Cavitand-Functionalized Porous Silicon as an Active Surface for

Dec 20, 2011 - This paper reports on the preparation of a porous silicon-based .... of the Entrapment of a Series of Organophosphonates in Aqueous Med...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/Langmuir

Cavitand-Functionalized Porous Silicon as an Active Surface for Organophosphorus Vapor Detection Cristina Tudisco,† Paolo Betti,‡ Alessandro Motta,† Roberta Pinalli,‡ Luigi Bombaci,† Enrico Dalcanale,*,‡ and Guglielmo G. Condorelli*,† †

Dipartimento di Scienze Chimiche, Università di Catania and INSTM UdR di Catania, v.le A. Doria 6, 95125 Catania, Italy Dipartimento di Chimica Organica e Industriale, Università di Parma and INSTM UdR di Parma, Parco Area delle Scienze 17/A, 43124 Parma, Italy

‡

S Supporting Information *

ABSTRACT: This paper reports on the preparation of a porous silicon-based material covalently functionalized with cavitand receptors suited for the detection of organophosphorus vapors. Two different isomeric cavitands, both containing one acid group at the upper rim, specifically designed for covalent anchoring on silicon, were grafted on H-terminated porous silicon (PSi) by thermal hydrosilylation. The covalently functionalized surfaces and their complexation properties were characterized by combining different analytical techniques, namely X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), and mass spectroscopy analysis coupled with thermal desorption experiments. Complexation experiments were performed by exposing both active surfaces and a control surface consisting of PSi functionalized with a structurally similar but inactive methylene-bridged cavitand (MeCav) to dimethyl methylphosphonate (DMMP) vapors. Comparison between active and inactive surfaces demonstrated the recognition properties of the new surfaces. Finally, the nature of the involved interactions, the energetic differences between active and inactive surfaces toward DMMP complexation, and the comparison with a true nerve gas agent (sarin) were studied by DFT modeling. The results revealed the successful grafting reaction, the specific host−guest interactions of the PSi-bonded receptors, and the reversibility of the guest complexation.

■

INTRODUCTION Organophosphonates (OPs) are a family of organic compounds bearing phosphorus−oxygen bonds that are often highly toxic because they can inhibit the activity of the critical enzyme acetylcholinesterase.1 In recent years, OPs have aroused interest in environmental and security fields due to their use in pesticides and as chemical warfare agents, thus spurring a great effort toward the detection, monitoring, and decontamination of environments containing toxic OPs. Many studies have been reported on the detection of these OPs, adopting a wide range of transduction signals based on electric properties,2,3 luminescent dyes4 reflectivity changes in photonic crystals based on multistructured porous silicon,5−8 surface acoustic waves,9 quartz crystal microbalances,10,11 smectic liquid crystals,12 microcantilevers,13 and using various organic and inorganic sorbents as active2,4 or preconcentration14 materials such as various metal oxides,3,11 polysiloxanes,10,15−17 organic polymers, 4 functionalized carbon nanotubes,18−20 SAM,21 and MOFs.22,23 Although all these schemes allow possible routes to detect OPs in the parts-per-million (ppm) range or lower, most surface-based techniques suffer from low selectivity and, in some cases, lack of reversibility.14,24 In most of the mentioned sorbents, OP complexation is mainly driven by the H bonding between OH sorbent groups and the PO © 2011 American Chemical Society

moiety in OPs, but this interaction is strongly affected by the presence of H2O,25 which reduces selectivity and sensitivity. Recently, cavitands have been used as efficient recognition films deposited on a surface plasmon resonance transducer to detect dimethyl methylphosphonate (DMMP),26 a nontoxic homologue of the nerve agents, which is used as a model compound to mimic their reactivity. In that study, the proposed combined effects due to the cavity and the H-bond interaction improved the complexation efficiency and reduced water interferences compared to that of commercial sorbents. An interesting approach for harnessing the full potential of molecular receptors consists of their arrangement in monolayers hosted on the silicon surface27−29 which, compared to both thin films and bulk materials, have the advantages of improving miniaturization and integration and of reducing nonspecific interactions which often mask the recognition events.30 In this context, cavitands have been recently bonded to flat Si(100) to perform molecular recognition tasks31 at the silicon−liquid32 and at the silicon−air33 interfaces. One of the limitations of monolayers on flat surfaces is related to the low surface area and, in turn, to Received: September 28, 2011 Revised: December 16, 2011 Published: December 20, 2011 1782

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

of the experimental results to sarin, an effective nerve gas agent used during the terroristic attack on a Tokyo subway in 1995.

the low amount of material involved in the recognition task thus requiring high sensitivity techniques32,34 or nanoscale devices13,20,35,36 to detect analyte signals. To overcome this limitation, sensing devices can be fabricated, combining the monolayer approach with high surface area substrates.7,37 Among these systems, porous silicon (PSi) is a material with unique properties, including, besides the mentioned high surface area, convenient surface chemistry, easy integration in silicon-based devices, and characteristic optical properties.5−8,38 Therefore, functionalization of PSi substrates with organic monolayers represents a theme of great interest for the synthesis of new functional materials for various applications39−42 and, in particular, for gas sensing in which high surface areas represent a significative advantage.5−8,43,44 In this paper, we report on a strategy for the detection of organophosphorus vapors which combines the supramolecular recognition properties of two cavitand receptors with the potentialities of PSi. The two adopted isomeric receptors, denoted AcOUT and AcIN (Figure 1), differ only in the orientation of the

