Monitoring on-chip pictet-spengler reactions by integrated analytical separation and label-free time-resolved fluorescence

Article abstract of DOI:10.1002/chem.201101768

High-throughput screening for optimal reaction conditions and the search for efficient catalysts is of eminent importance in the development of chemical processes and for expanding the spectrum of synthetic methodologies in chemistry. In this context we report a novel approach for a microfluidic chemical laboratory integrating organic synthesis, separation and time-resolved fluorescence detection on a single microchip. The feasibility of our integrated laboratory is demonstrated by monitoring the formation of tetrahydroisoquinoline derivatives by Pictet-Spengler condensation. After on-chip reaction the products and residual starting material were separated enantioselectively on the same chip. On-chip deep UV laser-induced fluorescence detection with time-correlated single photon counting was applied for compound assignment. The system was utilized to screen reaction conditions and various substrates for Pictet-Spengler reactions on-chip. Finally, the microlab was successfully applied to investigate enantioselective reactions using BINOL-based phosphoric acids as organocatalysts. Chip trick: An integrated chip-based approach for rapid monitoring of organic synthesis at the microscale level is presented. This is achieved by the integration of microfluidic channels for reaction and electrophoresis on a single device together with time-correlated single-photon counting for compound identification (see figure). Copyright

Full text of DOI:10.1002/chem.201101768

DOI: 10.1002/chem.201101768
Monitoring On-Chip Pictet–Spengler Reactions by Integrated Analytical
Separation and Label-Free Time-Resolved Fluorescence
Stefan Ohla,[a] Reinhild Beyreiss,[a] Stefanie Fritzsche,[a] Petra Glaser,[a] Stefan Nagl,[a]
Kai Stockhausen,[b] Christoph Schneider,[b] and Detlev Belder*[a]
Abstract: High-throughput screening
for optimal reaction conditions and the
search for efficient catalysts is of emi-
nent importance in the development of
chemical processes and for expanding
the spectrum of synthetic methodolo-
gies in chemistry. In this context we
report a novel approach for a micro-
fluidic chemical laboratory integrating
organic synthesis, separation and time-
resolved fluorescence detection on a
single microchip. The feasibility of our
integrated laboratory is demonstrated
by monitoring the formation of tetra-
enantioselectively on the same chip.
On-chip deep UV laser-induced fluo-
rescence detection with time-correlated
single photon counting was applied for
compound assignment. The system was
utilized to screen reaction conditions
and various substrates for Pictet–Spen-
gler reactions on-chip. Finally, the mi-
crolab was successfully applied to in-
vestigate enantioselective reactions
using BINOL-based phosphoric acids
as organocatalysts.
hydroisoquinoline
derivatives
by
Pictet–Spengler condensation. After
on-chip reaction the products and re-
sidual starting material were separated
Keywords: asymmetric catalysis
·
condensation reactions electro-
·
phoresis · fluorescence · microreac-
tors
Introduction
occur in microcavities or microchannels of the size of a
human hair, they feature numerous advantages like the ac-
celerated reaction speed and especially the potential for
system integration.[3] The most fascinating perspective of the
development of an integrated chip laboratory, where reac-
tion, sample transfer and analytical measurements are per-
formed on a single device, is the ability to create entirely
new systems that surpass the performance of conventional
chemical devices.[4] While extensive work has been done
either on microreaction technology or on chip-based analyti-
cal devices, fully integrated approaches are rare. Analytical
characterization of chemical synthesis performed in micro-
channels is still mostly performed offline with classical
equipment, not taking advantage of the promises of lab-on-
a-chip technology.
We reported previously a first step towards an integrated
chip laboratory incorporating biocatalysis, separation, and
detection on a single device.[5] High-throughput screening of
enantioselective biocatalysts was demonstrated in this first
proof of concept of an integrated chip laboratory. As enzy-
matic conversion and electrophoresis both occurred in aque-
ous media it was quite facile to combine the synthesis and
analysis steps by interconnecting channels on a microfluidic
device. For a more general application of this technology it
would be desirable to extend the approach to organic reac-
tions, typically using non-aqueous solvents, and to integrate
a more powerful detection technique for substance identifi-
cation. The predominant detection technique in chip-based
microfluidics is fluorescence due to unsurpassed sensitivity
and ease of technical implementation. However, a major
drawback of common fluorescence techniques is the require-
ment to tag analytes with fluorescent labels prior to analysis.
