Etching of Crystalline ZnO Surfaces upon Phosphonic Acid Adsorption

Research Article www.acsami.org

Etching of Crystalline ZnO Surfaces upon Phosphonic Acid Adsorption: Guidelines for the Realization of Well-Engineered Functional Self-Assembled Monolayers Alexandra Ostapenko,† Tobias Klöffel,‡ Jens Eußner,§ Klaus Harms,§ Stefanie Dehnen,§ Bernd Meyer,*,‡ and Gregor Witte*,† †

Fachbereich Physik, Molekulare Festkörperphysik and Wissenschaftliches Zentrum für Materialwissenschaften (WZMW), Philipps-Universität Marburg, Renthof 7, 35032 Marburg, Germany ‡ Interdisciplinary Center for Molecular Materials (ICMM) and Computer-Chemistry-Center (CCC), Department of Chemistry and Pharmacy, Friedrich-Alexander-Universität Erlangen-Nürnberg, Nägelsbachstrasse 25, 91052 Erlangen, Germany § Fachbereich Chemie and Wissenschaftliches Zentrum für Materialwissenschaften (WZMW), Philipps-Universität Marburg, Hans-Meerwein-Strasse 4, 35043 Marburg, Germany S Supporting Information *

ABSTRACT: Functionalization of metal oxides by means of covalently bound self-assembled monolayers (SAMs) offers a tailoring of surface electronic properties such as their work function and, in combination with its large charge carrier mobility, renders ZnO a promising conductive oxide for use as transparent electrode material in optoelectronic devices. In this study, we show that the formation of phosphonic acid-anchored SAMs on ZnO competes with an unwanted chemical side reaction, leading to the formation of surface precipitates and severe surface damage at prolonged immersion times of several days. Combining atomic force microscopy (AFM), X-ray diffraction (XRD), and thermal desorption spectroscopy (TDS), the stability and structure of the aggregates formed upon immersion of ZnO single crystal surfaces of different orientations [(0001̅), (0001), and (101̅0)] in phenylphosphonic acid (PPA) solution were studied. By intentionally increasing the immersion time to more than 1 week, large crystalline precipitates are formed, which are identified as zinc phosphonate. Moreover, the energetics and the reaction pathway of this transformation have been evaluated using density functional theory (DFT), showing that zinc phosphonate is thermodynamically more favorable than phosphonic acid SAMs on ZnO. Precipitation is also found for phosphonic acids with fluorinated aromatic backbones, while less precipitation occurs upon formation of SAMs with phenylphosphinic anchoring units. By contrast, no precipitates are formed when PPA monolayer films are prepared by sublimation under vacuum conditions, yielding smooth surfaces without noticeable etching. KEYWORDS: self-assembled monolayer, ZnO, precipitation, etching, phosphonic acid, phosphonate

1. INTRODUCTION

A versatile approach to tailor interfacial properties is the use of molecular contact primer layers,6 such as self-assembled monolayers (SAMs).7 Prototypical systems are organothiolbased SAMs absorbed on noble metal surfaces, which can be easily prepared by the immersion technique and, therefore, have been studied in great detail.8 Similarly, covalently fixated SAMs have also been reported for metal oxides. Next to thiols,9−13 other anchoring units, for example, silanes,14−16 amines, 17,18 and carboxylic 19−24 or phosphonic acid (PA),10,13,25−36 are used to functionalize such oxidic surfaces. Especially for ZnO it has been shown that PA forms strong covalent bonds with the substrate, exhibiting higher stability than other functional units.10,13,35 PA-anchored SAMs have

Interfaces between inorganic and organic semiconductors have become a focus of intense research because of their importance for the realization of electronic hybrid devices. Of particular interest for such applications are transparent conductive metal oxides, which are widely used as transparent electrodes in light emission and display devices or solar cells.1 Among these materials, ZnO has gained special interest since its large charge carrier mobility offers an alternative to amorphous silicon for the fabrication of thin film transistors (TFT), while its lowtemperature processability also enables processing on flexible substrates.2,3 In addition, ZnO-based organic/inorganic heterostructures have been used for the fabrication of hybrid diodes.4 An important aspect for all such applications is the control of electronic interface properties, as they affect the energy level alignment and charge injection barriers, which is decisive for the performance of low-voltage electronic devices.5 © XXXX American Chemical Society

Received: February 22, 2016 Accepted: May 9, 2016

A

DOI: 10.1021/acsami.6b02190 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces been successfully used for specific electronic surface modifications of ZnO25,26,31,32,35,36 and also enable a covalent attachment of dye molecules. However, details of the microstructure, interface chemistry, and stability of corresponding SAMs on metal oxides are by far less thoroughly studied and understood than for metals. In part this is due to the overall low conductivity of metal oxides, which hampers highresolution STM measurements, as well as the common use of polycrystalline oxide substrates, i.e., sintered powders, in real devices. Aside from its ease, the preparation of organothiol-SAMs on gold substrates by immersion has the advantage to be a selfterminating process that ensures formation of monolayer films only. When providing sufficient thermal energy and time during preparation, this yields exceptionally well ordered SAMs of one phase only.37 By contrast, on oxidic substrates undesirable side reactions, such as partial dissolution of substrate surfaces, have been reported,13,20,25,26,38,39 which are made responsible for the failure and improper behavior of electronic devices.40,41 Therefore, a closer analysis of the immersion chemistry of such hybrid systems is required. Understanding the chemical robustness of ZnO surfaces is also of interest in the field of corrosion, since Zn/ZnO coatings are commonly used as protection layers for steel sheets, hence raising questions about their chemical stability and possible delamination.42 Here we have studied the formation of PA-based SAMs on differently oriented, well-defined ZnO single crystals with a particular emphasis on substrate integrity upon adsorption. Beside the development of monolayer films, a notable dissolution of all ZnO substrates was found, which leads to the growth of crystalline precipitates during prolonged immersion. By combining energy dispersive X-ray (EDX) analysis, thermal desorption spectroscopy (TDS), atomic force microscopy (AFM), and X-ray diffraction (XRD), we have identified this precipitate as zinc phenylphosphonate (ZnPP). In a second step, the energetics of this surface-etching process was studied using density-functional theory (DFT). The calculations show that the transformation of the PPA/SAMs into ZnPP precipitates is a thermodynamically favorable reaction. In addition, we have examined the influence of surface roughness and pH value of the solvent on the etching process. Finally, a reduced etching is observed for phosphinic acids (HPA), suggesting their use as alternative anchoring units.

