Synthesis and Characterization of Phenylnitroxide-Substituted Zinc (II) Porphyrins

Article abstract of DOI:10.1021/jo971736q

The syntheses and fluid-solution EPR spectra of nitroxide-substituted porphyrins 1-3 are described. The phenylnitroxide is attached to the meso-position of three porphyrins: (a) directly through a single bond (1); (b) through an ethenyl group (2); and (c) through an ethynyl group (3). The spin distribution in each of the three radicals is discussed and compared to model compounds 4-9. It is suggested that delocalization increases in the order 2, 1, 3.

Full text of DOI:10.1021/jo971736q

J . Org. Chem. 1998, 63, 769-774
769
Syn th esis a n d Ch a r a cter iza tion of P h en yln itr oxid e-Su bstitu ted
Zin c(II) P or p h yr in s
David A. Shultz,* Kevin P. Gwaltney, and Hyoyoung Lee
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204
Received September 16, 1997
The syntheses and fluid-solution EPR spectra of nitroxide-substituted porphyrins 1-3 are described.
The phenylnitroxide is attached to the meso-position of three porphyrins: (a) directly through a
single bond (1); (b) through an ethenyl group (2); and (c) through an ethynyl group (3). The spin
distribution in each of the three radicals is discussed and compared to model compounds 4-9. It
is suggested that delocalization increases in the order 2, 1, 3.
Ch a r t 1
We are preparing radical-substituted metalloporphy-
rins for construction of coordination polymers with
interesting magnetic properties.1,2 There are several
design motifs for such materials, including coupling the
organic radical spin with the unpaired electron of an
oxidized porphyrin, coupling with a transition metal spin,
or coupling with both. Conjugation of a radical with the
porphyrin ring (and therefore coupling with any para-
magnetic metal bound therein) will occur provided that
the torsion angle between the radical and the porphyrin
is not 90°. To avoid a severe torsion angle,3 a meso-aryl
ring can be separated from the porphyrin ring by, for
example, an ethenyl or an ethynyl group.4
In such cases, the conjugative effectiveness of the
fragment that links the porphyrin and the aryl group is
critical for sufficient spin-spin communication. Herein,
we describe the preparation and EPR spectral properties
of Zn(II) porphyrins 1-3, which will be used to examine
linker effectiveness and other structure-property rela-
tionships. Porphyrins 1-3 each have one phenylnitrox-
ide bound to a meso-position: directly (1); through a
carbon-carbon double bond (2); and through a carbon-
carbon triple bond (3). We chose to occupy the remaining
meso-positions with mesityl groups, which will not affect
nitroxide conjugation since mesityl torsion angles are
nearly 90°.3 The fluid solution EPR spectra of 1-3 are
compared to model compounds 4-9 (Chart 1).
Sch em e 1
(1) Shultz, D. A.; Knox, D. A.; Morgan, L. W.; Sandberg, K.; Tew,
G. N. Tetrahedron Lett. 1993, 34, 3975.
(2) For studies of porphyrinic magnetic materials, see the follow-
ing: (a) Conklin, B. J .; Sellers, S. P.; Fitzgerald, J . P.; Yee, G. T. Adv.
Mater. 1994, 6, 836. (b) Kamachi, M.; Cheng, X.; Aota, H.; Mori, W.;
Kishita, M. Chem. Lett. 1987, 2331. (c) Kitano, M.; Koga, N.; Iwamura,
H. J . Chem. Soc., Chem. Commun. 1994, 447. (d) Kitano, M.; Ishimaru,
Y.; Inoue, K.; Koga, N.; Iwamura, H. Inorg. Chem. 1994, 33, 6012. (e)
Koga, N.; Iwamura, H. Nihon Kagaku Naishi 1989, 1456. (f) Miller, J .
S.; Vazquez, C.; Epstein, A. J . J . Mater. Chem. 1995, 5, 707. (g) Miller,
J . S.; Bohm, A.; Vazquez, C. Inorg. Chem. 1996, 35, 3083. (h) Miller,
J . S.; Calabrese, J . C.; McLean, R. S.; Epstein, A. J . Adv. Mater. 1992,
4, 498.
(3) Typically, meso-aryl groups are twisted 60-90°: Scheidt, W. R.;
Lee, Y. J . In Structure and Bonding, Vol. 64; Buchler, J . W., Ed.;
Springer-Verlag: Berlin, 1987; pp 1-70.
Model compound 4 was prepared according to a modi-
fied version of the published procedure,5 while 5-TBS was
prepared from the (4-formylphenyl)-TBS-nitroxide 106
using the Wittig reaction, as shown in Scheme 1. Acety-
lene 6-TBS was prepared from Pd-catalyzed coupling of
(4) For meso-alkenyl- and -alkynylporphyrin syntheses, see the fol-
lowing: (a) Boyle, R. W.; J ohnson, C. K.; Dolphin, D. J . Chem. Soc.,
Chem. Commun. 1995, 527. (b) Higuchi, H.; Shimizu, K.; Ojima, J .
Tetrahedron Lett. 1995, 36, 5359. (c) LeCours, S. M.; Guan, H.-W.;
DiMagno, S. G.; Wang, C. H.; Therien, M. J . J . Am. Chem. Soc. 1996,
118, 1497. (d) Milgrom, L. R.; Yahioglu, G. Tetrahedron Lett. 1996,
37, 4069. (e) Yashunsky, D. V.; Ponomarev, G. V. Tetrahedron Lett.