■

EXPERIMENTAL SECTION

Materials. All chemicals, unless otherwise noted, were commercially available and used as received. Water used for porous silicon and monolayer preparations was a Milli-Q grade (18.2 MΩ cm) with a final filtering step through a 0.22 μm filter. MeCav with ω-decenyl feet was prepared according to a published procedure.47 Synthesis of Methyl Esters of AcIN and AcOUT Cavitands. AcIN and AcOUT cavitands suited for silicon anchoring and with the acid functions protected as methyl esters were synthesized according to a modified previously reported procedure.26 Briefly, decenyl-footed resorcinarene47 was tribridged with methylene units by reaction with a limited amount of bromochloromethane. The desired carboxylic acid was then introduced in the form of the methyl ester on the remaining pairs of phenolic OHs using methyl dichloroacetate as bridging reagent, to give isomeric cavitands AcOUT and AcIN methyl esters in pure form after chromatographic separation. The detailed synthetic procedure and compound characterizations are reported in the Supporting Information. Porous Silicon Preparation. Porous silicon (PSi) has been obtained by wet metal-assisted chemical etching according to the procedure reported by Chartier et al.48 Briefly, Czochralski-grown, (100)oriented, p-type, boron-doped Si slides with resistivity 1.5−4 Ω cm were dipped for 5 min in an aqueous solution of HF (0.14 M) and AgNO3 (5 × 10−4 M) to deposit Ag particles. The slides were then etched in aqueous solutions containing HF, H2O2, and ultrapure H2O (40% HF:30% H2O2:H2O 25:10:4 V/V) for 1 min, and after etching, samples were rinsed with ultrapure water and dried with prepurified N2. Thickness and morphology of the etched layer, checked by SEM cross-section (Supporting Information Figure S1) are consistent with those reported in the literature for Ag-assisted chemical etching.40,48 This method allows the reproducible preparation of PSi with easily accessible columnar pores formed according to the literature48−51 along the (100) direction. Monolayer Preparation. For grafting on porous substrates, PSi was etched in 10% HF solution for 1 h, washed with ultrapure water for 20s, dried with N2, and immediately placed in the methyl esterprotected cavitand solution in mesitylene. The solution was then refluxed at 200 °C for 5 h, under slow N2 bubbling to prevent bumping. After grafting, the sample was removed from the solution (after cooling at room temperature) and cleaned by two rinsing cycles of ultrasonic cleaner (5 min each) in dichloromethane. Note that SEM cross-section analysis of samples after grafting shows a morphology similar to that observed before grafting, indicating that the anchoring route does not degrade the structure of porous silicon, as previously reported.40 Cavitand Complexation Tests. Before the complexation tests, methyl-protected carboxylic functionalities of PSi-grafted AcIN and AcOUT cavitands were deprotected by hydrolysis of the methyl ester (details in Supporting Information) following a published procedure.52,53 For complexation experiments, samples were placed for 20 min in a closed chamber saturated with DMMP vapors. The saturated DMMP vapor was obtained by placing a beaker with DMMP (3 mL) in the closed chamber (total volume about 500 mL) at 20 °C overnight, thus allowing vapor saturation. Under this condition, the vapor pressure of DMMP is about 0.3 Torr, which was estimated from thermodynamic data, taking into account the relative humidity.54,55 After DMMP exposure, samples were transferred (within 5−10 min of air exposure) either to the FT-IR spectrometer or to the XPS chamber. For XPS analysis, samples were furthermore kept in a prechamber under turbomolecular pumping for 20 min before their introduction to the main chamber for analysis. Sample Characterization. XPS spectra were run with a PHI 5600 multitechnique ESCA-Auger spectrometer equipped with a monochromated Al Kα X-ray source. Analyses were carried out with a photoelectron angle of 45° (relative to the sample surface) with an acceptance angle of ±7°. The XPS binding energy (BE) scale was calibrated

Figure 1. Chemical structure of cavitands and guests.

acid groups at the upper rim, because both cavitands contain one acid group at the upper rim, pointing outward or inward with respect to the cavity. The capacity of thin films of these receptors to detect DMMP has been previously demonstrated.26 In the present work, each of these cavitands was functionalized in the lower rim with four undecylenic feet for surface anchoring and then grafted on H-terminated PSi surfaces via thermal hydrosilylation of the double bonds. In addition, to evaluate the role of unspecific interactions, control samples were prepared by grafting on PSi the methylene-bridged cavitand MeCav (Figure 1) which is structurally similar to both AcOUT and AcIN, but it lacks the carboxylic acid group and, for this reason, it is inactive in DMMP complexation (previous papers have already demonstrated the negligible complexation properties of MeCav toward various classes of organic vapors45,46). X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR) were adopted to probe the cavitand-grafting route on PSi and to evaluate the sensing properties of the functionalized PSi surfaces. Mass spectroscopy coupled with thermal desorption was also used to monitor the products desorbing from the surface. Host−guest interactions and energetic differences among DMMP complexes with active and inactive cavitands have been explored by DFT calculation. Theoretical modeling also allowed extension 1783

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

Figure 2. FTIR spectral regions between (a) 3100−2750 cm−1 (C−H stretching region), (b) 2300−1900 cm−1 (Si−H stretching region), and (c) 1350−600 cm−1(SiOx stretching and bending modes) of HF-etched PSi, PSi−MeCav, and PSi−AcIN. The transmission spectrum of Si(100) has been subtracted from these spectra as background.

