Ionization for Matrix-Free Surface

Oct 11, 2008 - Fax: +82-42-868-5032. ... Abstract. We propose a new scheme of matrix-free laser desorption/ionization with cation assistance for surfa...
1 downloads 0 Views 618KB Size
Download PDF
Anal. Chem. 2008, 80, 8526–8531

Cation-Assisted Laser Desorption/Ionization for Matrix-Free Surface Mass Spectrometry of Alkanethiolate Self-Assembled Monolayers on Gold Substrates and Nanoparticles Tae Kyung Ha,†,‡ Tae Geol Lee,† Nam Woong Song,† Dae Won Moon,† and Sang Yun Han*,† Nanobio Fusion Research Center, Korea Research Institute of Standards and Science, Daejeon 305-340, Republic of Korea, and Department of Chemistry, Sogang University, Seoul 121-742, Republic of Korea We propose a new scheme of matrix-free laser desorption/ ionization with cation assistance for surface mass spectrometry of self-assembled monolayers (SAMs) of alkanethiolates on gold substrates and gold nanoparticles (NPs). In a proof-of-concept experiment, a simple treatment using an aqueous salt solution such as NaI(aq) was shown to lead to a significant laser desorption/ionization, producing the characteristic (disulfide) ions of alkanethiolate molecules from the monolayers. Further efforts to understand the mechanism were also given, including laser power and salt concentration dependence studies. In the power dependence study, the characteristic ions were found to be produced at low laser power where no gold substrate species was seen. At high laser power, the generation of gold species, Au+-Au5+, resulted in a saturation behavior in the characteristic mass peak for alkanethiolate molecules. In addition, characteristic ions with gold adducts were not observed at any laser power. With increasing salt concentration, the characteristic mass peak was gradually increased. The results suggest that the adduct formation of a cation with alkanethiolates in the monolayers provide a facile pathway to supply a charge to UV laser-desorbed secondary neutrals for mass spectrometric detection. This cation-assisted laser desorption/ ionization (CALDI) mass spectrometry was further examined with the SAMs and mixed SAMs with various terminals such as -OH, -OCH3, -NH2, -ethylene (-CHdCH2), and -acetylene (-CtCH). The CALDI method was also successfully applied to surface mass spectrometry of monolayer-protected gold NPs (∼16 nm diameter) with OH- and COOH-terminated SAMs. The unique advantages of the matrix-free CALDI method may extend our capability in investigations of interfacial chemistry at SAMs as well as mass spectrometric applications using biochips and nanoparticles. In this article, we report a new strategy for matrix-free laser desorption/ionization (LDI) of self-assembled monolayers (SAMs) * To whom correspondence should be addressed. Phone: +82-42-868-5716. Fax: +82-42-868-5032. E-mail:[email protected]. † Korea Research Institute of Standards and Science. ‡ Sogang University.

