Fatty acid capped, metal oxo clusters as the smallest conceivable nanocrystal prototypes

Metal oxo clusters of the type M6O4(OH)4(OOCR)12 (M = Zr or Hf) are valuable building blocks for materials science. Here, we synthesize a series of zirconium and hafnium oxo clusters with ligands that are typically used to stabilize oxide nanocrystals (fatty acids with long and/or branched chains). The fatty acid capped oxo clusters have a high solubility but do not crystallize, precluding traditional purification and single-crystal XRD analysis. We thus develop alternative purification strategies and we use X-ray total scattering and Pair Distribution Function (PDF) analysis as our main method to elucidate the structure of the cluster core. We identify the correct structure from a series of possible clusters (Zr3, Zr4, Zr6, Zr12, Zr10, and Zr26). Excellent refinements are only obtained when the ligands are part of the structure model. Further evidence for the cluster composition is provided by nuclear magnetic resonance (NMR), infrared spectroscopy (FTIR), thermogravimetry analysis (TGA), and mass spectrometry (MS). We find that hydrogen bonded carboxylic acid is an intrinsic part of the oxo cluster. Using our analytical tools, we elucidate the conversion from a Zr6 monomer to a Zr12 dimer (and vice versa), induced by carboxylate ligand exchange. Finally, we compare the catalytic performance of Zr12-oleate clusters with oleate capped, 5.5 nm zirconium oxide nanocrystals in the esterification of oleic acid with ethanol. The oxo clusters present a five times higher reaction rate, due to their higher surface area. Since the oxo clusters are the lower limit of downscaling oxide nanocrystals, we present them as appealing catalytic materials, and as atomically precise model systems. In addition, the lessons learned regarding PDF analysis are applicable to other areas of cluster science as well, from semiconductor and metal clusters, to polyoxometalates.

: Theoretically calculated X-ray PDF for the Zr12-acetate structure, which features bridging and chelating acetate ligands. S1 The hydrogen atoms were not included in any model since hydrogen atoms have an extremely low scattering cross section for X-rays. We used the following atomic displacement parameters (ADP) for the calculation. Zr: 0.007 A 2 , O: 0.02Å 2 , C: 0.02Å 2 . The structures used for calculations are shown.
S-3 Figure S3: PDF refinement of the Zr12-acetate PDF with the Zr12-acetate cluster model, including the hydrogen bonded ligands. The refined parameters are given in Table S1.
S-4 Figure S4: PDF refinement of the Zr12-acetate PDF with the Zr12-acetate cluster model, including an exponentially dampening sine wave contribution. The refined parameters are given in Table S1.
S-5 Figure S5: PDF refinement of the Zr12-acetate PDF with various cluster structures reported in literature (see main text). For each of the structural models, we removed the excess carbon atoms to arrive at a model with acetate ligands. The refined parameters are given in Table  S2. S-6 Figure S6: PDF refinement for Zr12-butanoate (using the Zr6-acetate structural model), Zr6-methylbutanoate (using the Zr12-acetate structural model), Zr12-octanoate (using the Zr6-acetate structural model), and Zr12-oleate (using the Zr6-acetate structural model). The contribution of the exponentially dampening sine wave is shown (orange dotted lines). The refined parameters are given in Table S4. S-7 Figure S7: PDF fit for Zr12-acetate/oleate cluster with Zr12-acetate model without/without the exponentially dampening sinusoidal contribution. The refined parameters are given in Table S5. Table S1: Refined parameters after fitting our synthesized Zr12-acetate cluster with various models, see Figure 2, Figure S3 and S4. All the models were derived from the crystal structure of Zr12-acetate, S1 and in all models, the hydrogen atoms were removed. Relative Amplitude = Amplitude / Scale. S-8 Table S2: Refined parameters after fitting our synthesized Zr12-acetate cluster with various models, see Figure S5. In all models, some carbon atoms were removed in order to form a structure equivalent to an acetate capped cluster and all hydrogen atoms were removed.    Table S5: Refined parameters after fitting Zr12-oleate cluster with Zr12-propionate model without the exponentially dampening sinusoidal contribution.
3.32 Rw 0.33 3 The organic ligand shell Figure S8: IR of the all bottom up synthesized clusters.
S-11 Figure S9: Zoom of the Zr12-acetate cluster crystal structure to display the H-bonded acids coordinated to the cluster. S1 Note that some ligands are removed for clarity. Figure S10: 1 H-NMR spectrum of the Zr12-acetate cluster. There is no observable signal at 4 ppm indicating that the amount of ester impurity is very low.
In an effort to remove all the hydrogen bonded (protonated) acid from a Zr12-oleate cluster a non-nucleophilic base (1,.0]undec-7-ene or DBU) was added. The base can accept a proton from the hydrogen bonded oleic acid and thus detach it from the cluster.
Afterwards, size exclusion chromatography (SEC) was performed to separate the clusters from the DBU coordinated acid. Only the first fraction of the SEC was obtained as a pure S-12 compound (absence of the peak around 1700 cm -1 ) with a very low yield of 0.8%. From the second fraction onwards, H-bonded acid can be seen in the spectra, see figure S11).

