Femtogram Detection and Quantitation of Residues Using Laser

A newly developed high irradiance laser ionization orthogonal time-of-flight ... The residues were first characterized in terms of shape and elemental...
0 downloads 0 Views 1MB Size
Download PDF
Anal. Chem. 2009, 81, 8623–8626

Femtogram Detection and Quantitation of Residues Using Laser Ionization Orthogonal Time-of-Flight Mass Spectrometry Quan Yu,† Zhiyu Cao,† Lingfeng Li,† Bin Yan,† Wei Hang,*,†,‡ Jian He,§ and Benli Huang† Department of Chemistry, Key Laboratory of Analytical Science, 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 A newly developed high irradiance laser ionization orthogonal time-of-flight mass spectrometer (LI-O-TOFMS) was employed for the elemental analysis of residues, which were prepared by evaporating mixed salt solutions. The residues were first characterized in terms of shape and elemental distribution. In TOFMS detection, all of the metal elements in the residue can be observed in the spectra. Relative sensitivity coefficients for different elements were within 1 order of magnitude, which meets semiquantitative analysis criteria. By calculating the individual masses from the ablated area due to a single laser shot, the absolute detection limit reached 7 × 10-15 g for most metal elements. The results indicate that LI-O-TOFMS is capable of performing ultratrace elemental qualification and quantification, with an absolute limit of detection and an absolute limit of quantitation at the femtogram level. Techniques for direct solid analysis are receiving increasing interest and are poised to replace the digestion-based techniques that are routinely used.1,2 Some well-known methods have been developed, including laser-induced breakdown spectroscopy (LIBS),3,4 glow discharge mass spectrometry (GDMS),5,6 laser ablationinductivelycoupledplasmamassspectrometry(LA-ICPMS),7,8 and secondary ion mass spectrometry (SIMS),9,10 as well as laser ionization mass spectrometry (LIMS).11,12 These techniques can perform solution analysis after suitable sample preparation, * Corresponding author. E-mail: [email protected]. † Department of Chemistry, Key Laboratory of Analytical Science, College of Chemistry and Chemical Engineering. ‡ State Key Laboratory of Marine Environmental Science. § Department of Mechanical and Electrical Engineering. (1) Rusak, D. A.; Castle, B. C.; Smith, B. W.; Winefordner, J. D. Trends Anal. Chem. 1998, 19, 453–469. (2) Sturgeon, R. E. J. Anal. At. Spectrom. 1998, 13, 351–361. (3) Winefordner, J. D.; Gornushkin, I. B.; Correll, T.; Gibb, E.; Smith, B. W.; Omenetto, N. J. Anal. At. Spectrom. 2004, 19, 1061–1083. (4) Omenetto, N. J. Anal. At. Spectrom. 1998, 13, 385–399. (5) Pisoneroa, J.; Ferna´ndeza, B.; Pereiroa, R.; Bordela, N.; Sanz-Medel, A. Trends Anal. Chem. 2006, 25, 11–18. (6) Harrison, W. W.; Barshick, C. M.; Klingler, J. A.; Ratliff, P. H.; Mei, Y. Anal. Chem. 1990, 62, 943–949. (7) Russo, R. E.; Mao, X. L.; Mao, S. S. Anal. Chem. 2002, 74, 70 A77 A. (8) Gu ¨ nther, D.; Hattendorf, B. Trends Anal. Chem. 2005, 24, 255–265. (9) Griffiths, J. Anal. Chem. 2008, 80, 7194–7197. (10) Walker, A. V. Anal. Chem. 2008, 80, 8865–8870. (11) Sysoev, A. A.; Sysoev, A. A. Eur. J. Mass Spectrom. 2002, 8, 213–232. 10.1021/ac901615k CCC: $40.75  2009 American Chemical Society Published on Web 09/10/2009

Table 1. Examples of ALODs

b

method

ALOD (g)

ref

LA-ICPMS ETA-LIFa GDMS LA-IS-TOFMSb LIBS SIMS LAMMA

10-12 4 × 10-15 1.5 × 10-12 10-12-10-15 10-12 3 × 10-18 10-20

13 19 20 21 22 23 24

a ETA-LIF: electrothermal atomization-laser induced fluorescence. LA-IS-TOFMS:laserablationionstoragetime-of-flightmassspectrometer.