The optimization of reaction conditions and the search for
new synthetic routes and catalysts are key challenges in syn-
thetic chemistry. Traditionally, this research and develop-
ment conducted in ordinary macroscopic flasks or tubes is
rather tedious and cost intensive. To accelerate the inven-
tion process, lab automation is of increasing importance, es-
pecially in combinatorial chemistry. While such chemical
robots can significantly speed up the process, further minia-
turization onto integrated chemical chip laboratories[1] or
microflow reactors[2] is about to revolutionize chemical de-
velopment. Such microfluidic reaction and analysis systems
are expanding the chemistꢀs toolbox, not only in academia
but also in industrial research. One promise of miniaturiza-
tion in chemistry is the drastic reduction of reagent and sol-
vent consumption with obvious cost and safety advantages.
Furthermore, when chemical systems are shrunken into so-
called lab-on-a-chip devices, in which chemical processes
[a] S. Ohla, R. Beyreiss, S. Fritzsche, P. Glaser, Dr. S. Nagl,
Prof. Dr. D. Belder
Institute of Analytical Chemistry
University of Leipzig
Linnꢁstr. 3, 04103 Leipzig (Germany)
Fax : (+)49-341-9736115
E-mail: belder@uni-leipzig.de
[b] K. Stockhausen, Prof. Dr. C. Schneider
Institute of Organic Chemistry
University of Leipzig
Johannisalle 29, 04103 Leipzig
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101768.
1240
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1240 – 1246

FULL PAPER
Especially for analysis of products from organic synthesis
lacking appropriate functional groups, an on-chip labeling
procedure can be troublesome or even impossible. An ap-
proach to achieve label-free optical detection in microflu-
ACHTUNGTRENNUNGidics is to use native analyte fluorescence with excitation in
the deep UV spectral region.[6] A lot of organic compounds,
especially aromatics, exhibit intrinsic fluorescence when ex-
cited with light below 300 nm. Furthermore, when fluores-
cent signals are recorded in the time domain the individual
fluorescence lifetimes (t) can be used to distinguish between
molecules.[7] Fluorescence lifetime spectroscopy is
a
common technique, widely utilized for physicochemical in-
vestigations as well as biomedical applications or in imaging
microscopy.[8] It has been employed for oligonucleotide de-
tection labeled with fluorescent dyes in microchips.[9] While
most studies deal with fluorescently tagged compounds,
label-free fluorescence lifetime detection in the deep UV
has rarely been realized in microscopy and microfluidics.
The chromophores are typically rather weak, so a dedicated
optoelectronic assembly is required for lifetime determina-
tion in microfluidics with adequate sensitivities and accura-
cies in this spectral region. Seegerꢀs group investigated pro-
tein molecules on quartz slides by time-resolved fluores-
cence detection with excitation at 266 nm,[10] while readout
of protein microarrays in the deep UV was also accom-
plished.[11] Several groups reported studies on intrinsic UV
fluorescence from small aromatics in living cells by two- and
three-photon excitation.[12]
Recently, we reported some preliminary data that showed
how this approach can be attractive to monitor on-chip or-
ganic syntheses.[13] We now report on such an integrated ap-
proach, combining a chip reactor followed by a separation
unit where detection is performed by deep UV laser-induced
fluorescence detection with time-correlated single photon
counting for a more reliable peak assignment. Integrating a
synthetic and a powerful analytical lab on a single microflu-
Figure 1. a) Photograph of the employed microchip and illustration of the
working principle with designation of the main components of the micro-
fluidic glass chip. b) Schematic drawing of the experimental procedure
exemplarily shown for the on-chip formation of salsolinol.
ACHTUNGTRENNUNGidic substrate allows the study of complex reactions and
chemical processes on-chip. This is especially interesting for
high-throughput screening and process optimization with
minimal reagent consumption. As a model reaction we in-
vestigated the formation of tetrahydroisoquinoline deriva-
tives by Pictet–Spengler condensation,[14] which is an impor-
tant route in alkaloid synthesis.
phoresis separation structure with a cross injector layout. At
the end of each channel, access holes were implemented
which serve as reagent inlets and as electrode reservoirs.
The whole chip is made of fused silica to ensure sufficient
transmission properties for light below 300 nm. Figure 1b
exemplarily illustrates the working principle of the chip for
an on-chip Pictet–Spengler condensation of dopamine with
acetaldehyde followed by chiral microchip electrophoresis
with deep UV time-resolved fluorescence detection.