1995, 36, 8485. (f) J iang, X.; Nurco, D. J .; Smith, K. M. Chem. Commun.
1996, 1759. (g) Arnold, D. P.; J ames, D. A. J . Org. Chem. 1997, 62,
3460. (h) Anderson, H. L. Tetrahedron Lett. 1992, 33, 1101.
(5) Calder, A.; Forrester, A. R. J . Chem. Soc., Chem. Commun. 1967,
682.
S0022-3263(97)01736-2 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/22/1998

770 J . Org. Chem., Vol. 63, No. 3, 1998
Shultz et al.
Sch em e 2
Sch em e 3
the latter are directly proportional to spin densities.15-19
We reasoned that torsion of the phenyl ring bearing the
nitroxide group in 1 would dramatically diminish inter-
action with the porphyrin, and CdC should be more
effective than CtC for spin delocalization. Therefore,
interaction of the nitroxide with the porphyrin might
increase from 1 to 3 to 2. However, steric interactions
between â-pyrrole hydrogens and CdC hydrogens in 2
(as with phenyl hydrogens in 1) that are absent in 3 could
alter the proposed ordering. Our reasoning that CdC
should be more effective than CtC for spin delocalization
in the absence of torsion is based on the resonance
structures A-D. Conjugation through CtC results in a
cumulenic resonance form, D, which is less favorable
than the corresponding noncumulenic form, B.
TMS-acetylene and (4-bromophenyl)-TBS-nitroxide 11,6
followed by removal of the TMS group. Radicals 4-6
were generated by fluoride deprotection of the TBS ethers
followed by PbO2 oxidation.
Nitroxide radicals 75-9 were prepared as shown in
Scheme 2. Boronic ester 12 was synthesized by trans-
metalation of 11, reaction with trimethylborate, and
subsequent condensation with pinacol.7 Suzuki coupling8
of 12 and commercially available bromides 13-15 pro-
vided the TBS ethers. Radical formation was accom-
plished as described for 4-6.
The syntheses of nitroxide porphyrins 1-3, shown in
Scheme 3, feature Pd-catalyzed coupling reactions of
5-iodo-10,15,20-trimesitylporphyrin with the appropriate
TBS-protected phenylnitroxide boronate,8 alkene,9 and
alkyne.10 Fluoride-promoted TBS deprotection and PbO2
oxidation provided nitroxide porphyrins 1-3. 5-Iodo-
10,15,20-trimesitylporphyrin was prepared by reacting
dipyrromethane,11 mesityldipyrromethane12 with 2 equiv
of mesitaldehyde according to the method of Lindsey12
and subsequently iodinating according to a modified
procedure of Dolphin.13,14
The fluid solution EPR spectra of 1-6, and spectral
simulations20 are shown in Figure 1, while EPR spectra
of model compounds 7-9 and simulations20 are shown
in Figure 2. Proton and nitrogen hfcc for 1-9 obtained
from the simulations are listed in Table 1. Unfortu-
nately, no hfcc due to porphyrinic nuclei were observed.
Experimental spectra for 1-3 exhibit asymmetric line
The spin distributions in 1-9 can be evaluated by
comparing the hyperfine coupling constants (hfcc), since
(6) Inoue, K.; Iwamura, H. Angew. Chem., Int. Ed. Engl. 1995, 34,
927.
(7) Lamba, J . J . S.; J ames, M.; Tour, J . M. J . Am. Chem. Soc. 1994,
116, 11723.
(8) Miyaura, N.; Yanagi, T.; Suzuki, A. Synth. Commun. 1981, 11,
513.
(9) The Heck conditions used were those reported by Yu: Bao, Z.;
Chen, Y.; Cai, R.; Yu, L. Macromolecules 1993, 26, 5281.
(10) LeCours, S. M.; Guan, H.-W.; DiMagno, S. G.; Wang, C. H.;
Therien, M. J . J . Am. Chem. Soc. 1996, 118, 1497.
(11) Wang, Q. M.; Bruce, D. W. Synlett 1995, 1267.
(12) Lee, C.-H.; Lindsey, J . S. Tetrahedron 1994, 50, 11427.
(13) Boyle, R. W.; J ohnson, C. K.; Dolphin, D. J . Chem. Soc., Chem.
Commun. 1995, 527.
(14) We found that using 0.5 equiv of bis(acetoxy)phenyl iodide:
iodine resulted in better yields of product, according to the procedure
of Merkushev: Merkushev, E. B.; Simakhina, N. D.; Koveshnikova,
G. M. Synthesis 1980, 486.
(15) Weissman, S. I. J . Chem. Phys. 1956, 25, 890.
(16) McConnell, H. M. J . Chem. Phys. 1956, 24, 764.
(17) McConnell, H. M.; Chestnut, D. B. J . Chem. Phys. 1958, 28,
107.
(18) Bersohn, R. J . Chem. Phys. 1956, 24, 1066.
(19) Wertz, J . E.; Bolton, J . R. Electron Spin Resonance; Chapman
and Hall: New York, 1986.
(20) Fluid solution EPR spectra were simulated using the follow-
ing: Duling, D. EPR Calculations for MS.-Windows NT/95, Version
0.96, Public EPR Software Tools, National Institute of Environmental
Health Sciences, National Institutes of Health, Research Triangle Park,
NC, 1996.