namely, a phase-change process from liquid to solid. In practice, evaporating the solution from the residue is a simple and practical strategy, which has been described in some previous literature.13-15 The absolute limit of detection (ALOD) is one of the most important figures of merit to evaluate the analytical capability of a given technique.16 It is defined as the minimum amount of analyte required for detection. Currently, ultratrace element analysis is commonly available using various instrumental approaches, and even atom-level detection has been achieved in some cases.17,18 Table 1 summarizes the ALODs for different techniques. As expected, most laser-related techniques have outstanding analytical capability in the microanalysis arena. Compared with optical detection schemes, mass spectrometry offers simple spectra that can be easily interpreted. Laser ablation (12) He, J.; Huang, R.; Yu, Q.; Lin, Y.; Hang, W.; Huang, B. J. Mass Spectrom. 2009, 44, 780–785. (13) Fittschen, U. E. A.; Bings, N. H.; Hauschild, S.; Forster, S.; Kiera, A. F.; Karavani, E.; Fromsdorf, A.; Thiele, J.; Falkenberg, G. Anal. Chem. 2008, 80, 1967–1977. (14) Gondal, M. A.; Hussain, T. Talanta 2007, 71, 73–80. (15) Yang, L.; Sturgeon, R. E.; Mester, Z. Anal. Chem. 2005, 77, 2971–2977. (16) Omenetto, N.; Petrucci, G. A.; Cavalli, P.; Winefordner, J. D. Fresenius J. Anal. Chem. 1996, 355, 878–882. (17) Glick, M.; Smith, B. W.; Winefordner, J. D. Anal. Chem. 1990, 62, 157– 161. (18) Hurst, G. S.; Nayfeh, M. H.; Young, J. P. Phys. Rev. A 1977, 15, 2283. (19) Masera, E.; Mauchien, P.; Lerat, Y. Spectrochim. Acta, Part B 1996, 51, 543–548. (20) Sola-Vazquez, A.; Martin, A.; Costa-Fernandez, J. M.; Pereiro, R.; Sanz-Medel, A. Anal. Chem. 2009, 81, 2591–2599. (21) Klunder, G. L.; Grant, P. M.; Andresen, B. D.; Russo, R. E. Anal. Chem. 2004, 76, 1249–1256. (22) Xu, L.; Bulatov, V.; Gridin, V. V.; Schechter, I. Anal. Chem. 1997, 69, 2103– 2108. (23) Cox, X. B.; Bryan, S. R.; Linton, R. W.; Griffis, D. P. Anal. Chem. 1987, 59, 2018–2023. (24) Guest, W. H. Int. J. Mass Spectrom. Ion Processes 1984, 60, 189–199.

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

8623

is a versatile tool because any sample with appropriate light absorptivity can be ablated.4,25 At present, LA-ICPMS is probably the most widely used technique for direct solid analysis. Although ICPMS has a quite low LOD, it has not shown superior performance in ALOD due to the “dilution” of the ablated mass packet in transportation and plasma expansion before the sampler. The coupling of a laser ionization source directly with a timeof-flight mass spectrometer (TOFMS) has been accepted as a powerful tool for solid analysis since its appearance in the 1960s.26 The laser beam was focused onto a microarea of the target surface, which minimized sample preparation and consumption. Traditional laser ionization time-of-flight mass spectrometers, such as LAMMA27 or LIMA,28 have their ion sources directly coupled to the mass analyzers in the same high vacuum chamber. Although they exhibited excellent ALODs, as shown in Table 1, their spectra had poor resolution and severe chemical background due to multicharged ions at high irradiance (109 W/cm2 and above). Thus, they were rarely used for quantitation. A compact high-irradiance LI-O-TOFMS system has been developed recently in our group, providing satisfactory performance in the multielemental analysis of solids, such as alloys and ore samples.29 In this paper, the instrument was utilized in direct analysis of dried droplets to explore the ALOD and ALOQ (absolute limit of quantitation) of the system. With suitable sample preparation and instrument optimization, high sensitivity and spatial resolution have been achieved using our instrument. This study presents not only a practical methodology for analysis of volatilizable solution samples but also a technique for ultratrace microanalysis with an ALOD and ALOQ at the femtogram level. EXPERIMENTAL SECTION The LI-O-TOFMS system used in this study has been described previously.29 Briefly, a Nd:YAG laser (NL303G, EKSPLA) with 355 nm wavelength and a 4.4 ns pulse duration was employed. The laser beam was focused using a laser-focusing objective (LMU3X, OFR division of Thorlabs) that has a theoretical focus spot of 5 µm in diameter. A compact TOFMS unit was operated in orthogonal repelling mode, providing a typical resolving power of 3000. In the experiments presented here, the signal was acquired using a digital storage oscilloscope (Lecroy, 42Xs) with a 2.5 Gs/s rate at 400 MHz bandwidth. The major advantages of this system are the usage of buffer gas in the source for cooling the energetic ions, charge reduction of multicharged ions through collisional recombination, and orthogonal TOF geometry to enhance the resolution.29 LA-ICPMS was used for the elemental scanning study of the residues. A quadrupole-based ICP mass spectrometer (Agilent 7500a) was coupled with a laser ablation system (CETAC LSX 213). The Nd:YAG laser was operated at wavelength of 213 nm and a pulse duration of 5 ns. A standard solution containing 1000 ppm of Na, Mg, Al, K, Ca, Mn, Ni, Cu, Zn, Rb, Sr, Cd, Cs, Ba, and Bi was prepared by (25) Russo, R. E.; Mao, X.; Borisov, O. V. Trends Anal. Chem. 1998, 17, 461– 469. (26) Fenner, N. C.; Daly, N. R. Rev. Sci. Instrum. 1966, 37, 1068–1070. (27) Vogt, H.; Heinen, H. J.; Meier, S.; Wechsung, R. Fresenius J. Anal. Chem. 1981, 308, 195–200. (28) Ruckman, J. C.; Davey, A. R.; Clarke, N. S. Vacuum 1984, 34, 911–924. (29) Yu, Q.; Huang, R.; Li, L.; Lin, L.; Hang, W.; He, J.; Huang, B. Anal. Chem. 2009, 81, 4343–4348.

