Article Cite This: Acc. Chem. Res. XXXX, XXX, XXX−XXX
pubs.acs.org/accounts
Nanoscale Studies of Organic Radicals: Surface, Interface, and Spinterface M. Benedetta Casu* Institute of Physical and Theoretical Chemistry, University of Tübingen, Auf der Morgenstelle 18, 72076 Tübingen, Germany S Supporting Information *
CONSPECTUS: In the last decade technology has brought significant changes to our lives, including new habits and a new view on social relationships. These technological innovations are based on several factors, one of which is miniaturization. This was made possible also due to the discovery and synthesis of new materials with characteristics at the nanoscale that are designed for specific purposes. This “on purpose” approach, joined to the development of preparation and growth methods, has led to use of thin films rather than bulk materials in devices. Using thin films makes devices easier to produce, and using films for coating protects the devices and gives specific properties to surfaces. For several decades thin films, surfaces, and interfaces have been intensively investigated. Indeed, device performances rely on the optimized match of thin films of different natures, such as organic and inorganic semiconductors and metals for contacts. Surprisingly, in comparison, little attention has been devoted to the deposition of organic radicals on a substrate. This might be because these materials are considered not stable enough for evaporation. In this work, we demonstrate that it is possible to evaporate and deposit organic radicals onto well-defined surfaces under controlled conditions, without degradation. Using soft X-ray spectroscopies, performed also at synchrotrons, we investigate thin film processes, surfaces, and interfaces at the nanoscale, when organic radicals are deposited on metal and metal oxide surfaces. We suggest how to design organic radicals bearing in mind the thermodynamic factors that govern thin film stability, with the purpose of obtaining not only a chemically stable radical, but also stable thin films. We investigate the thermal and air stability of the deposited films, and we explore the influence of the surface/radical chemical bond and the role of surface defects on the magnetic moment at the interface. We find that organic radicals are physisorbed and keep their magnetic moment on inert and passivated surfaces such as Au(111) and Al2O3(112̅0) single crystals, SiO2, and ideal TiO2(110) single crystals, while defective sites such as oxygen vacancies or the presence of OH groups lead to chemisorption of the organic radicals on the surface with quenching of their magnetic moment. Our work shows that the use of X-ray based techniques represents a powerful approach to reveal the mechanisms governing complex interfaces, such as radical/metal and radical/metal-oxide, where it is important to describe both charge and spin behavior (spinterfaces). It also makes it possible to conceive new experiments to investigate the magnetic character of the thin films versus their structural properties, toward tuning the arrangement of the molecules in films. Controlling the molecular arrangement will give the opportunity to tune the mutual position and orientation of the molecules, that is, of the single magnetic moments in the films, “imprinting” their magnetic properties. A deep understanding of stable radical/inorganic spinterfaces may open the way to use radicals in solid state devices or as quantum bits with dedicated configurations, as proposed for other molecular quantum bits, and in spin-based electronics.
■
carbon, and oxygen.10−22 Among the critical issues in advancing their implementation there are the processes involving their thin film formation. This Account reviews the recent results on radical thin films obtained by using organic molecular beam deposition (OMBD), the ultrahigh vacuum (UHV) technique typically used for growing organic semiconductor thin films. Similar to the process that governs molecular beam epitaxy (MBE) in inorganic materials,23 OMBD provides precise control of the growth parameters, the critical feature that made MBE the method of choice for fabrication of a wide range of new devices that have gained substantial advantages from the high control of film thickness and structure.
INTRODUCTION Miniaturization and technological progress are largely based on the comprehension of the surfaces and interface between different materials, for example, metallic contacts and organic/ inorganic semiconductors. Consequently, thin films and interfaces remained in the focus of extensive investigations for several decades.1−5 However, little attention has been devoted to the controlled nanoscale growth of thin films of organic radicals.6−8 This might be because these materials are considered unstable and may undergo degradation during evaporation or adsorption onto a surface.9 Organic molecules with a radical site are emerging candidates for groundbreaking applications from energy storage to quantum computing, due to the presence of a magnetic moment coupled to only light elements, such as nitrogen, © XXXX American Chemical Society
Received: December 8, 2017
A
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research OMBD allows growth reproducibility and control of film fabrication.23 The organic material is evaporated under UHV by using a Knudsen cell. The molecules are collimated by passing through an orifice and then deposited on a substrate, atomically cleaned and fully characterized. The flux is controlled by the cell temperature and a shutter. The evaporation rate is monitored by a quartz microbalance and, if available, a quadrupole mass spectrometer. We have shown that OMBD can provide excellent thin films of organic radicals, which opens the way to investigate their controlled thin films, interfaces, and spinterfaces. This approach supports the use of a range of spectroscopies and microscopies, much larger than the ones traditionally used to characterize organic radicals, and it allows accessing, describing, and controlling the radical/inorganic interface. A combination of surface-sensitive spectroscopies, such as X-ray photoemission (XPS)24 and microscopies with atomic force microscopy and first-principles calculations are the notable techniques.25−35 XPS, a very powerful application of the photoelectric effect,36 can be used in a purely analytical way, assessing the presence and the concentration of the various elements on a substrate, because each electron in an atomic shell has a specific binding energy that leads to a fingerprint line in a XPS spectrum. This allows determining whether the deposited films have the expected chemical composition and correlation with the stoichiometry of the evaporated molecule. The XPS main lines are also sensitive to the chemical environment of the emitting element, providing not only the possibility to check the integrity of the adsorbate and its interaction with the substrate but also to use best fit procedures based on chemical and physical arguments.25,37,38 The choice of a substrate for the first evaporation of a new molecule is a challenging process. A well-defined substrate, such as Au(111), TiO2(110), and Al2O3(112̅0) single crystals, and native SiO2/Si(111) wafers, greatly facilitates both the film preparation and characterization. In fact, these substrates are among the most investigated surfaces; thus, as we have observed, all “surprising” effects have been related to the adsorbate and to the newly formed interface. Another aspect to consider is the nature of the surfaces, such as an inert noble metal (Au(111)), a metal oxide (TiO2), a well-organized stepped surface (sapphire), and a flat and amorphous thin film oxide (native SiO2). The first investigated radicals, nitronyl nitroxides, were chosen among those that are known to be chemically stable. Nitroxides are also very promising candidates for sensors.39−41 Noteworthy, the first purely organic ferromagnet ever reported is the p-nitrophenyl nitronyl nitroxide radical.42
Figure 1. (a) 4,4,5,5-Tetramethyl-2-(pyrenyl)-imidazoline-1-oxy-3oxide (NitPyn) and (b) 1-[4′-(3-Oxide-1-oxyl-4,4,5,5-tetramethylimidazolin-2-yl)pyrazol-1′-yl]-pyrene (PPN).