Results and Discussion
The experimental setup used for detection is shown in
Figure 2. It is based on a time-resolved confocal fluores-
cence microscope setup which was modified to enable laser-
induced fluorescence detection with excitation in the deep
UV spectral region. This was realized by incorporating a pi-
cosecond Nd:YAG laser emitting at 266 nm and integration
of optical components which ensure transmission and achro-
maticity throughout in the ultraviolet region (from about
200 to 400 nm). The fluorescence emission was detected by
a photon counting photomultiplier tube synchronized to the
laser frequency. This allows recording photon timing with
A key point in the realization of a chip laboratory integrat-
ing chemical reaction and analysis is the design and optimi-
zation of the microfluidic network. Previous work showed
that a rather long serpentine channel as mixing and reaction
structure is quite challenging for enabling mass transport
only by electric forces.[5] On the other hand, an efficient
mixing of the reactants is crucial to ensure efficient conver-
sions.[15] We developed an improved chip design which is
shown in Figure 1a. It mainly consists of two parts: a ser-
pentine micromixer is combined with a microchip electro-
Chem. Eur. J. 2012, 18, 1240 – 1246
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1241

D. Belder et al.
Figure 2. Schematic drawing of the instrumental setup used for native
time-resolved fluorescence detection. FS=fused silica, TCSPC=time-
correlated single photon counting.
respect to the laser pulse with subnanosecond precision.
Thereby, fluorescence decays can be constructed for the de-
termination of native fluorescence lifetimes of each sub-
stance in the electropherogram.
With the setup described above we studied on-chip
Pictet–Spengler condensations to form tetrahydroisoquino-
lines. For this purpose about 5 mL of each reactant are
placed in the inlet microvials. By application of an electric
field the solutions are merged through the microreactor
where mixing and reaction occurs. After conversion, a small
plug of the reaction mixture is transferred into the electro-
phoresis channel by a voltage-controlled pinched injec-
tion.[16] The reaction products and residual starting material
are subsequently separated electrophoretically within the
separation channel and detected with time-correlated single
photon counting. For validation and optimization of the
system we first investigated the reaction of dopamine with
acetaldehyde. In the first step, we optimized the analysis
part of the chip in order to obtain adequate separation and
detection conditions. A 40 mmolLÀ1 phosphate buffer at
pH 3.0 and an applied separation field strength of
315 VcmÀ1 were found optimal as electrophoresis condi-
tions. The effective laser power was adjusted to around
110 mW yielding sufficient sensitivity and negligible photo-
bleaching during the transit time of the chromophores.
Thereafter, the chemical reaction conditions which had to
be compatible with the electrophoretic buffer system were
optimized.
Figure 3. Electropherograms with respective fluorescence decay curves
obtained after on-chip Pictet–Spengler condensation of dopamine with
acetaldehyde. a) Separation performed without a chiral selector. b) Sepa-
ration performed using b-cyclodextrin as additive in the separation
buffer.
of residual starting material and product is achieved in less
than 120 s. Data from on-the-fly fluorescence lifetime mea-
AHCTUNGTREGsNNNU urements were used for peak assignment. The fluorescence
decay obtained from time-correlated single photon counting
can be constructed for each electrophoresis signal; the cor-
responding average fluorescence lifetimes were determined
via bi-exponential tail fitting. These data together with the
electrophoretic mobilities allow a more reliable assignment
of the compounds present in the reaction mixture. The
signal marked with an asterisk derives from salsolinol and is
most likely attributed to an oxidation product of this sub-
stance.[18]
The presented reaction leads to a racemic mixture of the
chiral product. These two enantiomers can be separated on-
chip using an appropriate chiral selector as an additive to
the electrophoresis buffer. After optimization of the type
and concentration of the additive, we could resolve both
To improve the synthesis, various reaction solvent systems
were screened while the conversion was monitored electro-
phoretically, thus revealing that the reaction rate can be sig-
nificantly accelerated by pH increase, which is in accordance
with the literature.[17] The optimized result for an on-chip
Pictet–Spengler reaction within 60 s using a 100 mmolLÀ1
phosphate buffer at pH 8.0 followed by subsequent separa-
tion is shown in Figure 3a. As depicted, baseline separation
1242
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1240 – 1246

Monitoring On-Chip Pictet–Spengler Reactions
FULL PAPER
enantiomers after on-chip reaction. The corresponding elec-
tropherogram shown in Figure 3b was obtained using
12 mmolLÀ1 b-cyclodextrin as chiral selector. Compared to
Figure 3a the resulting migration times are slightly higher
but the enantiomers are now baseline separated. Further-
more, the determined fluorescence lifetimes are considera-
bly higher, which however is not surprising as inclusion com-
pounds tend toward slower fluorescence decays compared
with the non-complexed forms.[19] The fluorescence lifetimes
of the enantiomers vary by a difference of 0.19 ns. A series
of three identical experiments resulted in a standard devia-
tion of less than 0.05 ns for the determined lifetimes, which
indicates that this difference is significant. Such a chiral dis-
crimination based on the alteration of fluorescence lifetimes
caused by chiral additives has only recently been reported in
the literature[20] and might in principle be used to judge the
enantiomeric excess even without preceding separation.