8624

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

dissolving the chloride or nitrate salts in ultrapure water. This stock solution was diluted into a series of solutions with different concentrations from 100 to 1 ppm for further study. It must be emphasized that, in this paper, the concentration of the residue refers to the elemental concentration in the solution before drying. The residues were prepared by evaporating 10 µL of the solution on a special quartz glass plate, which was precoated with a hydrophilic molecular layer on the surface (treated with 3-aminopropyltriethoxysilane, followed by a remodification process using glutaraldehyde).30 This layer allows little shrinkage of the droplet when the sample is evaporated and does not introduce any metal elements when the layer is ablated and sampled. In order to fix the size of the residue, a shallow circular fluting was made on the glass with a diameter of 7 mm. In this way, the droplet was confined in the circle covering the whole inner area. The sample was then heated at 70 °C for 2 h at atmospheric pressure, leaving a homogeneous solute surface on the residue. The glass plate was mounted on a microposition stage in the ion source and moved a distance of 50 µm after each laser shot during the experiment. RESULTS AND DISCUSSION The accuracy of the analysis was significantly correlated to the condition of the sample under investigation. Hence, before the LI-O-TOFMS measurements, the elemental content and the homogeneity of the residues were evaluated using other analytical techniques, including LA-ICPMS, scanning electron microscopy (SEM), and metallurgical microscopy. As mentioned previously, the solution was restricted within the circular fluting on the glass, producing a homogeneous surface. Since the area of the samples was constant, it is obvious that the thickness of the residue was determined by the solute concentration. A roughly linear relationship was found between the height of the residue and solute concentration. Height values of 140, 15, and 2 µm were observed for 1000, 100, and 10 ppm residues, respectively, which were measured by determining the focus deviation between the sample and the substrate surface using metallurgical microscope imaging. In addition to the residue shape, elemental distribution was measured by LA-ICPMS for the 10 ppm residue. The scan was carried out with a sampling diameter of 100 µm and scanning rate of 100 µm/s. The results are shown in Figure 1, which indicates uniform elemental distribution. For the LI-O-TOFMS experiment, the crater generated by each laser shot was investigated to show the result of the laser-solid interaction. The profiles provided important information for calculating the absolute detection limit. Figure 2a shows the image of a typical crater with a diameter of about 6 µm and a depth of about 8 µm, which was generated on a 50 ppm residue under a laser energy of 11.6 mJ. The image of the same spot, after the residue was rinsed away, is presented in Figure 2b, clearly showing a hole penetrating the glass substrate. The LI-O-TOFMS system is capable of performing rapid multielemental analysis. A typical mass spectrum of the residue is shown in Figure 3. All the elements in the solution and their isotopes are clearly visible with little noise. The existence of silicon peaks was attributed to ablation of the glass substrate. Figure 4 shows the variation of the ion intensity acquired under different (30) Huang, R. M.S. Thesis, Xiamen University, Xiamen, 2006.

Figure 1. The LA-ICPMS scanning results of some of the elemental contents in the 10 ppm residue.

Figure 4. Elemental intensity of the 50 ppm residue acquired using different laser energies. The corresponding RSC values are shown in the upper half of the plot.

Figure 2. SEM image of (a) a typical crater and (b) the penetration of the crater in the substrate.

Figure 5. Mass spectra of consecutive single shots on the residues with different solution concentrations.