■

by centering the C 1s peak due to hydrocarbon moieties and “adventitious” carbon at 285.0 eV.56,57 SEM analysis was performed with a LEO SUPRA 55VP equipped with a field emission gun. Transmission FTIR measurements were recorded on a JASCO FTIR 430, with 100 scans collected per spectrum (scan range 560−4000 cm−1, resolution 4 cm−1). Thermal desorption experiments have been performed in a UHV chamber (10−9 Torr) equipped with a resistive heated sample holder (Vacuum Science, Italy) and a quadrupole mass spectrometer Smart-IQ+ (Thermo Electron Corporation). During experiments, the holder was heated with a ramp of 10 °C/min from 30 °C to 130 °C, and the evolved gases were monitored with the quadrupole spectrometer operated with an electron filament ion source and a multiplier detector (mass range 1−300). Computational Details. DFT calculations used the Gaussian and plane wave mixed-basis method (GPW), as implemented in the QUICKSTEP module58 within the CP2K simulation package.59 A triple quality TZVP Gaussian basis set was employed for all the atoms. The Goedecker−Teter−Hutter pseudopotentials60 together with a 320 Ry plane wave cutoff were used to expand the densities obtained with the Perdew−Burke−Ernzerhof (PBE)61 exchange-correlation density functional. This parametrization allows a reliable prediction of the investigated host−guest interaction. Stabilization energies associated with the host−guest adduct formation are corrected, taking into account the BSSE that in all cases represents a very small contribution to the total interaction energy spanning the 2.1−2.7 kcal/mol range. The introduction of van der Waals (VdW) forces in the theoretical approximation is fundamental for the simulation of the cavitand−guest recognition process. In fact, when the VdW forces are neglected, the DMMP guest goes away from the cavity and the interaction of the guest with the inner part of the cavity is totally lost. This interaction is thus mainly produced by the VdW contribution, and it is estimated to be about 25 kcal/mol. VdW forces are taken into account with the Grimme 2D Method.62 Molecular geometry optimization of stationary points used the Broyden−Fletcher−Goldfarb−Shannon (BFGS) method without any geometry constraint. For the nonperiodic calculation of the system, the wavelet method63 was adopted as the cell decoupling procedure. Vibrational analysis was carried out to obtain IR information within the harmonic approximation approach. Anharmonic effects are taken into account by adding an anharmonic constant term of 80 cm−1.64,65 Ethyl groups replace the decyl chains in the bottom side of the cavitand. Natural bond orbital (NBO) analysis was performed by Gaussian code (6-311G**/PBE)66 to obtain information regarding charge distribution.

RESULTS AND DISCUSSION Cavitand-Functionalized PSi. FTIR and XPS were carried out to characterize PSi functionalized with AcOUT and AcIN (PSi−AcOUT and PSi−AcIN) and with MeCav (PSi−MeCav). Note that carboxylic groups in PSi−AcOUT and PSi−AcIN were methyl protected to avoid interaction between the upper cavity and the surface.67 Carboxylic groups were deprotected before the complexation tests. Because FTIR spectra of PSi−AcOUT and PSi−AcIN are almost identical, only PSi−AcIN spectra are reported. FT-IR spectra of HF-etched PSi, PSi−AcIN, and PSi−MeCav samples are compared in Figure 2. In particular, three spectral regions are reported: (a) the C−H stretching region between 3100 and 2750 cm−1, (b) the Si−H stretching region between 2300 and 1900 cm−1, and (c) the region between 1350 and 600 cm−1 which contains SiOx stretching vibrations and SiHx bending modes. The appearance in PSi−AcIN and PSi−MeCav spectra of intense bands due to the CH2 symmetric (νs(CH2)) and antisymmetric (νa(CH2)) stretches at 2858 cm−1 and 2926 cm−1 and a characteristic smaller band at 3005 cm−1 assigned by Friggeri et al.68 to the stretches of aromatic C−H (ν(CH)) in cavitands (Figure 2a), combined with the absence of bands in the same spectral region of HF-etched PSi, confirms the covalent grafting of the cavitands onto the porous silicon surface. In the region between 2300 and 1900 cm−1, three absorption peaks at 2140 cm−1, 2114 cm−1, and 2087 cm−1 due to Si−H3, Si−H2, and Si−H stretching modes, respectively,69,70 are present in the spectrum of HF-etched PSi (Figure 2b, bottom). In all cavitand-grafted samples, SiHx peaks decrease and appear as a single broad band (Figure 2b, upper and middle curves), due to the hydrosilylation reaction which determines a partial replacement of Si−H bonds by Si−C bonds. A low broad band at 2251 cm−1 corresponding to OSi−Hx stretches was, however, observed due to a partial silicon oxidation.40 The 1350−600 cm−1 region of the FTIR spectrum of HFetched PSi (Figure 2c, bottom curve) shows two intense bands at 660 cm−1 and 625 cm−1 that are due to the wagging vibration modes of SiH2 and SiH, respectively, and the SiH2 bending vibrations (δSiH2) at 910 cm−1.71 In the spectrum of grafted samples (Figure 2c, middle and upper curves), the intensity of 1784

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

The C 1s region of the as-prepared PSi sample (Figure 3a) consists of two components. The first component is centered at 285.0 eV, and it represents aliphatic and aromatic hydrocarbons.72 The second component is centered at 286.5 eV, and it is attributed to carbon bonded to one oxygen. The latter component of this “adventitious” carbonaceous layer was attributed to the formation of Si−O−C groups due to the reaction between Si−H and Si−OH groups on the silicon surface with oxidized carbon species.27 The C 1s spectra of both PSi−MeCav and PSi−AcIN (Figure 3b and 3c, respectively) consist of three main components. The first two components at 285.0 and 286.3 eV are analogous to those observed in PSi, even though they have higher intensity. The first component is due to the aliphatic and aromatic hydrocarbon backbone of cavitands, while the component at 286.3 eV is due to oxidized carbons in the resorcinarene phenyl rings bonded to one oxygen atom.27,73 A third component at 288.4 eV, only observed in cavitand-decorated surfaces, can be used as a fingerprint of the presence of cavitand molecules, and it represents the four methylene groups bridging the oxygen atoms at the upper rim.27 In the PSi−AcIN sample, a component due to a carboxylic carbon can be added at 289.2 eV, but it is too close to the more intense component at 288.4 eV to single out its contribution. O1s signal observed in PSi substrates consists of a single component at 532.1 eV, which is due to a partial oxidation of the substrates. The O1s band of functionalized PSi (Figure 4) consists instead of two components. The first at 532.1 eV is due to Si surface partial oxidation,73 similar to the PSi sample. The second is centered at 533.5 eV and represents the Ph−O−C arrangement in cavitands.27 DMMP Complexation: XPS, FTIR, and Thermal Desorption Study. Gas−solid complexation tests were performed, exposing PSi−AcIN and PSi−AcOUT active surfaces and PSi− MeCav as a complexation-inactive reference to DMMP vapors (about 0.3 Torr). DMMP uptake was, then, evaluated by a surface-sensitive technique (XPS) and by transmission FTIR, taking advantage of the high surface area of the porous samples. Figure 5 shows FTIR difference spectra (before and after DMMP exposure) of PSi−AcIN, PSi−AcOUT, and PSi−MeCav

