Synthesis and photophysical properties of light-harvesting arrays comprised of a porphyrin bearing multiple perylene-monoimide accessory pigments

Syn th esis a n d P h otop h ysica l P r op er ties of Ligh t-Ha r vestin g

Ar r a ys Com p r ised of a P or p h yr in Bea r in g Mu ltip le

P er ylen e-Mon oim id e Accessor y P igm en ts

Kin-ya Tomizaki,^†Robert S. Loewe,^†Christine Kirmaier,^‡J ennifer K. Schwartz,^‡

J ennifer L. Retsek,^‡David F. Bocian,*^,§Dewey Holten,*^,‡and J onathan S. Lindsey*^,†

Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204,

Department of Chemistry, Washington University, St. Louis, Missouri 63130-4899, and Department of

Chemistry, University of California, Riverside, California 92521-0403

jlindsey@ncsu.edu

Received April 8, 2002

We present the synthesis and characterization of new light-harvesting arrays containing two, four,

or eight perylene-monoimide accessory pigments attached to a zinc porphyrin. Each perylene is

substituted with one or three 4-tert-butylphenoxy substituents. A 4,3′- or 4,2′-diarylethyne linker

joins the perylene N-imide position and the porphyrin meso-position, affording divergent or

convergent architectures, respectively. The architectures are designed to provide high solubility in

organic media and facile perylene-to-porphyrin energy transfer, while avoiding charge-transfer

quenching of the excited porphyrin product. For the array containing four perylenes per porphyrin

in both nonpolar (toluene) and polar (benzonitrile) media and for the array containing eight perylenes

per porphyrin in toluene, the photoexcited perylene-monoimide dye (PMI*) decays rapidly (∼3.5

ps) and predominantly (g90%) by energy transfer to the zinc porphyrin to form the excited zinc

porphyrin (Zn*), which has excited-state characteristics (lifetime, fluorescence yield) comparable

(within ∼10%) to those of the isolated chromophore. For the array containing eight perylenes in

benzonitrile, PMI* decays ∼80% by energy transfer (forming Zn*) and ∼20% by hole transfer

(forming PMI^-Zn⁺); Zn* subsequently decays ∼20% by electron transfer (also forming PMI^-Zn⁺)

and ∼80% by the normal routes open to the porphyrin monomer (intersystem crossing, internal

conversion, fluorescence). In addition to rapid and efficient perylene-to-porphyrin energy transfer,

the broad blue-green to yellow absorption of the perylene dyes complements the blue absorption of

the porphyrin, resulting in excellent light harvesting across a significant spectral region. Collectively,

the work described herein identifies multiperylene-porphyrin arrays that exhibit suitable

photochemical properties for use as motifs in larger light-harvesting systems.

In tr od u ction

approach to increase the efficiency of porphyrin-based

light-harvesting arrays is to introduce accessory pigments

that absorb in regions of the spectrum where porphyrins

are relatively transparent. A number of accessory pig-

ments have been examined in conjunction with porphy-

rins, including carotenoids,²cyanine dyes,³xanthene

dyes,⁴boron-dipyrrin dyes,⁵coumarins,⁶and perylene

dyes.^7-13Perylenes display many attractive photophysical

and chemical features that make them good candidates

Photosynthetic systems deploy large numbers of pig-

ments in antenna complexes to harvest light and funnel

the resulting excited-state energy to the reaction centers.

Major challenges concerning the design of synthetic light-

harvesting arrays entail the choice of light-absorbing

components and the appropriate three-dimensional ar-

rangement of the pigments. The resulting array of

pigments must absorb strongly across the spectrum of

incident light and enable excited-state energy migration

among the pigments without competing electron-transfer

reactions. A large number of synthetic light-harvesting

arrays based on porphyrins have been prepared,¹wherein

the porphyrin serves as a synthetically malleable sur-

rogate for chlorophyll.

(2) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34,

40-48.

(3) Lindsey, J . S.; Brown, P. A.; Siesel, D. A. Tetrahedron 1989, 45,

4845-4866.

(4) Lindsey, J . S.; Prathapan, S.; J ohnson, T. E.; Wagner, R. W.

Tetrahedron 1994, 50, 8941-8968.

(5) Li, F.; Yang, S. I.; Ciringh, Y.; Seth, J .; Martin, C. H., III; Singh,

D. L.; Kim, D.; Birge, R. R.; Bocian, D. F.; Holten, D.; Lindsey, J . S. J .

Am. Chem. Soc. 1998, 120, 10001-10017.

Porphyrins absorb intensely in the blue region but only

weakly across the remainder of the visible spectrum. One

(6) Hecht, S.; Vladimirov, N.; Fre´chet, J . M. J . J . Am. Chem. Soc.

2001, 123, 18-25.

(7) Miller, M. A.; Lammi, R. K.; Prathapan, S.; Holten, D.; Lindsey,

J . S. J . Org. Chem. 2000, 65, 6634-6649.

^†North Carolina State University. Fax: 919-513-2830.

^‡Washington University.

(8) Prathapan, S.; Yang, S. I.; Seth, J .; Miller, M. A.; Bocian, D. F.;

Holten, D.; Lindsey, J . S. J . Phys. Chem. B 2001, 105, 8237-8248.

(9) Yang, S. I.; Prathapan, S.; Miller, M. A.; Seth, J .; Bocian, D. F.;

Lindsey, J . S.; Holten, D. J . Phys. Chem. B 2001, 105, 8249-8258.

^§University of California at Riverside.

(1) Burrell, A. K.; Officer, D. L.; Plieger, P. G.; Reid, D. C. W. Chem.

Rev. 2001, 101, 2751-2796.

Published on Web 08/14/2002

J . Org. Chem. 2002, 67, 6519-6534

6519

Tomizaki et al.

CHART 1

for accessory pigments with porphyrins. From a photo-

physical standpoint, perylenes have moderately large

extinction coefficients in the region between the porphy-

rin Soret and Q bands (450-500 nm),¹⁴very high

fluorescence quantum yields,^14,15and long singlet excited-

state lifetimes.¹⁶From a chemical standpoint, perylenes

are amenable to a modular building block synthesis, have

good solubility in organic solvents, and are uncharged,

thereby facilitating chromatographic purification.

transfer to an adjoining porphyrin but that the perylene-

bis(imide) dyes are much more prone to competitive

excited-state electron-transfer reactions.^8-11Accordingly,

for light-harvesting applications, we have focused on the

use of perylene-monoimide dyes as accessory pigments

with porphyrins. A major objective is to design soluble

perylene-porphyrin building blocks that provide good

spectral coverage, undergo energy transfer without com-

peting electron-transfer reactions in nonpolar or polar

media, and can be elaborated into oligomeric structures.

To begin identifying suitable perylene-porphyrin motifs,

we initially prepared and characterized dyads having a

common zinc porphyrin and differing only in the nature

of the substituted perylene, linker, and site of attachment

(o-, m-, p-) on the meso-aryl ring of the porphyrin (Chart

1).¹²Suitable solubility and photochemical properties

were obtained with perylenes that (1) bear one or three

4-tert-butylphenoxy substituents and (2) are linked to the

porphyrin via a diarylethyne linker that bridges the

perylene N-imide position and the porphyrin meso-

position.

In our prior studies of perylene-porphyrin dyads, we

found that both perylene-bis(imide) dyes and perylene-

monoimide dyes undergo rapid excited-state energy

(10) Yang, S. I.; Lammi, R. K.; Prathapan, S.; Miller, M. A.; Seth,

J .; Diers, J . A.; Bocian, D. F.; Lindsey, J . S.; Holten, D. J . Mater. Chem.

2001, 11, 2420-2430.

(11) Kirmaier, C.; Yang, S. I.; Prathapan, S.; Miller, M. A.; Diers,

J . A.; Bocian, D. F.; Lindsey, J . S.; Holten, D. Res. Chem. Intermed.,

in press.

(12) Loewe, R. S.; Tomizaki, K.-Y.; Thamyongkit, P.; Chevalier, F.;

Kirmaier, C.; Schwartz, J . K.; Diers, J . R.; Bocian, D. F.; Holten, D.;

Lindsey, J . S. Manuscripts in preparation.

(13) Hayes, R. T.; Wasielewski, M. R.; Gosztola, D. J . Am. Chem.

Soc. 2000, 122, 5563-5567.

In this paper, we present the synthesis and selected

physical properties of several perylene-porphyrin light-

harvesting arrays using the types of perylenes and

linkers found to be viable in the dyad studies. The

(14) Langhals, H. Chem. Ber. 1985, 118, 4641-4645.

(15) (a) Rademacher, A.; Ma¨rkle, S.; Langhals, H. Chem. Ber. 1982,

115, 2927-2934. (b) Ebeid, E. M.; El-Daly, S. A.; Langhals, H. J . Phys.

Chem. 1988, 92, 4565-4568.

(16) Ford, W. E.; Kamat, P. V. J . Phys. Chem. 1987, 91, 6373-6380.

6520 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

systems described contain multiple (2, 4, or 8) perylene-

monoimide accessory pigments joined to a zinc porphyrin.

In each array, the perylenes are attached to the 2,6- or

3,5-positions of the meso-phenyl ring of the porphyrin via

an arylethynylphenyl (aep) linker. The resulting projec-

tion of the perylene out of the plane of the porphyrin

should suppress cofacial aggregation of the porphyrins

in solution, thereby affording highly soluble light-

harvesting arrays. Isopropyl groups on the 2,6-positions

of the aryl unit attached to the N-imide site cause the

perylene and the N-aryl unit to be essentially perpen-

dicular, thereby enhancing the solubility of the perylene

dye. The photochemical properties of the arrays have

been surveyed by static absorption and fluorescence

measurements to identify multiperylene-porphyrin con-

structs suitable for further study. Fluorescence lifetime

and time-resolved absorption measurements were per-

formed on the largest architectures, which contain four

or eight perylenes per porphyrin. This work is a prelude

to the development of multiperylene-porphyrin building

blocks for elaboration into oligomeric light-harvesting

arrays.

tion.²¹We decided to employ approach B in the synthesis

of this target perylene-porphyrin array. The Sonogashira

coupling of commercially available 2,6-dichlorobenzalde-

hyde and (trimethylsilyl)acetylene was performed under

the conditions developed by Buchwald and Fu,²²which

allow for the Sonogashira coupling of aryl bromides at

room temperature (Scheme 1). We employed the catalytic

system of Pd(PhCN)₂Cl₂, CuI, and P(t-Bu)₃in dioxane

containing diisopropylamine. The reaction was performed

at 70 °C with excess (trimethylsilyl)acetylene (1.75 equiv

per chloride) to overcome the low reactivity of the aryl

chlorides. Analysis of the crude reaction mixture by GC

after 24 h revealed no starting material or monocoupled

product. Purification by filtration through silica, Kugel-

rohr distillation, and column chromatography afforded

1 in 37% yield.

Aldehyde 1, mesitaldehyde, and pyrrole (71 mM) were

then subjected to a mixed-aldehyde condensation with

BF₃‚O(Et)₂-ethanol cocatalysis¹⁹in CHCl₃at room tem-

perature for 1.5 h, followed by oxidation with DDQ. The

resulting mixture of porphyrins was treated to the

standard conditions for zinc metalation (5 equiv of Zn-

(OAc)₂‚2H₂O in CHCl₃/MeOH at room temperature).⁴

However, these conditions failed to furnish the zinc

chelate. Therefore, the metalation reaction was per-

formed at 100 °C using DMF as solvent. Purification by

column chromatography furnished meso-tetramesitylpor-

phinatozinc(II) as the dominant porphyrin product and

the desired porphyrin 2 in only 1.1% yield. The very low

yield with 2,6-diethynylbenzaldehyde is in accord with

the low yields obtained with other aldehydes bearing

large groups at the 2,6-positions such as 2,6-bis(trifluo-

romethyl)benzaldehyde,¹⁹2,4,6-triphenylbenzaldehyde,²³

2,6-dibromobenzaldehyde,²⁴or 9-anthraldehyde.²⁵Treat-

ment of 2 with TBAF in CHCl₃/THF (1:1) afforded the

2,6-diethynyl-substituted zinc porphyrin 2′ in 82% yield.

