios were used for data analysis. Table 5 shows both criteria used for data validation. The “extensive” criteria ensure inclusion of all suspected peaks, whereas the “selective” filter the results with higher certainty (further elaborated in chapter 5.4). To minimise contamination, the shield and plug of the APCI source were cleaned daily by placing these in an ultrasonic bath in 100 ml
miliQ/isopropanol (1:1) for 5 minutes (2x). If needed, the ionization chamber and heater were cleaned with MeOH and the vaporization temperature of the sources was increased to 450 ◦ C for 3 minutes in between measurements to remove contaminants or polymer matrix.
Table 5: Criteria used for data analysis of DIP-MS measurements
Data analysis Minimum peak intensity Minimal signal-to-noise ratio
Extensive 100 3 times individual blank (10 times
for phthalates)
Selective 400 3 times highest blank (10 times for
phthalates)
3.3.4 Solvent extraction
Extraction procedure was based on solid-liquid extractions of polymers used by earlier studies 20,50. All
plastic samples were cut into small pieces of ∼ 3mm (Figure 7) with use of scissors and/or a scalpel. 0.3 g of the cut plastic sample (if available) was weighted in a glass tube. After addition of 4-5 mL DCM and 50μL internal standard, the suspension was sonicated for 1 hr and left in the fridge overnight. The solution was transferred to a clean glass vial, to remove the solid plastic particles. The residue was rinsed with 3 x 2 mL DCM and transferred to the glass vial after shaking with use of a vortex. The solutions were evaporated to ∼1 mL, with a stream of N2 gas. After addition of 4 mL hexane, the
samples were mixed with vortex.
Figure 7: Picture of cut beach samples
3.3.5 Clean-up by SPE
In order to remove polymers and other potential interfering compounds from the extracts, a clean-up was performed using a silica based solid phase extraction (SPE) method based on a publication of Ma
and co-workers (2013)51. Four grams of 3% H2O deactivated silica gel was transferred to a glass column
(10x300 mm i.d.) followed by 1 g of Na2SO4. The dry column was conditioned with 25 mL hexane,
followed by the addition of the sample. The column was eluted with 20 mL hexane (fraction 1), followed by 40 mL DCM/hexane (3:7 v/v) (fraction 2) and 40 mL EtAc (fraction 3). Fraction 1 and 2 were collected together in a clean vial and discharged after finishing the analysis. Fraction 3, containing the OPFR’s, was collected in a clean ASE vial and evaporated to 1 mL under a gentle stream of N2, heating the
samples at 50◦ C. 4 mL toluene was used to rinse the sides of the ASE vials and the samples were evaporated to 0.5 mL with N2. Samples were stored in the fridge for at least 12 hours, to settle down s
polymers left in the solution. If necessary, 100 μL of the top solution was transferred to a vial with insert to remove solid particles and minimise the insertion of matrix in the GC-MS.
3.3.6 GC-MS analysis
Analysis was done using an Agilent 6890 Series GC system coupled to an Agilent Technologies 5973 Mass Selective Detector. A DB-5 MS column (30 m x 0.25 mm, 0.25 mm film thickness) was installed. A 1 μL sample of the extracts was automatically injected in the GC-MS with a total run time of 23.33 min per sample. The injection temperature was set at 250◦C. The oven temperature was set at 110◦C for the first 4 minutes, increasing to 190◦C with a rate of 15◦C per minute and finally increased to 310◦C with a rate of 10◦C per minute and held for 4 minutes. Helium gas was used as carrier gas, with a constant flow of 1 mL/min. Selected ion monitoring (SIM) mode was used for analysis with the MS. Data processing was done with Aglient ChemStation. Peak areas were calculated with manual integration. Monitored ion mass and qualifier mass from isotope patterns (obtained from sisweb, Scientific Instrument Services isotope calculator) used for qualification are shown in Table 6.
Table 6: GC-MS monitoring ions of selected OPFR's
Compound Quantifier Qualifier Retention time
[min] Triphenyl phosphate d15 (TPHP-d15)* 341 339, 239 14.54 Tris(2-chloroethyl)phosphate -d12 (TCEP-d12)* 261 263 9.04 Tris(2-chloroethyl)phosphate (TCEP) 249 251 9.14 Triphenyl phosphate (TPHP) 326 325, 215 14.61 Tris(2-chloroisopropyl)phosphate (TCPP) 277 279, 201 9.42 *Internal standards
3.3.7 Method validation
In order to validate the developed extraction and clean-up method, recovery, precision, blanks and detection limits were calculated. Recovery of the clean-up were measured with use of spiking with stock solutions of target OPFR’s and internal standards. Recoveries are shown in Table 7.