Figure 1. Comparison of AFM (phase image) micrographs showing the morphology of ZnO−O samples immersed in ethanolic PPA solutions (a) using dried ethanol (immersion time 4 days) and (b) absolute ethanol (without drying, immersion time 9 h) and (c) a bare ZnO−O surface, together with corresponding topographic line scans (red curves). Note that the line profile for a bare sample is tilted for better visualization of the substrate surface steps. phosphonic acid (PPA, purity 98%, Sigma-Aldrich), dodecylphosphonic acid (DPA, purity 97%, ABCR), phenylphosphinic acid (PHPA, purity 99%, Sigma-Aldrich), and p-(trifluoromethyl)phenylphosphonic acid (TFPPA). The latter compound was synthesized as described in a previous work.44 The molecular adlayers were prepared by immersion of ZnO samples in 0.1 mM ethanolic solutions kept at 75 °C in a sealed jar. For this purpose dried ethanol as well as absolute ethanol (purity 99.9%, Sigma-Aldrich) was used. Preparation of dry ethanol, as well as subsequent immersion, was performed under an argon atmosphere with the exclusion of external moisture. After removal from solution, the substrates were thoroughly rinsed with pure ethanol and dried in a nitrogen stream. The pH value of ethanolic solutions was measured by nonbleeding pH-indicator strips (pH 0−6.0, Acilit, Merck KGaA, Darmstadt, Germany). To vary the pH value upon SAM fabrication, buffer solutions were prepared by dissolving sodium acetate (C2H3NaO2, purity 99.5%, Grüssing GmbH) and ethanoic acid (C 2 H 4 O2 , purity 99.8%, AnalaR NORMAPUR ACS) in distillated water (RiOs-DI-System). The buffer solution composition was selected according to pH values calculated by the Henderson−Hasselbalch equation, and the resulting pH values were verified by means of pH-indicator strips. The thermal stability of the SAMs and the formed surface precipitates was characterized by thermal desorption spectroscopy (TDS), using a quadrupole mass spectrometer (Balzer QMG 220) equipped with a Feulner cup. All mass spectra were acquired by linearly increasing the sample temperature at a rate of β = 0.5 K/s, which was measured by a thermocouple directly attached to the sample surface. The morphology of the surface precipitates was characterized by means of optical microscopy (Leitz, Metalloplan), scanning electron microscopy (SEM, JEOL, JSM-7500F), and atomic force microscopy (AFM, Agilent SPM 5500) operated in tapping mode under ambient conditions. Spatially resolved stoichiometric analysis of the surface precipitates was performed by energy dispersive X-ray (EDX) analysis using a scanning electron microscope (CamScan 4DV) equipped with an EDX microanalysis system (Noran Instruments Voyager 4.0). The crystalline phase and orientation of the surface precipitates were characterized by X-ray diffraction (Bruker AXS, Discover D8) using monochromatized Cu Kα radiation and a LynxEye silicon strip detector. In addition, a unit cell determination of the precipitates was performed with a single-crystal diffractometer (Bruker AXS, Quest D8) equipped with a microfocused Mo anode and a PHOTON 100 CMOS detector. 2.2. Computational Methods. Density-functional theory calculations were carried out with the periodic plane wave code PWscf of

2. METHODOLOGY 2.1. Experimental Details. For the present study, hydrothermally grown ZnO single crystals (CrysTec GmbH) were used that had been oriented and polished (roughness: rms 300 μm), obtained upon intentionally extended immersion times up to 1 month. They could be well-separated from the supporting ZnO crystals in an ultrasonic bath. The analysis yielded an orthorhombic crystal structure and unit cell parameters that are in close agreement with those reported for zinc phenylphosphonate (C6H5O3PZn)51 (cf. Table 1). A closer inspection reveals actually somewhat smaller lattice parameters and a slightly reduced cell volume for the precipitates as compared to the ZnPP bulk crystal structure. We attribute this difference to the presence of crystal water, which was reported for ZnPP bulk crystals,51−53 while it is essentially absent in the presently observed precipitations. Metal phenylphosphonates, such as ZnPP, feature a layered structure of (010) planes where metal cations are coordinated by the oxygen atoms from phosphonic acid PO3 groups, while the phenyl units act as an interlayer spacer (cf. Figure S5 in the Supporting Information). The rhombic crystallites, which are formed upon prolonged immersion, adopt a (010) orientation,

Figure 2. Morphology of surface precipitates formed on various ZnO surfaces upon immersion in PPA solutions using (absolute) ethanol: (a) on mixed terminated ZnO−M after 4 days of immersion (optical micrograph), (b) for ZnO−O and (d) for ZnO−Zn in each case after 2 days of immersion (SEM micrographs, top and side view). Panel (c) depicts an AFM phase image of an individual recumbent needle with corresponding topographic line scan.

precipitates was examined and compared with that of PPASAMs and crystalline PPA powder (cf. Figure 3). PPA powder

Figure 3. Series of thermal desorption spectra recorded for PPA powder (blue curve, detected mass signal m/z = 158 amu) and a PPASAM (red curve, mass signal m/z = 77 amu) during a linear heating rate of 0.5 K/s.

that was reamed onto the oxide sample starts to desorb at about 375 K and reveals a distinct desorption signal around 420 K. The largest signal was found for the mass of the molecule ion (m/z = 158 amu), which is indicative of an intact desorption, as expected for van der Waals bound PPA solids. By contrast, the PPA-SAM prepared by immersion remains rather stable up to more than 600 K and eventually desorbs dissociatively, causing a distinct desorption peak at the mass of the phenyl ring (m/z = 77 amu) around 690 K. Interestingly, the newly formed precipitates are thermally rather stable and still can be found on the surface, even after extensive heating up to 900 K in UHV. This enhanced thermal stability directly excludes crystalline PPA or any pure organic compound as a possible reason for the D


Research Article

ACS Applied Materials & Interfaces

Figure 4. Comparison of XRD θ/2θ scans of precipitates formed after immersion of ZnO samples in PPA solution: (a) for a ZnO−O sample after immersion in water-free ethanolic PPA solution for 4 days and (b) for a ZnO−M oriented crystal after different immersion times in (absolute) ethanolic PPA solution, which are compared with (c) a powder diffraction diagram calculated on the basis of the zinc phenylphosphonate (C6H5O3PZn, ZnPP) single crystal structure data.51 The optical micrographs in the inset show the morphological change of the precipitates with increased immersion time.

Figure 5. Top and side view of (a) a 0.5 ML of adsorbed PPA on the nonpolar ZnO−M surface and (b) a single layer of the ZnPP crystal. (c) Side view of the nonpolar (top) and polar (bottom) surfaces and a top view of a detached ZnO bilayer before and after relaxation of the lattice parameters.

