Analytical and Bioanalytical Chemistry

, Volume 406, Issue 11, pp 2495–2502 | Cite as

Using ICP-qMS to trace the uptake of nanoscale titanium dioxide by microalgae–potential disadvantages of vegetable reference material

  • Theodoros Potouridis
  • Johannes Völker
  • Heiko Alsenz
  • Matthias Oetken
  • Wilhelm Püttmann
Research Paper

Abstract

As nanoscale materials have gained in economic importance over recent years, concerns about accumulation in the environment and, consequently, analysis of nanoparticles in biological material have increasingly become the focus of scientific research. A nanomaterial used in a wide range of food, consumer and household products is titanium dioxide (nTiO2). Monitoring of nTiO2 via determination of elemental titanium (Ti) can be very challenging because of a variety of possible interferences. This work describes problems during the development of a quantification method for titanium dioxide (TiO2) using inductively coupled plasma-quadrupole mass spectrometry (ICP-qMS). To evaluate the analytical method, certified vegetable reference material NCS DC 73349 was used. Interestingly, measurements of NCS DC 73349 seemed to result in acceptable recovery values—however, this was without considering interferences or conceivable differences in the natural isotopic abundance of the certified titanium calibration solution and NCS DC 73349. Actually, recoveries were lower than initially assumed. The potential interferences causing augmented recovery could be attributed to the presence of the elements sulfur (S) and phosphorus (P), which were able to form oxide ions and nitrogen-interfering species. The effect of such interfering cluster ions could be prevented by dry ashing as a sample preparation step, to evaporate S and P, before digestion with aqua regia in a high-pressure asher (HPA). Final practicability of the analysis method was proved by monitoring the uptake of nTiO2 by the microalgae Scenedesmus acutus in an environmental exposure study.

Keywords

Nanoscale titanium dioxide Algae Scenedesmus acutus Matrix interferences High-pressure asher (HPA) Inductively coupled plasma-quadrupole mass spectrometer (ICP-qMS) 

Introduction

In the year 2009 the global production of TiO2 was approximately 4.68 million tons [1]. Nanoscale TiO2 contributes only a small amount (5000 tons in the year 2010) to total global TiO2 production, but the proportion is expected to increase continuously [2]. TiO2 nanomaterials are used in a wide range of food, consumer, and household products [3] and are partly discharged into sewage, which is treated in wastewater treatment plants (WWTP) [4]. WWTP are not able to completely remove TiO2 from sewage, so that small TiO2 particles (4–30 nm) have been detected in treated effluent [5]. Once in the aquatic environment, TiO2 nanoparticles can affect organisms, and in-vivo and in-vitro experiments on the ecotoxicity of nanosize TiO2 have been reviewed [6, 7]. It is generally believed that toxicity of nanoparticles in biological systems is largely caused by the particle size, shape, and surface properties, with smaller particles being more toxic [8]. Therefore, many studies have been dedicated to the morphological characterization of engineered nanoparticles [9, 10]. However, Warheit et al. [11] concluded from mechanistically oriented investigations that particle size is only one (and perhaps a minor) factor in the environmental safety of nanomaterials. To resolve this discrepancy, the combined use of quantitative and qualitative methods for studying nanoscale TiO2 in the environment might be helpful. Detailed descriptions of analytical methods based on inductively coupled plasma-quadrupole mass spectrometry (ICP-qMS) for the determination of Ti in biological matrices are scarce. Foltête et al. [12] used an ICP-MS method to analyse Ti content in roots of the plant Vicia faba after exposure to TiO2 nanoparticles.

Digestion methods using sulfuric acid may enable good recovery of Ti in food by use of ICP with optical emission spectroscopic detection [13], but this is not possible in combination with subsequent use of ICP-qMS for quantification, because the S oxide species has the same mass-to-charge ratio (m/z = 48) as the primary Ti isotope [3]. Therefore, other digestion methods based on nitric acid, hydrogen peroxide, and hydrofluoric acid have been used when applying ICP-qMS to analysis of Ti in ceramic material, which was apparently free from S [14]. However, biological material, including higher land-plants and algae, cannot be expected to be free from S, because sulfur-containing amino acids are widely distributed in the biosphere. For example, in biomass of Scenedesmus quadricauda, approximately 0.4 % of S (on the basis of dry weight) was determined [15]. An additional problem might arise from the presence of organically bound P in biomass, because the 31P18O cluster ion might interfere with the 49Ti isotope [16]. Therefore, before using ICP-qMS methods for determination of Ti in biological material, the organically bound S and P species have to be removed.