After we could show that the employed integrated reac-
tion and analysis chip is working as intended, we applied the
device for on-chip studies of the scope of this reaction utiliz-
ing different structural variants of the substrates (Scheme 1).
The obtained data are summarized in Table 1. In the case of
chiral products we also performed enantioselective separa-
tions, the corresponding results are displayed in Table 2.
These data show that the device can be successfully utilized
to screen reaction conditions and different starting materials
in a very fast, economic and environmentally friendly fash-
ion as only tiny amounts of chemicals are needed. While the
Figure 4. Reaction time course monitoring. Electropherogram of repeti-
tive injections obtained after condensing dopamine with acetaldehyde in
a microvial.
whole reaction and analysis time is below five minutes, the
step determining the throughput is the rinsing procedure of
the chips which takes several minutes. In a more advanced
chip the overall throughput should be significantly improved
by integrating several reaction channels on a single chip
which could be run in serial.
Owing to the fast analysis process the chip can also be
used to monitor reaction kinetics. For this purpose, the reac-
tions were performed by pipet-
ting the starting materials to-
gether into the same microvial.
By applying
a voltage-con-
trolled gated injection portions
of the reaction mixture are se-
quentially injected into the
separation channel.[21] This ap-
proach allows tracing of the re-
action time course of a chemi-
cal synthesis virtually in real
Scheme 1. Pictet–Spengler reactions with structural variants of the substrates. The structures of the amines and
aldehydes are shown in Table 1.
Table 1. Conversions and average fluorescence lifetimes obtained with
time. An obtained stacked electropherogram recorded after
condensation of dopamine with acetaldehyde with repetitive
injections in 30 s intervals is shown in Figure 4. Since we are
able to distinguish between different product enantiomers,
our approach is ideally suited to monitor asymmetric reac-
tions. As shown recently, several synthetic strategies have
been developed to perform enantioselective Pictet–Spengler
reactions which are, however, entirely limited to indole-
based substrates.[22]
To realize such a reaction on-chip, we first had to evaluate
the compatibility of the reaction conditions with the separa-
tion part. As a proof of principle we chose the condensation
of dopamine and acetaldehyde as model reaction with dif-
ferent BINOL-derived phosphoric acid derivatives as orga-
nocatalysts.[23] Method optimization resulted in the use of
10 mol% (R)-TRIP ((R)-3,3’-Bis(2,4,6-triisopropylphenyl)-
1,1’-binaphthyl-2,2’-diyl hydrogen phosphate) as catalyst and
iPrOH/tBuOH/2-Me-2-BuOH (1:1:1) as solvent for realiza-
on-chip Pictet–Spengler reactions.
Entry
R1
R2
Product
t (1)
[ns]
t (3)
[ns]
Conversion
[%]
1
2
3
H
CH3
H
CH3
CH3
H
3a
3b
3c
0.87
0.94
1.54
1.58
1.65
1.63
33
21
30
Table 2. Conversions and average fluorescence lifetimes obtained with
on-chip Pictet–Spengler reactions with chiral separation.[a]
Entry R1
R2
Product
t (1) t (3[a]
)
t (3[b]
[ns]
)
Conversion
[%]
[ns]
CH3 3a[a] +3a[b] 1.80
CH3 CH3 3b[a] +3b[b] 1.87
[ns]
4
5
H
2.61
2.93
2.76
3.06
30
20
[a] (S)-enantiomer. [b] (R)-enantiomer. t (1) displays the fluorescence
lifetime of compound 1. Data are mean values from triplicate experi-
ments except for entry 5 with duplicate experiments. Standard deviations
of determined fluorescence lifetimes were less than 0.05 ns.
Chem. Eur. J. 2012, 18, 1240 – 1246
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1243

D. Belder et al.
tion of the first asymmetric Pictet–Spengler reaction on-
chip.
the first on-chip enantioselective Pictet–Spengler reaction
using a chiral organocatalyst. For future work, an on-chip
implementation of additional microreactors appears con-
ceivable. This would allow high-speed screening of large
compound libraries with minimal reagent consumption.
Such integrated chip laboratories are fascinating tools allow-
ing to establish a more economic and environmentally
friendly chemistry.