we had started from a fully hydroxylated and water-covered ZnO−M surface54 and had adsorbed the PPA molecules via condensation reactions and displacement of water molecules. All PPA molecules are deprotonated and all surface O ions are converted to OH groups.55 Two O atoms of the PO3 unit are bound to surface Zn, and the third O atom is oriented across the trench of the ZnO−M surface and forms H-bonds with the adjacent OH-groups. Furthermore, this O atom sits right above a subsurface Zn ion, but no bond can form since this subsurface Zn ion is already 4-fold coordinated. If we compare the structure of the ZnO−M surface having adsorbed PPA molecules with that of a layer of ZnPP crystals (cf. Figure 5b), we can identify some similarities: the Zn ions form a centered rectangular lattice in both the ZnPP layer and the ZnO bilayers, the basic building block of the ZnO−M surface structure (cf. Figure 5c). Only the distances between the Zn ions differ. Thus, if a ZnO bilayer with adsorbed PPA molecules is detached from the ZnO−M surface and PPA molecules are adsorbed at the bottom in the same configuration as at the top (thereby introducing a mirror glide plane as is present in the ZnPP layer), then a structure is obtained which is already quite similar to the layers in the ZnPP crystal. The most important difference is, however, the PPA coverage. While the ZnO:PPA ratio is 1:1 in ZnPP layers, it

indicating that this is the most thermodynamically stable surface. We note that the (010) interlayer spacing of ZnPP of 14.3 Å is in excellent agreement with the step height of 1.4 nm obtained in the AFM micrograph for vertically oriented, crystalline, precipitated fibers (cf. Figure 2c). 3.2. Computational Results. Density-functional theory calculations were performed to analyze why the precipitates are formed. We first focus on the nonpolar, mixed-terminated ZnO−M surface and then generalize our results to the two polar ZnO−O and ZnO−Zn terminations. The most favorable binding mode of single PPA molecules on the ZnO−M surface is a bidentate configuration. The large distance between the Zn surface ions on ZnO−M prevents the formation of tridentate species. The interaction of the PPA molecules with the ZnO− M surface is very strong. With our PBE+D3 calculations we obtain a binding energy of 3.04 eV. Since each PPA molecule blocks two Zn surface sites, the highest coverage that can be achieved on the basis of a bidentate PPA binding mode is 0.5 ML. One of the most favorable configurations of PPA molecules on ZnO−M at 0.5 ML coverage is shown in Figure 5a. The binding energy per PPA molecule has increased to 3.54 eV. Note that the same structure would have been obtained if E


Research Article

ACS Applied Materials & Interfaces is 2:1 in the ZnO bilayer structure with adsorbed bidentate PPA molecules. Thus, to obtain a layer of the ZnPP crystal, the short Zn−Zn distance has to expand from the ZnO bulk lattice constant of a = 3.27 Å to 5.65 Å (the long Zn−Zn distance slightly shrinks from the ZnO bulk lattice constant of c = 5.27 Å to 4.84 Å) and an additional PPA molecule has to be inserted between neighboring PPA adsorbates. Simultaneously, deprotonation of the added PPA molecules converts the OH groups of the ZnO bilayer to water molecules as they are present in the ZnPP crystal (cf. Figure 5b). Are some of the additional PPA molecules already present on the ZnO−M surface? In the 0.5 ML structure of PPA on ZnO− M the available area for each PPA molecule is two surface unit cells (34.4 Å2). In the ZnPP crystal it is 27.4 Å2 and in the PPA crystal it is only 21.9 Å2. Thus, the packing density of the PPA molecules in the 0.5 ML structure is much lower than in the ZnPP crystal. So the 0.5 ML structure of PPA molecules on ZnO−M exhibits a certain degree of frustration: the phenyl rings prefer to be more closely packed; however, as long as the topmost ZnO bilayer is not detached from the ZnO surface, a higher PPA coverage can only be achieved if some of the PPA molecules forfeit their preferred bidentate binding mode and adopt a monodentate configuration. In the next step, we investigated the thermodynamic stability of such closely packed adsorbate layers. By using surface unit cells of size (4 × 2), (5 × 2), (6 × 2), and (7 × 2), we determined low-energy structures of PPA layers with 0.57, 0.6, 0.66, 0.71, 0.8, and 1 ML surface coverage. Indeed, the binding energy per molecule Eb decreases to 3.42, 3.38, 3.18, 3.11, 2.80, and 1.44 eV, respectively. All structures contain a certain fraction of monodentate-bound PPA molecules. These molecules are still protonated and the OH groups form very stable H-bonds to neighboring O atoms (cf. Figure 6). The thermodynamic stability of the PPA layers is analyzed by converting the binding energy Eb to the interface energy Δγ according to eq 2. Δγ is plotted in Figure 6 as a function of the PPA chemical potential ΔμPPA. At the chemical potential of ΔμPPA = −1.7 eV, which reflects the conditions in the immersion experiments, the thermodynamically most favorable surface coverage is 0.6−0.7 ML. At this coverage, the packing density of the phenyl rings is almost the same as in the ZnPP crystal. Higher coverages become unfavorable because the phenyl groups start to repel each other. In order to be able to accommodate a full PPA monolayer, as it would be required for the ZnPP crystal, the lattice constant a of the ZnO−M surface would have to expand from 3.27 to 5.65 Å as in the ZnPP crystal (cf. Figure 5). However, this can only happen if the upper ZnO bilayer detaches from the ZnO−M surface. The energy cost for cleaving a ZnO bilayer from the ZnO−M surface in vacuum is +1.59 eV per surface unit cell. In a vacuum, the detached ZnO bilayer structure is not stable. After relaxation of the lattice parameters the bilayer adopts a planar hexagonal honeycomb structure with lattice constant 3.28 Å, thereby lowering the energy by 0.14 eV per initial unit cell (two ZnO units, cf. Figure 5c). However, if we first saturate the ZnO−M surface with 0.6 ML of PPA molecules and then detach the topmost ZnO bilayer, saturating simultaneously its bottom and the newly created ZnO−M surface with PPA molecules as would occur in immersion or in a PPA vapor, the energy cost reduces to +0.36 eV per surface unit cell (cf. Figure 7). To determine this energy difference, a ZnO bilayer with a (5 × 2) unit cell and 12 adsorbed PPA molecules (6 on each side) was relaxed, and the energies of the PPA molecules and a

Figure 6. Surface phase diagram for the adsorption of PPA on ZnO− M. The chemical potential of ΔμPPA = −1.7 eV is indicated by the vertical solid line.

Figure 7. Schematic energy diagram for the transformation of a PPASAM on ZnO−M into a ZnPP crystal.

bilayer in a vacuum were subtracted, thereby assuming a PPA chemical potential of ΔμPPA = −1.7 eV. While the energy for detaching a ZnO bilayer in a PPA environment is still slightly positive, it is much smaller than in a vacuum and also significantly smaller than in liquid water. Repeating the same procedure for adsorbed monolayers of water instead of 0.6 ML of PPA yields an energy cost for detachment of +0.80 eV per ZnO−M surface unit cell. More important, the detached PPA/ ZnO bilayers can now expand to the ZnPP lattice constant and F