1s, and O 1s core level lines) of the XPS curves nicely agreed with the theoretical percentages expected because of NitPyn stoichiometry, and the XPS results obtained from the films were comparable to those obtained from the powder, that is, without NitPyn undergoing evaporation, indicating that the deposition occurred without degradation.25 The C 1s core level spectra show two lines: One is due to the electrons emitted from the aromatic carbons and from the methyl group, while carbons bound to nitrogen contribute to the higher binding energy line. A further indication that the radical molecules were evaporated without degradation is the shape of the N 1s and O 1s core level spectra, which are very important to ascertain the radical character of the films, because the unpaired electron is delocalized over the two equivalent NO groups. According to the equivalent chemical environment of the two N or O atoms, only one feature is expected and observed in the N 1s and the O 1s core level spectra at 402.2 and at 531.3 eV, respectively (Figure 2). This single line excludes the degradation to iminonitroxide or to the di-imino derivatives45 upon deposition. The presence of the imino group would imply an XPS line at lower binding energies in the N 1s core level spectra, because breaking the NO bond hinders the electron withdrawing action of oxygen and lowers the binding energy of the photoelectrons emitted from nitrogen. In fact, we observed this feature by using X-ray photoelectron emission microscopy (XPEEM), that is, the analogue of XPS in microscopy.30 In XPEEM, the photoemission signal is laterally correlated with a specific point at the surface that is imaged through the photoelectrons.46 During these experiments, our goal was to “measure” the radiation sensitivity of the radical organic films30 and to investigate its possible signature on the spectra. Micro-XPS operation mode was used to monitor the time evolution, in situ and in real time, of the C 1s and N 1s core level emission. We observed that the peak intensity decreases with time and the peak progressively shifts toward lower binding energies. After 800 s of beam exposure, the N 1s core level line experienced a clear narrowing and a signal appeared in the lower binding energy range (Figure 2e), while the pyrene substituent stays intact (Figure 2d), indicating the switching under the beam of some NitPyn molecules to the imino nitroxide.30 The XPS characterization gives an undoubtedly positive indication that OMBD can be applied to organic radicals. However, the radical character is usually investigated by using electron spin resonance (ESR) spectroscopy. Thus, we investigated by using ESR on not only the NitPyn films but also the residual powder left in the Knudsen cell after several cycles of heating and evaporation. The resulting spectra indicated without any doubt that the species after heating and deposition was an intact nitronyl nitroxide radical.25,29 We established a qualitative25 and quantitative31 correspondence between XPS features and ESR patterns in radicals:
■
THE CONTROLLED EVAPORATION OF A PYRENE DERIVATIVE OF THE NITRONYL NITROXIDE RADICAL Our first experiment25 was focused on a pyrene derivative of the nitronyl nitroxide radical (NitPyn, Figure 1) on Au(111) single crystals. We deposited NitPyn thin films with various thicknesses from monolayer up to several layers (Figure 2). The XPS data did not show any thickness-dependent changes. This allowed us to safely conclude that the NitPyn is physisorbed on gold and the molecules feel a very weak interaction with the substrate, because stronger interactions would perturb the transition matrix element and, consequently, the photoemission line shape at the interface.43,44 The ratio of the integrated signal intensities of the different lines (C 1s, N B
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research
Figure 2. Thickness dependent (a) C 1s, (b) N 1s, and (c) O 1s core level photoemission spectra of NitPyn thin films deposited on Au(111) single crystals. Photon energy 620 eV. Time-dependent (d) C 1s and (e) N 1s signals. Color scale: Green represents the background signal and white the initial peak intensity. Further details in ref 30. Photon energy 600 eV.