The last step in the preparation of the 2,6-disubstituted

perylene-porphyrin array 4 employed Sonogashira cou-

pling of 2′ and 2 equiv of bromoperylene building block

3.¹²The conditions for this coupling reaction were identi-

cal to those developed for the Sonogashira coupling of

iodoporphyrins and ethynylporphyrins²⁶with the follow-

ing modifications: (1) the reaction temperature was 60

°C to facilitate coupling of the less reactive bromo

substituent;²⁷(2) a 10:1 ratio of toluene/TEA was em-

ployed; (3) the reaction was performed at 10 mM perylene;

and (4) the ratio of [bromoperylene]:[Pd₂(dba)₃]:[P(o-tol)₃]

) 1:0.1:0.6.¹²Purification using a three-column proce-

Resu lts a n d Discu ssion

(1) Syn th etic Str a tegies. The preparation of the

perylene-porphyrin arrays follows one of two general

approaches. In approach A, a bis(perylene)aldehyde is

prepared by the Sonogashira coupling of a bromo- or

ethynylperylene with a diethynyl- or dibromobenzalde-

hyde. The resulting bis(perylene)aldehyde can then be

used as a starting material in a mixed-aldehyde conden-

sation (forming an A₃B-porphyrin), can be reacted with

a dipyrromethane (forming a trans-A₂B₂-porphyrin), or

can be reacted with pyrrole (forming an A₄-porphyrin).

In approach B, a mixed-aldehyde condensation affords a

diethynyl A₃B-porphyrin, which upon Sonogashira cou-

pling with a suitable bromoperylene affords the perylene-

porphyrin light-harvesting array. The two approaches

differ only in the order of the Sonogashira and porphyrin-

forming reactions. We note that the mixed-aldehyde route

to A₃B-porphyrins is somewhat unattractive, typically

producing a mixture of six porphyrins that is separated

by chromatography. However, the existing rational ap-

proaches for the preparation of A₃B-porphyrins have not

proved compatible with three mesityl groups,¹⁷as are

used in the perylene-porphyrin arrays described herein.

The mixed-aldehyde condensations¹⁸with mesitaldehyde

were performed using BF₃‚O(Et)₂-ethanol cocatalysis¹⁹

at a higher concentration²⁰(∼70 mM concentrations of

pyrrole and total ArCHO) than typically employed (10

mM).

(21) In preliminary work, a trans-A₂B₂porphyrin bearing two

perylenes was prepared (wherein an o-ethynylphenyl (ep) unit bridges

the 9-position of the perylene-monoimide and the meso-position of the

porphyrin). Two atropisomers were obtained. However, each atrop-

isomer proved unsuited for photochemical applications because of the

strong electronic coupling afforded by the ethynylphenyl linker (see

the Supporting Information).

(2) Con ver gen t P er ylen e-P or p h yr in Ar r a y. The

positioning of a perylene-monoimide at the 2,6-positions

of the meso-aryl ring of the tetraarylporphyrin was quite

attractive as a means of covering the face of the porphy-

rin, thereby suppressing porphyrin-porphyrin aggrega-

(22) Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fu, G. C. Org.

Lett. 2000, 12, 1729-1731.

(23) (a) Suslick, K. S.; Fox, M. M. J . Am. Chem. Soc. 1983, 105,

3507-3510. (b) Sugimoto, H.; Aida, T.; Inoue, S. Macromolecules 1990,

23, 2869-2875.

(17) Rao, P. D.; Dhanalekshmi, S.; Littler, B. J .; Lindsey, J . S. J .

Org. Chem. 2000, 65, 7323-7344.

(24) Collman, J . P.; Hampton, P. D.; Brauman, J . I. J . Am. Chem.

Soc. 1990, 112, 2986-2998.

(18) Lindsey, J . S.; Prathapan, S.; J ohnson, T. E.; Wagner, R. W.

Tetrahedron 1994, 50, 8941-8968.

(25) Cense, J .-M.; Le Quan, R.-M. Tetrahedron Lett. 1979, 3725-

3728.

(19) Lindsey, J . S.; Wagner, R. W. J . Org. Chem. 1989, 54, 828-

836.

(26) Wagner, R. W.; Ciringh, Y.; Clausen, C.; Lindsey, J . S. Chem.

Mater. 1999, 11, 2974-2983.

(27) Loewe, R. S.; Lammi, R. K.; Diers, J . R.; Kirmaier, C.; Bocian,

D. F.; Holten, D.; Lindsey, J . S. J . Mater. Chem. 2002, 12, 1530-1552.

(20) Wagner, R. W.; Li, F.; Du, H.; Lindsey, J . S. Org. Process Res.

Dev. 1999, 3, 28-37.

J . Org. Chem, Vol. 67, No. 18, 2002 6521

Tomizaki et al.

SCHEME 1

nances of the two isopropyl groups and the hydrogens

on the aryl group attached to the N-imide position of the

perylene. This architecture is designed to suppress co-

facial aggregation of porphyrins in solution.

(3) Diver gen t P er ylen e-P or p h yr in Ar r a ys. The

attachment at the 3,5-positions of the porphyrin meso-

aryl ring causes each perylene-monoimide to project away

from and out of the plane of the porphyrin.

Ar r a ys In cor p or a tin g Mon o-Ar yloxyp er ylen es.

The reaction of aldehyde 5,⁵mesitaldehyde, and pyrrole

(72 mM) with BF₃‚O(Et)₂-ethanol cocatalysis¹⁹in CHCl₃

at room temperature followed by oxidation with DDQ

afforded a mixture of porphyrins (Scheme 2). Treatment

of the partially purified porphyrin mixture with Zn-

(OAc)₂‚2H₂O followed by column chromatography af-

forded the zinc porphyrin building block 6 in 14% yield.

The reaction of porphyrin 6 with 2 equiv of bromo-

perylene 3 under Pd-coupling conditions afforded bis-

(perylene)porphyrin 7 in 13% yield.

We attempted to prepare a trans-A₂B₂-porphyrin array

bearing four perylene accessory pigments using approach

B. Aldehyde 5 and 5-mesityldipyrromethane (8)²⁸were

condensed using TFA in CH₂Cl₂at room temperature for

30 min, followed by oxidation with DDQ (Scheme 3).

These conditions are known to proceed without detectable

formation of byproducts derived from acidolysis of 5-mes-

ityldipyrromethane, followed by undesired recombination

(i.e., scrambling).²⁹Metalation with Zn(OAc)₂‚2H₂O af-

forded zinc porphyrin 9 in 15% yield. The tetraethyn-

ylporphyrin 9 serves as a benchmark compound for

spectroscopic characterization. Attempts to react 9 with

bromoperylene 3 under Pd-coupling conditions resulted

in a crude mixture containing the expected tetrakis-

(perylene)porphyrin array, but insufficient solubility in

toluene and THF precluded purification.

Ar r a ys In cor p or a tin g Tr is(a r yloxy)p er ylen es. The

higher solubility of perylenes bearing three aryloxy

groups (1-, 6-, and 9-positions) versus only one aryloxy

group (9-position) prompted us to employ the tris-

(aryloxy)perylenes 10¹²and 11¹²as precursors to per-

ylene-porphyrin arrays. We explored two routes for the

Sonogashira coupling leading to a suitable perylene-

aldehyde (13) for use in approach A (Scheme 4). The first

route employs bromoperylene 10 and 3,5-diethynylbenz-

aldehyde (5) (2:1 ratio) as the starting materials. The

reaction under copper-free conditions, similar to those

employed for the Sonogashira coupling of two porphy-

rins²⁶but with the ratios employed for 4, afforded only a

small amount (8.9%) of perylene-aldehyde 13 (entry 1,

Table 1). The same reaction in the presence of CuI failed

to produce perylene-aldehyde 13 (entry 2). We next

examined the reaction with the reverse ethyne-bromo

substitution pattern on the perylene and benzaldehyde.

The reaction of ethynylperylene 11 and 3,5-dibromobenz-

aldehyde (12) (2:1 ratio) with CuI cocatalysis afforded 13

in 31% yield (entry 3). A further increase in the amount

of perylene 11 (2.4:1 ratio) gave 13 in 66% yield (entry

4). Repetition of this reaction furnished 13 in 83% yield.

While of limited scope, the survey of the Sonogashira

dure²⁶afforded 4 in 27% yield. Proton NMR analysis

confirmed that the perylenes project over the face of the

porphyrin macrocycle, as evidenced by the upfield reso-

(28) Littler, B. J .; Miller, M. A.; Hung, C.-H.; Wagner, R. W.; O’Shea,

D. F.; Boyle, P. D.; Lindsey, J . S. J . Org. Chem. 1999, 64, 1391-1396.

(29) Littler, B. J .; Ciringh, Y.; Lindsey, J . S. J . Org. Chem. 1999,

64, 2864-2872.

6522 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

SCHEME 2

SCHEME 3

hyde, and pyrrole (71 mM) with BF₃‚O(Et)₂-ethanol

cocatalysis¹⁹in CHCl₃(Scheme 4). Oxidation with DDQ

followed by zinc metalation afforded perylene-porphyrin

14 in 18% yield, following preparative-scale size exclusion

chromatography (SEC) and one silica gel column. This

array contains two perylenes attached at the 3,5-positions

of the porphyrin meso-aryl ring.

The condensation of perylene-aldehyde 13 with 5-mes-

ityldipyrromethane (8) was carried out under nonscram-

bling reaction conditions²⁹[(1) TFA, CH₂Cl₂, room tem-

perature, 30 min; (2) DDQ]. Metalation of the free base

porphyrin using Zn(OAc)₂‚2H₂O followed by purification

afforded perylene-porphyrin 15 in 28% yield (Scheme

5). This array contains four perylenes attached at the 3,5-

positions of the trans-porphyrin meso-aryl rings. A space-

filling model of 15 is shown in Figure 1.

The condensation of perylene-aldehyde 13 and pyrrole

was performed in the presence of BF₃‚O(Et)₂and NaCl

in CH₂Cl₂at room temperature for 1 h (Scheme 6). The

addition of NaCl to the porphyrin-forming reaction has

been shown to increase the yield of porphyrin by up to

2-fold in some cases.³⁰Oxidation with DDQ followed by

zinc insertion afforded perylene-porphyrin 16 in 31%

yield. This perylene-porphyrin light-harvesting array

contains eight perylene accessory pigments, with exhaus-

tive substitution at the 3,5-positions of each of the four

meso-phenyl rings of the porphyrin. The presence of eight

perylenes and 24 aryloxy substituents results in a sizable

mass (8264 Da) for the array to which the porphyrin

makes a minor contribution.

(4) Ch em ica l Ch a r a cter iza tion . Each perylene-

porphyrin array was characterized by analytical SEC,

thin-layer chromatography, laser desorption mass spec-

trometry (LDMS), ¹H NMR spectroscopy, absorption

spectroscopy, and fluorescence emission spectroscopy.

SEC Beh a vior . Each purified array exhibited a single

sharp peak in the analytical SEC chromatogram. The

retention times were consistent with molecular sizes: 4

reaction of perylene 10 or 11 with aldehyde 5 or 12

revealed important reactivity differences between a bro-

moperylene and an ethynylperylene. Compound 13 proved

to be very soluble, thereby enabling porphyrin-forming

reactions.

An A₃B-porphyrin was prepared using a mixed-alde-

hyde condensation of perylene-aldehyde 13, mesitalde-

(30) (a) Li, F.; Yang, K.; Tyhonas, J . S.; MacCrum, K. A.; Lindsey,

J . S. Tetrahedron 1997, 53, 12339-12360. (b) Geier, G. R., III; Riggs,

J . A.; Lindsey, J . S. J . Porphyrins Phthalocyanines 2001, 5, 681-690.

J . Org. Chem, Vol. 67, No. 18, 2002 6523

Tomizaki et al.

