A rapid and reliable assay for regioselectivity using fluorescence spectroscopy

Article abstract of DOI:10.1002/adsc.200606047

The first report of a fluorescence-based assay for the direct measurement of the regioselectivity of a reaction is described herein, developed from the desire to construct a quicker analytical method for the determination of the ratio of regioisomers obtained in the tandem hydroformylation/Fischer indole synthesis. The method allows for extremely rapid acquisition times, as the use of crude reaction mixtures is allowed. The assay is also shown to be overall very reliable, tolerating the presence of various functional groups and proceeding on average with a standard error of measurement comparable to that of NMR. As fluorescence is the only requirement for the employment of this analytical method, countless numbers of target-specific assays can undoubtedly be developed based upon this initial finding.

Full text of DOI:10.1002/adsc.200606047

FULL PAPERS
DOI: 10.1002/adsc.200606047
A Rapid and Reliable Assay for Regioselectivity Using Fluores-
cence Spectroscopy
Goran Angelovski,a, b,* Mark D. Keränen,a, c,* Petra Linnepe,a
Stefan Grudzielanek,d and Peter Eilbrachta,*
a
Fachbereich Chemie, Organische Chemie I, Universität Dortmund, Otto-Hahn-Strasse 6, 44227 Dortmund, Germany
Fax : (+49)-231-755-5363; e-mail: goran.angelovski@udo.edu or peter.eilbracht@udo.edu
Current address: Department for Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics,
b
Spemannstrasse 38, 72076 Tübingen, Germany
Fax : (+49)-7071-601-652; e-mail: goran.angelovski@tuebingen.mpg.de
Department of Math, Science & Technology, University of Minnesota, Crookston, 2900 University Ave., Crookston, MN
c
56716, USA
Fax : (+1)-218-281-8250; e-mail: keran004@umn.edu
Fachbereich Chemie, Physikalische Chemie I, Universität Dortmund, Otto-Hahn-Strasse 6, 44227 Dortmund, Germany
d
Received: February 8, 2006; Accepted: May 9, 2006
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The first report of a fluorescence-based various functional groups and proceeding on average
assay for the direct measurement of the regioselec- with a standard error of measurement comparable to
tivity of a reaction is described herein, developed that of NMR. As fluorescence is the only require-
from the desire to construct a quicker analytical ment for the employment of this analytical method,
method for the determination of the ratio of re- countless numbers of target-specific assays can un-
gioisomers obtained in the tandem hydroformyla- doubtedly be developed based upon this initial find-
tion/Fischer indole synthesis. The method allows for ing.
extremely rapid acquisition times, as the use of crude
reaction mixtures is allowed. The assay is also shown Keywords: fluorescence spectroscopy; hydroformyla-
to be overall very reliable, tolerating the presence of tion; indoles; rapid assay; regioselectivity
Introduction
quently employed homogenously catalyzed processes
in the chemical industry and is also a very important
The ability to quickly and easily monitor the selectivi- tool in organic synthesis.[4] This transformation of an
ty of any chemical reaction forming a carbon-carbon olefin to an aldehyde is extremely versatile as it can
bond on a small scale via spectroscopic means is an be used alone or in sequence with subsequent trans-
indispensable tool to synthetic chemists.[1] The ap- formations.[5] A large amount of effort has gone into
pearance of a number of high-throughput screening developing bidentate ligands such as BIPHEPHOS
methods in the past years has resulted in various pos- and XANTPHOS designed to address one of the in-
sibilities for monitoring chemical reactions and pro- herent properties of the reaction, namely the produc-
cesses.[2] Among these are techniques based on tion of regioisomeric aldehydes when unsymmetrical
HPLC, colorimetry, MS, IR or fluorescence spectros- olefins are employed as starting materials.[6] Styrenes
copy. However, none of these methods are universally and their derivatives have proven to be somewhat
applicable, especially when the screening of catalytic challenging substrates as they are known to produce
transformations is desired, where many different cata- mixtures of regioisomers that are less controllable by
lyst:ligand:substrate ratios may be explored producing the addition of a ligand to the reaction.[6b,c,7] When
varying ratios of isomeric products. Examples of this screening ligands on a new substrate, another chal-
ꢀ
type of strategy have been reported for C C bond lenge is frequently encountered in the analysis of the
forming reactions such as aldol additions,[3] but exam- various mixtures of regioisomers generated. Many al-
ples involving analysis of transition metal-catalyzed dehydes and adducts formed via hydroformylation are
processes such as hydroformylation have not been re- sensitive and do not lend themselves to even standard
ported. Hydroformylation is one of the most fre- methods of analysis such as gas chromatography.