Phenylnitroxide-Substituted Zinc(II) Porphyrins
J . Org. Chem., Vol. 63, No. 3, 1998 771
F igu r e 1. X-band EPR spectra of porphyrins 1-3 (left) and model compounds 4-6 (right) in toluene recorded at 298 K. Each
plot includes a simulated spectrum, top (---), and an experimental spectrum, bottom (s). Best fit simulations were achieved
using hfcc listed in Table 1. The simulations were accomplished using a Simplex fitting routine, and all correlation coefficients
exceeded 0.98.20
broadening presumably due to either slow tumbling of
the large porphyrin molecules or anisotropic diffusion.21
Thus, the high-field lines are broadened relative to the
low-field lines, which in turn are broader than the central
lines, as is observed for viscous solutions of di-tert-
butylnitroxide.19
The hfcc values in Table 1 show that the aN decreases
and the ring-aH increases from 1-3, and from 4-6,
consistent with increased conjugation in the same
orderscontrary to our prediction based on resonance
structures A-D. However, this trend is expected on the
basis of the known correlation of arylnitroxide aN with
the substituent parameter, σp:22-26 -0.04 for CHdCH2
and +0.23 for CtCH.27 The more electron-withdrawing
CtC draws more spin density into the phenyl ring away
from the nitrogen than does CdC, not necessarily because
of increased conjugation, but because the electronega-
tivity of an sp-hybridized carbon is greater than an sp2-
hybridized carbon. Moreover, comparison of the dCH2
H-hfcc of 5 (aH ) 1.01 G) to the H-hfcc of the ≡CH proton
of 6 (aH ) 0.95 G) shows that slightly more spin density
reaches CdC than reaches CtC; thus CtC is better than
CdC at withdrawing spin into the phenyl ring, but not
quite as effective at further delocalization. It appears
that the supposed intrinsic difference in ability to con-
jugate between CdC and CtC is attenuated, and in this
case the two groups are nearly equal in their ability to
conjugate.
Examination of the phenyl ring H-hfcc and N-hfcc of 2
compared to 5 and 3 compared to 6 show that additional
spin density is delocalized away from the nitrogens in
porphyrins 2 and 3. Whether this additional spin density
(21) Hudson, A.; Luckhurst, G. R. Chem. Rev. 1969, 69, 191.
(22) Strom, E. T.; Bluhm, A. L.; Weinstein, J . J . Org. Chem. 1967,
32, 3853.
(23) Church, D. F. J . Org. Chem. 1986, 51, 1138.
(24) Lemaire, H.; Marechal, Y.; Ramasseul, R.; Rassat, A. Bull. Soc.
Chim. Fr. 1965, 372.
(26) Zhang, Y. H.; J iang, B.; Zhou, C. M.; J iang, X. K. Chin. J . Chem.
1994, 12, 516.
(27) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(25) Neugebauer, F. A.; Fischer, P. H. H. Z. Naturforsch 1966, 21b,
1036.

772 J . Org. Chem., Vol. 63, No. 3, 1998
Shultz et al.
compound 4 an appropriate model compound? Probably
not. Regardless of the torsion angle between the phenyl
and porphyrin rings, the carbon attached to the phenyl
ring in 1 is sp2 hybridized while the carbon bound to the
phenyl ring in 4 is sp3 hybridized, and therefore, aN(4) is
greater than aN(1), consistent with the cited substituent
effect (σp(t-Bu) ) -0.20).27 Perhaps compounds 7-9 are
better models of 1 since the aryl rings in 7-9 all have
sp2 carbons attached to the phenyl ring of the nitroxide.
However, comparing 1 to 7-9 ignores differences in
electron demand between phenyl and porphyrinyl. Un-
fortunately, to our knowledge there is no σp for porphy-
rinyl.
MM2 calculations28,29 on biphenyl, 2-methylbiphenyl,
and 2,4,6-trimethylbiphenyl give phenyl-aryl torsion
angles of 45°, 54°, and 62°, respectively. It is evident that
the aN values decrease as torsion increases. In fact, the
calculated torsion for 9 is quite close to measured phenyl
torsions in tetraphenylporphyrins.3 Nevertheless, aN(1)
is closest to aN(7), suggesting similar delocalization in 1
and 7. However, comparison of the spectrum of 7 with
that of 1 indicates that additional hf-induced line broad-
ening is absent in 1, and therefore, conjugation in 1 is
less than that in 7. Since the spin density delocalized
into the phenyl substituent of 7 is small, and this small
spin density is spread over several atoms, the substitu-
ent-hfcc are quite small and are manifested only in the
line width. We believe therefore that delocalization in 1
is closest to that in 9 and that the differences in aN
between 1 and 9 are because of the differences in
electronegativity between porphyrin and phenyl.
Because of the lack of porphyrin-hfcc and the absence
of a “linker” in 1, our hfcc analysis does not permit us to
distinguish with absolute certainty delocalization in 1 vs.
2 or 3. Moreover, we cannot ascertain whether porphyrin
rings have identical geometries in 1-3 and how porphy-
rin ring geometry might affect spin delocalization. Nev-
ertheless, we propose the following: (a) phenyl torsion
in 1 is substantial3 and we surmise that ethenyl torsion
in 2 is similarly large, such that delocalization in 2 is
less than in 1; (b) since the ethenyl and ethynyl frag-
ments are nearly equivalent at attracting spin density
to the carbon attached to the porphyrin, and ethenyl is
twisted relative to the porphyrin ring, 3 will be more
delocalized than 2; (c) since 3 can achieve planarity,
delocalization will be greater than in 1. Proposition (a)
is sensible since there is less spin density at the ethenyl
carbon attached to the porphyrin meso-position in 2 than
spin density at the para-carbon attached to the porphyrin
meso-position in 1. Thus, we propose that delocalization
increases in the order 2 < 1 < 3.
