Resin bead micro-UV-visible absorption spectroscopy

Article abstract of DOI:10.1021/ac049319i

The construction and design of a microscope coupled with a miniature UV-vis spectrometer is described. This was applied to the study of dyes linked to solid supports and displayed good correlation in spectral shape and λ_max values when compared

Full text of DOI:10.1021/ac049319i

Anal. Chem. 2005, 77, 2247-2251
Resin Bead Micro-UV-Visible Absorption
Spectroscopy
Lu Shin Wong,† Fabrice Birembaut,†,‡ William S. Brocklesby,§ Jeremy G. Frey,† and Mark Bradley*,†
School of Chemistry, University of Edinburgh, Edinburgh EH9 3JJ, U.K., and School of Chemistry and Opto-Electronics
Research Centre, University of Southampton, Southampton SO17 1BJ, U.K.
Colored beads have been generated for a number of other
solid-phase and combinatorial reasons. Thus, a range of solid-
phase strategies have been developed for the discovery and
preparation of dyes,15,16 colored beads have also been produced
in a number of assays in which libraries have been screened for
specific metal binding,17 while host-guest chemistry has long
exploited the localization of a dye-labeled guest onto beads
containing a specific high-affinity host.18 More recently, resin-
bound indicators have been employed in monitoring the progress
of solid-phase chemistries.19,20
In the case of dye synthesis, the ability to conduct UV-vis
spectrometry on beads would enable a number of direct on-bead
analyses to be carried out without requiring the additional step
of isolating the chromophores. In the screening of receptors, on-
bead quantitative measurements of spectrometric intensity would
be a good indicator of the binding strength of the receptor. There
are several advantages to the use of dyes over fluorophores in
combinatorial screening; nonfluorescent dyes are generally cheaper
and available in a wider variety of spectral and structural
characteristics. Additionally, quantitative absorbance information
of resin-bound dyes would be a step in the application of these
materials as miniature sensor devices. Apart from this, the
investigation of the colorimetric properties of indicator dyes on
solid supports would provide insights into the physicochemical
microenvironment within beads.
The construction and design of a microscope coupled with
a miniature UV-vis spectrometer is described. This was
applied to the study of dyes linked to solid supports and
displayed good correlation in spectral shape and λmax
values when compared to the dyes in solution, as well as
showing a linear relationship between dye loading and
UV-vis absorbance. The spectral profiles of these dyes
at various pH’s were measured and used to determine
the pKa of the dyes on the beads, which were compared
with the pKa values of the dyes in solution, thus enabling
the dye-loaded beads to act as pH sensors.
The expansion of combinatorial chemistry has vastly increased
the development of organic chemistry on solid supports,1 and
concurrently, a number of techniques have been developed for
the analysis of compounds on these supports.2-5 In the area of
optical microspectroscopy, this includes Fourier transform infrared
(FT-IR) microscopy2,6 and single-photon, single-photon confocal,7-9
and two-photon10,11 fluorescence microscopy, as well as Raman
confocal microscopy,12-14 which have allowed the spatial mapping
of the interior of a variety of supports, providing information
regarding the functional site distribution within beads.
* Corresponding author. Telephone: +44 (0) 131 650 7546. E-mail:
chemistry@ed.ac.uk.
† School of Chemistry, University of Southampton.
‡ Current address: School of Chemistry, University of Reading, Reading, U.K.
§ Opto-Electronics Research Centre, University of Southampton.
(1) Nicolaou, K. C.; Hanko, R.; Hartwig, W. In Handbook of Combinatorial
Chemistry: Drugs, Catalysts, Materials; Nicolaou, K. C., Hanko, R., Hartwig,
W., Eds.; Wiley-VCH: Weinheim, Germany, 2002; pp 3-9.
(2) Gremlich, H.-U. Biotechnol. Bioeng. 1999, 61, 179-187.
(3) Dal Cin, M.; Davalli, S.; Marchioro, C.; Passarini, M.; Perini, O.; Provera,
S.; Zaramella, A. Farmaco 2002, 57, 497-510.
