Oxidation state-specific fluorescent method for palladium(II) and platinum(IV) based on the catalyzed aromatic claisen rearrangement

Article abstract of DOI:10.1021/ja8065539

Contamination of active pharmaceutical ingredients and healthcare products with oxidized forms of palladium and platinum are important problems. We report a convenient fluorogenic probe based on the catalyzed aromatic Claisen rearrangement capable of dete

Full text of DOI:10.1021/ja8065539

Published on Web 11/14/2008
Oxidation State-Specific Fluorescent Method for Palladium(II) and
Platinum(IV) Based on the Catalyzed Aromatic Claisen Rearrangement
Amanda L. Garner and Kazunori Koide*
Department of Chemistry, UniVersity of Pittsburgh, 219 Parkman AVenue, Pittsburgh, PennsylVania 15260
Received August 18, 2008; E-mail: koide@pitt.edu
Pd contamination in active pharmaceutical ingredients (APIs)
and their synthetic intermediates is a severe problem in the
pharmaceutical industry, because substantial efforts are needed to
analyze and remove the residual metal.1 Since Pd0 and PdII bind to
scavengers differently, it is informative to separately quantify these
Pd species in synthetic samples to effectively remove these
impurities.2 Currently available analytical methods are not capable
of quantifying residual Pd species in APIs in native oxidation
states;3 therefore, it is difficult, if not impossible, to troubleshoot
the nonreproducible Pd scavenging without addressing oxidation
states during API purifications. Thus, the development of methods
to monitor Pd in an oxidation state-specific manner without altering
oxidation states is warranted. Moreover, such methods would
facilitate studies on Pd materials including colloid and polymer-
bound catalysts.4
We previously demonstrated that the transformation of nonfluo-
Figure 1. Fluorescence analysis after 4 h at 50 °C. [1] ) 12.5 µM. The
y-axis is fluorescence intensity (au × 105) at 535 nm. For example, “2”
rescent compound 1 to green fluorescent compound 2 is highly
means 2 × 105. (a) Metal specificity. [metal] ) 10 µM. (b) Pd species at
intensity and [PdII]. [ ) in buffer ([K+] ) 114 mM). 9 ) in buffer ([K+]
) 11.4 mM).
specific for Pd and Pt and capable of sensitively detecting these
metals by means of Tsuji-Trost type reactions.5 We also noted
various oxidation states. [Pd] ) 10 µM. (c) Correlation between fluorescence
the noncatalyzed Claisen rearrangement from 1 to fluorescent green-
yellow compound 3 at 150 °C in organic solvents6 and 100 °C in
Scheme 1. Pd/Pt Species-Dependent Deallylation or Claisen
Rearrangement of 1
water (unpublished results). Since 3 is nearly as fluorescent as 2
and distinct spectral differences exist between them (2, λmax ) 523
nm; 3, λmax ) 535 nm), we asked if this transformation could be
catalyzed by metals in water at a lower temperature.
Although numerous metal species catalyze the aromatic Claisen
rearrangement in organic solvents,7 only in rare examples have
metal species been shown to catalyze this transformation in water;
therefore, we were unable to expect particular metal species to do
so a priori. As such, we proceeded to screen for metals. Among
the metal reagents tested, only PdCl2 promoted this rearrangement
at 50 °C after 4 h in 1:4 DMSO/pH 10 buffer, while none of the
other reagents afforded 2 or 3 (Figure 1a). This reaction may
proceed through the mechanism shown in Scheme 1,8 indicating
that this detection method is fundamentally different from our
previous method. Although Pd0 species should afford 2 rather than
3, this needed to be confirmed.9 Toward this end, we screened Pd
to-background ratio (S/B) of 3. The S/B increased at a lower
reagents with various oxidation states. As Figure 1b shows, our
detection method is oxidation state-specific and each PdII reagent
and a PdIV reagent successfully performed the conversion from 1
to 310 while Pd0 and insoluble Pd species did not.11
buffer concentration (Figure 1c).
We next examined the detection of PdII contamination in the
presence of Pd0 in functionalized organic compounds. Since this
process requires the detection of PdII in the presence of a large
excess of synthetic compound (500, 50, 5 ppm ) 2000, 20 000,
200 000 equiv of compound with respect to PdII), we were originally
skeptical about our own method because of such stoichiometry.
