Synthesis, photophysical behavior, and electronic structure of push - Pull purines

Article abstract of DOI:10.1021/ja806348z

"Push-pull" purines have been synthesized by the introduction of electron-accepting functional groups (A = CN, CO₂Me, and CONHR) to the heterocyclic C(8) position to complement typical electron-donating substituents at C(2) (D¹) and

Full text of DOI:10.1021/ja806348z

Published on Web 12/29/2008
Synthesis, Photophysical Behavior, and Electronic Structure
of Push-Pull Purines
Roslyn S. Butler, Pamela Cohn, Phillip Tenzel, Khalil A. Abboud, and
Ronald K. Castellano*
Department of Chemistry, UniVersity of Florida, P.O. Box 117200, GainesVille, Florida 32611-7200
Received August 18, 2008; E-mail: castellano@chem.ufl.edu
Abstract: “Push-pull” purines have been synthesized by the introduction of electron-accepting functional
groups (A ) CN, CO2Me, and CONHR) to the heterocyclic C(8) position to complement typical electron-
donating substituents at C(2) (D1) and C(6) (D2). The donor-acceptor purines show significantly altered,
and overall improved photophysical properties relative to their acceptor-free precursors (A ) H); these
include red-shifted (20-50 nm) absorption maxima, highly solvatochromic emission profiles (em λmax from
355-466 nm depending on substitution pattern and solvent) with excellent linear correlations between
emission energy and solvent polarity (ETN), improved photochemical stability upon continuous irradiation,
and enhanced (up to 2500%) fluorescence quantum yields. Comprehensive structure-property studies
show how the absorption/emission maxima and quantum yields depend on donor and acceptor structure,
relative donor position (C(2) or C(6)), and solvent (1,4-dioxane, dichloromethane, acetonitrile, methanol,
and in some cases water). Further insight regarding electronic structure comes from a quantitative treatment
of the solvent-dependent emission data (that provides ∆µge values ranging from 1.9 to 3.4 D) and DFT
(B3LYP/6-311++G**) electronic structure calculations. X-ray crystal structures of several derivatives
showcase the molecular recognition capabilities of the donor-acceptor chromophores that overall have
photophysical and structural properties suitable for applications in biosensing and materials.
Unlike other heterocycles,6 some with biological relevance (e.g.,
porphyrins7), the nucleobases have sparsely been considered for
Introduction
Simple nucleobase heterocycles are attractive as “functional”
π components in organic materials1 wherein their built-in
molecular recognition features can be brought to bear on
solution-phase, solid-state, and surface architectures,2,3 and even
bulk transport or magnetic properties and device performance.4,5
materials applications that rely on their photophysical proper-
ties.8 This is in part due to the exceedingly low fluorescence
quantum yields (ΦF) of the purines and pyrimidines; values on
the order of 0.01% in water are typical.9 Remarkable improve-
ments to the intrinsic fluorescence properties of nucleobases
have, however, been achieved for biological applications,10
where designer molecules routinely play a role in sensing and
reporting.11,12 Some recent examples that are particularly
(1) For relevant reviews see:(a) Sessler, J. L.; Lawrence, C. M.; Jayaw-
ickramarajah, J. Chem. Soc. ReV. 2007, 36, 314–325. (b) Davis, J. T.;
Spada, G. P. Chem. Soc. ReV. 2007, 36, 296–313. (c) Weck, M. Polym.
Int. 2007, 56, 453–460. (d) Hoeben, F. J. M.; Jonkheijm, P.; Meijer,
E. W.; Schenning, A. Chem. ReV. 2005, 105, 1491–1546. (e) Sivakova,
S.; Rowan, S. J. Chem. Soc. ReV. 2005, 34, 9–21. (f) Araki, K.;
Yoshikawa, I. Top. Curr. Chem. 2005, 256, 133–165.
(4) (a) Lena, S.; Brancolini, G.; Gottarelli, G.; Mariani, P.; Masiero, S.;
Venturini, A.; Palermo, V.; Pandoli, O.; Pieraccini, S.; Samor`ı, P.;
Spada, G. P. Chem. Eur. J. 2007, 13, 3757–3764. (b) Maruccio, G.;
Visconti, P.; Arima, V.; D’Amico, S.; Blasco, A.; D’Amone, E.;
Cingolani, R.; Rinaldi, R.; Masiero, S.; Giorgi, T.; Gottarelli, G. Nano
Lett. 2003, 3, 479–483. (c) D’Amico, S.; Maruccio, G.; Visconti, P.;
D’Amone, E.; Cingolani, R.; Rinaldi, R.; Masiero, S.; Spada, G. P.;
Gottarelli, G. Microelectron. J. 2003, 34, 961–963. (d) Rinaldi, R.;
Maruccio, G.; Biasco, A.; Arima, V.; Cingolani, R.; Giorgi, T.;
Masiero, S.; Spada, G. P.; Gottarelli, G. Nanotechnology 2002, 13,
398–403. (e) Rinaldi, R.; Branca, E.; Cingolani, R.; Masiero, S.; Spada,
G. P.; Gottarelli, G. Appl. Phys. Lett. 2001, 78, 3541–3543.
(2) Representative recent examples:(a) Numata, M.; Sugiyasu, K.; Kishida,
T.; Haraguchi, S.; Fujita, N.; Park, S. M.; Yun, Y. J.; Kim, B. H.;
Shinkai, S. Org. Biomol. Chem 2008, 6, 712–718. (b) Jatsch, A.;
Kopyshev, A.; Mena-Osteritz, E.; Ba¨uerle, P. Org. Lett. 2008, 10, 961–
964. (c) Yoshikawa, I.; Sawayama, J.; Araki, K. Angew. Chem., Int.
Ed. 2008, 47, 1038–1041. (d) Spada, G. P.; Lena, S.; Masiero, S.;
Pieraccini, S.; Surin, M.; Samor`ı, P. AdV. Mater. 2008, 20, 2433–
2438. (e) Mather, B. D.; Baker, M. B.; Beyer, F. L.; Berg, M. A. G.;
Green, M. D.; Long, T. E. Macromolecules 2007, 40, 6834–6845. (f)
´
Arnal-Herault, C.; Barboiu, M.; Pasc, A.; Michau, M.; Perriat, P.; van
(5) (a) Murata, T.; Saito, G.; Enomoto, Y.; Honda, G.; Shimizu, Y.; Matsui,
S.; Sakata, M.; Drozdova, O. O.; Yakushi, K. Bull. Chem. Soc. Jpn.
2008, 81, 331–344. (b) Tanaka, H.; Shiomi, D.; Ise, T.; Sato, K.; Takui,
T. CrystEngComm 2007, 9, 767–771.
der Lee, A. Chem. Eur. J. 2007, 13, 6792–6800. (g) Sreenivasachary,
N.; Hickman, D. T.; Sarazin, D.; Lehn, J.-M. Chem. Eur. J. 2006, 12,
8581–8588. (h) Spijker, H. J.; Dirks, A. J.; van Hest, J. C. M. J. Polym.
Sci., Part A: Polym. Chem. 2006, 44, 4242–4250. (i) Park, T.;
Zimmerman, S. C.; Nakashima, S. J. Am. Chem. Soc. 2005, 127, 6520–
6521. (j) Norsten, T. B.; Jeoung, E.; Thibault, R. J.; Rotello, V. M.
Langmuir 2003, 19, 7089–7093.
(6) Forrest, S. R.; Thompson, M. E. Chem. ReV. 2007, 107, 923–925.
(7) See, for example: Barker, C. A.; Zeng, X.; Bettington, S.; Batsanov,
A. S.; Bryce, M. R.; Beeby, A. Chem. Eur. J. 2007, 13, 6710–6717.
(8) Kim, S. J.; Kool, E. T. J. Am. Chem. Soc. 2006, 128, 6164–6171.
(9) (a) Andre´asson, J.; Holme´n, A.; Albinsson, B. J. Phys. Chem. B 1999,
103, 9782–9789. (b) Wilson, R. W.; Callis, P. R. Photochem.
Photobiol. 1980, 31, 323–327. (c) Ward, D. C.; Reich, E.; Stryer, L.
J. Biol. Chem. 1969, 244, 1228–1237. (d) Drobnik, J.; Augenstein, L.
Photochem. Photobiol. 1966, 5, 13–30.
(3) (a) Kumar, A. M. S.; Sivakova, S.; Marchant, R. E.; Rowan, S. J.
Small 2007, 3, 783–787. (b) Sivakova, S.; Wu, J.; Campo, C. J.;
Mather, P. T.; Rowan, S. J. Chem. Eur. J. 2006, 12, 446–456. (c)
Sivakova, S.; Bohnsack, D. A.; Mackay, M. E.; Suwanmala, P.;
Rowan, S. J. J. Am. Chem. Soc. 2005, 127, 18202–18211.
9
10.1021/ja806348z CCC: $40.75
2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 623–633 623

A R T I C L E S
Butler et al.
relevant to the current work show how extension of the
nucleobase conjugated system by even a vinyl group12i or
aromatic ring11e,f,12d is sufficient to produce highly emissive
species that are also capable of base pairing.
2-Aminopurine (2-AP), an isomer of adenine, remains perhaps
the simplest functional nucleobase variant that is also fluores-
cent; a high quantum yield (68% in water for the riboside)9c
underlies its widespread usage as an optical probe13 and
continued mechanistic and structural study.14 Even so, molecules
based on this platform have sparsely filtered into traditional
organic materials and related sensing applications,8,15 and in
surprisingly few cases has their optimization and broader
photophysical evaluation (e.g., studies in nonaqueous solution
or the bulk) toward this goal been performed.
Figure 1. Generic structure of the donor-acceptor purines considered in
this work (D ) donor; A ) acceptor; Bn ) benzyl). Important atoms have
been numbered in the conventional way around the purine core.
upon introduction of acceptor substituents to the purine C(8)
position that complement typical donor groups at C(2) and C(6).
