J. Phys. Chem. B 2008, 112, 9729–9735
9729
Grafting Derivatives of Mn6 Single-Molecule Magnets with High Anisotropy Energy Barrier on Au(111) Surface F. Moro,*,†,‡ V. Corradini,‡ M. Evangelisti,‡ V. De Renzi,†,‡ R. Biagi,†,‡ U. del Pennino,†,‡ C. J. Milios,§ L. F. Jones,§ and E. K. Brechin§ Dipartimento di Fisica, UniVersita` di Modena e Reggio Emilia, Modena, Italy, CNR-INFM National Center on nanoStructures and bioSystems at Surfaces (S3), Modena, Italy, and School of Chemistry, The UniVersity of Edinburgh, EH9 3JJ Edinburgh, United Kingdom ReceiVed: March 13, 2008; ReVised Manuscript ReceiVed: May 06, 2008
We study the magnetic properties of two new functionalized single-molecule magnets belonging to the Mn6 family (general formula [MnIII6O2(R-sao)6(O2C-th)2L4-6], where R ) H (1) or Et (2), HO2C-th ) 3-thiophene carboxylic acid, L ) EtOH, H2O and saoH2 is salicylaldoxime) and their grafting on the Au(111) surface. Complex 1 exhibits spin ground-state S ) 4, as the result of ferromagnetic coupling between the two antiferromagnetic MnIII3 triangles, while slight structural changes in complex 2, switch the dominant magnetic exchange interactions from anti- to ferromagnetic, enhancing the spin ground-state to S ) 12 and, consequently, the effective energy barrier for the relaxation of magnetization. Direct-current and alternating-current magnetic susceptibility measurements show that the functionalized complexes preserve the main magnetic properties of the corresponding not-functionalized Mn6 clusters (i.e., total spin value and magnetic behavior as a function of temperature), though a reduction of the anisotropy barrier is observed in complex 2. For both complexes, the -O2C-th functionalization allows the direct grafting on Au(111) surface by liquid-phase deposition. X-ray photoemission spectroscopy demonstrates that the stoichiometry of the molecular cores is preserved after grafting. Scanning tunneling microscopy (STM) reveals a sub-monolayer distribution of isolated clusters with a slightly higher coverage for complex 1. The cluster stability in the STM images and the S-2p energy positions demonstrate, for both derivatives, the strength of the grafting with the gold surface. Introduction Nanoscale magnetic systems are currently the object of intense research activity. In recent years, remarkable progress has been made in the nanofabrication, control and study of such systems, leading to significant advances in our ability to test fundamental theoretical concepts in quantum mechanics and magnetism. Moreover, the control of magnetic phenomena on this scale may foster enormous technological impact, in particular concerning the development of nanoscale memory elements and qubits for quantum computation. To this end, a unique and very promising route is provided by the study and exploitation of single-molecule magnets (SMMs).1–6 Among them a large class is based on transition-metal (TM) ions. They are characterized by a large spin groundstate S, which is the net result of mutual interactions between their constituent TM ions and, by a negative Ising anisotropy D, which leads to an energy barrier U between the two fundamental states S and -S, equal to S2 D or (S2 - 1/4) D for integer or half-integer spin ground-states, respectively. The barrier hinders the reversal of the molecular spins below the blocking temperature TB, resulting in the magnetic bistability of each molecule.7 The first and best studied SMMs were the mixed-valence members of the Mn12 family, typically with S ) 10 spin ground-state, TB around a few kelvin and an effective energy barrier of magnetization * Corresponding author. E-mail address:
[email protected] (F. Moro). † Universita ` di Modena e Reggio Emilia. ‡ CNR-INFM National Center on nanoStructures and bioSystems at Surfaces (S3). § The University of Edinburgh.