Although the alcoholic reaction media and the aqueous
separation buffer did not match perfectly, we were finally
able to integrate the asymmetric reaction with chiral separa-
tion on the chip. An unsmoothed electropherogram ob-
tained after chiral catalysis for 60 s and subsequent separa-
tion of the reaction products is shown in Figure 5. The enan-
Experimental Section
Microchip specifications: Microchips were manufactured by standard
photolithographic techniques. As chip material fused silica was utilized in
order to ensure the required transmission properties for light below
300 nm. The outer chip dimensions were 90ꢃ15ꢃ1.8 mm. The lengths of
the reaction channel and of the separation channel were 3.4 cm and
8.0 cm, respectively. The microchannels were 20 mm in depth and 50 mm
in width at the top. Unsealed access holes at the end of each channel
were used to introduce the compounds and for electrical contact utilizing
platinum electrodes. These microvials could up to 10 mL of solution and
are designated sample inlet (SI), sample outlet (SO), buffer inlet (BI)
and buffer outlet (BO).
Reaction conditions: Starting materials were purchased from Sigma–Al-
drich (Steinheim, Germany) and used without further purification. Dopa-
mine and epinine were used as hydrochlorides. For enantioselective syn-
thesis the hydrochloride was neutralized by adding NaHCO3-solution fol-
lowed by isolation of the free amine, which was verified by thin layer
chromatography and NMR. Racemic on-chip conversions were carried
out for 60 s in phosphate buffer (100 mmolLÀ1, pH 8). Due to acetalde-
hydeꢀs boiling point below room temperature the aldehyde components
were used in excess (>55 equiv) in all syntheses. Experiments without
chiral separation were carried out using 3.25 mmolLÀ1 of amine com-
pound. In experiments with chiral separation solutions with an amine
concentration of 0.65 mmolLÀ1 were utilized. Monitoring of the reaction
time course (Figure 4) was accomplished by conducting an in-vial reac-
tion for 60 s using 6.50 mmolLÀ1 dopamine and injecting the reaction
mixture every 30 s. For enantioselective synthesis (Figure 5) all reactants
were dissolved in iPrOH/tBuOH/2-Me-2-BuOH (1:1:1). Chiral catalysis
was performed by adding 10 mol% (R)-TRIP to the 6.50 mmolLÀ1
amine solution. The catalyst has been characterized previously.[24] The in-
vial reaction was carried out for 60 s followed by 1:1 dilution with separa-
tion buffer and subsequent injection of the mixture. The reaction mixture
was merged through the reaction channel by applying 500 VcmÀ1. Prod-
uct identities of all reactions were verified by ESI-MS (see the Support-
ing Information).
Figure 5. On-chip chiral catalysis. Electropherogram obtained after enan-
tioselective Pictet–Spengler reaction of dopamine with acetaldehyde
using a chiral organocatalyst.
tiomeric excess was determined in five experiments yielding
30% ee for the R-enantiomer. The method was quite robust
as indicated by a standard deviation of less than 0.7% (n=
5). The ee-values determined on-chip were in good agree-
ment with those obtained in bench-scale synthesis and anal-
ysis with classical capillary electrophoresis. Although the ee-
values are rather moderate, these first data show that the
system is well suited for high-throughput screening of enan-
tioselective catalysts.
Conversion percentages were calculated via peak area ratios of products
and starting material corrected by a factor that reflects the differences in
quantum yield. This parameter was determined by microchip electropho-
resis using phosphate buffer (40 mmolLÀ1, pH 3.0) and equimolar con-
centrations of dopamine and salsolinol. The resulting five point calibra-
tion plot (n=4) within a concentration range of 1.5 mmolLÀ1 and 9.0
mmolLÀ1 yielded in a multiplier of 1.66 (salsolinol/dopamine) with a rela-
tive standard deviation of less than 1.2% and no correlation to the con-
centration. Standard deviations for resulting percentage conversions were
less than 8% in all cases.
Conclusion
In summary, we have developed a synthesis and analysis
tool based on the integration of chemical synthesis and anal-
ysis on a single microchip. The feasibility of the system was
demonstrated by monitoring the on-chip formation of tetra-
hydroisoquinoline derivatives by Pictet–Spengler condensa-
tion. The reaction products were subsequently separated by
chiral microchip electrophoresis on the same chip resulting
in rather short process times of only a few minutes. For
peak assignment we applied on-the-fly deep UV time-
Microchip electrophoresis: Chip electrophoresis was performed using a
bipolar four-channel high-voltage power supply (model HCV 40M-10000,
FuG Elektronik GmbH, Rosenheim, Germany) providing up to
+
/À 10 kV per channel. The electrical contact to the microfluidic structure
was established via a homemade PMMA plate with integrated platinum
electrodes. The chip was mounted within a homemade low conductivity
carrier plate on the x,y-translational stage of the microscope. Prior to
electrophoresis the chip was rinsed with 0.1m NaOH, deionized water
and the respective separation buffer. For chip preparation, the whole mi-
crofluidic structure was filled with separation buffer. Thereafter the
ACHTUNGTRENNUNGresolved fluorescence detection allowing the differentiation
of analytes according to their native fluorescence lifetimes.