In summary, we prepared three new zinc porphyrins
bearing nitroxide radicals that are conjugated with the
porphyrin ring. The factors that affect spin delocalization
were examined, and a relative ordering of interaction of
the radical with the porphyrin ring was suggested. It
appears that CdC is only a slightly better conjugating
group than CtC for nitroxide radicals. As an addition
to the known substituent effect on aN, we showed that
increased torsion measurably alters aN. The limitations
of using hfcc for determining spin distributions in ni-
troxide radicals were pointed out. Further studies of spin
delocalization in molecules 1-3 are underway.
F igu r e 2. X-band EPR spectra of toluene solutions of model
compounds 7-9 recorded at 298 K. Each plot includes a simu-
lated spectrum, top (---), and an experimental spectrum, bottom
(s). Best fit simulations were achieved using hfcc listed in
Table 1. The simulations were accomplished using a Simplex
fitting routine, and all correlation coefficients exceeded 0.98.20
Ta ble 1. EP R 14N- a n d 1H-Hyp er fin e Cou p lin g Con sta n ts
for 1-9a
compd |aN|/G |ao-H|/G |am-H|/G
|aH|/Gb
line width/G
1
2
3
4
5
6
7
8
9
11.99
11.44
11.23
12.46
11.72
11.43
11.94
12.10
12.18
2.08
2.16
2.24
1.97
2.17
2.19
2.12
2.03
2.03
0.87
0.94
0.99
0.87
1.01
1.04
0.93
0.88
0.88
0.36
0.38
0.32
0.33
0.23
0.21
0.46
0.31
0.31
0.99, 1.31
0.77, 1.01c
0.95
a
0.1 mM toluene solutions at 298 K; hfcc (( 0.01 G) and line
widths from spectral simulations.20 Ethenyl, ethynyl, or aryl H.
b
c Terminal H.
is due to a change in electron demand of the substituents
(ethenyl vs. 2-porphyrinylethenyl and ethynyl vs. 2-por-
phyrinylethynyl) or due to extended conjugation awaits
further studies.
What about porphyrin 1, in which the phenyl nitroxide
is directly bound to the porphyrin meso-position? Is
(28) Allinger, N. L. J . Am. Chem. Soc. 1977, 99, 8127.
(29) Allinger, N. L.; Yuh, Y. H. QCPE 1981, 13, 395.

Phenylnitroxide-Substituted Zinc(II) Porphyrins
J . Org. Chem., Vol. 63, No. 3, 1998 773
J ) 4.7 Hz), 8.67 (s, 4H), 7.95 (d, 2H, J ) 8.6 Hz), 7.48 (d, 2H,
J ) 7.8 Hz), 7.32 (s, 4H), 7.29 (s, 2H), 2.67 (s, 6H), 2.64 (s,
3H), 1.89 (bs, 18H), 1.24 (s, 9H), 1.01 (s, 9H), 0.03 (bs, 6H);
13C NMR (CDCl3) δ 152.3, 151.8, 150.4, 150.1, 150.0, 149.6,
139.4, 139.0, 137.7, 131.6, 131.5, 131.4, 131.2, 130.9, 127.9,
125.6, 120.6, 120.4, 120.0, 100.0, 96.5, 92.3, 61.6, 26.5, 26.4,
21.9, 21.8, 21.7, 18.3, -4.3; UV-vis (CHCl3) λmax (log ꢀ) 623
(4.35), 615 (4.34), 571 (4.22), 535 (3.65), 444 (5.59), 313 (4.37);
IR (film) νmax 2960, 2927, 2857, 1610, 1500, 1450 cm-1. Anal.
Calcd for C65H69N5OSiZn: C, 75.81; H, 6.75. Found: C, 76.02;
H, 7.01.
Exp er im en ta l Section
Solvent distillations, synthetic procedures, and EPR sample
preparation were carried out under an argon or nitrogen
atmosphere. THF and toluene were distilled from sodium
benzophenone-ketyl prior to use. Chloroform for porphyrin
synthesis was distilled from K2CO3. Anhydrous DMF was
purchased from Aldrich. Boron trifluoride etherate was
purchased from Lancaster. NMR spectra were recorded on a
Varian 300 MHz spectrometer using either deuteriochloroform
as solvent and referenced to protiochloroform at 7.26 ppm for
1H spectra and 77.0 ppm for 13C spectra or deuteriomethylene
chloride as solvent and referenced to protiomethylene chloride
at 5.32 ppm for 1H spectra and 54.0 ppm for 13C spectra.
Elemental analysis was performed by Atlantic Microlab, Inc,
Norcross, GA. X-band EPR spectra were recorded on an IBM-
Bru¨ker E200SRC spectrometer. Compound 4 was prepared
as described in ref 5, except that the aryllithium was used
instead of the Grignard.
