Communication
C
tion product ArÀH. The resultant radical DMF regenerates PPO
effected by the metal-free dyes BD (sensitizer) and PPO (triplet
by an exergonic back-electron transfer to give highly electro-
philic DMF+, which hydrolyzes upon work-up to volatile prod-
ucts.[16]
annihilator). Transient spectroscopy and quenching studies
1
support the formation of PPO* as high-energy intermediate.
1
The single-electron transfer from PPO* to the aryl bromides is
To explain the differences in reactivity between aryl bro-
mides 1–3, we have also investigated kinetic aspects of the re-
action mechanism. It is well established that the fragmentation
rate limiting and in combination with the corresponding back-
electron transfer determines the overall efficiency of the reac-
tion, as was evidenced by DFT calculations. Further optimiza-
tion of the photophysical and chemical steps by employing
longer lived excited states, lower light power, and more effec-
tive quenchers are currently being investigated.[18] The general
concept of visible-to-UV TTA and other photon-upconversion
processes holds great potential for challenging bond activa-
tions while retaining the benefit of mild reaction conditions by
the use of lower-energy visible light.
CÀ
of aryl bromide radical anions (ArBr ) to the respective aryl
À
C
radicals Ar and Br proceeds rapidly under the experimental
conditions.[9a] Further, we assume that the low activation ener-
C
gies of HAT from DMF to Ar and the large excess amounts of
DMF as solvent make this step very rapid.[16] This suggests that
the SET from 1PPO* to the aryl bromides is most likely the
rate-determining step. Therefore, we used the Marcus theory
to determine the activation barriers of the SET steps DG°
,
SET
which can be obtained from the free energy of the reaction
DG0 and the nuclear reorganization energy l [Eq. (6) and Experimental Section
SET
Table 3].[17]
Chemicals
2
ðDG0SET þ lÞ
Biacetyl (2,3-butanedione), 2,5-diphenyloxazole, 4-bromoacetophe-
¼
ð6Þ
DGSET
¼
none, 4-bromobenzotrifluoride, 4-bromoanisole, acetophenone, tri-
fluorotoluene, and anisole were commercially available. DMF, extra
dry over molecular sieves, acetonitrile (gradient grade), and metha-
nol (for analysis) were used as solvents without further purification.
4l
Table 3. Calculated data for the SET from 1PPO* to the quenchers 1–3.
[c]
ET
[d]
DG°
BET
[e]
S[a]
DGET (calcd)[b]
DG°
DG°
GSÀBET
General procedure for the steady-state irradiations
1
2
3
À7
+9
+22
0.1
9.5
29.9
6.8
0.4
7.8
119
104
272
A solution of aryl bromide (10 mm) with 2,3-butanedione (40 mm)
and 2,5-diphenyloxazole (13 mm) in dry DMF (1 mL) containing n-
pentadecane (10 mm) as GC internal standard was placed in
a 1.5 mL transparent vial with a 8 mm screw cap with butyl/PTFE
septum. Then, the solution was irradiated under continuous N2
purging for 30 min by a pulsed Nd:YAG laser at 430 nm (see the
Supporting Information). Then, ethyl acetate (3 mL) was added,
and the mixture was washed with brine (5 mL). The organic phase
was separated and filtered through MgSO4 for further analysis. The
course of the reaction was followed by quantitative GC-FID analysis
on a 7820 A Agilent versus internal n-pentadecane. Control experi-
ments showed that photoreduction of aryl bromides 1–3 did not
proceed in the dark or in the absence of PPO.
1
[a] Substrate (=quencher); [b] free energy of SET from PPO* (S1) to S cal-
culated by DFT in kcalmolÀ1; [c] activation energy of SET from 1PPO* (S1)
CÀ
to S calculated by DFT in kcalmolÀ1; [d] activation energy of SET from S
to PPO to form PPO* (S1), calculated by DFT in kcalmolÀ1; [e] activation
+
1
C
+
CÀ
C
energy of SET from S to PPO to form ground-state PPO (S0), calculat-
ed by DFT in kcalmolÀ1
.
The free energies of the SET process derived from DFT calcu-
lations DGET (calcd) were in very good agreement with the ex-
perimental values obtained from CV measurements (Table 2).
The calculated energy barrier for the SET reduction of 1 by
1PPO* is negligible, and the energy barrier for the back elec-
tron transfer (BET) is 6.8 kcalmolÀ1. This translates to the high-
est reactivity of 1 with a third of all molecules engaging in
Gas chromatography (GC)
GC was calibrated by using a four-point calibration versus 10 mm
of the internal standard n-pentadecane (Std.). GC oven tempera-
ture program: initial temperature 508C was kept for 0.5 min, the
temperature was increased at a rate of 258CminÀ1 over a period of
9.7 min, until the final temperature (2808C) was reached and kept
for 0.3 min.
1
a quenching process (Q=33) with PPO* (Table 2, 4th column,
1st row). The energy barrier for the SET reduction of 2 by
1PPO* is still relatively low, but the activation energy for the
BET is below 1 kcalmolÀ1, which results in much lower quench-
ing efficiency and low overall reactivity of 2. For p-methoxy de-
rivative 3, the SET activation energy is so high that other deac-
tivation mechanism of 1PPO* become dominant (Q=1,
UV/Vis and fluorescence spectroscopy
UV/Vis analyses were performed on a Varian Cary 50 UV/Vis spec-
trophotometer. Steady-state fluorescence measurements were per-
formed on a Horiba FluoroMax-4 fluorimeter. Excitation and emis-
sion slit widths were 1 nm. Hellma quartz SUPRASIL cuvettes (10
10 mm; 117.100F-QS) with a screw cap with PTFE-coated silicon
septum were used. The PPO concentration was 0.1 mm in both
methanol and DMF.
Table 2). Interestingly, even though the BET from all intermedi-
+
CÀ
C
ate S radical anions to PPO to give PPO (S0) are exergonic,
the activation barriers for this process are prohibitively high
(Marcus inverted region).
In summary, we have demonstrated a new combination of
visible-to-UV photon upconversion and an SET-initiated reduc-
tive activation of aryl bromides. The underlying TTA process is
The fluorescence quantum yield in DMF was measured with refer-
ence to 2,5-diphenyloxazole (FF0 =0.5 in methanol)[19] by compar-
Chem. Eur. J. 2015, 21, 15496 – 15501
15499
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim