ARTICLE pubs.acs.org/Langmuir
Adlayers of Dimannoside Thiols on Gold: Surface Chemical Analysis Paul M. Dietrich,*,† Tim Horlacher,‡ Pierre-Luc Girard-Lauriault,† Thomas Gross,† Andreas Lippitz,† Hyegeun Min,† Thomas Wirth,† Riccardo Castelli,‡ Peter H. Seeberger,‡,§ and Wolfgang E. S. Unger† †
Surface and Thin Film Analysis WG, BAM Federal Institute for Materials Research and Testing, D-12203 Berlin, Germany Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, D-14424 Potsdam, Germany § Institute of Chemistry and Biochemistry, Freie Universit€at Berlin, D-14195 Berlin, Germany ‡
bS Supporting Information ABSTRACT: Carbohydrate films on gold based on dimannoside thiols (DMT) were prepared, and a complementary surface chemical analysis was performed in detail by X-ray photoelectron spectroscopy (XPS), time-of-flight secondary ion mass spectrometry (ToF-SIMS), near-edge X-ray absorption fine structure (NEXAFS), FT-IR, and contact angle measurements in order to verify formation of ω-carbohydrate-functionalized alkylthiol films. XPS (C 1s, O 1s, and S 2p) reveals information on carbohydrate specific alkoxy (CO) and acetal moieties (OCO) as well as thiolate species attached to gold. Angle-resolved synchrotron XPS was used for chemical speciation at ultimate surface sensitivity. Angle-resolved XPS analysis suggests the presence of an excess top layer composed of unbound sulfur components combined with alkyl moieties. Further support for DMT attachment on Au is given by ToF-SIMS and FT-IR analysis. Carbon and oxygen K-edge NEXAFS spectra were interpreted by applying the building block model supported by comparison to data of 1-undecanethiol, poly(vinyl alcohol), and polyoxymethylene. No linear dichroism effect was observed in the angle-resolved C K-edge NEXAFS.
’ INTRODUCTION In the past decade, carbohydrate research has gained increased interest since glycans are involved in many fundamental biological processes. Glycans are present on virtually all cells and regulate interactions of the cell with the environment.14 Spotted arrays with thousands of oligosaccharides are considered essential tools to investigate carbohydrate-based interactions and might be exploited for diagnosis of diseases in the future.512 For an effective utilization and application of carbohydrate-based microarrays, further research related to immobilization and surface chemistry as well as density, conformation, and accessibility of glycans is indispensable. Most of the published work about carbohydrate-functionalized films on gold had focused on specific biochemical applications, e.g., binding studies or high-throughput screening, but often a detailed surface characterization of the respective surfaces and interfaces has not been undertaken.1323 However, a more detailed chemical characterization of relevant species seems to be another important prerequisite for a deeper understanding of carbohydrate-based biointerfaces and, based thereon, improved performance and stability in applications. Herein, we report on surface chemical analysis of a specific carbohydrate adlayer, prepared by self-assembly of dimannoside thiol (DMT) molecules on a gold substrate (cf. Scheme 1). This carbohydrate adlayer is a first trial to model more complex carbohydrate surfaces, which are used to mimic surfaces of r 2011 American Chemical Society
biological structures. Another aspect is to learn more about the individual measurement capabilities of the different methods used in this study to characterize a carbohydrate surface. DMT/Au films were characterized by a complementary set of techniques of chemical analysis in laboratory and synchrotron radiation X-ray photoelectron spectroscopy (SR XPS) in angleresolved mode, angle-resolved near-edge X-ray absorption fine structure (NEXAFS) analysis, and time-of-flight secondary ion mass spectrometry (ToF-SIMS) focusing on a detailed chemical speciation of the immobilized alkylthiol-functionalized carbohydrate molecules. Additionally FT-IR spectroscopy and water contact angle (WCA) measurements were used to characterize the DMT/Au films. SR XPS has been used in addition to Al KR radiation laboratory XPS because its surface sensitivity can be tuned in a way that exclusively the dimannoside thiol film contributes to the spectra. For C 1s photoelectrons the information depth24 (from which 95% of the signal originates) can be substantially decreased from 11 down to 2.1 nm by changing the excitation energy from 1486.6 eV (Al KR) to 385 eV synchrotron radiation, respectively.25,26
Received: October 7, 2010 Revised: February 4, 2011 Published: March 21, 2011 4808
dx.doi.org/10.1021/la104038q | Langmuir 2011, 27, 4808–4815
Langmuir Scheme 1. Idealized Immobilization Reaction of the Dimannoside Thiol (DMT) on Gold Using the Respective Disulfide DMT2a
a
Different chemical moieties are color-coded: acetal (red), alkoxy (green), alkyl (blue), thiolate (orange), and disulfide (brown).