1-[N-ter t-bu tyl-N-(ter t-bu tyld im eth ylsiloxy)a m in o]-4-
eth en ylben zen e (5-TBS). A 50 mL Schlenk flask containing
methyltriphenylphosphonium bromide (1.58 g, 4.42 mmol) and
THF (25 mL) was cooled in an ice bath. n-Butyllithium (2.25
M in pentane, 1.96 mL, 4.41 mmol) was added slowly via
syringe. After being stirred for 3 h, the mixture was cooled to
-78 °C and a solution of 10 (1.30 g, 4.24 mmol, in 5 mL THF)
was added. Once stirred overnight, petroleum ether was
added and the resulting mixture was washed twice with
saturated aqueous NaCl. The organic mixture was dried (Na2-
SO4) and concentrated under reduced pressure. 5-TBS (0.969
g, 75%) was isolated by flash chromatography (SiO2, petroleum
ether): 1H NMR (CDCl3) δ 7.27 (d, 2H), 7.19 (d, 2H, J ) 8.1
Hz), 6.68 (dd, 1H, J ) 17.6 Hz, 11.0 Hz), 5.69 (d, 1H, J ) 17.8
Hz), 5.17 (d, 1H, J ) 10.8 Hz), 1.09 (s, 9H), 0.90 (s, 9H), -0.13
(bs, 6H); 13C NMR (CDCl3) δ 136.8, 134.4, 126.6, 125.6, 125.4,
112.9, 48.84, 26.35, 18.21, -4.40; IR (film) νmax 3088, 3037,
Zin c(II) 5-[4-[N-ter t-bu tyl-N-(ter t-bu tyldim eth ylsiloxy)-
a m in o]p h en yl]-10,15,20-tr im esitylp or p h yr in (1-TBS). A
25 mL flask containing compound 19 (70 mg, 0.082 mmol),
Pd(PPh3)4 (4.7 mg, 4.1 µmol), and compound 12 (44.1 mg, 0.123
mmol) in anhydrous DMF (4 mL) was stirred while sparging
with argon for 10 min. K3PO4 (0.1778 g, 0.123 mmol) was
added quickly with argon purging and heated for 7 h. Once
cool, the solvent was removed by vacuum transfer. The
mixture was dissolved in petroleum ether and filtered through
silica. The solvent was removed under reduced pressure. The
residue was purified by radial chromatography (1% ether-
pentane) to give a pink solid, 1-TBS (0.049 g, 60%): 1H NMR
(CDCl3) δ 8.89 (d, 2H, J ) 4.5 Hz), 8.74 (d, 2H, J ) 4.6 Hz),
8.70 (s, 4H), 8.09 (d, 2H, J ) 8.1 Hz), 7.65 (d, 2H, J ) 8.0 Hz),
7.29 (bs, 6H), 2.62 (s, 9H), 1.85 (s, 6H), 1.84 (s, 12H), 1.38 (s,
9H), 1.05 (s, 9H), -0.17 (bs, 6H); 13C NMR (CDCl3) δ 151.2,
150.7, 150.4, 150.4, 139.8, 139.8, 139.7, 139.6, 138.1, 134.0,
132.7, 131.6, 131.5, 130.9, 128.2, 124.1, 120.9, 119.3, 119.1,
61.6, 26.8, 26.6, 22.0, 21.9, 21.8, 18.6, -4.2; UV-Vis (CHCl3)
λmax (log ꢀ) 594 (3.33), 553 (4.06), 509 (3.39), 482 (3.42), 423
(5.39), 403 (4.37), 309 (4.08). IR (film) νmax 3116, 2958, 2927,
2857, 1610, 1525, 1496, 1480, 1458 cm-1; MS-FAB C63H69N5O-
SiZn calcd exact mass 1003.4563, obsd 1003.4571. Anal. Calcd
for C63H69N5OSiZn: C, 75.23; H, 6.91. Found: C, 75.38; H, 7.00.
Zin c(II) 5-[4-[N-ter t-bu tyl-N-(ter t-bu tyldim eth ylsiloxy)-
a m in o]p h en yleth en yl]-10,15,20-tr im esitylp or p h yr in (2-
TBS). In a 25 mL round-bottom flask fitted with a condensor
were heated 19 (100 mg, 0.117 mmol), 5-TBS (44 mg, 0.144
mmol), Pd(OAc)2 (2.3 mg, 0.010 mmol), tri-o-tolylphosphine
(11.9 mg, 0.039 mmol), triethylamine (29.6 mg, 0.293 mmol)
and DMF (10 mL) to 100 °C for 5 h. Once cool, CH2Cl2 was
added followed by washing with aqueous NH4Cl, drying with
Na2SO4, and removal of all solvent at reduced pressure. Radial
SiO2 chromatography (SiO2, 1% Et2O-petroleum ether) re-
sulted in 22 mg (19%) of pure 2-TBS: 1H NMR (CD2Cl2) δ
9.67 (d, 1H, J ) 16 Hz), 9.59 (d, 2H, J ) 4.7 Hz), 8.78 (d, 2H,
J ) 4.6 Hz), 8.63 (s, 4H), 7.85 (d, 2H, J ) 8.7 Hz), 7.46 (d, 2H,
J ) 8.1 Hz), 7.39 (d, 1H, J ) 16 Hz), 7.30 (s, 4H), 7.28 (s, 2H),
2.63 (s, 6H), 2.61 (s, 3H), 1.85 (bs, 18H), 1.22 (s, 9H), 1.00 (s,
9H), 0.02 (bs, 6H); 13C NMR (CD2Cl2) δ 151.9, 150.5, 150.3,
150.1, 142.6, 139.8, 139.6, 138.1, 135.5, 131.5, 131.0, 129.3,
128.2, 126.4, 126.3, 119.5, 119.1, 117.8, 61.8, 26.7, 26.6, 22.0,
21.9, 21.8, 18.5, -4.2; UV-vis (CHCl3) λmax (log ꢀ) 608 (4.13),
563 (4.28), 521 (3.86), 434 (5.44), 311 (4.44); IR (film) νmax 3099,
2957, 2927, 2856, 2733, 1609, 1498, 1477, 1460 cm-1; MS-FAB
2960, 2929, 2890, 2857, 1631, 1605, 1504, 1472, 1462 cm-1
.