In this paper, the construction and design of a microscope used
in conjunction with a miniature UV-vis spectrometer is reported,
and initial testing and application in the study of dye and sensors
linked to solid supports is described.
MATERIALS AND METHODS
(4) Gallop, M. A.; Fitch, W. L. Curr. Opin. Chem. Biol. 1997, 1, 94-100.
(5) Egner, B. J.; Bradley, M. Drug Discovery Today 1997, 2, 102-109.
(6) Haap, W. J.; Walk, T. B.; Jung, G. Angew. Chem., Int. Ed. 1998, 37, 3311-
3314.
(7) Rademann, J.; Barth, M.; Brock, R.; Egelhaaf, H.-J.; Jung, G. Chem. Eur. J.
2001, 7, 3884-3889.
(8) McAlpine, S. R.; Schreiber, S. L. Chem. Eur. J. 1999, 5, 3528-3532.
(9) Moss, M. C.; Veiro, J. A.; Singleton, S.; Gregory, D. P.; Birmingham, J. J.;
Jones, C. L.; Cummins, P. G.; Cummins, D.; Miller, R. M.; Sheppard, R. C.;
Howard, V. C.; Bhaskar, N. Analyst 1993, 118, 1-9.
UV-Vis Microscope Construction. A photograph and sche-
matic plan of the combined microscope and UV-vis spectrometer
system are displayed in Figure 1. The light source for the
microscope and spectrometer was a Mini-D2T deuterium-
tungsten fiber optic light source (Ocean Optics, Dunedin, FL) and
the deuterium lamp had a spectral range of 200-410 nm, while
(10) Farrer, R. A.; Copeland, G. T.; Previte, M. J. R.; Okamoto, M. M.; Miller, S.
J.; Fourkas, J. T. J. Am. Chem. Soc. 2002, 124, 1994-2003.
(11) Ulijn, R. V.; Brazendale, I.; Margetts, G.; Flitsch, S. L.; McConnell, G.; Girkin,
J.; Halling, P. J. J. Comb. Chem. 2003, 5, 215-217.
(15) Isacsson, J.; Westman, G. Tetrahedron Lett. 2001, 42, 3207-3210.
(16) Nguyen, T.; Francis, M. B. Org. Lett. 2003, 5, 3245-3248.
(17) Francis, M. B.; Finney, N. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1996,
118, 8983-8984.
(12) Kress, J.; Rose, A.; Frey, J. G.; Brocklesby, W. S.; Ladlow, M.; Mellor, G.
(18) Weingarten, M. D.; Sekanina, K.; Still, W. C. J. Am. Chem. Soc. 1998, 120,
9112-9113.
W.; Bradley, M. Chem. Eur. J. 2001, 7, 3880-3883.
(13) Kress, J.; Zanaletti, R.; Amour, A.; Ladlow, M.; Frey, J. G.; Bradley, M. Chem.
Eur. J. 2002, 8, 3769-3772;
(19) Cho, J. K.; White, P. D.; Klute, W.; Dean, T. W.; Bradley, M. J. Comb. Chem.
2003, 5, 632-636.
(14) Kress, J.; Zanaletti, R.; Rose, A.; Frey, J. G.; Brocklesby, W. S.; Ladlow, M.;
(20) Cho, J. K.; White, P. D.; Klute, W.; Dean, T. W.; Bradley, M. Chem. Commun.
Bradley, M. J. Comb. Chem. 2003, 5, 28-32.
2004, 502-503.
10.1021/ac049319i CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/02/2005
Analytical Chemistry, Vol. 77, No. 7, April 1, 2005 2247

Figure 1. Schematic diagram of the custom-built UV-vis microspectrometer (left) and photograph (right).
the tungsten lamp covered 360-1100 nm (ignition delay < 2 s).