relaxation (Ueff) as large as 74 K.7–10 At the moment, any practical use of SMMs is hindered by the low TB which does not exceed a few kelvin in the most fortunate cases. In recent years, great research efforts have therefore been devoted in the synthesis of molecular systems with larger magnetic anisotropy. To this end, a very interesting example is provided by a S ) 12 member of the cluster family of general formula [MnIII6O2(R-sao)6(O2CR)2(ROH)x(H2O)y] (hereafter Mn6), which possesses the new Ueff record-value of 86 K, leading to a blocking temperature of about 4.5 K.11,12 The representative core of the [MnIII6O2(sao)6(O2CR)2(ROH)4] derivative13,14 contains a nonplanar [MnIII6(µ3-O2-)2(µ2-OR)2]12+ unit of two off-set, stacked [MnIII3(µ3- O2-)]7+ triangular subunits linked by two central oximato O-atoms (see the core of the structure in Figure 1, left panel). This class of compounds shows a ferromagnetic (FM) exchange between the two antiferromagnetically (AF) coupled MnIII3 triangles. Fitting of the magnetization data collected in fields up to 7 T affords S ) 4 for the spin ground state.15 However a number of important structural changes leading to different values of S and Ueff take place in the related complexes [MnIII6O2(R-sao)6(O2CR)2 (EtOH)4(H2O)2] with R ) Et, Me, Ph,11 where the increased steric bulk and nonplanarity of the R-sao2- ligands (see Scheme 1) cause a shortening (by ∼1 Å) of the phenolato oxygen-square pyramidal Mn distance and a severe twisting of the Mn-N-OMn moieties within each Mn3 subunit (see the core of the structure in Figure 1, right panel). The structural changes force a switch in the dominant magnetic exchange interactions from antiferromagnetic to ferromagnetic and thus the stabilization of an S ) 12 spin ground-state.11,12 These compounds however, are characterized by a weak exchange value (J < 1 cm-1), which
10.1021/jp802195x CCC: $40.75 2008 American Chemical Society Published on Web 07/23/2008
9730 J. Phys. Chem. B, Vol. 112, No. 32, 2008
Moro et al.
Figure 1. The Mn6-3tpc structures of complex 1 with S ) 4 (left panel) and complex 2 with S ) 12 (right panel). The large and medium circles depict the MnIII ions and the sulfur atoms respectively. While dark and clear small circles indicate nitrogen and oxygen. The lines link the carbon atoms. H-atoms are omitted for clarity.
SCHEME 1: The Structures of R-saoH2 (R ) H, Me, Et; Left) and HO2C-th (Right)
results in the population of low-lying excited states (S ) 11, S ) 10,..) and then tunneling involving excited-state multiplets, leading to a dramatic reduction in the energy barrier of magnetization relaxation.11 This problem has been successfully solved by further replacement of the carboxylates via a simple metathesis type reactions, which produces analogous complexes whose cores are identical except for the Mn-N-O-Mn torsion angles (Rv) that are modified according to the steric bulk of the ligand. The complexes with the largest torsion angles12 show an increase in the magnitude of the pairwise exchange parameter (J) and an increase in the effective Ueff up to 86 K, higher than that of Mn12-acetate. Beside the current efforts in the chemical synthesis of highUeff compounds, another fundamental issue that needs to be addressed for any practical exploitation of SMMs, is the development of strategies for an effective grafting of nanomagnets on surfaces obtaining bidimensional distributions of singly addressable units.16–23 One current approach is the SMM functionalization using groups that provide strong bonding with the substrate (typically Au or Si), with the aim to preserve the chemical stability and the magnetic properties of the cluster. This task is nontrivial as functionalization may influence the physical properties of the nanomagnets. For example, hysteresis has been reported to disappear in single layers of Mn12 grafted on gold surfaces, thus showing that the interaction with the substrate induces profound modifications in the magnetic and energetic properties of Mn12 core.24 In the case of Mn6 clusters, it is well known that even small modifications in the structure
can cause large changes in the magnetic response.25 Nevertheless, the small size of Mn6 molecules, which translates into fewer exchange-couplings within the clusters and the fact that they are formed by only one kind of manganese ions (MnIII), likely make them easier to characterize with respect to the Mn12 class. This fact makes this family of SMM particularly promising to verify the hypothesis that the molecules preserve their corestructure and physical properties upon surface grafting. In this work, we present the results of a combined chemical-physical characterization of two novel Mn6-3tpc derivatives:[Mn(III)6O2(sao)6(O2C-tpc)2(EtOH)4](1)and[Mn(III)6O2(Et-sao)6(O2C-tpc)2(EtOH)4(H2O)2] (2) functionalized with two organic ligands containing a S-atom (3-thiophenecarboxylic acid, HO2C-tpc) for direct grafting on gold surfaces by exploiting the strong S-Au affinity (see Scheme 1). In Figure 1, the functionalized compounds are shown; the 3-thiophenecarboxylate ligands, which present the S-atoms in the outer position (3) of the thiophene (TP) ring, are pointing out in opposite direction roughly perpendicular to the [Mn3] planes. We have investigated the bulk magnetic properties of the two functionalized complexes by means of variable-temperature direct-current (DC) and alternating-current (AC) magnetic susceptibility measurements, checking for a possible influence of the functionalization on the magnetic properties of the derivatives. Afterward, the effectiveness of the functionalization for surface grafting has been verified, obtaining for both complexes 1 and 2, a thin (submonolayer) distribution of SMMs on the Au surface. The cluster stoichiometry after functionalization and after grafting, the nature of the ligands-gold bonding and the integrity of the grafted molecules have been studied by X-ray photoemission spectroscopy (XPS). Moreover, the morphology, surface coverage and grafting strength of the adlayers have been investigated by scanning tunneling microscopy (STM). The results presented here show that these compounds are promising candidates for investigating the magnetic properties of thin Mn6 grafted adlayers.