The arrangement affords increased quantum yields (in many
cases near unity) and significantly red-shifted absorption/
emission spectra in organic solution (and even water) relative
to the acceptor-free molecules; methyl ester functionalization
of 2-amino-9-benzylpurine, for example, imparts a 13-fold
quantum yield increase in 1,4-dioxane (other derivatives show
up to 25-fold increases). Photophysical studies in multiple
solvents, X-ray crystallography, and theoretical analysis show
how the nature (and position) of the donor and acceptor groups
contribute to the optical parameters, solid-state ordering, and
electronic structure of the purines.
Shown here is how the donor-π-acceptor design, commonly
used to tailor the optical and electronic properties of π-conju-
gated systems,16-19 can be usefully extended to 2-AP and related
purines (Figure 1).20 Significant photophysical changes occur
(10) For recent reviews see:(a) Venkatesan, N.; Seo, Y. J.; Kim, B. H.
Chem. Soc. ReV. 2008, 37, 648–663. (b) Tor, Y. Tetrahedron 2007,
63, 3425–3426. (c) Asseline, U. Curr. Org. Chem. 2006, 10, 491–
518. (d) Wilson, J. N.; Kool, E. T. Org. Biomol. Chem. 2006, 4, 4265–
4274. (e) Rist, M. J.; Marino, J. P. Curr. Org. Chem. 2002, 6, 775–
793.
The current investigation includes synthesis of among the first
carboxamide derivatives 3 (Figure 1), that despite bearing a
(11) (a) Srivatsan, S. G.; Weizman, H.; Tor, Y. Org. Biomol. Chem. 2008,
6, 1334–1338. (b) Tor, Y.; Del Valle, S.; Jaramillo, D.; Srivatsan,
S. G.; Rios, A.; Weizman, H. Tetrahedron 2007, 63, 3608–3614. (c)
Srivatsan, S. G.; Tor, Y. Tetrahedron 2007, 63, 3601–3607. (d)
Srivatsan, S. G.; Tor, Y. J. Am. Chem. Soc. 2007, 129, 2044–2053.
(e) Greco, N. J.; Tor, Y. Tetrahedron 2007, 63, 3515–3527. (f) Greco,
N. J.; Tor, Y. J. Am. Chem. Soc. 2005, 127, 10784–10785.
(12) For additional recent examples:(a) Wilson, J. N.; Cho, Y.; Tan, S.;
Cuppoletti, A.; Kool, E. T. ChemBioChem 2008, 9, 279–285. (b) Seela,
F.; Sirivolu, V. R. Org. Biomol. Chem. 2008, 6, 1674–1687. (c) Sandin,
P.; Bo¨rjesson, K.; Li, H.; Mårtensson, J.; Brown, T.; Wilhelmsson,
L. M.; Albinsson, B. Nucleic Acids Res. 2008, 36, 157–167. (d) Mitsui,
T.; Kimoto, M.; Kawai, R.; Yokoyama, S.; Hirao, I. Tetrahedron 2007,
63, 3528–3537. (e) Mizuta, M.; Seio, K.; Miyata, K.; Sekine, M. J.
Org. Chem. 2007, 72, 5046–5055. (f) Firth, A. G.; Fairlamb, I. J. S.;
Darley, K.; Baumann, C. G. Tetrahedron Lett. 2006, 47, 3529–3533.
(g) Mart´ı, A. A.; Jockusch, S.; Li, Z.; Ju, J.; Turro, N. J. Nucleic Acids
Res. 2006, 34, e50. (h) Sharon, E.; Le´vesque, S. A.; Munkonda, M. N.;
Se´vigny, J.; Ecke, D.; Reiser, G.; Fischer, B. ChemBioChem 2006, 7,
1361–1374. (i) Gaied, N. B.; Glasser, N.; Ramalanjaona, N.; Beltz,
H.; Wolff, P.; Marquet, R.; Burger, A.; Me´ly, Y. Nucleic Acids Res.
2005, 33, 1031–1039. (j) Okamoto, A.; Saito, Y.; Saito, I. J.
Photochem. Photobiol. C 2005, 6, 108–122.
(16) (a) Meier, H. Angew. Chem., Int. Ed. 2005, 44, 2482–2506. (b)
Gompper, R.; Wagner, H.-U. Angew. Chem., Int. Ed. Engl. 1988, 27,
1437–1455.
(17) Recent examples involving small molecules oligomers Davies, J. A.;
Elangovan, A.; Sullivan, P. A.; Olbricht, B. C.; Bale, D. H.; Ewy,
T. R.; Isborn, C. M.; Eichinger, B. E.; Robinson, B. H.; Reid, P. J.;
Li, X.; Dalton, L. R. J. Am. Chem. Soc. 2008, 130, 10565–10575. (b)
Jordan, B. J.; Pollier, M. A.; Ofir, Y.; Joubanian, S.; Mehtala, J. G.;
Sinkel, C.; Caldwell, S. T.; Kennedy, A.; Rabani, G.; Cooke, G.;
Rotello, V. M. Chem. Commun. 2008, 1653–1655. (c) Ort´ız, A.;
Insuasty, B.; Torres, M. R.; Herranz, M. A.; Mart´ın, N.; Viruela, R.;
Ort´ı, E. Eur. J. Org. Chem. 2008, 99–108. (d) Zhou, Y.; Xiao, Y.;
Chi, S.; Qian, X. Org. Lett. 2008, 10, 633–636. (e) Yamaguchi, Y.;
Shimoi, Y.; Ochi, T.; Wakamiya, T.; Matsubara, Y.; Yoshida, Z-i J.
Phys. Chem. B 2008, 112, 5074–5084. (f) Khrarnov, D. M.; Bielawski,
C. W. J. Org. Chem. 2007, 72, 9407–9417. (g) Yuan, M.-S.; Liu,
Z.-Q.; Fang, Q. J. Org. Chem. 2007, 72, 7915–7922. (h) Meier, H.;
Mu¨hling, B.; Gerold, J.; Jacob, D.; Oehlhof, A. Eur. J. Org. Chem.
2007, 625–631. (i) Dufresne, S.; Bourgeaux, M.; Skene, W. G. J.
Mater. Chem. 2007, 17, 1166–1177. (j) Sakai, N.; Sisson, A. L.;
Bhosale, S.; Fu¨rstenberg, A.; Banerji, N.; Vauthey, E.; Matile, S. Org.
Biomol. Chem. 2007, 5, 2560–2563. (k) Oliva, M. M.; Casado, J.;
Raposo, M. M. M.; Fonseca, A. M. C.; Hartmann, H.; Herna´ndez,
V.; Navarrete, J. T. L. J. Org. Chem. 2006, 71, 7509–7520. (l) Lu,
Z.; Lord, S. J.; Wang, H.; Moerner, W. E.; Twieg, R. J. J. Org. Chem.
2006, 71, 9651–9657. (m) Sonoda, Y.; Goto, M.; Tsuzuki, S.; Tamaoki,
N. J. Phys. Chem. A 2006, 110, 13379–13387. (n) Lin, J.-H.;
Elangovan, A.; Ho, T.-I. J. Org. Chem. 2005, 70, 7397–7407. (o)
Andersson, A. S.; Qvortrup, K.; Torbensen, E. R.; Mayer, J.-P.;
Gisselbrecht, J.-P.; Boudon, C.; Gross, M.; Kadziola, A.; Kilså, K.;
Nielsen, M. B. Eur. J. Org. Chem. 2005, 3660–3671. (p) Murase, T.;
Fujita, M. J. Org. Chem. 2005, 70, 9269–9278. (q) Sumalekshmy, S.;
Gopidas, K. R. J. Phys. Chem. B 2004, 108, 3705–3712.
(13) See, for example: Ballin, J. D.; Bharill, S.; Fialcowitz-White, E. J.;
Gryczynski, I.; Gryczynski, Z.; Wilson, G. M. Biochemistry 2007, 46,
13948–13960.
(14) (a) Kodali, G.; Kistler, K. A.; Matsika, S.; Stanley, R. J. J. Phys. Chem.
B 2008, 112, 1789–1795. (b) Bharill, S.; Sarkar, P.; Ballin, J. D.;
Gryczynski, I.; Wilson, G. M.; Gryczynski, Z. Anal. Biochem. 2008,
377, 141–149. (c) Neely, R. K.; Magennis, S. W.; Parsons, S.; Jones,
A. C. ChemPhysChem 2007, 8, 1095–1102. (d) Serrano-Andre´s, L.;
Mercha´n, M.; Borin, A. C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
8691–8696. (e) He, R.-X.; Duan, X.-H.; Li, X.-Y. Phys. Chem. Chem.
Phys. 2006, 8, 587–591. (f) Perun, S.; Sobolewski, A. L.; Domcke,
W. Mol. Phys. 2006, 104, 1113–1121. (g) Borin, A. C.; Serrano-
Andre´s, L.; Ludwig, V.; Coutinho, K.; Canuto, S. Int. J. Quantum
Chem. 2006, 106, 2564–2577. (h) Hardman, S. J. O.; Thompson, K. C.
Biochemistry 2006, 45, 9145–9155. (i) Seefeld, K. A.; Plu¨tzer, C.;
Lo¨wenich, D.; Ha¨ber, T.; Linder, R.; Kleinermanns, K.; Tatchen, J.;
Marian, C. M. Phys. Chem. Chem. Phys. 2005, 7, 3021–3026. (j)
Larsen, O. F. A.; van Stokkum, I. H. M.; Groot, M.-L.; Kennis,
J. T. M.; van Grondelle, R.; van Amerongen, H. Chem. Phys. Lett.
2003, 371, 157–163. (l) Rachofsky, E. L.; Ross, J. B. A.; Krauss, M.;
Osman, R. J. Phys. Chem. A 2001, 105, 190–197. (m) Jean, J. M.;
Hall, K. B. J. Phys. Chem. A 2000, 104, 1930–1937. (n) Holme´n, A.;
Norde´n, B.; Albinsson, B. J. Am. Chem. Soc. 1997, 119, 3114–3121.