A fused silica lens (40 mm focal length) was placed 80 mm from
the light source, and the sample stage was located 80 mm above
the lens. Above this was placed the objective lens, an OFR LMU-
40×-NUV (numerical aperture 0.50) with a working distance of 1
mm and an effective focal length of 5 mm. The beam exiting the
objective lens was then either refocused on to a fiber-optic cable
leading to the entrance slit of a USB2000-UV-vis spectrometer
(spectral range, 200-850 nm; typical overall spectral resolution,
1.5 nm; from Ocean Optics Inc.) or diverted to a closed circuit
television camera using a flipping mirror. Another fused silica lens
with a focal length of 40 mm was used to refocus the beam onto
the fiber-optic cable. The magnification at the detector end was
×8 and at the camera end, ×18 while the microscope had a
Figure 2. UV-vis transmission spectra of the unloaded beads: (a)
sampling area with a diameter of 25 µm. The spectrometer was
connected to a PC which used OOIBase32 software (ver. 1.0.3,
Ocean Optics Inc.) configured to measure UV-vis transmission
intensity. The optic fibers used throughout had core diameters
of 200 µm.
glass slide only, (b) polystyrene dry, (c) TentaGel dry, (d) DMF on
glass slide, (e) polystyrene swollen in DMF, and (f) TentaGel swollen
in DMF.
microscope slide well and the well filled with the solution. The
measured transmittance spectra were then converted and further
processed with the statistical software package Origin version 6.1
(OriginLab, Northampton, MA) to produce the absorbance spectra
using Beer’s law (Abs ) -log(I/Io) ) ꢀcl). In the calculations,
the concentration of the dye within the beads was used (c, was
calculated from the known loading and volume of the beads) and
the diameter of the bead was used for l.
Materials and Equipment. Aminomethylpolystyrene (PS;
PolymerLabs, Church Stretton, U.K.), TentaGel-S Amino21 (TG;
Rapp Polymere, Tu¨bingen, Germany), aminopropyl-controlled pore
glass (CPG; VitraBio, Steinach, Germany), p-methyl red (Tokyo
Kasei Kogyo, Tokyo, Japan), N-Fmoc-3-nitrotyrosine (Advanced
ChemTech, Louisville, KY), alizarin yellow GG (Sigma-Aldrich,
Gillingham, U.K.), and mordant orange 1 (Sigma-Aldrich) were
all used as supplied. (4-Carboxyphenyl)bromophenol blue19 was
donated by Dr. J. K. Cho (Centre for Combinatorial Excellence,
University of Southampton). Dyes were coupled to the beads using
standard amide coupling conditions. Full details of all the
experimental procedures are given in the Supporting Information.
Solid-Phase UV-Vis Spectra Measurements. The acquisi-
tion time used was 1 s, and each processed spectrum was the
average of eight different beads. The beads were soaked in the
solution of interest for at least 1 h before being placed in a
RESULTS AND DISCUSSION
Preliminary Testing of the UV-Vis Microspectrometry
System. Initially, the spectra of two types of beads were measured,
aminomethyl-functionalized PS and poly(ethylene glycol) grafted
polystyrene (TentaGel, TG). Using dry beads, reduced transmit-
tance was observed since the opaque beads scattered the light.
When swollen in solvent, beads typically swell up to 5 times their
initial volume (volume of the swollen bead is ∼90% solvent), and,
as can be seen, the beads became “optically transparent” and
allowed the transmission of light (Figure 2). Therefore, the work
(21) Rapp, W. Methods Org. Chem. (Houben-Weyl), 4th ed.; Thieme: Stuttgart,
Germany, 2002; Vol. E22a, pp 672-684.
2248 Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

Figure 3. UV-vis absorption spectra of p-methyl red in 1.6 M HCl in 1,4-dioxane: (a) on PS (7% loading), (b) in solution in the microspectrometer
(100 µM, 3 mm cell), and (c) in solution on a conventional spectrometer (25 µM, 10 mm cell).
presented here was carried out using only swollen beads in
solvents.
supports was then examined and compared with the dyes in
solution, (4-carboxyphenyl)bromophenol blue19 on TG, alizarin
yellow GG on TG, 3-nitrotyrosine on CPG, and mordant orange 1
on TG, although larger differences in both the shape and λmax of
the spectra were observed. (See Supporting Information for full
spectra.) Overall, the new system achieved good sensitivity with
spectra being measured from beads with 10% or less of the
functional sites loaded. A step further was then taken toward
quantitative analyses.