TABLE 1. Con d ition s for Son oga sh ir a Rea ction

SCHEME 4

Yield in g Bis(p er ylen e)a ld eh yd e 13^a

copper

entry source ligand temp (°C) reactants (mM) yield (%)

1

2

3

4

none P(o-tol)₃

60

50

10 (10)

11 (10) 12 (5)

11 (12) 12 (5)

5 (5)

8.9

0

31

CuI

PPh₃

66, 83

a

Toluene/TEA (5:1) as solvent; Pd₂(dba)₃as palladium source

for all entries.

SCHEME 5

For example, 4 and 7 have identical molecular weight

yet showed a difference in retention times of 0.2 min,

which stems from the convergent versus divergent ar-

chitecture of the perylene-porphyrin arrays. A similar

observation was noted for the RR- versus Râ-atropisomers

of a perylene-porphyrin joined via a 2-ethynylphenyl

linker (see the Supporting Information).

Solu bility. High solubility in a variety of solvents is

critical for purification, characterization, and photochemi-

cal studies. We have found each of the perylene-

porphyrin arrays to be sufficiently soluble for routine

chemical processing. The highest solubility was noted for

arrays 14-16, wherein each perylene bears three 4-tert-

butylphenoxy substituents. These arrays are soluble in

hexanes/CH₂Cl₂(1:1) for preparative chromatography,

are soluble in minimal amounts of CDCl₃for NMR

(9.7 min), 7 (9.5 min), 14 (9.2 min), 15 (8.7 min), and 16

(8.4 min). In addition to providing evidence for purity,

in a few instances, evidence also was obtained concerning

the relative hydrodynamic volumes of isomeric arrays.

6524 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

SCHEME 6

F IGURE 1. Space-filling model of perylene-porphyrin array

15 showing the projection of the perylenes with respect to the

porphyrin.

spectroscopy, and afford 10 mM solutions in toluene,

CHCl₃, CH₂Cl₂, THF, or benzonitrile. The high solubility

of these perylene-porphyrin model compounds augurs

well for the use of these motifs in larger light-harvesting

arrays.

(5) P h otop h ysica l Ch a r a cter iza tion of Ben ch -

m a r k P er ylen e Dyes. The 4-tert-butylphenoxy substit-

uents were introduced as a means of imparting greater

solubility to the perylene-porphyrin arrays. The pres-

ence of the aryloxy substituents causes significant changes

in the absorption spectrum of the perylene dye. Bench-

mark perylene-monoimide dyes bearing zero, one, or

three 4-tert-butylphenoxy substituents were prepared

and characterized; these dyes are PMI-1,³¹PMI-2,³²PMI-

(OR),¹²and 11,¹²with the latter hereafter termed PMI-

(OR)₃′ for comparison purposes (Chart 2). A summary of

the photophysical properties of these three reference

compounds is given in Table 2.

tion to the spectral shift stems from substitution at the

9-position of the perylene; further substitution of two

aryloxy groups at the flanking positions (1-, 6-) of the

perylene has comparatively little effect on the spectral

properties.

The absorption spectra of the benchmark perylene dyes

are shown in Figure 2 (solid). The perylene-monoimides

display a broad absorption manifold (fwhm ∼ 75 nm)

consisting of a series of substantially overlapping vibronic

transitions with a modest molar extinction coefficient

(ꢀ_max∼ 35 000 M^-1cm^-1). The unsubstituted perylene-

monoimide exhibits the (0,0) band at 507 nm and the

comparably intense (1,0) band at 479 nm with progres-

sively weaker (2,0) and (3,0) features at approximately

455 and 430 nm. The presence of one or three 4-tert-

butylphenoxy substituents results in a bathochromic shift

of approximately 30 nm, such that the origin band moves

from 506 nm in PMI-2 to 532 nm in PMI(OR) and 536

nm in PMI(OR)₃′ (Table 2). Thus, the dominant contribu-

The larger substituent effect at the perylene 9-position

derives from the high electron density at this site relative

to other sites.³³In addition to the substantial spectral

shifts shown in Figure 2 for the dyes that employ the

solubilizing aryloxy groups, we have found that incorpo-

ration of ethyne or ethynylphenyl groups at the 9-position

causes 12 or 25 nm red shifts, respectively, in the origin

transition of perylene-monoimide PMI-1, which differs

from PMI-2 only in the substituents on the aryl ring at

the N-imide position (Chart 2).¹⁰Even larger shifts have

been seen in a perylene-monoimide dye bearing an

N-pyrrolidinyl substituent at the 9-position (along with

4-tert-butylphenoxy substituents at the 1- and 6-posi-

tions), which moves the origin band by about 95 nm to

601 nm.³⁴Such a large shift in this band and the

(31) Feiler, L.; Langhals, H.; Polborn, K. Liebigs Ann. 1995, 1229-

1244.

(32) Quante, H.; Mu¨llen, K. Angew. Chem. 1995, 107, 1487-1489.

(33) Adachi, M.; Murata, Y.; Nakamura, S. J . Phys. Chem. 1995,

99, 14240-14246.

J . Org. Chem, Vol. 67, No. 18, 2002 6525

Tomizaki et al.

CHART 2

F IGURE 2. Absorption spectra (solid) and fluorescence

spectra (dashed) of benchmark perylene monomers PMI-2,

PMI(OR), and PMI(OR)₃′ in toluene at room temperature.

mirror image to the absorption manifold (Figure 2,

dashed). The (0,0), (0,1), (0,2), and (0,3) emission features

for PMI-2 are at approximately 530, 570, 617, and 680

nm, respectively. In analogy to the absorption spectrum,

the fluorescence spectrum red shifts upon incorporation

of 4-tert-butylphenoxy substituents. The origin fluores-

cence band of PMI-2 at 529 nm moves to 567 nm for PMI-

(OR) and 577 nm for PMI(OR)₃′ (Table 2). On the other

hand, the presence of solubilizing aryloxy groups does

not significantly affect the fluorescence quantum yield

of the perylene-monoimide dye (Table 2). The high

fluorescence yields of Φ_f) 0.91, 0.82, and 0.86 for PMI-

2, PMI(OR), and PMI(OR)₃′, respectively, indicate that

the excited singlet state of each of these dyes has

negligible nonradiative decay competing with the emis-

sion process. These perylene benchmark compounds also

were found to display long and monophasic fluorescence

lifetimes (τ ∼ 5 ns; Table 2). These data demonstrate that

the perylene-monoimide building blocks employed in this

study (bearing one or three aryloxy groups) are suitable

accessory pigments for porphyrins.

(6) P h otop h ysica l Ch a r a cter iza tion of P er ylen e-

P or p h yr in Ar r a ys. Absor p tion Sp ectr oscop y. Figure

3 shows the absorption spectra of multiperylene-por-

phyrin arrays 14, 15, and 16, along with the spectra of

reference compounds porphyrin 6 and perylene 11. There

are several noteworthy points: (1) A small but systematic

bathochromic shift is observed in the Soret band as the

number of perylenes is increased, with an overall shift

of 8 nm upon going from an unsubstituted porphyrin to

a porphyrin bearing eight perylenes. A similar effect was

TABLE 2. Absor p tion a n d F lu or escen ce P r op er ties of

P er ylen e Ben ch m a r k Dyes^a

fwhm

compound

λ_abs(log ꢀ)

(nm)

λ

_em(nm)

Φ_f

τ (ns)

PMI-2

PMI(OR)

PMI(OR)₃′ 511 (4.5), 536 (4.6)

479 (4.5), 506 (4.5)^b

507 (4.6), 532 (4.5)

75

82

71

529, 569 0.91^c5.0^d

567, 611 0.82

577, 623 0.86

4.8

5.0

a

All data for toluene at room temperature with λ_exc) 490 nm.

Only the wavelengths of the (0,0) and (1,0) absorption bands and

the (0,0) and (0,1) emission bands are listed, although (at least)

four overlapping vibronic bands are observed in each spectrum.

b

The errors in the fluorescence lifetimes are (0.2 ns. Values from

literature ) 484 (4.5), 506 (4.5).⁴⁷^cValue from literature ) 0.90.⁴⁸

Value from literature ) 4.6 ns.⁴⁸The value for

compound (PMI-1) is 4.8 ( 0.1 ns.

a related

d

corresponding emission features would be too large for

our purposes, as this would reduce spectral overlap for a

through-space contribution to energy transfer with the

porphyrin (which supplements the through-bond contri-

bution¹⁰). Perylene-monoimide dyes each bearing four

aryloxy substituents (at the 1-, 6-, 7-, and 12-positions)

have also been prepared, but spectral data have not been

reported.³⁵

Each of the benchmark perylene-monoimide dyes has

a broad fluorescence contour that is approximately a

(34) Gosztola, D.; Niemczyk, M. P.; Wasielewski, M. R. J . Am. Chem.

Soc. 1998, 120, 5118-5119.

(35) Holtrup, F. O.; Mu¨ller, G. R. J .; Quante, H.; De Feyter, S.;

DeSchryver, F. C.; Mu¨llen K. Chem.sEur. J . 1997, 3, 219-225.

6526 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

TABLE 3. Absor p tion a n d F lu or escen ce P r op er ties of P er ylen e-P or p h yr in Ar r a ys a n d Refer en ce P or p h yr in s

Φ_f^c

τ^d(ns)

no. of perylene

pigments

a

b

compound

λ_max(nm)

ꢀ_perylene/ꢀ_Soret(%)

toluene

PhCN

0.050

toluene

PhCN

ZnTPP

0

2

4

8

423

424

e

0.033^f

0.039ⁱ

0.034^l

0.034

0.028

0.036

0.044

0.050

2.1^g

2.4^j

2.4ⁿ

2.2

1.9

1.8

2.0^o

2.1

ZnTMP^h

ZnU^k

9

0.050^m

0.041

0.013

0.045

0.047

0.070

0.050

e

4

7

423, 509, 541

424, 511, 544

424, 508, 543

426, 508, 538

431, 507, 533

16

15

16

40

73

14

15

16

2.1

2.0

1.7

a

b

Absorption maxima in toluene at room temperature. ꢀ_perylene) ꢀ_maxfrom 509 to 544 nm in toluene. ^cFluorescence emission yields

((10%) were determined upon excitation (423 nm) of solutions in toluene or benzonitrile at room temperature by ratioing to that of

ZnTPP in toluene. A small fraction of the observed emission stems from the perylene. Fluorescence lifetime ((0.2 ns). ^eNot applicable.

d

^fReference 37. ^gValues in the range 2.0-2.1 ns were determined here and in ref 38. ZnTMP is zinc(II) tetramesitylporphyrin. Reference

h

i

j

k

39. Average of 2.3 ns found here and 2.4 and 2.5 ns found in ref 39. ZnU is zinc(II) 5-[4-[2-(trimethylsilyl)ethynyl]phenyl]-10,15,20-

l

m

trimesitylporphyrin. References 38 and 39a. Values in the range 0.04-0.06 with an average of 0.05 have been measured for this

compound in a number of polar coordinating solvents. Agrees with the values 2.4 and 2.5 ns reported in ref 38. ^oValues in the range

n

2.3-2.6 ns are reported in ref 38 for other polar coordinating solvents, including acetonitrile.

F IGURE 4. Schematic state diagram for the key processes

open to the perylene (PMI) and porphyrin (Zn) in the arrays.

These processes are energy transfer (ENT), hole transfer (HT),

and electron transfer (ET).

compound 9 were analyzed by fluorescence emission

spectroscopy. Measurements were performed in both

toluene and benzonitrile to garner information as to the

excited-state behavior of these systems in nonpolar and

polar media, respectively. To facilitate the presentation

of the photophysical data, in Figure 4, we depict the most

likely energy-transfer (ENT), hole-transfer (HT), and

electron-transfer (ET) processes in the perylene-porphy-

rin arrays that compete with the emissive and nonra-

diative decay pathways operative in the isolated chromo-

phores.^7-12

Excitation of a perylene-porphyrin array at 490 nm,

where the perylene dye absorbs preferentially, results in

F IGURE 3. Absorption spectra for perylene-porphyrin arrays

an emission characteristic of both the perylene (500-700

14, 15, and 16, reference porphyrin 6, and perylene 11. All

nm) and the zinc porphyrin [Q(0,0) and Q(0,1) bands at

spectra were collected at room temperature in toluene.