Grafting Derivatives of Mn6 Single-Molecule Magnets
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9731
Experimental Section General Synthetic Methodology for Synthesizing Mn6 Derivatives. All manipulations were performed under aerobic conditions using materials as received (reagent grade). Caution! Although we encountered no problems, care should be taken when using the potentially explosive perchlorate anion. The derivatized oximes were synthesized as described elsewhere.26 The complexes 1 and 2 were prepared as previously described.25 Method 1. To pale pink solutions of Mn(ClO4)2 · 6H2O in EtOH were added equivalent amounts of the appropriate oxime, the corresponding carboxylic acid, and base (CH3ONa or NEt4OH). The solutions were left stirring for ∼30 min, filtered, and then left to slowly evaporate. In each case, suitable crystals grew after a period of 3-5 days. Method 2. The sodium salt of the corresponding carboxylic acid was treated with equivalent amounts of Mn(ClO4)2 · 6H2O, the appropriate oxime, and base (CH3ONa or NEt4OH) in MeOH (or EtOH). Single crystals were grown upon slow evaporation. For both complexes the yields vary from minimum of 30% to a maximum of 50%. Magnetic Measurements. Variable temperature DC and AC magnetic susceptibility data down to 1.8 K were collected on polycrystalline samples (PC) of 1 and 2 by a calibrated commercial magnetometer. Diamagnetic corrections were applied to the observed paramagnetic susceptibilities using Pascal’s constants. Surface Studies. The Au(111) surface was cleaned by repeated cycles of Ar+ sputtering-annealing in ultrahigh vacuum (UHV) environment. Thin films of both compounds were obtained by dipping for 10 min the Au single crystal in a 1 mM CH2Cl2 solution of complex 1 or 2, rinsing in CH2Cl2, and rapidly introducing it into the UHV experimental chambers. As directly verified by STM, for both complexes this procedure gave the highest coverage attainable, which corresponds to less than a complete Monolayer (s-ML). For comparison, XPS spectra were also taken on thick films (TF) of 1 and 2, obtained by drop casting the saturated solutions on freshly cleaved highly oriented pyrolityc graphite (HOPG). STM measurements were carried out by an Omicron UHV VT-STM system. The tips used were electrochemically etched tungsten wires. The adlayer morphology and cluster size distributions were evaluated from a set of images obtained independently in identical experimental conditions. Room temperature STM acquisitions were carried out in constantcurrent mode with typical imaging conditions of 2.0 V and 30 pA in order to minimize dragging and damaging of the soft organic materials with the scanning tip. XPS measurements were performed using an Omicron hemispherical analyzer (EA125) and a Mg KR X-ray source (hν ) 1253.6 eV). Results and Discussion Magnetic Properties. DC measurements of the magnetic susceptibility (χ), performed on PC of complexes 1 and 2 in the 2-300 K range with an applied field of 0.1 T, are plotted as χT versus T in Figure 2. The χT values at 300 K are 17.7 and 17.3 emu K mol-1 for 1 and 2, respectively. These values are close to the spin-only (g ) 2) value of 18 emu K mol-1 expected for six high-spin MnIII ions at room temperature. For 1, χT decreases monotonically with T, indicating the dominant antiferromagnetic character of the interactions between metal centers. The χT value at low temperature (2 K) is 7.0 emu K mol-1, a value consistent with an S ) 4 spin ground-state, if the effect of magnetic anisotropy is taken into account. In the case of compound 2 instead, the χT dependence on temperature
Figure 2. Temperature-dependence of χT for both complexes in PC form, measured at 0.1 T.