Nonetheless, each compound (12.5 mg/mL) was spiked with Pd0
(6 µM in solution) and varying amounts of PdII ([PdII] ) 5-500
ppm relative to each compound; 0.6-60 µM in solution), treated
with 1 and heated at 50 °C for 4 h in 1:4 DMSO/pH 10 buffer.
Figure 2a shows that, although the absolute fluorescence fluctuated
among samples (Figures S3a and S3c), for each organic compound
the relative contents of PdII can be rapidly monitored to prioritize
Using PdCl2, we examined the initial rate and the sensitivity
of our method. The initial rate was measured in 1:1 DMSO/pH
10 buffer (Figure S2). The reaction continued to proceed even
after 24 h (data not shown), indicating that longer incubation
time would increase the sensitivity of this detection method
proportionally since the fluorescence signal is generated catalyti-
cally with respect to the analyte.10 The fluorescence intensity
correlated to the concentration of PdII in the 0.5-50 µM (50
ppb-5 ppm) range (Figure 1b). The detection limit under these
conditions was calculated as 3.9 µM (390 ppb) with a signal-
9
16472 J. AM. CHEM. SOC. 2008, 130, 16472–16473
10.1021/ja8065539 CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
drink. As Figure 2c shows, spiked PtIV was successfully detected
by fluorescence in a concentration-dependent manner with a
detection limit of 0.54 nM (0.11 ppb) with S/B of 3 in the presence
of Pt0 at 250 µM, which is 3-orders of magnitude more sensitive
than that currently employed.16
In summary, we have demonstrated that fluorogenic probe 1 can
detect PdII/IV and PtIV via Claisen rearrangement to 3 even in
functionalized compounds and Pt0-water, each without sample
preparation steps. This method may find application in the
pharmaceutical industry, the environment, and Pd/Pt quality control.
Figure 2. PdII-specific detection in the presence of synthetic samples and
PtIV detection in water. For details, see text and Figure S3. (a) The y-axis
is log10(fluorescence intensity (au × 105) at 535 nm) (normalized).12 B,
D-J: [PdII] ) 60 µM (red), 6 µM (blue), 0.6 µM (green). A ) background,
B ) PdII only, C ) Pd0 only, D ) PdII + Pd0. All organic compounds
contain both PdII and Pd0: E ) thioanisole, F ) cholesterol, G ) 2-carboxy-
7-hydroxycoumarin, H ) morpholine, I ) indole, J ) N-methylephedrine.
(b, c) The y-axis is fluorescence intensity (au × 105) at 535 nm. (b)
Monitoring of PtIV (1 mM) reduction to Pt0. Reduction time ) 0 (A), 10
(B), 20 (C), 30 min (D). (c) Detection of PtIV in Pt0-containing drinking
water ([Pt0] ) 250 µM). [PtIV] ) 0 (A), 0.5 () 0.0975 ppb) (B), 5 (C), 50
(D), 500 nM (E).
Acknowledgment. We thank Mr. Minenobu Okayama (Apt Co.,
LTD) for providing platinum samples. This project was funded by
the National Science Foundation (Grant CHE-0616577).
Supporting Information Available: Details of all fluorescence
analyses. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Garrett, C. E.; Prasad, K. AdV. Synth. Catal. 2004, 346, 889.
(2) (a) Galaffu, N.; Man, S. P.; Wilkes, R. D.; Wilson, J. R. H. Org. Process
Res. DeV. 2007, 11, 406. (a) Welch, C. J.; Albaneze-Walker, J.; Leonard,
W. R.; Biba, M.; DaSilva, J.; Henderson, D.; Laing, B.; Mathre, D. J.;
Spencer, S.; Bu, X.; Wang, T. Org. Process Res. DeV. 2005, 9, 198. (b)
Urawa, Y.; Miyazawa, M.; Ozeki, N.; Ogura, K. Org. Process Res. DeV.
2003, 7, 191.
(3) Electrochemical methods require chromatographic separation, during which
Pd0 can be oxidized to PdII. ICP-MS analysis and a method by the Anslyn
group require acid treatment of samples. Houk, R. J. T.; Wallace, K. J.;
Hewage, H. S.; Anslyn, E. V. Tetrahedron 2008, 64, 8271.
scavenging methods and optimize the protocol in a high throughput
manner without pretreatment. While indole at 12.5 mg/mL was
found to quench the fluorescence signal of 3 (note: it does not
quench at <1.3 mg/mL; see Figure S4), an electron-deficient indole
was found to be compatible with this method (Figure S5).