Moreover, NEXAFS spectra, both carbon and oxygen K-edge, were measured in order to identify carbohydrate-specific features. A detailed interpretation of NEXAFS spectra was achieved by comparison to data of well-known reference specimens and application of a building block model.
’ EXPERIMENTAL DETAILS Materials. If not noted otherwise, all chemicals were purchased from Sigma-Aldrich and were of highest available purity. Gold substrates (Georg Albert PVD, Germany) used for the deposition of self-assembled monolayers were prepared by thermal evaporation of 30 nm of Au (purity 99.99%) onto 9 nm titanium precoated polished single-crystal Si(100) wafers. Hexanethiol-terminated disaccharide (DMT) was prepared following methods analogous to those described in the literature.2729 2-O-Acetyl-3,4,6-tri-O-benzylmannosyl trichloroacetimidate30 was coupled to S-benzyl-6-mercaptohexanol; after deprotection of the acetyl moiety, a second gylcosylation with the same mannosyl trichloroacetimidate furnished the fully protected dimannoside. Complete cleavage of the protecting groups was performed by dissolving metal reduction and the DMT compound was recovered in the form of disulfide after dialysis against water. Film Preparation. The gold substrates were cleaned with super piranha solution31 (1/6/10 v/v/v, 96% HNO3/96% H2SO4/30% H2O2) by dipping for less than 30 s and rinsing successively with water and ethanol. (Caution: “super piranha” solution reacts violently with organic materials and must be handled with extreme care.) Directly after cleaning the films were prepared in a sealed glass container under an atmosphere of nitrogen by immersing wafer pieces into an ethanolic solution (0.4 mM) of the dimannoside disulfide (DMT2) at room temperature for 48 h. After immersion, samples were rinsed with ethanol, sonicated in ethanol (15 mL) for 3 min, and rinsed again with ethanol. Immediately after drying in a stream of nitrogen, the samples were measured by laboratory XPS. For transport to BESSY II, IR, and WCA analysis, the samples were stored in a sealed container under nitrogen atmosphere in the dark. X-ray Photoelectron Spectroscopy (XPS). XPS survey spectra were measured with an AXIS Ultra DLD electron spectrometer manufactured by Kratos Analytical (Manchester, U.K.) using monochromated Al KR excitation at 80 eV pass energy. The electron emission angle was 0° and the source-to-analyzer angle was 60°. The binding energy scale of the instrument was calibrated following a Kratos Analytical procedure which uses ISO 1547232 binding energy data. Spectra were taken by setting the instrument to the hybrid lens mode and the slot mode, providing approximately a 300 700 μm2 analysis area. Synchrotron radiation XPS (SR XPS) measurements were carried out with an Scienta
ARTICLE
3000 electron energy analyzer at the endstation of the HE-SGM monochromator dipole magnet beamline at the synchrotron radiation source BESSY II (Berlin, Germany). High-resolution core-level spectra were recorded in FAT (fixed analyzer transmission) mode at pass energies of 20 eV (C 1s, S 2p, Au 4f) or 50 eV (O 1s) and the following excitation energies: survey scan 700 eV, O 1s 620 eV, C 1s 385 eV, and S 2p 285 eV, respectively. The respective excitation energy is given for each spectrum in the Results and Discussion section. Binding energy (BE) scales were referenced to the Au 4f7/2 peak at 84.0 eV using the Au 4f spectra measured at the respective excitation energy. The electron emission angle was 10°, except for angle-resolved SR XP spectra that were measured at three different emission angles (0°, 60°, and 70°). All spectra were analyzed using the CasaXPS peak fit program, version 