Research Article

ACS Applied Materials & Interfaces can increase their PPA coverage to 1 ML, thereby lowering the energy by −0.86 eV per initial ZnO−M surface unit cell (cf. Figure 7). Finally, vertical stacking of the ZnPP layers to form the ZnPP crystal gives an additional energy gain of −0.29 eV per unit cell (cf. Figure 7). Overall, the schematic energy diagram of Figure 7 illustrates that the formation of the ZnPP crystal is a thermodynamically favorable process and that the ZnPP crystal is the thermodynamic ground state and not the PPA/ZnO−M self-assembled monolayer. The PPA-saturated bilayer can be regarded as a fictitious, idealized transition structure that only serves to demonstrate that no high activation barriers have to be overcome when PPA/ZnO is detached from the surface and transformed into ZnPP crystals. On the basis of our calculations we can identify two driving forces that are responsible for the transformation of the PPASAMs to the ZnPP crystals: First, the frustration in the adsorbed PPA layer. The most favorable binding mode of the PPA molecules on ZnO is a bidentate or tridentate configuration. However, to achieve the preferred packing density of the phenyl rings, PPA molecules have to adsorb in a monodentate geometry, since not enough Zn sites are available. The PPA molecules can only recover their preferred bidentate and tridentate binding mode if the surface bilayer restructures by detaching from the substrate. The second driving force is the very strong interaction of PPA with ZnO. The binding energy for PPA is much larger than for water.54 The PPA molecules stabilize the ZnO−M surfaces as well or better than a monolayer of water even at coverages well below 0.5 ML. This strong interaction makes the detachment of the ZnO layers thermodynamically favorable. These arguments are not specific for the nonpolar, mixedterminated ZnO−M surface. Also the polar O- and Znterminated surfaces ZnO−O and ZnO−Zn are composed of ZnO bilayer building blocks (cf. Figure 5c). The only difference is that the bilayers from the polar surfaces consist of a Zn and an O layer and the unit cell is hexagonal with lattice constant a, whereas the bilayer from the nonpolar ZnO−M surface contains an equal number of Zn and O atoms in both layers and the unit cell is rectangular with lattice constants a and c = 1.614 a (cf. Figure 5c). A hexagonal lattice, however, can also be described by a rectangular unit cell with lattice constants a and √3a = 1.732a. After relaxation of the atoms and the lattice constants, the bilayers from the polar surfaces adopt exactly the same planar honeycomb structures as the relaxed bilayers from the nonpolar ZnO−M surface (cf. Figure 5c). Thus, all arguments given above for the stabilization of the bilayers by PPA adsorption and their transformation to ZnPP can be applied to the bilayers stemming from the polar surfaces as well. 3.3. Influence of Surface Roughness on Precipitation. Surface defects like step edges are likely to be active nucleation sites for ZnO delamination and the formation of ZnPP precipitates. Therefore, we have also investigated the influence of surface morphology on the etching efficiency by comparing the density of surface precipitates formed on well-defined and initially roughened ZnO samples. The single crystal preparation yields (cf. section 2.1) smooth surfaces consisting of extended terraces separated by monatomic steps (cf. Figure 1c), while significant surface roughness and an increased amount of defects are induced by extensive Ar+ ion beam bombardment (800 eV, 2 h). Figure 8 compares the density of precipitates that are formed at the differently prepared ZnO samples after immersion in PPA solution for 2 days under equivalent conditions: the amount of formed ZnPP precipitates on the

Figure 8. Optical micrographs of surface precipitates formed upon immersion of (a) smooth and (b) initially roughened ZnO−O surfaces for 2 days together with corresponding schemes representing the primary surface morphology of the samples.

damaged surface clearly exceeds the number of precipitates observed on the well-defined ZnO surface and thus corroborates the idea that precipitations nucleate at defects. 3.4. Influence of pH Value on Etching. Another important factor that affects the ZnO stability and its surface etching upon immersion is the pH value of the immersion solution, which has been mentioned in previous studies. Being a strong acid, PPA has the ability to partially dissolve metal oxides.56,57 On the basis of previous studies, the following pH regimes have been reported for ZnO: pH < 3.8, severe etching; 3.8 < pH < 5.5, slow surface dissolution; 5.5 < pH < 10, no evidence of etching.57,58 In our experiments a pH value of 4.5 was measured for the 0.1 mM ethanolic PPA solution. Therefore, additional SAMs were prepared from a solution for which the pH value had been adjusted to 5.75 by using an acetic acid/sodium acetate aqueous buffer system. However, immersing ZnO in such a buffer solution does not suppress surface etching. Quite the opposite, a dense layer of large crystalline lamellar precipitates has been formed already after 6 h of immersion, as depicted in Figure 9, while lamellas of similar size have been found only for sample that were kept for 1 month in ethanolic PPA solution. The precipitates also consist of ZnPP, as confirmed by XRD measurements. This severe surface etching is attributed to the presence of water in the solution, which expedites the precipitations, as reported in

Figure 9. SEM micrographs of a ZnO−O sample immersed for 6 h in an aqueous acetic acid/sodium acetate buffer solution (pH 5.75) with (a) small and (b) large magnification revealing the formation of large crystalline ZnPP precipitates. G


Research Article


extended fluorinated phenyl spacer, as depicted schematically in Figure 10b. Interestingly, when using phosphonic acid molecules with aliphatic backbone for SAM preparation, such as DPA, no visible precipitates were found, even after immersion for 1 week. At first glance this suggests an absence of etching. However, on a microscopic level, an increased surface roughness as compared to the bare surface was observed by means of AFM, again showing an etching of the ZnO surface. The absence of sizable precipitates indicates that alkane chains reduce the stability (i.e., lattice energy) of crystalline zinc dodecylphosponate precipitates but does not exclude a surface etching. In this context, we would like to emphasize that aliphatic PA-SAMs are rather undesired for modifications of electrodes in device applications because of their poor transport properties. In the case of gold electrodes it has been demonstrated that SAMs of (partly) fluorinated alkanethiols allow one to tailor their work function over a wide range.59 However, the large band gap of such aliphatic SAMs hampers an efficient charge transport across such films, which form an insulating layer on the electrode and yield rather poor device properties.60 Replacement of phosphonic acid with phosphinic acid is expected to lead to a different etching behavior, as both are chemically different regarding stability, reducing power, and acidity. In contrast to the thermally more stable triprotic phosphoric acid, phosphonic and phosphinic acid are both diprotic compounds. While the acidity for the first deprotonation step is very similar for both acids (with a pKs value of 2.0), phosphinic acid is particular in that is shows roughly the same acidity for the second deprotonation, with pKs = 2.2, thus favoring the cleavage of a P−H bond, whereas for phosphonic acid the pKs value for the second deprotonation increases to 6.6. Organic derivatives of phosphinic acid are formed by replacing one, two, or all three H atoms with organic groups (alkyl, aryl), with the H atoms bonded directly to the P atom being replaced first. The properties of the organophosphinic acids develop similarly as those of the inorganic free acid. In fact, a reduced etching was observed upon immersing a ZnO−O crystal in (absolute) ethanolic phenylphospinic acid (PHPA) solution, and even after 1 week only few, but extended, crystalline needles are formed while large parts of the ZnO substrate remain free of precipitates and exhibit only slightly increased surface roughness (cf. Figure S7 in the Supporting Information). On the basis of a single crystal XRD analysis, these precipitates are identified as zinc phenylphosphinate (ZnPHPP, cf. Table 2), of which crystal structure had been characterized already in previous work.61

previous studies.25,26 Indeed, the surface area of a ZnO sample intentionally marked with a droplet of distillated water before immersion in PPA solution exhibits a distinctly higher density of surface precipitates than the untreated surface (cf. Figure S3 in the Supporting Information). On the basis of these findings, also different solvents (isopropyl alcohol, acetonitrile) have been tested, but in all cases they showed an etching behavior similar to that of ethanolic solutions. 3.5. ZnO Surface Etching by Other SAMs. To complement our study, also the influence of other backbone units on the resulting structure of precipitates as well as the impact of a slightly changed anchoring unit on the etching of the ZnO surface has been addressed. Of particular interest for the tailoring of electronic interface properties (for example, the work function) are (partially) fluorinated SAMs (such as TFPPA), because they exhibit an intrinsic dipole moment. Also for this case, we found the formation of distinct, star-shaped precipitates after immersion of a ZnO−O crystal in ethanolic TFPPA solution (for 1 week), as depicted in Figure 10a.