Table 7: Measured recoveries of silica-based clean-up method. Measured with use of spiking of stock solutions of selected OPFR's and internal standards.
Compound Recovery clean-up
TPHP 78-79%
TCPP 102-116%
TCEP 92-104%
Precision (reproducibility) was validated by calculating the relative standard deviation (RSD) of extraction and clean-up of sample 31 in triplo. The formula for the RSD is shown in equation 2, in which s is the standard deviation of the values and 𝑥̅ the mean of the values
.
Figure 8 shows the results of the replica measurement, RSD of 2,8%. This value was determined to be an acceptable variation. Validation of the extraction procedure was measured by repeating the extraction, no compounds were measured in the second extraction.𝑅𝑆𝐷 =
s
𝑥̅
× 100%.
Figure 8: Levels of TPHP measured in three replicate samples of beach sample 31. Error bars indicate standard deviation of the mean. Results are used to validate reproducibility of the extraction method and calculation of measurement error. 0 10 20 30 40 50 60 70 80 BS31 -1 BS31 -2 BS31 -3 TP hP m ea su re d [n g/ g]
Calibration was performed by plotting analyte concentrations versus peak areas and calculating the gradient of this line, shown in appendix v. Before calculating results, all signals were corrected for blank values and detection limits, as shown in Table 8. Mean blank values were calculated based 10 blank measurements from different days of lab work.
For TPHP and TCEP the limit of detection (LOD) and limit of quantification (LOQ) were determined by signal-to-noise ratios of 3 and 10, respectively. For TCPP however, the LOD and LOQ were calculated as 3 and 10 times the standard deviation of the blanks, since the blank signals for TCPP were relatively high and varied more over time compared to those of TPHP and TCEP.
Table 8: LOD and LOQ values used for data processing of GC-MS results
Compound LOD [ng] LOQ [ng] TCPP 14,55 48,51 TPHP 1,11 3,70 TCEP 0,98 3,25
3.3.8 Contamination and background measurement
Since OPFR’s are present in dust and air 32, special attention should be paid
to avoiding contamination. All glassware was cleaned with acetone and EtOH prior to use. After cleaning and during the experiments, all samples and glassware was covered with aluminium foil (Figure 9). Contact between laboratory gloves and plastic samples has been avoided, samples were handled with cleaned metal tweezers. In addition, all materials such as tweezers, scissors, scalpel and grater have been cleaned with acetone prior to use and in between samples.
Since relatively high blank signals of TCPP were found in the GC-MS measurements, solvents used in the extraction and clean-up procedure were measured for presence of contaminants, but no target chemicals were
found in the solvents. Figure 9: Picture showing aluminum foil wrapped around
glass column and vial to minimize contamination during clean-up
4 Results
4.1 Identification product types
Figure 10 shows the distribution of product types identified of all collected samples from beaches of Mumbai (Dadar beach, Girgaon, Juhu beach, Mahin Beach and Versova Beach). In total 1564 samples were collected, of which 45% remained unidentified and were labelled as fragment. The unidentified samples are not included in the distribution calculations shown in Figure 10. In Figure 11 the
distribution of only the single-use products are shown, together representing 84% of the plastic items identified in this study.
Figure 10: Distribution of product types identified from all samples collected at beaches of Mumbai. Data collected by Dr. Saha1
1 Chemical Oceanography Division, National Institute of Oceanography, Dona Paola, Panjim,
Packaging material 29%
Caps 27% Straws & lollipop
sticks 17% Spoons 10% Fibres 6% Pens 4% Film 3%
Beads & pellets
Figure 11: Distribution of single-use plastic items identified from all samples collected at beaches of Mumbai.
4.2 Identification polymers
Table 9Table 10 show the count of polymer types identified per product category for both the selection of beach samples and the newly bought samples. The distribution of polymer types identified in the beach samples is shown in Figure 12. In these results, copolymers are not listed; in case of a copolymer or polymer mix, the polymer type with the highest resemblance in the absorption spectrum was listed. More detailed information of the co-polymers and polymer mixed identified are listed in the table with sample pictures, found in appendix i and ii. Infrared absorption spectra of three samples are show in Figure 13, representing absorption characteristics of the different polymer types identified.