these features decreases, analogous to the Si−H stretching signals, due to the hydrosilylation reaction. The 1350− 600 cm−1 region of the PSi−AcIN spectrum is dominated by a broad band around 1100 cm−1 that is assigned to the vibrational modes of SiO2.40 Although not all Si−H sites are replaced by Si−C bonds and SiOx is also formed as expected for silicon grafting of molecules with large steric hindrances,27,31 overall FTIR results indicate the successful grafting of cavitands on the Si surface. Cavitand-functionalized PSi-based materials (PSi−AcIN, PSi−AcOUT, and PSi−MeCav) were also characterized by XPS. Elemental compositions of as-grafted samples and of HF-etched PSi substrates are reported in Table 1. Table 1. Elemental Composition of Porous Silicon Samples before and after the Grafting of AcIN, AcOUT, and MeCav Cavitands atomic concentration HF-etched PSia PSi−AcIN PSi−AcOUT PSi−MeCav

Si 2p

O 1s

C 1s

76.7 25.4 21.7 29.0

7.6 27.8 28.1 21.2

15.7 46.8 50.2 49.8

a

A small amount of residual Ag (Ag < 0.2%) was found in the PSi substrates.

XPS data show that C 1s signals in cavitand-decorated surfaces increase significantly compared to C 1s-related signal of the HF-etched PSi sample, as expected after the grafting of a cavitand monolayer.31,40 High resolution C 1s and O 1s photoelectron spectral regions of HF-etched PSi, as-grafted PSi− AcIN, and PSi−MeCav are compared in Figures 3 and 4.

Figure 3. High-resolution C 1s XPS spectra (takeoff angle 45°) of (a) HF-etched PSi, (b) PSi−MeCav, and (c) PSi−AcIN.

Figure 4. High-resolution O 1s XPS spectra (takeoff angle 45°) of (a) PSi−MeCav and (b) PSi−AcIN. Figure 5. Difference FTIR spectra (between the spectrum after DMMP exposure minus the spectrum before DMMP exposure) of PSi−AcIN, PSi−AcOUT, and PSi−MeCav samples in the 1380−780 cm−1 region.

Analogous XPS spectra of as-grafted PSi−AcOUT are similar to those of PSi−AcIN. 1785

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

in the 1380−780 cm−1 region, where the main DMMP features are present.74,75 Spectra of PSi−AcIN and PSi−AcOUT show a characteristic strong doublet at 1035 cm−1 and 1058 cm−1 due to the stretching ν(C−O−P) and a band at 1202 cm−1 assigned to the stretching of the PO group. Other bands at 787 cm−1 and 828 cm−1 due to the stretching in the O−P−O group (ν(PO2)) and at 918 cm−1 and 1314 cm−1, respectively, assigned to rocking of the CH3P group (ρ(CH3P)) and the antisymmetric deformation of CH3P group (δ(CH3P))17 can be observed in the spectra. Much less intense signals were detected on PSi− MeCav because MeCav is inactive toward DMMP complexation and the residual adsorption is likely due to unspecific interactions. It is worthy of note that in all spectra the ν(PO) is significantly shifted at lower wavenumber compared to the frequencies of free gaseous or liquid DMMP (1275 cm−1 or 1240 cm−1, respectively), likely due to the formation of hydrogen bonds during DMMP complexation.76 These results suggest that DMMP uptake is much higher on PSi−AcIN and PSi−AcOUT in which the presence of a COOH group in the cavity upper rim allows the formation of an H-bond with the PO group of the DMMP molecule inside the cavity (vide infra DFT modeling). On PSi−MeCav, a negligible DMMP uptake occurs due to the presence of residual acid OH groups40 on the PSi-oxidized surface. The complexation process on the active surfaces has been also demonstrated by XPS. Figure 6 compares P 2p spectral

cycles. A strong decrease of ν(C−O−P) and ν(PO) bands in the FTIR spectra of DMMP-exposed PSi−AcIN and PSi−AcOUT samples is observed after heating (60 °C) or N2 rinsing for 15 min, thus indicating that decomplexation can be obtained either by thermal or by rinsing treatments. Figure 7 shows the evolution of

Figure 7. Trend of the absorbance values of the characteristic ν(C− O−P) band (at 1035 cm−1) during DMMP complexation/decomplexation cycles on PSi−AcIN.

the intensity of the ν(C−O−P) band in the FTIR spectra of PSi−AcIN (or PSi−AcOUT) samples undergoing various complexation/decomplexation cycles of DMMP exposure (about 0.3 Torr) followed by heating at 60 °C for 15 min. Although within the adopted conditions DMMP removal was not complete, the significant difference between FTIR signals of complexed and decomplexed surfaces is an indication of a good process reversibility and, hence, of a possible use of these materials in detectors for onsite field OP monitoring. Thermal desorption of DMMP molecules from PSi−AcIN, PSi−AcOUT, and PSi−MeCav samples was also monitored by in situ mass spectroscopy. Various typical ions associated with fragmentation of DMMP78 in the mass spectrometer were observed during the heating ramp (Figure 8).