2.3.15. In curve fitting of C 1s and O 1s spectra, full widths at halfmaximum (fwhms) were constrained to be equal, resulting in fwhm values ranging from 0.9 to 1.5 eV. A Gaussian/Lorentzian product function peak shape model GL(30) (70% Gaussian, 30% Lorentzian) was used in combination with a Shirley background. S 2p core level spectra were fitted with a minimum set of doublets of equal fwhm, a doublet separation of 1.2 eV, and an S 2p3/2/S 2p1/2 peak area ratio of 2:1.33,34 XPS spectra were simulated using the Simulation of Electron Spectra for Surface Analysis (SESSA) software Version 1.2 [National Institute of Standards and Technology (NIST), Gaithersburg, MD].35 For the SESSA simulation, inelastic mean free path values calculated with the TPP-2 M formula,25 Scofield photoionization cross sections,36 and appropriate density data [gold, 19.3 g/cm3; 1-hexanethiol, 0.84 g/ cm3; dimannose, 1.76 g/cm3 (calculated using the freeware ACD/ ChemSketch 12.0)] were used. For the SESSA simulations, a threelayer model was used that is based on an energy-minimized structure of the DMT molecule generated with the CambridgeSoft ChemBio3D Ultra 12.0 software.
Time-of-Flight Secondary Ion Mass Spectrometry (ToFSIMS). Static SIMS spectra were obtained using the time-of-flight secondary ion mass spectrometer ToF-SIMS IV, made by Ion-Tof GmbH (M€unster, Germany). Primary ion bombardment was done by 25 keV Gaþ ions with a pulsed current of 0.68 pA (repetition rate 10 kHz). The ion beam was digitally scanned with a 128 128 array over an analysis area of 101 μm 101 μm, respectively. The total acquisition time was fixed to 100 s.
Near-Edge X-ray Absorption Fine Structure (NEXAFS). NEXAFS spectroscopy was carried out at the HE-SGM monochromator dipole magnet beamline at the synchrotron radiation source BESSY II (Berlin, Germany). Spectra were acquired at C and O K-edges in the PEY (partial energy electron yield) mode using a channel plate detector with a retarding field of 150 eV.37 The resolution E/ΔE of the monochromator at the carbonyl π* resonance (hν = 287.4 eV) was on the order of 2500. Raw spectra were divided by the monochromator transmission function, which was obtained with a freshly sputtered Au sample.37 C and O K-edges were recorded at an angle of 55° measured between the surface plane of the sample and the direction vector of the incident linearly polarized light beam. Energy alignment of the energy scale was achieved by using an I0 feature referenced to a C 1sf π* resonance measured with a fresh surface of HOPG (highly ordered pyrolytic graphite, Advanced Ceramic Corp., Cleveland, OH) at 285.4 eV.38 Normalized spectra are shown in units of the absorption edge jump after subtraction of the pre-edge count rate.37 FT-IR Spectroscopy. FT-IR spectra of modified gold samples were collected at grazing incidence (80°) using a Bruker Vertex70 system equipped with a horizontal reflection unit A518/Q with polarized light (90°). Cleaned, unmodified gold samples were used to generate background spectra. Spectra were acquired with an aperture of 6 mm; a total of 64 scans with 2 cm1 resolution were averaged for each spectrum. All FT-IR spectra were analyzed using the OPUS software. 4809
dx.doi.org/10.1021/la104038q |Langmuir 2011, 27, 4808–4815
Langmuir
ARTICLE
Water Contact Angle (WCA). Static contact angle measurements were performed on a Kr€uss contact angle measuring system G2 using the sessile drop method. All measurements were done at ambient temperature and humidity. Six independent samples were analyzed. At least four drops of Milli-Q water with a volume of 3 μL each were placed on the samples. The standard deviation of the mean contact angle data was