Table 9: Polymer types identified per product category of beach plastic
Category PE PP LDPE PS PVC Polyester
film Total samples Cap 10 3 13 Cotton bud 1 1 Cup 1 1 Fibre 2 2 4 Film 1 1 2 Foam 1 1 Fragment 8 1 9 Hair comb 1 1 Lollipop stick 2 1 3 Packaging 2 3 1 1 7 Pen 4 4 Spoon 10 1 11 Straw 10 10 Syringe 1 1
Straws & lollipop sticks 20% Packaging 35% Caps 32% Cups 1% Spoons 12%
Table 10: Polymer types identified per product type of newly bought samples
Category PE PP LDPE PS Total
samples Cap 3 2 5 Cotton bud 1 1 Cup 1 1 Fibre 1 1 Fork 1 1 Hair comb 1 1 Packaging 3 1 4 Pen 1 3 4 Spoon 3 2 5 Straw 1 1 1 3 Syringe 1 1
Figure 12: Distribution of polymer types identified in beach samples
PP 53% PE 41% PS 3% PVC 2% Polyester film1%
Figure 13: FTIR spectra of three beach samples. From top to bottom: BS 6, BS 32, BS 22.
4.2 Direct Probe Mass Spectrometry
Table 11 summarises the results of the screening with DIP-MS. Only results that met the selective criteria (Table 5) are mentioned. Chemical compounds that were not found are excluded from this table as well. The tables of results including samples meeting the criteria for the extensive data analysis are shown in appendix vi and can be used for further analysis.
Abs PP (Lollipop stick, BS6) Abs LDPE (Packaging, BS32) Abs PS (Spoon, BS22)
Table 11: Results of DIP-MS screening, only includes results meeting the "selective" criteria.
Compound Beach sample no. New product no.
Biocides: Antimicrobials
3-Iodo-2-propynyl N-butylcarbamate 15 -
5-Chloro-2-(2,4 -dichlorophenoxy)phenol,
Triclosan 31, 37 -
Isocyanates 4,4ʹ-Methylenebis(phenyl isocyanate) (MDI), 2 4'-Methylenebis(phenyl isocyanate), Diphenylmethane-2,2'-diisocyanate
10, 26, 52 3, 7
Phthalates Bis(2-ethylhexyl) phthalate (DEHP),
Di-n-octyl phthalate (DNOP) 1, 3, 5, 7, 25, 28, 31, 34, 37, 39, 40, 42, 64 -
Dibutyl phthalate (DBP) 3, 31, 39, 42, 45, 58, 62 5
Diisoheptyl phthalate (DIHP) 28, 54
Diisononyl phthalate (DINP) 5 -
Diisodecyl phthalate (DIDP) 45 -
Dihexyl phtalate 53, 54, 55, 59, 62, 65 -
Flame retardants
TPhP 10 -
Nonylphenol and Nonylphenolethoxylates
4-nonylphenol, branched 2, 4, 5, 10, 11, 12, 21,26, 34,
39, 61 16
Nonylphenol, ethoxylated 31, 61 12
PAHs (EU REACH 8 PAHs) Benzo[a]pyrene (BaP), Benzo[e]pyrene (BeP), Benzo[b]fluoranthene (BbFA), Benzo[j]fluoranthene (BjFA), Benzo[k]fluoranthene (BkFA)
4, 23, 52 -
Benzo[a]anthracene (BaA) , Chrysen (CHR) 48, 52, 64 5, 7
Adhesives (surfactant)
Bisphenol A (BPA) 23, 65 -
4.3 Extraction and GC-measurement selected OPFR’s
Concentrations measured with GC-MS of selected OPFR’s in beach samples and new products are listed in
Table 12 Table 13 respectively. All results are normalised to ng/g, after correction for LOD and blanks. Levels above LOQ were only found for TPHP, comparison of concentrations of TPHP in beach samples (BS) and their new products are shown in Figure 14. Figure 15 shows the variation of TPHP levels between the different sampling locations. Since the extraction method wasn’t suitable for polystyrene samples, is was not possible to measure new product sample 1 and 6.