Anal. Calcd for C18H31NOSi: C, 70.76; H, 10.23. Found: C,
70.49; H, 10.27.
1-[N-ter t-bu tyl-N-(ter t-bu tyld im eth ylsiloxy)a m in o]-4-
eth yn ylben zen e (6-TBS). A flask containing 11-TBS (0.844
g, 2.35 mmol), (trimethylsilyl)acetylene (0.67 mL, 0.463 g, 4.70
mmol), and triethylamine (10 mL) was sparged with argon for
30 min. Pd(PPh3)4 (0.277 g, 0.24 mmol) was added quickly
with argon purging succeeded by heating to reflux for 5 h. Once
cool, the mixture was filtered through SiO2 with petroleum
ether rinsing. The solvent was removed in vacuo, and the
residue was purified by flash chromatography (SiO2, petroleum
ether). The resulting trimethylsilyl-protected 6-TBS (0.612
g, 1.63 mmol) was stirred with K2CO3 (0.478 g, 3.46 mmol) in
THF-MeOH (3:1, 12 mL) for 16 h. Petroleum ether was
added, and the solution was washed with saturated aqueous
NaCl followed by drying (Na2SO4) and concetration under
reduced pressure. The residue was subjected to reversed-
phase medium-pressure liquid chromatography (LiChroprep
RP-18, methanol) with UV (254 nm) isolating 6-TBS (0.486
g, 68%): 1H NMR (CDCl3) δ 7.36 (d, 2H, J ) 8.7 Hz), 7.19 (d,
2H, J ) 8.1 Hz), 3.03 (s, 1H), 1.08 (s, 9H), 0.90 (s, 9H), -0.13
(bs, 6H); 13C NMR (CDCl3) δ 151.9, 131.3, 125.0, 118.1, 83.9,
76.4, 61.2, 18.0, -4.7; IR (film) νmax 3318, 3090, 3039, 2956,
2930, 2884, 2856, 2110, 1602, 1494, 1471 cm-1. Anal. Calcd
for C18H29NOSi: C, 71.23; H, 9.63. Found: C, 71.41; H, 9.52.
1-[N-ter t-bu tyl-N-(ter t-bu tyld im eth ylsiloxy)a m in o]-4-
p h en ylben zen e (7-TBS). A flask containing bromobenzene
(0.026 mL, 0.251 mmol) and Pd(PPh3)4 (0.0087 g, 0.0075 mmol)
in THF (10 mL) was stirred while sparging with argon for 20
min. Compound 12 (0.100 g, 0.279 mmol) and Na2CO3 (2 M,
0.28 mL) were added quickly with argon purging, and the
mixture was refluxed for 12 h. Once cool, the mixture was
filtered through a glass filter with petroleum ether rinsing to
remove inorganic solids. The resulting mixture was washed
twice with saturated aqueous NaCl followed by drying with
MgSO4. The solvent was removed under reduced pressure.
The remaining oil was purified by radial chromatography with
pentane to give 7-TBS (0.030 g, 30%): 1H NMR (CD2Cl2) δ
7.62 (d, 2H), 7.50 (d, 2H), 7.43 (t, 3H), 7.33 (d, 2H,), 1.14 (s,
9H), 0.94 (s, 9H), -0.07 (bs, 6H); 13C NMR (CD2Cl2) δ 151.0,
141.2, 137.7, 129.1, 127.3, 127.1, 126.1, 126.0, 61.3, 26.4, 26.3,
18.3, -4.5; IR (film) νmax 3032, 2958, 2930, 2857, 1605, 1485,
1472, 1462 cm-1; MS m/z 355 (M+, 17), 299 (38), 242 (62), 224
(100), 210 (25), 167 (87).
C
65H71N5OSiZn calcd exact mass 1029.4719, obsd 1029.4697.
Zin c(II) 5-[4-[N-ter t-bu tyl-N-(ter t-bu tyldim eth ylsiloxy)-
a m in o]p h en yleth yn yl]-10,15,20-tr im esitylp or p h yr in (3-
TBS). A flask containing 6-TBS (132 mg, 0.433 mmol), 20
(131 mg, 0.153 mmol), THF (6 mL), and triethylamine (1 mL)
was sparged with argon for 30 min. Pd(PPh3)4 (27 mg, 0.023
mmol) and CuI (9 mg, 0.046 mmol) were added quickly with
argon purging succeeded by stirring for 20 h. The solvent was
removed in vacuo, and the residue was purified by flash
chromatography (SiO2, 1% Et2O-petroleum ether): 157 mg;
99%; 1H NMR (CDCl3) δ 9.81 (d, 2H, J ) 4.7 Hz), 8.85 (d, 2H,
2′-Meth yl-4-[N-ter t-bu tyl-N-(ter t-bu tyld im eth ylsiloxy)-
a m in o]bip h en yl (8-TBS). A 50 mL flask containing 2-bro-

774 J . Org. Chem., Vol. 63, No. 3, 1998
Shultz et al.
motoluene (0.067 mL, 0.558 mmol) and Pd(PPh3)4 (0.0193 g,
0.0167 mmol) in THF (20 mL) was stirred while sparging with
argon for 5 min. Compound 12 (0.2 g, 0.558 mmol) and
aqueous Na2CO3 (2M, 0.56 mL) were added quickly with argon
purging and refluxed for 2 days. Once cool, the solvent was
removed by rotary evaporation, and the mixture was dissolved
in petroleum ether, filtered through a plug of Celite, and
concentrated under reduced pressure. The resultant oil was
purified by chromatography on a Chromatotron, eluting with
pentane to give 8-TBS (0.0507 g, 25%): 1H NMR (CD2Cl2) δ
7.28-7.14 (m, 8H), 2.23 (s, 3H), 1.14 (s, 9H), 0.92 (s, 9H), -0.09
(bs, 6H); 13C NMR (CD2Cl2) δ 150.0, 142.1, 138.6, 130.5, 130.01,
128.2, 125.9, 125.1, 61.1, 26.4, 20.7, 18.2, -4.5. Anal. Calcd
for C23H36NOSi: C, 74.53; H, 9.79. Found: C, 74.57; H, 9.52.