Figure 3. Field-dependence of the molar magnetization M for 1 (top) and 2 (bottom) in PC form. The solid lines are the fit to the experimental data. See text.
presents a maximum (57.5 emu K mol-1) at 3 K, before decreasing at lower temperatures. This nonmonotonic behavior is indicative of dominant ferromagnetic interactions between the metal centers. Anticipating the results inferred from M(H), the relatively large values of the low-temperature χT can be well understood in terms of the ferromagnetic (S ) 12) spin ground-state combined with the magnetic anisotropy responsible for the decrease of χT below 3 K. In order to further detail the nature of both ground states, we collected DC-field magnetization data M(H) in the range 0-7 T for several temperatures between 2-25 K (Figure 3). Fitting of the experimental data with an axial zero-field splitting plus Zeeman Hamiltonian, H ) D(Sˆz2 - S(S + 1)/3) + µBgHSˆ, over the whole field and temperature ranges, afforded the best fit parameters: S ) 4, g ) 2, D ) -1.76 K for 1, and S ) 12, g ) 2, D ) -0.6 K for 2. For the latter sample, only the data collected for T e 15 K were fitted, since the presence of excited spin states invalidates the use of the single-spin approximation at higher temperatures. Because the samples are polycrystalline, all calculated fits were obtained taking into account spin random
9732 J. Phys. Chem. B, Vol. 112, No. 32, 2008
Moro et al.
Figure 4. Temperature-dependencies of the in-phase χ′ (top) and out-of-phase χ′′ (bottom) AC-susceptibilities for 1 (left) and 2 (right) in PC form, at the indicated temperature and frequency ranges.
orientations. Remarkably, the fit to S ) 4 and S ) 12, respectively, does confirm that the main characteristics of Mn6 were not significantly altered by their functionalization. AC-susceptibility measurements were performed on both complexes in the 1.8-15 K T-range in zero-applied field and a 3.5 G AC-excitation oscillating at frequencies ranging between 50 and 9300 Hz. The in-phase χ′ and out-of-phase χ′′ signals for both complexes are shown in Figure 4. For 2, we clearly see the appearance of two peaks in χ′′ indicating the presence of Jahn-Teller (JT) isomers,27 e.g., two molecules differing in the relative orientation of one or more JT axes of the MnIII ions. Though the two isomers show a S ) 12 spin ground-state, as unambiguously determined by the value of the magnetization saturation, they are not magnetically equivalent. This nonequivalence is likely due to different values of the anisotropy barrier and results in a slower (higher temperature) and a faster (lower temperature) relaxing species, as already seen in the prototype Mn12 SMM.28 Indeed, the fit of the frequencydependence of the χ′′ peak maxima to an Arrhenius law τ ) τ0 exp (Ueff/kBT) provides τ0 ) 2.04 × 10-8 s and Ueff ) 28 K for 1; τ0 ) 1.17 × 10-10 s and Ueff ) 55 K for 2FR; and τ0 ) 1.46 × 10-10 s and Ueff ) 67 K for 2SR. Presumably, the two species originate from the structural disorder observed in 2, whereby the S-atoms occupy two equivalent positions (see CIF of 2 in Supporting Information for full details). These results prove that the Mn6 molecule, though slightly affected by the functionalization, retains very interesting magnetic properties. STM and XPS Investigation. Grafting of Mn6 derivatives on the Au(111) surface was investigated by means of XPS and STM. XPS provides information on the coverage, stoichiometry, and chemical state of the s-ML distributions. XPS measurements were performed on single layers obtained after immersion of the gold surface in a solution of 1 and 2 (indicated as s-ML 1 and s-ML 2, respectively), as well as on drop-casted thick films (indicated as TF 1 and TF 2), which were used as reference standards for both stoichiometric ratio and core-level lineshapes. In Figure 5 the Mn-2p, N-1s, and O-1s core level spectra of s-MLs are compared with the corresponding TFs spectra. The core-level binding energies and lineshapes are essentially the same for both the monolayers and the thick films. This is particularly relevant in the case of Mn core level (see Figure 5
Figure 5. Core level XPS spectra of the s-MLs and the corresponding TFs. All core-level intensities have been normalized in order to facilitate the comparison. In the case of the s-MLs the Mn-2p3/2 core level line is superimposed on the Au-4p1/2 feature. In order to obtain the isolated Mn-2p core level component, the attenuated gold contribution has been subtracted.