Table 12: Selected OPFR's in beach samples; concentrations measured with GC-MS
Sample TCPP TPHP TCEP
Beach sample no. (polymer type) Concentration [ng/g] Concentration [ng/g] Concentration [ng/g] LOD 49 ng/g 4 ng/g 3 ng/g
12 cap (PP) < LOD < LOD < LOD
24 cap (PE) < LOD < LOD < LOD
30 cap (PE) < LOD < LOD < LOD
37 cap (PE) 79 a 14 < LOD
48 cup (PP) < LOD < LOD < LOD
28 hair comb (PP) < LOD 17 < LOD
4 packaging (PP) < LOD 54 < LOD
32 packaging (LDPE) < LOD < LOD < LOD
35 packaging (PE) 64 a < LOD < LOD
58 packaging (PE) < LOD < LOD < LOD
65 pen (PP) < LOD < LOD < LOD
31 spoon (PP) 63 a 73 < LOD
38 spoon (PP) < LOD 48 7 a
44 spoon (PP) < LOD 7 a < LOD
51 spoon (PP) < LOD < LOD < LOD
1 straw (PP) < LOD 38 < LOD
23 straw (PP) 83a 8 a < LOD
42 straw (PP) < LOD 6 a < LOD
67 straw (PP) < LOD 31 < LOD
47 syringe (PP) 69 a 24 < LOD
aBelow LOQ, concentration uncertain
Table 13:Selected OPFR's in new products; concentrations measured with GC-MS
Sample TCPP TPHP TCEP
New product no. (polymer type) Concentration [ng/g] Concentration [ng/g] Quantification [ng/g] LOD 49 ng/g 4 ng/g 3 ng/g
8 cap (PE) < LOD < LOD < LOD
18 cap (PE) < LOD 14 < LOD
20 cap (PE) < LOD 12 a < LOD
17 hair comb (PP) < LOD 44 < LOD
3 packaging (PE) 84 a 32 < LOD
16 packaging (PE) 69 a < LOD < LOD
25 packaging (PE) 69 a < LOD < LOD
26 pen, inside (PE) < LOD 16 < LOD
6 spoon (PP) < LOD 7 a < LOD
24 spoon (PP) 55 a 20 < LOD
5 straw (LDPE) < LOD 5 a < LOD
13 straw (PE) 89 a < LOD < LOD
23 syringe (PP) < LOD 10 a < LOD
Figure 14: Comparison of TPHP concentrations [ng/g] found in beach samples and resembling new products. Concentrations <LOD are considered 0. Horizontal red line indicates LOQ. On the x-axis, product types are mentioned, with corresponding sample codes (BS= beach sample no., NP= resembling new product no.) corresponding to pictures in appendix iii. *concentration below LOQ
Figure 15: Variation of TPHP [ng/g] found on beach samples at different sampling locations. Samples with concentrations < LOD are not mentioned in this figure.
0 10 20 30 40 50 60 70 80 Cap (BS24*,
NP20*) BS7, NP18)Cap (BS30, (BS28, NP17)Hair comb (BS4, NP3)Packaging Pen (BS65*,NP26) (BS31/44*,Spoon NP24) Spoon (BS38, NP6*) Straw (BS67,NP13) Syringe(BS47, NP23*) Tr ip he ny l p ho sp ha te [ ng /g ]
TPHP in beach samples and their new products
beach sample beach sample resembling new product LOQ
0 10 20 30 40 50 60 70 80
1 straw 4 packaging 23 straw 28 hair comb 31 spoon 37 cap 38 spoon 42 straw 44 spoon 47 syringe 65 pen 67 straw
Tr ip he ny l p ho sp ha te [ ng /g ]
TPHP in beach samples, variation between beaches
5 Discussion
5.1 Method development
The literature describes several methods and equipment to cut the plastics in small fragments prior to extraction4. Usage of a knife and scalpel proved to be more user-friendly compared to using of a grater
or metal file since grating produced small static plastic particles, which can induce cross-contamination between samples and experiments.
Substituting the Ace:EtAc (3:7 v/v) eluent of the clean-up method for OPFR’s51 by onli EtAc, reduced
the amount of polymer matrix in the final fraction without lowering the recovery of selected OPFR’s. However, this method showed to be unsuitable for polystyrene samples, which completely dissolved in EtAc, forming a sticky solution unusable for SPE and GC-MS injection.
5.2 Product type
Among the samples collected, the most frequently identified items are packaging materials (29%) followed by caps (27%), straws and lollipop sticks (17%). From the identified samples, 84% consisted of single-use plastic products. Since a large amount of the samples collected remained unidentified, there is a possibility that the product types presented in these results do not fully cover the sources of beach plastics. In addition, some categories, such as ‘films’ and ‘packaging’ are more general terms and could be more narrowly specified to substation this identification of product types.
Moreover, a study on identification of product types of beach plastic from a recreational beach at Chennai, a city on the east coast of India, found that plastics bags (33%) were most common,
followed by cups (16%), food packaging (18%) and (bottle)caps (8%)23. In our samples, no plastic bags
were identified. However, fragments of plastic bags could have been labeled as films. This lack of identified plastic bags could indicate a successful implementation of the Maharashtra state ban on plastic bags, which went into effect in June 201852. Identifying product types can improve knowledge
about sources of plastic found in the (marine) environment, which is important for designing
strategies to tackle plastic pollution. This study shows that the majority of the plastic items are single-use plastics, mainly packaging materials. Therefore, it is recommended to focus on single-single-use plastic items in efforts tackling plastic pollution on the beaches of Mumbai.