The objective of this study was to evaluate the effect of interferences from different biological matrices on the quantification of titanium, and to develop a sample preparation step and ICP-qMS measurement method that overcome these problems. The practicability of the method was tested with samples from the microalgae Scenedesmus acutus exposed to nTiO2 under controlled conditions.

Experimental

Materials and chemicals

A multi-element ICP-MS tuning solution from Varian (Darmstadt, Germany), containing barium (Ba), beryllium (Be), cerium (Ce), cobalt (Co), indium (In), magnesium (Mg), lead (Pb), and thorium (Th), was diluted with 1 % (w/w) aqueous nitric acid solution (aq. HNO3) to obtain final concentrations of 1 and 5 μg L−1 per element. The 1 μg L−1 solution was used for torch alignment and the 5 μg L−1 solution for mass calibration. Argon gas to generate the ICP-qMS plasma was of purity 5.0 and was purchased from Praxair (Düsseldorf, Germany). Certified single-element standard solutions of In, used as the internal standard element, and Ti, both obtained from Roth (Karlsruhe, Germany), were used for external calibration of the method. Nitric acid 69 % (HNO3) and hydrochloric acid 35 % (HCl) of supra-grade quality, obtained from Roth (Karlsruhe, Germany), were used for sample digestion. To avoid cross-contamination, all laboratory equipment, including Teflon evaporating dishes, quartz glass vessels, porcelain crucibles, and polypropylene containers (e.g. centrifuge tubes), were soaked in a diluted acid bath (5 % w/w aq. HNO3) for at least 24 h and rinsed several times with ultrapure water before use.

Instrument tuning solutions, calibration solutions, internal standard solution, and sample solutions were prepared with ultrapure water with a minimum resistivity of 18.2 MΩ cm (conductivity 0.055 μS cm−1), obtained from the Astacus Analytical system from membraPure (Bodenheim, Germany). The purification was performed using the TwinPak cartridge (with organic scavenger) from Purefekt (Karlsruhe, Germany).

Samples

The bush, branches, and leaves certified reference material NCS DC 73349, from China National Analysis Center for Iron and Steel (NACIS, Beijing, China), was used for evaluating the accuracy of the analytical method.

Samples of microalgae Scenedesmus acutus var. acutus (formerly Scenedesmus obliquus) were obtained from a research project investigating adverse effects of nTiO2 on aquatic organisms, conducted by the Institute for Ecology, Evolution and Diversity (J.W. Goethe-University Frankfurt am Main, Germany). The algae were treated with Aeroxide TiO2 P25, an uncoated nanoscale material (primary particle size of 21 nm) obtained from Evonik Industries (Essen, Germany), via water exposure (algae medium according to OECD, 2011 [17]) at three different concentration levels: 0.3 mg L−1, 1.2 mg L−1, and 4.8 mg L−1. In addition, one control group of algae was grown without nTiO2. The algae were prepared after 72 hours of exposure by the following steps. First, 400 mL of each algae suspension was filtered, using a pleated filter with 4 μm mesh size (grade 595 ½, Whatman, Dassel, Germany), under a gentle vacuum. After filtration, the algae were resuspended in fresh M4-medium without EDTA (OECD, 2008 [18]). Next, the algae suspension was centrifuged (model 5702, Eppendorf, Hamburg, Germany) for 5 min at 4400 rpm, and subsequently the pellet was freeze-dried (ALPHA 1-4 LSCplus, Christ, Osterode am Harz, Germany) for 24 h to enable determination of the Ti concentration by ICP-qMS.

Sample digestion

For analysis of the bush, branches, and leaves certified reference material NCS DC 73349, a dry ashing procedure combined with an acid digestion was used. Dry ashing was performed with approximately 0.5 g sample material in an open porcelain crucible. Muffle furnace temperature was programmed as following: from room temperature up to 400 °C at 10 °C min−1; hold time 15 min; up to 475 °C at 5 °C min−1; hold time at end temperature 8 h (model LS 30/13 equipped with a TC 504 controller, Rohde, Prutting, Germany). Subsequently the sample was quantitatively transferred into a quartz glass vessel and pre-digested for 1 h with 6 mL aqua regia (1.5 HNO3 69 %, 4.5 mL HCl 35 %) at room temperature. Then the sample was placed in an HPA (HPA-S, Anton Paar, Graz, Austria) and digested for 1.5 h at 300 °C and 128 ± 8 bar (starting at room temperature, heating at maximum speed up to 80 °C, then within 30 min up to 160 °C, then at maximum speed up to 300 °C). Afterwards the sample was quantitatively transferred into a Teflon dish and placed on a heating block to remove the HCl. To achieve this, the sample volume was evaporated close to dryness at 230 °C (residual volume 2–3 mL), then 10 mL 1 % (w/w) aq. HNO3 was added and the sample was again evaporated close to dryness. This step was repeated twice. Finally, the sample was quantitatively transferred to a 15 mL centrifuge tube and filled up to the 10 mL mark with 1 % (w/w) aq. HNO3. For ICP-qMS analysis, the sample was further diluted by a factor of 50 with 1 % (w/w) aq. HNO3.