left panel), whose line shape is quite sensitive to the oxidation state of the metal ion.29 This observation suggests therefore that the grafting process and the interaction with surface do not significantly alter the chemical state of the MnIII ions. Information on the layer stoichiometry was derived by quantitative analysis of XPS spectra and are summarized in Table 1. All core level intensities were normalized accounting for the atomic sensitivity factors and for the attenuation of the electronic signal due to the presence of an overlayer. The N-1s/ Mn-2p and O-1s/Mn-2p ratios measured for s-ML 1 and s-ML 2 are in agreement with the corresponding values obtained for the TFs and fit well with the stoichiometric values (reported in brackets in Table 1), further indicating the stability of the Mn6
Grafting Derivatives of Mn6 Single-Molecule Magnets
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9733
TABLE 1: Stoichiometric Values Derived from the Core Level Intensities for the s-MLs and TFs of the complexes 1 and 2a Mn6 complexes N1s/Mn2p [1] s-ML 1 (S)4) s-ML 2 (S)12) TF 1 (S)4) TF 2 (S)12)
1.1 ( 0.2 0.9 ( 0.2 1.1 ( 0.2 1.1 ( 0.2
O1s /6Mn 30 ( 5 [22] 33 ( 5 [24] 25 ( 5 [24] 30 ( 5 [28]
S2p/ 6Mn [2] Au4f/ S1+S2 area/S Å2 free TP coverage Au4f/ Mn2p 11 ( 2 12 ( 2 3(1 2(1
90 ( 10 100 ( 10
80-100 90-110
25-35% 20-30%
140 ( 30 170 ( 30
Mn6 coverage XPS
STM
26-38% 20-35% 20-30% 15-25%
a The expected stoichiometric values are reported in brackets. In columns 6 and 7 are reported, respectively, the average area occupied by each S atom bond to gold and the free TP ligand coverage, as estimated considering a molecular density in TP SAMs ranging from 1S-at./(22 Å2) to 1S-at./(34 Å2).34 In the last two columns the Mn6 coverages derived by XPS and STM are compared.
core in the deposition process. The observed small excess of oxygen is likely due to air exposure. The average surface density for each Mn6 cluster was derived from the Au-4f/Mn-2p ratio obtaining a value of one cluster every (8 ( 2 nm2) in s-ML 1 and one cluster every (10 ( 2 nm2) in s-ML 2, respectively. As the actual dimension of each cluster is about 2.5 nm2, these values correspond to a surface coverage of 26-38% and 20-30% respectively for both complexes. The S-2p/Mn-3p ratio reported in Table 1 provides important information about the number of S atoms present for each Mn6 cluster. While in both TFs this value only slightly exceeds the expected stoichiometric value of 2, in both s-MLs it is five or six times higher. This fact indicates that the sulfur is present in excess relative to Mn and suggests that a fraction of TP ligands cleave their bond to the Mn6 cluster core and adsorb as free molecules on the surface. Similar behavior was in fact also reported in the case of functionalized Cr7Ni clusters grafted on gold surfaces.21 More insight into the adsorption process can be obtained from a detailed analysis of the S core-level line shape. In Figure 6, the S-2p core levels of both s-MLs 1 and 2 are shown along with their best fit curves. The core-level of TF 1, reported for comparison, is quite broad and located at around 163-164 eV binding energy, which is typical of TP groups in thick films.30,31 The S-2p spectra of the s-MLs present a completely different line shape, where contributions around 161-162 eV are observed. A more detailed analysis of the S-2p MLs core levels is obtained by a fitting procedure, using a set of different spin-orbit split doublet components (Voigt functions, see figure caption for details). For both s-MLs, four components were used, located at 161.3 ( 0.1 eV (S2), 162.3 ( 0.1 (S1), 163.3 ( 0.1 (S3), and 164.3 ( 0.1 (S4), respectively. The presence of the S1 and S2 components has been reported in several papers on TP deposition on gold surfaces from solution and assigned to TP molecules chemisorbed to gold through sulfur. The exact origin of the binding energy difference between S1 and S2 is still a matter of debate. Some attribute both of them to two different adsorption geometries,30,32 while others suggest that S2 indicates the presence of alkyl-thiolate species derived from the cleavage of TP rings.33 On the other hand, attribution of S1 to chemisorbed atomic sulfur is generally excluded due to the well-known stability of sulfur inside the TP molecule. On the basis of reported literature, we attribute both S1 and S2 to different sulfur molecular species bonded to gold. The ratio between the (S1 + S2) and the 6 Mn atoms intensities is 9 ( 2, clearly showing that 80-90% of the sulfur in excess derives from chemisorbed thio-ligands which are not bonded to the Mn6 clusters. The values of the Au/(S1 + S2) ratio (reported in Table 1), correspond to a very similar surface density for both MLs (one chemisorbed S-atom every 80-110 Å2). Since typical values of molecular density in TP SAMs range from 1S-at./(22 Å2) to 1S-at./(34 Å2),34 we estimate that in our case about a
Figure 6. S-2p core level fit for the s-ML 1 and s-ML 2. Fitting parameters: spin-orbit splitting of 1.2 eV, branching ratio of 0.5, Lorentzian width of 574 meV, and Gaussian width of 780 meV for all components. The S-2p core-level of TF 1 is also reported for comparison. The TF 2 spectrum (not shown) presents a very similar line shape.