An interesting feature observed in the initial testing was the
apparent increase in the intensity of transmitted light through the
swollen beads compared to the light beam passing through solvent
alone. Above 400 nm there is approximately a 3-fold increase in
the transmitted light due to the spherical beads acting as a lens
and focusing the light into the collecting objective. Below 400 nm
the absorption of the polymer support itself (or more accurately,
the polymer and trace impurities within it from its manufacturing
processes) reduces the transmitted light; this is a limitation
inherent to working with polymer supports. A plot of the
transmitted intensity through the bead in dimethylformamide
(DMF) against the intensity through DMF alone for both PS and
TG beads gave a near-linear relationship between 400 and 900
nm (with a slope of ∼3; see Supporting Information for plots).
Solvents commonly used in solid-phase organic synthesis were
evaluated to determine the extent of this effect and if it was
possible to eliminate it. However, complete matching of the
transmittance through the bead and through solvent alone was
not possible using standard solvents. Depending on the optical
properties of the solvent and its effect on the swelling of the resin,
the lens effect was apparent with an increase in transmittance or
with “poor” solvents; transmittance was reduced. To compensate
for this effect, measurements were performed on swollen undyed
beads and used as the reference. (See Supporting Information
for the list of solvents used and results obtained.)
Qualitative UV-Vis. Spectra of Solid-Supported Dyes. Initially
p-methyl red (Dabcyl) was coupled to amino-functionalized PS and
TG resins and the UV-vis spectra of the beads were measured
and compared with the dye in solution (Figure 3). The overall
shapes of the spectra for the dye in solution and attached to the
resins were observed to be similar although there was some
alteration in the value of λmax. This change may be the result of
alteration of the chromophore by its attachment to the solid phase
since this involved the conversion of the carboxylic acid to an
amide. Additionally, the physicochemical environment around the
chromophore was altered due to the presence of the polymer from
the bead. A range of structurally diverse chromophores on various
Study of Dye Loaded Beads at Various Loadings. UV-vis spectra
of PS beads, coupled with varying amounts of p-methyl red, were
recorded and absorbance was plotted as a function of loading
(Figure 4). Attempts were then made to quantify, in absolute
terms, the relationship between loading and absorbance by
applying the Beer-Lambert Law. The actual amount of dye
coupled to the resins could not be accurately measured by either
elemental analysis or other well-known tests due to the low loading
levels involved (1-10%). To address this issue, the amount of
coupled dye was determined indirectly through UV-vis analysis
of the reaction solutions. Thus, for selected samples, the amount
of dye in the reaction was measured pre- and postcoupling, and
the difference was then used to calibrate Figure 4. By applying
the Beer-Lambert law, ꢀmax for the dye attached to the beads
was calculated to be ∼7700 L mol-1 cm-1. There was a similar
linear relationship between loading and absorbance for (4-
carboxyphenyl)bromophenol blue with a calculated ꢀmax ∼ 8500
L mol-1 cm-1. (See Supporting Information for full spectra.)
The values of ꢀmax calculated were significantly lower than those
of the solution-phase analogues for both dyes (p-methyl red, 18 266
at 548 nm; (4-carboxyphenyl)bromophenol blue, 21 233 at 595 nm;
in the same solvents as used for the beads). The reason for this
is unclear, but ꢀmax may be affected by the lower polarity of the
surrounding environment within the polymer matrix and the high
concentration of dye within the bead and may also be the result
of the previously mentioned “lensing” optical effect on the beads.
