60th Annual Report on Research 2015

Figure   1. ESI-MS(−) spectrum of the phthalocyanine derived from pyromellitic dianhydride, showing peaks assignable to CuPc, and not to (CuPc*)₄.

Next, we focused our attention on CuPc, and prepared MOF building blocks by installing various donor groups (4-carboxyphenyl, 3,5-dicarboxyphenyl, 4-pyridyl, 4-pyrazolyl) on the four sides of the phthalocyanine core (Figure 2). Similarly, we attached various chromophores (anthracene, pyrene, 4-(trifluoromethyl)coumarin) onto the phthalocyanine core, to prepare new materials with possible applications in organic solar-cells. Figure 2. Synthesis of the functionalized phthalocyanine compounds.

The chromophore functionalized materials were studied by femtosecond fluorescent measurements to understand the interaction of the chromophores with the phtahlocyanine core, as well as inter-Pc interactions. Also, two-photon absorption measurements were carried out to study their non-linear optical properties. All new materials were characterized by electrospray ionization mass spectrometry (ESI-MS), UV-vis and fluorescence spectroscopy (examples shown in Figures 3–5).  ESI-MS confirms that the products are fully substituted on all four sides. To prepare MOFs based on the phthalocyanine building-blocks shown in Figure 2 (including the parent CuPc octacarboxylic acid, as well as the 4-carboxyphenyl, 3,5-dicarboxyphenyl, 4-pyridyl and 4-pyrazolyl substituted tetrakis(phthalimide) derivatives), a large variety of crystal growing set-ups were employed, including solvothermal synthesis using a variety of metal nodes (Fe, Ni, Cu, Zn, Pb). Only amorphous materials and no single crystals suitable for X-ray diffraction experiments could be obtained to date. These results suggest that stacking interactions between the phthalocyanine cores (aggregation) likely dominate over the coordination bonds formed between the terminal donor groups and the metal ions, favoring an amorphous coordination polymer over a well-ordered, crystalline metal-organic framework.

Figure 3. Electrospray ionization mass spectrum of tetrakis-(N-(4-pyridyl)phthalimide) copper phthalocyanine, showing the molecular peak (M) as well as fragments resulting from the successive loss of the terminal groups (X).

Figure 4. Absorption features suggest aggregation of coumarin-decorated copper phthalocyanine. Fluorescence quenching is assigned to the aggregation-induced energy transfer between chromophores, as observed from ultrafast fluorescence and anisotropy decay.

Figure 5. Absorption features suggest aggregation of anthracene-decorated copper phthalocyanine. Fluorescence decay suggests efficient charge transfer, as the anisotropy does not decay faster.

The catalytic activity of metal phthalocyanine octacarboxylic acid (MPc, M = Cu, Fe, Zn) for selectively oxidizing organic sulfides to sulfoxides has been investigated. The eight carboxylic acid groups on the periphery impart solubility in polar solvents, such as water, methanol, dimethylsulfoxide and dimethylformamide, offering an advantage over simple copper-phthalocyanine, which is insoluble in all organic solvents. We have monitored the catalytic oxidation of various sulfide substrates (R = Me, Et, ⁱPr, Bu, Ph) by aqueous H₂O₂ in CD₃OD using Nuclear Magnetic Resonance (NMR), and gas chromatography using mass spectrometric detection (GC-MS). Examples are shown in Figures 6–7, which show a much improved conversion of dibutylsulfide to dibutylsulfoxide, compared to the uncatalyzed reaction. Of the different MPc catalysts used, FePc and CuPc prove to be the most efficient, while ZnPc is the least efficient, as indicated by the different rate constants (k) shown in Figure 8. Figure 9 shows that increasing amount of oxidizing agent (from 20% to 300% excess) reduces the time required for reaction completion significantly, e.g. from ~5 hours to ~ 2 hours in the case of di-n-butylsulfide. Furthermore, we have also shown that the soluble CuPc catalyst is selective for the oxidation of sulfides to sulfoxides, as no overoxidation product (sulfone) was obtained. Even after 14 hours of reaction time for di-n-butylsulfoxide with a 300% excess of H₂O₂, only 0.7% di-n-butylsulfone could be detected (Figure 10). Figure 6. GC-MS analysis of the reaction mixture obtained by the uncatalyzed (bottom) and 0.5 mol% CuPc-catalyzed (top) oxidation of di-n-butylsulfide (at 3.25 min) to di-n-butylsulfoxide (at 5.41 min) by hydrogen peroxide after 2 hours. No di-n-butylsulfone byproduct is detected in either case.

Figure 7. Oxidation of di-n-butyl sulfide with H₂O₂ monitored by ¹H NMR spectroscopy in CD₃OD. The catalyzed reaction (0.5 mol% CuPc) is complete in ~ 5 hours, while the uncatalyzed reaction is still incomplete (90%) after 13 hours. No di-n-butylsulfone byproduct is detected in either case.

Figure 8. Rate constants determined for the oxidation of various sulfides with H₂O₂ by ¹H NMR spectroscopy in CD₃OD, in the presence of 0.5 mol% MPc.

Figure 9. Oxidation of di-n-butyl sulfide with H₂O₂ (300 % excess) monitored by ¹H NMR spectroscopy in CD₃OD, catalyzed by 0.5 mol% CuPc.

Figure 10. Oxidation of di-n-butylsulfoxide with H₂O₂ (300 % excess), catalyzed by 0.5 mol% CuPc, shows only 0.7% di-n-butylsulfone after 14 hours.

            This ACS-PRF grant supported and greatly benefited a full-time graduate research assistant, who learned new synthesis and purification techniques, as well as instrumentation methods, including NMR, UV-vis and fluorescence spectroscopy, mass spectrometry and GC-MS analysis, and kinetics. He already presented his results at various research symposia and conferences, and is preparing two manuscripts for publication. Working on this project has also impacted my own career: for the first time in my research group, we studied kinetics of catalyzed reactions, and dealt with the challenges of preparing metal-organic frameworks. These new avenues of research introduced me to previously unexploited techniques, and helped in obtaining a new grant from NSF.