The algae samples were prepared similarly; however, dry ashing was omitted because the limited amounts of sample material meant there was a risk of severe losses during dry ashing and subsequent transfer into HPA digestion vessels. Thus, direct acid digestion by HPA was performed with approximately 80–170 mg freeze-dried sample material. Digested samples were analysed by ICP-qMS after appropriate dilution with 1 % (w/w) aq. HNO3.

In addition, because of the modified sample preparation of the algae, the analysis of NCS DC 73349 was repeated as described above but without dry ashing.

Method calibration

ICP-qMS calibration solutions were freshly prepared on each day of measurement from a certified Ti single-element standard solution in 1 % (w/w) aq. HNO3. Concentration levels were 5 μg L−1, 10 μg L−1, 25 μg L−1, 50 μg L−1, 100 μg L−1, 250 μg L−1, and 500 μg L−1. Each level and an additional blank standard (1 % w/w aq. HNO3) were measured in single analysis. The limit of detection (LOD) and the limit of quantification (LOQ) were estimated according to DIN 32645 [19].

Instrumental set-up and operating conditions of the ICP-qMS system

The measurements were performed using a Varian 820-MS (Varian, Mulgrave, Australia) equipped with a Varian SPS-3 autosampler (Varian, now part of Bruker Daltonics, Fremont, CA, USA). Sample introduction was performed with a peristaltic pump. Data acquisition and instrumental operating control were performed using the ICP-MS Expert software from Varian. Detailed instrumental set-up, operating conditions, and data acquisition parameters are summarized in Table 1. Calibration was performed with 4 μg L−1 In in 1 % (w/w) aq. HNO3 as internal standard, to correct possible instrumental drifts. The In standard was continuously introduced into the instrument, together with each external calibration standard or sample, via a plastic mixing T-connector. Peristaltic pump tubes were chosen to achieve a mixing ratio of 1:1 for In standard and calibration standard or sample, resulting in a dilution factor of two for external calibration standards and samples. After each nine analyses, two blank standard measurements (1 % w/w aq. HNO3) were performed to control for possible saturation effects within the instrumentation set-up.
Table 1

ICP-qMS set-up, operating conditions, and data acquisition parameters for titanium analysis

Instrumental configuration and conditions

 Nebuliser

MicroMist, glass concentric (sample uptake: 0.4 mL min−1; gas flow: 1 L min−1, with EzyFit connector)

 Spray chamber

Double-pass Scott-type quartz spray chamber

 Torch

One-piece quartz torch with ball connection, 2.4 mm injector

 Sampler cone

Nickel, 1 mm orifice diameter

 Skimmer cone

Nickel, 0.5 mm orifice diameter

 Sample uptake (mL min−1)

0.19

 Sample uptake time (s)

40

 Rinse time 1 % (w/w) aq. HNO3 (s)

15

 Fast pump during sample uptake and rinse

On

Plasma and nebuliser conditions

 RF Generator frequency (MHz)

27.12

 RF power (kW)

1.4

 Plasma sampling depth (mm)

6.5

 Argon plasma gas flow (L min−1)

16.8

 Auxiliary gas flow (L min−1)

1.63

 Sheath gas flow (L min−1)

0.16

 Nebuliser gas flow (L min−1)

0.94

 Spray chamber temperature (°C)

3 (Peltier-cooled device)

 Stabilization delay (s)

60

Data acquisition

 Isotopes monitored

47Ti+, 48Ti+, 49Ti+, 44Ca+, 115In+

 Dwell time/amu (ms)

47Ti+: 50, 48Ti+: 10, 49Ti+: 50, 44Ca+: 10, 115In+: 50

 Scan mode

Peak hopping

 Mass resolution high/low isotope (amu)

0.80

 Number of replicates per sample

6

 Number of scans per replicate

25

 Points per peak

1

Results and discussion

Performance of ICP-qMS titanium analysis of certified vegetable reference material after dry ashing and high-pressure ashing sample preparation