5.3 Polymer type
Most items analysed contained PP (53%), followed by PE (41%). This is in line with the polymer types identified as marine macroplastics in earlier studies20 and reported consumption patterns of plastics.
showing that microplastics pellets found along the coast of Goa, India, mainly consist of PE8. This
discrepancy could be ascribed to the different source of debris on the Indian beaches. The source of micoplastic pellets found in Goa was most likely ocean-based, originating from ocean transport of raw manufacturing material8. The composition of polymer types in this study resembles the consumption
pattern of plastic in India22 and most likely originating from land-based sources21.
Most commercial polymer products consist of multiple polymers53. With FT-IR it is hard to identify all
polymers that are present in low percentages53. In addition, the differences between low and high
density PE (LDPE and HDPE) is not always detected by ATR FTIR54. It is therefore likely that most items
analysed in this study contain more polymer types than identified and information about specific PE typeis lacking.
5.4 DIP-MS results
The most frequently detected chemicals are phthalates and NPs. These are known additives used in plastic production13. BPA and TPHP are commonly used as antioxidant and plasticizer in plastics,
respectively13,31. Other chemicals detected with the DIP-MS are less commonly found in PP and PE
consumer products and packaging. These results might be false positives, caused by interferences of peaks and the polymer matrix or could be allocated to following reasons. First, the presence of isocyanates was indicated in several packaging sachets. Isocyanates are used as antimicrobial substance in polyurethane materials55 and are not reported as additive in PP packaging. So although it is unlikely,
the samples could consist of a PU adhesive layer, which is being used in food packaging56. Secondly,
the presence of PAHs in the beach plastic might be allocated to adsorbed PAHs. Even though concentrations of adsorbed chemicals are very low, below detection limit of DIP-MS, earlier studies reported up to 16000 ng/g PAHs adsorbed to marine plastic from Mumbai57. These levels were
correlated with oil pollution in the metropolitan city and are above the detection limit of the DIP-MS.
Another observation around the results of the screening with DIP-MS, is that it did not detect the levels of TPHP found after extraction and GC-MS analysis. This is in line with the LOD of >100 µg/g for DIP-MS determined by previous research29. Ballesteros-Góomez et al. (2016) compared detection sensitivity of
chemicals in plastic with DIP-APCI and solvent extraction with LC-based analysis methods50. They also
found a lower sensitivity for TPHP and TCPP with DIP-APCI compared to the extraction and analysis with LC-based methods. This demonstrates that the LOD of DIP-MS is suitable when looking for chemical additives in plastic, since these are often present between 1 and 70% w:w30,31. However DIP-MS is not
suitable for detection of chemical additives. Results of the DIP-MS screening in this study provide a good starting point for further research about phthalates and NPs in plastic beach litter.
The selection of TPHP for target analysis was done on the base of data obtained with another method in Task. The mass error was accidentally set to +/- 0.1, instead of +/- 0.004. This resulted in many false positive results, and explains the low values later found after extraction and GC-MS analysis. Figure 16 andFigure 17 illustrate the peaks selected by Task, and the resulting outcome of peak intensity. Since the mass error is above 0.004, the peak intensities cannot be assigned to the selected components. This shows the importance of using a high-resolution mass detection with a small window when conducting nontarget screening.
In addition to using of a high-resolution mass accuracy, peak intensities should be reviewed carefully before concluding about the presence of the regarding compound. In this study, a second set of “selective” criteria was added to standard (“extensive”) criteria, for consulting the results of the DIP-MS analysis. Selection of peaks with the selective criteria minimises the possibility of reporting false blanks but increase the possibility of excluding significant results. When using the DIP-MS analysis for screening purposes only, the extensive data analysis can be desirable, since all suspected samples will be included for further research which will give definitive results. The selective criteria are added in this study to provide a second, more trustworthy dataset for the compounds and samples which are not further analysed. For the selective criteria the minimum peak intensity was set at 400 (instead of 100), since noise ratios often occurred above intensities of 400. In addition, for the signal-to-noise ratio, the highest blank measured (instead of individual blanks per measurement) was selected since the blank values showed high fluctuations.
Moreover, isotopic patterns were checked for only a few results of the DIP-MS. This is rather time consuming since it is important to look at the shape and retention time of the individual peaks, due to possible interferences and background noise. When selling DIP-MS as a ‘fast screening method’, the usage of a high-resolution mass accuracy together with the data analysis criteria (without isotopic patterns) counter-balances accuracy with time consumption. Since qualitative measurements should follow after the screening, it was decided not to include isotopic patterns in the data analysis method of the DIP-MS in this study. The DIP-MS results should therefore only be regarded as indicative. In the qualitative GC-MS analysis, isotopic patterns were included.