8526

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

of alkanethiolates on gold. SAMs of thiolates on metals are of great importance in many areas of nanotechnology and bioapplications.1 Among many surface analysis methods that are used to examine SAMs, mass spectrometry has been known to be particularly versatile, being capable of characterizing the molecular entities of SAMs by their molecular masses.2-9 In particular, matrixassisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS) is well suited for the characterizations of biochemically tailored SAMs of alkanethiolates on gold.9-15 The MALDI method has been also known to be useful for label-free detection of biomolecules adsorbed on the surfaces of SAMs.9,16-21 In this field, Mrksich and colleagues have led various investigations using MALDI-TOF mass spectrometry combined with SAMs, including rapid evaluations of interfacial reactions at SAMs12-15 as well as of specific interactions of proteins and oligonucleotides with SAM surfaces,16-18,21 which is referred to as the SAMDI (selfassembled monolayers for MALDI) approach. In the SAMDI (1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103–1169. (2) Leggett, G. J. In TOF-SIMS: Surface Analysis by Mass Spectrometry; Vickerman, J., Briggs, D., Eds.; IM Publications, SurfaceSpectra Limited: West Sussex, Manchester, U.K., 2001; pp 573-593. (3) Wolf, K.; Cole, D.; Bernasek, S. Anal. Chem. 2002, 74, 5009–5016. (4) Hanley, L.; Kornienko, O.; Ada, E.; Fuoco, E.; Trevor, J. J. Mass Spectrom. 1999, 34, 705–723. (5) Trevor, J. L.; Lykke, K. R.; Pellin, M. J.; Hanley, L. Langmuir 1998, 14, 1664–1673. (6) Trevor, J. L.; Mencer, D. E.; Lykke, K. R.; Pellin, M. J.; Hanley, L. Anal. Chem. 1997, 69, 4331–4338. (7) Kpegba, K.; Spadaro, T.; Cody, R. B.; Nesnas, N.; Olson, J. A. Anal. Chem. 2007, 79, 5479–5483. (8) Shibue, T.; Nakanishi, T.; Matsuda, T.; Asahi, T.; Osaka, T. Langmuir 2002, 18, 1528–1534. (9) Mrksich, M. ACS Nano 2008, 2, 7–18. (10) Su, J.; Mrksich, M. Angew. Chem., Int. Ed. 2002, 41, 4715–4718. (11) Su, J.; Mrksich, M. Langmuir 2003, 19, 4867–4870. (12) Min, D.; Tang, W.; Mrksich, M. Nat. Biotechnol. 2004, 22, 717–723. (13) Min, D.; Su, J.; Mrksich, M. Angew. Chem., Int. Ed. 2004, 43, 5973–5977. (14) Min, D.; Yeo, W.; Mrksich, M. Anal. Chem. 2004, 76, 3923–3929. (15) Li, J.; Thiara, P.; Mrksich, M. Langmuir 2007, 23, 11826–11835. (16) Yeo, W.; Min, D.; Hsieh, R.; Greene, G.; Mrksich, M. Angew. Chem., Int. Ed. 2005, 44, 5480–5483. (17) Patrie, S.; Mrksich, M. Anal. Chem. 2007, 79, 5878–5887. (18) Marin, V.; Bayburt, T.; Sligar, S.; Mrksich, M. Angew. Chem., Int. Ed. 2007, 46, 8796–8798. (19) Griessera, H. J.; Kingshott, P.; McArthur, S. L.; McLean, K. M.; Kinsele, G. R.; Timmonse, R. B. Biomaterials 2004, 25, 4861–4875. (20) Evans-Nguyen, K. M.; Tao, S.; Zhu, H.; Cotter, R. J. Anal. Chem. 2008, 80, 1448–1458. (21) Tsubery, H.; Mrksich, M. Langmuir 2008, 24, 5433–5438. 10.1021/ac801405k CCC: $40.75  2008 American Chemical Society Published on Web 10/11/2008

Figure 1. Schematic illustration of a proof-of-concept experiment for CALDI.

Figure 2. LDI mass spectra showing the characteristic disulfide ions produced by treatment with 10 mM aqueous solutions of (a) NaI, (b) NaCl, (c) KI, and (d) KCl, respectively.

strategies, background SAMs with ethylene glycol functionalities are generally employed as a key component preventing nonspecific adsorption of protein on the SAMs. The immobilized biomolecules among the background SAMs provide reaction centers for an assay using a biochip.22 In the surface mass spectrometry of SAMs of alkanethiolates on gold (Au-S-R), MALDI has been found to produce characteristic ions in a disulfide, or in some cases intact thiolate, form of SAM molecules (R-S-S-R:Na+), with a cation adduct such as Na+. The formation of such characteristic ions makes it possible to monitor interfacial chemistry occurring at the monolayers in a straightforward way.9-15 Although the MALDI method gives highly qualitative information, the use of the matrix itself imposes certain limitations on its applications. The choice of matrix and optimization of matrix deposition conditions are of primary concern. In addition, the application of solid matrixes on the surface of SAM results in poorly reproducible ion signals at different positions on a sample chip. This further limits MALDI MS applications toward chemical mapping and quantitative assays on biochips.9 (22) Houseman, B.; Gawalt, E.; Mrksich, M. Langmuir 2003, 19, 1522–1531.

