Laser Ionization Orthogonal Time-of-Flight Mass Spectrometry for

Anal. Chem. 2010, 82, 1949–1953

Laser Ionization Orthogonal Time-of-Flight Mass Spectrometry for Simultaneous Determination of Nonmetallic Elements in Solids Lingfeng Li,† Bochao Zhang,† Rongfu Huang,† Wei Hang,*,†,‡ Jian He,§ and Benli Huang† Key Laboratory of Analytical Science, Department of Chemistry, College of Chemistry and Chemical Engineering, State Key Laboratory of Marine Environmental Science, and Department of Mechanical and Electrical Engineering, Xiamen University, Xiamen 361005, China The simultaneous determination of nonmetallic elements in solid samples is difficult owing to their discrepant physical and chemical properties. We developed a highirradiance laser ionization orthogonal time-of-flight mass spectrometry (LI-O-TOFMS) system and applied it for the determination of nonmetallic elements in solids. Helium was used as the buffer gas at 250 Pa in the source chamber; the laser irradiance was about 7 × 1010 W/cm2. A series of artificial standards containing B, C, N, O, F, Si, P, S, Cl, As, Br, Se, I, and Te were used. Explicit spectra were obtained with only a little interference from gas species and doubly charged matrix ions. Standardless semiquantitative analysis could be accomplished with a novel sampling methodology to obtain near-uniform sensitivity coefficients for different elements. Limits of detection (LOD) at microgram per gram level and a dynamic range of 6 orders of magnitude were achieved for most nonmetallic elements. Great progress has recently been made in the development of atomic spectroscopic techniques.1,2 However, the determination of nonmetallic elements still remains difficult. Most nonmetallic elements are still determined by traditional chemical methods or by single-element analyzers using gravimetry, colorimetry, or spectrophotometry techniques. The determination of nonmetallic elements is of great importance in materials, biological, environmental, and geological research. Most conventional methods used today, which can easily detect metallic elements, are unable to determine some nonmetallic elements. Due to their relatively high ionization potential, most nonmetallic elements need to absorb a large amount of energy for detection using absorption methods, or these elements must be excited or ionized at high temperatures for detection using emission spectroscopy or mass spectrometry. The commonly used atomic absorption spectrometry (AAS) has traditionally been * To whom correspondence should be addressed. E-mail: weihang@ xmu.edu.cn. † College of Chemistry and Chemical Engineering. ‡ State Key Laboratory of Marine Environmental Science. § Department of Mechanical and Electrical Engineering. (1) Broekaert, J. A. C. Nature 2008, 455, 1185–1186. (2) Staack, D.; Fridman, A.; Gutsol, A.; Gogotsi, Y.; Friedman, G. Angew. Chem., Int. Ed. 2008, 47, 8020–8024. 10.1021/ac9026912  2010 American Chemical Society Published on Web 02/04/2010