Adv. Synth. Catal. 2006, 348, 1193 – 1199
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1193

Goran Angelovski et al.
FULL PAPERS
Other techniques requiring large amounts of handling the regioselectivity of the hydroformylation. The in-
and purification prior to analysis generally prove to creased amount of conjugation present in the rear-
be more problematic. Our goal was to bypass these ranged products relative to the 3-benzylindoles en-
issues standing in the way of a smooth analysis and to sures that they will exhibit different spectral proper-
develop a method for the rapid and reliable determi- ties – namely that the highly conjugated system will
nation of the regioselectivity of the hydroformylation be a stronger fluorophore than the non-conjugated
reaction using fluorescence spectroscopy, a technique system – which made this system an ideal one for ini-
widely used due to its high sensitivity and simplicity. tial measurements.[11]
Here we report a rapid fluorescence assay to monitor
Several experiments were performed to test our
and determine the regioselectivity of the sequential methodology as a means of determining the regiose-
Rh-catalyzed hydroformylation/Fischer indole synthe- lectivity of the reaction. To establish a reference, a
sis.
collection of standard spectra was acquired on known
ratios (10:0, 9:1, etc…) of indoles 5a and 6a. In all of
these cases, the indolization was performed stepwise
using H2SO4(cat.)/THF.[9] After removal of the reac-
tion solvent and purification by silica gel chromatog-
Results and Discussion
Our group has been investigating hydroformylation raphy, samples were dissolved to 20 mM in DMSO. In
coupled with tandem/sequential reactions for several order to ensure equal probability of excitation for
years.[5b–e] Much effort has gone into studying the re- both compounds in the mixture (lmax5a =284 nm,
gioselectivity of the hydroformylation under these
lmax6a =311 nm), a wavelength of 275 nm was used
types of reaction conditions. Recently, significant re- since the extinction coefficients of both isomers are
sults have been observed in the sequential hydrofor- roughly the same on this wavelength. The fluores-
mylation/Fischer indole synthesis, where an easy and cence spectrum of each mixture was recorded, and
straightforward synthesis of pharmaceutically active the dependence of the ratio of emission intensities at
indoles has been discovered.[8] The development of the respective emission maximum as well as the
this new analytical method came out of this most center of spectral mass on the ratio of isomers were
recent study, where the need for an improved assay calculated. The difference of nearly 30 nm in the
for regioselectivity was demonstrated. Our methodol- emission maxima and the presence of an isosbestic
ogy was inspired by the phenomenon occurring point around 350 nm supported our predictions of
during the indolization of aldehydes with phenylhy- two fluorescent regioisomers in the mixture, clearly
drazines, a reaction well explored and recently de- showing the strong dependence of the fluorescence
scribed by our group elsewhere.[9] In either a stepwise spectra on the ratio of these regioisomers (Figure 1).
or tandem fashion, the linear n-aldehydes 2a,b arising
To determine whether the analysis of crude samples
from styrene hydroformylation proceed to give 3-ben- was possible, aliquots were simply removed from the
zylindoles 5a–c while the regioisomeric branched iso- reaction and evaporated prior to acquiring the fluo-
aldehydes 3a,b result in 3-methyl-2-phenylindoles 6a– rescence spectra. Amazingly, there was no difference
c as the products of Wagner–Meerwein rearrange- observed between crude and pure products, lending
ment[10] (Scheme 1). As the ratio of n-:iso-aldehydes greatly to this methodꢁs employment as a rapid assay
is not changed by the indolization reaction conditions, (Figure 2).[12]
the ratio of the final indole products directly reflects
Scheme 1. Synthesis of indoles via tandem or stepwise hydroformylation/Fischer-indolization.
1194
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1193 – 1199

Assay for Regioselectivity Using Fluorescence Spectroscopy
FULL PAPERS
Figure 1. Fluorescence spectra of 5a and 6a and their mix-
tures (5a:6a=10:0, 9:1, …, 1:9, 0:10) in DMSO, lex =
275 nm, c=20 mM at 258C.
Figure 3. Fluorescence spectra of 5b, 6b and their mixtures
(lex =275 nm, DMSO, c=5 mM, 258C).
proaches are expected to be rather insensitive to the
particular instrumental set-up and are independent of
the fluorophore concentration, as long as inner filter
effects or self-quenching due to complex formation
can be ruled out. Hence, data from different spec-
trometers are likely to be comparable.
To further determine the scope and limitations of
this assay, other novel substrates that bore substitu-
tion on either the styrene (5b and 6b) or the phenyl-
hydrazine (5c and 6c) were run through the tandem
reaction and their fluorescence spectra were mea-
sured. When a p-chlorostyrene was employed in the
reaction, an even greater change (ca. 60 nm) in the
maxima of the fluorescence spectra than was present
in the initial pair was noticed (Figure 3). As expected,
similar results were noticed on both standard curves
for the determination of the isomeric ratio (Figure 4).
An additional test of the breadth of this assay came
from the use of indoles 5c and 6c, prepared from sty-
rene and p-chlorophenylhydrazine (Figure 5). These
compounds exhibit different spectral properties than
the previous pairs in that they have no noticeable dif-
ference in fluorescence maxima. Nevertheless, they
show great changes in their fluorescence intensity (6c
is approximately 60-fold stronger than 5c). Taking ad-
vantage of this fact, the intensities of 5c and 6c and
their mixtures can be plotted directly, easily allowing
for the determination of the ratio of regioisomers
present (Figure 6).
Figure 2. Regioselectivity profiles of both crude and pure 5a
and 6a and their mixtures in DMSO, lex =275 nm, c=20 mM
at 258C.
As should be expected, all three pairs of indoles –
Dual calculations were performed for the determi- those lacking any substitutents or those bearing chlor-
nation of the isomeric ratio. First, the dependence of ine as a substituent anywhere on the indole or arene
the ratio of the fluorescence intensities at the emis- system – exhibited different spectral characteristics. In
sion maxima of the respective pure isomers on the spite of these differences, we were able in all cases to
isomeric ratio was calculated. For comparison, the de- equate the spectra and calculate the dependence of
pendence of the center of spectral mass (COM, hn˜i) fluorescence intensities at the emission maxima with
on the isomeric ratio was also calculated. Both ap- the isomeric ratio.
Adv. Synth. Catal. 2006, 348, 1193 – 1199
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1195

Goran Angelovski et al.
FULL PAPERS
Figure 6. Maximum emission intensity (lem =361 nm) vs.
ratio of the regioisomers (lex =275 nm) of the 5c/6c pair.
mers (obtained from the hydroformylation of styrene
or by mixing known amounts of the corresponding
commercial products), was trapped with phenylhydra-
zine and the corresponding pair of indoles is obtained.
The ratio of the products was determined by obtain-
ing the fluorescence spectra and calculating the ratio
from the standard curves. Finally, the result obtained
from the fluorescence assay was compared with the
1
ratio determined via H NMR spectroscopy. The re-
Figure 4. Ratio of the fluorescence intensity at the two emis-
sion maxima (of n- and iso-regioisomer, respectively) (top)
and COM of the emission spectra (bottom) vs. the molar
ratio of the two regioisomers of 5b/6b pair at lex =275 nm.
sults indicate that further investigation and develop-
ment of this method is most certainly warranted.
Over 65% of the trials deviated by less than ꢁ5%
from the result shown by NMR. Less than one run in
ten deviated by more than ꢁ10% from the integrated
NMR spectra.
Conclusions
In conclusion, we have reported a rapid and reliable
fluorescence-based assay for the measurement of re-
gioselectivity. Notably, the assay allows for the analy-
sis of crude reaction mixtures and has proven itself
superior to gas chromatography in the handling of un-
stable reaction products. In addition, the accuracy of
the assay is comparable to that of NMR spectroscopy.
This method brings together two seldom-paired disci-
plines of chemical science and opens a new realm of
applications of fluorescence spectroscopy as an ana-
lytical tool in organic synthesis. We demonstrated its
use in the determination of regioisomers produced in
the tandem hydroformylation/Fischer indole synthesis,
and reliable results were obtained in all cases. As
Figure 5. Fluorescence spectra of 5c, 6c and their mixtures
(lex =275 nm, DMSO, c=20 mM, 258C).