595 and 650 nm, respectively]. The yield of emission from

noted in the architecturally similar array bearing eight

boron-dipyrrin dyes.⁵(2) The absorption due to the

perylene in the arrays increases essentially linearly upon

going from two to four to eight perylenes (on the basis of

the normalized porphyrin Soret band). (3) The absorption

spectrum in the visible region of each perylene-porphy-

rin array is well approximated by the sum of the spectra

of the component parts. This observation is indicative of

the relatively weak electronic coupling between the

perylenes and porphyrin. The absorption properties of the

arrays are summarized in Table 3.

the perylene is diminished at least 30-fold from that of

the benchmark perylene monomer. These results are

consistent with efficient energy transfer to the porphyrin.

In each case, the observation of a significant level of

emission from the perylene and from the porphyrin

reflects the fact that the inherent quantum yield of

fluorescence for the perylene donor is very high (Φ_f)

0.8-0.9; Table 2),¹²whereas that for the zinc porphyrin

is relatively low (Φ_f) 0.03-0.05, depending on substit-

uents and solvent^8,36-39). The energy-transfer yields in

F lu or escen ce Sp ectr oscop y. Each perylene-por-

phyrin array (4, 7, 14-16) and porphyrin reference

(36) Seybold, P. G.; Gouterman, M. J . Mol. Spectrosc. 1969, 31,

1-13.

J . Org. Chem, Vol. 67, No. 18, 2002 6527

Tomizaki et al.

the multiperylene-porphyrin arrays are determined

more accurately from the time-resolved measurements

described below.

meso-aryl ring of the porphyrin, reducing the distance

from the Zn atom of the porphyrin to the middle of the

central benzenoid ring of the perylene from 17.5 to 11.1

Å.

The fluorescence yield of the zinc porphyrin in each

array also was examined (λ_exc) 423 nm) in order to

identify any quenching interactions due to the presence

of the appended perylene dyes. Quantum yield measure-

ments were performed in both toluene and benzonitrile

because the rates of charge-transfer quenching processes

are influenced by the effect of the solvent polarity on the

free energies of charge-separated product states, such as

PMI^-Zn⁺(Figure 4). In toluene, each perylene-porphy-

rin array has a fluorescence yield Φ_f) 0.028-0.050. Up

to 10% of the emission could be due to residual perylene

fluorescence elicited by direct excitation of this chro-

mophore. Accordingly, the fluorescence from the porphy-

rin in the array is comparable to or slightly greater than

We also measured the fluorescence lifetimes of the

largest arrays (and several reference compounds) in both

toluene and benzonitrile in order to obtain a second

measure of the extent to which the excited porphyrin may

be affected by the presence of multiple appended acces-

sory pigments. These excited-state lifetimes (all (0.2 ns)

are given in Table 3. The lifetimes for 9 (2.2 ns), 15 (2.1

ns), and 16 (2.1 ns) in toluene are virtually the same.

The values for 9 (2.1 ns) and 15 (2.0 ns) in benzonitrile

may be slightly shorter than those in toluene, consistent

with the expected slightly increased k_fdue to metal

coordination. The lifetime for 16 in benzonitrile (1.7 ns)

shows the only clear (yet modest) reduction in value

compared to that of the reference porphyrin. Analysis of

the lifetimes for 15 and 9 using standard methods^8,38,39

indicates that there is little quenching of the excited zinc

porphyrin in 15 due to the presence of the four appended

perylenes in either toluene or benzonitrile. A similar

analysis for 16 indicates little effect on the porphyrin due

to the presence of eight perylenes in toluene. For ex-

ample, the quenching yield for 15 and 16 in toluene is

Φ_Q) (1-2.1 ns/2.2 ns) × 100 ∼ 5% and that for 15 in

benzonitrile is Φ_Q) (1-2.0 ns/2.1 ns) × 100 ∼ 5%;

however, in each of these cases, the quenching yield could

be ∼10% given experimental uncertainty. The eight

perylenes in 16 give a somewhat larger (yet still modest)

quenching of the excited porphyrin in benzonitrile (Φ_Q

) (1-1.7 ns/2.1 ns) × 100 ∼ 20%). In each case, the

quenching is likely due to PMI Zn* f PMI^-Zn⁺electron

transfer, as illustrated in Figure 4.^8-12

Tim e-Resolved Absor p tion Sp ectr oscop y. Arrays

15 and 16 were examined by time-resolved absorption

spectroscopy in order to determine the rate and yield of

perylene-to-porphyrin energy transfer and to further

assess the contribution of any non-energy-transfer pro-

cesses to the observed photodynamics. Perylene-porphy-

rin arrays 15 and 16 in toluene or benzonitrile were

excited with a 130-fs flash at 485 nm, which is primarily

absorbed by perylene. The excitation-light intensity was

attenuated to avoid pumping more than one of the four

perylenes in 15 or eight perylenes in 16. Representative

spectral and kinetic data for 15 in toluene are shown in

Figure 5, and some of the data are reproduced in Figure

6B.

that for

a given zinc porphyrin reference mono-

mer^8,36-39(Table 3). Similar analysis for each array

(except 4) in benzonitrile also results in a fluorescence

yield (Φ_f) 0.045-0.070) generally comparable to those

for the reference porphyrins in this solvent or others that

can coordinate to the central zinc ion. The slightly

increased fluorescence yields and (less) slightly reduced

excited-singlet-state lifetimes of the zinc porphyrins in

metal-coordinating solvents can be traced to an increased

radiative rate constant k_f. The increased k_fis also

manifested in the intensity of the origin transition in the

optical absorption and emission spectra. Increased k_f

values are also often observed, depending on the sub-

stituents (ethyne, mesityl, halogen, etc.) on the aryl rings

of the porphyrin. In general, these effects derive from

modulation of the porphyrin frontier molecular orbitals

and, thus, the wavefunction of the lowest excited singlet

state.^8,10,38,39In addition, steric-electronic effects of the

multiple perylenes present in 15 and 16 may slightly

alter the electronic interactions involving the ethyne-

containing aryl groups and the porphyrin macrocycle,

causing a small increase in the radiative rate constant

k_fand, thus, the fluorescence yield. However, such effects

represent a relatively minor perturbation on the proper-

ties of the excited porphyrin in these architectures.

Analysis of the fluorescence yields using standard

methods^8,38,39suggests that there is virtually no quench-

ing or, at worst, only modest quenching of the excited

porphyrin due to the presence of multiple perylene

accessory pigments in the arrays. The only clear excep-

tion based on the emission yields is 4 in benzonitrile,

where the significantly reduced fluorescence likely stems

from competing electron transfer (forming PMI^-Zn⁺;

Figure 4) derived from the shorter perylene-porphyrin

distance compared to those of the other arrays. Com-

pound 4 is unique in that the perylenes are attached at

the 2,6-positions rather than the 3,5-positions of the

The transient absorption difference spectrum im-

mediately after excitation of 15 in toluene shows es-

sentially exclusively characteristics of the excited perylene

PMI* (0.5 ps spectrum in Figure 5A, dotted). One of the

two main features is a broad trough at ∼530 nm that

contains bleaching of the PMI (0,0) ground-state absorp-

tion band together with PMI* excited-state stimulated

emission.⁴⁰The second main feature is a broad PMI*

excited-state absorption with a maximum at ∼700 nm.

The PMI* spectrum for 15 is similar to those that we

have observed previously for excited perylene-monoimide

dyes in isolation and in dyads with porphyrins, with the

(37) Gouterman, M. In The Porphyrins; Dolphin, D., Ed.; Academic

Press: New York, 1978; Vol. III, pp 1-165.

(38) (a) Hsiao, J .-S.; Krueger, B. P.; Wagner, R. W.; J ohnson, T. E.;

Delaney, J . K.; Mauzerall, D. C.; Fleming, G. R.; Lindsey, J . S.; Bocian,

D. F.; Donohoe, R. J . J . Am. Chem. Soc. 1996, 118, 11181-11193. (b)

Li, F.; Gentemann, S.; Kalsbeck, W. A.; Seth, J .; Lindsey, J . S.; Holten,

D.; Bocian, D. F. J . Mater. Chem. 1997, 7, 1245-1262.

(39) (a)Yang, S. I.; Seth, J .; Strachan, J .-P.; Gentemann, S.; Kim,

D.; Holten, D.; Lindsey, J . S.; Bocian, D. F. J . Porphyrins Phthalocya-

nines 1999, 3, 117-147. (b) Yang, S. I.; Lammi, R. K.; Seth, J .; Riggs,

J . A.; Arai, T.; Kim, D.; Bocian, D. F.; Holten, D.; Lindsey, J . S. J .

Phys. Chem. B 1998, 102, 9426-9436.

(40) Emission from the excited chromophore is stimulated by the

white-light probe pulse, appears as negative ∆A in the absorption

difference spectra, and occurs at approximately the same wavelengths

as the spontaneous emission.

6528 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

F IGURE 5. Time-resolved absorption spectra (A) and kinetic

profiles (B) for perylene-porphyrin array 15 at room temper-

ature in toluene. The fits to the kinetic data in part B are to

the instrument response and dual-exponential decays plus a

constant, giving time constants of 2.7 and 9.5 ps (535 nm, main

figure) and 3.3 and 12 ps (635 nm, inset). The amplitude-

weighted average value from these and other wavelengths is

3.5 ps. The PMI* spectrum was constructed by concatenating

several spectra at successive delay times, owing to the distor-

tions in the raw difference spectra of this rapidly decaying

state due to the dispersion in the arrival times at the sample

of the different wavelengths in the probe pulse.

F IGURE 6. Time-resolved absorption spectra for porphyrin

9 and perylene-porphyrin arrays 15 and 16 in toluene and

benzonitrile. The spectra for 15 in panel B are reproduced from

Figure 5. The solid spectra are at pump-probe delay times of

1 ps (9) or 50 ps (15, 16), and the dashed spectra are at 3.4

ns.

cant difference indicates the involvement of at least one

process open to PMI* in the array that is not operable in

the isolated chromophore. On the basis of our previous

work with perylene-porphyrin arrays,^7-12the most likely

processes are PMI* Zn f PMI Zn* energy transfer and

PMI* Zn f PMI^-Zn⁺hole transfer (Figure 4).

The spectral evolution for 15 in toluene indicates that

the PMI* decay occurs predominantly (>90%) by the

energy-transfer channel. In particular, by ∼50 ps, the

PMI* spectrum has been fully replaced by the spectrum

of the excited zinc porphyrin (Figures 5A and 6B, solid).

The Zn* spectrum shows all the expected features, which

following caveat.^10-12Like the ground-state absorption

and fluorescence spectra described above, the PMI*

excited-state optical characteristics in 15 are affected by

the presence of the three aryloxy groups on each perylene.

In particular, these groups broaden and red-shift all the

features compared to the cases of the arrays containing

perylene-monoimides with fewer or no solubilizing sub-

stituents.

(41) The time-evolution of PMI* in either 15 or 16 in either toluene

or benzonitrile indicates the presence of two components whose time

constants and relative contributions to the spectra depend on detection

wavelength. The major component generally has a lifetime of 2-5 ps,

and the second component has that of 6-16 ps. The second component

is much less prominent for either array in benzonitrile as compared

to toluene. From analysis across the spectral region shown in Figure

6, the amplitude-weighted average time constants are 3.9 ps for 15 in

toluene, 3.5 ps for 16 in toluene, 3.4 ps for 15 in benzonitrile, and 2.7

ps for 16 in benzonitrile. Considering the complexities of the decays,

we used a mean PMI* lifetime of 3.5 ps in each case. The presence of

two components indicates that the observed PMI* dynamics involve

(at least) two excited-state forms (vibrational, conformational, elec-

tronic) along with relevant relaxation processes. These effects are more

prominent with aryloxy-substituted perylenes.¹²

The PMI* spectrum decays rapidly, as is seen from the

representative kinetic data and fits in Figure 5B. The

figure shows decay of the PMI ground-state bleaching at

535 nm (main panel) and PMI* excited-state absorption

at 635 nm (inset). At each wavelength, the data are best

fit with a dual-exponential function.⁴¹Nevertheless, the

kinetic data across the spectral region shown in Figure

5A when taken as a whole are reasonably represented

by an average PMI* lifetime of ∼3.5 ps. This lifetime is

about 1000-fold shorter than the PMI* lifetime of 5 ns

for reference perylene PMI(OR)₃′ (Table 2). This signifi-

J . Org. Chem, Vol. 67, No. 18, 2002 6529

Tomizaki et al.

are also seen in the spectra for the isolated porphyrin

reference compound 9 in toluene (compare Figure 6A and

B, solid). These Zn* features include Q(1,0) bleaching at

∼550 nm, combined Q(0,0) bleaching and stimulated

emission at 595 nm, and Q(0,1) stimulated emission at

650 nm.⁴⁰All of these features are embedded on a

relatively broad excited-state absorption that increases

in strength in the blue-green region.

