A reliable detection method for nano-silver in functional textiles

Materials

Nitric acid (trace metal grade) and sulfuric acid (trace metal grade) were purchased from Fisher Chemical. PerkinElmer Pure IV (1000 mg L⁻¹ of Ag, Al, B, Ba, Bi, Ca, Cd, Co, Cr, Cu, Fe, Ga, In, K, Li, Mg, Mn, Na, Ni, Pb, Sr, TI, and Zn in 10% HNO₃) and NexION Setup Solution (1 μg L⁻¹ of Be, Ce, Fe, In, Li, Mg, Pb, and U in 1% HNO₃) were obtained from Perkin Elmer. Rhodium internal standard (1000 ± 10 μg L⁻¹ of in 15% (v/v) HCl) was purchased from Inorganic Ventures. Silver nanoparticles (60–120 nm) were purchased from Aladdin. Deionized water was used in all experiments.

Samples

The two functional textile samples were a sport T-shirt (ST) and a pair of sport socks (SS). For detection limit investigation, two non-functional textiles, a conventional T-shirt (CT) and a UV arm cover (AC), were also used in this study. All of them were purchased in local markets (Beijing, China). During the analysis, all samples would be washed and ashed. Corresponding marks and characteristics of the samples are summarized in Table 1.

OPEN IN VIEWER

Table 1.

Sample information and marks

Code	Sample	Characteristics	Mark	Washing	Ashing	Washing and ashing
1	Sport T-shirt	63% nylon 37% polyesterInhibits the growth of odor-causing bacteria	ST	STW	STA	STWA
2	Sport sock	96% nylon 4% elastaneInhibits the growth of odor-causing bacteria	SS	SSW	SSA	SSWA
3	Conventional T-shirt	White, 100% cotton	CT	CTW	CTA	CTWA
4	UV arm cover	White, 86% polyamide 13% elastane 1% polyester, UPF50+	AC	ACW	ACA	ACWA

UV: ultraviolet; ST: sport T-shirt; SS: pair of sport socks; CT: conventional T-shirt; AC: UV arm cover.

Sample pretreatment

Washing: ST/SS (about 8 cm × 8 cm, w₁ g) were rinsed by 400 mL deionized water three times and dried at 100°C. After drying, the sample was marked as ST_W/SS_W (w₂ g). Ashing: The ST_W/SS_W (w₃ g) was cut up and calcined in a muffle furnace at T°C for t h (denoted by T-t) with a heating rate of 7.5°C/min, marked as ST_WA/SS_WA (w₄ g).

Sample preparation for XRD analysis

ST/SS/ST_W/SS_W (about 5 cm × 5 cm) were cut and placed on the XRD sample holder for measurement. ST_A/SS_A/ST_WA/SS_WA were ground, then used to fill the XRD sample holder and flattened by a glass slide.

Sample preparation for SEM-EDS analysis

The fine fiber of ST/SS/ST_W/SS_W were drawn out and stuck to a sample stage through carbon conductive tape. Ground ST_A/SS_A/ST_WA/SS_WA were stuck to the sample stage via carbon conductive tape. In order to increase conductivity, all samples were treated by spray-platinum.

Sample preparation for transmission electron microscopy analysis

To wash out the nanoparticles in the textile samples: ST/SS (1.5 cm × 1.5 cm) were immersed in 15 mL of deionized water, ultrasonically treated for 30 min, and then stirred for 30 min at 400 r/min. After repeating the previous operation twice, the samples were filtered and liquid filtration was concentrated to about 200 μL in a 60°C oven. The results were tested using transmission electron microscopy (TEM) analysis.

Sample preparation for ICP-MS analysis

Total content analysis of silver in ST/SS: 0.1 g of ST/SS were cut into scraps and digested in 10 mL of concentrated nitric acid at 180°C, and then adding 1 mL of concentrated sulfuric acid until the ST/SS dissolved completely. The resultant was diluted to 50 mL and filtered to apply in analyzing the total amount of silver in the samples by ICP-MS for proving the accuracy of TGA quantitative detection.