(o) Evans, K.; Xu, D.; Kim, Y.; Nordlund, T. M. J. Fluoresc. 1992,
2, 209–216.
(18) (a) Buresˇ, F.; Schweizer, W. B.; Boudon, C.; Gisselbrecht, J.-P.; Gross,
M.; Diederich, F. Eur. J. Org. Chem. 2008, 994–1004. (b) Buresˇ, F.;
Schweizer, W. B.; May, J. C.; Boudon, C.; Gisselbrecht, J.-P.; Gross,
M.; Biaggio, I.; Diederich, F. Chem. Eur. J. 2007, 13, 5378–5387. (c)
Michinobu, T.; Boudon, C.; Gisselbrecht, J.-P.; Seiler, P.; Frank, B.;
Moonen, N. N. P.; Gross, M.; Diederich, F. Chem. Eur. J. 2006, 12,
1889–1905. (d) Moonen, N. N. P.; Pomerantz, W. C.; Gist, R.; Boudon,
C.; Gisselbrecht, J.-P.; Kawai, T.; Kishioka, A.; Gross, M.; Irie, M.;
Diederich, F. Chem. Eur. J. 2005, 11, 3325–3341.
(19) (a) Spitler, E. L.; Haley, M. M. Org. Biomol. Chem. 2008, 6, 1569–
1576. (b) Spitler, E. L.; Monson, J. M.; Haley, M. M. J. Org. Chem.
2008, 73, 2211–2223. (c) Marsden, J. A.; Miller, J. J.; Shirtcliff, L. D.;
Haley, M. M. J. Am. Chem. Soc. 2005, 127, 2464–2476.
(15) D’Souza, F.; Gadde, S.; Islam, D.-M. S.; Pang, S.-C.; Schumacher,
A. L.; Zandler, M. E.; Horie, R.; Araki, Y.; Ito, O. Chem. Commun.
2007, 480–482.
(20) Preliminary results have been published: Butler, R. S.; Myers, A. K.;
Bellarmine, P.; Abboud, K. A.; Castellano, R. K. J. Mater. Chem.
2007, 17, 1863–1865.
9
624 J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009

Push-Pull Purines
A R T I C L E S
Scheme 1. Synthesis of Donor-Acceptor Purines 1 and 2
Table 1. Yields for the Preparation of Bromides 6 and
Donor-Acceptor Purines 1 and 2a
yield (%)b
derivative
R1
R2
6
1
2
a
b
c
d
e
f
NH2
H
H
H
24
55
80
77
78
68
89
80
26
39
42
91
77
93
60
73
71
82
86
75
64
NHCH3
N(CH3)2
NH2
NH2
N(CH3)2
N(CH3)2
OBnc
N(CH3)2
NH2
N(CH3)2
g
a The syntheses of derivatives a, d, e, and f have been reported in ref
20. b Isolated yields shown. c Bn ) benzyl.
generally high yields (Table 1). The bromides were subsequently
converted to their corresponding nitriles 1 using Gundersen’s
cyanation conditions (Zn(CN)2, Pd(PPh3)4, NMP) in moderate
yields and with good reproducibility.29b Several precautions
seem to improve the outcome of the reactions including (a)
degassing and drying the solvent (NMP), (b) slow addition of
a Zn(CN)2 solution to an already prepared purine/Pd mixture
(see the Supporting Information for details) in an attempt to
keep the concentration of excess cyanide low,30 and (c) use of
significant excesses of purified Pd(PPh3)4 (20-50 mol %) given
what are likely highly coordinating purine species. While other
Pd catalysts can be used, we and others29a,b have found that
Pd(PPh3)4 is generally reliable. Finally, conversion of the nitriles
1 to carboxymethyl esters 2 comes in generally good yield
through treatment with methanolic ammonia and then acidic
hydrolysis of the intermediate methyl imidates (not shown).28c
Worth noting, other (more direct) approaches to 8-carboxyesters
2 are conceivable. Among these, we have found that lithiation31
or lithium-halogen exchange32 followed by an electrophilic
quench (e.g., with methyl chloroformate) are impractically
capricious for substrates 5.
Design and Synthesis of Carboxamides 3. Given access to 1
and 2, we envisioned that an amide linkage at the C(8) position
could mutually serve as an electron-accepting function and the
basis for later connecting the purines to biomolecules or surfaces.
Along these lines, while access to primary (CONH2) and simple
tertiary (e.g., CON(CH3)2)28c,31,33 amides is reasonably well-
described for purines, routes to diverse amides (including
secondary amides) are largely limited to the palladium-catalyzed
carboxyamidation chemistry of Eaton and co-workers (reagents/
conditions: 8-bromopurine, amine, CO (50 psi), Pd(PPh3)4, NEt3,
“weak” acceptor group maintain high (>80%) quantum yields
over a range of solvent polarities. Preparation of 3 through
conventional amide-bond forming reactions (i.e., DCC coupling)
underlies an attractive strategy for introduction of the fluoro-
phores to peptidic architectures or for their connection to other
molecules or surfaces. Overall, the work presents (a) the
straightforward extension of the “push-pull” concept to purines
to afford useful photophysical properties and (b) fundamental
insight, from structure-property studies, prerequisite for the
broader application of purines in biosensing and materials.
Results and Discussion
SynthesisofDonor-AcceptorPurines1and2.Thedonor-acceptor
purines (Figure 1 and Table 1) are accessible from a common
intermediate, 2-amino-9-benzyl-6-chloropurine, 4, made from
benzylation of commercially available (or readily made21)
2-amino-6-chloropurine.22 The benzyl group at N(9) serves as
a protecting and solubilizing group (for organic solvents), and
to impart tautomeric stability. Hydrogenolysis of 4 provides 5a
(see Table 1 for derivative letter codes);23 subsequent diazoti-
zation and chlorination gives known 9-benzyl-2-chloropurine
(not shown),24 precursor to 2-methylamino 5b25 and 2-dim-
ethylamino 5c that are formed upon appropriate nucleophilic
aromatic substitution (see the Supporting Information for
synthetic details). DABCO-catalyzed alcoholysis or direct
amination of 4 affords 5d26 and 5e,27 respectively. Finally,
diazotization and chlorination of 4 gives 9-benzyl-2,6-dichlo-
ropurine (not shown), and access to 5f (in two additional steps)
and 5g (in one additional step) (Scheme 1).
With the donor groups installed, the C(8) position is available
for acceptor group introduction. The cyano group was selected
due to its standard use in simple donor-π-acceptor molecules,16-19
known introduction to halopurines through displacement28 or
metal-mediated cross-coupling reactions,29 and amenability to
further functionalization. Bromination at the C(8) position of 5
(28) (a) Butora, G.; Schmitt, C.; Levorse, D. A.; Streckfuss, E.; Doss, G. A.;
MacCoss, M. Tetrahedron 2007, 63, 3782–3789. (b) Ho, C. Y.; Kukla,
M. J. Bioorg. Med. Chem. Lett. 1991, 1, 531–534. (c) Kato, T.;
Arakawa, E.; Ogawa, S.; Suzumura, Y.; Kato, T. Chem. Pharm. Bull.
1986, 34, 3635–3643. (d) Matsuda, A.; Nomoto, Y.; Ueda, T. Chem.
Pharm. Bull. 1979, 27, 183–192.
25
(29) (a) Dola´kova´, P.; Masoj´ıdkova´, M.; Holy´, A. Heterocycles 2007, 71,
1107–1115. (b) Gundersen, L.-L. Acta Chem. Scand. 1996, 50, 58–
63. (c) Nair, V.; Purdy, D. F. Tetrahedron 1991, 47, 365–382. (d)
Tanji, K.; Higashino, T. Heterocycles 1990, 30, 435–440.
(30) The sensitivity of Pd catalysts to excess cyanide is well documented:
(a) Erhardt, S.; Grushin, V. V.; Kilpatrick, A. H.; Macgregor, S. A.;
Marshall, W. J.; Roe, D. C. J. Am. Chem. Soc. 2008, 130, 4828–
4845. (b) Chidambaram, R. Tetrahedron Lett. 2004, 45, 1441–1444.
(c) Marcantonio, K. M.; Frey, L. F.; Liu, Y.; Chen, Y.; Strine, J.;
Phenix, B.; Wallace, D. J.; Chen, C.-y. Org. Lett. 2004, 6, 3723–
3725. (d) Sundermeier, M.; Mutyala, S.; Zapf, A.; Spannenberg, A.;
Beller, M. J. Organomet. Chem. 2003, 684, 50–55.
was accomplished with Br2 to afford intermediates 6 in
(21) Igi, M.; Hayashi, T. Process for preparing 2-amino-6-halopurines and
intermediates. EP 543095, 1993.
(22) Holy´, A.; Gu¨nter, J.; Dvora´kova´, H.; Masoj´ıdkova´, M.; Andrei, G.;
Snoeck, R.; Balzarini, J.; De Clercq, E. J. Med. Chem. 1999, 42, 2064–
2086.
(23) Liu, F.; Dalhus, B.; Gundersen, L.-L.; Rise, F. Acta Chem. Scand.
1999, 53, 269–279.
(24) Tobrman, T.; Dvora´k, D. Org. Lett. 2006, 8, 1291–1294.
(25) Kurimoto, A.; Ogino, T.; Ichii, S.; Isobe, Y.; Tobe, M.; Ogita, H.;
Takaku, H.; Sajiki, H.; Hirota, K.; Kawakami, H. Bioorg. Med. Chem.
2003, 11, 5501–5508.
(31) Hayakawa, H.; Haraguchi, K.; Tanaka, H.; Miyasaka, T. Chem. Pharm.