Unlike the spectrum at 50 ps for 15 in toluene that

was assigned primarily (>90%) to Zn*, the spectrum at

this time delay for 16 in toluene has characteristics of

PMI^-Zn⁺in addition to the dominant contribution of Zn*

(both formed from PMI*; Figure 4). The formation of the

charge-separated state in the latter array is deduced from

several characteristics of the transient difference spectra.

The primary signature is a diminished amplitude of the

transient absorption due to Zn* between 500 and 540 nm

for 16 in toluene relative to array 15 or porphyrin 9 in

this solvent (compare solid spectra in Figure 6A-C). The

diminished blue-green (positive) transient absorption

for 16 derives primarily from the compensating (nega-

tive) bleaching of the modestly strong ground-state

absorption of PMI in this region (Figure 2C) that occurs

when state PMI^-Zn⁺forms (but which is not present

for Zn*).

To estimate the yield of the hole-transfer process, the

reasonable assumption is made that the PMI bleach in

the blue-green region for PMI^-Zn⁺is approximately the

same as that for PMI*. This analysis takes into account

the estimated contribution of PMI*-stimulated emission

in the 500-540-nm region for state PMI* and ignores

possible small differences in the excited-state absorption

of this state and PMI^-Zn⁺in this region.^10-12Using these

assumptions along with the 0.5-ps PMI* spectrum for 15

in toluene in Figure 5A (dotted) and the companion 50-

ps Zn* spectrum (Figure 5A and 6B, solid) as guides, it

is estimated that the 50-ps spectrum for 16 in toluene

(Figure 6C, solid) contains ∼10% PMI^-Zn⁺and ∼90%

Zn*. Such a small amount of the charge-separated state

is discernible because of the “pulling down” effect on the

Zn* transient absorption between 500 and 540 nm caused

by even a small amount of perylene bleaching when one

considers the much larger inherent amplitude of the

latter feature compared to the former (e.g., Figure 5A

dotted and solid). In addition to the altered spectral

characteristics in the 500-540-nm region, another indi-

cation that PMI^-Zn⁺contributes to the 50-ps spectrum

for 16 in toluene is increased transient absorption in the

vicinity of 630 nm compared to that of array 15 or

porphyrin 9 in this solvent (Figure 6A-C, solid). We have

previously found that the anions of the appropriate

perylene-monoimide dyes have prominent transient ab-

sorption bands centered in the 620-650-nm region.^10-12

The small amount of PMI^-Zn⁺formed from PMI* for

16 in toluene appears to largely decay by the 3.4-ns limit

of the transient absorption experiment, given that the

spectra at this time are very similar to those for array

15 and porphyrin 9 in this solvent (Figure 5), all of which

have been assigned to Zn^T. This observation is consistent

with the finding of similar porphyrin fluorescence life-

times for these three compounds in toluene (Table 3),

which indicates that the decay characteristics of Zn* for

16 in toluene are comparable (within ∼10%) to those of

the isolated chromophore.

Over the next several nanoseconds (and to the ∼3.5-

ns limit of the transient absorption spectrometer), the

Zn* stimulated emission decays away (e.g., the feature

at 650 nm), with comparatively small changes in the

ground-state bleaching when referenced to the broad

excited-state absorption. This spectral evolution seen for

array 15 in toluene (Figures 5A and 6B, dashed) is also

observed for reference porphyrin 9 in the same solvent

(Figure 6A, dashed). These spectral changes are expected

for formation of the porphyrin excited triplet state (Zn^T)

via intersystem crossing in high yield⁴²(>80%). This

triplet-state formation occurs with the relatively long (∼2

ns) Zn* lifetime measured for array 15 and porphyrin 9

via fluorescence decay. Consistent with the emission

yields and lifetimes, the transient absorption spectra for

15 in toluene reveal no significant (e10%) formation of

charge-separated states such as PMI^-Zn⁺from either

PMI* (by hole transfer) or Zn* (by electron transfer). The

production of a charge-separated state would have given

PMI ground-state bleaching (and PMI^-absorption) either

in parallel with Zn* formation or during Zn* decay, which

would have distorted the porphyrin excited-state spectra

in array 15 from those expected and observed for the

isolated reference porphyrin.^8-12

The time-resolved absorption spectra described above

for array 15 in toluene (Figures 5 and 6B) and porphyrin

monomer 9 in toluene (Figure 6A) provide a framework

for assessing the extent to which hole/electron-transfer

reactions supplement the primary energy-transfer pro-

cesses for 16 in toluene and for 15 and 16 in benzonitrile.

To this end, Figure 6C-F shows key transient absorption

spectra for the latter three cases along with data for 9

in benzonitrile. All the data in Figure 6 were acquired

and are presented so that the spectral shapes and

amplitudes can be compared among the different com-

pounds/solvents. Furthermore, the amplitude and general

shape of the subpicosecond PMI* spectrum for 15 in

toluene (Figure 5A, dotted) are relevant not only for the

data at longer times reproduced for that compound in

Figure 6B but also for the spectra at longer times for 16

in toluene and 15 and 16 in benzonitrile presented in

Figure 6 (50 ps, solid; 3.4 ns, dashed). In analogy to the

case of 15 in toluene, the PMI* decay profiles for 16 in

toluene and for 15 and 16 in benzonitrile are complex

(more so in toluene) but are reasonably described by an

average time constant of 3.5 ps.⁴¹The most significant

differences that occur for arrays/solvents are seen in the

spectra (and states) formed from PMI* decay, and

subsequently from Zn* decay, as described below.

Similar analyses can be performed for 15, 16, and

reference porphyrin 9 in benzonitrile (Figure 6D-F). In

short, the transient difference spectra for 9 and 15 in

benzonitrile (Figure 6D and E) are similar to one another,

just as was found in comparing the spectra for these two

compounds in toluene (Figure 6A and B). The only

difference is in the positions and relative amplitudes of

the bleaching and stimulated-emission features in ben-

(42) (a) Gradyushko, A. T.; Sevchenko, A. N.; Solovyov, K. N.;

Tsvirko, M. P. Photochem. Photobiol. 1970, 11, 387-400. (b) Hurley,

J . K.; Sinai, N.; Linschitz, H. Photochem. Photobiol. 1983, 38, 9-14.

(c) Kajii, Y.; Obi, K.; Tanaka, I.; Tobita, S. Chem. Phys. Lett. 1984,

111, 347-349. (d) Kikuchi, K.; Kurabayashi, Y.; Kokubun, H.; Kaizu,

Y.; Kobayashi, H. J . Photochem. Photobiol., A 1988, 45, 261-263. (e)

Bonnett, R. D.; McGarvey, D.; Harriman, A.; Land, E. J .; Truscott, T.

G.; Winfield, U. J . Photochem. Photobiol. 1988, 48, 271-276.

6530 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

zonitrile versus toluene (as expected from the comparable

changes in the static optical spectra due to coordination

of the benzonitrile solvent molecules to the central zinc

ion of the porphyrin). Thus, 15 in benzonitrile has a rapid

PMI* decay (∼3.5 ps)⁴¹that occurs primarily (g90%) by

energy transfer to the porphyrin, and subsequently, Zn*

decays with characteristics comparable (within ∼10%) to

those for the isolated chromophore.

shorter than τ^m) 5 ns in the isolated perylene PMI(OR)₃′

(Table 2). These data give a rate constant for the

quenching of PMI* by the porphyrin of k_Q) (1/τ - 1/τ^m)

∼ (3.5 ps)^-1. On the basis of the yields estimated above,

the rate constant for PMI* Zn f PMI Zn* energy transfer

is k_ENT) 0.95 × (3.5 ps)^-1∼ (3.7 ps)^-1for 15 in toluene.

The corresponding microscopic rate constant for hole

transfer is k_HT) 0.05 × (3.5 ps)^-1∼ (70 ps)^-1

.

The spectrum at 50 ps for array 16 in benzonitrile

(Figure 6F, solid) suggests that PMI* decays in part by

hole transfer to form PMI^-Zn⁺. In benzonitrile, the Zn*

absorption between 500 and 540 nm is “pulled down”

even further by the opposing perylene bleaching than it

is in toluene, indicating that the amount of the charge-

separated state has increased in yield from ∼10% in

toluene to ∼20% in benzonitrile. At 3.4 ns (Figure 6F,

dashed), the Zn^Tspectrum in benzonitrile is compromised

by apparent formation of some PMI^-Zn⁺via electron

transfer from Zn* back to the perylene. On the basis of

the fluorescence lifetime data presented above (Table 3),

we estimated this Zn* electron-transfer quenching to

have a yield of ∼20%, which is consistent with the

transient absorption spectra.

Once the excited porphyrin is formed (by energy

transfer from perylene or direct excitation), Zn* decays

with characteristics (lifetime, emission yield) comparable

(within ∼10%) to those operable in the isolated porphyrin

(Figure 4). Within the series of analogous arrays 7, 14,

15, and 16, porphyrin quenching rises above the ∼10%

level only for 16 in benzonitrile, for which Zn* decays

∼20% via electron transfer back to PMI with an overall

rate constant of 0.2 × (2.1 ns)^-1∼ (10 ns)^-1. Since the

porphyrin can transfer an electron to any one of the eight

perylenes in 16, the microscopic rate constant for each

process is (10 ns)^-1/8 ) (80 ns)^-1. Thus, electron transfer

is significantly slower than the normal routes (intersys-

tem crossing, internal conversion, fluorescence) operable

in the isolated porphyrin. These results suggest that the

PMI^-Zn⁺charge-separated state lies above Zn* (but

below PMI*) in toluene and moves close to or possibly

slightly below Zn* in benzonitrile because of solvent

stabilization.

Thus, the modest increase in the PMI Zn* f PMI^-Zn⁺

electron-transfer yield to ∼20% for 16 in benzonitrile as

compared to ∼10% in toluene can be understood by a

solvent-polarity effect on the rate, as we have found for

other perylene-porphyrin arrays.^9,11,12In turn, a small

increase in the overall Zn* electron-transfer yield for 16

versus 15 derives from the presence of twice as many

perylenes in the former array. The observations are

consistent with the microscopic energy-transfer rate

constant being the same for both arrays (and for the

arrays containing fewer perylenes) in a given solvent.

However, as noted earlier for the decay of PMI*, the

perylene-porphyrin interactions may be perturbed slightly

with an increasing number of perylenes because of

structural-electronic effects, although these effects ap-

pear to be relatively small.

In summary, the extensive blue-green absorption of the

eight perylenes appended to the porphyrin in 16 affords

a suitable light-harvesting system that has relatively

minor competing charge-transfer quenching of the excited

perylene in nonpolar solvents and modest competing

quenching of both the perylene and porphyrin in highly

polar media. Charge-transfer quenching is even less for

15, for which the four perylenes still give substantial

absorption complementary to that of the porphyrin to

make an excellent light-harvesting motif in both polar

and nonpolar media.