Figure 10. (a) Optical micrographs and (c) θ/2θ scan of surface precipitate formed after immersion of a ZnO−O sample in TFPPA solution for 1 week. A comparison with the powder diffraction diagram calculated on the basis of the single crystal structure data of ZnPP51 allows a tentative assignment of the diffraction peaks, which suggests an enhanced interlayer distance d(010) due to the substitution of −CF3 groups to the backbone, as shown in part (b).

Table 2. Unit Cell Parameter Derived from a Single Crystal XRD Analysis of Isolated Precipitates and Literature Data of Zinc phenylphosphinate (ZnPHP) Single Crystal61

Although a large area of the oxide substrate is covered by precipitates, only small needles are formed (even after prolonged immersion time), which hampered a single crystal analysis. However, θ/2θ measurements have been possible and confirmed the crystalline nature of such precipitates. By comparison of the acquired X-ray powder diffraction diagram with a powder diffraction diagram of ZnPP calculated on the basis of single crystal data,51 a tentative assignment of the observed reflexes was made (cf. Figure 10c). This enabled us to identify the leading diffraction peaks as (0n0) reflexes with a corresponding interlayer distance of d(010) = 18.6 Å. Compared to the nonfluorinated ZnPP precipitates, this means an increased layer separation that can be rationalized by the H

unit cell parameters

precipitates

ZnPHP single crystal

a (Å) b (Å) c (Å) α/β/γ (deg) crystal system space group Z cell volume (Å3)

15.56 10.96 8.21 90/108.96/90 monoclinic C2/c 4 1332

15.61 10.95 8.23 90/108.62/90 monoclinic C2/c 4 1332


Research Article


4. DISCUSSION The starting point in preventing etching of the ZnO surfaces is an understanding of the underlying mechanisms and parameters that affect the instability of this surface. The monolayer formation by immersion is always accompanied by the attachment of hydroxyl groups that are formed on the surface of an oxide during contact to air. This occurs for two main reasons: the presence of remaining water in the applied solution or a hydroxylation due to air humidity during sample treatment at ambient conditions (sample transfer in air after UHV preparation). By means of TDS, we have proven in a recent study that even upon short exposure to ambient air for a few minutes, almost complete hydroxylation of the ZnO surface takes place.62 Etching of ZnO crystals has been reported in a number of previous studies.13,25,26,38,63,64 The etching process is commonly described as a dissolution reaction and formation of Zn2+ cations,57,58 which subsequently may react with organic molecules from the solution, thereby forming complexes and precipitates on the surface by recrystallization.65,66 The general concept of ZnO dissolution involves the following steps: first, a protonation of surface oxo species; second, a rupture of metal− oxygen bonds; and, finally, the transfer of metal ions into the solution.65 ZnO as an amphoteric oxide can react with both strong acids and bases to form salts and water. Depending on the pH value of the solution, the surface hydroxyl groups can act as proton donors or acceptors. The resulting phase stability of metal oxides as a function of the pH value of the solution is generally described by Pourbaix diagrams.57 In all these processes an aqueous solution plays a vital role. In fact, it was pointed out by Lange et al.25 that the concentration of residual water in the immersion solution is crucial for the efficient precipitation, which is also corroborated by our present findings for immersion in aqueous buffer solutions. For the case of PPA interaction with ZnO, however, our DFT calculations have shown that the bonding of the PPA molecules to the Zn cations is exceptionally strong. PPA deprotonates and converts surface O atoms to OH groups and water. Because of the strong interaction, the ZnO units with the protonated O species are likely to remain bound to the PPA molecules. Furthermore, the layers of the ZnPP crystal and the bilayers of the ZnO structure share the same structural element of a centered-rectangular arrangement of the Zn ions. Thus, the precipitate formation can be well explained by a direct transformation of the PPA-SAMs to ZnPP layers by delamination of ZnO units without any substantial ion dissolution and ZnPP recrystallization. Instead, this transformation is thermodynamically driven by the large gain of lattice energy of the formed ZnPP crystal. The occurrence of such a mechanism in the case of ZnPP precipitation is also supported by the following observation: In general, the polar O-terminated ZnO surface is found to be less stable against etching in oxidizing solvents than the Zn- or mixed-terminated surfaces.63 This difference is usually exploited to distinguish both polar terminations. Surprisingly, however, such a dependency has not been observed in our experiments. Instead, rather similar precipitation is found for both polar and the nonpolar ZnO surface terminations on a time scale of several days when using water-free immersion solutions, hence evidencing another transformation mechanism than by conventional surface etching via Zn dissolution. The direct transformation of PPA-SAMs to ZnPP layers by ZnO delamination is a solvent-assisted process. The detach-

ment of ZnO units is favorable only because of the strong stabilization by solvation and PPA adsorption. Thus, the pH value and the presence of additional water are expected to have a large impact, as also observed in our experiments using aqueous buffer solutions. Nevertheless, the direct transformation does not strictly require the presence of additional water. Indeed, precipitate formation was also observed in nominally water-free environments, although now on a time scale of days instead of hours. In this context, we note again that due to the protonation of the surface-bound hydroxyl groups, water is formed even if completely dry solvents are used. A solvent-assisted detachment of ZnO units from the surfaces is expected to start most easily at step edges and other surface defects. In fact, this assumption is well-corroborated by our observation of an increased density of precipitates formed on intentionally roughened ZnO surfaces. In a recent work, the addition of Zn2+ ion solutions to the phosphonic acid solution was reported to prevent any precipitation upon immersion of nanocrystalline ZnO powder.67 As another approach, the usage of Al-doped ZnO was proposed to avoid the etching effect,68 although a surface degradation was still found in the case of doped samples, which raises questions about the resulting film quality. In a previous study, we demonstrated an alternative preparation of highly ordered PPA monolayer films based on organic molecular beam deposition (OMBD).62 By first depositing PPA films with a thickness of a few nanometers and subsequent thermal desorption of the excess multilayers, firmly bound and welloriented PPA monolayer films were obtained. Accompanying AFM data (cf. Figure S4 in the Supporting Information) reveal no evidence for precipitation on the ZnO surfaces, hence indicating that this approach is a vital alternative for the preparation of phosphonic acid-anchored monolayer films, which enables a rigorous exclusion of water and humidity.