With the rapidity and tolerance of this assay dem- fluorescence is the sole requirement for its employ-
onstrated, comparing its reliability and accuracy to ment, this method allows for the creation of specific
1
the standard technique of H NMR spectroscopy was assays for other bond-forming reactions, target com-
performed as a final test. A mixture of aldehyde iso- pounds or isomers.
1196
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1193 – 1199

Assay for Regioselectivity Using Fluorescence Spectroscopy
FULL PAPERS
1 equiv. of 4a were diluted in 4% H2SO4/THF (w/w). The
reaction mixture was refluxed for 3 h, cooled to the room
temperature, washed with aqueous ammonia and dried over
MgSO4. The solvent was evaporated and the crude product
was purified by flash chromatography on silica gel.
Experimental Section
General Remarks
All chemicals were purchased from commercial sources and
were used without further purification. Solvents were puri-
fied using standard procedures.[13] 3-Phenylpropionaldehyde
(hydrocinnamaldehyde, 2a) and 2-phenylpropanal (hydratro-
paldehyde, 3a) were purchased from Fluka. All reactions
with air-sensitive compounds were carried out in dry reac-
tion vessels under an atmosphere of dry argon. 1H and
13C NMR spectra were recorded at room temperature with
Bruker DRX 400 and DRX 500 spectrometers. The 1D-
NOESY experiment was performed on a Varian Inova 600.
If not otherwise specified, CDCl3 was used as solvent with
TMS as an internal standard. 13C NMR spectra were refer-
enced to CDCl3 (77.00 ppm). Mass spectra (FAB) were re-
corded using a JEOL JMS-SX 102 A spectrometer. FT-IR
spectra were recorded as thin films on salt plates or as KBr
pellets. The peak intensities are defined as very strong (vs),
strong (s), middle (m) or weak (w). Fluorescence measure-
ments were performed with a Perkin–Elmer LS 50B spectro-
fluorometer at 258C. Column chromatography was per-
formed using silica gel 60 (70–230 mesh ASTM) from Ma-
cherey–Nagel GmbH & Co. KG, Düren, with a mixture of
cyclohexane and MTBEor EtOAc as eluent. Pressure reac-
tions were carried out in autoclaves of type A (250 mL,
PTFEinsert) from Berghof, Eningen or similar house-made
autoclaves (100 mL, stainless steel) with specially designed
heating and stirring mantles.
Preparation of 3-Benzyl-1H-indole (5a) and
3-Methyl-2-phenyl-1H-indole (6a)
Method 1: Following general procedure A, 0.32 g
(3.1 mmol) of 1a and 0.32 g (3.0 mmol) of 4a were stirred in
anhydrous THF for 2 d. After purification 0.05 g (8%) of 5a
were isolated along with 0.19 g (31%) of 6a. The spectro-
scopic data matched that reported in the literature.[14]
Method 2: Following general procedure C, 5a and 6a were
synthesized in quantitative amounts when starting from 2a
or 3a and phenylhydrazine.
Preparation of 3-(4-Chlorobenzyl)-1H-indole (5b)
and 2-(4-Chlorophenyl)-3-methyl-1H-indole (6b)
Following general procedure A, 0.28 g (2.0 mmol) of 1b and
0.23 g (2.1 mmol) of 4a were stirred in anhydrous THF for 2
d. After purification 0.09 g (20%) of 5b were isolated along
with 0.21 g (48%) of 6b, both as yellow solids.
5b: 1H NMR (500 MHz, CDCl3): d=4.00 (s, 2H, CH2),
6.83 (s, 1H, CH), 6.98–7.41 (8H, 8 ” CH), 7.90 (bs, 1H,
NH); 13C NMR (125 MHz, CDCl3): d=31.0 (CH2), 111.1
(CH), 115.3 (Cq), 119.0 (CH), 119.5 (CH), 122.2 (CH), 122.3
(CH), 126.5 (Cq), 128.4 (2 ” CH), 130.0 (2 ” CH), 131.6
(Cq), 139.7 (Cq), 148.0 (Cq); FAB-MS: m/z (%)=242 (M+
H+, 49), 241 (M+, 100), 74 (29); IR: n˜ =3408 (m), 3051 (w),
2920 (w), 1491 (s), 1090 (s), 744 (vs) cmꢀ1; FAB-HR-MS:
General Procedure for Reactions in the Autoclave
After charging the autoclave with the starting material, the
catalyst precursor and the solvent, the reactor was flushed
with argon, pressurized with gases (carbon monoxide and
hydrogen), and heated to the required reaction temperature.