(7) Effective Accessor y P igm en ts. The perylene-

monoimide dyes examined herein appear to be ideal for

use as accessory pigments with porphyrins. The perylene

dyes provide the following attributes: (1) absorption in

the trough between the porphyrin B and Q bands; (2)

rapid and efficient energy transfer to the porphyrin; (3)

limited or no competing electron-transfer quenching

processes with the photoexcited porphyrin in polar or

nonpolar solvents; (4) long, monophasic excited-state

lifetime, high fluorescence yield, and very low yield of

Su m m a r y of Excited -Sta te Dyn a m ics. The photo-

physical data for arrays containing two (7 and 14), four

(15), and eight (16) perylenes attached at the meta-

positions of the porphyrin indicate that these are excel-

lent light-harvesting motifs, with the optimum being four

perylenes. The collective data are internally consistent

and indicate that the yield of PMI* Zn f PMI Zn* energy

transfer is >90% for arrays with two or four perylenes

in toluene or benzonitrile, ∼90% for 16 in toluene, and

∼80% for 16 in benzonitrile. The remaining fraction of

the PMI* decay in each case primarily involves PMI* Zn

f PMI^-Zn⁺hole transfer (Figure 4), which has a yield

of at most ∼20% (for 16 in benzonitrile). Regarding these

results, we should note that, in principle, the same yield

for perylene-to-porphyrin energy transfer should be found

for all the analogous arrays (7, 14, 15, 16) in a given

solvent, since only one perylene in each array should be

excited at any time and each perylene should behave

independently in energy (or hole) transfer to the central

porphyrin. The energy-transfer yield will increase slightly

for a given array in benzonitrile as compared to toluene

because of stabilization of the PMI^-Zn⁺state in the polar

solvent. These expectations are generally borne out by

the results. The only apparent deviation is a small

increase in the hole-transfer yield for 16 as compared to

15 (especially in benzonitrile). This finding may derive

from small (structural) effects of the eight perylenes in

16 on the electronic interactions with the porphyrin, as

noted earlier regarding the static optical spectra, or from

the assumptions made in analyzing the time-resolved

optical spectra that may lead to overestimates of the hole-

transfer yields. Nevertheless, deviations from the expec-

tation of similar perylene-to-porphyrin energy-transfer

yields among the arrays are small. Furthermore, the

collective results indicate that energy transfer to the

porphyrin dominates the decay of PMI* in all the arrays.

For each dyad, the rate constants for PMI* energy and

hole transfers (Figure 4) can be estimated from the time-

resolved data using standard analysis.^8,38,39For example,

the average PMI* lifetime of τ ∼ 3.5 ps⁴¹for 15 in toluene

(like the other arrays and solvents) is considerably

J . Org. Chem, Vol. 67, No. 18, 2002 6531

Tomizaki et al.

intersystem crossing; (5) compatibility with a building

block synthesis; (6) amenability to architectures with

multiple accessory pigments; and (7) high solubility and

nonpolar nature, affording facile purification. Very few

classes of dyes meet this set of criteria.⁴³For example,

xanthenes or cyanine dyes provide superior laser dyes

and/or biolabels but are difficult to purify given their

intrinsic charge. Carotenoids absorb strongly but have

an exquisitely short excited-state lifetime requiring very

close proximity to the porphyrin for efficient energy

transfer. Boron-dipyrrin dyes have two excited-state

conformers with short lifetimes, cause quenching of the

photoexcited porphyrin in polar solvents, and, in addition,

present purification challenges upon incorporation in

multipigment arrays because of their intrinsic polarity.

The perylene-monoimides provide about 2.5-times the

integrated absorption (ꢀ_max× fwhm) of that of boron-

dipyrrin dyes, which we have used previously for light-

harvesting applications. All factors considered, the

perylene and linker motifs examined herein are very well

suited for use with porphyrins in light-harvesting archi-

tectures.

used in a building block approach to construct light-

harvesting rods consisting of a central chain of porphy-

rins decorated by closely packed perylene accessory

pigments. The preparation and study of such architec-

tures based on the results obtained in the present paper

are now underway.

Exp er im en ta l Section

Gen er a l. ¹H (300 or 400 MHz) and ¹³C (75 MHz) NMR

spectra were recorded in CDCl₃unless noted otherwise. Mass

spectra of porphyrins and perylene-porphyrin arrays were

obtained by high-resolution fast atom bombardment (FABMS)

and/or by laser desorption mass spectrometry (LDMS)⁴⁴or

matrix-assisted LDMS (MALDI-MS) using the matrix dithra-

nol. Absorption and emission spectra were collected in toluene

unless noted otherwise. Elemental analyses were performed

by Atlantic Microlab, Inc. Melting points are uncorrected.

Silica gel (40 µm average particle size) and alumina (80-200

mesh) were used for column chromatography. Preparative SEC

was performed using BioRad Bio-Beads SX-1 (200-400 mesh)

beads. Analytical SEC was performed using an HPLC (column

size ) 1000 Å; flow rate ) 0.800 mL/min; solvent ) THF;

quantitation at 420 and 510 nm; reference at 670 nm; oven

temperature at 25 °C).⁴⁵All Pd-mediated reactions were

performed using a Schlenk line. The conditions for Sonogashira

reactions with porphyrins use tris(dibenzylideneacetone)-

dipalladium(0) (Pd₂(dba)₃) and P(o-tol)₃in the absence of any

copper reagents.²⁶Palladium insertion and transmetalation

have not been observed with these conditions. Toluene and

triethylamine were freshly distilled from CaH₂and sparged

of oxygen prior to use. Chloroform contained 0.8% ethanol as

a stabilizer. PC Model for Windows 7.50.00 (Serena Software,

Inc.) was used to generate the space-filling model in Figure 1

and to calculate intramolecular distances.

Con clu sion

Several new perylene-porphyrin light-harvesting ar-

rays have been synthesized and characterized in order

to identify suitable multiperylene-porphyrin motifs for

use in larger light-harvesting arrays. The preparation of

the arrays followed two synthetic strategies: (1) an

ethynylporphyrin building block was coupled with a

bromoperylene (affording arrays 4 and 7), and (2) a

perylene-aldehyde was employed in a porphyrin-forming

reaction (affording arrays 14-16). All arrays examined

exhibit virtually quantitative (g90%) excited-state energy

transfer from perylene to porphyrin except for the array

with eight perylenes in polar media, which gave a 20%

loss due to competing charge-transfer quenching. The

arrays with tris(aryloxy)perylene-monoimide dyes posi-

tioned at the 3,5-positions of one, two, or four aryl rings

of the porphyrin have high solubility and unperturbed

porphyrin fluorescence lifetimes and yields in nonpolar

(toluene) solvents and, in most cases, in polar (benzoni-

trile) solvents as well. Only with eight perylenes per

porphyrin in benzonitrile is there a clear (>10%) quench-

ing of the excited porphyrin, and again at a modest level

of ∼20%. By comparison, the arrays with perylene-

monoimide dyes positioned at the 2,6-positions provide

a more compact geometry but are synthesized in lower

yields and also are less suited for light-harvesting ap-

plications in polar media.

Non com m er cia l Com p ou n d s. Aldehyde 5;⁵5-mesityl-

dipyrromethane (8);²⁸and perylenes PMI-1³¹and PMI-2³²were

prepared as described in the literature. The syntheses of

perylenes 3, 10, PMI(OR), and 11 will be described elsewhere.¹²

Op tica l Sp ectr oscop y. Absorption spectra (HP 8453, Cary

3) and fluorescence spectra (Spex FluoroMax) were collected

as described previously using nondegassed solutions (typically

<1 × 10^-5M).³⁹Fluorescence quantum yields were determined

by ratioing the integrated corrected emission spectrum of the

sample to that of ZnTPP (0.033)³⁷or N,N′-bis(2,5-di-tert-

butylphenyl)-3,4:9,10-perylenedicarboximide (0.97).¹⁵Tran-

sient absorption data were obtained at room temperature as

described elsewhere.⁴⁶Samples (∼0.1 to 0.2 mM in toluene or

benzonitrile) in 2-mm path length cuvettes were excited at 10

Hz with 130-fs pump pulses (20-35 µJ at 475-490 nm) and

probed with white light pulses of comparable duration.

2,6-Bis[2-(tr im eth ylsilyl)eth yn yl]ben zaldeh yde (1). Fol-

lowing the method described by Buchwald and Fu,²²samples

of 2,6-dichlorobenzaldehyde (3.56 g, 20.4 mmol), Pd(PhCN)₂Cl₂

(468 mg, 1.22 mmol), and CuI (155 mg, 814 µmol) were placed

in a Schlenk flask. Diisopropylamine (6.7 mL), dioxane (30

mL), and P(t-Bu)₃(494 mg, 2.44 mmol) were placed in a second

Schlenk flask and sparged of oxygen for 10 min, and then the

mixture was added to the first Schlenk flask. The flask was

Overall, the array containing four perylenes attached

at the 3,5-positions of two trans-aryl rings of the por-

phyrin provides the optimal characteristics in both polar

and nonpolar media in terms of light harvesting with

little if any (e10%) deleterious charge-transfer quenching

of either the excited perylene or porphyrin. This motif

leaves the other two aryl rings of the porphyrin free for

attaching synthetic handles. These functionalities can be

(44) (a) Fenyo, D.; Chait, B. T.; J ohnson, T. E.; Lindsey, J . S. J .

Porphyrins Phthalocyanines 1997, 1, 93-99. (b) Srinivasan, N.; Haney,

C. A.; Lindsey, J . S.; Zhang, W.; Chait, B. T. J . Porphyrins Phthalo-

cyanines 1999, 3, 283-291.

(45) del Rosario Benites, M.; J ohnson, T. E.; Weghorn, S.; Yu, L.;

Rao, P. D.; Diers, J . R.; Yang, S. I.; Kirmaier, C.; Bocian, D. F.; Holten,

D.; Lindsey, J . S. J . Mater. Chem. 2002, 12, 65-80.

(46) (a) Yang, S. I.; Li, J .; Cho, H. S.; Kim, D.; Bocian, D. F.; Holten,

D.; Lindsey, J . S. J . Mater. Chem. 2000, 10, 283-296. (b) Kirmaier,

C.; Holten, D. Biochemistry 1991, 30, 609-613.

(47) Bo¨hm, A.; Helfer, W. U.S. Patent 5,808,073.

(43) (a) Wagner, R. W.; Lindsey, J . S. Pure Appl. Chem. 1996, 68,

1373-1380. (b) Wagner, R. W.; Lindsey, J . S. Pure Appl. Chem. 1998,

70 (8), p. i.

(48) Hofkens, J .; Latterini, L.; De Belder, G.; Gensch, T.; Maus, M.;

Vosch, T.; Karni, Y.; Schweitzer, G.; De Schryver, F. C.; Hermann, A.;

Mu¨llen, K. Chem. Phys. Lett. 1999, 304, 1-9.

6532 J . Org. Chem., Vol. 67, No. 18, 2002

Synthesis and Properties of Perylene-Porphyrin Arrays

placed in an oil bath heated to 70 °C. After 5 min, (trimeth-

ylsilyl)acetylene (7.19 mL, 50.9 mmol) was added and the

mixture was stirred at 70 °C for 4 h. Removal of an aliquot

and analysis by GC showed mono- and dicoupled products in

a nearly 1:1 ratio. An additional equivalent of (trimethylsilyl)-

acetylene (2.9 mL, 20.4 mmol) was added. After 24 h at 70 °C,

GC analysis showed no monocoupled product. The mixture was

cooled to room temperature and then filtered through a pad

of silica. Kugelrohr distillation (110 °C, 0.1 Torr) followed by

column chromatography (silica, toluene) afforded a yellow oil

(2.22 g, 37%): ¹H NMR δ 0.27 (s, 18H), 7.40-7.43 (m, 1H),

7.51 (d, J ) 8.0 Hz, 2H), 10.63 (s, 1H); ¹³C NMR δ -0.0, 101.5,

102.5, 125.2, 132.3, 134.4, 137.2, 190.6. FABMS obsd, 299.1267;

calcd, 299.1287. Anal. Calcd for C₁₇H₂₂OSi₂: C, 68.40; H, 7.43.

Found: C, 67.94; H, 7.36.