Sample preparation for the detection limit study of the method

Carbon materials: 5.0 g of glucose was dissolved in 50 mL of deionized water, and then sealed in a Teflon-lined autoclave. The autoclave was kept at 180°C for 20 h. After natural cooling to room temperature, the resulting products were cleaned ultrasonically three times with deionized water and dried at 60°C for 18 h. The carbon materials as a matrix to apply in the detection limit investigation was obtained.

Ashing residual rate: CT/AC were washed and ashed in a muffle furnace at 450°C for 1 h with a heating rate of 7.5°C/min, referring to the Sample pretreatment section. The ashing residual rate is the mass ratio after ashing to that before ashing (w₄/w₃).

Characterization

The XRD spectra were obtained by a Brucker D8 FOCUS X-ray diffractometer using a CuKα radiation (λ = 0.154056 nm) source with a scanning speed of 2°/min at 40 kV and 40 mA. The SEM images and corresponding EDS spectrum were recorded on a Hitachi field-emission scanning electron microscope (S-4800) with a Horiba 7593-41 at acceleration voltage of 15.0 kV. High-resolution TEM images were obtained through a JEM-F200 at 200 kV. The TGA profile was obtained using a Mettler Toledo TGA/DSC 1 analyzer under air flow of 40 mL/min with a heating rate of 10°C/min. The total silver content verification experiment was evaluated by ICP-MS (Perkin Elmer NexIon 2000) with a radio frequency (RF) power of 1600 W, atomization gas flow of 0.8 L/min, auxiliary gas flow of 0.75 L/min, plasma gas flow of 12 L/min, and Rh (10 ug/L) as the internal standard.

Conventional direct detection of nano-silver in textiles

It is well-known that the XRD can be used to identify the form and crystal of silver and, meanwhile, SEM-EDS observation provides the size, morphology, and component information of the materials. Herein, two approaches were combined to detect nano-silver in textiles for more comprehensive structural results. The typical XRD patterns and SEM-EDS images of ST/SS are shown in Figures 1 and 2, respectively.

OPEN IN VIEWER

Figure 1.

X-ray diffraction patterns of ST, ST_W, SS, and SS_W.

OPEN IN VIEWER

Figure 2.

(a) Scanning electron microscopy (SEM) image of the sport T-shirt. (b) Energy-dispersive X-ray spectroscopy (EDS) spectra of areas a and b in Figure 2(a). (c) SEM image of the pair of sport socks and (d) EDS spectrums of areas c and d in Figure 2(c).

In Figure 1, the XRD patterns of all samples clearly exhibited the characteristic diffraction peaks at 2θ of 20.286° and 23.434°, which could be related to textile matrix. ²⁵ Other diffraction peaks at 2θ of 38.174°, 44.307°, 64.479°, 77.416° could be assigned as (111), (200), (220), (311) diffraction of silver, respectively, indicating the presence of elemental silver (PDF#04-0783) ²⁶ in all samples. It also suggested that washing process before ashing had little impact on XRD detection.

The SEM image of ST in Figure 2(a) shows that the sample was a fiber with the diameter of about 15 μm; meanwhile, irregular particle aggregates around 2 μm in Figure 2(a)-a and circular particles of 3 μm in Figure 2(a)-b were also observed. Relevant element components analysis of the EDS spectrum is demonstrated in Figure 2(b). The signals at 0.277, 0.523, 1.040, 2.048, 2.622, 4.510 keV represent the C, O, Na, Pt, Cl, Ti elements, respectively. The Pt element came from the coating materials to enhance the conductivity of the sample in the SEM observation preparation process, and the C and O elements showed the organic property of the textile matrix. The obvious signal of the Ti element in area Figure 2(a)-a might be involved with common functional nanomaterial titanium dioxide. The Na and Cl elements of the area in Figure 2(a)-b indicated inorganic salt in fresh ST. The SS sample (Figure 2(c)) exhibited a fiber diameter of around 15 μm, particle aggregation of about 1 μm in Figure 2(c)-c and 500 nm in the area in Figure 2(c)-d. The EDS results of those particles were similar to those of ST, except for the peak of silver at 2.984 keV in Figure 2(c)-d.