2′,4′,6′-Tr im eth yl-4-[N-ter t-bu tyl-N-(ter t-bu tyld im eth -
ylsiloxy)a m in o]bip h en yl (9-TBS). A 25 mL flask contain-
ing 2-bromomesitylene (0.086 mL, 0.558 mmol), Pd(PPh3)4
(0.0193 g, 0.0167 mmol), compound 12 (0.2 g, 0.558 mmol),
anhydrous DMF (4 mL), and distilled toluene (2 mL) was
stirred while sparging with argon for 10 min. K3PO4 (0.1778
g, 0.8376 mmol) was added quickly with argon purging and
heated for 2 days. Once cool, the solvent was removed by bulb-
to-bulb vacuum distillation and the residue dissolved in
petroleum ether, filtered through a plug of Celite column, and
evaporated under reduced pressure. The resultant oil was
purified by chromatography on a Chromatotron, eluting with
pentane to give, after evaporation of solvent, a white solid,
9-TBS (0.1229 g, 55%): 1H NMR (CD2Cl2) δ 7.29 (d, 2H), 6.96
(d, 2H), 6.91 (s, 2H), 2.29 (s, 3H), 1.97 (s, 6H), 1.14 (s, 9H),
0.92 (s, 9H), -0.08 (bs, 6H); 13C NMR (CD2Cl2) δ 150.0, 137.9,
136.6, 136.3, 128.5, 128.2, 125.8, 61.0, 26.3, 21.1, 20.7, 18.3,
(trifluoroacetoxy)iodobenzene (0.201 g, 0.467 mmol), and iodine
(0.096 g, 0.380 mmol) were stirred for 1 h. The mixture was
poured into CH2Cl2, washed with aqueous Na2S2O3, and dried
with MgSO4. The residue after solvent evaporation was
chromatographed on alumina (1-3% Et2O-petroleum ether).
20 (0.364 g, 73%): 1H NMR (CDCl3) δ 9.61 (d, 2H, J ) 4.9
Hz), 8.72 (d, 2H, J ) 4.7 Hz), 8.60 (s, 4H), 7.30 (s, 4H), 7.27 (s,
2H), 2.65 (s, 6H), 2.62 (s, 3H), 1.86 (bs, 18H), -2.47 (bs, 2H);
13C NMR (CD2Cl2) δ 147.2 (broad), 139.9, 139.8, 138.6, 138.5,
138.2, 137.4, 132.1, 131.3, 128.4, 119.7, 119.5, 77.80, 21.9, 21.8;
IR (film) νmax 3057, 2957, 2857, 2160, 2108, 1602, 1480 cm-1
.
Anal. Calcd for C47H43N4I: C, 71.38; H, 5.48. Found: C, 71.36;
H, 5.87.
Zin c(II) 5-Iod o-10,15,20-tr im esitylp or p h yr in (19).
A
mixture of 5-iodo-10,15,20-trimesitylporphyrin (0.363 g, 0.459
mmol), chloroform (10 mL), and zinc(II) acetate (0.202, 0.918)
was heated to a gentle reflux for 2 h. The mixture was poured
into CH2Cl2, washed with aqueous NaHCO3, and dried with
MgSO4. The residue after solvent evaporation was chromato-
graphed on alumina (1-3% Et2O-petroleum ether). 21 (0.351
g, 89%): 1H NMR (CDCl3) δ 9.72 (d, 2H, J ) 4.6 Hz), 8.80 (d,
2H, J ) 4.6 Hz), 8.68 (bs, 4H), 7.30 (s, 4H), 7.28 (s, 2H), 2.65
(s, 6H), 2.63 (s, 3H), 1.86 (bs, 18H); 13C NMR (CDCl3) δ 152.1,
151.0, 150.5, 150.4, 139.5, 139.0, 138.2, 137.8, 132.3, 131.8,
131.6, 127.9, 119.9, 80.0, 22.0, 21.9, 21.7; LD-MS C47H41N4IZn
calcd mass 852.2; obsd 852.3.
Gen er a l P r oced u r e for TBS-Dep r otection a n d Su bse-
qu en t Oxid a tion of Hyd r oxyla m in e P or p h yr in s. The
TBS-protected metalloporphyrin (>0.01 mmol) was dissolved
in 1 mL of THF and subsequently treated with 1 equiv of
Bu4NF (1 M in THF). Upon completion of the reaction (as
detected by TLC, usually <3 h), the solution was washed with
aqueous NH4Cl that had been sparged with argon for 30 min,
dried with Na2SO4, and concentrated with a stream of argon
providing the hydroxylamine. EPR samples were prepared in
a glovebox by dissolution of the hydroxylamine in an appropri-
ate amount of toluene (0.2 mM) and transfer to a quartz tube
containing an excess of PbO2. Oxidation for UV-vis and IR
samples was accomplished by stirring a THF solution of the
hydroxylamine (as prepared above) with 20 equiv of PbO2 for
1-3 h (completion detected by TLC), evaporating the THF with
a stream of argon and purifying with a small SiO2 column.