Figure 6. P 2p spectral regions of PSi−AcIN (above), PSi−AcOUT (middle), and PSi−MeCav (bottom) after exposure to DMMP vapors. Spectra have been normalized to the total Si 2p intensity.

regions of PSi−AcIN, PSi−AcOUT, and PSi−MeCav after exposure to DMMP vapors. In both active surfaces, the presence of the P 2p band centered at 134.8 eV indicates, according to FTIR experiments, the presence of molecular DMMP on the surface.77 No P 2p signals are evident in PSi−MeCav, thus indicating that DMMP is not strongly retained on this surface. Note that the observed differences between response signals from complexation-active PSi−AcIN and PSi−AcOUT and from structurally analogous but complexation-inactive PSi− MeCav indicate that on PSi−AcIN and PSi−AcOUT DMMP uptake is strongly enhanced by the specific interactions with AcIN and AcOUT receptors. Unspecific adsorptions such as capillary condensation, physisorption phenomena, and DMMP H-bonding on residual SiOH terminations, which can be considered similar on both active and inactive surfaces, lead to the poor FT-IR signals observed for PSi−MeCav and to nondetectable XPS peaks. Because onsite field monitoring demands a repeatable uptake/ release of organophosphorus vapors, process reversibility has been investigated by FTIR monitoring of complexation/decomplexation

Figure 8. Ion intensity during thermal desorption from (a) PSi−AcIN and from (b) PSi−MeCav samples exposed to DMMP vapors. Thermal desorption from PSi−AcOUT is found comparable to that from PSi−AcIN.

In particular, the parent DMMP ion (124 amu) and the other intense cracking fragments, such as 79 (PO3), 94 (CH3PO3), and 109 (CH3)2PO3) amu, were observed with similar temperature profiles, thus suggesting the desorption of intact DMMP from the surface. A similar desorption experiment, carried out on PSi−MeCav after exposure to DMMP, showed a much lower 1786

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

DMMP is involved in a 3-fold CH−π interaction with the benzene fragments of the cavitand. This interaction results from the distance between methyl H atoms of the DMMP and the benzene (Be) centroid (2.38−2.58 Å) and from the C−H···Becentr angles (141−157.0°). Geometrical constraints do not allow optimized CH−π interactions of all methyl C−H bonds with the cavitand benzene fragments. From a dynamical point of view, this arrangement can be explained by the geometrical availability of all cavitand benzene fragments with respect to the CH−π interactions due to both the mean distance of the DMMP methyl group with the inside cavitand walls and the possibility for the methyl group to turn around the P−C(Me) single bond. All these interactions yield to an overall stabilization of 32.7 kcal/mol in the adduct formation between DMMP and AcOUT. Similar interactions are observed in the case of AcIN. The computed stabilization associated to the adduct formation is also comparable (33.7 kcal/mol). The computational data are in agreement with both SPR26 and present porous silicon sensing experiments, indicating that AcIN and AcOUT are comparable as receptors, i.e., the orientation of the COOH group with respect to the cavity does not affect DMMP binding. Vibrational analysis has been computed to acquire information about the effect of the H-bonding on the PO stretching of the DMMP in the host−guest adduct. FTIR spectroscopic measurements revealed an evident frequency shift (Δ = −40 cm−1) passing from 1242 cm−1 for the liquid DMMP to 1202 cm−1 for DMMP−AcOUT (or AcIN) adduct. A similar shift (Δ = 60.2 cm−1) of the PO stretching vibrational frequency upon passing from free DMMP (νPO = 1253.6 cm−1) to the DMMP hosted in the cavitand (νPO = 1193.4 cm−1) is computed as a consequence of the H-bonding interaction. A similar adduct has been computed with the MeCav host as inactive reference in the perspective to confirm the role of the acid group in the recognition of the DMMP molecule. In this case, similar interactions of the O2 and O3 DMMP atoms with the acid methylene C−H groups of the cavitand as well as the CH−π interactions are active. The computed energy stabilization of the DMMP−MeCav adduct is, however, much lower (24.2 kcal/mol) than that of the DMMP−AcOUT (or AcIN) adduct. The difference (Δ ∼ 10 kcal/mol) is ascribable to the H-bond promoted by the acid functionality on the upper rim of the active cavitands. The acid functionality is, hence, responsible for the observed affinity of the AcOUT and AcIN cavitands for the DMMP guest. Finally, to verify the performance of these receptors with respect to real nerve gases (experimentally very dangerous), the adduct formation between AcIN (or AcOUT) and a sarin molecule has been modeled. In this case, a slightly greater stabilization (35.6 kcal/mol and 37.3 for sarin−AcIN and sarin− AcOUT, respectively) with respect to both the DMMP−AcIN and the DMMP−AcOUT adducts has been computed as a consequence of a greater affinity between sarin and AcIN (or AcOUT) cavitands. This result gave us confidence regarding the correct usage of DMMP as a suitable and harmless model for the study of nerve gas agent sensing.

amount of desorbed molecules from the surface (Figure 8b), further confirming the poor DMMP uptake on PSi−MeCav surfaces. Organophosphorus Vapor Complexation: Theoretical Modeling. DFT modeling was performed to investigate the nature of the interactions involved in the DMMP uptake in the cavitands. The analysis allowed the explanation of the observed similar behavior of AcIN and AcOUT receptors and the estimation of energetic differences between these receptors and the reference MeCav. Charge analysis (Table 2) shows that the interaction between DMMP and AcOUT in the host−guest adduct involves a Table 2. NBO Atomic Charge Assignment on the Free DMMP Molecule and in the DMMP−AcOUT Adducta free DMMP