Furthermore, the ST and SS were treated as details in the Sample preparation for TEM analysis section to observe silver nanoparticles by TEM. As seen in Figure 3(a), the TEM image of ST showed nanoparticles of around 30–100 nm. The interplanar crystal spacing was measured to be 0.23 nm (Figure 3(a) inset), which could be assigned as the (111) lattice. The selected area electron diffraction (SAED) characterization suggested a typical silver diffraction of (311), (220), (200), (111) patterns (Figure 3(b)), which confirmed the 3C syn structures of silver combined with the mapping images in Figures 3(c) and (d). ²³ Similarly, the TEM image of SS in Figure 3(e) presented nanoparticles with diameters of about 10–100 nm. The lattice spacing of 0.23 nm (Figure 3(e) inset) and SAED pattern (Figure 3(f)) along with the mapping measurements (Figures 3(g) and (h)) of SS displayed clearly the characteristics for nanoscale silver particles.

OPEN IN VIEWER

Figure 3.

(a) Transmission electron microscopy (TEM) and high-resolution (inset) images. (b) Selected area electron diffraction (SAED). (c), (d) Mapping of the sport T-shirt. (e) TEM and high-resolution (inset) images. (f) SAED. (g) and (h) Mapping of the pair of sport socks.

The new detection pathway

All of the above results are typical nanomaterial characterization measures, which proved to be still suitable for textiles with a complex matrix. However, those measures showed poor sensitivity with the organic matrix as the major part of the samples, and moreover, these results provide only the structure properties, specifically the nanostructure information. When it comes to quantitative data of the nanoparticles in the entire samples, those characterizations are not adequate to draw a conclusion. Therefore, we introduced the washing and ashing process as the pretreatment to increase detection sensitivity and TGA study as a new quantitative analysis method.

Study on pretreatment conditions

Study on ashing condition

In textiles, even inorganic additives are very complicated for fabrication consideration. Therefore, washing treatment is necessary to eliminate components that may influence the detection of silver. Besides, as seen in XRD and SEM characterization, an organic matrix was the main content of the samples. During TGA measurement, carbon residues might be formed and covered the silver nanoparticles, which would obviously influence the detection of the nanoparticles. In order to increase the detection sensitivity, an ashing pretreatment was employed. However, inappropriate ashing conditions showed a strong impact on the detection results. Specifically, too high a temperature causes aggregation or deformation of silver particles due to the nanoscale effect of the material. Otherwise, inadequate temperature causes low ashing efficiency and high carbon residues to decrease the detection sensitivity.^27,28 Herein, the best ashing condition was explored through the weight loss of the heating process under an air atmosphere.

The TGA-derivative thermogravimetric (DTG) curves of ST/SS are presented in Figure 4(a), and the weight loss processes of the two samples showed the same three stages. At the first stage, the weight losses of around 1.26% of ST and 1.47% of SS at 25–100°C were related to the absorbed water. ²⁹ The second stage was in the temperature range of 200–500°C, and the weight losses were about 75.85% of ST and 82.70% of SS, corresponding to the devolatilization reaction. ²⁹ The third stage at 500–600°C revealed weight losses of about 20.91% of ST and 12.93% of SS, due to further carbonization and oxidation of the matrix.^30,31 Based on the DTG curves in Figure 4(a), it could be seen that the major weight loss of both samples took place at 450°C, and therefore the ashing temperature at 450°C was considered as an optimal choice.

OPEN IN VIEWER

Figure 4.

(a) Thermogravimetric analysis-derivative thermogravimetric (DTG) curves of the sport T-shirt (ST) and the pair of sport socks (SS) and (b) The weight loss rates of ST and SS at different ashing conditions (calcination at T°C for t h is denoted by T-t).

Subsequently, the ashing weight loss rates of different conditions by calcination in the muffle furnace are presented in Figure 4(b). When fixing 1 h as the heating time, the weight loss rates of ST were 72.3% at 400°C, 97.4% at 450°C, and 98.3% at 500°C, indicating 450°C as an optimal temperature to ensure a high ashing rate at a relatively gentle condition. At 450°C, the weight loss rate of 1 h (97.4%) was close to 1.5 h (97.3%), larger than that of 0.5 h (90.7%). These results indicated that calcinations at 450°C for 1 h were preferred as the optimal ashing condition, which was also the same case for the ST sample (Figure 4(b)).