Figure 16: DIP result of beach sample 1. Above the mass spectrum of the DIP-MS measurement, between 229.05 - 229.25 m/z. Below the peak intensity of BPA obtained with use of data analysis with Task.
Figure 17: DIP result of beach sample 23. Above the mass spectrum of the DIP-MS measurement, between 326.96 – 327.45 m/z. Below the peak intensity of TPHP obtained with use of data analysis with Task.
5.5 Extraction and GC-MS results
TPHP was found most frequently with concentrations between 4 and 73 ng/g in beach samples and 4-44 ng/g in new products. The TCPP concentrations were all below LOQ: 83 ng/g in beach samples and 49-89 ng/g in new products. TCEP was only detected once above the LOD, at 7 ng/g in one of the beach samples. The levels of TPHP and TCPP found are relatively low. Table 14 shows the comparison between levels of OPFR’s usually added during manufacturing of plastic products and levels found in this study. The levels found in this study are substantially lower than concentrations of additives and proportionate to levels found by studies measuring POP’s adsorbed to marine plastics45,46, shown in Table 15. It is therefore
likely that the concentrations found are chemicals adsorbed to the plastic samples, rather than additives in the plastic.
Table 14: Amounts of additives typically added during manufacturing of plastics in mass percentage [% w/w]
Additive Typical amount added [%w/w] Concentrations measured [%w/w]
OPFR 1-3031 6.3 x 10-5 – 8.9 x 10-5 (TCPP) 0.4 x 10-5 – 7.3 x 10-5 (TPHP) 0.7 x 10-5 (TCEP) Plasticizers 10 - 70 13,30 BS 1 BS 23
Table 15: Reported levels of chemicals adsorbed on marine plastics
Chemical type Levels reported adsorbed on marine plastic particles [ng/g]
Concentrations measured [ng/g] OPFRs 8.5 – 78.6 (TCPP)58 8.5 – 10.6 (TPHP)58 4.6 – 9.8 (TCEP)58 49 - 83(TCPP) 4 – 73 (TPHP) 7 (TCEP) PAH’s 27 - 120046 Nonylphenol 24.9 - 266059 BPA 5 - 30059
Another explanation could be that the extraction procedure does not fully extract all the chemicals from the polymer matrix. Even though the selected OPFR’s are not chemically bound to the polymer matrix30, shaking with DCM might not have extracted all the chemicals60, especially the larger plastic
particles. Furthermore, the variety of polymer composition and size of the cut samples, might have caused a different recovery for the extraction procedure. Therefore, comparison of concentration levels between the different products is inaccurate. To compare beach samples and resembling new products it was, however, tried to cut the sample in the exact same size during sample preparation. However, the results of the exhaustive extraction suggest that all chemicals have been extracted from the particles and extraction with DCM by sonication has been validated in earlier studies50. For future
research it is desirable to develop a method to grind the plastic samples to even smaller particles, to increase the contact area for solvent extraction and optimise consistency of the particle size. In addition, the exact extraction recovery could be validated using plastic samples with known concentrations of target chemicals. Availability of plastic particles with known concentrations of additives could improve the innovation of novel analytical methods and ease validation.
The amount of TPHP detected in beach samples (4-73 ng/g) is marginally higher compared to the new products (4-44 ng/g), but this difference was not significant. Comparison of concentration levels between matching products do not show a uniform difference. This is in line with the assumption that the measured chemicals are adsorbed to the plastic rather than present in the polymers as additives. When measuring additives, one would expect to find higher values of chemicals in the new products compared to matching beach samples, but adsorption of chemicals can even occur on the beach and also during the lifetime of a product. including transportation to The Netherlands. For measuring adsorbed chemicals, the brand or manufacture origin are less influential. The polymer type and sampling location can, however, affect the amount of chemicals adsorbed on the beach samples. Since almost all samples analysed with GC-MS were composed of PP, differences between adsorption of TPHP
at PP and PE could not be concluded from this study. Furthermore, since the samples selected for the target analysis mainly originated form Mahim beach, more samples from the other locations should be analysed before differences between these locations can be concluded. For further research on adsorbed chemicals, samples from different polymer types could be examined and equal amounts of samples from different locations could be compared.
For measuring leaching effects, the sample collection method used in this study is unreliable. Even though some beach samples and new products were from the exact same brand and polymer type, there is a lack of information about the similarity of the products and their origin. The composition of the plastic can be different between production batches and manufacturing sites, even from the same brand. In addition, the origin of the beach samples is unknown. Some could have drift ashore from the ocean, others might have been dropped on the beach shortly before sample collection. This complicates the interpretation of the results.
