Metal Content in Textile and (Nano)Textile Products

Metals, metallic compounds, and, recently, metallic nanoparticles appear in textiles due to impurities from raw materials, contamination during the manufacturing process, and/or their deliberate addition. However, the presence of lead, cadmium, chromium (VI), arsenic, mercury, and dioctyltin in textile products is regulated in Europe (Regulation 1907/2006). Metal determination in fabrics was performed by inductively coupled plasma-mass spectrometry (ICP-MS) after microwave-assisted acid digestion. The ICP-MS procedure has been successfully validated; relative standard deviations were up to 3% and analytical recoveries were within the 90–107% range. The developed method was applied to several commercial textiles, and special attention has been focused on textiles with nanofinishing (fabrics prepared with metallic nanoparticles for providing certain functionalities). Arsenic content (in textile T4) and lead content (in subsamples T1-1, T1-2, and T3-3) were found to exceed the maximum limits established by the European Regulation 1907/2006. Although impregnation of yarns with mercury compounds is not allowed, mercury was quantified in fabrics T1-2, T5, and T6. Further speciation studies for determining hexavalent chromium species in sample T9 are necessary (hexavalent chromium is the only species of chromium regulated). Some textile products commercialised in Europe included in this study do not comply with European regulation 1907/2006.


Introduction
Textile products can be contaminated with metals during both production and storage. In addition, manufacturers add intentionally metals or metallic compounds to textile products to achieve or improve specific functionalities. As example, approximately 80% of leathers are chrome-tanned [1,2], while vanadium, chromium, barium, lead, copper, cobalt, and nickel are used in textile dyeing [3,4], and iron and manganese are used in bleaching processes [5,6]. In addition, antimony compounds have been demonstrated to enhance the flame retardant properties of halogen compounds and commonly flameretardant compounds used in textiles preparation [7,8]. Similarly, titanium can be present in some textile products after the Zirpro process with hexafluorotitanate salts for achieving fire-resistant textiles [8,9]. Furthermore, antimony can be in poly-ethylene terephthalate (PET) fibres since antimony compounds are used as catalysts in PET manufacture [10].
Currently, the textile industry pays great attention to the possibilities of using metallic nanoparticles (NPs) to improve the manufacture of fibres and to obtain fabrics with new or improved properties [11]. A new category has, therefore, been added to textile finishing, called "nanofinishing" [12]. The use of titanium dioxide nanoparticles (TiO 2 NPs) is an example of nanofinishing to improve the existing textile manufacturing procedures, specifically the scouring and bleaching processes of fibres (nanoscouring and nanobleaching).

Instrumentation
Samples and reagents were weighed using an analytical balance ML 204T (Mettler Toledo, Columbus, OH, USA). The microwave-assisted acid digestion was performed in 100 mL closed Teflon vessels using an ETHOS PLUS microwave lab-station (Milestone, Sorisole, Italy). The total metal content was assessed using a NexION ® 300X ICP-MS (Perkin Elmer, Waltham, MA, USA) equipped with a SeaFastSC2 DX autosampler (Elemental Scientific, Omaha, NB, USA). The nebulisation system consisted of a Meinhard ® nebuliser and a cyclonic spray chamber thermostated by a Peltier refrigerator.

Samples
Type and composition of the studied clothing are listed in Table 1. The table also shows the number of subsamples for some textiles, as some of them were visually different (made with different fibres). The textile products studied were bought from online marketplaces except for sample T6 (men's cycling culotte) and sample T9 (compression socks with silver ions for diabetic people) which were purchased from a local shop and a pharmacy, respectively. All studied garments were manufactured in Europe, except T1, T2, and T3 (textile products fabricated in Bangladesh by the same company).
Each garment subsample was cut in pieces (5 × 5 mm approximately) with ceramic scissors and stored in plastic bags at room temperature in the dark.

Microwave-Assisted Acid Digestion
Clothing samples (0.2000 g) were placed into PTFE digestion vessels and 8 mL of 69% nitric acid, 0.5 mL of 33% hydrogen peroxide, 0.5 mL of 37% hydrochloric acid, and 1 mL of 40% hydrofluoric acid were added. Operating conditions of the microwave-assisted acid digestion procedure are detailed in Table 2. Digested samples and blanks were made up to 25 mL with ultrapure water. The samples were digested in triplicate and one blank was performed at least for each set of digestions.