DMMP−AcOUT

2.167 −1.003 −0.792 −0.792

2.235 −1.070 −0.799 −0.808

P O1 O2 O3 a

Label refers to Figure 9.

general displacement of the electron density distribution in the DMMP molecule. In particular, the negative charge centered on the O1 atom of the DMMP increases (Δ = 0.67) as a consequence of the H-bonding formation with the cavitand HAc atom (Figure 9). The O1−HAc distance (1.73 Å) and the O−HAc···O1

Figure 9. Host−guest adducts modeled in the present work. Carbon in gray, hydrogen in blue, oxygen in red, phosphorus in green, and fluorine in pink. The host−guest interaction energy (in kcal/mol) is placed under each structure. Ethyl groups replace the alkyl chains in the bottom side of the cavitand.

angle (172.1°) confirm the presence of the H-bond between DMMP and the cavitand. A lower increase in the negative charge (Δ = 0.02) is also observed upon O2 and O3 oxygen atoms of the DMMP upon passing from free to the DMMP−AcOUT adduct. Also in this case, the oxygen atoms O2 and O3 point toward the C−H bond of the cavitand methylene groups (O···H−C ∼ 167°). A tenuous interaction with the methylene H atoms (weakly acid in character) of the cavitand upper rim can explain the adduct charge distribution and the geometrical arrangement. Finally, the methyl group directly bonded to the P atom of the

■

CONCLUSIONS In this paper, we described the synthesis, characterization, and complexation properties of functional hybrid materials consisting of high area porous silicon substrates modified with cavitand receptors. The recognition properties of both AcIN and AcOUT, first used as solid film deposited for SPR detection on 1787

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

gold substrates,26 have been successfully transferred to a siliconbased material relevant for device integration by covalent functionalization with suitable cavitands possessing undecylenic feet. Material characterization showed the success of the cavitand grafting on a porous substrate, thus obtaining a new active surface which combines the typical robustness of Si−C covalent bonds79 with the high surface area of PSi. Reversible host−guest complexation of DMMP on a functionalized porous surface has been evaluated by XPS, FTIR, and desorption experiments, while DMMP/surface nonspecific physisorption is proven negligible by blank experiments adopting a nonactive methylene-bridged cavitand. Results showed that the developed cavitand-decorated surfaces retain the characteristic features of the supramolecular host− guest interactions such as complexation properties and reversibility observed for bulk cavitands. Note that the preparation of high surface area materials, compared to monolayers on flat surfaces, allows simpler analytical tools (i.e., transmission IR techniques) for both DMMP detection and study of the DMMP−AcIN (and DMMP−AcOUT) complexation modes. In this context, DFT modeling supported by FT-IR characterization showed the nature of the involved interactions and estimated energetic differences between complexation on different surfaces. Moreover, the theoretical approach also pointed out the efficiency of the cavitand recognition properties toward a real nerve gas (sarin). Finally, the transfer of the cavitand recognition properties into a silicon framework can be considered a starting point toward the development of an integrated sensing device for the recognition of organophosphorus toxic agents.