Figure 3. (a) Plots of ion yields for the characteristic disulfide ion, (HO-(EG)6-(CH2)11-S-S-(CH2)11-(EG)6-OH)Na+ (b) (× 2, for visual clarity), Au+(0), Au2+(O), Au3+(]), Au4+(∆), and Au5+(3), as a function of relative laser power. (b) Plots of ion yields for the characteristic ion using NaI(aq) solutions with different concentrations of 5-200 mM.

In pursuit of a matrix-free LDI method toward quantitative surface mass spectrometry for SAMs of alkanethiolates on gold, we first considered previous two-laser mass spectrometry.4-6 In the investigations carried out by Hanley and colleagues, a UV laser (337 nm) was used to induce desorption of thiolate molecules from Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8527

Figure 4. CALDI mass spectra of SAMs with various terminals; SAM(a) of Au-S-(CH2)11-(EG)3-OCH3, mixed SAM(b) of Au-S(CH2)11-(EG)3-OCH2-CtCH and Au-S-(CH2)11-(EG)3-OCH2CHdCH2, and mixed SAM(c) of Au-S-(CH2)11-(EG)6-NH2 and AuS-(CH2)11-(EG)3-OH. All mass spectra clearly show the corresponding characteristic disulfide ions, which are summarized in Table 1.

SAMs. A subsequent irradiation of VUV laser (118 nm) photoionized desorbed neutrals in a TOF mass spectrometer. Their results revealed that dimers (R-S-S-R+), that is disulfides, were dominant products from the SAMs, when they were prepared by adsorption of alkanethiol and disulfide reagents to gold. The study also suggested that dimerization into disulfide products occurred as a result of recombination of surface thiolates during the desorption process.5 In the study, it is to be pointed out that the only role of the second VUV laser (118 nm) was postionization of secondary neutrals in vacuum, offering them a charge needed for mass analysis. Thereby, we postulated that by supplying a charge to secondary neutrals instead of postionization, the first UV irradiation alone will lead to the formation of characteristic ions in the gas phase. Furthermore, we also noted that Na+ containing solutions such as NaOH(aq) are effective reagents that ionize polyethylene glycol (PEG) molecules in an electrospray ionization.23 The addition of NaI salt was also found to enhance LDI efficiency of molecules adsorbed on a SAM surface by cationization.24 These suggest that cation adducts such as Na+ and K+ to ethylene glycol moieties in SAM molecules may be stable enough to survive through a UV laserinduced desorption process. To prove the postulation, this work was carried out and discovered a new way of matrix-free LDI of SAMs of alkanethiolates on gold. In this article, we report cation-assisted laser desorption/ionization (CALDI) for matrix-free surface mass spectrometry of alkanethiolate SAMs on gold. (23) Maziarz, E. P., III; Baker, G. A.; Mure, J. V.; Wood, T. D. Int. J. Mass Spectrom. 2000, 202, 241–250.