Figure 3. Thickness dependent N 1s core level spectra (together with their peak-fit analysis28) of NitPyn thin films deposited on rutile TiO2(110) single crystals: (a) 1.0 nm nominally thick film; (b, c) 0.3 nm nominally thick film deposited on two surfaces with different Ti3+ 3d state densities (17% and 3%, respectively). The satellite S is due to electronic relaxation effects, occurring as a response of the system to the core−hole creation in the photoemission process.26,28 Its intensity varies with the different chemical environments. In panel a, the radical is intact; in panels b and c, the radical is chemisorbed, and the interaction, as discussed in the text, is different in the two cases. Photon energy 1486.6 eV (Al Kα).
possible to deposit these radicals without degradation, that is, the preparation conditions do not affect the paramagnetic character of the molecules. Thus, we deposited NitPyn and a second pyrene derivative of the nitronyl nitroxide radical (PPN, Figure 1) on rutile TiO2(110) single crystals.26,28,35 The C 1s core level spectra do not show appreciable differences with respect to those discussed in the previous section, while the O 1s core level signal is the convolution of the substrate and the molecule signals, making its analysis less reliable.28 Here we focus on the N 1s core level spectra that are the most interesting for our purposes (Figure 3). At first sight, the N 1s core level spectra seem more complicated than that of NitPyn on gold and show more features. A deeper look reveals a few analogies, such as the main line at around 402 eV, as in NitPyn on gold, that dominates the spectrum. However, it is
Stoichiometric XPS features correspond to the expected ESR radicals patterns; in contrast, “different” ESR patterns are mirrored by nonstoichiometric XPS spectroscopic lines.25,31 This is a big advantage when working with radicals, because we can directly access the electronic structure of the radical by XPS. When there is a change in the chemical environment of the radical, this can be readily observed with XPS and consequently mirrored by ESR. The opposite also holds true.25
■
TWO PYRENE DERIVATIVES OF THE NITRONYL NITROXIDE RADICAL AT THE INTERFACE WITH TiO2(110) SINGLE CRYSTALS With our first experiments, we established the exact OMBD recipe for growing thin films of derivatives of the nitronyl nitroxide radical. By using these controlled conditions, it is C
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research
perturbed transition matrix element for some molecules at the interface, due to chemisorptions via the NO group. The combination of photoemission and DFT calculations represents a powerful approach to shed light on the mechanisms governing the contact between complex systems such as a radical and the TiO2 surface, where it is important to fully describe the contact region between two materials (interface) also from the magnetic point of view (spinterface50).
accompanied by several features below 401 eV. These are also observed in the layers at the interface, and they dominate the spectra of the 0.3 nm thick films (Figure 3b,c). These features originate from NitPyn nitrogen atoms that see a chemical environment different than that in a free or physisorbed intact radical. We can understand the mechanisms behind this phenomenon looking at the substrate. The rutile TiO2(110) (1 × 1) surface is characterized by parallel rows of Ti atoms running along the [001] direction, alternated with rows of bridging oxygen atoms. The oxygen atoms protrude above the surface plane.47,48 The surface preparation in UHV consists of alternating cycles of sputtering and annealing. This treatment gives rise to an oxygen deficient surface, because some bridging oxygen atoms are removed.47,48 Some of the vacancies undergo fast hydroxylation for the reaction with OH groups due to the residual water partial pressure.49 Thus, the XPS spectra indicate that the first layer has a mixed nature with contributions due to physisorbed intact NitPyn molecules (the line at around 402 eV) and NitPyn molecules that interact with defect sites and are therein chemisorbed giving rise to the contributions at lower binding energies. These contributions are still visible in the thick film spectra because NitPyn follows the layer-plus-island growth mode, that is, the first layer is never completely covered. Note that we are able to tune the concentration of the chemisorbed species tuning the Ti3+ 3d states densities, using different surface preparation conditions, in agreement with the active role played by the defects in NitPyn chemisorptions (compare Figure 3b,c). We obtained similar results for PPN (Figure 4), though the N 1s core level spectra are complicated
■
“GOING THERMODYNAMICALLY”: DESIGNING RADICALS FOR STABLE THIN FILMS We monitored NitPyn and PPN thin films, with similar morphological characteristics, kept in UHV at room temperature (RT) after deposition, minimizing the beam exposure and switching off the X-ray source between measurements. Figure 5a,b shows the time dependence of the substrate (Au 4f and Si 2p) and the carbon signal (C 1s) for NitPyn deposited on Au(111) and PPN deposited on SiO2/Si(111). The time dependence of the NitPyn Au 4f signal versus the C 1s signal shows a strong increase of the gold signal (Figure 5a). This would suggest desorption of the molecules from the substrate, if it were accompanied by a simultaneous quasiequivalent decrease of the C 1s signal. This is not the case, thus, we can explain this phenomenon by film dewetting and island ripening. This means that the film leaves larger substrate regions free, and the islands reorganize on the substrate minimizing their free energy. To collect more information, we performed ex-situ AFM on the NitPyn films deposited on gold and sapphire. We obtained different root-mean-squared (RMS) roughness values (between 20 and 60 nm) measuring the same samples deposited on gold over several days.38 Sapphire is a stepped surface (0.3 nm step high). The stepped structure impacts the dewetting: We could observe a restrained diffusion over the step edges.27 Monitoring for three consecutive days the film surface exposed to ambient conditions, we obtained the step height distributions shown in Figure 5. They are wider over time, suggesting a reorganization process in which the islands or molecules gather into assemblies at the upper terraces of a step edge giving rise to larger step heights. A completely different scenario is observed in case of the PPN film: The signals from the substrate and from the film are substantially stable, with a slightly increase of the C 1s signal over 1000 min (Figure 5b), due to the adsorption of carbon from the environment, possible also in UHV. This result indicates that the films do not show postgrowth phenomena, that is, PPN films have much stronger vacuum stability than NitPyn films. The only structural difference between NitPyn and PPN is the presence of the pyrazole ring in PPN, but they show very dissimilar film stability. It is possible to understand this different behavior with thermodynamic arguments. The first step to deposit a film onto a substrate is based on a nonequilibrium phenomenon. It is necessary to overcome the thermodynamic equilibrium between adsorption and desorption.51,52 This is achieved by OMBD in supersaturation, Δμ. As a first approximation, supersaturation is a function of substrate temperature, Tsub, and deposition rate, Φ,51,53 and it is given by
Figure 4. Thickness dependent C 1s (left) and N 1s (right) core level spectra of PPN films deposited on TiO2(110) single crystals. Photon energy: 1486.6 eV (Al Kα).