1: IR (film) νmax 3098, 2961, 2925, 2734, 1456, 1342, 1272,
1000, 802 cm-1; UV-vis (CHCl3) λmax (log ꢀ) 580 (3.26), 541
(4.15), 502 (3.50), 417 (5.38), 306 (4.21).
-4.6. Anal. Calcd for
C25H40NOSi: C, 75.31; H, 10.11.
Found: C, 75.41; H, 10.05.
4-[N-ter t-bu tyl-N-(ter t-bu tyldim eth ylsiloxy)am in o]ph en -
ylp in a col Bor on a te (12). In a 100 mL Schlenk flask, 11-
TBS (0.991 g, 2.76 mmol) was dissolved in 15 mL of distilled
THF and cooled to -78 °C. tert-Butyllithium (3.72 mL, 5.58
mmol) was slowly added over 40 min. The reaction mixture
was stirred for 1 h at -78 °C. Trimethyl borate (1.10 mL,
9.67 mmol) was added and stirred overnight. Pinacol (0.980
g, 8.29 mmol) was added and stirred for 10 h. The reaction
mixture was washed with saturated NaCl twice. The organic
layer was dried over MgSO4. The solvent was removed under
reduced pressure. The remaining yellow oil was purified by
radial chromatography (1% Et2O-petroleum ether) and then
recrystallized from methanol to give 12 (0.721 g, 64%) as a
white solid: 1H NMR (CDCl3) δ 7.67 (d, 2H), 7.23 (d, 2H), 1.34
(s, 12H), 1.08 (s, 9H), 0.90 (s, 9H), -0.14 (s, 6H); 13C NMR
(CDCl3) δ 154.1, 133.9, 124.4, 83.6, 61.1, 26.2, 26.1, 25.0, 18.0,
-4.6; IR (film) νmax 2979, 2930, 2858, 1605, 1564, 1472, 1462,
1360 cm-1. Anal. Calcd for C22H40NBO3Si: C, 65.17; H, 9.94.
Found: C, 65.07; H, 9.86.
2: IR (film) νmax 3110, 2958, 2922, 2854, 2732, 1454, 1205,
997, 797 cm-1; UV-vis (CHCl3) λmax (log ꢀ) 614 (4.31), 562
(4.35), 433 (5.46), 311 (4.55).
3: IR (film) νmax 3112, 2959, 2923, 2954, 2733, 2187, 1205,
997, 798 cm-1; UV-vis (CHCl3) λmax (log ꢀ) 621 (4.64), 566
(4.43), 526 (4.01), 468 (4.95), 441 (5.46), 433 (5.46), 317 (4.57).
5,10,15-Tr im esitylp or p h yr in . Dipyrromethane (1.33 g,
9.0 mmol), meso-mesityldipyrromethane (2.38 g, 9.0 mmol,)
and mesitaldehyde (2.67 g, 18.0 mmol) were dissolved in 1800
mL of chloroform. Boron trifluoride etherate (0.66 mL, 5.4
mmol) was added, and the mixture was stirred for 30 min.
TEA (6.3 mL, 45 mmol) and DDQ (6.13 g, 27 mmol) were then
added, followed by stirring for 1 h. The mixture was poured
over silica (120 mm × 8 in) and eluted with CH2Cl2 until all
porphyrins eluted. The porphyrins were then separated by
gravity elution, silica chromatography (petroleum ether to 1%
Et2O-petroleum ether), and alumina chromatography (petro-
leum ether to 3% Et2O-petroleum ether). 19 (0.716 g, 12%):
1H NMR (CDCl3) δ 10.23 (s, 1H), 9.39 (d, 2H, J ) 4.6 Hz),
8.99 (d, 2H, J ) 4.6 Hz), 8.86 (s, 4H), 7.46 (s, 4H), 7.43 (s,
2H), 2.78 (s, 6H), 2.76 (s, 3H), 2.03 (bs, 18H), -2.60 (bs, 2H);
13C NMR (CDCl3) δ 146.3 (broad), 139.8, 138.9, 138.3, 138.0,
131.7, 130.8, 130.4, 130.3, 128.1, 128.0, 118.4, 117.7 104.3,
22.0, 21.7. Anal. Calcd for C47H44N4: C, 84.90; H, 6.67.
Found: C, 85.06; H, 6.77.
Ack n ow led gm en t. This work is supported by the
National Science Foundation (CHE-9501085) under the
CAREER Program and the Research Corporation
(CS0127) under the Cottrell Scholars Program. Mass
spectra were obtained at the Mass Spectrometry Labo-
ratory for Biotechnology. Partial funding for the Facil-
ity was obtained from the North Carolina Biotechnology
Center and the National Science Foundation. The
MALDI-TOF mass spectrometer was funded in part by
the North Carolina Biotechnology Center. The GC-MS
instrument was purchased with funds from Hewlett-
Packard through their Grants for Universities Program.
Su p p or tin g In for m a tion Ava ila ble: Spectral data (51
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
5-Iod o-10,15,20-tr im esitylp or p h yr in . 5,10,15-Trimesi-
tylporphyrin (0.421 g, 0.633 mmol), chloroform (10 mL), bis-
J O971736Q