■

(5) Jang, S.; Koh, Y.; Kim, J.; Park, J.; Park, C.; Kim, S. J.; Cho, S.; Ko, Y. C.; Sohn, H. Mater. Lett. 2008, 62, 552−555. (6) Sohn, H.; Lètant, S.; Sailor, M. J.; Trogler, W. C. J. Am. Chem. Soc. 2000, 122, 5399−5400. (7) Ruminsky, A. M.; Moore, M. M.; Sailor, M. J. Adv. Funct. Mater. 2008, 18, 3418−3426. (8) Jang, S.; Koh, Y.; Kim, J.; Jung, K.; Sohn, H. J. Korean Phys. Soc. 2009, 55, 294−298. (9) Du, X.; Ying, Z.; Jiang, Y.; Liu, Z.; Yang, T.; Xie, G. Sens. Actuators, B 2008, 134, 409−413. (10) Hea, W.; Liua, Z.; Dua, X.; Jianga, Y.; Xiao, D. Talanta 2008, 76, 698−702. (11) Zhao, Y.; Hea, J.; Yanga, M.; Gaoa, S.; Zuoa, G.; Yana, C.; Chenga, Z. Anal. Chim. Acta 2009, 654, 120−126. (12) Yang, K.-L.; Cadwell, K.; Abbott, N. L. Sens. Actuators, B 2005, 104, 50−56. (13) Yang, Y.; Ji, H.-F.; Thundat, T. J. Am. Chem. Soc. 2003, 125, 1124−1125. (14) Kanan, S. M.; Tripp, C. P. Langmuir 2001, 17, 2213−2218. (15) Du, X.; Wang, Z.; Huang, J.; Tao, S.; Tang, X.; Jiang, Y. J. Mater. Sci. 2009, 44, 5872−5876. (16) Huang, J.; Jiang, Y.; Du, X.; Bi, J. Sens. Actuators, B 2010, 146, 388−394. (17) Ferguson-McPherson, M. K.; Low, E. R.; Esker, A. R.; Morris, J. R. J. Phys. Chem. B 2005, 109, 18914−18920. (18) Lee, C. Y.; Sharma, R.; Radadia, A. D.; Masel, R. I.; Strano, M. S. Angew. Chem., Int. Ed. 2008, 47, 5018−5021. (19) Kong, L.; Wang, J.; Fu, X.; Zhong, Y.; Meng, F.; Luo, T.; Liu, J. Carbon 2010, 48, 1262−1270. (20) Delalande, M.; Clavaguera, S.; Toure, M.; Carella, A.; Lenfant, S.; Deresmes, D.; Vuillaumeb, D.; Simonato, J.-P. Chem. Commun. 2011, 47, 6048−6050. (21) Bertilsson, L.; Engquist, I.; Liedberg, B. J. Phys. Chem. B 1997, 101, 6021−6027. (22) Zou, R.; Zhong, R.; Han, S.; Xu, H.; Burrell, A. K.; Henson, N.; Cape, J. L.; Hickmott, D. D.; Timofeeva, T. V.; Larson, T. E.; Zhao, Y. J. Am. Chem. Soc. 2010, 132, 17996−17999. (23) Montoro, C.; Linares, F.; Procopio, E. Q.; Senkovska, I.; Kaskel, S.; Galli, S.; Masciocchi, N.; Barea, E.; Navarro, J. A. R. J. Am. Chem. Soc. 2011, 133, 11888−11891. (24) Tesfai, T. M.; Sheinker, V. N.; Mitchell, M. B J. Phys. Chem. B 1998, 102, 7299−7302. (25) Bermudez, V. M. Langmuir 2010, 26, 18144−18154. (26) Daly, S. M.; Grassi, M.; Shenoy, D. K.; Ugozzoli, F.; Dalcanale, E. J. Mater. Chem. 2007, 17, 1809−1818. (27) Condorelli, G. G.; Motta, A.; Favazza, M.; Fragalà, I. L.; Busi, M.; Menozzi, E.; Dalcanale, E.; Cristofolini, L. Langmuir 2006, 22, 11126−11133. (28) Bunimovich, Y. L.; Shin, Y. S.; Yeo, W.-S.; Amori, M.; Kwong, G.; Heath, J. R. J. Am. Chem. Soc. 2006, 128, 16323−16331. (29) Narducci, D.; Bernardinello, P.; Oldani, M. Appl. Surf. Sci. 2003, 212−213, 491−496. (30) Zampolli, S.; Elmi, I.; Mancarella, F.; Betti, P.; Dalcanale, E.; Cardinali, G. C.; Severi, M. Sens. Actuators, B 2009, 141, 322−328. (31) Motta, A.; Tudisco, C.; Condorelli, G. G. Sci. Adv. Mater. 2011, 3, 362−377. (32) Biavardi, E.; Favazza, M.; Motta, A.; Fragalà, I. L.; Massera, C.; Prodi, L.; Montalti, M.; Melegari, M.; Condorelli, G. G.; Dalcanale, E. J. Am. Chem. Soc. 2009, 131, 7447−7455. (33) Condorelli, G. G.; Motta, A.; Favazza, M.; Gurrieri, E.; Betti, P.; Dalcanale, E. Chem. Commun. 2010, 46, 288−290. (34) Dubey, M.; Bernasek, S. L.; Schwartz, J. J. Am. Chem. Soc. 2007, 129, 6980−6981. (35) Elfström, N.; Karlström, E. A.; Linnros, J. Nano Lett. 2008, 8, 945−949. (36) Yue, M.; Stachowiak, J. C.; Lin, H.; Datar, R.; Cote, R.; Majumdar, A. Nano Lett. 2008, 8, 520−524. (37) Chen, B.; Xiang, S.; Qian, G. Acc. Chem. Res. 2010, 43, 2115− 2124.

ASSOCIATED CONTENT

* Supporting Information S

Preparation and characterization of cavitand methyl esters. SEM cross-section of PSi. Experimental details of the hydrolysis of surface-bonded cavitand methyl esters and relative FTIR spectra. A complete list of Cartesian coordinates of the main structures presently analyzed. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (G.C.); enrico.dalcanale@ unipr.it. (E.D.).

■

ACKNOWLEDGMENTS The authors thank Ministero dell’Istruzione, dell’Università e della Ricerca (MIUR), for financial support through PRIN (2008) “Strutture molecolari e nanocrystalline con funzionalità magnetiche, foto-magnetiche e foto-emettitrici e loro organizzazione su superfici, in film polimerici o in sol-gel”, and FIRB “ITALNANONET” RBPR05JH2P projects and CINECA (grant HP10B0R1E4) for the availability of high performance computing resources and support.

■

REFERENCES

(1) Chambers, H. W. In Organophosphates: Chemistry, Fate, and Effects; Chambers, J. E., Levi, P. E., Eds.; Academic Press: San Diego, CA, 1992; p 10. (2) Huang, J.; Miragliotta, J.; Becknell, A.; Katz, H. E. J. Am. Chem. Soc. 2007, 129, 9366−9376. (3) Lee, H. J.; Park, K. K.; Oralkan, Ö .; Kupnik, M.; Khuri-Yakub, B. T. IEEE 2008, 434−439. (4) Levitsky, I.; Krivoshlykov, S. G. Anal. Chem. 2001, 73, 3441− 3448. 1788