by the presence of the pyrazole nitrogen. A comparison of the thick and thin film spectra gives a clear indication that some PPNs are chemisorbed on the TiO2 surface, because of the broadening of the spectra at the interface. Our first-principles density functional theory calculation results support the experimental observation.35 The nitronyl nitroxide radical does not interact chemically with the ideal TiO2(110) surface and maintains the same magnetic character as in the gas phase (1.0 Bohr magneton); that is, it is physisorbed, in agreement with the experimental finding that most of the molecules are physisorbed when the surface is prepared with very low densities of defective sites. Conversely, in the case of a surface oxygen vacancy, the molecules bind one of their oxygen atoms to the defective site. This decreases their calculated magnetic momentum to 0.2 Bohr magneton. The calculations are in agreement with our observation of a
Δμ = ΔHsub + RTsub[ln( 2πMRTsub Φ) − A]
(1)
ΔHsub is the heat of sublimation, R is the gas constant, M is the molecular weight, and A is an empirical constant related to the entropy of sublimation.53 Supersaturation can be controlled D
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research
Figure 5. Temporal evolution at RT of (a) the Au 4f and C 1s XPS signals for a NitPyn film and (b) the Si 2p and C 1s XPS signals for a PPN film after deposition. Photon energy 1486.6 eV (Al Kα). Time evolution of the step height distribution of a 2.9 nm nominally thick NitPyn film deposited on sapphire over three consecutive days. (c) Clean substrate and (d) first, (e) second, and (f) third day. In the background, the step heights of the film as measured the first day are given as a reference.
films. In Figure 6c, we report a comparison of N 1s core level spectra acquired upon deposition and after 12 h. The spectra do not show any change, indicating chemical stability of the Blatter radical derivative films in UHV. To monitor the film morphology upon air exposure, we measured the XPS core level spectra of a 6.1 nm thick film immediately after evaporation and after air exposure. The overlap of the N 1s core level spectra of a freshly evaporated film and after 45 h air exposure allowed us to confidently rule out relevant changes after 45 h (Figure 6d). The N 1s signal presents changes after keeping the sample 6 days (∼144 h) under air conditions (Figure 6e). Simultaneously, we performed AFM measurements for four consecutive days (Figure 6f). The roughness experienced a small change (from ∼8.7 nm to ∼9.4 nm, 7%; it can be hindered in devices by encapsulation), attesting once more the good stability of the films. Remarkable results were obtained by ESR that show constant ratios of the peak heights over 3 months of air exposure,33 confirming that the film characteristics fulfill our aim for synthesizing the Blatter radical derivative.
during growth by varying substrate temperature and deposition rate. In the case of organic radicals, the range of allowed values for these two parameters may be very limited; for example, high substrate temperatures and deposition rates may lead to decomposition of the radicals. A way to address this issue is designing the molecules adjusting their molecular weight (as in NitPyn versus PPN) and their heat and entropy of sublimation. This tunes the supersaturation giving good film forming properties and thermal stability in the temperature range of interest for devices. Considering this concept, we designed a derivative of the Blatter radical (1,3-diphenyl-1,4-dihydro-1,2,4-benzotriazin-4yl).33 The Blatter radical is a chemically (super) stable radical:54−56 The best starting “block” for developing our idea. Pyrene is a well-known stable chromophore. We reasoned that fusing a Blatter’s radical to pyrene would lower the overall vapor pressure, increase the molecular weight, and give a molecule with good film forming properties. To prove our ideas valid, first we used in situ XPS and ex-situ ESR spectra, confirming that the Blatter radical derivative films were deposited without molecular degradation;33 then we investigated their vacuum, air, and thermal stability. To check the vacuum stability, we monitored substrate and film XPS signals as a function of time, as previously done for NitPyn and PPN. Figure 6b shows the time dependence of the C 1s and Si 2p signals. The signals stay substantially constant. There is no evidence of dewetting or ripening processes in the
■
CONCLUSIONS AND OUTLOOK A point of paramount importance in electronics is the growth of materials as thin films at the nanoscale, which provides several advantages in device fabrication. With our work, we demonstrate that it is possible to evaporate and deposit organic radicals onto well-defined surfaces under controlled conditions E
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research
Figure 6. (a) The pyrene-Blatter radical derivative molecular structure. (b) Temporal evolution at RT of the Si 2p and C 1s XPS signals for a 6.6 nm nominally thick film after deposition. (c) Comparison of N 1s core level spectra of a freshly evaporated film and after 12 h, keeping the sample at RT in UHV. Comparison of N 1s core level spectra of a freshly evaporated film and after (d) 45 h and (e) 6 days of air exposure. (f) RMS roughness (10 μm × 10 μm images) evolution as a function of time for a 6.6 nm nominally thick film kept at ambient conditions. The time 0 indicates when the sample has been taken out of the vacuum chamber. Photon energy 1486.6 eV (Al Kα).