Bull. 1987, 35, 72–79.
(26) Linn, J. A.; McLean, E. W.; Kelley, J. L. J. Chem. Soc., Chem.
Commun. 1994, 913–914.
(32) Coˆng-Danh, N.; Beaucourt, J.-P.; Pichat, L. Tetrahedron Lett. 1979,
2385–2388.
(27) Martin, A. M.; Butler, R. S.; Ghiviriga, I.; Giessert, R. E.; Abboud,
K. A.; Castellano, R. K. Chem. Commun. 2006, 4413–4415.
(33) Gudmundsson, K. S.; Daluge, S. M.; Condreay, L. D.; Johnson, L. C.
Nucleosides, Nucleotides Nucleic Acids 2002, 21, 891–901.
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 625

A R T I C L E S
Butler et al.
Scheme 2. Synthesis of Purinyl Carboxamides 3
3f) the phenyl and purine planes are nearly perpendicular
( abcd ) 82°, 91°, 103°, and 89°, respectively) while the values
for bcde vary as the phenyl group responds to nearest
neighbors in the solid state.39
Key aspects of the packing structures of 2f and 3f are shown
(Figure 2). Ester 2f, like 1f, forms H-bonded dimers from its
Hoogsteen edge (Figure 2a), but through four hydrogen bonds
(N(7)· · ·N(10)′ ) 3.03 Å; O(15) · · ·N(10)′ ) 2.87 Å) instead
of two.40 The structure is reminiscent of other quadruply
H-bonded dimers derived from purines.26,27,41 Amide 3f (Figure
2b) instead forms infinite polar chains via intermolecular
H-bonding between amide carbonyl oxygens and amino group
hydrogens (N(10)· · ·O(15)′ ) 2.97 Å). This pattern bears
relevance to the “guanine ribbons” exploited by Gottarelli and
co-workers, the intrinsic dipole of which has been shown critical
to gel and liquid crystal formation and electronic device
performance.4 Ester 2f additionally shows (Figure 2a) strong
antiparallel (dipolar) π-stacking interactions (interplanar distance
) 3.30 Å), documented previously for 1e (3.32 Å) and 1f (3.38
Å),20 and common for donor-acceptor molecules.19a,c,42,43
Within this motif the ester acceptor of one molecule is positioned
directly above the pyrimidine ring of a neighboring purine; the
same ester-pyrimidine packing can be found in the structure of
3f. Extended packing diagrams are provided in the Supporting
Information that reveal the role of the benzyl substituents in
the longer-range organization of the molecules. Most impor-
tantly, the crystal structures highlight relationships between the
built-in molecular recognition functionality of the purines and
their solid-state organization; this knowledge could potentially
be used to optimize device performance and optoelectronic
properties.
Absorption Properties. In general, the photophysical proper-
ties of 2-amino- and 2-dimethylaminopurine derivatives remain
intensely studied, from both mechanistic and applied perspec-
tives (vide infra). In solution, the optical behavior of various
C(8)-unsubstituted compounds has been evaluated in water,9c,44
but equivalently comprehensive measurements in organic media
are mostly restricted to unsubstituted 2-aminopurines9c,14b,o,44a,b
and 2-dialkylaminopurines.44a,b Relevant data from these com-
pounds and 5 are discussed in the context of the donor-acceptor
purines 1-3 below. The basic premise is that knowing the
solvent-induced absorption and emission changes for 1-3 and
understanding the dependence of the changes on structure should
expose opportunities for the molecules in both biological and
materials applications. Discussion of the absorption and emission
Table 2. Yields for the Preparation of Carboxamides 3
entry
product
R1
R2
R3
yield (%)a
1
2
3
4
3g1
3g2
3g3
3f
N(CH3)2
N(CH3)2
N(CH3)2
N(CH3)2
N(CH3)2
N(CH3)2
N(CH3)2
NH2
(CH2)3CH3
42
82
45
51
CH2CO2CH3
(CH2)4CO2CH3
CH2CO2CH3
a Isolated yields shown.
and DMF).34,35 Eaton’s approach could, in theory, be entertained
beginning from bromides 6. We chose to alternatively explore
amidepreparationviatheintermediate(andgenerallyunstable28c,d,36
)
8-carboxypurine (generated from hydrolysis of either 1 or 2)
using appropriate amines and standard peptide coupling condi-
tions.
Four examples (Scheme 2 and Table 2) illustrate the approach.
Hydrolysis of bis(dimethylamino) derivative 1g using 10%
aqueous sodium hydroxide in ethanol first provides the car-
boxylate salt (not shown), an apparently stable species that does
not require special handling or storage. Following protonation
with HCl at 0 °C, the carboxylic acid is kept cold and used
immediately37 in a DCC-mediated coupling reaction with
various primary amines. Entry 1 shows the result using simple
butylamine,38 while entries 2-4 that use the methyl esters of
glycine and 5-aminovaleric acid are more relevant to incorpora-
tion of the purines within peptide sequences. Entry 4 further
shows, expectedly, that the aromatic amino group at C(6) is
relatively non-nucleophilic under these conditions.
X-ray Crystallographic Analysis. The X-ray crystal structures
of ester 2f and amide 3f are presented in Figure 2 (additional
crystallographic data, including ORTEP plots, are presented in
the Supporting Information), as they highlight some of the
features common to the donor-acceptor purines in the solid
state. The structures of two nitriles, isomers 1e and 1f, have
already been reported.20 At the molecular level, the donor and
acceptor substituents are fully conjugated with the purine cores.
The dimethylamino groups are planar (sum of the three C-N-C
bond angles ) 360°) and the nonhydrogen atoms defining the
methyl ester (Figure 2a) or amide (Figure 2b) substituents
deviate from the mean purine plane by <0.1 Å. The -NH of
the amide group of 3f is expectedly positioned on the same
side as N(7), a consequence of electrostatics (N(7) · · ·N(16) )
2.74 Å). Finally, important for how the purines arrange in the
solid state, the geometries of the N(9) benzyl substituents can
be analyzed through two torsion angles, abcd and bcde
(labeled in Figure 2a). For all four structures (1e, 1f, 2f, and
(39) A comparison of the bond lengths for 1e, 1f, 2f, 3f, and 2-AP is
presented in Table S2 (Supporting Information). The values agree
within ∼0.05 Å.
(40) The compound at best only weakly dimerizes in CDCl3; poor solubility
at high concentrations (e.g., 10 mM) has prevented a more detailed
analysis.
(41) (a) Ong, H. C.; Zimmerman, S. C. Org. Lett. 2006, 8, 1589–1592. (b)
Park, T.; Todd, E. M.; Nakashima, S.; Zimmerman, S. C. J. Am. Chem.
Soc. 2005, 127, 18133–18142. (c) Nowick, J. S.; Chen, J. S.; Noronha,
G. J. Am. Chem. Soc. 1993, 115, 7636–7644.
(42) (a) Boiadjiev, S. E.; Lightner, D. A. J. Org. Chem. 2005, 70, 688–
691. (b) Moonen, N. N. P.; Gist, R.; Boudon, C.; Gisselbrecht, J.-P.;
Seiler, P.; Kawai, T.; Kishioka, A.; Gross, M.; Irie, M.; Diederich, F.
Org. Biomol. Chem. 2003, 1, 2032–2034.
(34) Tu, C.; Keane, C.; Eaton, B. Nucleosides Nucleotides 1997, 16, 227–
237.
(35) There are, however, no examples of C(8) purinyl carboxamides where
C(2) is substituted with an NR′R′′ group.
(43) (a) Yao, S.; Beginn, U.; Gress, T.; Lysetska, M.; Wu¨rthner, F. J. Am.
Chem. Soc. 2004, 126, 8336–8348. (b) Wu¨rthner, F.; Yao, S.;
Debaerdemaeker, T.; Wortmann, R. J. Am. Chem. Soc. 2002, 124,
9431–9447. (c) Wu¨rthner, F.; Yao, S. Angew. Chem., Int. Ed. 2000,
39, 1978–1981.
(36) Naka, T.; Honjo, M. Chem. Pharm. Bull. 1976, 24, 2052–2056.
(37) The putative acid formed upon hydrolysis of 1a readily decarboxylates
at room temperature in solution or the solid state; the process can be
1
observed by TLC and H NMR.
(38) Also tested was conversion of the carboxylic acid first to its acid
chloride (SOCl2, 80 °C, 1 h) and then to the amide (butylamine, NEt3,
THF, rt). The overall yield for the two-step process was a lower 23%.
(44) (a) Smagowicz, J.; Wierzchowski, K. L. J. Lumin. 1974, 8, 210–232.
(b) Drobnik, J.; Augenstein, L. Photochem. Photobiol. 1966, 5, 83–
97. (c) Mason, S. F. J. Chem. Soc. 1954, 2071–2081.
9
626 J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009

Push-Pull Purines
A R T I C L E S
Figure 3. Absorption data for the 2,6-bis(dimethylamino)purine series g
in CH2Cl2 (2 × 10-5 M).
9-methylpurine,14d,m,44a,45 and verified by preliminary calcula-
tions (vide infra), we assign the lowest energy band to the 1(ππ*
La)46 transition (S0 f S1), and the higher-energy band to the
1(ππ* Lb) transition (S0 f S3).47 In all cases the maximum
absorption wavelengths (λmax) are concentration independent
over a typical range for the measurements (5-60 µM) and plots
of absorbance versus concentration for the low-energy band are
linear. Moreover, the extinction coefficients are moderately high
(log ε ≈ 4), consistent with other 2-AP derivatives.14n,44c
The remaining results and discussion will focus primarily on
the low-energy absorption for 5 and 1-3 since it leads to
fluorescence emission. Earlier optical and theoretical
studies9d,14m,n,44 with purines help to rationalize the response
of this band first to donor structure at C(2) and C(6), and then
to acceptor substitution at C(8). The low-energy transition
moment (S0 f S1, π f π*) for 2-amino-9-methylpurine is
polarized at an angle ∼55° with respect to the purine C(4)-C(5)
bond (in the direction of N(1); Figure 1), a value relatively
insensitive to “inert” substituents at N(9).14n Donor substituents
at C(2), positioned ∼30° from the long axis (∼65° from the
transition dipole moment) significantly affect this transition. Data
from the acceptor-free purines 5 are illustrative (see Table S3).