On the basis of the calibration curve settings used for the isotopes 48Ti and 49Ti (Table 2), the linearity of isotope 47Ti calibration (data not shown) was insufficient. Therefore, quantification of the certified vegetable reference material NCS DC 73349 was performed measuring the isotopes 48Ti and 49Ti. Nevertheless, with altered calibration curve settings for isotope 47Ti, good linearity and recovery results could be achieved for analysis of Ti in water samples (unpublished results). The calibration for 48Ti and 49Ti revealed good linearity in the specified concentration range of 5–500 μg L−1 (for 48Ti, the concentration of 5 μg L−1 was excluded as an outlier). The difference between standard and calculated concentration was smaller than 5 % for each level. Calibration curve data, with LOD and LOQ values for 48Ti and 49Ti, are summarized in Table 2.
Table 2

Titanium calibration curve data measuring 48Ti and 49Ti by ICP-qMS

Isotope

Calibration equation

R 2

Calibration curve settings

LODa

LOQb

48Ti

y = (483.8 + 4020x)*[I/S ratioc]

0.9999

Curve fit: linear,

0.07 μg L−1

0.26 μg L−1

Weighted fit: No,

Fit through blank: Yes

49Ti

y = (446.7 + 309.6x)*[I/S ratioc]

0.9999

Curve fit: linear,

0.16 μg L−1

0.57 μg L−1

Weighted fit: No,

Fit through blank: Yes

48Ti: six calibration points with blank standard; n=1; single measurement; calibration ranged 10–500 μg L–1; y=counts/second; x=concentration. 49Ti: seven calibration points with blank standard; n=1; single measurement; calibration ranged 5–500 μg L–1; y=counts/second; x=concentration

aLimit of detection, details in section Method calibration

bLimit of quantification, details in section Method calibration

cInternal standard ratio, details in section Instrumental set-up and operating conditions of the ICP-qMS system

dDetails in section Method calibration and Instrumental set-up and operating conditions of the ICP-qMS system

Recovery and repeatability were calculated from five independent sample digestions of the certified reference material NCS DC 73349 (Table 3). Each sample was measured in triplicate. The recovery for 48Ti ranged from 102 % to 143 %, with a mean concentration of 114 ± 16 μg g−1, which is within the standard deviation of the certified value (95 ± 20 μg g−1 Ti). The recovery for 49Ti ranged from 51 % to 85 % and revealed a mean concentration of 63 ± 13 μg g−1, which was still acceptable compared with the range of the certified value. One explanation for the discrepancy between recovery results of the 48Ti and 49Ti isotopes might be a possible difference in the natural isotopic abundance of the certified single-element Ti standard solutions used for external calibration and of the certified vegetable reference material NCS DC 73349 selected for method evaluation. Particularly, different relative abundances of the low-abundance 49Ti (5.41 % [20]) in the single-element standard and the certified reference material might substantially affect the quantification, resulting in a calibration curve unsuitable for the certified reference material NCS DC 73349.
Table 3

Recovery and repeatability data for titanium (Ti), measuring 48Ti and 49Ti for certified reference material NSC DC 73349 by ICP-qMS

Isotope

NSC DC 73349 repetition:

 

Certified value of Tia,

1

2

3

4

5

Mean ± SD, n = 5

Mean ± SD

48Ti

 Conc. (μg g−1)

101 ± 1.4

114 ± 1.3

97 ± 0.3

136 ± 1.3

124 ± 0.5

114 ± 16

95 ± 20

 Recovery (%)

106 ± 1.4

120 ± 1.4

102 ± 0.3

143 ± 1.3

130 ± 0.6

120 ± 17

/

49Ti

 Conc. (μg g−1)

49 ± 1.0

61 ± 0.3

54 ± 0.4

81 ± 0.7

69 ± 0.2

63 ± 13

95 ± 20

 Recovery (%)

51 ± 1.1

64 ± 0.3

57 ± 0.5

85 ± 0.8

73 ± 0.2

66 ± 13

/

Five sample repetitions, i.e. independent sample digestions; each sample measured in triplicate. Repeatability is expressed as standard deviation (SD) of the mean

aAccording to the certificate of analysis, the value was obtained by the following analytical methods: atomic absorption spectrometry, colorimetry, inductively coupled plasma-emission spectrometry, and X-ray fluorescence spectrometry

The repeatability measurements revealed an acceptable analytical precision for 48Ti, with a relative standard deviation (RSD) of 14 %. The RSD for 49Ti was 21 %, which can be regarded as sufficient to enable use of the method, at least for the environmental analysis described in this work. Each triplicate analysis revealed a good analytical precision, with an RSD below 5 %. All scan replicates (n = 6) by ICP-qMS for external calibration standards and certified reference material NCS DC 73349 also had an RSD smaller than 5 %.