restricted to the determination of metals and metalloids3,4 because of the lack of suitable light sources to provide the vacuum ultraviolet light needed for the detection of nonmetals. Popular inductively coupled plasma mass spectrometry (ICPMS) and glow discharge mass spectrometry (GDMS) techniques are used successfully for metal element determination, but these techniques cannot always detect nonmetallic elements because of low source temperatures and severe spectral interference problems.5,6 Another widely used technique for elemental analysis, X-ray fluorescence (XRF), is a nondestructive analytical method, but it is usually insensitive to light elements.7 Laser-produced plasma can reach a temperature of 20 000-50 000 K8 at which most nonmetallic elements can be excited and ionized. Laser-induced breakdown spectroscopy (LIBS) is capable of simultaneous multielemental analysis with no special sample preparation.9,10 However, it is difficult to detect nonmetallic elements owing to the fact that most of the strongest emission lines of nonmetallic elements lie in the vacuum ultraviolet region (125-190 nm).11 In comparison with LIBS, a laser plasma source coupled to a mass spectrometer provides simple spectra and avoids the difficulty of acquiring the ultraviolet emission lines. Traditional laser ionization mass spectrometers with on-axis geometry (such as the LAMMA-1000 from Leybold-Heraeus and LIMA-2A from Cambridge Mass Spectrometry Ltd.) can be used for multielemental analysis, but the use of this technique has been limited in qualitative analysis because the large kinetic energy distribution of the ions would lead to a very poor mass resolution.12,13 Herein, we present a laser ionization orthogonal time-of-flight mass (3) Bencs, L.; Szakacs, O.; Kantor, T.; Varga, I.; Bozsai, G. Spectrochim. Acta, Part B 2000, 55B, 883–891. (4) Zhang, Y.; Adeloju, S. B. Talanta 2008, 76, 724–730. (5) Broekaert, J. A. C. Fresenius’ J. Anal. Chem. 2000, 368, 15–22. (6) Harrison, W. W.; Yang, C.; Oxley, E. In Glow Discharge Plasmas in Analytical Spectroscopy; Marcus, R. K., Broekaert, J. A. C., Eds.; Wiley: New York, 2003; pp 71-96. (7) Beckhoff, B.; Langhoff, N.; Kanngießer, B.; Wedell, R.; Wolff, H. Handbook of Practical X-Ray Fluorescence Analysis; Springer: Berlin, Germany, 2006. (8) Bogaerts, A.; Chen, Z. Y.; Bleiner, D. J. Anal. At. Spectrom. 2006, 21, 384– 395. (9) Gornushkin, I. B.; Mueller, M.; Panne, U.; Winefordner, J. D. Appl. Spectrosc. 2008, 62, 542–553. (10) Tsai, S. J. J.; Chen, S. Y.; Chung, Y. S.; Tseng, P. C. Anal. Chem. 2006, 78, 7432–7439. (11) Asimellis, G.; Giannoudakos, A.; Kompitsas, M. Anal. Bioanal Chem. 2006, 385, 333–337. (12) Becker, J. S.; Dietze, H. J. Fresenius’ J. Anal. Chem. 1992, 344, 69–86. (13) Van Vaeck, L.; Van Roy, W.; Gijbels, R. Analusis 1993, 21, 53–75.

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1949

Figure 1. Schematic diagram of the LI-O-TOFMS.

spectrometer (LI-O-TOFMS) capable of rapid multielemental analysis with minimal sample preparation. Unique features of the instrument include a buffer-gas-assisted source and an orthogonal TOF geometry. In contrast to traditional on-axis geometry, orthogonal TOF sharply enhances the resolving power. The use of the buffer gas in the ion source cools the energetic ions from the plasma and suppresses multiply charged ions through collisional recombination. The pressure used for nonmetallic elements determination was lower than that for metallic element experiments.14 Though a relatively lower pressure brings the multiply charged ions rising slightly, it results in a higher electron temperature, which is beneficial for the ionization of nonmetallic elements. As a result, this technique can be used to acquire spectra that have lower background, lower interference, and higher resolution than spectra obtained using other conventional methods. EXPERIMENTAL SECTION The schematic setup of the LI-O-TOFMS system is shown in Figure 1. It consists of three vacuum stages: ion source stage (before the nozzle), transportation stage (between the nozzle and the slit), and analyzer stage (after the slit). A frequency-doubled (532 nm) Nd:YAG laser (NL303G, EKSPLA) with 4.4 ns pulse duration and 10 Hz repetition rate was employed. The laser beam was focused by a laser-focusing objective (LMU-3X, OFR division of Thorlabs Inc.) onto a spot of 20 µm in diameter. The focal spot was positioned approximately 6 mm from the nozzle tip in the X direction and 8 mm in the Y direction. The typical laser irradiation flux was 7 × 1010 W/cm2. High-purity helium (99.999%) was used as the buffer gas. A needle valve was used to control the pressure of the source chamber, which was optimized at 250 Pa. The signal was acquired using a digital storage oscilloscope (Lecroy, 42Xs) with a 2.5 Gs/s sampling rate in all experiment. The data were processed by an in-house-compiled program written in LabVIEW 8.0 (National Instruments Inc.). Other details about the system have been described previously.14,15 Because nonmetallic elements are difficult to be determined accurately, most of the commercial certified references do not (14) Yu, Q.; Huang, R.; Li, L.; Lin, L.; Hang, W.; He, J.; Huang, B. Anal. Chem. 2009, 81, 4343–4348. (15) He, J.; Huang, R.; Yu, Q.; Lin, Y.; Hang, W.; Huang, B. J. Mass Spectrom. 2009, 44, 780–785.