Purines 5a-5c, for example, bearing donors only in the C(2)
position, show the most red-shifted low-energy absorption bands
whose wavelength scales with the donor ionization potential
(NH2 (304 nm) < NHCH3 (317 nm) < N(CH3)2 (330 nm)).
The presence of a donor at C(6) incidentally blue shifts the low-
energy absorption wavelength ∼20-30 nm. The hypsochromic
shift magnitude is relatively insensitive to the nature of the C(6)
donating group such that the nature of the C(2) donor tunes the
Figure 2. X-ray crystal structures of 2f (a) and 3f (b). (a) Dimers of 2f
formed by quadruple H-bonding (left) and antiparallel dipolar π-stacking
(right). (b) Polar 1-D H-bonded chains of 3f in the solid state. H-bonding
interactions are indicated by dashed lines. Some hydrogens have been
omitted for clarity. Additional crystallographic details are provided in the
Supporting Information.
Table 3. Electronic Absorption and Emission Data for 1-3 in
a
CH2Cl2
lowest energy abs λmax [nm]
c
ΦF
purine
em λmax [nm]b
∆λmax
(log ε [M-1 cm-1])
1ad
1b
326 (3.2)
341 (3.8)
361 (4.2)
311 (4.3)
324 (4.2)
336 (4.1)
348 (4.4)
328 (4.1)
345 (3.2)
362 (4.1)
315 (4.3)
330 (4.2)
338 (4.1)
351 (4.3)
338 (4.2)
343 (4.3)
371
383
429
355
375
387
388
379
403
433
371
393
409
409
402
407
45
41
68
44
51
51
40
51
58
71
56
63
71
58
64
64
0.20
0.58
0.90
0.81
0.30
>0.95
0.20
0.42
0.65
1c
1dd
1ed
1fd
1g
2ad
2b
2c
0.81
2dd
2ed
2fd
2g
>0.95
>0.95
>0.95
0.90
0.87
0.91
3g1
3g2
a All measurements performed at room temperature. b All experiments
were performed using optical densities e0.1 at the excitation
wavelength (λex
) 320 nm; for 3g1 and 3g2 λex ) 360 nm).
abs λmax
.
c Fluorescence quantum yields are relative to the quantum yield of
quinine sulfate in 0.1 M H2SO4 (ΦF ) 0.577). d Data previously
reported in ref 20.
The introduction of acceptor substituents to C(8), on the
purine long axis, significantly affects the absorption maxima
(Figure 3); in general, the low-energy absorption band is red-
shifted by 20-50 nm (and the higher-energy absorption band
significantly less). Here, presumably, the transition dipole
direction is further aligned with the purine long axis. The
data will each begin with the behavior of 1-3 in dichlo-
romethane (CH2Cl2), followed by solvatochromic studies.
Absorption data for 1-3 in CH2Cl2 is provided in Table 3
and graphically for the g series in Figure 3; all tabular data for
the C(8)-H purines 5 is presented in the Supporting Informa-
tion. The absorption spectra of 5 show two intense low-energy
bands at ∼250 and ∼300 nm; both absorption bands are red-
shifted for 1-3 (Figure 3). By analogy to 2-AP and 2-amino-
(45) The optical behavior of the purines is relatively insensitive to the N(9)
substituent. See refs 20 and 14o.
(46) For nomenclature, see:(a) Platt, J. R. J. Chem. Phys. 1949, 17, 484–
495.
(47) Not discussed here is the n f π* (S0 f S2) transition that is weak
and found at an intermediate energy. See ref 14 for details.
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 627

A R T I C L E S
Butler et al.
Table 4. Electronic Absorption and Emission Data for 1-3 in
Table 6. Electronic Absorption and Emission Data for 1-3 in
1,4-Dioxanea
Methanola
lowest energy abs λmax [nm]
lowest energy abs λmax [nm]
(log ε [M-1 cm-1])
c
ΦF
c
ΦF
purine
em λmax [nm]b
∆λmax
purine
em λmax [nm]b
∆λmax
(log ε [M-1 cm-1])
1a
1bd
1c
1d
1e
331 (4.2)
342 (4.0)
355 (4.4)
313 (4.3)
334 (4.4)
339 (4.2)
345 (4.3)
331 (4.2)
345 (4.1)
358 (4.2)
313 (4.3)
334 (4.3)
340 (4.4)
344 (4.3)
332 (4.2)
336 (4.3)
381
390
424
362
373
382
383
386
405
426
376
391
399
402
392
397
50
48
69
49
39
43
38
55
60
68
63
57
59
58
60
61
0.34
0.61d
0.91
0.73
0.78
0.92
>0.95
0.64
1a
1ad
1be
1c
1d
1e
333 (3.2)
331 (2.8)
342 (3.9)
358 (4.0)
318 (4.1)
327 (4.2)
338 (4.2)
342 (4.4)
335 (4.1)
325 (4.1)
345 (3.9)
362 (4.1)
321 (4.3)
331 (4.2)
345 (4.1)
347 (4.2)
335 (4.2)
340 (4.2)
398
406
417
456
382
389
410
403
412
420
429
466
405
420
436
432
424
429
65
75
75
98
64
62
72
61
77
95
84
104
84
89
91
85
89
89
0.64
0.88
0.49e
0.14
0.14
0.01
0.80
0.03
0.87
>0.95
0.61
0.17
0.92
0.39
0.76
0.66
0.89
0.90
1f
1g
2a
2b
2c
2d
2e
1f
1g
2a
2ad
2b
2c
2d
2e
0.65
0.83
>0.95
>0.95
>0.95
>0.95
>0.95
>0.95
2f
2g
3g1
3g2
2f
2g
3g1
3g2
a All measurements performed at room temperature. b All experiments
were performed using optical densities e0.1 at the excitation
a All measurements performed at room temperature. b All experiments
were performed using optical densities e0.1 at the excitation
wavelength (λex
) 320 nm; for 3g1 and 3g2 λex ) 360 nm).
c Fluorescence quantum yields are relative to the quantum yield of
quinine sulfate in 0.1 M H2SO4 (ΦF ) 0.577). d Decomposition has been
observed upon prolonged irradiation (vide infra).
wavelength (λex
) 320 nm; for 3g1 and 3g2 λex ) 360 nm).
c Fluorescence quantum yields are relative to the quantum yield of
quinine sulfate in 0.1 M H2SO4 (ΦF ) 0.577). d Measurements taken in
water. e Decomposition has been observed upon prolonged irradiation
(vide infra).
Table 5. Electronic Absorption and Emission Data for 1-3 in
Acetonitrilea
lowest energy abs λmax [nm]
c
ΦF
purine
em λmax [nm]b
∆λmax
(log ε [M-1 cm-1])
insensitive to the solvent polarity, only fluctuating by a few
nanometers across the range of solvents. The result is similar
to 2-AP14l,o,44a and reflects a small change in dipole moment
magnitudeand/ordirectionbetweenthegroundandFranck-Condon
excited states. Again, plots of the optical density versus
concentration for each compound in each solvent show linearity.
That 1-3 do not appear to aggressively aggregate by dipolar
π-stacking42,43 can be partially attributed to their relatively
modest ground-state dipole moments (µgs ≈ 2-6 D, vide infra).
Emission Properties in CH2Cl2. The electronic emission
spectra for 1-3 are shown in Figure 4 and the data are
summarized in Tables 3-6; data for 5 is provided in the
Supporting Information. Irradiation at the low-energy absorption
wavelength of C(8) unsubstituted 2-amino- and 2-dimethylami-
nopurines 5 in CH2Cl2 (Table S3), for example, yields a broad
and featureless emission from 350-400 nm depending on the
substitution pattern. The results are consistent with the limited
data for purines in organic solvents.9c,44a,49 The emission band
is red-shifted by 14-36 nm upon conversion of 5 to 1, and an
additional 4-22 nm upon the nitrile’s transformation to the
methyl ester 2. For parent series a (5a, 1a, and 2a), for example,
the emission wavelengths increase from 357 f 371 f 379 nm;
even larger changes are found across series f (360 f 387 f
409 nm). Exceptions to this progression are found only for
conversion of 5b, 5d, and 5g to 1, where the emission maxima
show a small hypsochromic shift (∼5 nm). The emission of
amide 3g1 is slightly blue-shifted relative to ester 2g. Although
correlations between emission energies and parameters that
reflect electron-acceptor strength (substituent electron affinity)
are system dependent, these general trends have been observed
in stilbene-like D-A systems.48 For the donor-acceptor purines
in CH2Cl2, the Stokes shifts are 40-71 nm and the values are
generally highest for the esters 2 and amides 3, suggesting a
1a
1b
1c
1d
1e
330 (3.2)
342 (3.0)
357 (3.9)
313 (4.3)
326 (4.3)
338 (4.3)
344 (4.3)
330 (4.3)
345 (3.2)
359 (4.1)
316 (4.2)
332 (4.4)
342 (4.1)
347 (4.3)
333 (4.2)
336 (4.3)
381
405
440
366
378
396
394
388
413
444
382
399
414
415
405
409
51
63
83
53
52
58
50
58
68
85
66
67
72
68
72
73
0.31
0.58
0.53
0.65
0.08
0.55
0.26
0.55
0.66
0.60
0.87
0.83
0.75
0.86
>0.95
>0.95
1f
1g
2a
2b
2c
2d
2e
2f
2g
3g1
3g2
a All measurements performed at room temperature. b All experiments
were performed using optical densities e0.1 at the excitation
wavelength (λex
) 320 nm; for 3g1 and 3g2 λex ) 360 nm).
c Fluorescence quantum yields are relative to the quantum yield of
quinine sulfate in 0.1 M H2SO4 (ΦF ) 0.577).
maxima shifts are fairly similar among the acceptors chosen
(1g ) 348 nm; 2g ) 348 nm; 3g1 ) 338 nm), consistent with
acceptor group trends observed within other systems (λmax CN
≈ λmax CO2CH3 > λmax CONH2).48 Noteworthy, the “parent” a
series shows a red shift of ∼26 nm upon CN/CO2CH3 introduc-
tion (i.e., conversion of 5a to 1a/2a), a potentially useful
improvement over 2-AP for selective excitation in biological
settings.