Total Metal Determination by ICP-MS
Daily adjustment of ICP-MS parameters such as torch position, quadrupole voltages, and nebulisation flow was performed to maximise the sensitivity. This adjustment was carried out before the measurement of samples and using a solution of Be, Ce, Fe, In, Li, Mg, Pb, and U (1 µg L −1 ). Operating and acquisition parameters of ICP-MS, as well as monitored ions, are listed in Table 3. Due to the low sensitivity of arsenic, a longer dwell time (200 ms) than that used for the other elements (50 ms) was selected. Polyatomic interferences were minimised using helium as a collision gas (either 1.0 mL min −1 or 4.0 mL min −1 flow rate depending on the monitored ion, Table 3). A solution containing 10 µg L −1 of internal standards (Ge, Y, Rh, and In) in 1% nitric acid was mixed with standards and samples by a T-shaped plastic piece before their introduction into the nebuliser. Table 3. ICP-MS conditions for metal assessment in digested clothing samples.

Operating Parameters
Radiofrequency power (W) 1600 Plasma gas flow (L min −1 ) 16 Auxiliary gas flow (L min −1 ) 1.2 Nebulisation gas flow (L min −1 ) 0.9-1.1 Collision cell gas He Digested textiles were diluted 10-fold with ultrapure water before analysis except for Zn determination in T1, T2, and T3 samples (and their respective subsamples), and for Ti assessment in all samples (200-fold dilution). Multi-elemental standard addition calibration was performed to avoid matrix effects. The standard addition calibration covered the range of 0-100 µg L −1 , except for the quantification of aluminium and iron (range of 0-5000 µg L −1 ).

Microwave-Assisted Acid Digestion
Since nitric acid has been found adequate to decompose textiles based on cotton, wool, polyester, flax, and hemp [35,38,39,41], a mixture of 69% nitric acid 33% hydrogen peroxide was firstly used for microwave-assisted acid digestion of (nano)textiles (Table 1). However, fine white particles were observed in all digests, which impairs precision of the ICP-MS measurements. Therefore, other mineral acids such as hydrochloric acid and hydrofluoric acid were tested, and clear acid digests were obtained when using small volumes of both acids (0.5 and 1.0 mL of hydrochloric acid and hydrofluoric, respectively). Finally, a mixture of nitric acid/hydrogen peroxide/hydrochloric acid/hydrofluoric acid (8:0.5:0.5:1, volume ratio) was selected, and the microwave operating conditions are listed in Table 2.

Limit of Detection, Precision, and Analytical Recovery Assays
The 3σ/m and 10σ/m criteria were applied to calculate the limit of detection (LOD) and the limit of quantification (LOQ), respectively. Using these criteria, σ is the standard deviation of eleven measurements of a blank (1% nitric acid) by ICP-MS, and m is the slope of the standard addition calibration. The LOD and LOQ of developed method referred to the textile sample mass are shown in Table 4. LOQs ranged between 0.00427 (beryllium) and 6.33 µg g −1 (titanium). The high LOQ value for titanium was attributed to the high dilution required (1:200) for its analysis. The precision of ICP-MS determinations for each element was evaluated by the relative standard deviation (RSD) of eleven measurements of an acid digest. RSDs for metal concentrations were within the range of 1-3% (Table 4), showing good precision for ICP-MS quantification.
The accuracy of ICP-MS assessments was based on the analytical recovery approach at several concentration levels by spiking acid digests at the selected metal concentrations (Table 4). Each spiked digest was analysed eleven times and analytical recoveries ranging from 90 to 107% were obtained (Table 4), which implies accurate ICP-MS determinations.

Metal Content in the Analysed Textiles
Metal content in the studied textile products assessed by ICP-MS after a microwaveassisted acid digestion is shown in Table 5.