The certified reference material NCS DC 73349 contains 1.68 ± 0.11 % calcium (16.8 ± 1.1 mg g−1 Ca, certified value), which might interfere with 48Ti [21]. Good recoveries of isotope 48Ti reveal that the ICP-MS Expert software from Varian correctly calculated the concentration of 48Ti, by subtracting the percentage caused by mass overlap with the 48Ca isotope (isobaric interference) using the following equation:
$$ {}^{48}{\mathrm{Ti}}_{\mathrm{corrected}c/s\left(\mathrm{mean}\right)}{=}^{48}{\mathrm{Ti}}_{\mathrm{measured}c/s\left(\mathrm{mean}\right)}\hbox{--} 0.08965{\times}^{44}{\mathrm{Ca}}_{\mathrm{measured}c/s\left(\mathrm{mean}\right)} $$
where c/s is counts per second, and 0.08965 is the ratio of the relative isotopic abundance of 48Ca (0.187 % [20]) and 44Ca (2.086 % [20]).

Performance of ICP-qMS titanium analysis of certified vegetable reference material and algae samples, after high-pressure ashing sample preparation

Method application to certified reference material NCS DC 73349

To evaluate the accuracy of the modified analytical method without dry ashing, the bush, branches, and leaves certified reference material NCS DC 73349 was analysed again. Recovery and repeatability were calculated from three independent sample digestions (Table 4). Each sample was measured in triplicate. The RSD of the repeatability was 10 % and 13 % for 48Ti and 49Ti, respectively. Each triplicate analysis revealed good analytical precision, with RSD smaller than 5 %. All scan replicates (n = 6) by ICP-qMS for external calibration standards, certified reference material NCS DC 73349, and algae samples also had an RSD smaller than 5 %.
Table 4

Recovery and repeatability data for titanium (Ti), without dry ashing procedure, measuring 48Ti and 49Ti for certified reference material NSC DC 73349 by ICP-qMS

Isotope

NSC DC 73349 repetition:

 

Certified value of Tia,

 

1

2

3

Mean ± SD, n = 3

Mean ± SD

48Ti

 Conc. (μg g−1)

194 ± 1.4

159 ± 2.0

184 ± 1.5

179 ± 18

95 ± 20

 Recovery (%)

204 ± 1.4

168 ± 2.1

194 ± 1.6

189 ± 19

/

49Ti

 Conc. (μg g−1)

132 ± 1.4

103 ± 0.7

131 ± 1.9

122 ± 16

95 ± 20

 Recovery (%)

139 ± 1.4

108 ± 0.7

138 ± 2.0

128 ± 17

/

Three sample repetitions, i.e. independent sample digestions; each sample measured in triplicate. Repeatability is expressed as standard deviation (SD) of the mean

aAccording to the certificate of analysis the value was obtained by the following analytical methods: atomic absorption spectrometry, colorimetry, inductively coupled plasma-emission spectrometry, and X-ray fluorescence spectrometry

As can be seen from Table 4, recovery for 48Ti was not within the acceptable range. The increased mean value of 189 ± 19 % (n = 3) might be caused by the presence of sulfur (S), which can form a 32S16O+ oxide ion [21, 22, 23, 24] and 33S15N+ and 34S14N+ interfering species [21]. The natural element S comprises four stable isotopes (32S, 33S, 34S, and 36S), of which 32S is the most abundant (94.99 %) [20]. The certified reference material NCS DC 73349 contains 0.73 ± 0.06 % (certified value) S. This value is approximately 75-fold higher than the certified value of Ti; thus, resulting interferences might strongly affect Ti measurements. Furthermore, the certified reference material NCS DC 73349 contains 1000 ± 40 μg g−1 phosphorus (certified value, P). Therefore, formation of a 31P17O+ oxide ion is also conceivable; however, because of the low abundance of 17O, this would be a minor interfering species. With a sample preparation step of dry ashing at 475 °C (boiling point of S = 445 °C; of P = 280 °C), S and P are evaporated effectively. This is confirmed by the good recovery obtained by the analytical procedure using dry ashing. However, without dry ashing, the S and P content of the matrix might interfere with the 48Ti measurement and thus explain the recoveries exceeding 100 %.

The apparently slightly increased mean recovery of 128 ± 17 % (n = 3) of isotope 49Ti seems to be an acceptable result. However, when considering the mean concentration of 63 ± 13 μg g−1 (n = 5) after dry ashing, the difference seems remarkable. The reason for this discrepancy might again be the S isotopes, 32S, 33S, and 34S, and the element P, forming 32S17O+ [24], 33S16O+ [25], 32S16O1H+ [23, 24], 34S15N+ [21], and 31P18O+ [16] and thus interfering with 49Ti when the analytical procedure is used without dry ashing. Although these are minor interfering species, the percentage is very high in relation to the measured counts per second of the low-abundance 49Ti isotope (5.41 % relative isotopic abundance [20]). Thus, in summary, analysis of the certified reference material NCS DC 77349 cannot be performed by measuring the isotopes 48Ti and 49Ti after this sample preparation procedure.