8528

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

EXPERIMENTAL METHODS Preparation of Alkanethiolate SAMs on Gold Substrates and Nanoparticles. The gold-plated silicon wafer was cleaned with a superpiranha solution (H2O2, H2NO3, and H2SO4 in a 10: 1:6 ratio), which was found to give a higher reproducibility than a piranha treatment.25 The wafer was thoroughly rinsed with deionized water and ethanol, then cut into ∼ 1 × 1 cm2 pieces. A SAM was prepared by immersing the chip in a 1 mM ethanol solution of thiol or disulfide reagent, typically for 12 h. The reagents for SAMs were commercially obtained and used without further purifications (>95%, CosBiotech, Daejeon, Korea). The SAM with OH terminal (Au-S-(CH2)11-(EG)6-OH), which was mainly used for the evaluations of the CALDI method in this work, was prepared using the thiol reagent of HS-(CH2)11-(EG)6-OH (P/N C3-0006). The EG notation stands for ethylene glycol (-OCH2CH2-) moiety. The methoxy-terminated SAM (Au-S(CH2)11-(EG)3-OCH3) was prepared using HS-(CH2)11-(EG)3OCH3 (P/N C3-0007). The mixed SAM with acetylene and ethylene terminals was prepared using the disulfide reagent of H2CdCH-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2CtCH (P/N C3-0036). The preparation of the mixed SAM with NH2 and OH terminals used a 1:1 solution of HS-(CH2)11(EG)6-NH2 (P/N C3-0026) and HS-(CH2)11-(EG)3-OH (P/N C3-0003) in ethanol (a total concentration of 2 mM). The SAMs were then rinsed and dried under a stream of nitrogen. As for the monolayer-protected gold nanoparticles (NPs), the ligand exchange method using citrate-stabilized gold NPs was employed. Typically, 250 mL of 0.01% HAuCl4 (99.9+%, Aldrich, P/N C520918) solution was heated to the boiling point with vigorous stirring. A volume of 8.75 mL of 34 mM sodium citrate tribasic dehydrate (Sigma, P/N C8532) solution was added to the vortex of the pale yellow solution, causing a color change to burgundy. The solution was boiled for 10 min and allowed to cool, while stirring was further continued for 15 min. The absorption maximum of the resulting nanoparticle solution was observed at 518 nm. The average size of 127 sampled gold NPs was determined to be 15.9 ± 0.9 nm by using transmission electron microscopy (TEM). With the use of the citrate-stabilized gold NPs, the ligand exchanges of surface citrates with respective thiol reagents of HS-(CH2)11-(EG)3-OH (P/N C3-0003) and HS-(CH2)11(EG)3-OCH2-COOH (P/N C3-0008) were done. For the synthesis of AuNP-S-(CH2)11-(EG)3-OH, 190 µL of 1 M HS-(CH2)11-(EG)3-OH solution was added to a 40 mL solution of citrate-stabilized gold NPs with an optical density (OD) of 0.9 at 518 nm. The mixture was stirred for 12 h at room temperature. A 1 mL portion of the mixture was taken and centrifuged at 10 000 rpm for 10 min. After supernatants were removed, 1 mL of ethanol was added to rinse out unbound ligands. Gold NPs were then resuspended by vortexing and centrifuged again. After another washing step using ethanol, 1 mL of deionized water was added and the solution was vortexed to a suspension. The suspension was stored in a 3.5 kDa cutoff membrane cassette and dialyzed in 1 L of deionized water to further remove unreacted reagents such as citrates and thiols. For the preparation of (24) Bounichou, M.; Sanguinet, L.; Elouarzaki, K.; Ale´veˆque, O.; Dias, M.; Levillain, E.; Rondeau, D. J. Mass Spectrom. In press, DOI: 10.1002/jms.1414. (25) Min, H.; Park, J.-W.; Shon, H.-K.; Moon, D.-W.; Lee, T.-G. Appl. Surf. Sci. In press, DOI: 10.1016/j.apsusc.2008.05.099.

Table 1. Expected Structures for the Characteristic Ion Peaks Produced in CALDI Experiments SAMs

characteristic disulfide ions

mass (amu)

(a) (b)

(H3CO-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH3)Na+ (HCtC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2-CtCH)Na+ (HCtC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2-CHdCH2)Na+ (H2CdHC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2-CHdCH2)Na+ (HO-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OH)Na+ (HO-(EG)3-(CH2)11-S-S-(CH2)11-(EG)6-NH2)Na+ (H2N-(EG)6-(CH2)11-S-S-(CH2)11-(EG)6-NH2)Na+

721.5 769.5 771.5 773.5 693.4 824.5 955.6

(c)

