Determination of phthalate esters in children ’ s toys

A validated analytical method for the determination of phthalates (DBP, BBP, DEHP, DNOP, DINP and DIDP) in plastics was described and applied to 20 toys in Ho Chi Minh city, Vietnam. Phthalates in the sample were extracted in a mixture of acetone and hexane and analyzed by GC-FID. Limits of detection (LOD) and limits of quantification (LOQ) of the analysis method for DBP, BBP, DEHP và DNOP in the PVC plastic were from 0.0011% to 0.0014% and from 0.0028 to 0.0046%, respectively. Recoveries of DBP, BBP, DEHP và DNOP in PVC and PP at two spiked concentrations (0.1% and 0.01%) ranged from 86.6% to 91.7% with the relative standard deviation (%RSD) were from 2.5% to 8.7%. Wide ranges of linearity of standard curves were obtained from 0.5 mg/L to 100 mg/L for DBP, BBP, DEHP and DNOP and from 25 mg/L to 1000 mg/L for DINP and DIDP with R > 0.999. The studied method was a reliable, rapid, easy and cheap method and permitted to determine the restricted phthalates in plastic toys in compliance with European Regulation (No 1907/2006). The method was applied to determine phthalates in 20 toys in HoChiMinh city, Vietnam. Phthalates were found in almost soft PVC toys with high concentrations.


INTRODUCTION
Phthalate or phthalate ester is a general term for esterified substances of phthalic acid and alcohol.Phthalate esters are added to plastics to make these materials more flexible and elastic.

Phthalates have toxicity such as carcinogenesis, immunization failure or reproduction failure…
The high concentrations of phthalates in toys will harm children's health.
The REACH ("Registration, Evaluation, Authorisation and Restriction of Chemicals") Directive (EC No 1907/2006) restricts dibutyl phthalate (DBP), butyl benzyl phthalate (BBP), di-2-ethylhexyl phthalate (DEHP) to 0.1% in all toys and di-octylphthalate (DNOP), diisononyl phthalate (DINP), and di-isodecyl phthalate (DIDP) to the same percentage in mouthing toys [1].Moreover, the restricted substances are not Trang 80 limited to synthetic resins of which the principal constituent is polyvinyl chloride but have been expanded to other materials that may include plasticizers.
Phthalates in plastic toys can be extracted by different organic solvents and measured by GC-FID or GC-MS [2][3][4][5][6][7][8][9].Recently, in Vietnam, many restricted phthalates with high concentrations were found in plastic toys.Cheap recycled plastic toys manufactured by local companies can also contain phthalates.In this study, 6 restricted phthalates DBP, BBP, DEHP, DNOP, DINP and DIDP were investigated in PVC (polyvinyl chloride) and recycled PP (polypropylene) children's toys in Vietnam.In this study, sample (0.5 g) was added only 15 mL of a mixture of acetone: hexane (3:7, v/v) and let stand overnight at room temperature.The sample was centrifugated for 15 minutes at 4000 rpm and the extract was transferred to a 10-mL volumetric flask and added hexane to 10 mL.Before injecting to GC-FID, the extract was filtered using PTFE filter (0.45 µm) and diluted 5 times with hexane.For the GC-MS confirmation, the extract may be diluted over 10 times with hexane.

Recovery study
PVC blank sample (without studied phthalates) was kindly provided by a laboratory in HoChiMinh city.The PVC blank sample and a raw PP material were used as blank samples for recovery studies of phthalates in PVC and PP matrices.
Blank samples were added standard phthalates at two different concentrations (0.1% and 0.01%), let stand 2 hours before solvent extraction and GC-FID analysis.

Analysis of Reference Material (RM)
RM sample (labelled #13044) [3] was a children's booklet with a plastic witch inside containg DBP and DINP.The solvent extraction and GC-FID analysis were applied to determine DBP in the RM sample.

Method LOD and LOQ
Blank PVC sample (0.5 g) was added a mixture of DBP, BBP, DEHP and DNOP at 0.005% or was added DINP or DIDP at 0.1% and let stand for 2 hours at room temperature before solvent extraction and GC-FID analysis.LOD and LOQ of the method were the concentrations with the Signal/Noise (S/N) equal 3 and 10, respectively.

Real samples
A total of 20 plastic children's toy samples

GC-FID determination
GC-FID and GC-MS conditions in this study were reported elsewhere [4,5].

Detection limit and linear correlation
Calibration curves for all of the studied phthalates were linear (R 2 ≥ 0.999) in the investigated concentration range (0.5 -100 mg/L for the first four phthalates (Table 2) and from 25 -1000 ppm for DINP and DIDP.The LODs for the GC-FID were approximately 0.10 mg/L and LOQs were from 0.22 to 0.43 mg/L for the first four phthalates (Table 2).The LODs of DINP and DIDP by GC-FID were approximately 10 ppm and it was not easy to determine LODs due to broad unresolved peaks of many isomers in these phthalates.Analysis of DINP and DIDD was omitted from determinations of restricted phthalates in toys [8].