Discussion
Despite the fact that the fabrics were manufactured in several European countries and some of them (T1, T2, and T3) in Bangladesh, most of the metal levels were found to be similar in all textile samples. The presence of some metals can be directly related to their use (or their compounds' usage) during fabrics production and/or functionalisation. However, some metals such as beryllium, mercury, and cadmium were detected at trace levels in some samples even though they have not been reported to have a specific role in the textile industry. Beryllium (concentrations lower than 0.0300 µg g −1 ), mercury (concentrations between 0.0818 and 0.375 µg g −1 ), and cadmium (concentrations from 0.103 to 0.319 µg g −1 ) were found, and their presence in some garments could imply contamination during the manufacturing, storage and/or selling of textiles. Beryllium concentrations in the analysed samples were lower than those reported by Rezyć et al. [38] in samples processed in the Croatian textile industry.
The presence of other metals in textiles can be related to their use in some stages of the production. This is the case of lithium (concentrations within the range of 0.109-0.699 µg g −1 ) since lithium grease is one of the most widely used lubricants in industry worldwide, and the use of lithium grease in mechanical devices during the manufacturing of garments could contaminate textiles with this metal.

Metals Used in the Dyeing Process
The presence of metals in textiles can be directly related to the dyeing process. That is the case of vanadium, since vanadium compounds are used as catalysts in the production of aniline black dye (CI 50440) [42,43]. Vanadium was detected in most of samples, and the range of concentrations was 0.0160-0.256 µg g −1 where samples made of black fibres showed the highest vanadium content [T4 (0.123 ± 0.0167 µg g −1 ), T6 (0.256 ± 0.0294 µg g −1 ), and T9 (0.143 ± 0.0000797 µg g −1 )]. Another example is chromium, because several chromiumbased dyes, such as chrome complexes with salicylic acid-based dyes (Mordant Yellow 1 and Mordant Yellow 5) give rise to mainly yellow and orange colours. In addition, Mordant Red 7, Mordant Red 19, Mordant Black 3, Mordant Black 9, Mordant Black 11, and Mordant Blue 13 dyestuffs, obtained by chroming aryl azo compounds, are also used in the textile industry [3,44]. Chromium was, therefore, quantified in all samples, except T1-1, and chromium concentration was within the range of 0.156-1.03 µg g −1 , except in textile coded as T9, which showed a very high chromium content (974 ± 6.01 µg g −1 ). Except for this sample (T9), similar concentrations were obtained by Menezes et al. [37] in textile samples produced in Brazil and China.
In addition, barium sulphate (whitening colourant) and barium pigments can be used in the textile industry. Most of the analysed samples were found to contain barium, such as textiles T2-3 (53.9 ± 4.82 µg g −1 ) and T13 (0.758 ± 0.117 µg g −1 ) that contained the highest and lowest concentrations, respectively. Similarly, lead white and lead chromate pigments are also used in the textile industry [45]. Lead concentrations varied between 0.165 and 1.39 µg g −1 .
Other metals such as copper, cobalt, and nickel are also used during the dyeing process as mordants [3]. This is the case of cobalt and nickel chlorides, as well as copper (I, II) sulphates [46]. These elements were in the studied samples at the concentration ranges as follows: cobalt from 0.00561 to 3.21 µg g −1 , nickel between 0.0383 and 0.341 µg g −1 , and copper in the range of 0.361-41.1 µg g −1 . The concentrations of these elements were lower than those previously reported in the literature [37][38][39]. Similarly, stannous chloride is also used as a mordant in dyeing procedures [46], and certain organotin species have been reported as biocides in textile products [47]. Sixteen studied samples were found to contain tin which ranged from 0.0556 to 0.255 µg g −1 , except for textile T4 (5.34 ± 0.199 µg g −1 ) and textile T9 (1.26 ± 0.0630 µg g −1 ).