Zin c(II) 5-[2,6-Bis[2-(tr im eth ylsilyl)eth yn yl]p h en yl]-

10,15,20-tr im esitylp or p h yr in (2). Following a standard

procedure for mixed-aldehyde condensations¹⁸at high concen-

tration²⁰with BF₃‚O(Et)₂-ethanol cocatalysis,¹⁹samples of 1

(299 mg, 1.00 mmol), mesitaldehyde (445 mg, 3.00 mmol), and

pyrrole (280 µL, 4.00 mmol) were condensed in CHCl₃(56 mL)

in the presence of BF₃‚O(Et)₂(129 µL, 1.02 mmol) at room

temperature for 1.5 h. Then, DDQ (680 mg, 3.00 mmol) was

added. After 1 h, TEA was added and the crude mixture was

passed through a silica column [CHCl₃/hexanes (1:1)], affording

a mixture of porphyrins. The resulting porphyrin mixture was

treated with Zn(OAc)₂‚2H₂O (550 mg, 2.50 mmol) in DMF (20

mL) overnight at 100 °C. Water was added to the reaction

mixture, and the resulting solid was filtered. Column chro-

matography [silica, CH₂Cl₂/hexanes (1:1)] afforded a purple

solid (10.7 mg, 1.1%): ¹H NMR δ -1.28 (s, 18H), 1.84-1.88

(m, 18H), 2.62 (m, 9H), 7.26 (m, 6H), 7.62-7.68 (m, 1H), 7.81-

7.84 (m, 2H), 8.67-8.74 (m, 8H); LDMS obsd, 996.33 [M⁺].

FABMS obsd, 994.3788; calcd, 994.3805 (C₆₃H₆₂N₄Si₂Zn). λ_abs

) 421, 548 nm; λ_em(λ_ex) 550 nm) 593, 642 nm.

Zin c(II) 5-(2,6-Dieth yn ylph en yl)-10,15,20-tr im esitylpor -

p h yr in (2′). A sample of 2 (10.0 mg, 10.0 µmol) in CHCl₃/THF

[5.0 mL, (1:1)] was treated with TBAF on silica gel (22.0 mg,

1.0-1.5 mmol/g) for 5 h at room temperature. The reaction

mixture was washed with 10% NaHCO₃and water and dried

over Na₂SO₄. Column chromatography [silica, CHCl₃/hexanes

(1:1)] afforded a purple solid (7.0 mg, 82%): ¹H NMR δ 1.84

(m, 18H), 2.08 (s, 2H), 2.62 (m, 9H), 7.25-7.26 (m, 6H), 7.69-

7.74 (m, 1H), 7.90-7.93 (m, 2H), 8.67-8.74 (m, 8H); MALDI-

MS obsd, 852.71 [M⁺]. FABMS obsd, 850.3042; calcd, 850.3014

(C₅₇H₄₆N₄Zn). λ_abs) 423, 550 nm; λ_em(λ_ex) 550 nm) 595, 644

nm.

Zin c(II) 5-[2,6-Bis[2-[4-[9-(4-ter t-bu tylph en oxy)per ylen e-

3,4-d ic a r b oxim id o]-3,5-d iisop r op ylp h e n yl]e t h yn yl]-

p h en yl]-10,15,20-tr im esitylp or p h yr in (4). A mixture of 2′

(7.0 mg, 8.2 µmol), 3 (14.5 mg, 20.5 µmol), Pd₂(dba)₃(1.9 mg,

2.1 µmol), and P(o-tol)₃(3.7 mg, 12 µmol) in toluene/triethyl-

amine [2.0 mL (10:1)] was stirred at 60 °C under argon. After

2.5 h, another identical batch of catalyst was added. After 18

h, the mixture was cooled and passed through a silica column

(CHCl₃). Preparative SEC (toluene) followed by column chro-

matography [silica, CHCl₃/hexanes (4:1)] and trituration with

hexanes afforded a red solid (4.6 mg, 27%): ¹H NMR δ -0.16

(d, J ) 6.6 Hz, 24H), 1.35 (s, 18H), 1.79-1.88 (m, 22H), 2.56,

2.60 (m, 9H), 6.61 (s, 4H), 6.88 (d, J ) 8.1 Hz, 2H), 7.06 (d, J

) 8.8 Hz, 4H), 7.18 (brs, 2H), 7.23 (brs, 4H), 7.42 (d, J ) 8.8

Hz, 4H), 7.58-7.63 (m, 2H), 7.80-7.85 (m, 1H), 7.99 (d, J )

8.1 Hz, 2H), 8.09 (d, J ) 8.1 Hz, 2H), 8.19-8.30 (m, 8H), 8.37

(d, J ) 8.1 Hz, 2H), 8.41 (d, J ) 8.1 Hz, 2H), 8.60 (d, J ) 4.4

Hz, 2H), 8.63 (d, J ) 4.4 Hz, 2H), 8.78 (d, J ) 5.1 Hz, 2H),

8.87 (d, J ) 4.4 Hz, 2H). LDMS obsd, 2110.42 [M⁺]; calcd av

mass, 2107.93 (C₁₄₅H₁₂₀N₆O₆Zn). λ_abs) 423, 509, 541 nm; λ_em

(λ_ex) 510 nm) 568, 600, 642 nm.

mmol), mesitaldehyde (289 mg, 1.95 mmol), and pyrrole (182

µL, 2.60 mmol) were condensed in CHCl₃(36 mL) in the

presence of BF₃‚O(Et)₂(80 µL, 0.633 mmol) at room temper-

ature for 1.5 h. Then, DDQ (441 mg, 1.95 mmol) was added.

After 1 h, TEA was added and the crude mixture was passed

through a silica column [CHCl₃/hexanes (1:1)], affording a

mixture of porphyrins. The resulting porphyrin mixture was

treated with Zn(OAc)₂‚2H₂O (712 mg, 3.25 mmol) in CHCl₃

(40 mL) and methanol (10 mL) overnight at room temperature.

The organic phase was washed with water and dried over Na₂-

SO₄. Two column chromatography procedures [silica, CH₂Cl₂/

hexanes (1:1) and (2:3)] afforded a purple solid (78.6 mg,

14%): ¹H NMR δ 1.84 (s, 12H), 1.85 (s, 6H), 2.64 (m, 9H), 3.16

(s, 2H), 7.28 (m, 6H), 8.02 (m, 1H), 8.32-8.33 (m, 2H), 8.70-

8.73 (m, 4H), 8.77 (d, J ) 5.1 Hz, 2H), 8.80 (d, J ) 5.1 Hz,

2H); LDMS obsd, 852.49 [M⁺]. FABMS obsd, 850.3041; calcd,

850.3014 (C₅₇H₄₆N₄Zn). λ_abs) 423, 549 nm; λ_em(λ_ex) 550 nm)

593, 644 nm.

Zin c(II) 5-[3,5-Bis[2-[4-[9-(4-ter t-bu tylph en oxy)per ylen e-

3,4-d ic a r b oxim id o]-3,5-d iisop r op ylp h e n yl]e t h yn yl]-

p h en yl]-10,15,20-tr im esitylp or p h yr in (7). A mixture of 6

(40.0 mg, 46.9 µmol), 3 (73.0 mg, 103 µmol), Pd₂(dba)₃(9.4 mg,

10 µmol), and P(o-tol)₃(18.8 mg, 61.8 µmol) in toluene/

triethylamine [5.0 mL (10:1)] was stirred at 60 °C under argon.

After 3.5 h, another identical batch of catalyst was added. After

15.5 h, the mixture was cooled and passed through a silica

column (CHCl₃). Preparative SEC (THF) and two column

chromatography procedures [silica, CHCl₃/hexanes (9:1) and

CHCl₃] followed by trituration with hexanes afforded a red

solid (12.8 mg, 13%): ¹H NMR δ 1.18 (d, J ) 6.6 Hz, 24H),

1.37 (s, 18H), 1.87 (m, 18H), 2.63-2.64 (m, 9H), 2.71-2.80 (m,

4H), 6.97 (d, J ) 8.8 Hz, 2H), 7.12 (d, J ) 8.8 Hz, 4H), 7.28

(m, 2H), 7.29 (m, 4H), 7.47 (d, J ) 8.8 Hz, 4H), 7.53-7.58 (m,

4H), 7.67-7.72 (m, 2H), 8.24 (m, 1H), 8.30 (d, J ) 8.1 Hz, 2H),

8.43 (d, J ) 8.8 Hz, 2H), 8.42 (m, 2H), 8.456 (d, J ) 8.1 Hz,

2H), 8.464 (d, J ) 8.8 Hz, 2H), 8.55 (d, J ) 7.3 Hz, 2H), 8.61

(d, J ) 8.1 Hz, 2H), 8.65 (d, J ) 8.1 Hz, 2H), 8.70-8.72 (m,

4H), 8.82 (d, J ) 4.4 Hz, 2H), 8.95 (d, J ) 5.1 Hz, 2H). MALDI-

MS obsd, 2108.87 [M⁺]; calcd av mass, 2107.93 (C₁₄₅H₁₂₀N₆O₆-

Zn). λ_abs) 424, 511, 544 nm; λ_em(λ_ex) 510 nm) 593, 644 nm.

Zin c(II) 5,15-Bis(3,5-d iet h yn ylp h en yl)-10,20-d im esit -

ylp or p h yr in (9). Samples of 5 (80.0 mg, 0.519 mmol) and 8

(137 mg, 0.519 mmol) were condensed in CH₂Cl₂(52 mL) in

the presence of TFA (71.0 µL, 0.928 mmol) at room tempera-

ture for 30 min. Then, DDQ (118 mg, 0.519 mmol) was added.

After 1 h, TEA was added and the crude mixture was passed

through a silica column (CH₂Cl₂), affording a mixture of

porphyrins. The resulting porphyrin mixture was treated with

Zn(OAc)₂‚2H₂O (570 mg, 2.60 mmol) in CHCl₃(20 mL) and

methanol (5.0 mL) overnight at room temperature. The organic

phase was washed with water and dried over Na₂SO₄. Column

chromatography [silica, CH₂Cl₂/hexanes (1:1)] afforded a purple

solid (33.2 mg, 15%): ¹H NMR δ 1.82 (s, 12H), 2.64 (s, 6H),

3.16 (s, 4H), 7.29 (m, 4H), 8.03 (m, 2H), 8.33 (m, 4H), 8.80 (d,

J ) 4.4 Hz, 4H), 8.83 (d, J ) 4.4 Hz, 4H); LDMS obsd, 858.86

[M⁺]. FABMS obsd, 856.2563; calcd, 856.2544 (C₅₈H₄₀N₄Zn).

λ_abs) 424, 549 nm; λ_em(λ_ex) 550 nm) 596, 645 nm.

3,5-Bis[2-[4-[1,6,9-t r is(4-ter t-b u t ylp h en oxy)p er ylen e-

3,4-d ica r boxim id o]-3,5-d iisop r op ylp h en yl]eth yn yl]ben z-

a ld eh yd e, Meth od A (13). A mixture of 10 (281 mg, 280

µmol), 5 (22.0 mg, 140 µmol), Pd₂(dba)₃(43.0 mg, 46.8 µmol),

and P(o-tol)₃(86.0 mg, 281 µmol) in toluene/triethylamine [28

mL (5:1)] was stirred at 60 °C under argon. After 3 h, another

identical batch of catalyst was added. After 2 h, the mixture

was cooled and passed through a silica column (CHCl₃).

Preparative SEC (THF) and column chromatography (silica,

CHCl₃) afforded a magenta solid (25.0 mg, 8.9%): mp > 230

°C; ¹H NMR δ 1.17 (d, J ) 6.6 Hz, 24H), 1.32 (s, 18H), 1.34 (s,

18H), 1.35 (s, 18H), 2.68-2.75 (m, 4H), 6.90 (d, J ) 8.8 Hz,

2H), 7.02 (d, J ) 8.8 Hz, 4H), 7.08 (d, J ) 8.8 Hz, 4H), 7.09 (d,

J ) 8.1 Hz, 4H), 7.37 (d, J ) 8.8 Hz, 4H), 7.41 (d, J ) 8.8 Hz,

4H), 7.42 (d, J ) 8.8 Hz, 4H), 7.48 (m, 4H), 7.62-7.67 (m, 2H),

Zin c(II) 5-(3,5-Dieth yn ylph en yl)-10,15,20-tr im esitylpor -

p h yr in (6). Following a standard procedure for mixed-

aldehyde condensations¹⁸at high concentration²⁰with BF₃‚

O(Et)₂-ethanol cocatalysis,¹⁹samples of 5 (100 mg, 0.649

J . Org. Chem, Vol. 67, No. 18, 2002 6533

Tomizaki et al.