Figure 3. A typical mass spectrum of the 50 ppm residue acquired using the LI-O-TOFMS system.

laser energies. To evaluate the precision, RSDs were calculated with eight consecutive shots at different locations using the same laser irradiance. The results clearly demonstrate the increase in ion generation with the increase in laser energy due to the enhanced ablation and, more importantly, enhanced ionization efficiency at high irradiance. Relative sensitivity coefficients

(RSCs) were also calculated and are plotted in Figure 4, which indicates elemental fractionation in the measurement.11 Different elemental properties and transportation bias could be the cause of the fractionation. However, RSC values for different elements are within 1 order of magnitude using Ni as the reference element, which are inside the range of semiquantitation.31 Mass spectra of consecutive shots on the residues with different concentrations are given in Figure 5. Each spectrum was acquired with a single laser shot of 11.6 mJ energy. The spectra show less variation for the sample at 100 ppm concentration. The intensities were reduced and their variation increased with the decrease in the solute concentrations. RSDs for the total ion currents of 100, 10, and 1 ppm spectra were 7%, 15%, and 28%, (31) Herrera, K. K.; Tognoni, E.; Omenetto, N.; Smith, B. W.; Winefordner, J. D. J. Anal. At. Spectrom. 2009, 24, 413–425.

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

8625

respectively. For the 1 ppm residue, most elements can be observed stably in most laser shots. However, signals of Mg, Al, and Ca could not be stably observed due to the low intensities, which are caused by either their elemental properties or the transmission bias against low-mass ions. Further experiments were performed using a 0.1 ppm residue. Most of the elements could not be stably observed in the spectra, with the exception of alkali metal ions. Therefore, the residue made from 1 ppm solution reached the LOD of the LI-O-TOFMS system. The ALOD and ALOQ were the primary targets under this investigation. Although the volume of the crater generated in each laser shot could be acquired, the sample density was difficult to estimate. Hence, directly calculating the mass removed by the ablation crater was not feasible. The ALOD was then determined through a simple methodology. As mentioned above, the residue has a homogeneous elemental distribution and each laser shot can penetrate through the thin residue layer with low elemental concentration. The ablated mass of a certain element was determined by using the following equation mi ) (Sc /Sr) × mt

(1)

where mi is the mass ablated for a specific element in a single shot, mt is the total mass of the specific element in the residue, and Sc and Sr are the area of the crater and the residue, which are about 28 µm2 and 38 mm2, respectively. Meanwhile, mt can be easily derived from the elemental concentration in the original solution. Figure 6 shows Si, K, Cu, Cs, and Bi intensities versus concentrations under the laser energy of 11.6 mJ. Other elements exhibited the same trend but are not shown in the figure in order to avoid overcrowding of the plot. Generally, signal intensities of metal elements increase with the solution concentrations up to 50 ppm, after which they level off. In contrast, the Si signal decreases with solution concentration. This phenomenon indicates that the crater depth (about 8 µm) for 50 ppm residue reaches the interface of the residue and the glass. A further increase of residue thickness by high concentration solution will not increase the signal intensity because of the limitation on crater depth. The data in Figure 6 show that the calibration curves can be plotted for metal elements down to the 1 ppm residue. Such a low concentration produced a considerably thinner residue in the substrate, and the ablated mass of each element was about 7 × 10-15 g, calculated using eq 1. The signal-to-noise ratios are 12 for K, 6 for Cu, 23 for Cs, and 11 for Bi, on average. Thus, it can

8626

Analytical Chemistry, Vol. 81, No. 20, October 15, 2009

Figure 6. The signal intensity versus the residue concentration.

be concluded that the ALOD and ALOQ can reach the femtogram level with the LI-O-TOFMS system. CONCLUSIONS The analytical methodology for residue analysis has been demonstrated with a newly built LI-O-TOFMS. The evaporation procedure was performed on a modified glass substrate to achieve homogeneous elemental distribution. Reliable elemental information was acquired in single shot detection. The variation of RSCs was in the appropriate range to realize semiquantitative criteria. Most importantly, femtogram ALOD and ALOQ were achieved in our experiment. A practical and simple protocol is provided to perform elemental measurement in solution by the laser ionization mass spectrometric detection of residues. This methodology could pave the way for novel microanalysis, such as deposited particulates, single cells, and forensic samples, with an ultralow absolute limit of detection and absolute limit of quantitation. ACKNOWLEDGMENT Financial support was provided by the National 863 project, Natural Science Foundation of China, and Fujian Province Department of Science & Technology. This work has also been supported by NFFTBS (No. J0630429).

Received for review July 21, 2009. Accepted August 25, 2009. AC901615K