• No results found

Catalytic C–N coupling enabled by complex 3.1 and light

Chapter 3. Copper complexes supported by tridentate bis(phosphino)carbazole ligands

3.2.3. Catalytic C–N coupling enabled by complex 3.1 and light

The viability to transform the decomposition product of the one-electron oxidation of 3.1 (e.g. 3.8) back to 3.1 is advantageous in the context of developing catalytic transformations using 3.1. Specifically, we envisioned that upon photoexcitation, the excited- state complex (3.1*) can deliver an electron to the alkyl bromide substrate, releasing an alkyl

91 radical and the oxidized copper complex (3.7; Figure 3.7). Complex 3.7 may then be trapped by an external nucleophile (Nu) in solution, or participate in an electron transfer event with copper(I)–nucleophile species ([LnCuINu]) to enter a productive bond-forming cycle.14 Albeit at a slow rate, 3.7 may decompose in solution to 3.8 as observed in the attempted isolation, but 3.8 can re-enter the productive cycle in the presence of an external base to form

3.1.

Figure 3.7. Potential photocatalytic cycle (e.g. left half of the full cycle shown in Figure 3.1) that involves complex 3.1.

Gratifyingly, we found that 3.1 can be used as an effective catalyst for the photoinduced copper-catalyzed couplings of carboxamide-type nucleophiles with unactivated alkyl halides. In particular, the coupling of t-butyl carbamate with an unactivated

92 secondary bromide using blue LED lamps as the light source can be achieved in the presence of 3.1 although the yield of the observed C–N coupling is modest (33%, Scheme 3.5). Similar reactivity is observed when the same reaction is performed using ligand L3.1 and CuBr (34%). In the absence of light, no reaction is observed, and the alkyl bromide can be recovered quantitatively.

Scheme 3.5. Catalytic C–N coupling of t-butyl carbamate with 2-bromo-4-phenylbutane. Ar = 4-t-Bu-phenyl.

To further confirm the importance of the design principles embodied in L3.1, related tridentate ligands featuring a carbazole backbone have also been prepared (Figure 3.8). When ligand SNS-Ar (L3.7) or NNN-iPr (L3.8) is employed instead of L3.1 under related conditions, carbamate alkylation does not proceed. The use of 3,6-di-t-butyl-carbazole and tri-t-butylphosphine in place of L3.1 also fails to yield substantial amount of the C–N coupled product (<5 % yield).

93

Figure 3.8. Ligand structures and combinations that are ineffective for the coupling of t- butyl carbamate with 2-bromo-4-phenylbutane under conditions related to that shown in

Scheme 3.5.

The observed catalytic C–N coupling is also unique to copper. If CuBr is replaced by other transition metals commonly utilized in cross-couplings, no photocatalysis is observed. For example, even at 10 mol% loadings, carbamate alkylation is not observed using CoBr2, NiCl2(glyme), PdCl2(cod), or Au(PPh3)Cl in place of CuBr. Thus, the original design principles outlined in Figure 3.2 remain appropriate for the development of photocatalysts for the C–N couplings of alkyl bromides.

3.3. Conclusions

Inspired by the mechanistic conclusions made in Chapter 2, we designed a new photoredox catalyst 3.1, based on copper and a tridentate bis(phosphino)carbazole ligand

L3.1, that can serve as an electron donor upon irradiation by blue LED or Hg lamps. We demonstrated that the new photocatalyst features a highly-reducing excited state that can elicit a mesolytic cleavage of an unactivated secondary alkyl bromide. Upon one-electron oxidation, 3.1 forms a metastable radical 3.7 which thermally decomposes to the diamagnetic

94 cation 3.8 via a formal hydrogen atom transfer. The decomposition product 3.8 can be recycled to give the starting material 3.1 under cross-coupling conditions. We also articulated that 3.1 can catalyze the C–N coupling reaction between t-butyl carbamate and 2-bromo-4- phenylbutane. The design embodied in complex 3.1 is crucial in eliciting the observed photochemical transformation, as copper, bis(phosphino)carbazole, and light are all required for the formation of C–N bonds. The methodological development of this carbamate alkylation using complex 3.1 or ligand L3.1 is discussed in detail in Chapter 4.