quarter of the surface is covered by a layer of free thio-ligands, forming a largely incomplete monolayer. These findings are fully confirmed by the STM observation discussed later. Concerning the S3 and S4 components, as mentioned previously, they are assigned in the literature to TP molecules which are not directly bound to the gold surface.30,31 As for both compounds, the two 3-tpc-ligands stem in opposite directions from the cluster core (see Figure 1), only one of them can bond to the surface. On the basis of results previously obtained on a similar system,21 we tentatively attribute the S4 component to sulfur atoms in 3-tpc ligands which are bound to the cluster, but not to the gold surface. This attribution is supported by the value of the S4/6Mn ratio (0.95 for the s-ML 1 and 0.85 for the s-ML 2), which is in good agreement with the value of 1 expected if the deposited clusters preserve the bond to both TP ligands. The S3 component is then tentatively assigned to 3-tpc ligands which are separated from the core of the cluster and are weakly adsorbed on top of the first adlayer.21 The S3/(S1
9734 J. Phys. Chem. B, Vol. 112, No. 32, 2008
Figure 7. STM images (70 × 70 nm2) in constant-current mode of the Au(111) surface immersed in a solution of complex 1 (s-ML 1, top panel) and complex 2 (s-ML 2, down panel). Large features, small features, and impurities are indicated with A, B, and C, respectively. Tunneling conditions: bias 2.0 V, current 30 pA.
+ S2) ratio indicates that about 10% in s-ML 1 (20% in s-ML 2) of the first layer is covered by a second layer of unbound free TP ligands. Further important information on the adsorption process and monolayer morphology has been derived by STM investigation. The topography of the Au(111) surface after dipping in a solution of 1 (top panel, s-ML 1) and of 2 (bottom panel, s-ML 2) are shown in Figure 7. In both cases the surface appears as only partially covered by a sub-ML distribution of clusters, and it is important to note that no presence of 3D aggregates was observed. In Figure 7, one can notice that the clusters fall into two main classes having very different sizes. Large and small clusters are indicated with A and B, respectively. Furthemore, we observe a third and smaller class (C) whose height is lower than 0.2 nm. We ascribe its origin to solvent impurities, since they are present also when the gold substrate is dipped in the pure solvent (CH2Cl2) alone. In order to clearly distinguish these aggregates, a high-resolution STM image of features A and B is shown in Figure 8, together with their line-profile. For both s-MLs, the large feature A has a diameter of 3.5 ( 0.5 nm in fair agreement with the value expected for a single Mn6-3tpc molecule (1.8 nm), accounting for the non-negligible curvature radius of the tip. While, for both s-MLs, the height of feature A is 1.1 ( 0.2 nm, a value slightly smaller than the geometric height of Mn6-3tpc (1.5 nm). This difference could be ascribed to the lower conductivity of the organic molecule with respect
Moro et al.
Figure 8. High-resolution STM image (15 × 15 nm2) of s-ML 1 and the corresponding 3D view (upper panel). STM line profile of large (A) and small (B) features (lower panel).