Figure S11: Fractions collected with size exclusion chromatography
To quantify the amount of hydrogen bonded acid, we measured TGA of our synthesized clusters, see Figure S12. Assuming a pure cluster, the theoretical mass loss can be calculated from the molecular S-13 formula, considering that the end product is zirconia.
Starting from 100 g of clusters, we can calculate the mass of zirconia at the end: Where M cluster and M ZrO 2 are the molecular weights of the cluster and zirconia, respectively.
Note that the calculations are done with the monomeric species and we use the molecular weight of the monomer (which is exactly half of the dimer) for all our calculations. This value is reported in Table S7 as the theoretical value. The experimental value is consistently lower than the theoretical one (Table S7), indicating an extra organic fraction that is assigned to mostly hydrogen bonded ligands. We quantify its amount by the following procedure. Again assuming that we start from 100 g of clusters, we calculated the molar amount of zirconia in the residual mass (experimental value).
We determine the molar amount of monomeric cluster that this corresponds to: We calculate the apparent molecular weight of the cluster by using the molar amount and the starting mass (100 g): The difference with the theoretical molecular weight is calculated and assigned to the extra S-14 organic fraction: By dividing ∆M by the molecular weight of the carboxylic acid, we get the number of carboxylic acids that is present per monomer.
S-17 Figure S16: 1 H NMR of the Zr12-acetate cluster in CD 3 OD. All signals disappear after multiple cycles of dissolving/evaporation indicating degradation of the cluster. Figure S17: HRMS of the Zr12-acetate cluster in MeOH, no signals are close to the target mass (2775.17 g/mol) or could be matched with degradation products. Figure S18: HRMS of the bottom up synthesized Zr12-propionate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S19: HRMS of the bottom up synthesized Zr12-butanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S20: HRMS of the bottom up synthesized Zr6-methylbutanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S21: HRMS of the bottom up synthesized Zr12-hexanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S22: HRMS of the bottom up synthesized Zr12-octanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S23: HRMS of the bottom up synthesized Zr6-methylheptanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S24: HRMS of the bottom up synthesized Zr12-decanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated). Figure S25: HRMS of the bottom up synthesized Zr12-dodecanoate cluster in THF, the grey spectrum (experimental) is compared with the blue spectrum (simulated).

S-22
6 Clusters synthesized by ligand exchange Figure S26: Overlay C-NMR for the Zr12-hexanoate, -oleate and Zr6-methylheptanoate synthesized via exchange reaction. Zr12-acetate is added as a reference to prove that the signals of acetate are gone after exchange.
S-23 Figure S27: PDF spectra of the clusters synthesized via the exchange reaction. Note that the acetic acid spectrum is added as a reference spectrum as it is not synthesized via an exchange reaction.
S-24 Figure S28: IR spectra of the clusters synthesized via the exchange reaction.    S-28   256-11.956 11.256-11.956 S-29 Figure S36: Thermogravimetric analysis of our bottom up synthesized hafnium clusters. 8 Zirconium oxide nanocrystals Figure S37: 1 H NMR spectrum of ZrO 2 nanocrystals capped with oleic acid. Figure S38: TEM image of ZrO 2 nanocrystals capped with oleic acid, with the histogram as inset. S-31