95

3.4. Experimental section

3.4.1. General information

Unless otherwise noted, materials were either purchased from commercial suppliers and used as received, or prepared via literature procedures. L3.1 was synthesized according to a literature procedure15 and recrystallized from a cold, saturated solution in Et2O/CH3CN for use in photoinduced cross-couplings. L3.216 and L3.817 were prepared following reported procedures. Solvents were deoxygenated and dried by thoroughly sparging with argon, followed by passage through an activated column in a solvent purification system.

All manipulations of air‐sensitive materials were carried out in oven-dried glassware using standard Schlenk, or glovebox techniques, under an N2 atmosphere. Silicycle

SiliaFlash® P60 silica gel (particle size 40–63 μm) was used for flash chromatography.

Analytical thin layer chromatography was conducted with glass TLC plates (silica gel 60 F254), and spots were visualized under UV light or after treatment with standard TLC stains.

X-band EPR measurements were made with a Bruker EMX spectrometer at 77 K. Simulation of EPR data was conducted using the software EasySpin.18

IR measurements were recorded on a Bruker ALPHA Diamond ATR or using a Perkin Elmer Paragon 1000 spectrometer using thin films deposited on KBr plates.

1H, 13C, and 31P NMR spectra were recorded on a Bruker Ascend 400 MHz, a Varian

300 MHz, a Varian 400 MHz, a Varian 500 MHz, or a Varian 600 MHz spectrometer with CHCl3 (1H, δ = 7.26) and CDCl3 (13C, δ = 77.0) as internal references and with 85% H3PO4

96 (31P) as an external reference. Multiplicity and qualifier abbreviations are as follows: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad, app = apparent). GC analyses were carried out on an Agilent 6890 Series system with an HP-5 column (length 30 m, I.D. 0.25 mm).

X-ray crystallography studies were carried out at the Beckman Institute Crystallography Facility either Bruker Kappa Apex II diffractometer or a Bruker D8 Venture kappa duo photon 100 CMOS instrument (Mo Kα radiation). Structures were solved using SHELXT and refined against F2 by full-matrix least squares with SHELXL and OLEX2. Hydrogen atoms were added at calculated positions and refined using a riding model. The crystals were mounted on a glass fiber or a nylon loop with Paratone N oil.

Steady-state fluorimetry was performed in the Beckman Institute Laser Resource Center (BILRC; California Institute of Technology). Steady-state emission spectra were collected on a Jobin S4 Yvon Spec Fluorolog-3-11 with a Hamamatsu R928P photomultiplier tube detector with photon counting. Absorbance spectra were acquired on a Cary 50 UV-Vis spectrophotometer with a Unisoku Scientific Instruments cryostat to maintain temperature.

Measurements of the excited-state lifetimes were conducted in BILRC. A Q- switched Nd:YAG laser (SpectraPhysics Quanta-Ray PRO-Series; 355 nm) was used as the source of the excitation pulse Transmitted light from the sample was detected with a photomultiplier tube (Hamamatsu R928). All instruments and electronics in these systems were controlled by software written in LabVIEW (National Instruments).

97 Electrochemical measurements were carried out in a thick-walled one-component electrochemical cell fitted with a Teflon stopcock and tungsten leads protruding from the top of the apparatus in a nitrogen-filled glovebox. A CH Instruments 600B electrochemical analyzer or a CHS-3600B potentiostat was used for data collection. A freshly-polished glassy carbon electrode was used as the working electrode, and platinum wire was used as the auxiliary electrode. Solutions (THF) of electrolyte (0.1 M tetra-n-butylammonium hexafluorophosphate) contained ferrocene (~1 mM, post measurement), to serve as an internal reference, and analyte (~1 mM).

Mass spectral data were collected on a Thermo LCQ or LTQ ion trap mass spectrometer, or on an Agilent 5973 mass spectrometer.

Photolytic reactions were performed using 34 W Kessil H150 Blue LED lamps, a 100 W Blak-Ray Long Wave Ultraviolet Lamp (Hg), or a 100-W Blak-Ray B-100Y High Intensity Inspection Lamp (Hg). If required, the temperature was maintained with an isopropanol bath cooled by an SP Scientific cryostat.

98

Related documents