AuNP-S-(CH2)11-(EG)3-OCH2-COOH, 190 µL of 1 M HS-(CH2)11-(EG)3-OCH2-COOH solution was added to a 40 mL solution of citrate-stabilized gold NPs with 0.9 OD at 518 nm with stirring. The mixture was stirred for 12 h at room temperature. Then, a small portion of solution mixture (500 µL) was taken and dialyzed in 500 mL of deionized water using a 3.5 kDa cutoff membrane cassette to remove residues in the nanoparticle solution. LDI Experiments. In CALDI experiments, KCl (99.99+%; P/N 204099), KI (99.99+%; P/N 204102), NaI (99.99+%; P/N 229911), and NaCl (99.999%; P/N 204439) salts were purchased from Sigma-Aldrich and used without further purifications. The aqueous solutions were prepared by dissolving respective salts in deionized water to a particular concentration. For pretreatments, SAM chips were immersed in the aqueous salt solutions for 10 min to induce adsorption of alkali cations to the

monolayers. A 10 mM concentration was used in this work, except for the concentration dependence study. Pipetting or spraying of aqueous salt solutions onto the surface of SAMs also yielded the same mass spectra with good quality. However, we used only dipping for all the experiments given in the text for reproducibility. After the pretreatments with salt solutions, the SAM chips were air-dried, and then mass spectrometry was carried out. LDI experiments utilized a commercial MALDITOF mass spectrometer (Autoflex III, Bruker Daltonics, Germany), which is equipped with a 355 nm laser with a repetition rate up to 200 Hz. Because of the nature of a commercial instrument, however, it was difficult to measure the laser power density at the sample surface. Instead, relative laser power (%) read from the control software (flexControl) was employed. For information, the maximum laser peak power measured before the laser exit outside the mass spectrometer

Figure 5. TEM images of gold NPs protected by alkanethiolate monolayers of (a) AuNP-S-(CH2)11-(EG)3-OH and (b) AuNP-S-(CH2)11-(EG)3-OCH2-COOH. Their respective CALDI mass spectra are given in parts c and d. The CALDI mass spectrum (c) shows the formation of characteristic disulfide ions at mass 693.4, (HO-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OH)Na+. However, the second peak (/) with a mass spacing of 32 from the characteristic peak was seen at mass 661.5, which is likely to indicate the formation of sulfide ions, (HO-(EG)3-(CH2)11-S-(CH2)11-(EG)3-OH)Na+, probably due to synthetic failures or photochemistry at play for the NPs with a high surface curvature. The spectrum (d) shows the three peaks of characteristic disulfide ions from the COOH-terminated monolayer on gold NPs, but the peak for sulfide formation was not seen. The observed three mass peaks stand for (HOOC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2COOH)Na+ (mass 809.5), (NaOOC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2-COOH)Na+ (mass 831.4), and (NaOOC-CH2O(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2-COONa)Na+(mass 853.4). Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8529

