Chemistry Reference
In-Depth Information
42
Figure 16.22 Compound 42 was produced by a Suzuki-Miyaura-type reaction with-
out the addition of transition metals.
Adapted from Ref. 71.
16.4.2.3.5 Applications. Whether ''palladium-free'' reactions that nor-
mally require palladium are truly devoid of palladium remains contro-
versial.
68,69
A well-known case was a Suzuki-Miyaura coupling that proved to
be catalyzed by trace amounts of palladium present in Na
2
CO
3
.
70
This re-
port by Leadbeater and Marco prompted others to analyze cross-coupling
reagents more rigorously before claiming ''palladium-free''. Inamoto's group
reported a cross-coupling between 4-iodobenzonitrile and phenylboronic
acid in the presence of K
3
PO
4
in dimethyl carbonate.
71
Without the addition
of palladium reagents, the cross-coupling proceeded to form the biphenyl
product 42 (Figure 16.22). The Inamoto-Koide team analyzed the crude mix-
ture of the reaction, treated it with nitric acid and subjected the resulting
sample to the deallylation-based fluorimetric method to detect palladium. It
was concluded that 0.78 ppb of palladium was present in the original re-
action solution, which was in reasonable agreement with an independent
ICP-MS analysis of a closely related sample with 0.42 ppb of palladium.
72
As the previous cases illustrate,
70,72,73
commercial reagents may contain
trace amounts of palladium. Koide's group tested chemicals that might be
relevant to cross-coupling reactions and found that particular batches of CuI
and Cs
2
CO
3
contained palladium. In some instances, the fluorimetric
method could be directly applied to solid samples. For example, round-
bottomed flasks in the Koide laboratory were subjected to the fluorimetric
method and some were found to be contaminated with palladium.
63
Williams and Koide were also able to quantify palladium in ore samples
without metal extraction with acids.
74
Although the APE method was suciently robust and accurate in the
Koide laboratory, a Merck team wished to make it more accurate and broadly
applicable with their API samples. Pretreatment of samples with aqua regia
proved to be beneficial for accuracy, although HCl suced in many cases.
67
The deallylation can proceed without NaBH
4
but its use improved the per-
centage signal recovery with complex samples produced at Merck.
67
By in-
creasing the APE concentration and shortening the time for deallylation to
10-15 min, the linear range was extended to 1-30 ppb.
67
Therefore, by
measuring fluorescence signals first at 15 and then at 60 min, 0.1-30 ppb of
palladium can be quantified, meeting criterion 7 above. Importantly, the
APE-based fluorogenic method was rigorously compared with ICP-OES with
12 API samples and proved to be accurate.
67
Following these studies, the APE method now meets criteria 1-6
above, with comments warranted for criterion 5. An ICP-MS instrument
costs $100 000-250 000 and requires frequent maintenance and repair;
Search WWH ::
Custom Search