Following the reaction, the solvent was removed by rotary
evaporation and the products were purified by column chro-
matography.
calculated for C15H12NCl: 241.0658 gmolꢀ1
241.0661 gmolꢀ1
;
found:
.
6b: 1H NMR (500 MHz, CDCl3): d=2.35 (s, 3H, CH3),
7.03–7.17 (2H, 2 ” CH), 7.25–7.53 (6H, 6 ” CH), 7.88 (bs,
1H, NH); 13C NMR (125 MHz, CDCl3): d=9.6 (CH3), 109.2
(Cq), 110.7 (CH), 119.1 (CH), 119.7 (CH), 122.6 (CH), 128.9
(2 ” CH), 129.0 (2 ” CH), 129.9 (Cq), 131.7 (Cq), 132.8
(Cq), 133.2 (Cq), 135.9 (Cq); FAB-MS: m/z (%)=242 (M+
H+, 43), 241 (M+, 100), 77 (89); IR: n˜ =3419 (vs), 3050 (w),
2925 (w), 1643 (w), 1493 (w), 1092 (m), 744 (vs) cmꢀ1; FAB-
HR-MS: calculated for C15H12NCl: 241.0658 gmolꢀ1; found:
General method A (indoles directly from olefins): 1 equiv.
olefin, 1 equiv. arylhydrazine, 1 equiv. p-toluenesulfonic acid
and Rh(acac)(CO)2 (0.5 mol%) were diluted in anhydrous
G
THF, filled in an autoclave and pressurized with CO and
H2. After stirring at 1008C for 2 or 3 days, the mixture was
washed with aqueous ammonia and dried over MgSO4. The
solvent was evaporated and the crude product was purified
by flash chromatography on silica gel.
241.0670 gmolꢀ1
.
General method B (indoles via hydrazones): 1 equiv.
olefin, 1 equiv. arylhydrazine and Rh(acac)(CO)2 (0.5
A
Preparation of 1-Chloro-4-iodobenzene
mol%) were diluted in anhydrous THF, filled in an auto-
clave and pressurized with CO and H2. After stirring for 3 d
at 1008C, the solvent was evaporated. The crude hydrazone
was dissolved in 4% H2SO4/THF (w/w). The reaction mix-
ture was refluxed for 3 h, cooled to the room temperature,
washed with aqueous ammonia and dried over MgSO4. The
solvent was evaporated and the crude product was purified
by flash chromatography on silica gel.
4-Chloroaniline (8.23 g, 88 mmol) was suspended in 12 mL
H2SO4 and 100 mL water and cooled to 08C. NaNO2
(6.10 g, 88 mmol) dissolved in 20 mL water was added. The
reaction mixture was filtered into a solution consisting of
22.53 g (150 mmol) NaI and 75 mL water. After stirring at
room temperature for 15 min the mixture is extracted with
MTBE, dried over MgSO4 and the solvent is removed to
afford 1-chloro-4-iodobenzene; yield: 12.70 g (60%).
General method C (indoles from aldehydes): 1 equiv. of
the pure or mixed 2a and 3a (n- and iso-aldehydes) and
Adv. Synth. Catal. 2006, 348, 1193 – 1199
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1197

Goran Angelovski et al.
FULL PAPERS
3
1H NMR (500 MHz, CDCl3): d=7.08 (d, J=8.7 Hz, 2H, 2
” CHar), 7.61 (d, 3J=8.7 Hz, 2H, 2 ” CHar); 13C NMR
(125 MHz, CDCl3): d=109.6 (Cq), 130.5 (2 ” CHar), 133.0
(Cq), 138.7 (2 ” CHar). The spectroscopic data matched
those reported in the literature.[15]
Determination of the Regioisomeric Ratio
Stock solutions of pure n- (5a) and iso- (6a) products were
prepared and diluted to the concentration of 20 mM. Nine
mixtures with the same concentration of this pair were pre-
pared (n:iso=1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7:3, 8:2 and 9:1).