Moreover, it was demonstrated that the system is not limit-
ed to aqueous reaction media and is applicable to complex
catalytic transformations. This was shown exemplarily by
1244
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1240 – 1246

Monitoring On-Chip Pictet–Spengler Reactions
FULL PAPER
buffer in SI was substituted with sample solution. Afterwards the correct
position of the microfluidic chip was ensured and the desired detection
point within the channel was optically focused. Electrophoretic separa-
tions were performed using phosphate buffer (40 mmolLÀ1, pH 3.0) and
an effective field strength of 315 VcmÀ1. For chiral separations b-cyclo-
dextrin (12 mmolLÀ1) was added as chiral selector. In experiments for
monitoring the reaction time course and for enantioselective synthesis an
applied field strength of 296 VcmÀ1 was used. Effective separation
lengths were 7.3 cm and 5.6 cm when working with pinched and gated in-
jection, respectively. Migration order of enantiomers was determined ac-
cording to literature data.[25]
911–915; Angew. Chem. Int. Ed. 2010, 49, 6195–6198; k) A. J. de-
Mello, R. C. Wootton, Nat. Chem. 2009, 1, 28–29.
[2] a) O. Trapp, S. K. Weber, S. Bauch, W. Hofstadt, Angew. Chem.
2007, 119, 7447–7451; Angew. Chem. Int. Ed. 2007, 46, 7307–7310;
b) C. Lang, U. Gartner, O. Trapp, Chem. Commun. 2011, 47, 391–
393; c) J. P. McMullen, M. T. Stone, S. L. Buchwald, K. F. Jensen,
Angew. Chem. 2010, 122, 7230–7234; Angew. Chem. Int. Ed. 2010,
49, 7076–7080; d) T. Noꢅl, S. Kuhn, A. J. Musacchio, K. F. Jensen,
S. L. Buchwald, Angew. Chem. 2011, 123, 6065–6068; Angew. Chem.
Int. Ed. 2011, 50, 5943–5946; e) K. Geyera, T. Gustafsson, P. H. See-
berger, Synlett 2009, 15, 2382–239; f) D. Webb, T. F. Jamison, Chem.
Sci. 2010, 1, 675–6801; g) M. Rasheed, T. Wirth, Angew. Chem.
2011, 123, 374–376; Angew. Chem. Int. Ed. 2011, 50, 357–358; h) W.
Verboom, Chem. Eng. Technol. 2009, 32, 1695–1701; i) M. Rasheed,
T. Wirth, Chim. Oggi 2011, 29, 53–55; j) B. Ahmed-Omer, J. C.
Brandt, T. Wirth, Org. Biomol. Chem. 2007, 5, 733–740; k) F.
Benito-Lopez, A. J. Mettelalj, R. J. M. Egberink, A. H. Velders,
R. M. Tiggelaar, H. J. G. E. Gardeniers, D. N. Reinhoudt, W. Ver-
boom, Chim. Oggi 2010, 28, 56–59; l) F. Costantini, E. M. Benetti,
D. N. Reinhoudt, J. Huskens, G. J. Vancso, W. Verboom, Lab Chip
2010, 10, 3407–3412.
[3] D. Janasek, J. Franzke, A. Manz, Nature 2006, 442, 374–380.
[4] D. Belder, Angew. Chem. 2009, 121, 3790–3791; Angew. Chem. Int.
Ed. 2009, 48, 3736–3737.
[5] D. Belder, M. Ludwig, L. W. Wang, M. T. Reetz, Angew. Chem.
2006, 118, 2523–2526; Angew. Chem. Int. Ed. 2006, 45, 2463–2466.
[6] a) Q. Li, S. Seeger, Appl. Spectrosc. Rev. 2010, 45, 12–43; b) P.
Schulze, D. Belder, Anal. Bioanal. Chem. 2009, 393, 515–525; c) P.
Schulze, M. Ludwig, F. Kohler, D. Belder, Anal. Chem. 2005, 77,
1325–1329.