TPHP, TCPP and TCEP have been reported to be present in polymer compositions of computers, cars interiors, carpets and building materials31. The presence of these compounds as additives in the
products found on the beach, such as food packaging and spoons, is less likely. These OPFR’s were selected for quantitative analysis since they were detected frequently with DIP-MS. However, as explained above, due to erratic settings too many false positive values were found. Later checks proved reliable results but detected OPFR’s less frequently. In future research, other chemicals such as phthalates and 4-NP are more likely to be found in beach plastic.
6 Conclusion
6.1 Outcome research
Beach plastics from three beaches of Mumbai, India were examined. The most frequently identified items were packaging materials (29%), followed by caps (27%) and straws and lollipop sticks (17%). The beach samples analysed contain mostly PP (53%), followed by PE (40%). This composition of polymer types resembles the consumption pattern of plastic in India22.
Screening with DIP-MS indicated the presence of toxic additives such as phthalates, nonylphenol and bisphenol-A in several beach plastic items. Screening results also indicate the presence of chemicals not reported as additive, such as PAHs and isocyanates. Complimentary analysis is required to definitively conclude about the presence of indicated chemicals in beach and new samples. DIP-MS provides a valuable fast screening method for speeding up analysis of chemical additives in plastic. However, this screening method is not sensitive enough to measure traces of chemical pollutants adsorbed to (marine) plastic. To minimise false positive results of screening with DIP-MS, extra attention should be paid to the blank measurements and a high-resolution mass accuracy should be used for data analysis.
With solvent extraction and GC-MS analysis TCPP, TPHP and TCEP were detected. Only TPHP was measured above the LOQ at 4-73 ng/g in beach samples and 4-44 ng/g in the new products. No significant difference in TPHP concentrations was found between beach plastics and matching new products. The concentrations found in this study are relatively low and are more likely adsorbed from the environment rather than being added during manufacturing of the products.
6.2 Recommendations for future research
• Methods for data analysis of DIP-MS should be further optimised to balance accuracy with time consumption.
• Following the screening results, qualitative analysis of phthalates or nonylphenol could be performed on listed beach samples to provide more insight into the occurrence and potential harm of chemical additives in plastic beach litter.
• Validation of extraction methods for qualitative determination of additives in plastic could be improved with the use of standard plastic particles with known additive concentrations. • The relatively high blank signals of TCPP in GC-MS should be further investigated and minimised
7 References
(1) Schnurr, R. E. J.; Alboiu, V.; Chaudhary, M.; Corbett, R. A.; Quanz, M. E.; Sankar, K.; Srain, H. S.; Thavarajah, V.; Xanthos, D.; Walker, T. R. Reducing Marine Pollution from Single-Use Plastics (SUPs): A Review. Mar. Pollut. Bull. 2018, 137 (October), 157–171.
https://doi.org/10.1016/j.marpolbul.2018.10.001.
(2) Worm, B.; Lotze, H. K.; Jubinville, I.; Wilcox, C.; Jambeck, J. Plastic as a Persistent Marine Pollutant. Annu. Rev. Environ. Resour. 2017, 42 (1), 1–26. https://doi.org/10.1146/annurev-environ-102016-060700.
(3) Jambeck, J. R.; Geyer, R.; Wilcox, C.; Siegler, T. R.; Perryman, M.; Andrady, A.; Narayan, R.; Law, K. L. Plastic Waste Inputs from Land into the Ocean. Science (80-. ). 2015, 347 (6223), 768 LP – 771. https://doi.org/10.1126/science.1260352.
(4) Fries, E.; Dekiff, J. H.; Willmeyer, J.; Nuelle, M. T.; Ebert, M.; Remy, D. Identification of Polymer Types and Additives in Marine Microplastic Particles Using Pyrolysis-GC/MS and Scanning Electron Microscopy. Environ. Sci. Process. Impacts 2013, 15 (10), 1949–1956.
https://doi.org/10.1039/c3em00214d.
(5) Conkle, J. L.; Báez Del Valle, C. D.; Turner, J. W. Are We Underestimating Microplastic Contamination in Aquatic Environments? Environ. Manage. 2018, 61 (1), 1–8. https://doi.org/10.1007/s00267-017-0947-8.
(6) Wang, J.; Zheng, L.; Li, J. A Critical Review on the Sources and Instruments of Marine
Microplastics and Prospects on the Relevant Management in China. Waste Manag. Res. 2018, 36 (10), 898–911. https://doi.org/10.1177/0734242X18793504.