Optical data for 1-3 in 1,4-dioxane, acetonitrile, and
methanol (and H2O for series a) is provided in Tables 4-6 to
complement the CH2Cl2 data; tabular data for C(8)-H purines
5 is presented in the Supporting Information. The low-energy
absorption maximum for the purines (collectively) is relatively
(48) Anstead, G. M.; Carlson, K. E.; Kym, P. R.; Hwang, K.-J.; Katzenel-
(49) Balu, N.; Gamcsik, M. P.; Colvin, M. E.; Colvin, O. M.; Dolan, M. E.;
lenbogen, J. A. Photochem. Photobiol. 1993, 58, 785–794.
Ludeman, S. M. Chem. Res. Toxicol. 2002, 15, 380–387.
9
628 J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009

Push-Pull Purines
A R T I C L E S
Table 7. Change in Dipole from the Ground to Excited State
(∆µge) Based on Plots of Stokes Shift versus ET for 1-3a
N
c
)
d
c
)
d
purine a (Å)b slope (cm-1
Rc
∆µge
slope (cm-1
Rc
∆µge
including 1,4-dioxane data excluding 1,4-dioxane data
1a
1b
1c
1d
1e
1f
1g
2a
2b
2c
2d
2e
2f
4.0
4.1
4.2
4.5
4.2
4.2
4.4
4.1
4.2
4.3
4.6
4.4
4.3
4.5
4.8
2170
3260
2690
1880
2580
3070
2770
3330
2540
3060
2260
3350
2610
2730
2930
0.951 2.04
0.902 2.60
0.937 2.45
0.882 2.27
0.915 2.40
0.996 2.61
0.979 2.66
0.952 2.63
0.952 2.38
0.978 2.70
0.832 2.57
0.984 2.93
0.957 2.50
0.949 2.73
0.984 3.12
2720
4200
3390
2730
1590
2820
3120
4070
3270
3490
3630
3560
2190
3530
3380
0.999 2.29
0.935 2.95
0.968 2.75
0.982 2.73
0.954 1.88
1.00
2.51
0.983 2.83
0.989 2.90
0.997 2.70
0.984 2.89
0.997 3.26
0.978 3.02
0.930 2.29
0.995 3.11
0.997 3.35
2g
3g1
a See the Results and Discussion section for details. b Onsager radii
corresponding to volumes determined by eq 2. See the text for details.
c The slopes and correlation factors (R) for fitted lines from Stokes shift
for examples). d Values
N
(cm-1
)
versus ET plots (see Figure
5
determined using eq 1. See the text for details.
quantum yields to 50% or higher across the board (Table 3).
High values also appear to characterize amides 3. These trends
are difficult to predict solely on the basis of optical energy gaps
(vide infra) or apparent acceptor “strength”. Finally, as shown
for selected compounds before,20 the fluorescence lifetimes (τF)
determined in CH2Cl2 scale, in general, with the quantum yields
and the best emitters have τF values of ∼2.5-4 ns (larger than
2-AP in 1,4-dioxane (1.5 ns)14b). All decays could be fit, with
ꢀ2 values close to 1, to a single exponential, and are consistent
with fluorescence emission from the lowest singlet excited state
(S1) (see the Supporting Information for additional data and
experimental details).
Fluorescence Solvatochromism. The emission properties for
5 and 1-3 have been evaluated in 1,4-dioxane (ET(30) ) 36.0
kcal mol-1), CH2Cl2 (ET(30) ) 40.7 kcal mol-1), acetonitrile
(ET(30) ) 45.6 kcal mol-1), methanol (ET(30) ) 55.4 kcal
mol-1), and additionally for the a series, water (ET(30) ) 63.1
kcal mol-1).52,53 The data are represented in Tables 3-7, Figures
5 and 6, and the Supporting Information. For the acceptor-
bearing purines 1-3, the emission wavelength maxima generally
increase with increasing solvent polarity; the Stokes shifts
correspondingly increase to the tune of 16-32 nm for 1, 29-40
nm for 2, and ∼30 nm for 3g over the range of solvent polarities
(from 1,4-dioxane to methanol or water). Thus, em λmax ranges
from 355 to 466 nm for the purines 1-3 depending on
substitution pattern and solvent. Upon the suggestion of Tor
and co-workers,53a we have used plots of the Stokes shift versus
Reichardt’s normalized microscopic solvent polarity parameter
Figure 4. Normalized emission spectra (based on quantum yield) for 1
(a), 2, and 3g1 (b) in CH2Cl2 (concentration ) 5 × 10-6 M; λex as indicated
in Table 3).
greater degree of structural reorganization in the excited state
for these compounds.
Quantum yields for 1-3 were determined using quinine
sulfate in 0.1 M H2SO4 (ΦF ) 0.577)50 as a standard. In CH2Cl2,
the quantum yields of 5 parallel what has been reported for
9-ethyl-2-aminopurine (ΦF ) 0.085 in CHCl3), 2-aminopurine,
and 2-dimethylaminopurine in organic solvents.9c,44a The values
are generally highest for purines bearing only a C(2) donor,
and increase with the donor strength (a < b < c). Quantum
yields are consistently low (e12%) when a C(6) donor is present
(again, the ΦF values are vanishing in the absence of a donor
at C(2)9,51). The addition of the C(8)-CN acceptor leads to
dramatic quantum yield increases from 5; compounds 1d (ΦF
) 0.033 f 0.81), 1e (ΦF ) 0.013 f 0.30), and 1g (ΦF ) 0.033
f 0.20) show increases of 2 orders of magnitude while the
values for compounds 1c and 1f increase to near unity.
Comparison of isomeric species 1e (ΦF ) 0.30) and 1f (ΦF )
> 0.95) highlights the contribution of the C(2) donor, and its
position, to the quantum yield. Although the enhancements are
considerably more modest across the a series, further conversion
of the nitriles 1 to the methyl esters 2 serves to increase the
ET (or equivalently, ET(30))52 to compare the purines more
N
quantitatively in terms of their solvent-dependent optical
N
behavior (slopes for fitted lines are given in Table 7). The ET
polarity scale, unlike methods that use bulk solvent polarity
functions such as the dielectric constant (ε) and refractive index
(n), takes specific solvent-solute interactions into account. Even
with notoriously anomalous dioxane53c and H-bonding solvents54
(52) Reichardt, C. Chem. ReV. 1994, 94, 2319–2358.
(53) (a) Sinkeldam, R. W.; Tor, Y. Org. Biomol. Chem. 2007, 5, 2523–
2528. (b) Katritzky, A. R.; Fara, D. C.; Yang, H.; Ta¨mm, K.; Tamm,
T.; Karelson, M. Chem. ReV. 2004, 104, 175–198. (c) Suppan, P.;
Ghoneim, N. SolVatochromism; The Royal Society of Chemistry:
Cambridge, 1997.
(50) Eastman, J. W. Photochem. Photobiol. 1967, 6, 55–72.
(51) Even when a dimethylamino group is at C(6): Albinsson, B. J. Am.
Chem. Soc. 1997, 119, 6369-6375.
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 629

A R T I C L E S
Butler et al.
Figure 6. Normalized emission spectra (based on quantum yield) for 1a
(a) and 2a (b) in various solvents (concentration ) 5 × 10-6 M; λex ) 320
nm).
(average correlation factor, R, equals 0.94). The plots for the g
series are given as insets in Figure 5. Removing the 1,4-dioxane
data improves the d values substantially and boosts the overall
average correlation factor to 0.98 (Table 7). Incidentally, plots
of the Stokes shifts versus ∆f (the orientation polarizability)
show, on average, poor linear correlations as do Stokes shift
N
versus ET plots constructed using the data from acceptor-free
5. Three important conclusions emerge from the studies: (a)
The solvatochromic behavior is consistent with stabilization of
an emissive excited-state that has polar character (i.e., µe > µg);56
(b) the positive solvatochromic response is generally larger
(based on the slopes) for the esters than the nitriles, consistent
with some literature observations;48 (c) given the usefulness of
Stokes shift changes for sensing applications, purines 1-3 are
importantly more responsive than the acceptor-free purines 5
or 2-AP due to their enhanced charge separation.
Figure 5. Normalized emission spectra (based on quantum yield) for 1g
(a), 2g (b), and 3g1 (c) in various solvents (concentration ) 5 × 10-6 M;
λex as indicated in Tables 3-6). The insets show plots of the Stokes shift
(cm-1) versus Reichardt’s normalized solvent polarity measure, ETN. Solvent
values are as follows. 1,4-dioxane ) 0.164; dichloromethane ) 0.309;
acetonitrile ) 0.460; methanol ) 0.762; water ) 1.000 (ref 52). See Table
7 for additional data.
(54) Specific H-bonding interactions may (ref 14o) or may not (refs 14b
and 14l) significantly influence 2-AP spectral shifts.
(55) Mannekutla, J. R.; Mulimani, B. G.; Inamdar, S. R. Spectrochim. Acta,
Part A 2008, 69, 419–426.
like methanol (and water, for series a) included,55 plots for
1a-g, 2a-g, and 3g1 give very good linear correlations
(56) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer: New York, 2006.