Method application to microalgae Scenedesmus acutus

To analyse the microalgae Scenedesmus acutus, both 48Ti and 49Ti were used for quantification. Respective results are summarized in Tables 5 and 6. To assess the uptake of nTiO2 by Scenedesmus acutus, 48Ti and 49Ti results were calculated as nTiO2. In addition, to compare nTiO2 concentrations of the three nTiO2 treatment groups (0.3 mg L−1; 1.2 mg L−1; 4.8 mg L−1; n = 3 each) and the control group, the findings were normalized to dry weight. These results reveal a good correlation between adsorption of nTiO2 by Scenedesmus acutus and the used concentrations of Aeroxide TiO2 P25. Furthermore, using both isotopes 48Ti and 49Ti for calculation of the nTiO2 concentration resulted in nearly the same mean values: the background Ti concentration (control group), calculated as nTiO2, was 228 ± 100 mg kg−1 (mean value of n = 3) when performing the analysis with isotope 48Ti, and 217 ± 79 mg kg−1 (n = 3) using isotope 49Ti.
Table 5

Nanoscale titanium dioxide (nTiO2) concentrations for microalgae Scenedesmus acutus, related to dry weight, calculated from mass scan 48Ti using ICP-qMS

Algae suspension:

nTiO2 treatment groups

Sample repetition:

nTiO2 concentration in algae (dry weight) calculated from 48Ti (mg kg−1)

1

2

3

Mean ± SDa, n = 3

Control groupb

140

207

337

228 ± 100

(1.6)c

(0.2)c

(5.8)c

 

0.3 mg L−1

654

420

395

490 ± 143

(1.7)c

(1.2)c

(4.1)c

 

1.2 mg L−1

1716

2201

1225

1714 ± 488

(22.4)c

(21.2)c

(1.1)c

 

4.8 mg L−1

2839

9097

5168

5701 ± 3163

(25.2)c

(30.8)c

(13.0)c

 

Three sample repetitions; each sample measured in duplicate

aStandard deviation

bGrown without Aeroxide TiO2 P25

cConcentration difference of duplicate analysis

Table 6

Nanoscale titanium dioxide (nTiO2) concentrations for microalgae Scenedesmus acutus, related to dry weight, calculated from mass scan 49Ti using ICP-qMS

Algae suspension:

nTiO2 treatment groups

Sample repetition:

nTiO2 concentration in algae (dry weight) calculated from 49Ti (mg kg−1)

1

2

3

Mean ± SDa, n = 3

Control groupb

144

206

302

217 ± 79

(2.1)c

(2.1)c

(0.1)c

 

0.3 mg L−1

625

425

395

482 ± 125

(0.5)c

(5.3)c

(1.9)c

 

1.2 mg L−1

1677

2201

1207

1695 ± 497

(13.8)c

(4.3)c

(0.4)c

 

4.8 mg L−1

2834

9112

5174

5707 ± 3173

(52.9)c

(45.6)c

(67.7)c

 

Three sample repetitions; each sample measured in duplicate

aStandard deviation

bGrown without Aeroxide TiO2 P25

cConcentration difference of duplicate analysis

A possible reason for this relatively high background Ti concentration might be the laboratory and experimental environment, e.g. glassware contact during the growing of the algae, or the reagents used. For example, HNO3 69 % and HCl 35 % (supra-grade quality) for HPA digestion contained Ti at trace level (< 1 μg L−1), which most possibly remains in the sample material after the evaporation procedure. Since the calibration does not account for the acid digestion treatment, this might lead to the detection of additional Ti. Furthermore, the formation of S and P oxide ions and SN ions in the algae Scenedesmus acutus is also conceivable: the algae medium and the M4-Medium consists of several salts, including magnesium sulfate and potassium dihydrogen phosphate (OECD 2008, 2011 [17, 18]). Moreover, the algae itself contains P: Kylin [26] reported a P concentration in normal Scenedesmus acutus cells of 3–5 mg g−1 fresh weight.

Conclusion

Using certified vegetable reference materials, for example NCS DC 73349 (bush, branches, and leaves matrix), to evaluate the accuracy of a method for analysis of Ti at trace levels in biological material can be critical, because of the formation of S and P oxide ions and SN ions. These interferences might occur with any sample containing S and P. However, in this study S and P were eliminated via evaporation by a dry ashing procedure. The described method of sample preparation by dry ashing and subsequent high-pressure ashing before ICP-qMS analysis obtained good, or at least acceptable, recovery and an appropriate repeatability for the measurement of 48Ti and 49Ti isotopes in the certified reference material NCS DC 73349. Alternatively, the problem of S and P cluster-ion interferences could also be solved through introduction of helium collision gas, injected through the skimmer cone into the argon plasma to destroy the S and P cluster ions that interfere with the Ti measurements.