properties. As a matter of fact, a characteristic of molecular films is polymorphism. By different preparation conditions, that is, by tuning deposition rate and substrate temperature, using specific substrates, it is possible to tune the arrangement of the molecules in films.62 Magnetism in “classical magnets” is based on specific interactions between the magnetic moments; thus, controlling the molecular arrangement gives the opportunity to tune the mutual position and orientation of the molecules, that is, of the single magnetic moments. Pushing this idea to its extreme, understanding how the molecules are arranged in thin films of organic radicals and tuning their arrangement will make it possible to “imprint”, so to say, the wished intermolecular coupling, setting film properties including magnetism. A deep understanding of stable radical/inorganic spinterfaces may also open the way to really use radicals in spin-based electronics63 or as molecular quantum bits10,14,22,64 using dedicated configurations or attaching them to a surface, as proposed for other molecular quantum bits65,66
without degradation. Using soft X-ray spectroscopies, performed also at synchrotrons, we shed light on organic radical thin film processes, interfaces, and spinterfaces at the metal/ metal oxide surfaces. We investigated the vacuum, thermal, and air stability of the films. What it is also important is that we identify the concepts necessary to design organic radicals bearing in mind the thermodynamic factors that govern thin film stability, with the purpose of obtaining stable thin films for applications. In this respect, a fundamental point is starting from a stable radical: A large variety of novel, stable radicals were synthesized in recent years, such as nitroxides, nitronyl and imino nitroxides, aminyls, verdazyls, and Blatter radicals (Figure 1),19,45,56−61 using two main strategies for the stabilization of organic radicals: Protection of the spin centers with steric hindrance and delocalization of the unpaired electron. Comparing the stability of the various radical films that we investigated, we noticed that molecular structure and delocalization of the unpaired electron play a role: Our less stable films were based on a TEMPO derivative,34 while the most stable were based on the Blatter-radical derivative.33 Our work combines methods and concepts brought together from different fields that were never considered complementary. An unconventional multidisciplinary approach not only sheds light on physical and chemical properties of materials but suggests new ideas and solutions for future applications. Our work allows foreseeing new experiments to investigate the magnetic character of the thin films versus their structural
■
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.accounts.7b00612. Experimental methods and Blatter-pyrene N 1s spectrum assignment (PDF) F
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research
■
(8) Grillo, F.; Mugnaini, V.; Oliveros, M.; Francis, S. M.; Choi, D. J.; Rastei, M. V.; Limot, L.; Cepek, C.; Pedio, M.; Bromley, S. T.; Richardson, N. V.; Bucher, J. P.; Veciana, J. Chiral Conformation at a Molecular Level of a Propeller-Like Open-Shell Molecule on Au(111). J. Phys. Chem. Lett. 2012, 3, 1559−1564. (9) Huang, Z.; Zhang, Y.; He, Y.; Song, H.; Yin, C.; Wu, K. A Chemist’s Overview of Surface Electron Spins. Chem. Soc. Rev. 2017, 46, 1955−1976. (10) Nakazawa, S.; Nishida, S.; Ise, T.; Yoshino, T.; Mori, N.; Rahimi, R. D.; Sato, K.; Morita, Y.; Toyota, K.; Shiomi, D.; Kitagawa, M.; Hara, H.; Carl, P.; Höfer, P.; Takui, T. A Synthetic Two-Spin Quantum Bit: G-Engineered Exchange-Coupled Biradical Designed for ControlledNot Gate Operations. Angew. Chem., Int. Ed. 2012, 51, 9860−9864. (11) Sproules, S. Molecules as Electron Spin Qubits. Electron Paramagnetic Resonance; The Royal Society of Chemistry: 2017; Vol. 25, pp 61−97. (12) Oyaizu, K.; Nishide, H. Radical Polymers for Organic Electronic Devices: A Radical Departure from Conjugated Polymers? Adv. Mater. 2009, 21, 2339−2344. (13) Suga, T.; Konishi, H.; Nishide, H. Photocrosslinked Nitroxide Polymer Cathode-Active Materials for Application in an OrganicBased Paper Battery. Chem. Commun. 