Sample pretreatment and recovery study
Firstly, the CPSC-CH-C1001-09.3pretreatment sample method, prescribed by the US Consumer Product Safety Commission was applied for soft PVC plastic toys [2].THF (tetrahydrofuran) (5 mL) followed by hexane (10 mL) was used for the extraction of restricted phthalates in only 0.050 g of sample.However, the results of recovery studies were unrepeatable (data not shown) when the mass of sample was increased from 0.050 g to 0.2 or 0.5 g.Small amounts of the sample (0.050 g) may not show the representation of the whole analysis sample.
In addition, this sample pretreatment was generally applied for PVC plastic when many other plastics may be used in children's toys.
Therefore, in this study, sample pretreatment was conducted according to "Japanese Misnistry of Health, Labour and Welfare Notification No.336, 2010" with some minor modifications [2].The use of only 15 mL of a mixture of acetone and hexane (3:7) and the extraction at room temperature were advantages of the method.
Recoveries were from 86.6% to 91.7% for DBP, BBP, DEHP and DNOP at both spiked concentrations (0.1% and 0.01% (ten times lower than EU Regulation)) with % RSD from 2ị to 8.7% (Table 3) in PVC and PP matrices.Recoveries were 92.1 (% RSD = 5.8) and 89.7 (%RSD = 11) for DINP and DIDP, respectively in PVC at the spiked level of 0.1%.DINP and DIDP were not detected in the lower spiked concentration (0.01%) in both PVC and PP matrices (Table 3).

LOD and LOQ study of the method
The LODs of the method were from 0.0010 -0.0014% for DBP, BBP, DEHP and DNOP (Table 2) in PVC plastic, which all satisfy the limitation requirement of 0.1%.For DINP and DIDP, the LOQ of the method was approximately 0.07%.

Analysis of the RM
The studied method was applied to the RM sample and DBP were found 0.062 and 0.066% (2 replicates).The true concentration of DBP in the RM sample was 0.072% (mass/mass) [3].The studied method was applied to real samples.

Real sample
The real samples were focused on 2 groups: (1) animal coloured soft PVC plastic children's toys (15 samples) and ( 2) coloured hard recycled PP children's toys (5 samples including car, cooking or fruit toys).Toys in group 1 were bought in supermarkets or markets in Ho Chi Minh city.They were made in Vietnamese and Chinese and had higher price than toys in group 2 (approximately over 5-folds higher in price).
Toys in group 2 were cheap, recycled PP toys and made by noname local companies (no information of the manufacturers on products).The results of real samples were showed in Table 4.
Because phthalate esters were used as a plastiscizers in PVC, many researches focused on soft PVC toys.In 2000, a total of 72 toys were purchased in 17 countries and many restricted phthalates were found with high concentrations [7].DEHP was found in 48% of the analysed PVC toys (n=72) in concentrations of 0.003-35.5% followed by DBP (12.5% detection frequency with lower levels 0.002-0.18%).

(
15 soft PVC toys, 5 recycled PP toys) was analyzed for studied phthalates.Samples were focus on soft coloured PVC and hard coloured recycled PP children's toys for children under 3 year olds.PVC toys were bought in supermarkets or markets and recycled PP toys were bought from local manufacturers in HoChiMinh city in 2014 -2015.
Figure 1 showed the chromatograms of studied phthalate standard solutions by GC-FID in 15 minutes.As DINP and DIDP consist of different isomers, low resolutions of peaks of these phthalates were reported [2-9].The peak of DNOP are overlapped with peaks of DINP.Peaks of DINP (C9 isomers) are partial overlapped with peaks of DIDP (isomers of C10).Column temperature programs were changed to achieve the separation of DNOP from DINP (data not shown) but the problem sill remained.Therefore, separate standard curves of DINP and DIDP were performed together with the standard curve of a mixture of DBP, BBP, DEHP and DNOP.The retention time of DNOP is so close to DINP.Therefore, GC-MS confirmation may be used when DNOP is detected in real samples[2].

Figure 2 .
Figure 2. GC-FID chromatograms of phthalate ester in a standard solution (a) and in a PVC-toy sample solution with DBP and DEHP detected (sample Ti14-4) (b).

Figure 3 .Fig. 2 Figure 4 .Figure 5 .
Figure 3. TIC chromatogram of a PVC toy (sample Ti14-4) Fig. 2 showed the GC-FID chromatograms of studied phthalates in a standard solution (a) and in a PVC toy sample solution (b).Fig. 3 showed the TIC chromatogram obtained from analysis of an extract of a PVC toy (sample Ti14-4) by GC-MS.Peaks in the GC-MS chromatogram (Fig. 3)

Table 1 .
Analytical conditions

Limit of detection (LOD) and limit of quantitation (LOQ) of the GC-FID instrument
g) was added 50 mL of a mixture of acetone: hexane (3:7, v/v) and let stand overnight at 40 o C.

Table 2 .
Linearity of phthalate ester calibration curves and LOD and LOQ of the GC-FID instrument and the method

Table 3 .
% Recoveries (R) of phthalate esters in PVC and PP

Table 4 .
Real sample