to the gold surface, as it is well known that the height of the STM profile depends not only on sample topography, but also on the electronic density of states. Hence, the measured dimensions of features A allow us to attribute them to Mn63tpc clusters grafted on the surface. A statistical analysis of the STM images where only the areas covered by the A features are taken into account, indicates that about 20-35% of the surface is occupied by a 2D distribution of Mn6 clusters for the s-ML 1, and only 15-25% in the case of s-ML 2. These results are in agreement with the values of the Mn6 coverage derived by XPS (see Table 1), thus supporting our attribution. On the other hand, the small features B observed in Figures 7 and 8 have a mean diameter and height lower than 2 and 0.6 nm, respectively, and can be ascribed, following the XPS indications, to free organic ligands detached from the cluster cores and grafted to the gold surface. The overall coverage measured with STM can be roughly estimated at around 50% in both MLs. Also these values are in agreement with the results of XPS analysis (see Table 1), thus further supporting the attribution of features B to TP ligands grafted on the surface. In summary, by combining STM and XPS measurements we are able to provide a consistent and comprehensive picture of the grafting process, showing that both s-ML 1 and s-ML 2 are formed by a dense 2D distribution of well-isolated clusters strongly grafted on the surface via the covalent bond between S and Au atoms. The grafting stability is confirmed by the fact that in STM measurements the clusters are not shifted by the
Grafting Derivatives of Mn6 Single-Molecule Magnets scanning tip, even when increasing the tunneling current up to 0.5 nA. The clusters are surrounded by free TP ligands chemisorbed on the gold, forming a SAM which partially covers the surface. On top of the first layer the presence of a relatively small amount of weakly adsorbed free ligands is revealed by XPS; though it cannot be observed in STM images presumably because weakly bonded species cannot be easily imaged by the STM tip. The different values of coverage for the two complexes may be tentatively ascribed to a different strength of the bond between the TP rings and the Mn6 core. In fact, as shown in the Figure 1, the TP rings of complex 1 are linked to the core by two oxygen bridges, whereas in complex 2 there is only one oxygen bridge suggesting a weaker bond. As a consequence complex 2 has a higher probability of losing its functional groups so compromising its grafting capability. Conclusions We have reported on two novel Mn6-based complexes (1 and 2), having S ) 4 and 12 spin ground-states, respectively, that are suitably functionalized to enable their grafting on the Au surface. After verifying that the functionalization did not deteriorate their magnetic properties, we have shown that, under optimized experimental conditions, submonolayers of 1 and 2 can be firmly grafted on Au(111) surfaces. Their packing level and grafting strength have been investigated by STM. We have found a larger surface coverage for 1 (26-38%) rather than for 2 (20-30%). Finally, by XPS we showed that the core stoichiometry of both derivatives is preserved while an excess of sulfur indicates the presence on the surface of a small amount of free organic ligands detached from the clusters in solution. Direct measurements of the magnetic properties of the grafted layers by X-ray magnetic circular dichroism, planned for the future, will enable us to characterize completely these new functionalized Mn6 derivatives. We are confident that the intelligent, targeted structural distortion of compounds similar to those presented here will permit the design of new SMMs with even higher blocking temperatures, and we strive to implement new deposition methods enabling larger control and reproducibility of surface coverage. This should open the way toward the practical exploitation of the magnetic properties of SMMs. Acknowledgment. This work was supported by the EU Network of Excellence “MAGMANet”, and the Leverhulme Trust (U.K.). Supporting Information Available: This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Khan, O. Molecular Magnetism; VHC: New York, 1993. (2) Joachim, C.; Gimzewski, J. K.; Aviram, A. Nature 2000, 408, 541. (3) Sessoli, R.; Gatteschi, D. Angew. Chem., Int. Ed. 2003, 42, 268. (4) Christou, G.; Gatteschi, D.; Hendrickson, D. N.; Sessoli, R. MRS Bull. 2000, 25, 66.