N-Methylephedrine presumably binds to PdII strongly and retards
the metal-catalyzed Claisen rearrangement (example J).10 Even with
these types of compounds, the relative Pd concentrations can still
be monitored during Pd scavenging because the relative fluorescence
signal decreases as the Pd content decreases.10
On the basis of the similar π-electrophilicity between cationic
Pd and Pt species, we asked if this method could be extended to
Ptn+ detection. Pt0 has been shown to be beneficial for human health
due to its ability to catalytically quench reactive oxygen species to
less toxic materials and is used in many health-related products
including commercially bottled drinking water.13 However, a major
concern in manufacturing these products is contamination with PtIV
because it is produced through the reduction of the more stable
PtIV species and PtIV is highly toxic.
(4) (a) MacQuarrie, S.; Horton, J. H.; Barnes, J.; McEleney, K.; Loock, H.-P.;
Crudden, C. M. Angew. Chem., Int. Ed. 2008, 47, 3279. (b) Ko¨hler, K.;
Kleist, W.; Pro¨ckl, S. S. Inorg. Chem. 2007, 46, 1876. (c) Crudden, C. M.;
Sateesh, M.; Lewis, R. J. Am. Chem. Soc. 2005, 127, 10045.
(5) (a) Song, F.; Garner, A. L.; Koide, K. J. Am. Chem. Soc. 2007, 129, 12354.
For the absorption and emission spectra of 1 and 2, see ref 6 and (b) Garner,
A. L.; Koide, K. Chem. Commun., in press. For the application of 1 for Pt
detection, see: (c) Garner, A. L.; Koide, K. Chem. Commun., in press.
(6) Koide, K.; Song, F.; de Groh, E. D.; Garner, A. L.; Mitchell, V. D.;
Davidson, L. A.; Hukriede, N. A. ChemBioChem 2008, 9, 214.
(7) Majumdar, K. C.; Alam, S.; Chattopadhyay, B. Tetrahedron 2008, 64, 597.
(8) (a) van der Baan, J. L.; Bickelhaupt, F. Tetrahedron Lett. 1986, 27, 6267.
(b) Schenck, T. G.; Bosnich, B. J. Am. Chem. Soc. 1985, 107, 2058.
(9) Itami, K.; Yamazaki, D.; Yoshida, J.-I. Org. Lett. 2003, 5, 2161.
(10) See Supporting Information.
(11) A similar observation was also made in organic solvent.10
(12) The data are normalized for 500 ppm PdII analogously to ICP-MS analysis,
in which a standard curve is generated for Pd in each organic molecule
separately. For more detail, see Supporting Information.
(13) (a) http://www.apt-ca.com. (b) Kajita, M.; Hikosaka, K.; Iitsuka, M.;
Kanayama, A.; Toshima, N.; Miyamoto, Y. Free Radical Res. 2007, 41,
615.
Although in our metal screening studies PtII did not produce
fluorescence signal, we hypothesized that PtIV would be a more
efficient catalyst because it is presumably more π-electrophilic.14
Indeed, unlike Pt0/II, PtIV catalyzed the Claisen rearrangement in
water.15 We applied this reaction for the Pt0 manufacturing process
to monitor the progress of the electrochemical reduction of PtIV to
Pt0 in water. As Figure 2b shows, our detection method is successful
for fluorescently monitoring this reduction. We next used this
fluorescence method to detect PtIV contamination in a Pt0-containing
(14) Barluenga, J.; Die´guez, A.; Ferna´ndez, A.; Rodr´ıguez, F.; Fan˜ana´s, F. J.
Angew. Chem., Int. Ed. 2006, 45, 2091.
(15) There is one example from the literature reporting the PtIV-catalyzed Claisen
rearrangement: Stewart, H. F.; Seibert, R. P. J. Org. Chem. 1968, 33, 4560.
(16) In a manufacturing facility, PtIV can be detected at 500 nM and above in
the presence of Pt0 at 250 µM. Minenobu Okayama, APt Co., personal
communication; April 1 and May 16, 2008.
JA8065539
9
J. AM. CHEM. SOC. VOL. 130, NO. 49, 2008 16473