2007, 1730−1732. (14) Lehmann, J.; Gaita-Arino, A.; Coronado, E.; Loss, D. Quantum Computing with Molecular Spin Systems. J. Mater. Chem. 2009, 19, 1672−1677. (15) Olson, J.; Cao, Y.; Romero, J.; Johnson, P.; Dallaire-Demers, P.L.; Sawaya, N.; Narang, P.; Kivlichan, I.; Wasielewski, M. R.; AspuruGuzik, A. Quantum Information and Computation for Chemistry. 2016, arXiv:1706.05413, https://arxiv.org/abs/1706.05413. (16) Ratera, I.; Veciana, J. Playing with Organic Radicals as Building Blocks for Functional Molecular Materials. Chem. Soc. Rev. 2012, 41, 303−349. (17) Mas-Torrent, M.; Crivillers, N.; Rovira, C.; Veciana, J. Attaching Persistent Organic Free Radicals to Surfaces: How and Why. Chem. Rev. (Washington, DC, U. S.) 2012, 112, 2506−2527. (18) Mas-Torrent, M.; Crivillers, N.; Mugnaini, V.; Ratera, I.; Rovira, C.; Veciana, J. Organic Radicals on Surfaces: Towards Molecular Spintronics. J. Mater. Chem. 2009, 19, 1691−1695. (19) Caneschi, A.; Gatteschi, D.; Sessoli, R.; Rey, P. Toward Molecular Magnets: The Metal-Radical Approach. Acc. Chem. Res. 1989, 22, 392−398. (20) Miller, J. S. Magnetically Ordered Molecule-Based Materials. Chem. Soc. Rev. 2011, 40, 3266−3296. (21) Rajca, A. Organic Diradicals and Polyradicals: From Spin Coupling to Magnetism? Chem. Rev. 1994, 94, 871−893. (22) Boulon, M.-E.; Fernandez, A.; Moreno Pineda, E.; Chilton, N. F.; Timco, G.; Fielding, A. J.; Winpenny, R. E. P. Measuring Spin··· Spin Interactions between Heterospins in a Hybrid [2]Rotaxane. Angew. Chem., Int. Ed. 2017, 56, 3876−3879. (23) Forrest, S. R. Ultrathin Organic Films Grown by Organic Molecular Beam Deposition and Related Techniques. Chem. Rev. 1997, 97, 1793−1896. (24) Hüfner, S. Photoelectron Spectroscopy, 3rd ed.; Springer-Verlag: Berlin Heidelberg, 2003. (25) Savu, S.-A.; Biswas, I.; Sorace, L.; Mannini, M.; Rovai, D.; Caneschi, A.; Chassé, T.; Casu, M. B. Nanoscale Assembly of Paramagnetic Organic Radicals on Au(111) Single Crystals. Chem. Eur. J. 2013, 19, 3445−3450. (26) Kakavandi, R.; Savu, S.-A.; Caneschi, A.; Casu, M. B. Paramagnetic Character in Thin Films of Metal-Free Organic Magnets Deposited on TiO2(110) Single Crystals. J. Phys. Chem. C 2013, 117, 26675−26679. (27) Abb, S.; Savu, S.-A.; Caneschi, A.; Chassé, T.; Casu, M. B. Paramagnetic Nitronyl Nitroxide Radicals on Al2O3(11−20) Single Crystals: Nanoscale Assembly, Morphology, Electronic Structure, and Paramagnetic Character toward Future Applications. ACS Appl. Mater. Interfaces 2013, 5, 13006−13011. (28) Kakavandi, R.; Savu, S.-A.; Caneschi, A.; Chasse, T.; Casu, M. B. At the Interface between Organic Radicals and TiO2(110) Single
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
M. Benedetta Casu: 0000-0002-5659-7040 Funding
This work was supported by German Research Foundation (DFG, Grants CA852/5-1 and CA852/5-2), HelmholtzZentrum Berlin (HZB), and Wilhelm-Schuler-Stiftung with their research award. Notes
The author declares no competing financial interest. Biography Maria Benedetta Casu received her Ph.D. in Physics at the University of Potsdam (2001). After a postdoctoral stay at the University of Würzburg, in 2006 she joined the Institute of Physical and Theoretical Chemistry at the University of Tübingen where she is presently the head of the “Organic Nano&Magnetic” group. Her research interests include organic/inorganic interfaces and thin films investigated using photoemission spectroscopy and microscopy.
■
ACKNOWLEDGMENTS I thank BESSY @HZB and Elettra-Sincrotrone Trieste for providing beamtime, HZB User Coordination for support, R. Ovsyannikov, M. Oehzelt, and S. Pohl for technical support at BESSY, Andrea Locatelli and Onur Mentes for beamtime support and helpful discussions at Elettra, Hilmar Adler, Matteo Lucian, Elke Nadler, and Wolfgang Neu for technical support, Thomas Chassé for the access to the photoelectron spectroscopy laboratory, Dante Gatteschi for the very first stimulating discussion (back in 2008), Arrigo Calzolari for the DFT calculations, and Andrea Caneschi and Martin Baumgarten for providing their radicals and for helpful discussions. I am deeply indebted to Andrzej Rajca for the synthesis of the Blatter radical derivative and for stimulating discussions.