Native time-resolved fluorescence detection: Deep UV fluorescence life-
time detection (Figure 3 and Table 1) was accomplished using a setup
based on the MicroTime 200 platform (PicoQuant, Berlin, Germany)
with an Olympus IX 71 microscope. As excitation source a Nd:YAG
laser (Cougar, Time-Bandwidth Products, Zꢄrich, Switzerland) was incor-
porated utilized in the fourth harmonic via two b-barium borate crystals
yielding in Gaussian output pulses of around 10 ps and 5 nJ at a repeti-
tion rate of 20 MHz and an operating wavelength of 266 nm. The laser
beam passed a half-wave plate and a polarizer before being expanded via
quartz lenses. The excitation light was reflected by a dichroic mirror and
focused onto the microfluidic channel by backfilling a 40x, 0.8 NA fused
silica objective lens (Partec, Mꢄnster, Germany). Emitted light was col-
lected by the same objective. After passing the dichroic mirror the colli-
mated emission was sent through several UV lenses and a confocal pin-
hole (diameter 150 mm) to a photon counting photomultiplier tube (PMA
165-N-M, PicoQuant, Berlin, Germany) synchronized to the laser fre-
quency. A Brightline HC 280/20 excitation filter (implemented at an
angle of around 108 to allow for 266 nm transmission), a Z266RDC di-
chroic mirror and a Schott DUG11x filter (bandpass 285–365 nm) was
used for emission filtering (ITOS, Mainz, Germany).
[7] K. Suhling, P. M. W. French, D. Phillips, Photochem. Photobiol. Sci.
2005, 4, 13–22.
Prior to a series of measurements the focal plane position was adjusted.
Therefore, the chip was filled with the compound of interest and the mi-
crochannel was focused through the microscope eyepiece. The detected
fluorescence signal was monitored as a function of the z-position in order
to account for the focal plane shift in the UV. The position yielding maxi-
mal fluorescence intensity was used in the subsequent electrophoretic ex-
periments.
[8] For an overview for fluorescence lifetime spectroscopy, see: a) F. V.
Bright, C. A. Munson, Anal. Chim. Acta 2003, 500, 71–104; for an
overview for lifetime-based imaging, see: b) J. W. Borst, A. J. W. G.
Visser, Meas. Sci. Technol. 2010, 21, 1–21; c) P. Herman, H.-J. Lin,
J. R. Lakowicz, in CRC Biomedical Photonics Handbook, 1 edition
(Ed: T. Vo-Dinh), CRC Press, New York, 2003, pp. 9/1–9/30;
d) C. Y. Dong, T. French, P. T. So, C. Buehler, K. M. Berland, E.
Gratton, Methods Cell Biol. 2003, 72, 431–464; e) R. M. Clegg, O.
Holub, C. Gohlke, Methods Enzymol. 2003, 360, 509–542.
[9] a) L. Zhu, W. J. Stryjewski, S. A. Soper, Anal. Biochem. 2004, 330,
206–218; b) S. D. Llopis, W. Stryjewski, S. A. Soper, Electrophoresis
2004, 25, 3810–3819.
The binning time interval for electropherogram construction was set to
10 ms corresponding to a data acquisition rate of 100 Hz. To discriminate
short-lived background fluorescence the integration frame was delayed
by 0.4 ns with respect to the laser pulse.
For steady-state native fluorescence detection (Figure 4 and Figure 5) we
utilized a setup described previously.[6c] Intrinsic fluorescence was in-
duced by a fiber-coupled frequency quadrupled Nd:YAG laser (type
FQSS266-Q, Crystal GmbH, Berlin, Germany) with 3 mW quasi-cw
power at 266 nm.
[10] Q. Li, S. Seeger, Anal. Chem. 2006, 78, 2732–2737.
[11] a) H. M. Striebel, P. Schellenberg, P. Grigaravicius, K. O. Greulich,
Proteomics 2004, 4, 1703–1711; b) M. Schꢄttpelz, C. Mꢄller, H. Neu-
weiler, M. Sauer, Anal. Chem. 2006, 78, 663–669.
[12] a) W. R. Zipfel, R. M. Williams, R. Christie, A. Y. Nikitin, B. T.
Hyman, W. W. Webb, Proc. Natl. Acad. Sci. USA 2003, 100, 7075–
7080; b) J. Balaji, R. Desai, S. Maiti, Microsc. Res. Tech. 2004, 63,
67–71; c) S. W. Botchway, A. W. Parker, R. H. Bisby, A. G. Crisosto-
mo, Microsc. Res. Tech. 2008, 71, 267–273.
Acknowledgements
[13] a) S. Nagl, P. Schulze, S. Ohla, R. Beyreiss, L. Gitlin, D. Belder,
Anal. Chem. 2011, 83, 3232–3238; b) S. Fritzsche, S. Ohla, P. Glaser,
D. S. Giera, M. Sickert, C. Schneider, D. Belder, Angew. Chem.
2011, 123, 9639–9642; Angew. Chem. Int. Ed. 2011, 50, 9467–9470.
[14] a) R. Deitrich, V. Erwin, Annu. Rev. Pharmacol. Toxicol. 1980, 20,
55–80; b) A. Pictet, T. Spengler, Ber. Dtsch. Chem. Ges. 1911, 44,
2030–2036; c) J. Stçckigt, A. P. Antonchick, F. Wu, H. Waldmann,
Angew. Chem. 2011, 123, 8692–8719; Angew. Chem. Int. Ed. 2011,
50, 8538–8564.