Verification of ashing treatment

To understand the impact of washing and ashing treatment on the samples, XRD and SEM characterization were also carried out for the treated samples if the calcination treatment destroyed the nanostructure of the silver particles.

The XRD patterns of ashed samples are displayed in Figure 5. There were three components obtained through the characteristic diffraction peak. The 2θ peaks of around 38.174°, 44.307°, 64.479°, and 77.416° were attributed to elemental silver (PDF#04-0783). ²⁶ The 2θ peaks of 27.39°, 31.70°, 45.50°, and 56.52° belonged to NaCl (PDF#05-0628) ³² and the 2θ peaks of 25.33°, 48.09°. 53.94°, and 55.09° related to TiO₂ as the anatase phase (PDF#21-1272). ³³ Compared with ST (Figure 1), the intensity of diffraction peaks for ST_A increased clearly for NaCl and TiO₂ and weakly for silver. On the other hand, the diffraction peak intensity of silver in ST_WA was enhanced significantly along with the dissolution of NaCl via washing treatment. The above results demonstrated that both washing and ashing were necessary to improve the detection sensitivity of silver in textiles. The same case was also found in SS_A and SS_WA (Figure 5), which further confirmed the importance of pretreatment.

OPEN IN VIEWER

Figure 5.

X-ray diffraction patterns of ST_A, ST_WA, SS_A, and SS_WA.

The morphology and composition of ST_WA/SS_WA were characterized by SEM-EDS. As presented in Figure 6(a) (ST_WA) and Figure 6(c) (SS_WA), particles with the size of about 20–100 nm were clearly observed, coinciding with the untreated samples (Figure 3), which proved that the calcination was gentle enough to keep the nanostructure of silver. The main composition analysis of areas a, b in Figure 6(a) and c, d in Figure 6(c) via EDS spectra (Figures 6(b) and (d)) exhibited the same signals: C at 0.277 keV, O at 0.525 keV, Al at 1.486 keV, and Ag at 2.634, 2.984, and 3.151 keV. The C and O came mainly from the matrix. A little Al may be brought in during the dyeing process of textiles. ³⁴ Compared with ST/SS (Figures 2(b) and (d)), the intensity of silver signals clearly increased, which confirmed the importance of washing and ashing for improving detection sensitivity.

OPEN IN VIEWER

Figure 6.

(a) Scanning electron microscopy (SEM) image of ST_WA. (b) Energy-dispersive X-ray spectroscopy (EDS) spectra of areas a, b in Figure 6(a). (c) SEM image of SS_WA and (d) EDS spectra of areas c, d in Figure 6(c).

Quantitative analysis and verification

Figure 7 illustrates the weight loss process of ST_WA and SS_WA through TGA. As can be seen in Figure 7, the weight loss was divided mainly into three stages, namely 25–150°C, 300–630°C, and 1050–1500°C. For the first stage, the weight losses of around 10.00% of ST_WA and 6.35% of SS_WA could be attributed to adsorbed water. At the second stage, the weight losses were about 78.07% of ST_WA and 55.55% of SS_WA, corresponding to further oxidation of the carbon residues in treated textiles under an air condition. The weight losses in the third stage were 5.01% of ST_WA and 24.36% of SS_WA, which were related to the volatilization of silver particles around the melting point of silver at 961.78°C.

OPEN IN VIEWER

Figure 7.

The thermogravimetric analysis and derivative thermogravimetric (DTG) curves of (a) ST_WA and (b) SS_WA.

In conclusion, after washing and ashing pretreatment to eliminate the interference of inorganic and organic components in the textile samples, TGA data could be used to quantify the content of silver, and the specific calculation was carried out using the following formula:

c = ( w 2 / w 1 ) × ( w 4 / w 3 ) × r × 1000

(1)

where c is the content of silver in functional textiles (mg/g), w₁ is the initial weight (g), w₂ is the weight after washing and drying (g), w₃ is the weight before ashing (g), w₄ is the weight after ashing at 450°C for 1 h (g), and r is the weight loss rate at 1050–1500°C. The content of silver particles in ST and SS were 4.68 and 7.55 mg/g with errors of 0.21 and 0.17 mg/g, respectively, using the Equation (1) calculation via three parallel measurements. In order to verify the accuracy of the TGA quantitative method, ST and SS were treated as described in the Sample preparation for ICP-MS analysis section and analyzed through ICP-MS to obtain the content for 4.73 and 7.59 mg/g with errors of 0.27 and 0.25 mg/g, respectively, in accordance with TGA analysis to suggest the accuracy of the TGA quantitative method.