9
630 J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009

Push-Pull Purines
A R T I C L E S
The solvatochromic Stokes shift data could further be used
to estimate the dipole moment change on excitation of the
purines from the ground to excited-state using Samanta and
Radhakrishnan’s eq 1:57
among c-g (e.g., 1f), but the behavior shown in Figure 5a (for
1g) is typical. Importantly, though, is that purines 1-3 do not
respond equivalently to solvent polarity. The methyl esters 2
show only moderate decreases in fluorescence (see Figure 5b
for 2g), by comparison to 5 or 1, as solvent polarity is increased;
most derivatives even maintain high values regardless of the
solvent. The same is true for amides 3 (see Figure 6c for 3g1),
the quantum yields of which remain near unity in all four
solvents.
3
√
∆µge ) µe-µg ) [(slope × 81) ⁄ (11307.6 × (6.2 ⁄ a) ] (1)
a(in Å) ) 3 [(3 × 1.33 × V ) ⁄ 4π]
(2)
√
w
The estimated ∆µge depends on a, the Onsager radius, the
determination of which is notoriously difficult. Various protocols
to estimate a are available in the literature;58 for coumarin dyes,
for example, the value has been taken as half the distance (in
Å) between the donor amine nitrogen and acceptor carbonyl
oxygen.57a We instead have used eq 2, where Vw is the calculated
volume of the van der Waals surface created for each of 1-3
using Discovery Studio Visualizer v. 2.0 (Accelyrs Software,
Inc.);59 multiplying this volume by 1.33 approximates the molar
volume in the liquid state.58a,60 The structures used for the
volume determinations (except 1d and 2d) could be produced
directly in the program from atomic modifications of the crystal
structure data presented earlier (for 1e, 1f, 2f, and 3f). For 1d
and 2d, simple energy-minimized structures were used (Mac-
roModel v. 9.1). The Onsager radii for the purines are given in
Table 7 and fall between 4.0 and 4.8 Å. Worth noting, this
method applied to PRODAN offers a value (4.0 Å) close to its
accepted radius of 4.2 Å but gives a smaller value (3.2 Å) than
the one generally used for 2-AP (3.8 Å).14m,o,61 Together with
A notable and inversed quantum yield trend emerges within
the “parent” a series (5a, 1a, and 2a), where the fluorescence
efficiency increases as the solvent polarity is increased (Figure
6). For example, the quantum yield of 1a increases from 0.34
in 1,4-dioxane to 0.88 in water (Figure 6a). The trend parallels
what is known for 2-AP9c,44a and 2-amino-9-methylpurine,62a
where less-efficient π-π* quenching by vibronic coupling to
the n-π* state accompanies more polar solvents.62a Series b
shows behavior between a and c, with reasonably high quantum
yields across the range of solvent polarities. How this behavior
is linked to the amino group (vide supra), for example through
hydrogen bonding or excited-state pyramidalization, is currently
unclear.
Chemical and Photochemical Stability. Purines 1-3 show
no significant decomposition by TLC or 1H NMR analysis after
months of being stored as solids in vials or in solutions of 1,4-
dioxane, dichloromethane, acetonitrile, or methanol with expo-
sure to air and ambient light. The photochemical stability of
5a, 5b, 1a, 1b, 1g, and 2b was further evaluated in nondegassed
N
the calculated slopes from the Stokes shift versus ET plots,
CH2Cl2 by monitoring the fluorescence intensity (at em λmax
)
∆µge values across 15 purines emerge between 2.0-3.1 D with
1,4-dioxane data included and 1.9-3.4 D without this data
(Table 7). These values are (a) similar to experimental values
for structurally related and extensively studied coumarin laser
dyes,55,57b (b) consistent with the degree of charge separation
suggested by theoretical analysis (vide infra), and (c) reasonable
based on values already determined for 2-aminopurine
experimentally14a,o,44a,62 (e.g., 1.6,44a 2.8,14o 2.9 D61b) and
computationally.14a,f,l,m Closer inspection reveals some expected
trends, many of which have been identified earlier. The ∆µge
values, for example, are, on average, larger for the esters 2 than
the nitriles 1 within any one series.
The quantum yield values for donor-acceptor purines 1-3
generally decrease with increasing solvent polarity (Figure 5)
with the exception of series a, which increases (Figure 6), and
b, which is quite constant. The decrease in quantum yield with
increasing polarity for donor-acceptor molecules is generally
likened to stabilization of the polar excited state that introduces
competing deactivation pathways.56 There are some exceptions
during constant irradiation (5a and 5b: λex ) 290 nm; 1a, 1b,
1g, and 2b: λex ) 320 nm) over 300-600 s. In each case, except
for 5b, the fluorescence intensity changed by less than 1% (see
the Supporting Information for details). For this unsubstituted
purine, even in thoroughly degassed CH2Cl2, the peak intensity
decreased ∼5% during each of three consecutive 5-min expo-
sures. The decomposition of the compound was confirmed by
1
TLC and H NMR, although the mixture was too complex to
interpret. Worth noting, the absorption spectra of both the
original compound and the decomposition mixture are similar
and the mixture maintains its fluorescence (although λem is red-
shifted to 377 nm); in other words, simply monitoring the
absorbance does not reliably report on the integrity of the
sample. In the remaining three solvents, all tested compounds,
except for 5b and 1b, are photochemically stable. The former
shows decomposition in all three, while 1b is sensitive in 1,4-
dioxane and methanol (ester 2b is stable in these solvents).
Electronic Structure Calculations. Both semiempirical and
high-level DFT calculations have been used to shed light on
the electronic structure of the donor-acceptor purines. The
ground-state geometries, dipoles, and orbital energies have been
obtained from DFT calculations at the B3LYP/6-311++G**
level (as implemented in Gaussian 0363). In all cases benzyl
groups (Bn) have been abbreviated as methyl substituents (Me)
to reduce computational cost (the truncated compounds are
referred to as 1b-Me, 2c-Me, etc.), and frequency calculations
have been performed to assign the optimized geometries as
energy minima or transition states. For esters 2-Me and the
amide 3 g-Me, the initial acceptor geometries used were the
ones identified by X-ray crystallography (Figure 2). To dem-
onstrate the accuracy of the calculations, Table S2 compares
the bond lengths for 2-AP14c and 1e from their X-ray crystal
(57) (a) Ravi, M.; Soujanya, T.; Samanta, A.; Radhakrishnan, T. P. J. Chem.
Soc., Faraday Trans. 1995, 91, 2739–2742. (b) Ravi, M.; Samanta,
A.; Radhakrishnan, T. P. J. Phys. Chem. 1994, 98, 9133–9136.
(58) (a) Wong, M. W.; Wiberg, K. B.; Frisch, M. J. J. Comput. Chem.
1995, 16, 385–394. (b) Karelson, M. M.; Zerner, M. C. J. Phys. Chem.
1992, 96, 6949–6957. (c) Wong, M. W.; Frisch, M. J.; Wiberg, K. B.
J. Am. Chem. Soc. 1991, 113, 4776–4782. (d) Edward, J. T. J. Chem.
Educ. 1970, 47, 261–270.
(59) Available free from http://accelrys.com.
(60) Similar values (within 0.1 Å) come from an alternative equation that
does not scale the calculated volume by 1.33, but rather adds 0.5 Å to
the final value of a.
(61) (a) Broo, A.; Holme´n, A. Chem. Phys. 1996, 211, 147–161. (b)
Gryczyn´ski, I.; Kawski, A. Bull. Acad. Pol. Sci., Ser. Sci., Math.,
Astron. Phys. 1977, 25, 1189–1196.
(62) (a) Rachofsky, E. L.; Osman, R.; Ross, J. B. A. Biochemistry 2001,
40, 946–956. (b) Kawski, A.; Bartoszewicz, B.; Gryczyn´ski, I.;
Krajewski, M. Bull. Acad. Pol. Sci., Ser. Sci., Math., Astron. Phys.
1975, 23, 367–372.
(63) Frisch, M. J. T., et al. Gaussian, Inc.: Wallingford, CT, 2004.
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 631

A R T I C L E S
Butler et al.
Table 8. Electronic Structure Data for 5a-Me and 1-3-Mea
HOMO
calcd (eV)
LUMO
calcd (eV)
∆E
calcd (eV)
∆E optical (eV)
[λcutoff (nm)]b
purine
µg
5a-Me
1a-Me
1b-Me
1c-Me
1d-Me
1e-Me
1f-Mec
1 g-Me
2a-Me
2b-Me
2c-Me
2d-Me
2e-Me
2f-Me
2 g-Me
3 g-Med
3.85
5.45
5.96
6.13
6.60
5.00
5.66
5.52
5.19
5.31
5.42
4.74
2.43
3.16
2.81
3.36
-6.03
-6.52
-6.24
-6.08
-6.29
-5.88
-5.75
-5.64
-6.24
-5.97
-5.82
-6.03
-5.62
-5.53
-5.40
-5.36
-1.23
-2.27
-2.21
-2.14
-1.87
-1.68
-1.68
-1.60
-2.04
-1.99
-1.94
-1.69
-1.55
-1.56
-1.48
-1.25
4.80
4.25
4.03
3.94
4.42
4.20
4.07
4.04
4.20
3.98
3.88
4.34
4.07
3.97
3.92
4.11
3.7 [332]
3.5 [352]
3.2 [385]
3.0 [410]
3.5 [352]
3.4 [368]
3.1 [396]
3.2 [385]
3.4 [370]
3.2 [385]
3.0 [410]
3.5 [355]
3.3 [380]
3.2 [390]
3.1 [405]
3.2 [386]
a See the Supporting Information for computational details. All
benzyl groups have been replaced by methyl groups for the calculations.
b Determined for 1-3 based on UV absorption data in CH2Cl2. The
λcutoff is defined here as the wavelength after the lowest energy λmax
where there is less than a 200 M-1 cm-1 variation over a 5 nm range
(procedure according to ref 19a). c One imaginary frequency was
identified for this structure indicating a transition state. The frequency is
associated with minor inversion of the C(2) amino group based on
analysis of the structure coordinate file. d Acceptor ) CONHCH3.