Fluorescence spectra of every mixture were recorded, and
the dependency of the ratio of emission intensities and the
center of spectral mass (COM) of the ratio of isomers were
calculated, respectively. The standard curves thus obtained
were each fitted with a double-exponential function:[17]
Preparation of N-(4-Chlorophenylhydrazine)-
carboxylic Acid tert-Butyl Ester (4b)
1-Chloro-4-iodobenzene (2.98 g, 13 mmol), BocNHNH2
(1.98 g, 15 mmol), cesium carbonate (5.70 g, 18 mmol), CuI
(0.12 g, 1 mmol) and phenanthroline (0.23 g, 1 mmol) were
dissolved in 13 mL DMF and heated to 808C for 3 d. The re-
action mixture was filtered through silica with ethyl acetate,
washed with water and dried over MgSO4. After column
chromatography 4b was isolated; yield: 1.60 g (56%).
1H NMR (500 MHz, CDCl3): d=1.50 (s, 9H, 3 ” CH3), 4.39
where y=measured fluorescence signal (ratio of intensities
at the emission maxima of the pure isomers/center of spec-
tral mass/emission intensity at a fixed wavelength, respec-
tively), x=percentage of iso-regioisomer, y0 =signal for
pure n-isomer, y1, y2 =amplitudes of the double-exponential
function, and t1, t2 =corresponding exponential factors.
For the purpose of the fluorescence experiment, samples
with “unknown” ratios were prepared with the same con-
centration (20 mM in DMSO) as with pure products. Upon
recording of the fluorescence spectra and calculating rele-
vant values for the Inmax/Iisomax and COM, the n:iso ratio
was calculated according to exponential decay fit [Eq. (1)]
and values were compared with ones calculated from
1H NMR (see Table 1 in the Supporting Information).
Samples with “unknown” ratios were synthesized via gen-
eral procedure B or C:
3
(bs, 2H, NH2), 7.25 (d, J=8.7 Hz, 2H, 2 ” CHar), 7.42 (d,
3J=8.7 Hz , 2H, 2 ” CHar); 13C NMR (125 MHz, CDCl3):
d=28.3 (3 ” CH3), 82.2 (Cq), 124.4 (2 ” CHar), 128.2 (2 ”
CHar), 129.7 (Cq), 142.5 (Cq), 153.0 (Cq); MS (FAB): m/z
(%)=243 (M+H+, 24), 242 (M+, 22), 187 (100), 142 (45),
58 (79); IR (KBr/Film): n˜ =3310 (m), 3010 (w), 2982 (m),
1696 (vs), 1491 (s), 1335 (s), 1160 (m) cmꢀ1; HR-MS (FAB):
calculated for C11H15N2O2ClNa: 265.0720 gmolꢀ1; found:
265.0723 gmolꢀ1
.
Preparation of 3-Benzyl-5-chloro-1H-indole (5c) and
5-Chloro-3-methyl-2-phenyl-1H-indole (6c)
Via general procedure B: After the hydroformylation of
styrene (1a) and trapping of aldehydes with phenylhydra-
zine, the mixture of regioisomeric indoles 5a and 6a was ob-
tained in its crude form. The ratio of isomers was deter-
Following general procedure B, 0.22 g (2.1 mmol) of 1a and
0.49 g (2.0 mmol) of 4b[16] were stirred in anhydrous THF
for 3 d. After purification 0.13 g (26%) of 5c were isolated
along with 0.21 g (44%) of 6c, both as yellow solids.
1
mined via H NMR and compared with one obtained after
the fluorescence experiment.
Via general procedure C: The arbitrary mixture of com-
mercial aldehydes 2a and 3a was trapped with phenylhydra-
zone into the mixture of regioisomeric indoles 5a and 6a.
The obtained indole ratio was the same as that of the start-
5c: 1H NMR (500 MHz, CDCl3): d=4.07 (s, 2H, CH2),
6.95 (s, 1H, CH), 6.98 (d, J=8.7 Hz, 1H, CH), 7.13 (d, J=
8.7 Hz, 1H, CH), 7.16–7.22 (2H, 2 ” CH), 7.25–7.31 (3H, 3
” CH), 7.48 (s, 1H, CH), 7.90 (bs, 1H, NH); 13C NMR
(125 MHz, CDCl3): d=31.4 (CH2), 112.1 (CH), 115.6 (Cq),
118.6 (CH), 120.5 (Cq), 122.3 (CH), 122.7 (CH), 125.0 (Cq),
126.0 (CH), 128.4 (2 ” CH), 128.6 (2 ” CH), 131.8 (Cq),
136.5 (Cq); MS (FAB): m/z (%)=242 (M+H+, 49), 241 (M+,
100), 91 (63); IR: n˜ =3428 (m), 3034 (w), 2929 (w), 1601 (m),
1493 (s), 1265 (vs), 739 (vs) cmꢀ1; HR-MS (FAB): calculated
1
ing aldehydes (determined via H NMR of the crude indole
mixture), and was compared with the ratio obtained after
the fluorescence experiment.