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789

Langmuir

Article

(38) Cullis, A. G.; Canham, L. T. Nature 1991, 353, 335−338. (39) Buriak, J. M. Philos. Trans. R. Soc. A 2006, 364, 217−225. (40) Buriak, J. M.; Stewart, M. P.; Geders, T. W.; Allen, M. J.; Choi, H. C.; Smith, J.; Raftery, D.; Canham, L. T. J. Am. Chem. Soc. 1999, 121, 11491−11502. (41) Tudisco, C.; Trusso Sfrazzetto, G.; Pappalardo, A.; Motta, A.; Tomaselli, G. A.; Fragalà, I. L.; Ballistreri, F. P.; Condorelli, G. G. Eur. J. Inorg. Chem. 2011, 13, 2124−2131. (42) Sam, S.; Touahir, L.; Salvador Andresa, J.; Allongue, P.; Chazalviel, J.-N.; Gouget-Laemmel, A. C.; Henry de Villeneuve, C.; Moraillon, A.; Ozanam, F.; Gabouze, N.; Djebbar, S. Langmuir 2010, 26, 809−814. (43) Shenoy, D. K.; Feresenbet, E. B.; Pinalli, R.; Dalcanale, E. Langmuir 2003, 19, 10454−10456. (44) Di Natale, C.; Paolesse, R.; Macagnano, A.; Nardis, S.; Martinelli, E.; Dalcanale, E.; Costa, M.; D’Amico, A. J. Mater. Chem. 2004, 14, 1281−1287. (45) Feresenbet, E. B.; Dalcanale, E.; Dulcey, C.; Shenoy, D. K. Sens. Actuators, B 2004, 97, 211−220. (46) Feresenbet, E. B.; Busi, M.; Ugozzoli, F.; Dalcanale, E.; Shenoy, D. K. Sensor Lett. 2004, 2, 186−193. (47) van Velzen, T. E. U.; Engbersen, J. F. J.; Reinhoudt, D. N. Synthesis 1995, 8, 989−997. (48) Chartier, C.; Bastide, S.; Lèvy-Clèment, C. Electrochim. Acta 2008, 53, 5509−5516. (49) Fang, H.; Li, X. D.; Song, S.; Xu, Y.; Zhu, J. Nanotechnology 2008, 19, 255703. (50) Huang, Z.; Shimizu, T.; Senz, S.; Zhang, Z.; Geyer, N.; Gösele, U. J. Phys. Chem. C 2010, 114, 10683−10690. (51) Huang, Z.; Geyer, N.; Werner, P.; de Boor, J.; Gösele, U. Adv. Mater. 2011, 23, 285−308. (52) Condorelli, G. G.; Motta, A.; Bedoya, C.; Di Mauro, A.; Pellegrino, G.; Smecca, E. Inorg. Chim. Acta 2007, 360, 170−178. (53) Strother, T.; Cai, W.; Zhao, X.; Hamers, R. J.; Smith, L. M. J. Am. Chem. Soc. 2000, 122, 1205−1209. (54) Tevault, D. E.; Buchanan, J. H.; Buettner, L. C. Int. J. Thermophys. 2006, 27, 486−493. (55) Butrow., A. B.; Buchanan, J. H.; Teveault, D. E. J. Chem. Eng. Data 2009, 54, 1876−1883. (56) Swift, I. L. Surf. Interface Anal. 1982, 4, 47−51. (57) Briggs, D.; Beamson, G. Anal. Chem. 1992, 64, 1729−1736. (58) Vandevondele, J.; Krack, M.; Mohamed, F.; Parrinello, M.; Chassaing, T.; Hutter, J. Comput. Phys. Commun. 2005, 167, 103−128. (59) The CP2K developers group. http://cp2k.berlios.de/ (60) Goedecker, S.; Teter, M.; Hutter, J. Phys. Rev. B 1996, 54, 1703−1710. (61) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78, 1396 (E). (62) Grimme, S. J. Comput. Chem. 2006, 27, 1787−1799. (63) Genovese, L; Deutsch, T; Neelov, A; Goedecker, S; Beylkin, G. J. Chem. Phys. 2006, 125, 074105/1−074105/5. (64) Digne, M.; Sautet, P.; Raybaud, P.; Euzen, P.; Toulhoat, H. J. Catal. 2004, 226, 54−68. (65) Raybaud, P.; Digne, M.; Iftimie, R.; Wellens, W.; Euzen, P.; Toulhoat, H. J. Catal. 2001, 201, 236−246. (66) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Hada, N. H.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;

Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 09, Revision B.1; Gaussian, Inc., Pittsburgh, PA, 2003. (67) Sieval, A. B.; Demirel, A. L.; Nissink, J. M.; Linford, M. R.; van der Maas, J. H.; de Jeu, W. H.; Zuilhof, H.; Sudhölter, E. J. R. Langmuir 1998, 14, 1759−1768. (68) Friggeri, A.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Langmuir 1998, 14, 5457−5463. (69) Choi, H. C.; Buriak, J. M. Chem. Mater. 2000, 12, 2151−2156. (70) Mattei, G.; Valentini, V.; Yakovlev, V. A. Surf. Sci. 2002, 502− 503, 58−62. (71) Alekseev, S. A.; Lysenko, V.; Zaitsev, V. N.; Barbier, D. J. Phys. Chem. C 2007, 111, 15217−15222. (72) Cerofolini, G. F.; Galati, C.; Reina, S.; Renna, L.; Viscuso, O.; Condorelli, G. G.; Fragalà, I. L. Mater. Sci. Eng., C 2003, 23, 989−994. (73) Briggs, D. Practical Surfaces Analysis, 2nd ed.; Briggs, D., Seah, M. P., Eds.; Wiley-VCH: Weinheim, Germany, 1995; Vol. 1, p 444. (74) Bertilsson, L.; Engquist, I.; Liedberg, B. J. Phys. Chem. B 1997, 101, 6021−6027. (75) Panayotov, D. A.; Morris, J. R. Langmuir 2009, 25, 3652−3658. (76) Kanan, S. M.; Waghe, A.; Jensen, B. L.; Tripp, C. P. Talanta 2007, 72, 401−407. (77) Ratliff, J. S.; Tenney, S. A.; Hu, X.; Conner, S. F.; Ma, S.; Chen, D. A. Langmuir 2009, 25, 216−225. (78) Chen, D. A.; Ratliff, J. S.; Hu, X.; Gordon, W. O.; Senanayake, S. D.; Mullins, D. R. Surf. Sci. 2010, 604, 574−587. (79) Wang, X.; Landis, E. C.; Franking, R.; Hamers, R. J. Acc. Chem. Res. 2010, 43, 1205−1215.

1789

dx.doi.org/10.1021/la203797b | Langmuir 2012, 28, 1782−1789