8.00 (m, 3H), 8.31 (s, 2H), 8.34 (s, 2H), 8.49 (d, J ) 8.8 Hz,

2H), 9.25 (d, J ) 8.8 Hz, 2H), 9.45 (d, J ) 8.1 Hz, 2H), 10.03

(s, 1H); ¹³C NMR δ 24.1, 29.3, 29.9, 31.7, 34.6, 87.3, 92.1, 111.8,

118.4, 118.5, 120.0, 120.4, 121.4, 122.3, 123.6, 123.8, 124.5,

124.9, 125.1, 125.2, 126.0, 126.7, 127.1, 127.3, 127.4, 127.86,

127.93, 128.2, 130.0, 130.2, 130.9, 131.8, 132.06, 132.13, 136.8,

139.9, 146.6, 147.2, 147.3, 147.8, 152.7, 153.5, 153.6, 156.3,

163.4, 191.1. MALDI-MS obsd, 2005.14; calcd av mass, 2002.51

(C₁₃₉H₁₂₈N₂O₁₁). λ_abs) 415, 537 nm; λ_em(λ_ex) 540 nm) 579,

624 (sh) nm.

3,5-Bis[2-[4-[1,6,9-t r is(4-ter t-b u t ylp h en oxy)p er ylen e-

3,4-d ica r boxim id o]-3,5-d iisop r op ylp h en yl]eth yn yl]ben z-

a ld eh yd e, Meth od B (13). A mixture of 11 (412 mg, 434

µmol), 12 (48.0 mg, 181 µmol), Pd₂(dba)₃(17.0 mg, 18.1 µmol),

PPh₃(29.0 mg, 109 µmol), and CuI (8.0 mg, 43 µmol) in toluene/

triethylamine [36 mL (5:1)] was stirred at 50 °C under argon.

After 3 h, another identical batch of catalyst was added. After

20 h, the mixture was cooled and passed through a silica

column [CHCl₃/hexanes (4:1)]. Preparative SEC (THF) and

column chromatography [silica, CHCl₃/hexanes (9:1)] afforded

a magenta solid (240 mg, 66%): Analytical data were identical

to those as described above.

Zin c(II) 5-[3,5-Bis[2-[4-[1,6,9-tr is(4-ter t-bu tylp h en oxy)-

perylene-3,4-dicarboximido]-3,5-diisopropylphenyl]ethynyl]-

p h en yl]-10,15,20-tr im esitylp or p h yr in (14). Following a

standard procedure for mixed-aldehyde condensations¹⁸at high

concentration²⁰with BF₃‚O(Et)₂-ethanol cocatalysis,¹⁹samples

of 13 (100 mg, 49.9 µmol), mesitaldehyde (22.2 mg, 150 µmol),

and pyrrole (14 µL, 200 µmol) were condensed in CHCl₃(2.8

mL) in the presence of BF₃‚O(Et)₂(6.2 µL, 49 µmol) at room

temperature for 1.5 h. Then, DDQ (34 mg, 150 µmol) was

added. After 1 h, TEA was added and the crude mixture was

passed through a silica column [CH₂Cl₂/hexanes (2:1)], afford-

ing a porphyrin mixture. The resulting porphyrin was treated

with Zn(OAc)₂‚2H₂O (55 mg, 250 µmol) in CH₂Cl₂(10 mL) and

methanol (3.0 mL) at room temperature for 12 h. The organic

phase was washed with water and dried over Na₂SO₄. Pre-

parative SEC (THF) and column chromatography [silica, CH₂-

Cl₂/hexanes (1:1)] afforded a red solid (24.3 mg, 18%): ¹H NMR

δ 1.13 (d, J ) 6.6 Hz, 24H), 1.31 (s, 18H), 1.33 (s, 18H), 1.34

(s, 18H), 1.85-1.86 (m, 18H), 2.63-2.71 (m, 13H), 6.89 (d, J

) 8.8 Hz, 2H), 7.01 (d, J ) 8.8 Hz, 4H), 7.06 (d, J ) 8.8 Hz,

4H), 7.09 (d, J ) 8.8 Hz, 4H), 7.28 (m, 6H), 7.35 (d, J ) 8.8

Hz, 4H), 7.40 (d, J ) 8.1 Hz, 4H), 7.43 (d, J ) 8.8 Hz, 4H),

7.50 (m, 4H), 7.61-7.66 (m, 2H), 8.20 (m, 1H), 8.29 (s, 2H),

8.32 (s, 2H), 8.39 (m, 2H), 8.48 (d, J ) 8.1 Hz, 2H), 8.71 (m,

4H), 8.79 (d, J ) 5.1 Hz, 2H), 8.91 (d, J ) 4.4 Hz, 2H), 9.24 (d,

J ) 9.5 Hz, 2H), 9.44 (d, J ) 8.1 Hz, 2H). MALDI-MS obsd,

2694.31 [M⁺]; calcd av mass, 2700.74 (C₁₈₅H₁₆₈N₆O₁₀Zn). λ_abs

) 424, 508, 543 nm; λ_em(λ_ex) 540 nm) 593, 644 nm.

Zin c(II) 5,15-Bis[3,5-b is[2-[4-[1,6,9-t r is(4-ter t-b u t yl-

phenoxy)perylene-3,4-dicarboximido]-3,5-diisopropylphenyl]-

eth yn yl]p h en yl]-10,20-d im esitylp or p h yr in (15). Samples

of 13 (50.0 mg, 25.0 µmol) and 8 (6.6 mg, 25 µmol) were

condensed in CH₂Cl₂(2.5 mL) in the presence of TFA (3.5 µL,

45 µmol) at room temperature for 30 min. Then, DDQ (8.6 mg,

38 µmol) was added. After 1 h, TEA was added and the crude

mixture was passed through a silica column [CH₂Cl₂/hexanes

(3:1)]. The resulting porphyrin was treated with Zn(OAc)₂‚

2H₂O (14 mg, 63 µmol) in CH₂Cl₂(8.0 mL) and methanol (2.0

mL) at room temperature for 12 h. The organic phase was

washed with water and dried over Na₂SO₄. Column chroma-

tography [silica, CHCl₃/hexanes (3:1)] followed by trituration

with methanol afforded a red solid (16.1 mg, 28%): ¹H NMR

δ 1.12 (d, J ) 6.6 Hz, 48H), 1.30 (s, 36H), 1.33 (s, 36H), 1.34

(s, 36H), 1.85 (s, 12H), 2.64-2.71 (m, 14H), 6.89 (d, J ) 8.8

Hz, 4H), 7.00 (d, J ) 8.8 Hz, 8H), 7.06 (d, J ) 6.6 Hz, 8H),

7.09 (d, J ) 6.6 Hz, 8H), 7.29 (m, 4H), 7.35 (d, J ) 8.8 Hz,

8H), 7.39 (d, J ) 6.6 Hz, 8H), 7.42 (d, J ) 6.6 Hz, 8H), 7.49

(m, 8H), 7.61-7.66 (m, 4H), 8.20 (m, 2H), 8.28 (s, 4H), 8.32 (s,

4H), 8.40-8.41 (m, 4H), 8.47 (d, J ) 8.8 Hz, 4H), 8.83 (d, J )

4.4 Hz, 4H), 8.95 (d, J ) 5.1 Hz, 4H), 9.24 (d, J ) 8.8 Hz, 4H),

9.44 (d, J ) 8.1 Hz, 4H); MALDI-MS obsd, 4552.50 [M⁺],

4402.44 [(M - 4-t-Bu-Ph-O)⁺]; calcd av mass, 4555.07

(C₃₁₄H₂₈₄N₈O₂₀Zn). λ_abs) 426, 508, 538 nm; λ_em(λ_ex) 540 nm)

596, 645 nm.

Zin c(II) meso-Tetr a k is[3,5-bis[2-[4-[1,6,9-tr is(4-ter t-bu t-

ylp h en oxy)p er ylen e-3,4-d ica r boxim id o]-3,5-d iisop r op yl-

p h en yl]eth yn yl]p h en yl]p or p h yr in (16). Following a pro-

cedure for cocatalysis with BF₃‚O(Et)₂+ salt,³⁰samples of 13

(48.0 mg, 24.0 µmol) and pyrrole (1.7 µL, 24 µmol) were

condensed in CH₂Cl₂(2.4 mL) in the presence of BF₃‚O(Et)₂

[3.0 µL of a 0.81 mM stock solution of BF₃‚O(Et)₂in CH₂Cl₂,

2.4 µmol] and NaCl (35 mg, 600 µmol) at room temperature

for 1 h. Then, DDQ (4.1 mg, 18 µmol) was added. After 1 h,

TEA was added and the crude mixture was passed through a

silica column [CHCl₃/acetone (49:1)] followed by preparative

SEC (THF). The resulting porphyrin was treated with Zn-

(OAc)₂‚2H₂O (6.6 mg, 30.0 µmol) in CHCl₃(5.0 mL) and

methanol (1.0 mL) at room temperature for 12 h. The organic

phase was washed with water and dried over Na₂SO₄. Column

chromatography [silica, CHCl₃/hexanes (4:1)] followed by

trituration with methanol afforded a red solid (15.4 mg,

31%): ¹H NMR δ 1.11 (d, J ) 6.6 Hz, 96H), 1.27 (s, 72H), 1.29

(s, 72H), 1.36 (s, 72H), 2.62-2.71 (m, 16H), 6.87 (d, J ) 9.5

Hz, 8H), 6.98 (d, J ) 8.8 Hz, 16H), 7.04 (d, J ) 8.8 Hz, 16H),

7.07 (d, J ) 8.1 Hz, 16H), 7.32 (d, J ) 8.8 Hz, 16H), 7.36 (d,

J ) 8.8 Hz, 16H), 7.41 (d, J ) 8.8 Hz, 16H), 7.51 (m, 16H),

7.58-7.63 (m, 8H), 8.23 (m, 4H), 8.27 (s, 8H), 8.30 (s, 8H),

8.43-8.46 (m, 16H), 9.11 (m, 8H), 9.22 (d, J ) 8.8 Hz, 8H),

9.41 (d, J ) 7.3 Hz, 8H). MALDI-MS obsd, 8252.92 [M⁺],

8104.44 [[M - (4-t-Bu-Ph-O)]⁺], 7961.46 [[M - 2 × (4-t-Bu-

Ph-O)]⁺], 7483.56 [[M - 5 × (4-t-Bu-Ph-O)]⁺], 7334.92 [[M

- 6 × (4-t-Bu-Ph-O)]⁺]; calcd av mass, 8263.66 (C₅₇₂H₅₁₆N₁₂O₄₀-

Zn). λ_abs) 431, 507, 533 nm; λ_em(λ_ex) 530 nm) 564 (sh), 598,

648 nm.

Ack n ow led gm en t. This work was supported by the

NSF (Grant CHE-9988142). Mass spectra were obtained

at the Mass Spectrometry Laboratory for Biotechnology

at North Carolina State University. Partial funding for

the facility was obtained from the North Carolina

Biotechnology Center and the National Science Founda-

tion.

Su p p or tin g In for m a tion Ava ila ble: ¹H NMR and LDMS

(or MALDI-MS) of all new compounds; experimental proce-

dures for the synthesis of the atropisomers of a 9-ethynyl-

substituted bis(perylene)porphyrin; and energy-transfer simu-

lations. This material is available free of charge via the

Internet at http://pubs.acs.org.

J O0258002

6534 J . Org. Chem., Vol. 67, No. 18, 2002

Article Doi

Synthesis and photophysical properties of light-harvesting arrays comprised of a porphyrin bearing multiple perylene-monoimide accessory pigments

DOI: 10.1021/jo0258002

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo0258002

Full text of DOI:10.1021/jo0258002

Products guided by the article

R&D Labs maybe for 468082-98-6

Relevant to this article

Hot Product