For the application of measuring Ti in algae samples, the procedure for sample preparation had to be modified because of limited sample amounts. The respective method without dry ashing was also evaluated by the use of the certified reference material NCS DC 73349. The results confirmed the assumption that dry ashing is a critical step in obtaining correct values for Ti concentrations in the presence of S and P, because without dry ashing the recoveries were too high. The use of the modified method for sample preparation with the microalgae Scenedesmus acutus provided accurate results, indicated by equal values for quantifications on the basis of the different Ti isotopes 48Ti and 49Ti. Moreover, if cluster ion interferences were present they would be identified during measurement of the control group.

When using biological reference material, including NCS DC 73349, one should be aware that the elemental composition might differ substantially from the composition of the analysed sample material (in this case Scenedesmus acutus). Despite this difference, the use of the certified reference material NCS DC 73349 was of importance to discover the S and P cluster ion interferences with the 48Ti and 49Ti ions. It might thus be worth trying to apply an ICP-qMS method to specific sample material even if the method is not applicable to the reference material, as long as the accuracy of the results can be evaluated by another method without use of said reference material.

Notes

Acknowledgments

We would like to thank Evonik Industries for providing Aeroxide TiO2 P25 and the OECD for participation in the OECD sponsorship programme.

References

  1. 1.
    Steinbach C, Förster A Knowledge base nanomaterials, DaNa, Frankfurt am Main. URL: http://www.nanoobjects.info/cms/lang/en/Wissensbasis/Titandioxid
  2. 2.
    Landsiedel R, Ma-Hock L, Kroll A, Hahn D, Schnekenburger J, Wiench K, Wohlleben W (2010) Testing metal-oxide nanomaterials for human safety. Adv Mater 22:2601–2627CrossRefGoogle Scholar
  3. 3.
    Weir A, Westerhoff P, Fabricius L, Hristovski K, Goetz N (2012) Titanium dioxide nanoparticles in food and personal care products. Environ Sci Technol 46:2242–2250CrossRefGoogle Scholar
  4. 4.
    Kiser MA, Westerhoff P, Benn T, Wang Y, Pérez-Rivera J, Hristovski K (2009) Titanium nanomaterial removal and release from wastewater treatment plants. Environ Sci Technol 43:6757–6763CrossRefGoogle Scholar
  5. 5.
    Westerhoff P, Song G, Hristovski K, Kiser MA (2011) Occurrence and removal of titanium at full scale wastewater treatment plants: implications for TiO2 nanomaterials. J Environ Monit 13:1195–1203CrossRefGoogle Scholar
  6. 6.
    Menard A, Drobne D, Jemec A (2011) Ecotoxicity of nanosized TiO2. Review of in vivo data. Environ Pollut 159:677–684CrossRefGoogle Scholar
  7. 7.
    Luque-Garcia JL, Sanchez-Díaz R, Lopez-Heras I, Martin P, Camara C (2013) Bioanalytical strategies for in-vitro and in-vivo evaluation of the toxicity induced by metallic nanoparticles. Trends Anal Chem 43:254–268CrossRefGoogle Scholar
  8. 8.
    Navarro E, Baun A, Behra R, Hartmann NB, Filser J, Miao A-J, Quigg A, Santschi PH, Sigg L (2008) Environmental behavior and ecotoxicity of engineered nanoparticles to algae, plants and fungi. Ecotoxicology 17:372–386CrossRefGoogle Scholar
  9. 9.
    Lorenz C, Tiede K, Tear S, Boxall A, von Goetz N, Hungerbühler K (2010) Imaging and characterization of engineered nanoparticles in suncreens by electron microscopy, under wet and dry conditions. Int J Occup Environ Health 16:406–428CrossRefGoogle Scholar
  10. 10.
    Johnston BD, Scown TM, Moger J, Cumberland SA, Baalousha M, Linge K, van Aerle R, Jarvis K, Lead JR, Tyler CR (2010) Bioavailability of nanoscale metal oxides TiO2, CeO2 and ZnO to fish. Environ Sci Technol 44:1144–1151CrossRefGoogle Scholar
  11. 11.
    Warheit DB, Sayes CM, Reed KL, Swain KA (2008) Health effects related to nanoparticle exposures: Enivronmetal, health and safety considerations for assessing hazards and risks. Pharmacol Therapeut 120:35–42CrossRefGoogle Scholar
  12. 12.