[15] V. Hessel, H. Lçwe, F. Schçnfeld, Chem. Eng. Sci. 2005, 60, 2479–
2501.
[16] S. C. Jacobson, R. Hergenrçder, L. B. Koutny, R. J. Warmack, J. M.
Ramsey, Anal. Chem. 1994, 66, 1107–1113.
Financial support by the Deutsche Forschungsgemeinschaft (Bonn, Ger-
many) is gratefully acknowledged.
[1] a) K. Mae, Chem. Eng. Sci. 2007, 62, 4842–4851; b) J. P. McMullen,
K. F. Jensen, Annu. Rev. Anal. Chem. 2010, 3, 19–42; c) A. Arora,
G. Simone, G. B. Salieb-Beugelaar, J. T. Kim, A. Manz, Anal. Chem.
2010, 82, 4830–4847; d) A. J. deMello, Nature 2006, 442, 394–402;
e) P. Watts, S. J. Haswell, Chem. Soc. Rev. 2005, 34, 235–246; f) S. J.
Haswell, Nature 2006, 441, 705–705; g) R. L. Hartmann, K. F.
Jensen, Lab Chip 2009, 9, 2495–2507; h) A. van den Berg, T. S. J.
Lammerink, Top. Curr. Chem. 1998, 194, 21–49; i) J. Baek, P. M.
Allen, M. G. Bawendi, K. F. Jensen, Angew. Chem. 2011, 123, 653–
656; Angew. Chem. Int. Ed. 2011, 50, 627–630; j) R. L. Hartmann,
J. R. Naber, S. L. Buchwald, K. F. Jensen, Angew. Chem. 2010, 122,
[17] H. A. Bates, J. Org. Chem. 1981, 46, 4931–4935.
[18] Z. Fa, G. Dryhurst, Bioorg. Chem. 1991, 19, 384–397.
Chem. Eur. J. 2012, 18, 1240 – 1246
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1245

D. Belder et al.
[19] a) G. Nelson, G. Patony, I. M. Warner, Appl. Spectrosc. 1987, 41,
1235–1238; b) F. V. Bright, G. C. Catena, J. Huang, J. Am. Chem.
Soc. 1990, 112, 1343–1346.
[20] Y. Wei, S. Wang, S. Shuang, C. Dong, Talanta 2010, 81, 1800–1805.
[21] S. C. Jacobson, L. B. Koutny, R. Hergenrçder, A. W. Moore, J. M.
Ramsey, Anal. Chem. 1994, 66, 3472–3476.
[22] a) H. Waldmann, G. Schmidt, H. Henke, M. Burkard, Angew. Chem.
1995, 107, 2608–2610; Angew. Chem. Int. Ed. Engl. 1995, 34, 2402–
2403; b) M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126,
10558–10559; c) J. Seayad, A. M. Seayad, B. List, J. Am. Chem. Soc.
2006, 128, 1086–1087; d) M. J. Wanner, R. N. S. van der Haas, K. R.
de Cuba, J. H. van Maarseveen, H. Hiemstra, Angew. Chem. 2007,
119, 7629–7631; Angew. Chem. Int. Ed. 2007, 46, 7485–7487;
e) N. V. Sewgobind, M. J. Wanner, S. Ingemann, R. de Gelder, J. H.
van Maarseveen, H. Hiemstra, J. Org. Chem. 2008, 73, 6405–6408;
f) F. R. Bou-Hamdan, J. L. Leighton, Angew. Chem. 2009, 121,
2439–2442; Angew. Chem. Int. Ed. 2009, 48, 2403–2406.
[23] a) T. Akiyama, Chem. Rev. 2007, 107, 5744–5758; b) M. Terada,
Chem. Commun. 2008, 4097–4112; c) M. Sickert, C. Schneider,
Angew. Chem. 2008, 120, 3687–3690; Angew. Chem. Int. Ed. 2008,
47, 3631–3634; d) D. S. Giera, M. Sickert, C. Schneider, Org. Lett.
2008, 10, 4259–4262.
[24] M. Sickert, F. Abels, M. Lang, J. Sieler, C. Birkemeyer, C. Schneid-
er, Chem. Eur. J. 2010, 16, 2806–2818.
[25] Z. Quan, Y. Song, G. Peters, M. Shenwu, Y. Sheng, H. M. Hwang,
Y. M. Liu, Anal. Sci. 2005, 21, 115–119.
Received: June 9, 2011
Revised: September 8, 2011
Published online: December 16, 2011
1246
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1240 – 1246