1950

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

provide their concentrations. Even though some standards contain certified amount of nonmetallic elements, the variety of nonmetallic elements is very limited. To demonstrate the simultaneous multielemental capability of this system, a series of artificial standards containing B, C, N, O, F, Si, P, S, Cl, As, Br, Se, I, and Te elements were prepared. The samples were made using the ultrahigh-purity copper powder (99.99999%, Huaxi Institute for High-Purity Materials, certified by GDMS) as the matrix mixed with pure NaBF4, NaCl, KBr, KIO3, Li2CO3, NH4H2PO4, As4S4, SiO2, Se, and TeO2 powders (analytical grade, Sinopharm Chemical Reagent Co., Ltd.). The compounds were blended and crushed in an agate mortar for approximately 15 min. The low-concentration samples were made by diluting the highconcentration samples with the copper powder. The samples were loaded separately into a die and pressed by a hydraulic press machine with a pressure of 1 × 108 Pa for 5 min, forming discs with thickness of 1.5 mm and diameter of 8 mm. RESULTS AND DISCUSSION It has been demonstrated previously that the system is capable of qualitative and quantitative analysis of solid samples for metallic elements; even standardless semiquantitative analysis is possible.14 However, owing to the different physical and chemical properties of nonmetallic elements, further research is required. We investigated the differences between metallic and nonmetallic ion packets that are indicated by their temporal profiles in the repelling region. As shown in Figure 2, ion packets of nonmetallic elements have much shorter profiles than those of metallic elements. The ionization process for nonmetallic elements stops quickly after the laser shot because of the rapid drop in the plasma temperature and the high ionization potentials of nonmetallic elements. On the other hand, due to the presence of the buffer gas, frequent recombinations take place through collisional processes (mainly three-body, possibly charge-exchange or radiative two-body collisions) in the plasma. Recombination rates of nonmetallic ions are higher than those of metallic ions;16 thus, nonmetallic ions are preferentially neutralized. As a consequence, the duration of nonmetallic ions is less than that of metallic ions. The above result indicates that the sampling scheme for nonmetallic elements should be different from the previous (16) Namba, S.; Nozu, R.; Takiyama, K.; Oda, T. J. Appl. Phys. 2006, 99, 0733021–073302-9.

Figure 2. Temporal profile of 39K+ and 35Cl+ for a pure KCl sample.

Table 1. Summary of the RSCs, the Correlation Coefficients (R2) of the Calibration Curves, and the Limits of Detection (LODs) element

RSC (60 µs)a

RSC (200 µs)a

R2

LOD (µg/g)

Li B C N O F Na Si P S Cl K As Br Se I Te

0.18 ± 0.02 0.16 ± 0.02 0.19 ± 0.09 0.56 ± 0.13 0.35 ± 0.07 0.21 ± 0.03 1.00 ± 0.00 0.81 ± 0.12 0.47 ± 0.09 0.51 ± 0.07 0.38 ± 0.08 1.22 ± 0.19 0.54 ± 0.14 0.35 ± 0.05 0.55 ± 0.07 0.23 ± 0.07 0.31 ± 0.11

0.08 ± 0.01 0.03 ± 0.004 0.04 ± 0.02 0.12 ± 0.03 0.07 ± 0.01 0.04 ± 0.006 1.00 ± 0.00 0.20 ± 0.03 0.10 ± 0.02 0.22 ± 0.06 0.08 ± 0.02 3.02 ± 0.43 0.36 ± 0.08 0.34 ± 0.05 0.42 ± 0.06 0.54 ± 0.10 0.62 ± 0.13

0.9986 0.9947 0.9929b 0.9938b 0.9948b 0.9983 0.9806 0.9895 0.9945 0.9998 0.9962 0.9961 0.9984 0.9988 0.9993 0.9995 0.9917

0.2 0.5 185 974 469 5.7 0.1 12.6 1.6 9.8 0.4 0.2 0.5 3.0 2.0 2.3 4.9

a Signal accumulation time after the laser shot. concentration above 1500 µg/g.

b

Figure 3. (a) Spectrum constructed from the composition of the sample, assuming RSCs are equal for all elements; (b) spectrum from the experiment using the LI-O-TOFMS.