UV-Vis. Profiling of pH Indicator Dyes. p-Methyl red and
bromophenol blue are used as pH indicators, and the pKa of a
dye in solution can be determined by UV-vis spectrometry.22 To
test the applicability of this on the solid-supported indicators, the
Analytical Chemistry, Vol. 77, No. 7, April 1, 2005 2249

Figure 4. UV-vis absorbance spectra of p-methyl red PS beads at various loadings in 1M aqueous HCl in 1,4-dioxane: (left) with nominal
loadings of (a) 10, (b) 9, (c) 8, (d) 7, (e) 6, (f) 5, (g) 4, (h) 3, (i) 2, and (j) 1%. Linear fit of absorbance vs loading (right) and samples quantified
by analysis of reaction solutions (×).
Figure 5. UV-vis absorption spectra of (4-carboxyphenyl)bromophenol blue beads in aqueous solutions (left) at various pH’s: (a) 2, (b) 3, (c)
3.5, (d) 4, (e) 4.5, (f) 5, (g) 5.5, and (h) 6. Sigmoidal fit of absorbance vs pH (right).
spectra of (4-carboxyphenyl)bromophenol blue loaded on Tenta-
Gel (10% nominal loading) was measured at various pH’s. The
intensities of the two peaks corresponding to the acid and base
forms of the dye were measured and plotted against pH with the
best-fit Boltzmann sigmoidal curve (Figure 5).
From these plots, the pH at the midpoint between the
maximum and minimum absorbance intensity (x0) was deter-
mined. Thus, the pKa from Figure 8 was estimated to be between
4.2 (based on measurements at 440 nm) and 4.6 (based on
measurements at 607 nm).
A similar experiment was conducted with TentaGel-linked
p-methyl red (10% loading). The spectra in this case were more
difficult to interpret due to the close overlap of the two possible
absorbance bands so that in order to extract the desired informa-
tion from these overlapping bands, deconvolution calculations
were necessary. Using a model of two Gaussian peaks, the
calculated maxima for these deconvoluted peaks (at 480 and 520
nm, respectively) were then plotted in a manner similar to the
bromophenol blue derivative. (See Supporting Information for full
spectra.)
For comparison, the spectra of a solution of (4-carboxyphenyl)-
bromophenol blue were recorded at various pH’s and the pKa
values obtained (3.94 for the peak at 440 nm and 4.29 for the peak
at 594 nm) were very similar to literature values for unmodified
bromophenol blue in solution.22 Both were somewhat lower than
for the resin-bound indicator, signifying that the deprotonation
of the indicator on TentaGel required more basic conditions.
From the plots, the estimated pKa values in this case were 0.55
(from 520 nm) and 0.64 (from 480 nm). This was confirmed by
visual observations of the color changes in this pH range. This
differed significantly from the known pKa of 1.78 for p-methyl red
in solution.23 To determine if amide bond formation may have been
the cause of this, p-methyl red was coupled to methylamine. The
solution-phase spectra of this modified indicator were then
(22) Escolar, D.; Haro, M. R.; Saucedo, A.; Ayuso, J.; Jimenez, A.; Alvarez, J. A.
(23) S¸eclaˇman, E.; Sallo´, A.; Elenes¸, F.; Craˇs¸maˇreanu, C.; Wikete, C.; Timofei,
S.; Simon, Z. Dyes Pigm. 2002, 55, 69-77.
Appl. Spectrosc. 1996, 50, 1290-1294.
2250 Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

Figure 6. UV-vis absorbance at 550 nm plotted against time for three sample beads.
measured and the pKa determined to be 1.66, which was again
significantly different from the value for the solid-phase version.
This implies that much higher H3O+ concentrations are needed
to protonate the indicator moieties that are within the polymer
matrix of TentaGel. This in turn could be related to the lower
polarity of the matrix compared to a purely aqueous environment
and the resulting lowered stability of the ionized forms of the
indicators. Additionally, the high local concentration of indicator
moieties within the bead and repulsion between the charged
species may also affect their likelihood of protonation.