was ∼200 µJ/pulse at 200 Hz. The laser optics was set to focus the laser beam into a spot of ∼50 µm diameter on the sample. RESULTS AND DISCUSSION A Proof-of-Concept Experiment for CALDI. In order to corroborate our postulation, we carried out a proof-of-concept experiment that is illustrated in Figure 1. We prepared the OHterminated SAM of Au-S-(CH2)11-(EG)6-OH, where EG stands for the ethylene glycol moiety. Freshly prepared SAMs were immersed in the aqueous solutions of 10 mM NaI, NaCl, KI, and KCl for 10 min, respectively, dried in ambient conditions, and then LDI experiments were carried out using a MALDI-TOF mass spectrometer at 355 nm. The results are given in Figure 2. Indeed, as shown in Figure 2, the irradiation of UV laser alone on the SAMs treated with the four salt solutions all led to the formation of characteristic disulfide ions, (HO-(EG)6-(CH2)11-S-S-(CH2)11-(EG)6-OH)Na+ or K+ (mass 957.6 or 973.6, respectively) with a cation adduct according to treated solutions. However, without treatment of the solutions, no detectable amount of ions was observed. Accordingly, the simple treatment with cation-containing solutions was proved to assist LDI of the SAMs of alkanethiolates on gold. Among the four salt solutions, the NaI(aq) solution gave the best LDI efficiency, but the other solutions also gave the mass spectra with good quality. To the best of our knowledge, this study is the first that demonstrates matrix-free laser desorption/ionization of alkanethiolate SAMs on gold with cation assistance. Effects of Laser Power and Salt Concentrations. In an effort to look into the observed cation-assisted laser desorption/ ionization process more closely, first, we conducted a laser power dependence study using the OH-terminated SAM treated with 10 mM NaI(aq). Figure 3a shows the plots of the ion intensities of ionic species generated by LDI experiments as a function of relative laser power. As shown in Figure 3a, the characteristic disulfide ion (mass 957.6) for the SAM began to be produced at low laser power where any gold substrate species was not seen. Its mass peak gradually increased with increasing laser power until gold species apparently showed up, Au+ and Aun+ (up to n ) 5). However, above the minimum laser power for the smallest gold species of Au+ to be generated, the ion intensity of the disulfide ion displays a saturation behavior, while those for gold species still gradually increased with laser power. Besides, disulfide ions with gold adducts were not observed at any laser power. The lower laser power for the formation of characteristic ions suggests that photodestruction (ablation) of the gold substrate may not be deeply involved in the generation process. The observed saturation behavior may also indicate that the ablation of gold substrate in fact interrupts soft desorption of alkanethiolate molecules from the SAM surface. Second, the effect of cation concentrations was further examined at a fixed laser power, where gold ions were not produced. For this study, the OH-terminated SAM chips treated by NaI(aq) solutions with a range of concentrations of 5-200 mM were used. Figure 3b plots the yields for the characteristic disulfide ion (mass 957.6) produced with different concentrations. In the results, it was conceivable that the treatment with a higher concentration produced more characteristic ions, demonstrating that the Na+ concentration is directly related to the efficiency of ion formation. It supports our initial postulation that the needed charge can be 8530

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

supplied simply via the adduct formation of Na+, probably to the ethylene glycol moieties in the alkanethiolate molecules immobilized on gold. It is also to be noted that by increasing the concentration, the signal intensity, thus sensitivity, can be greatly increased. However, at the concentrations higher than about 50 mM, tiny salt crystals began to form at the SAM surface so that they interfered with reproducible measurements. The crystal usually served as a hot spot that gave a stronger ion intensity, perhaps due to increased local Na+ concentration. The treatment with less than 1 mM was also good enough to give a good signalto-noise ratio (>500) for mass analysis. Likewise, the mechanism of CALDI seems to be analogous to that observed in the earlier two-laser mass spectrometry.4-6 As a matter of fact, the CALDI method simply replaces the expensive VUV postionization by supplying a charge with cationization. In a separate experiment, MALDI using THAP matrix (5 mg/mL in 100% methanol solution) also gave the same characteristic ions for the above SAM. However, we noted that the threshold laser energy for the MALDI process was rather lower than that for CALDI, which may indicate that the MALDI process has a different mechanism from that for CALDI. The role of matrix assistance in the MALDI method for SAMDI-TOF MS may be still a subject of further study. CALDI of Various Alkanethiolate SAMs on Gold Substrates and Nanoparticles. The CALDI method was further examined for the self-assembled monolayers with different terminal functionalities; (a) SAM with -OCH3 terminal, Au-S(CH2)11-(EG)3-OCH3, (b) mixed SAM with ethylene (-CHdCH2) and acetylene (-CtCH), Au-S-(CH2)11-(EG)3OCH2-CtCH and Au-S-(CH2)11-(EG)3-OCH2-CHdCH2, and (c) mixed SAM with -OH and -NH2 terminals, Au-S(CH2)11-(EG)6-NH2 and Au-S-(CH2)11-(EG)3-OH. As shown in Figure 4, the obtained CALDI mass spectra clearly exhibited the corresponding characteristic disulfide ions, demonstrating that CALDI is applicable to SAMs with various functionalities. The observed characteristic disulfide ions are summarized in Table 1. In addition to the various SAMs on gold substrates, the alkanethioate monolayers on gold NPs were also examined. The TEM images in Figure 5 show the gold NPs with ∼16 nm diameter, which were protected with OH-terminated and COOHterminated alkanethiolate monolayers by adsorption of thiol reagents of HS-(CH2)11-(EG)3-OH and HS-(CH2)11-(EG)3OCH2-COOH, respectively. A volume of 0.5 µL of solution of the monolayer-protected gold NPs was dropped on a standard MALDI steel target (Bruker MTB plate). An additional 0.5 µL drop of 1 mM NaI(aq) was applied and dried as in the dried-droplet method for MALDI sample preparations. Then, the mass spectra were taken by a MALDI-TOF mass spectrometer. As shown in the LDI mass spectrum (Figure 5c), the cation-assistance gave the characteristic disulfide ions, (HO-(EG)3-(CH2)11-S-S-(CH2)11(EG)3-OH)Na+ at mass 693.4, from the surface of the gold NPs. More interestingly, even without the addition of NaI(aq) solution, the monolayer-protected gold NPs sample gave the same spectrum with a weaker intensity, which was probably caused by residual Na+ ions from the earlier synthetic steps. The COOH-terminated gold NPs gave three characteristic disulfide ions as well, which is in a good agreement with the previous MALDI experiment of