For elemental

approach.14,15 The relative sensitivity coefficient (RSC) is a good indicator of the atomization, ionization, and detection efficiency of each element.15 RSCs were determined using the following equation: RSCi ) (Cimes /Cist)/(Comes /Cost)

(1)

where Cimes is the concentration of an impurity element derived from the relative peak intensity, Comes is the concentration of the reference element derived from the relative peak intensity, Cist is the concentration of an impurity element in the reference sample, and Cost is the concentration of the reference element in the reference sample. Table 1 shows the RSCs of each element calculated from spectra accumulated within 60 and 200 µs after the laser shot. In practice, for every microsecond delay between the laser pulse and the repelling pulse, a spectrum was recorded with scans averaged 100 times. Sodium was used as the reference element. The RSCs of nonmetallic elements were very low compared to the RSCs of metallic elements when the spectra were

accumulated within the large delay range (e.g., 0-200 µs). However, if the spectra accumulation was in the range of 0-60 µs, in which signals for nonmetallic ions still exist, low-discrepancy RSCs for different elements could be obtained. Even though the RSCs of nonmetallic elements were lower than those of metallic elements, the variations among the nonmetallic and metallic elements were within 1 order of magnitude (Table 1). This means that LI-O-TOFMS would have the potential to be used for standardless semiquantitative determination of nonmetallic elements if we accumulate the signals within several tens of microseconds after the laser shots. Therefore, the 60 µs accumulation scheme was chosen for the following experiments. A typical spectrum of a sample mixture is shown in Figure 3b. The sample mixture (copper as the matrix) was composed of 1% NaBF4, NaCl, KBr, KIO3, Li2CO3, NH4H2PO4, SiO2, Se, TeO2, and 0.25% As4S4 in molar fraction. For comparison, the ideal spectrum of the sample is shown in Figure 3a, which assumes that the atomization, ionization, transmission, and detection efficiencies were equal for all elements. The composition unit in Figure 3a is atomic fraction (atom %) because the signal of each isotope corresponds to the atom number theoretically. Figure 3b shows that high-irradiance laser ionization offers a clean spectrum with only slight interference from doubly charged ions and gas species. All of the elements in the sample could be detected with reasonable signal intensities related to the composition of the sample. The spectrum from the experiment (Figure 3b) is in good agreement with the theoretical spectrum (Figure 3a). As seen in Figure 4 and Table 1, quantitative analysis could be performed for all nonmetallic elements in the samples. The matrix element copper was used as the internal standard to Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1951

Figure 4. Calibration curves for representative elements.

eliminate source fluctuations and instrument shift. The calibration curve for each element was plotted by the signal intensity ratio (Ii/ICu) versus the concentration ratio (Ci/CCu). The correlations between the intensity ratios and the concentration ratios were linear with a dynamic range of 6 orders of magnitude for most nonmetallic elements. The limits of detection (LOD) reached microgram per gram or submicrogram per gram in weight (listed in Table 1), calculated from 3σ based on the noise of the corresponding mass. Because of impurities in the helium gas and residue air in the ion source chamber, the background noise for C, N, and O was higher than that for other elements, which resulted in relatively high LODs for C, N, and O. The H2O+ and Cu2+ peaks were relatively high in the spectrum, but they did not interfere with the elements of interest. Polyatomic ions, 1952

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

such as O2+, N2+, and H3O+, might exist in the spectrum, but their intensities are too low to impact on the result significantly. Figure 5 shows the correlation between the elemental concentrations and the signal intensities for the series of artificial standards. The composition unit used in the calibration is atomic fraction (atom %). Except for the stray points representing low concentrations of C, N, and O, which are the result of the high background, the signal intensities of different nonmetallic elements generally varied linearly with concentration. This correlation demonstrates the detection unity for nonmetallic and metallic elements, which indicates that the LI-O-TOFMS has the potential to be used for semiquantitative analysis without the use of a standard.

neous and sensitive detection of nonmetallic elements. The prototype instrument offers LODs in the microgram per gram range and a dynamic range of 6 orders of magnitude for most nonmetallic elements. Spectra are relatively clean, with only a slight interference from gas species and doubly charged matrix ions. With these achievements, this technique can be used for direct quantitative determination of nonmetallic elements in solids and for standardless semiquantitative determination of nonmetallic and metallic elements. ACKNOWLEDGMENT This work was supported by the Natural Science Foundation of China, the National 863 program, the Fujian Province Department of Science and Technology, and the Young Talent Project of Fujian Province (No. 2008F3104). This work has also been supported by NFFTBS (No. J0630429). Figure 5. Element concentration vs signal intensity from a series of sample mixtures.

CONCLUSIONS A high-irradiance laser ionization source combined with a TOF mass spectrometer in an orthogonal geometry enables simulta-

Received for review November 25, 2009. Accepted January 25, 2010. AC9026912

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1953