J. Phys. Chem. B, Vol. 112, No. 32, 2008 9735 (5) (a) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789. (b) Meier, F.; Levy, J.; Loss, D. Phys. ReV. Lett. 2003, 90, 047901. (6) (a) Ishikawa, N.; Sugita, M.; Tanaka, N.; Ishikawa, T.; Koshihara, S.; Kaizu, Y. Inorg. Chem. 2004, 43, 5498. (b) Ishikawa, N.; Sugita, M.; Wernsdorfer, W. J. Am. Chem. Soc. 2005, 127, 3650. (7) Sessoli, R.; Gatteschi, D.; Caneschi, A.; Novak, M. A. Nature 1993, 365, 141. (8) Lis, T. Acta Crystallogr. 1980, B36, 2042. (9) Friedman, J. R.; Sarachik, M. P.; Tejeda, J.; Ziolo, R. Phys. ReV. Lett. 1996, 76, 3830. (10) Thomas, L.; Lionti, F.; Ballou, R.; Gatteschi, D.; Sessoli, R.; Barbara, B. Nature 1996, 383, 145. (11) Milios, C. J.; Vinslava, A.; Wood, P. A.; Parsons, S.; Wernsdorfer, W.; Christou, G.; Perlepes, S. P.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 8. (12) Milios, C. J.; Vinslava, A.; Wernsdorfer, W.; Moggach, S.; Parsons, S.; Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 2754. (13) Milios, C. J.; Vinslava, A.; Whittaker, A. G.; Parsons, S.; Wernsdorfer, W.; Christou, G.; Perlepes, S. P.; Brechin, E. K. Inorg. Chem. 2006, 45, 5272. (14) Milios, J. C.; Manoli, M.; Rajaraman, G.; Mishra, A.; Budd, L. E.; White, F.; Parsons, S.; Wernsdofer, W.; Christou, G.; Brechin, E. K. Inorg. Chem. 2006, 45, 6782. (15) Milios, C. J.; Raptopoulou, C. P.; Terzis, A.; Lloret, F.; Vicente, R.; Perlepes, S. P.; Escuer, A. Angew. Chem., Int. Ed. 2003, 43, 210. (16) Gomez-Segura, J.; Veciana, J.; Ruiz Molina, D. Chem. Comm. 2007, 3699. (17) Condorelli, G. G.; Motta, A.; Fragala`, I. L.; Giannazzo, F.; Raineri, C.; Caneschi, A.; Gatteschi, D. Angew. Chem., Int. Ed. 2004, 43, 408. (18) Cavallini, M.; Gomez-Segura, J.; Albonetti, C.; Ruiz-Molina, D.; Veciana, J.; Biscarini, F. J. Phys. Chem B 2006, 110, 11607. (19) Go´mez-Segura, J.; Dı´ez-Pe´rez, I.; Ishikawa, N.; Nakano, M.; Veciana, J.; Ruiz-Molina, D. Chem. Commun. 2006, 2866. (20) Cornia, A.; Fabretti, A. C.; Pacchioni, M.; Zobbi, L.; Bonacchi, D.; Caneschi, A.; Biagi, R.; del Pennino, U.; De Renzi, V.; Gurevich, L.; Van der Zant, H. S. J. Angew. Chem., Int. Ed. 2003, 42, 1645. (21) Corradini, V.; Biagi, R.; del Pennino, U.; De Renzi, V.; Gambardella, A.; Affronte, M.; Muryn, C. A.; Timco, G. A.; Winpenny, R. E. P. Inorg. Chem. 2007, 46, 4937. (22) Coronado, E.; Forment Aliaga, A.; Romero, F. M.; Corradini, V.; Biagi, R.; De Renzi, V.; Gambardella, A.; del Pennino, U. Inorg. Chem. 2005, 44, 7693. (23) del Pennino, U.; Corradini, V.; Biagi, R.; De Renzi, V.; Moro, F.; Boukhvalov, D. W.; Panaccione, G.; Hochstrasser, M.; Carbone, C.; Milios, C. J.; Brechin, E. K. Phys. ReV. B 2008, 77, 085419. (24) Bogani, L.; Cavigli, L.; Gurioli, M.; Novak, R. L.; Mannini, M.; Caneschi, A.; Pineider, F.; Sessoli, R.; Clemente-Leo´n, M.; Coronado, E.; Cornia, A.; Gatteschi, D. AdV. Mater. 2007, 19, 3906. (25) Milios, C. J.; Inglis, R.; Vinslava, A.; Bagai, R.; Wernsdorfer, W.; Parsons, S.; Perlepes, S. P.; Christou, G.; Brechin, E. K. J. Am. Chem. Soc. 2007, 129, 12505. (26) Dunsten, W. R.; Henry, T. A. J. Chem. Soc. 1899, 75, 66. (27) Sun, Z.; Ruiz, D.; Dilley, N. R.; Soler, M.; Ribas, J.; Guzei, I. A.; Rhiengold, A. L.; Folting, K.; Maple, M. B.; Christou, G.; Hendrickson, D. N. Chem. Commun. 1999, 1973. (28) Aubin, S. M. J.; Sun, Z.; Eppley, H. J.; Rumberger, E. M.; Guzei, I. A.; Folting, K.; Gantzel, P. K.; Rheingold, A. L.; Christou, G.; Hendrickson, D. N. Inorg. Chem. 2001, 40, 2127. (29) Nelson, A. J.; Reynolds, J. G.; Roos, J. W. J. Vac. Sci. Technol., A 2000, 18, 1072. (30) Noh, J.; Ito, E.; Araki, T. Surf. Sci. 2003, 532, 1116. (31) Ishida, T.; Choi, N.; Mizutani, W.; Tokumoto, H.; Kojima, I.; Azehara, H.; Hokari, H.; Akiba, U.; Fujihira, M. Langmuir 1999, 15, 6799. (32) Noh, J.; Ito, E.; Nakajima, K.; Kim, J.; Lee, H.; Hara, M. J. Phys. Chem. B 2002, 106, 7139. (33) Sako, E. O.; Kondoh, H.; Nakai, I.; Nambu, A.; Nakamura, T.; Ohta, T. Chem. Phys. Lett. 2005, 413, 267. (34) Esplandiu`, M. J.; Hagenstrom, H.; Kolb, D. M. Langmuir 2001, 17, 828.
JP802195X