Changes in pKa due to the altered environment caused by
different solvents are well-known,24,25 and changes due to the
incorporation of indicators into various sol-gel matrixes have been
described.26-28 It would appear that alterations of pKa can also be
as a result of the local environment within a polymeric solid
support commonly used in organic synthetic chemistry. Interest-
ingly, such a great effect on pKa was not observed with supported
bromophenol blue when compared to p-methyl red.
maximum absorbance, k ) rate constant, t ) time, and t0 ) time
when A is zero).
Overall, the data fitted very well with the model and the rate
constant k demonstrated good reproducibility despite differences
in bead size with an average value of 0.0065 s-1, confirming that
a reaction time of 5 min was sufficient for the positive determi-
nation of amines on-bead since the majority of the sites reacted
within this time.
CONCLUSIONS
The UV-vis microspectrometer has proven to be a useful tool
for the quantitative and qualitative measurement of a wide range
of dyes on solid supports. Additionally, real-time rate measure-
ments could also be conducted. This system can be employed
regardless of the refractive properties of the solvent(s) or polymer
support used and, in this respect, avoids the need to find index-
matched solvents for each sample prior to analysis.
This spectrometer was used to investigate the effect of
immobilization on the pKa of indicator dyes, and significant
differences were observed as a result of the attachment of the
dyes to the beads. In both cases, ionization of the dyes required
more extreme conditions, lower pH for protonation and higher
for deprotonation. Nevertheless, these solid-supported indicator
dyes continued to act as indicators, allowing beads to act as
sensors of pH.
On-bead UV-vis microspectrometry can now be expanded to
the study of solid-supported chromophores since the spectrometer
is able to give quantitative values for colorimetric data which would
have previously been simple qualitative (visual) examinations.
Thus, a wide variety of chromophores such as sensors for pH or
cations, or probes for various structural motifs on the solid-support,
can now be explored.
On-Bead Colorimetric Reaction Kinetics. The use of the
microspectrometer as a tool for real-time reaction monitoring was
also investigated since its solid-state design and associated
software enables spectra to be measured at a higher rate compared
to split-beam spectrometers. Here, the chloranil test for secondary
amines was studied where blue beads are produced in a pale
yellow solution.29 Proline, attached to 2-chlorotrityl-functionalized
PS, was reacted with acetaldehyde, and a UV-vis spectrum of a
single bead was measured every 1.0 s after addition of chloranil,
and the absorbance at 550 nm was plotted against time (Figure
6). The λmax value was not used in order to avoid saturating the
spectrometer and to maintain linearity. Since the amount of
acetaldehyde and chloranil used in each reaction was in excess,
the only limiting factor was the number of amine sites on the
beads, a pseudo-first-order reaction model could be applied. (At
SUPPORTING INFORMATION AVAILABLE
) Amax[1 - e-k(t-t0)], where A ) absorbance at time t, Amax
)
Table of relative signal intensity changes of beads in various
solvents, detailed experimental procedures for the synthesis of
dye-linked beads, p-methyl red methyl amide, and analysis of dyes
and beads, and UV-vis spectra for solid-supported dyes, indicator
dyes in solution, and pKa measurements. This material is available
free of charge via the Internet at http://pubs.acs.org.
(24) Lambert, J. B.; Shurvell, H. F.; Lightner, D. A.; Cooks, R. G. Organic
Structural Spectroscopy; Prentice Hall: Upper Saddle River, NJ, 1998; pp
274-303.
(25) Gonzalez, A. G.; Herrador, M. A.; Asuero, A. G. Anal. Chim. Acta 1991,
246, 429-434.
(26) Rottman, C.; Grader, G.; De Hazan, Y.; Melchior, S.; Avnir, D. J. Am. Chem.
Soc. 1999, 121, 8533-8543.
Received for review May 8, 2004. Accepted December 1,
2004.
(27) Rottman, C.; Anvir, D. J. Am. Chem. Soc. 2001, 123, 5730-5734.
(28) Liang, C.; Weaver, M. J.; Dai, S. Chem. Commun. 2002, 1620-1621.
(29) Vojkovsky, T. Pept. Res. 1995, 8, 236-237.
AC049319I
Analytical Chemistry, Vol. 77, No. 7, April 1, 2005 2251