the SAMs on the gold substrate.11 The three disulfide ions include (HOOC-CH2O-(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2COOH)Na+ at mass 809.5, (NaOOC-CH2O-(EG)3-(CH2)11S-S-(CH2)11-(EG)3-OCH2-COOH)Na+ at mass 831.4, and (NaOOC-CH2O(EG)3-(CH2)11-S-S-(CH2)11-(EG)3-OCH2COONa)Na+ at mass 853.4. It demonstrates that the CALDI method can be also applied to surface mass spectrometry of the monolayer-protected gold NPs. Figures of Merit for CALDI. The matrix-free condition achieved by CALDI gives many advantages for mass spectrometry of alkanethiolate SAMs on gold. Most of all, our method of using the aqueous solutions of alkali salts is simple. It does not require any complicated reagent preparation or optimization, and deposition procedures but dissolving alkali salts in water and soaking the sample with it. As shown in the concentration dependence study, sensitivity can be greatly enhanced and is even controllable. Moreover, the homogeneous solutions provide uniform pretreatments over the entire SAM surface, which as a result brings the high reproducibility required for quantitative mass analysis and imaging mass spectrometry of the SAMs. For example, a simple reproducibility test was done by taking the absolute CALDI intensities of characteristic disulfide ion from 16 different positions randomly chosen on a OH-terminated SAM chip. The results gave a standard deviation of 7% for the disulfide ion intensities, which is otherwise difficult to achieve using the MALDI method due to hot spots caused by matrix crystals (see Supporting Information). In addition, the aqueous solutions of Na+ or K+ with neutral pH offer more physiological environment than rather offensive matrix solutions, which may allow new applications in assays using biochips.

CONCULSIONS In summary, we describe the development of matrix-free cation-assisted laser desorption/ionization (CALDI) for selfassembled monolayers of alkanethiolates on gold substrates and nanoparticles. This matrix-free approach was found to give the same qualitative information that can be obtained by MALDI mass spectrometry. However, by avoiding the use of matrix, the CALDI method opens a new opportunity toward quantitative mass analysis and imaging mass spectrometry for the SAMs on gold. This new way of surface mass spectrometry may extend our capability in understanding interfacial chemistry occurring at the interface of alkanethiolate SAMs on gold as well as mass spectrometric applications using biochips and nanoparticles. ACKNOWLEDGMENT This work was supported by the Biosignal Analysis Technology Innovation Program (Grant M106450100002-06N4501-00210) of MEST via KOSEF and also partly by the Nano R&D Program (Grant 2008-04406) and KRISS. T.K.H. is thankful to the Brain Korea 21 (BK21) program for scholarship. SUPPORTING INFORMATION AVAILABLE Reproducibility tests for MALDI and CALDI using the SAMs of Au-S-(CH2)11-(EG)6-OH. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review July 8, 2008. Accepted September 3, 2008. AC801405K

Analytical Chemistry, Vol. 80, No. 22, November 15, 2008

8531