Scheme 1. Reactions of Cumyl Radicals
In an alternate mechanism, one could propose electron
transfer from AuNP to the peroxide to give RO + RO (R
cumyl); if this was the case the maximum yield of
acetophenone (reaction 2) would be 50%. However, in the
absence of hydrogen donors (e.g., methanol, see below) the
acetophenone:cumyl alcohol ratio is as high as 97:3, ruling
out electron transfer as the dominant mechanism.
Figure 3
mM of dicumyl peroxide in H
.
Graph of conversion vs number of shots per drop for 1
O with 5% of MeOH irradiated with
•
-
2
)
a 532 nm laser with different number of shots per drop. The power
of the laser used is 50 mJ/shot. In all cases the black bar is the
control experiment with no AuNP. The right panel in the graph
shows the results using a microwave (300 W, 20 min) and a 530
nm set of LED (4 × 15 W, 35 °C, 60 min).
The change of color noted during excitation persists for
milliseconds, but its blue tint is not present after the
experiment. These red shifts are usually attributed to changes
in surface molecules and/or aggregation.
of DCP under microwave conditions, bulk heating of the
drops cannot explain the chemistry observed.
To verify if the effect observed was a true photothermal
effect, we also exposed the solutions of DCP with and
without AuNP to microwave (MW) irradiation. No reac-
tion was observed after 20 min at 80 °C with a MW power
of 300 W with no AuNP and 47% conversion with AuNP.
This confirms that we are not simply observing bulk
heating of the solution; gold is also a good microwave
absorber and prolonged heating under these conditions
does lead to some conversion. Note that in the case of
the laser excitation, the conversion after 25 s (25 equal
laser shots) exceeds that obtained after 20 min of
microwave irradiation.
As another method of photoexcitation, we repeated the
experiment using a custom-designed LED irradiator consist-
ing of four LedEngin 10 W LZ4-40G110 emitters (λexc 530
nm, see the SI). The data obtained with this method are
compared with MW in Figure 3. Despite the lower LED
intensity compared with the laser, we still obtain 21%
conversion after 1 h of irradiation with AuNP. LED sources
may offer an inexpensive alternative to lasers and micro-
waves for organic reactions.
energy (J)
C (J/g K) × mass (g)
∆
T (K) )
(4)
p
0.05J
∆
T (K) )
= 1.7 K
4.2 (J/g K) × 0.007 mL × 1.0 g/mM
(
5)
In another extreme case, we assumed that during the 8 ns
laser pulse the AuNP themselves behave as adiabatic units
that in this time do not exchange heat with surrounding
molecules; this treatment resembles an ideal gas of nano-
particles. A typical 7 µL drop may contain 10 g of gold
(C ) 0.129 J/g K). Thus, a similar calculation as eq 4 leads
to ∆T >10 K, clearly not realistic. Even taking into account
incomplete light absorption (absorbance ∼0.05 in the drop
diameter, but enhanced by internal reflections), ∆T values
-
7
p
6
6
are in the 10 K range. The melting point of nanoparticles is
1
1
lower than that of the bulk material. For bulk gold its
melting point is 1337K, while for the particles used here it
1
1
is probably around 1200K. Laser excitation of AuNP can
lead to surface changes, suggesting that temperatures near
the AuNP melting point are achievable. Combining these
Reaction 1 is clearly possible, yet inefficient under SPB-
mediated photolysis, and may thus provide an idea of the
limitations of this methodology. A particularly interesting
aspect relates to the actual temperatures that can be achieved
on the surface of the AuNP.
(
5) Bakhtiari, A. B. S.; Hsiao, D.; Jin, G.; Gates, B. D.; Branda, N. R.
Angew. Chem., Int. Ed. 2009, 48, 4166–4169.
6) Adleman, J. R.; Boyd, D. A.; Goodwin, D. G.; Psaltis, D. Nano
Lett. 2009, 9, 4417–4423.
(
p
The calculation of eqs 4 and 5, where C corresponds to
(
7) Cherukuri, P.; Glazer, E. S.; Curley, S. A. AdV. Drug DeliVery ReV.
the heat capacity of water and ∼1.0 g/mL is its density, leads
2
010, 62, 339–345.
to a bulk heating of 1.7 K for a 50 mJ laser pulse (2.2 ×
(8) Banks, J. T.; Scaiano, J. C. J. Am. Chem. Soc. 1993, 115, 6409–
-7
6413.
1
0
einstein) and a 7 µL drop, even if the light was totally
(
9) Marin, M. L.; McGilvray, K. L.; Scaiano, J. C. J. Am. Chem. Soc.
absorbed (vide infra). The calculation assumes an adiabatic
drop, something that is probably a good approximation for
a single laser pulse, but not for the experiment lasting minutes
2
008, 130, 16572–16584.
(10) McGilvray, K. L.; Decan, M. R.; Wang, D.; Scaiano, J. C. J. Am.
Chem. Soc. 2006, 128, 15980–15981.
11) Dick, K.; Dhanasekaran, T.; Zhang, Z.; Meisel, D. J. Am.Chem.
Soc. 2002, 124, 2312–2317.
(
(at 1 Hz, as many seconds as laser pulses). Given the stability
206
Org. Lett., Vol. 13, No. 2, 2011