    Foltête A-S, Masfaraud J-F, Bigorgne E, Nahmani J, Chaurand P, Botta C, Labille J, Rose J, Férard J-F, Cotelle S (2011) Environmental impact of sunscreen nanomaterials: ecotoxicity and genotoxicity of altered TiO2 nanocomposites on Vicia faba. Environ Pollut 159:2515–2522CrossRefGoogle Scholar
  13. 13.
    Lomer MCE, Thompson RPH, Commisso J, Keen CL, Powell JJ (2000) Determination of titanium dioxide in foods using inductively coupled plasma optical emission spectrometry. Analyst 125:2339–2343CrossRefGoogle Scholar
  14. 14.
    Packer AP, Larivière D, Li C, Chen M, Fawcett A, Nielsen K, Mattson K, Chatt A, Scriver C, Erhardt LS (2007) Validation of an inductively coupled plasma mass spectrometry (ICP-MS) method for the determination of cerium, strontium, and titanium in ceramic materials used in radiological dispersal devices (RDDs). Anal Chim Acta 588:166–172CrossRefGoogle Scholar
  15. 15.
    Umysová D, Vítová M, Doušková I, Bišová K, Hlavová M, Čížková M, Machát J, Doucha J, Zachleder V (2009) Bioaccumulation and toxicity of selenium compounds in the green alga Scenedesmus quadricauda. BMC Plant Biol 9:58CrossRefGoogle Scholar
  16. 16.
    Yu L, Koirtyohann SR, Rueppel ML, Skipor AK, Jacobs JJ (1997) Simultaneous determination of aluminium, titanium and vanadium in serum by electrothermal vaporization-inductively coupled plasma mass spectrometry. J Anal At Spectrom 12:69–74CrossRefGoogle Scholar
  17. 17.
    Organisation for Economic Cooperation and Development (OECD) (2011) Guidelines for the testing of chemicals. Freshwater Alga and Cyanobacteria, Growth Inhibition Test (Guideline 201, Annex 5 corrected: 28 July 2011). OECD, ParisGoogle Scholar
  18. 18.
    Organisation for Economic Cooperation and Development (OECD) (2008) Guidelines for the testing of chemicals. Daphnia magna reproduction test (Guideline 211, adopted 3st October 2008). OECD, ParisGoogle Scholar
  19. 19.
    DIN 32645 (2008) Chemical analysis - Decision limit, detection limit and determination limit under repeatability conditions - Terms, methods, evaluation. Deutsches Institut für Normung. Beuth Verlag, BerlinGoogle Scholar
  20. 20.
    Berglund M, Wieser ME (2011) Isotopic compositions of the elements 2009 (IUPAC Technical Report). Pure Appl Chem 83:397–410CrossRefGoogle Scholar
  21. 21.
    Vaughan M-A, Baines AD, Templeton DM (1991) Multielement analysis of biological samples by inductively coupled plasma-mass spectrometry. II. Rapid survey method for profiling trace elements in body fluids. Clin Chem 37:210–215Google Scholar
  22. 22.
    Luck J, Siewers U (1988) Progress in analytical application of an inductively coupled plasma/mass spectrometer system. Fresenius’ Z Anal Chem 331:129–132CrossRefGoogle Scholar
  23. 23.
    Reed NM, Cairns RO, Hutton RC (1994) Characterization of polyatomic ion interferences in inductively coupled plasma mass spectrometry using a high resolution mass spectrometer. J Anal At Spectrom 9:881–896CrossRefGoogle Scholar
  24. 24.
    Marshall J, Franks J (1990) Multielement analysis and reduction of spectral interferences using electrothermal vaporization inductively coupled plasma-mass spectrometry. At Spectrosc 11:177–186Google Scholar
  25. 25.
    Nonose N, Ohata M, Narukawa T, Hioki A, Chiba K (2009) Removal of isobaric interferences in isotope dilution analysis of vanadium in silicon nitride fine ceramics powder by DRC-ICP-MS. J Anal At Spectrom 24:310–319CrossRefGoogle Scholar
  26. 26.
    Kylin A (1964) The influence of phosphate nutrition on growth and sulfur metabolism of Scenedesmus. Physiol Plant 17:384–402CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Theodoros Potouridis
    • 1
  • Johannes Völker
    • 2
  • Heiko Alsenz
    • 1
  • Matthias Oetken
    • 2
  • Wilhelm Püttmann
    • 1
  1. 1.Institute for Atmospheric and Environmental Sciences, Department of Environmental Analytical ChemistryJ.W. Goethe-University Frankfurt am MainFrankfurt am MainGermany
  2. 2.Department Aquatic EcotoxicologyJ.W. Goethe-University Frankfurt am MainFrankfurt am MainGermany

Personalised recommendations