5. CONCLUSIONS Here we demonstrate, that functionalization of ZnO samples via self-assembled monolayer films anchored by phosphonic acids exhibits a chemical side reaction upon prolonged immersion times. This leads to a marked surface degradation and formation of surface precipitates which have been identified as zinc phosphonate. Such a surface etching was observed for both polar and the mixed-terminated ZnO surfaces and for phosphonic acids with different aromatic backbones. Initial substrate surface roughness and the presence of remaining water in the solution are critical parameters and accelerate the precipitation. Remarkably, such phosphonate precipitates are also formed upon immersion in carefully dried (i.e., water-free) PPA solutions on a time scale of several days. This demonstrates that the formation of PA-based SAMs on ZnO is not a self-terminating process like found for organothiolbased SAMs on gold substrates. Therefore, prolonged immersion times should be avoided upon SAM formation for practical applications. In fact, a DFT-based energy analysis reveals that PPA-SAMs on ZnO represent a rather metastable configuration, while zinc phosphonate is the energetically more stable configuration. The theoretical analysis suggests further a reaction path leading to the structural transformation without addition of water. Moreover, it provides a detailed microstructural description of the molecule/ZnO interface, which is an important prerequisite for further theoretical studies of the electronic coupling and charge transport through this interface. The presently identified electrochemically driven precipitation I


Research Article


(9) Sadik, P. W.; Pearton, S. J.; Norton, D. P.; Lambers, E.; Ren, F. Functionalizing Zn- and O-terminated ZnO with thiols. J. Appl. Phys. 2007, 101, 104514. (10) Perkins, C. L. Molecular Anchors for Self-Assembled Monolayers on ZnO: A Direct Comparison of the Thiol and Phosphonic Acid Moieties. J. Phys. Chem. C 2009, 113, 18276−18286. (11) Jayalakshmi, G.; Saravanan, K.; Balasubramanian, T. Impact of thiol and amine functionalization on photoluminescence properties of ZnO films. J. Lumin. 2013, 140, 21−25. (12) Halevi, B.; Vohs, J. M. TPD study of the reaction of CH3CH2SH and (CH3CH2)2S2 on ZnO(0001) and ZnO. Catal. Lett. 2006, 111, 1− 2. (13) Chen, J.; Ruther, R. E.; Tan, Y.; Bishop, L. M.; Hamers, R. J. Molecular Adsorption on ZnO(101̅0) Single-Crystal Surfaces: Morphology and Charge Transfer. Langmuir 2012, 28, 10437−10445. (14) Kornherr, A.; French, S. A.; Sokol, A. A.; Catlow, C. R. A.; Hansal, S.; Hansal, W. E. G.; Besenhard, J. O.; Kronberger, H.; Nauer, G. E.; Zifferer, G. Interaction of adsorbed organosilanes with a polar zinc oxide surface: A molecular dynamics study comparing two models for the metal oxide surface. Chem. Phys. Lett. 2004, 393, 107−111. (15) Turgeman, R.; Gershevitz, O.; Palchik, O.; Deutsch, M.; Ocko, B. M.; Gedanken, A.; Sukenik, C. N. Oriented Growth of ZnO Crystals on Self-Assembled Monolayers of Functionalized Alkyl Silanes. Cryst. Growth Des. 2004, 4, 169−175. (16) Allen, C. G.; Baker, D. G.; Albin, J. M.; Oertli, H. E.; Gillaspie, D. T.; Olson, D. C.; Furtak, T. E.; Collins, R. T. Surface Modification of ZnO Using Triethoxysilane-Based Molecules. Langmuir 2008, 24, 13393−13398. (17) Ozawa, K.; Hasegawa, T.; Edamoto, K.; Takahashi, K.; Kamada, M. Adsorption State and Molecular Orientation of Ammonia on ZnO(101̅0) Studied by Photoelectron Spectroscopy and near-Edge Xray Absorption Fine Structure Spectroscopy. J. Phys. Chem. B 2002, 106, 9380−9386. (18) Dominguez, A.; Lorke, M.; Schoenhalz, A. L.; Rosa, A. L.; Frauenheim, T. H.; Rocha, A. R.; Dalpian, G. M. First principles investigations on the electronic structure of anchor groups on ZnO nanowires and surfaces. J. Appl. Phys. 2014, 115, 203720. (19) Persson, P.; Ojamäe, L. Periodic Hartree−Fock study of the adsorption of formic acid on ZnO(101̅0). Chem. Phys. Lett. 2000, 321, 302−308. (20) Zhang, B.; Kong, T.; Xu, W.; Su, R.; Gao, Y.; Cheng, G. Surface Functionalization of Zinc Oxide by Carboxyalkylphosphonic Acid SelfAssembled Monolayers. Langmuir 2010, 26, 4514−4522. (21) Cornil, D.; Van Regemorter, T.; Beljonne, D.; Cornil, J. Work function shifts of a zinc oxide surface upon deposition of selfassembled monolayers: a theoretical insight. Phys. Chem. Chem. Phys. 2014, 16, 20887−20899. (22) Moreira, N. H.; da Rosa, A. L.; Frauenheim, T. Covalent functionalization of ZnO surfaces: A density functional tight binding study. Appl. Phys. Lett. 2009, 94, 193109. (23) Tian, X.; Xu, J.; Xie, W. Controllable Modulation of the Electronic Structure of ZnO(101̅0) Surface by Carboxylic Acids. J. Phys. Chem. C 2010, 114, 3973−3980. (24) Liang, Y.; Thorne, J. E.; Kern, M. E.; Parkinson, B. A. Sensitization of ZnO Single Crystal Electrodes with CdSe Quantum Dots. Langmuir 2014, 30, 12551−12558. (25) Lange, I.; Reiter, S.; Pätzel, M.; Zykov, A.; Nefedov, A.; Hildebrandt, J.; Hecht, S.; Kowarik, S.; Wöll, C.; Heimel, G.; Neher, D. Tuning the Work Function of Polar Zinc Oxide Surfaces using Modifed Phosphonic Acid Self-Assembled Monolayers. Adv. Funct. Mater. 2014, 24, 7014−7024. (26) Timpel, M.; Nardi, M. V.; Krause, S.; Ligorio, G.; Christodoulou, C.; Pasquali, L.; Giglia, A.; Frisch, J.; Wegner, B.; Moras, P.; Koch, N. Surface Modification of ZnO(0001)−Zn with Phosphonate-Based Self-Assembled Monolayers: Binding Modes, Orientation, and Work Function. Chem. Mater. 2014, 26, 5042−5050. (27) Gliboff, M.; Sang, L.; Knesting, K. M.; Schalnat, M. C.; Mudalige, A.; Ratcliff, E. L.; Li, H.; Sigdel, A. K.; Giordano, A. J.; Berry, J. J.; Nordlund, D.; Seidler, G. T.; Brédas, J. L.; Marder, S. R.;

might also be helpful to understand frequently reported problems in the long-time operation of devices utilizing such interfaces. A less pronounced surface etching was found upon SAM formation by immersion when replacing phosphonic acid with phosphinic acid anchoring units. This might be due to differences in acidity for the second deprotonation step of both acids or in the lattice energies of phosphonates and phosphinates. By contrast, a completely dry, OMBD-based preparation of PA-anchored monolayer films was previously established, which allows a rigorous exclusion of water or humidity.62 First results reveal no evidence of any precipitate formation, which favors their use for device applications and will hopefully trigger corresponding studies in the near future.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b02190. Additional water contact angle measurements, SEM data on precipitates after heating, characterization of OMBDprepared monolayers, and visualization of the crystal structures of ZnPP and ZnPHP (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*B.M. e-mail: [email protected]. *G.W. e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge support from the German Research Foundation (DFG) in the framework of the Research Training Group RTG 1782 “Functionalization of Semiconductors” and the Cluster of Excellence EXC 315 “Engineering of Advanced Materials”.