Acknowledgements
for C15H12NCl: 241.0658 gmolꢀ1; found: 241.0677 gmolꢀ1
.
6c: 1H NMR (500 MHz, CDCl3): d=2.43 (s, 3H, CH3),
7.16 (d, J=8.4 Hz, 1H, CH), 7.28 (d, J=8.4 Hz, 1H, CH),
7.39 (dd, J=7.2; 7.2 Hz, 1H, CH), 7.50 (dd, J=7.2; 7.5 Hz,
2H, 2 ” CH), 7.55–7.60 (3H, 3 ” CH), 8.04 (bs, 1H, NH);
13C NMR (125 MHz, CDCl3): d=9.6 (CH3), 108.4 (Cq),
111.6 (CH), 118.5 (CH), 122.5 (CH), 125.2 (Cq), 127.7 (CH),
127.8 (2 ” CH), 128.9 (2 ” CH), 131.2 (Cq), 132.8 (Cq),
134.1 (Cq), 135.5 (Cq); MS (FAB): m/z (%)=242 (M+H+,
63), 241 (M+, 100); IR: n˜ =3399 (m), 3067 (w), 2924 (w),
1601 (m), 1495 (s), 1090 (m), 734 (vs) cmꢀ1; HR-MS (FAB):
The authors would like to thank DAAD and DFG for finan-
cial support. We also thank Degussa AG Düsseldorf for the
donation of chemicals.
References
[1] a) R. H. Crabtree, Chem. Commun. 1999, 1611–1616;
b) F. Tanaka, R. Thayumanavan, C. F. Barbas, III J.
Am. Chem. Soc. 2003, 125, 8523–8528.
[2] a) S. R. Stauffer, J. F. Hartwig, J. Am. Chem. Soc. 2003,
125, 6977–6985, and references 1–16 cited therein;
b) K. D. Shimizu, M. L. Snapper, A. H. Hoveyda,
calculated for C15H12NCl: 241.0658 gmolꢀ1
;
found:
241.0661 gmolꢀ1. The structure was clarified by 1D-NOESY
experiments.
1198
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1193 – 1199

Assay for Regioselectivity Using Fluorescence Spectroscopy
FULL PAPERS
Chem. Eur. J. 1998, 4, 1885–1889; c) J. G. de Vries,
A. H. M. de Vries, Eur. J. Org. Chem. 2003, 799–811;
d) M. T. Reetz, Angew. Chem. Int. Ed. 2001, 40, 284–
der Veen, P. H. Keeven, G. C. Schoemaker, J. N. H.
Reek, P. J. C. Kamer, P. W. N. M. van Leeuwen, M.
Lutz, A. L. Spek, Organometallics 2000, 19, 872–883;
e) G. D. Cuny, S. L. Buchwald, J. Am. Chem. Soc. 1993,
115, 2066–2068; f) C. J. Cobley, J. Klosin, C. Qin, G. T.
Whiteker, Org. Lett. 2004, 6, 3277–3280.
´
310; e) G. A. Korbel, G. Lalic, M. D. Shair, J. Am.
Chem. Soc. 2001, 123, 361–362; f) G. T. Copeland, S. J.
Miller, J. Am. Chem. Soc. 2001, 123, 6496–6502; g) B.
Jandeleit, D. J. Schaefer, T. S. Powers, H. W. Turner,
W. H. Weinberg, Angew. Chem. Int. Ed. 1999, 38,
2494–2532; h) J. P. Stambuli, J. F. Hartwig, Curr. Opin.
Chem. Biol. 2003, 7, 420–426; i) Y. Zhang, X. Gong, H.
Zhang, R. C. Larock, E. C. Yeung, J. Comb. Chem.
2000, 2, 450–452.