■
REFERENCES
(1) Koch, N. Organic Electronic Devices and Their Functional Interfaces. ChemPhysChem 2007, 8, 1438−1455. (2) Heimel, G.; Salzmann, I.; Duhm, S.; Koch, N. Design of Organic Semiconductors from Molecular Electrostatics. Chem. Mater. 2011, 23, 359−377. (3) Koch, N. Electronic Structure of Interfaces with Conjugated Organic Materials. Phys. Status Solidi RRL 2012, 6, 277−293. (4) Heimel, G.; Duhm, S.; Salzmann, I.; Gerlach, A.; Strozecka, A.; Niederhausen, J.; Burker, C.; Hosokai, T.; Fernandez-Torrente, I.; Schulze, G.; Winkler, S.; Wilke, A.; Schlesinger, R.; Frisch, J.; Broker, B.; Vollmer, A.; Detlefs, B.; Pflaum, J.; Kera, S.; Franke, K. J.; Ueno, N.; Pascual, J. I.; Schreiber, F.; Koch, N. Charged and Metallic Molecular Monolayers through Surface-Induced Aromatic Stabilization. Nat. Chem. 2013, 5, 187−194. (5) Koch, N.; Ueno, N.; Wee, A. T. S. The Molecule-Metal Interface; Wiley: 2013. (6) Shekhah, O.; Roques, N.; Mugnaini, V.; Munuera, C.; Ocal, C.; Veciana, J.; Wö ll, C. Grafting of Monocarboxylic Substituted Polychlorotriphenylmethyl Radicals onto a Cooh-Functionalized SelfAssembled Monolayer through Copper (Ii) Metal Ions. Langmuir 2008, 24, 6640−6648. (7) Grillo, F.; Fruchtl, H.; Francis, S. M.; Mugnaini, V.; Oliveros, M.; Veciana, J.; Richardson, N. V. An Ordered Organic Radical Adsorbed on a Cu-Doped Au(111) Surface. Nanoscale 2012, 4, 6718−6721. G
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
Article
Accounts of Chemical Research Crystals: Electronic Structure and Paramagnetic Character. Chem. Commun. (Cambridge, U. K.) 2013, 49, 10103−10105. (29) Kakavandi, R.; Savu, S.-A.; Sorace, L.; Rovai, D.; Mannini, M.; Casu, M. B. Core-Hole Screening, Electronic Structure, and Paramagnetic Character in Thin Films of Organic Radicals Deposited on SiO2/Si(111). J. Phys. Chem. C 2014, 118, 8044−8049. (30) Caneschi, A.; Casu, M. B. Substrate-Induced Effects in Thin Films of a Potential Magnet Composed of Metal-Free Organic Radicals Deposited on Si(111). Chem. Commun. (Cambridge, U. K.) 2014, 50, 13510−13513. (31) Kakavandi, R.; Ravat, P.; Savu, S. A.; Borozdina, Y. B.; Baumgarten, M.; Casu, M. B. Electronic Structure and Stability of Fluorophore−Nitroxide Radicals from Ultrahigh Vacuum to Air Exposure. ACS Appl. Mater. Interfaces 2015, 7, 1685−1692. (32) Casu, M. B. Growth, Structure, and Electronic Properties in Organic Thin Films Deposited on Metal Surfaces Investigated by Low Energy Electron Microscopy and Photoelectron Emission Microscopy. J. Electron Spectrosc. Relat. Phenom. 2015, 204, 39−48. (33) Ciccullo, F.; Gallagher, N. M.; Geladari, O.; Chasse, T.; Rajca, A.; Casu, M. B. A Derivative of the Blatter Radical as a Potential MetalFree Magnet for Stable Thin Films and Interfaces. ACS Appl. Mater. Interfaces 2016, 8, 1805−1812. (34) Arantes, C.; Chernick, E. T.; Gruber, M.; Rocco, M. L. M.; Chasse, T.; Tykwinski, R. R.; Casu, M. B. Interplay between SolutionProcessing and Electronic Structure in Metal-Free Organic Magnets Based on a Tempo Pentacene Derivative. J. Phys. Chem. C 2016, 120, 3289−3294. (35) Kakavandi, R.; Calzolari, A.; Borozdina, Y. B.; Ravat, P.; Chassé, T.; Baumgarten, M.; Casu, M. B. Unraveling the Mark of Surface Defects on a Spinterface: The Nitronyl Nitroxide/TiO2(110) Interface. Nano Res. 2016, 9, 3515−3527. (36) Siegbahn, K. M. Nobel Lecture: Electron Spectroscopy for Atoms, Molecules and Condensed Matter. http://www.nobelprize. org/nobel_prizes/physics/laureates/1981/siegbahn-lecture.html (3 Feb 2017). (37) Schöll, A.; Zou, Y.; Jung, M.; Schmidt, T.; Fink, R.; Umbach, E. Line Shapes and Satellites in High-Resolution X-Ray Photoelectron Spectra of Large Π-Conjugated Organic Molecules. J. Chem. Phys. 2004, 121, 10260−10267. (38) Savu, S.-A.; Casu, M. B.; Schundelmeier, S.; Abb, S.; Tonshoff, C.; Bettinger, H. F.; Chassé, T. Nanoscale Assembly, Morphology and Screening Effects in Nanorods of Newly Synthesized Substituted Pentacenes. RSC Adv. 2012, 2, 5112−5118. (39) Likhtenshtein, G. Novel Fluorescent Methods for Biotechnological and Biomedical Sensoring: Assessing Antioxidants, Reactive Radicals, No Dynamics, Immunoassay, and Biomembranes Fluidity. Appl. Biochem. Biotechnol. 2009, 152, 135−155. (40) Wang, H.; Zhang, D.; Guo, X.; Zhu, L.; Shuai, Z.; Zhu, D. Tuning the Fluorescence of 1-Imino Nitroxide Pyrene with Two Chemical Inputs: Mimicking the Performance of an ″and″ Gate. Chem. Commun. (Cambridge, U. K.) 2004, 670−671. (41) Hughes, B. K.; Braunecker, W. A.; Ferguson, A. J.; Kemper, T. W.; Larsen, R. E.; Gennett, T. Quenching of the Perylene Fluorophore by Stable Nitroxide Radical-Containing Macromolecules. J. Phys. Chem. B 2014, 118, 12541−12548. (42) Tamura, M.; Nakazawa, Y.; Shiomi, D.; Nozawa, K.; Hosokoshi, Y.; Ishikawa, M.; Takahashi, M.; Kinoshita, M. Bulk Ferromagnetism in the Β-Phase Crystal of the P-Nitrophenyl Nitronyl Nitroxide Radical. Chem. Phys. Lett. 1991, 186, 401−404. (43) Häming, M.; Schöll, A.; Umbach, E.; Reinert, F. AdsorbateSubstrate Charge Transfer and Electron-Hole Correlation at Adsorbate/Metal Interfaces. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85, 235132. (44) Savu, S. A.; Biddau, G.; Pardini, L.; Bula, R.; Bettinger, H. F.; Draxl, C.; Chasse, T.; Casu, M. B. Fingerprint of Fractional Charge Transfer at the Metal/Organic Interface. J. Phys. Chem. C 2015, 119, 12538−12544.