It is noticeable that the weight loss caused by silver species was not observed for the untreated samples, which further proved that the washing and ashing treatments were very important in the detection process.

Detection limit analysis

To explore the detection limit of the method, investigation was carried out through the XRD standard curve method. In this part, commercial silver nanoparticle powder was purchased and added to the carbon materials (as described in the Sample preparation for the detection limit study of the method section) as a matrix with different weight ratios to be used as standard samples for XRD recording.

The XRD patterns of standard samples with different contents of silver particles are shown in Figure 8(a). The intensity of diffraction peaks at 2θ of 38.125° for silver increased gradually along with the content increasing. The corresponding standard curve equation derived background correction was as follows (Figure 8(b)):

y = 0.0075 x + 0.0185 , R 2 = 0.995

(2)

OPEN IN VIEWER

Figure 8.

(a) X-ray diffraction patterns of adding silver particles in matrix materials of (a) 0 mg/g, (b) 1 mg/g, (c) 2.5 mg/g, (d) 5 mg/g, (e) 10 mg/g, and (f) 20 mg/g. (b) The standard curve equation of silver in carbon materials.

The XRD detection limit (LOD1) of silver in matrix materials was deduced by analyte concentration corresponding to three times the standard deviation of 11 consecutive determinations of intensity of the blank sample ³⁵ at a 2θ peak of 38.125°. After background correction, their intensities were 89, 99, 104, 94, 103, 99, 96, 122, 104, 109, and 109, respectively, with a standard deviation of 8.892. The LOD1 was calculated for 0.219 mg/g through three times standard deviation (8.892) as x of Equation (2). With the consideration of the ashing treatment, the detection limit of the method (LOD) was equal to LOD1 multiplying by the ashing residual rate (450°C for 1 h) of the textile. The ashing residual rate of different materials was different, related to the ingredients of the textiles. In this study, typical cotton clothes (CT) and chemical fiber textiles (AC) were selected as reference materials. The ashing residual rates were 0.602% for CT and 1.806% for AC. After calculation, the LOD of silver particles in CT and AC were 1.318 and 3.955 μg/g, respectively. Based on the above results, the detection limit of the method determined by XRD could achieve the μg/g level.

Pathway for the assessment of textiles functionalized with nanoparticles

A typical flow chart to assess the nanostructure and quantity information of textiles functionalized with silver nanoparticles is shown in Figure 9.

OPEN IN VIEWER

Figure 9.

A typical flow chart for the assessment pathway of silver nanoparticles in textiles. XRD: X-ray diffraction; SEM: scanning electron microscopy; EDS: energy-dispersive X-ray spectroscopy; TGA: thermogravimetric analysis.

For an unknown sample, the procedure should be carried out with the washing and ashing pretreatment as described before and then identification by XRD and SEM-EDS. When elemental silver and nanoparticles containing the silver element are detected by both XRD and SEM-EDS, the conclusion could be made that the sample was functionalized with silver in the nanoscale form, and otherwise, no silver nano-species is detected. Then, for samples in which silver nanoparticles were detected, quantitative analysis was conducted by TGA using ashed samples (e.g. ST_WA, SS_WA) to calculate the silver content by Equation (1). For faster detection, the pretreatment process might be skipped for only samples with high silver content where no quantitative data is needed (Figure 9(b), right-hand side).

This assessment pathway can be applied to an extensive kinds of nanoparticles with different kinds of components in different products, especially with the organic matrix for ashing benefit. It should be noticed that, for different matrixes, the ashing conditions might vary a little bit because the ashing resultant showed a great impact on both the detection sensitivity and detection limit.