■

REFERENCES

(1) Meyer, J.; Hamwi, S.; Kröger, M.; Kowalsky, W.; Riedl, T.; Kahn, A. Transition Metal Oxides for Organic Electronics: Energetics, Device Physics and Applications. Adv. Mater. 2012, 24, 5408−5427. (2) Fortunato, E. M. C.; Barquinha, P. M. C.; Pimentel, A.C.M.B.G.; Goncalves, A. M. F.; Marques, A. J. S.; Martins, R. F. P.; Pereira, L. M. N. Wide-bandgap high-mobility ZnO thin-film transistors produced at room temperature. Appl. Phys. Lett. 2004, 85, 2541−2543. (3) Park, J. S.; Maeng, W. J.; Kim, H. S.; Park, J. S. Review of recent developments in amorphous oxide semiconductor thin-film transistor devices. Thin Solid Films 2012, 520, 1679−1693. (4) Pal, B. N.; Sun, J.; Jung, B. J.; Choi, E.; Andreou, A. G.; Katz, H. E. Pentacene-Zinc Oxide Vertical Diode with Compatible Grains and 15-MHz Rectification. Adv. Mater. 2008, 20, 1023−1028. (5) Koch, N. Organic electronic devices and their functional interfaces. ChemPhysChem 2007, 8, 1438−1455. (6) Ma, H.; Yip, H. L.; Huang, F.; Jen, A. K. Y. Interface Engineering for Organic Electronics. Adv. Funct. Mater. 2010, 20, 1371−1388. (7) Heimel, G.; Romaner, L.; Zojer, E.; Bredas, J. L. The interface energetics of self-assembled monolayers on metals. Acc. Chem. Res. 2008, 41, 721−729. (8) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-assembled monolayers of thiolates on metals as a form of nanotechnology. Chem. Rev. 2005, 105, 1103−1170. J


Research Article

ACS Applied Materials & Interfaces Pemberton, J. E.; Ginger, D. S. Orientation of Phenylphosphonic Acid Self-Assembled Monolayers on a Transparent Conductive Oxide: A Combined NEXAFS, PM-IRRAS, and DFT Study. Langmuir 2013, 29, 2166−2174. (28) Guerrero, G.; Alauzun, J. G.; Granier, M.; Laurencin, D.; Mutin, H. Phosphonate coupling molecules for the control of surface/ interface properties and the synthesis of nanomaterials. Dalton Trans. 2013, 42, 12569−12585. (29) Li, H.; Ratcliff, E. L.; Sigdel, A. K.; Giordano, A. J.; Marder, S. R.; Berry, J. J.; Brédas, J. L. Modification of the Galium-Doped Zinc Oxide Surface with Self-Assembled Monolayers of Phosphonic Acid: A Joint Theoretical and Experimental Study. Adv. Funct. Mater. 2014, 24, 3593−3603. (30) Wood, C.; Li, H.; Winget, P.; Brédas, J. L. Binding Modes of Fluorinated Benzylphosphonic Acids on the Polar ZnO Surface and Impact on Work Function. J. Phys. Chem. C 2012, 116, 19125−19133. (31) Kedem, N.; Blumstengel, S.; Henneberger, F.; Cohen, H.; Hodes, G.; Cahen, D. Morphology-, synthesis- and dopingindependent tuning of ZnO work function using phenylphosphonates. Phys. Chem. Chem. Phys. 2014, 16, 8310−8319. (32) Koldemir, U.; Braid, J. L.; Morgenstern, A.; Eberhart, M.; Collins, R. T.; Olson, D. C.; Sellinger, A. Molecular Design for Tuning Work Functions of Transparent Conducting Electrodes. J. Phys. Chem. Lett. 2015, 6, 2269−2276. (33) Braid, J. L.; Koldemir, U.; Sellinger, A.; Collins, R. T.; Furtak, T. E.; Olson, D. C. Conjugated Phosphonic Acid Modified Zinc Oxide Electron Transport Layers for Improved Performance in Organic Solar Cells. ACS Appl. Mater. Interfaces 2014, 6, 19229−19234. (34) Hotchkiss, P. J.; Malicki, M.; Giordano, A. J.; Armstrong, N. R.; Marder, S. R. Characterization of phosphonic acid binding to zinc oxide. J. Mater. Chem. 2011, 21, 3107−3112. (35) Pawsey, S.; Yach, K.; Reven, L. Self-Assembly of Carboxyalkylphosphonic Acids on Metal Oxide Powders. Langmuir 2002, 18, 5205−5212. (36) Timpel, M.; Nardi, M. V.; Ligorio, G.; Wegner, B.; Pätzel, M.; Kobin, B.; Hecht, S.; Koch, N. Energy-Level Engineering at ZnO/ Oligophenylene Interfaces with Phosphonate-Based Self-Assembled Monolayers. ACS Appl. Mater. Interfaces 2015, 7, 11900−11907. (37) Cyganik, P.; Buck, M.; Strunskus, T.; Shaporenko, A.; WiltonEly, J. D. E. T.; Zharnikov, M. M.; Wöll, C. Competition as a design concept: Polymorphism in self-assembled monolayers of biphenylbased thiols. J. Am. Chem. Soc. 2006, 128, 13868−13878. (38) Nam, S.-H.; Kim, M.-H.; Yoo, D. G.; Jeong, S. H.; Kim, D. Y.; Lee, N.-E.; Boo, J.-H. Metal-doped ZnO thin films: synthesis, etching characteristic, and application test for organic light emitting diode (OLED) devices. Surf. Rev. Lett. 2010, 17, 121−127. (39) Palacios-Lidon, E.; Pickup, D. E.; Johnson, P. S.; Ruther, R. E.; Tena-Zaera, R.; Hamers, R. J.; Colchero, J.; Himpsel, F. J.; Ortega, J. E.; Rogero, C. Face-Selective Etching of ZnO during Attachment of Dyes. J. Phys. Chem. C 2013, 117, 18414−18422. (40) Sakthivel, S.; Neppolian, B.; Shankar, M. V.; Arabindoo, B.; Palanichamy, M.; Murugesan, V. Solar photocatalytic degradation of azo dye: comparison of photocatalytic efficiency of ZnO and TiO2. Sol. Energy Mater. Sol. Cells 2003, 77, 65−82. (41) Gouvêa, C. A. K.; Wypych, F.; Moraes, S. G.; Durán, N.; Nagata, N.; Peralta-Zamora, P. Semiconductor-assisted photocatalytic degradation of reactive dyes in aqueous solution. Chemosphere 2000, 40, 433−440. (42) Furbeth, W.; Stratmann, M. The delamination of polymeric coatings from electrogalvanised steel − a mechanistic approach: Part 1: delamination from a defect with intact zinc layer. Corros. Sci. 2001, 43, 207−227. (43) Götzen, J.; Witte, G. Rapid preparation of highly ordered ultraflat ZnO surfaces. Appl. Surf. Sci. 2012, 258, 10144−10147. (44) Bennett, J. A.; Hope, E. G.; Singh, K.; Stuart, A. M. Synthesis and coordination chemistry of fluorinated phosphonic acids. J. Fluorine Chem. 2009, 130, 615−620. (45) Giannozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.; Chiarotti, G. L.; Cococcioni, M.; Dabo, I.;