(7) Napper, I. E.; Bakir, A.; Rowland, S. J.; Thompson, R. C. Characterisation, Quantity and Sorptive Properties of Microplastics Extracted from Cosmetics. Mar. Pollut. Bull. 2015, 99 (1–2), 178– 185. https://doi.org/10.1016/j.marpolbul.2015.07.029.
(8) Veerasingam, S.; Saha, M.; Suneel, V.; Vethamony, P.; Rodrigues, A. C.; Bhattacharyya, S.; Naik, B. G. Characteristics, Seasonal Distribution and Surface Degradation Features of Microplastic Pellets along the Goa Coast, India. Chemosphere 2016, 159, 496–505.
https://doi.org/10.1016/j.chemosphere.2016.06.056.
(9) Dauvergne, P. Why Is the Global Governance of Plastic Failing the Oceans? Glob. Environ. Chang. 2018, 51 (December 2017), 22–31. https://doi.org/10.1016/j.gloenvcha.2018.05.002. (10) Rochman, C. M.; Hoh, E.; Kurobe, T.; Teh, S. J. Ingested Plastic Transfers Hazardous Chemicals to Fish and Induces Hepatic Stress. Sci. Rep. 2013, 3, 1–7. https://doi.org/10.1038/srep03263.
(11) Cole, M.; Lindeque, P.; Fileman, E.; Halsband, C.; Galloway, T. S. The Impact of Polystyrene Microplastics on Feeding, Function and Fecundity in the Marine Copepod Calanus
Helgolandicus. Environ. Sci. Technol. 2015, 49 (2), 1130–1137. https://doi.org/10.1021/es504525u.
(12) Frond, H. L.; Sebille, E.; Parnis, J. M.; Diamond, M. L.; Mallos, N.; Kingsbury, T.; Rochman, C. M. Estimating the Mass of Chemicals Associated with Ocean Plastic Pollution to Inform Mitigation Efforts. Integr. Environ. Assess. Manag. 2019, 15 (4), 596–606.
https://doi.org/10.1002/ieam.4147.
(13) Hermabessiere, L.; Dehaut, A.; Paul-Pont, I.; Lacroix, C.; Jezequel, R.; Soudant, P.; Duflos, G. Occurrence and Effects of Plastic Additives on Marine Environments and Organisms: A Review. Chemosphere 2017, 182, 781–793. https://doi.org/10.1016/j.chemosphere.2017.05.096. (14) Liu, C.; Wang, Q.; Liang, K.; Liu, J.; Zhou, B.; Zhang, X.; Liu, H.; Giesy, J. P.; Yu, H. Effects of
Tris(1,3-Dichloro-2-Propyl) Phosphate and Triphenyl Phosphate on Receptor-Associated MRNA Expression in Zebrafish Embryos/Larvae. Aquat. Toxicol. 2013, 128–129, 147–157.
https://doi.org/10.1016/j.aquatox.2012.12.010.
(15) Rochman, C. M. The Complex Mixture, Fate and Toxicity of Chemicals Associated with Plastic Debris in the Marine Environment. In Marine Anthropogenic Litter; Springer International Publishing: Cham, 2015; pp 117–140. https://doi.org/10.1007/978-3-319-16510-3_5.
(16) León, V. M.; García-Agüera, I.; Moltó, V.; Fernández-González, V.; Llorca-Pérez, L.; Andrade, J. M.; Muniategui-Lorenzo, S.; Campillo, J. A. PAHs, Pesticides, Personal Care Products and Plastic Additives in Plastic Debris from Spanish Mediterranean Beaches. Sci. Total Environ. 2019, 670, 672–684. https://doi.org/10.1016/j.scitotenv.2019.03.216.
(17) León, V. M.; García, I.; González, E.; Samper, R.; Fernández-González, V.; Muniategui-Lorenzo, S. Potential Transfer of Organic Pollutants from Littoral Plastics Debris to the Marine
Environment. Environ. Pollut. 2018, 236, 442–453. https://doi.org/10.1016/j.envpol.2018.01.114.
(18) Zhang, H.; Zhou, Q.; Xie, Z.; Zhou, Y.; Tu, C.; Fu, C.; Mi, W.; Ebinghaus, R.; Christie, P.; Luo, Y. Occurrences of Organophosphorus Esters and Phthalates in the Microplastics from the Coastal Beaches in North China. Sci. Total Environ. 2018, 616–617, 1505–1512.
https://doi.org/10.1016/j.scitotenv.2017.10.163.
(19) Van, A.; Rochman, C. M.; Flores, E. M.; Hill, K. L.; Vargas, E.; Vargas, S. A.; Hoh, E. Persistent Organic Pollutants in Plastic Marine Debris Found on Beaches in San Diego, California.