(45) Ullman, E. F.; et al. Stable Free Radicals. Viii. New Imino, Amidino, and Carbamoyl Nitroxides. J. Org. Chem. 1970, 35, 3623− 3631. (46) Bauer, E. Surface Microscopy with Low Energy Electrons; Springer: 2014. (47) Diebold, U. The Surface Science of Titanium Dioxide. Surf. Sci. Rep. 2003, 48, 53−229. (48) Lun Pang, C.; Lindsay, R.; Thornton, G. Chemical Reactions on Rutile TiO2(110). Chem. Soc. Rev. 2008, 37, 2328−2353. (49) Di Valentin, C.; Pacchioni, G.; Selloni, A. Electronic Structure of Defect States in Hydroxylated and Reduced Rutile TiO2(110) Surfaces. Phys. Rev. Lett. 2006, 97, 166803. (50) Sanvito, S. Molecular Spintronics: The Rise of Spinterface Science. Nat. Phys. 2010, 6, 562−564. (51) Ohring, M. Materials Science of Thin Films; Academic Press: San Diego, CA, 2002. (52) Venables, J. A. Introduction to Surface and Thin Film Processes; Cambridge University Press: Cambridge, U.K., 2000. (53) Verlaak, S.; Steudel, S.; Heremans, P.; Janssen, D.; Deleuze, M. S. Nucleation of Organic Semiconductors on Inert Substrates. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 68, 195409. (54) Zheng, Y.; Miao, M.-s.; Kemei, M. C.; Seshadri, R.; Wudl, F. The Pyreno-Triazinyl Radical − Magnetic and Sensor Properties. Isr. J. Chem. 2014, 54, 774−778. (55) Constantinides, C. P.; Koutentis, P. A.; Krassos, H.; Rawson, J. M.; Tasiopoulos, A. J. Characterization and Magnetic Properties of a “Super Stable” Radical 1,3-Diphenyl-7-Trifluoromethyl-1,4-Dihydro1,2,4-Benzotriazin-4-Yl. J. Org. Chem. 2011, 76, 2798−2806. (56) Blatter, H. M.; Lukaszewski, H. A New Stable Free Radical. Tetrahedron Lett. 1968, 9, 2701−2705. (57) Rajca, A.; Shiraishi, K.; Pink, M.; Rajca, S. Triplet (S = 1) Ground State Aminyl Diradical. J. Am. Chem. Soc. 2007, 129, 7232− 7233. (58) Rajca, A.; Olankitwanit, A.; Wang, Y.; Boratyński, P. J.; Pink, M.; Rajca, S. High-Spin S = 2 Ground State Aminyl Tetraradicals. J. Am. Chem. Soc. 2013, 135, 18205−18215. (59) Morgan, I. S.; Peuronen, A.; Hänninen, M. M.; Reed, R. W.; Clérac, R.; Tuononen, H. M. 1-Phenyl-3-(Pyrid-2-Yl)Benzo[E][1,2,4]Triazinyl: The First “Blatter Radical” for Coordination Chemistry. Inorg. Chem. 2014, 53, 33−35. (60) Demetriou, M.; Berezin, A. A.; Koutentis, P. A.; KrasiaChristoforou, T. Benzotriazinyl-Mediated Controlled Radical Polymerization of Styrene. Polym. Int. 2014, 63, 674−679. (61) Yan, B.; Cramen, J.; McDonald, R.; Frank, N. L. Ferromagnetic Spin-Delocalized Electron Donors for Multifunctional Materials: [Small Pi]-Conjugated Benzotriazinyl Radicals. Chem. Commun. 2011, 47, 3201−3203. (62) Jones, A. O. F.; Chattopadhyay, B.; Geerts, Y. H.; Resel, R. Substrate-Induced and Thin-Film Phases: Polymorphism of Organic Materials on Surfaces. Adv. Funct. Mater. 2016, 26, 2233−2255. (63) Cinchetti, M.; Dediu, V. A.; Hueso, L. E. Activating the Molecular Spinterface. Nat. Mater. 2017, 16, 507. (64) Sato, K.; Nakazawa, S.; Rahimi, R.; Ise, T.; Nishida, S.; Yoshino, T.; Mori, N.; Toyota, K.; Shiomi, D.; Yakiyama, Y.; Morita, Y.; Kitagawa, M.; Nakasuji, K.; Nakahara, M.; Hara, H.; Carl, P.; Hofer, P.; Takui, T. Molecular Electron-Spin Quantum Computers and Quantum Information Processing: Pulse-Based Electron Magnetic Resonance Spin Technology Applied to Matter Spin-Qubits. J. Mater. Chem. 2009, 19, 3739−3754. (65) Ghirri, A.; Candini, A.; Affronte, M. Molecular Spins in the Context of Quantum Technologies. Magnetochemistry 2017, 3, 12. (66) Jenkins, M. D.; Zueco, D.; Roubeau, O.; Aromi, G.; Majer, J.; Luis, F. A Scalable Architecture for Quantum Computation with Molecular Nanomagnets. Dalton Trans. 2016, 45, 16682−16693.
H
DOI: 10.1021/acs.accounts.7b00612 Acc. Chem. Res. XXXX, XXX, XXX−XXX
