Communications
arrows). The intensities of the cross-peaks for 7 increase
possibilities made accessible by the UF NMR methodology.
Real-time monitoring of the multistep reaction described
above revealed important data about its mechanistic and
kinetic aspects. Additional studies about new applications of
the ultrafast methodology are in progress in our laboratory.
exponentially and level off toward the end of the sequence.
The positions of the signals from the Z/E vinyl triflates 3 and
the pyrimidine 7 were confirmed through standard one-
dimensional and two-dimensional experiments (see the
Supporting Information).
We assigned the new correlations (magenta arrows) to the
nitrilium salt intermediates 4 and 5 (Scheme 1). Their
calculated chemical shifts (ACD/Labs 8.00 Release) are in
agreement with the new signals observed. The intermediate
nitrilum salts 4 and 5 appear as a mixture at an early stage of
the reaction (TOCSY-2, 0.68 min), prior to the appearance of
Experimental Section
A solution of Tf2O (29.6 mg; 150 mm) in CD3CN (0.5 mL) was
prepared and added to a 5 mm NMR tube. A fast mixing device was
devised and consisted of a long Teflon injection tube which connected
a syringe with a Luer-lock tip to the reaction mixture inside a simple
5 mm NMR tube (see diagram). The NMR tube was fitted with a cap
having a hole and a bearing to minimize oscillations of the injection
the signals from the Z/E vinyl triflates
3 (TOCSY-6,
3.55 min), confirming that the short lived species 2 is captured
faster by Tf2O than it loses a proton. Intermediate 5 easily
eliminates TfOH to afford the olefinic nitrilium salt inter-
mediate 6 (TOCSY-5, 1.86 min). Probably as a result of their
low intensity, no allylic correlations could be detected. The
presence of the double bond causes the cross-peaks from the
ethyl group of the intermediate 6 to appear at higher field
than those in intermediate 5. These cross-peaks persist until
TOCSY-11. An animation of this process was made using the
first 211 TOCSY experiments (see the Supporting Informa-
tion).
Figure 3 shows the intensities versus time of the various
species that were as part of the reaction mixture. The
intermediate characteristics of 4, 5, and 6 are shown with
their rise and fall during the course of the reaction. The t1/2
lifetimes of the starting ketone 1, intermediates 4, 5, and 6
were obtained by fitting the data points to the equation I(t)
=
I0exp(Àt/t1/2) + IÀ. Values shown in Figure 3 gave reaction
rate measurements which permit evaluation of the reactive
behavior of the intermediates. The points chosen were taken
from measurements starting at 0.0, 1.2, and 3.2 minutes
respectively.
tube. In the fully loaded position, the injection tube contained, in
order from the bottom tip upward: an air bubble (ca. 10 mL; about 1–
2 cm length), the solution to be injected [3-pentanone (6.03 mg;
100 mm) of in CD3CN (2.0 mL)], another air-bubble, about (20 mL;
about 3–4 cm in the injection tube). The upper part of the injection
tube was filled with CD3CN to efficiently propagate the pressure
throughout the injection tube. The bottom end of the injection tube
was 1–2 mm inside the solution (see figure). The plastic cap was
adjusted to maintain the injector at the correct height inside the NMR
tube. The vertical position of the NMR tube was adjusted with the
tube spinner and best results were achieved when the bottom of the
NMR tube was approximately 10 mm below the lower end of the
detection coil. Once the NMR tube with the spinner and the injection
tube was fully assembled and placed into the detection coil zone, the
bottom tip of the injection tube was well above the detection coil.
Standard NMR adjustments were carried out before starting the
TOCSY experiments. Acquisition of TOCSY experiments were
started 20 s (2 scans) before the injection of the 3-pentanone solution
and scans were recorded every 10 s. A total of 525 TOCSY scans were
recorded. The acquisition parameters were: bandwidth of chirp pulse:
60 kHz; Ge = 8 GcmÀ1; t1max/2 = 10 ms; Ga = 20 GcmÀ1; Ta = 0.246 ms;
N2 = 64; gradient switching time = 40 ms. These parameters corre-
spond to a spectral window of SW1 = 3.63 ppm and SW2 = 3.50 ppm.
A sinusoidal purge gradient of 16 GcmÀ1 during 200 ms was applied
before acquisition. The data were sampled every 1 ms. Time used for
the DIPSI sequence was 20 ms. For each experiment a suitable
shearing was carried out and data were zero filled before the T2
Fourier transformation. Spectra were represented in magnitude
Figure 3. The averaged integrated peak intensity as a function of time
^
^
^
for reactant 1 ( ), intermediates 4, 5 ( ), and 6 ( ), and final
^
^
products 3 ( ) and 7 ( ).
In conclusion, the reaction between a simple ketone,
3-pentanone, and triflic anhydride in the presence of
[D3]acetonitrile was monitored by two-dimensional
UF TOCSY using a standard 500 MHz spectrometer and a
5 mm NMR tube. The evolution of the reactants, presence of
intermediates, and generation of reaction products were
observed. These results represent another example of the
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 6274 –6277