[7] a) S. Kotha, Tetrahedron 1994, 50, 3639–3662; b) H. R.
Sonawane, N. S. Bellur, J. R. Ahuja, D. G. Kulkarni,
Tetrahedron: Asymmetry 1992, 3, 163–192.
[8] a) P. Kçhling, A. M. Schmidt, P. Eilbracht, Org. Lett.
2003, 5, 3213–3216; b) A. M. Schmidt, P. Eilbracht, J.
Org. Chem. 2005, 70, 5528–5535; c) A. M. Schmidt, P.
Eilbracht, Org. Biomol. Chem. 2005, 3, 2333–2343.
[9] P. Linnepe, A. M. Schmidt, P. Eilbracht, Org. Biomol.
Chem. 2006, 4, 302–313.
[3] F. Tanaka, N. Mase, C. F. , Barbas, III. J. Am. Chem.
Soc. 2004, 126, 3692–3693.
[4] M. Beller, C. Bolm, (Eds. ) Transition Metals for Or-
G
ganic Synthesis: Building Blocks and Fine Chemicals,
2nd edn., Vol. 1, Wiley-VCH, Weinheim, 2004, pp. 29–
55.
[10] a) G. J. Wagner, Russ. Phys. Chem. Soc., 1899, 31, 690;
b) H. Meerwein, Liebigs Ann. Chem. 1914, 405, 129.
[11] Recent applications of indoles as fluorophores, see:
a) Q. Wu, E. V. Anslyn, J. Am. Chem. Soc. 2004, 126,
14682–14683; b) G. Chen, D. J. Yee, N. G. Gubernator,
D. Sames, J. Am. Chem. Soc. 2005, 127, 4544–4545.
[12] It should be noted that if one desires to analyze reac-
tion mixtures containing fluorescent ligands, flash filtra-
tion through a small plug of neutral alumina is advised
in order to remove the unwanted fluorophore.
[5] a) P. Eilbracht, L. Bärfacker, C. Buß, C. Hollman, B. E.
Kitsos-Rzychon, C. L. Kranemann, T. Rische, R. Rog-
genbuck, A. Schmidt, Chem. Rev. 1999, 99, 3329–3365;
b) M. D. Keränen, P. Eilbracht, Org. Biomol. Chem.
2004, 2, 1688–1690; c) M. D. Keränen, K. Kot, C. Holl-
mann, P. Eilbracht, Org. Biomol. Chem. 2004, 2, 3379–
3384; d) G. Angelovski, P. Eilbracht, Tetrahedron 2003,
41, 8265–8274; e) F. KoÅ, P. Eilbracht, Tetrahedron
2004, 60, 8465–8476; f) M. Ahmed, A. M. Seayad, R.
Jackstell, M. Beller, J. Am. Chem. Soc. 2003, 125,
10311–10318; g) B. Breit, Acc. Chem. Res. 2003, 36,
264–275.
[13] D. D. Perrin, W. L. F. Armarego, Purification of labora-
tory chemicals, 3rd edn.; Butterworth Heinemann Ltd.,
Oxford, 1988.
[14] A. Fürstner, D. N. Jumbam, Tetrahedron 1992, 48,
5991–6010.
[6] a) C. P. Casey, G. T. Whiteker, M. G. Melville, L. M.
Petrovich, J. A. Gavney Jr., D. R. Powell, J. Am. Chem.
Soc. 1992, 114, 5535–5543; b) M. Kranenburg, Y. E. M.
van der Burgt, P. J. C. Kamer, P. W. N. M. van Leeuwen,
K. Goubitz, J. Fraanje, Organometallics 1995, 14, 3081–
3089; c) A. van Rooy, P. C. J. Kamer, P. W. N. M. van
Leeuwen, K. Goubitz, J. Fraanje, N. Veldman, A. L.
Spek, Organometallics 1996, 15, 835–847; d) L. A. van
[15] L. Rebrovic, G. F. Koser, J. Org. Chem. 1984, 49, 13,
2462–2472.
[16] The Boc-protected hydrazine is used instead of an un-
protected one since the compound is more stable
against oxidation due to the protection of the nitrogen.
[17] A double-exponential function was arbitrarily chosen
for curve fitting since this approach gave satisfactory
fits with high correlation coefficients (R2 >0.99).
Adv. Synth. Catal. 2006, 348, 1193 – 1199
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1199