Metals Used in the Bleaching Process
Iron can be used to form metal-complex dyes [3,46,48] but it is also used as a catalyst in the bleaching process [5,6] (hydrophobic components removal guarantees an optimum further dyeing). All studied textile samples have been found to contain iron, and textile T4 was the one with the highest level (57.1 ± 3.67 µg g −1 ). Furthermore, the presence of manganese in textiles is also related to the bleaching stage since several manganese complexes are used as catalysts to enhance this process [6]. Regarding manganese, textile samples coded as T6 and T4 showed the highest concentrations (85.1 ± 3.76 and 25.0 ± 1.10 µg g −1 , respectively). The concentration of manganese in the remaining samples was in the range of 0.171-7.20 µg g −1 .
Samples coded as T7 and T8 were purchased from the same manufacturer (e-market) who specified that TiO 2 NPs were added to these textile products (children's T-shirt) to obtain protection against ultraviolet radiation (Table 1). It was found that titanium concentrations were 2590 ± 20 and 2949 ± 67 µg g −1 for T7 and T8, respectively.
The manufacturers of the remaining samples did not specify the addition of titanium, which could mean that titanium was used to enhance fire retardation or opacity of fibres. Moreover, titanium could be present in these samples since they contain totally or partially white fibres (bulk titanium dioxide and TiO 2 NPs provide brightness and opacifying characteristics). Samples T10, T11, T12, T13, and T14 (T14-1, T14-2, and T14-3) were all-white clothes. Sample T6 was the sample with highest concentration of titanium, and it was half white and half black (one colour in each side of the piece of garment). In addition, titanium was present in T1-2, T2-2, T2-3, T3-2, and T3-3 which were subsamples that contain white fibres mixed with coloured fibres.
Regarding zinc, textiles coded as T1, T2, and T3 were purchased from the same manufacturer who specified that these garments contained ZnONPs (Table 1) to provide UV protection [59]. Therefore, these textiles showed very high zinc concentrations, ranging from 280 to 873 µg g −1 . Zinc content in the remaining textiles was slightly lower than those levels found in samples T1 to T3, varying between 0.668 and 5.68 µg g −1 .
Finally, several manufacturers of the studied textiles have specified the presence of silver ions or AgNPs in the garments since silver compounds offer biocide properties to the textiles [60]. In accordance with textile compositions given by the manufacturers, silver ions were present in the samples T4, T5, T6, and T9, whereas AgNPs were added to textiles T10, T11, T12, T13, and T14 (T14-1, T14-2, and T14-3) ( Table 1). Silver concentrations in the above-mentioned samples ranged between 0.436 (T11) and 31.2 µg g −1 (T5). In addition, another two samples (T1-1 and T2-1) were found to contain silver, although silver ions/NPs were not listed in the textile composition.

Other Uses of Metals in the Textile Industry
The presence of metals such as molybdenum and antimony is related to their use in flame retardant formulations [7,8,61]. Molybdenum was quantified in twelve samples at concentrations lower than 0.600 µg g −1 . The levels of Mo in the samples were lower than those reported by Rezyć et al. [38] whose found Mo concentrations between 0.04 and 9.93 µg g −1 in textile samples. However, antimony was present at high concentrations (0.301 to 218 µg g −1 ). As previously commented, high antimony concentrations in polyester fibres can be due to its use in the fabrication of polyethylene terephthalate (the most common type of polyester used in garments) [10]. A direct relationship was found between the presence of antimony and the percentage of polyester in the samples, and samples with higher polyester percentages than 50% (Table 1) were those with the highest antimony levels (higher than 100 µg g − 1 , samples T2-2, T2-3, T3-2, T3-3, T5, T6, T11, and T13).
Arsenic compounds are effective insecticides and herbicides and the use of organoarsenicals is a common practice to control weeds in cotton fields and as defoliants of cotton plants [62]. Arsenic was, therefore, found in six cotton-based textile samples (T2-3, T3-2, T3-3, T4, T14-1, and T14-3) and in one non-cotton-based fabric (T11) ( Table 1). The concentration of arsenic in the studied textiles varied between 0.0556 and 0.250 µg g −1 , except for sample T4 (15.8 ± 0.920 µg g −1 ), where high arsenic levels could be associated with high cotton percentage (80%) in the garment.

Conclusions
A microwave-assisted acid digestion procedure using a mixture of nitric acid, hydrogen peroxide, hydrochloric acid, and hydrofluoric acid has found adequate for decomposing a vast variety of textiles and (nano)textiles. ICP-MS multi-element determinations were found to be accurate (analytical recoveries within the range of 90-107%) and precise (RSDs varied between 1 and 3%).
On the other hand, speciation studies are necessary in the analysis of chromium. Hexavalent chromium is the only species of chromium regulated in textile products, and its maximum allowable concentration is 1 µg g −1 . Total concentrations of chromium in textiles T4 and T6 were just in the maximum legal limit of hexavalent chromium, whereas sample T9 was over it (974 ± 6.01 µg g −1 ).
Regarding metals such as silver, titanium, and zinc, which are mainly added as NPs (AgNPs, TiO 2 NPs, and ZnONPs) in functionalised (nano)textiles, high levels have been found in some garments which were reported to contain NPs by the manufacturers. Further studies would be welcome in order to characterise the particle size distribution as well as the release of these metallic NPs under conventional washing (laundry) conditions.

Conflicts of Interest:
The authors declare no conflict of interest.