Figure 7. Representative orbital density plots (Molden v. 4.6) for the D-A
purines calculated from B3LYP/6-31++G** optimized geometries. The
HOMO (left) and LUMO (right) levels for series g-Me are shown (Table
7).
modest, but consistent with the experimental ∆µge values (vide
supra) and similarly structured fluorophores.
structures with the calculated 9-methyl analogues, 5a-Me and
1e-Me, respectively. The values are the same within ∼0.02 Å.
Likewise, the calculated ground-state dipole moment for 5a-
Me (3.85 D at the B3LYP/6-31++G** level) is similar to the
literature calculated value of 3.64 D (method ) CASSCF (10,14
CAS)).14l Nearly all of the optimized geometries do show a
slight pyramidalization of the C(2) and C(6) amino (or dim-
ethylamino) nitrogens (overall C1 symmetry), consistent with
the X-ray crystal structure of 2-AP14c and other calculated purine
structures.14a,f,m Thedistortionhassmallenergeticconsequences;14f,m
constraining the geometry of the dimethylamino group of 1e,
for example, provides no change in ground-state dipole moment
or in the HOMO and LUMO energies.
Summarized in Table 8 are the dipole and orbital energy data
obtained for 1-3-Me from the DFT calculations. Complementary
orbital density plots of the HOMO and LUMO levels could be
generated using Molden v. 4.664 to show graphically the
electronic nature of the ground states (Figure 7 shows the g-Me
series; others can be found in the Supporting Information). The
orbital features are quite similar within and between the nitriles,
methyl esters, and amides. The HOMO consists of delocalized
π orbitals on the purine plane, with the greatest density in the
N(9)-C(4)-C(5)-C(6) region (see Figure 1 for atom labels)
consistent with what has been observed for parent 2-AP.14e,l
Localized pz orbitals are seen on the exocyclic donor heteroa-
toms and the acceptor functionality, and there is slightly more
charge separation in the HOMO for the nitriles (consistent also
with their greater ground-state dipole moments; average for
nitriles ) 5.77 D, esters ) 4.15 D). The LUMO is also of π
character but more concentrated on the acceptor substituent,
the cyano group, methyl ester, or amide. Generally the density
on the C(2) donor group diminishes most substantially in the
LUMO; the C(6) donor orbital coefficient is only slightly
affected. The degree of charge separation in the molecules is
Some familiar trends emerge from Table 8. Comparing a-c-
Me, the HOMO and LUMO energies and ground-state dipole
moments (µg) increase with increasing donor strength at C(2),
and 1c-Me has the lowest calculated HOMO-LUMO gap. The
computational data parallels the experimental results where 1c
(versus 1b or 1a) has the most red-shifted abs λmax and lowest
optical band gap. Comparison of 1a-c-Me and 2a-c-Me
further shows that the esters (2) generally bring slightly lower
HOMO-LUMO gaps and µg values in the gas-phase, differ-
ences that do not emerge in solution. There is overall good
agreement between the trends in the calculated HOMO-LUMO
gaps and the optical gaps calculated from the onset of the low-
energy absorption maximum, although the former tend to be
up to 0.9 eV larger. Preliminary cyclic voltammetry (CV) data
collected for 1e-g in CH2Cl2 suggests that the calculated
HOMO values may be a bit low (0.2-0.4 eV) and the calculated
LUMO values a bit high (0.5-0.9 eV).65 Finally, ZINDO/S CI
semiempirical calculations66 (see the Supporting Information
for details) using the optimized DFT ground-state geometries
show that the lowest energy transition for the D-A purines is
associated with promotion of an electron from the HOMO to
the LUMO in all cases (S0 f S1). Given the large CI coefficient
for this transition (∼0.6), it is an allowed process for which the
trend in energy gap (327-354 nm; 3.79-3.50 eV) correlates
well with the trend in λmax in CH2Cl2 (311-362 nm; 3.99-3.43
eV).
Conclusions
“Push-pull” purines have been prepared by the introduction
of electron-accepting functional groups (-CN, -CO2CH3,
-CONHR) to the heterocyclic C(8) position to complement
typical electron-releasing groups at C(2) and C(6). The chemical
(64) Available from www.cmbi.ru.nl/molden/molden.html (see Schaftenaar,
(65) Data will be reported separately.
(66) Ridley, J. E.; Zerner, M. C. Theor. Chim. Acta 1976, 42, 223–236.
G. N., J. H. J. Comput.-Aided Mol. Design 2000, 14, 123-134).
9
632 J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009

Push-Pull Purines
A R T I C L E S
modifications result in significantly improved photophysical
properties relative to the acceptor-free molecules, important with
respect to potential biosensing or materials applications. These
include (1) red-shifted (20-50 nm) absorption maxima; (2)
highly solvatochromic emission profiles (em λmax from 355-466
nm depending on substitution pattern and solvent); (3) signifi-
cantly enhanced (up to 2500%) fluorescence quantum yields
(providing among the highest values for purines known); and
(4) improved photochemical stability upon continuous irradia-
tion. Access to carboxamides 3 using conventional peptide-
coupling chemistry demonstrates one approach to elaborating
the fluorophores while maintaining, even enhancing, their optical
properties. Derivatives like 3f that bear primary amino groups
further speak to use of the molecules as unnatural amino acids,
where their interesting molecular recognition features and optical
properties could be used in synthetic peptide architectures.
to be tolerated by the DNA duplex.67 On the materials side we
are exploring applications wherein efficient small-molecule
fluorophores have traditionally been successful, with the ultimate
goal of drawing (biorelevant)structure-function relationships
in this context. Useful in both cases will be purines that mirror
the better of the donor-acceptor arrangements presented here
and preserve at least one hydrogen-bonding edge for molecular
recognition. 2,6-Diaminopurine derivatives are promising targets
in this regard given that they are compatible with DNA, RNA,
and PNA structure and complementary to uracil and thymine.68
Acknowledgement. We are grateful to the University of Florida,
the Donors of the American Chemical Society Petroleum Research
Fund (PRF Grant No. 42229-G4), and the Research Corporation
(Research Innovation Award to R.K.C.) for financial support. R.S.B.
is a University of Florida Alumni Graduate Fellow. We thank Priya
Vijapura, funded through the UF Student Science Training Program,
Andrea K. Myers, funded through the NSF REU program (CHE-
0139505), and Prabhu Bellarmine for help with initial spectroscopic
studies. We additionally thank Aubrey L. Dyer for preliminary CV
measurements and Dr. Bobby V. Sumpter (Center for Nanophase
Materials Sciences, Oak Ridge National Laboratory) for compu-
tational advice. K.A.A. thanks the National Science Foundation and
University of Florida for funding the X-ray crystallography equip-
ment. We especially thank Dr. Katsu Ogawa, Prof. Adrian Roitberg,
Prof. Kirk Schanze, and Prof. Linda Bloom for guidance and/or
instrument access.
Comprehensive structure-property relationships for the pu-
rines have emerged from analysis of their photophysical data
in various solvents. While the absorption and emission maxima
have been shown to depend on donor and acceptor structure,
relative donor position, and medium in understandable ways,
the quantum yields have varied less intuitively. Ester (b) and
amide (c) acceptor groups have served to partially “neutralize”
the quantum yield sensitivity to solvent polarity that is otherwise
observed within the nitriles a and many other donor-π-acceptor
systems. Given this feature and that the acceptor-modified
purines enjoy excellent linear correlations between emission
N
energy and solvent polarity (ET ), the molecules should be useful
Supporting Information Available: Experimental procedures
and characterization data, crystallographic data for 2f and 3f,
additional photophysical data (including representative photo-
stability studies), computational details (including additional
molecular orbital plots and Cartesian coordinates and energies
for geometry-optimized structures), copies of 1H NMR spectra
for all new compounds, and complete ref 63. This material is
available free of charge via the Internet at http://pubs.acs.org.
in microenvironment sensing applications. Lastly, a quantitative
treatment of the solvent-dependent emission data has provided
insight into electronic structure; the calculated ∆µge values
(2.1-3.4 D), for example, are consistent with the degree of
charge separation reflected in calculated molecular orbital plots.
There is much left to do to exploit the “information content”
of biology’s heterocycles within niche materials and sensing
applications, and this pursuit is likely to offer both fundamental
and practical advances. Two avenues are available now that we
have imparted useful photophysical properties to simple purines
using readily modified substituents. Toward biological and
sensing applications, we are currently developing water soluble
versions of the purines that will allow us to understand how
they perform in biomolecular microenvironments. Encouraging
is that C(8) substituents like -CN, -CO2CH3, and -CONHCH3
appear not much larger than the vinyl (CHdCH2) group known
JA806348Z
(67) Gaied, N. B.; Glasser, N.; Ramalanjaona, N.; Beltz, H.; Wolff, P.;
Marquet, R.; Burger, A.; Me´ly, Y. Nucleic Acids Res. 2005, 33, 1031–
1039.
(68) (a) Pasternak, A.; Kierzek, E.; Pasternak, K.; Turner, D. H.; Kierzek,
R. Nucleic Acids Res. 2007, 35, 4055–4063. (b) Haaima, G.; Hansen,
H. F.; Christensen, L.; Dahl, O.; Nielsen, P. E. Nucleic Acids Res.
1997, 25, 4639–4643. (c) Gaffney, B. L.; Marky, L. A.; Jones, R. A.
Tetrahedron 1984, 40, 3–13.
9
J. AM. CHEM. SOC. VOL. 131, NO. 2, 2009 633