Dal Corso, A.; de Gironcoli, S.; Fabris, S.; Fratesi, G.; Gebauer, R.; Gerstmann, U.; Gougoussis, C.; Kokalj, A.; Lazzeri, M.; Martin-Samos, L.; Marzari, N.; Mauri, F.; Mazzarello, R.; Paolini, S.; Pasquarello, A.; Paulatto, L.; Sbraccia, C.; Scandolo, S.; Sclauzero, G.; Seitsonen, A. P.; Smogunov, A.; Umari, P.; Wentzcovitch, R. M. QUANTUM ESPRESSO: A Modular and Open-Source Software Project for Quantum Simulations of Materials. J. Phys.: Condens. Matter 2009, 21, 395502. (46) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (47) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132, 154104. (48) Meyer, B. First-principles study of the polar O-terminated ZnO surface in thermodynamic equilibrium with oxygen and hydrogen. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 69, 045416. (49) Bauer, T.; Schmaltz, T.; Lenz, T.; Halik, M.; Meyer, B.; Clark, T. Phosphonate- and Carboxylate-Based Self-Assembled Monolayers for Organic Devices: A Theoretical Study of Surface Binding on Aluminum Oxide with Experimental Support. ACS Appl. Mater. Interfaces 2013, 5, 6073−6080. (50) Killian, M. S.; Seiler, S.; Wagener, V.; Hahn, R.; Ebensperger, C.; Meyer, B.; Schmuki, P. Interface Chemistry and Molecular Bonding of Functional Ethoxysilane-Based Self-Assembled Monolayers on Magnesium Surfaces. ACS Appl. Mater. Interfaces 2015, 7, 9006−9014. (51) Martin, K. J.; Squattrito, P. J.; Clearfield, A. The Crystal and Molecular Structure of Zinc Phenylphosphonate. Inorg. Chim. Acta 1989, 155, 7−9. (52) Gerbier, P.; Guerin, C.; Henner, B.; Unal, J. R. An organometallic route to zinc phosphonates and theirintercalates. J. Mater. Chem. 1999, 9, 2559−2565. (53) Scott, K. J.; Zhang, Y.; Wang, R.-C.; Clearfield, A. Synthesis, Characterisation, and Amine Intercalation Behavior of Zinc Phosphite Phenylphosphonate Mixed Derivatives. Chem. Mater. 1995, 7, 1095− 1102. (54) Meyer, B.; Marx, D.; Dulub, O.; Diebold, U.; Kunat, M.; Langenberg, D.; Wöll, C. Partial Dissociation of Water Leads to Stable Superstructures on the Surface of Zinc Oxide. Angew. Chem., Int. Ed. 2004, 43, 6641−6645. (55) We found one configuration with a slightly lower energy (see ref 62). If only half of the PPA molecules are fully deprotonated and the other half remain singly protonated, the binding energy per PPA molecule increases to 3.58 eV. (56) Taratula, O.; Galoppini, E.; Wang, D.; Chu, D.; Zhang, Z.; Chen, H.; Saraf, G.; Lu, Y. Binding Studies of Molecular Linkers to ZnO and MgZnO Nanotip Films. J. Phys. Chem. B 2006, 110, 6506− 6515. (57) Wagh, A. S.; Jeong, S. Y. Chemically Bonded Phosphate Ceramics: I, A Dissolution Model of Formation. J. Am. Ceram. Soc. 2003, 86, 1838−1844. (58) Han, J.; Qiu, W.; Gao, W. Potential dissolution and photodissolution of ZnO thin Films. J. Hazard. Mater. 2010, 178, 115−122. (59) Alloway, D. A.; Hofmann, M.; Smith, D. L.; Gruhn, N. E.; Graham, A. L.; Colorado, R., Jr.; Wysocki, V. H.; Lee, T. R.; Lee, P. A.; Armstrong, N. R. Interface Dipoles Arising from Self-Assembled Monolayers on Gold: UV-Photoemission Studies of Alkanethiols and Partially Fluorinated Alkanethiols. J. Phys. Chem. B 2003, 107, 11690− 11699. (60) Bock, C.; Pham, D. V.; Kunze, U.; Käfer, D.; Witte, G.; Wöll, Ch. Improved morphology and charge carrier injection in pentacene field-effect transistors with thiol-treated electrodes. J. Appl. Phys. 2006, 100, 114517. (61) Shieh, M.; Martin, K. J.; Squattrito, P. J.; Clearfield, A. New Low-Dimensional Zinc Compounds Containing Zinc-Oxygen-Phosphorus Frameworks: Two-Layered Inorganic Phosphites and a Polymeric Organic Phosphinate. Inorg. Chem. 1990, 29, 958−963. (62) Ostapenko, A.; Klöffel, T.; Meyer, B.; Witte, G. Formation and Stability of Phenylphosphonic Acid Monolayers on ZnO: Comparison K


Research Article

ACS Applied Materials & Interfaces of in-Situ and ex-Situ SAM Preparation. Langmuir 2016, DOI: 10.1021/acs.langmuir.6b00487 (in press). (63) Mariano, A. N.; Hanneman, R. E. Crystallographic Polarity of ZnO Crystals. J. Appl. Phys. 1963, 34, 384. (64) Valtiner, M.; Borodin, S.; Grundmeier, G. Stabilization and Acidic Dissolution Mechanism of Single-Crystalline ZnO(0001) Surfaces in Electrolytes Studied by In-Situ AFM Imaging and ExSitu LEED. Langmuir 2008, 24, 5350−5358. (65) Mutin, P. H.; Guerrero, G.; Vioux, A. Hybrid materials from organophosphorus coupling molecules. J. Mater. Chem. 2005, 15, 3761−3768. (66) Horiuchi, H.; Katoh, R.; Hara, K.; Yanagida, M.; Murata, S.; Arakawa, H.; Tachiya, M. Electron Injection Efficiency from Excited N3 into Nanocrystalline ZnO Films: Effect of (N3-Zn2+) Aggregate Formation. J. Phys. Chem. B 2003, 107, 2570−2574. (67) Pomorska, A.; Grundmeier, G.; Ozcan, O. Effect of Zn2+ Concentration on the Absorption of Organophosphonic Acids on Nanocrystzalline ZnO Surfaces. Colloids Interface Sci. Commun. 2014, 2, 11−14. (68) Gerischer, H. Neglected problems in the pH dependence of the flatband potential of semiconducting oxides and semiconductors covered with oxide layers. Electrochim. Acta 1989, 34, 1005−1009.

L


Etching of Crystalline ZnO Surfaces upon Phosphonic Acid Adsorption

Recommend Documents