Design, synthesis, and characterization of prototypical multistate counters in three distinct architectures

View Article Online / Journal Homepage / Table of Contents for this issue

Design, synthesis, and characterization of prototypical multistate

counters in three distinct architectures{

Karl-Heinz Schweikart,^aVladimir L. Malinovskii,^aJames R. Diers,^bAmir A. Yasseri,^b

David F. Bocian,*^bWerner G. Kuhr*^band Jonathan S. Lindsey*^a

^aDepartment of Chemistry, North Carolina State University, Raleigh, North Carolina,

27695-8204, USA. E-mail: jlindsey@ncsu.edu

^bDepartment of Chemistry, University of California, Riverside, California, 92521-0403, USA.

E-mail: david.bocian@ucr.edu; werner.kuhr@ucr.edu

Received 20th September 2001, Accepted 3rd January 2002

First published as an Advance Article on the web 25th February 2002

The storage of multiple bits of information in distinct molecular oxidation states is anticipated to afford

extraordinarily high memory densities. Among redox-active molecules, lanthanide porphyrinic triple-decker

complexes are attractive candidates for such molecular information storage elements because they provide at

least four cationic states over a modest potential range. Our approach toward a molecular multistate counter

involves covalently linking two triple deckers with interleaving oxidation potentials to achieve a commensurate

increase in the number of oxidation states within a single dyad. We report the design and synthesis of ﬁve

dyads comprised of triple-decker complexes of the form (Pc)Ln(Pc)Ln(Por) or (Por¹)Ln(Pc)Ln(Por²), where

Pc ~ phthalocyaninato, Por ~ porphyrinato, and Ln ~ Ce or Eu. The dyads were prepared in a modular

building-block fashion by Pd-mediated coupling of an iodophenyl-triple decker and an ethynylphenyl-triple

decker. The different dyad architectures examined include a vertical architecture with one linker (V1), a

horizontal architecture with one linker (H1), and a horizontal architecture with two linkers (H2). In each dyad,

one or two S-acetylthiomethyl groups incorporated on the porphyrinic moiety of the triple decker enables

attachment to an electroactive surface via in situ cleavage of the protected thiol linker(s). Three dyads

(Dyad1–3) have the V1 architecture but different compositions of triple deckers; three dyads (Dyad3–5) have

identical composition but different architectures (V1, H1, H2) for comparison of the properties in self-

assembled monolayers (SAMs) on gold. The SAM of each dyad exhibits robust reversible electrochemical

behavior; the redox waves are essentially the sum of the waves of the component triple deckers. In contrast,

mixtures of the component triple decker monomers form poor quality SAMs with inferior electrochemical

characteristics. Accordingly, all three dyad architectures are viable constructs for assembling a multistate

counter.

and porphyrinic lanthanide triple-decker sandwich com-

Introduction

plexes^6,7as the redox units prepared as self-assembled

monolayers (SAMs) on electroactive surfaces (e.g., gold).

Among these different classes of molecules, the triple-decker

sandwich complexes⁸have proved particularly attractive for

multibit information storage because they typically exhibit at

least four oxidation states in the range 0–1.4 V (vs. Ag/Ag¹),

corresponding to the formation of the mono-, di-, tri-, and

tetracations.^7,9The fact that the triple deckers possess four

readily accessible cationic oxidation states provides for

counting from one to four (where the neutral state is zero),

which enables the storage of two bits of information. A

complementary approach employs polynuclear cobalt com-

plexes of polypyridine ligands that afford up to 10 resolvable

anionic states in solution over the range of 20.5 to 22.9 V.¹⁰

Another approach employs an FeS cluster in the core of a

dendrimer.¹¹

One of our objectives was to extend the number of distinct

oxidation states from four to eight, thus enabling the storage of

three bits of information. In principle this can be achieved by

utilizing two different triple deckers with well resolved,

interleaved redox potentials. Toward this goal, we prepared

a library of triple-decker phthalocyaninato and porphyrinato

sandwich complexes of lanthanides (Eu, Ce) to elucidate how

different macrocycle substituents inﬂuence the redox potentials

of the complexes.^6,7Electron-donating groups attached to

the porphyrin and/or phthalocyanine building blocks were

Our groups have been investigating how multiple bits of

information can be stored at the molecular level.^1–7Our

approach toward molecular-based information storage utilizes

the distinct redox states of molecules. Information can be

stored either in cationic or anionic states; however, we have

chosen to work with molecules that afford a rich set of cationic

(rather than anionic) states, owing to their greater stability

under real-world (e.g., oxidizing) conditions. One of the

principal goals of our research program is to design molecular

systems that have a large number of resolvable oxidation states

whose potentials are less than 2 V. In addition to storing

multiple bits of information, our approach has attributes that

include (1) electrical writing/reading, (2) no moving parts,

(3) low power consumption, (4) scalability to molecular

dimensions, (5) operation under ambient conditions, and (6)

retention of stored charge for many minutes after removal of

applied potential.

In our prior investigations of multibit information storage, we

have employed thiol-functionalized porphyrins,^2–4ferrocenes,⁵

{Electronic supplementary information (ESI) available: ¹H NMR and

¹³C NMR spectra for each dipyrromethane; absorption, LD-MS,

and ¹H NMR spectra for each porphyrin and each triple decker;

absorption and LD-MS spectra for each triple-decker dyad. See http://

www.rsc.org/suppdata/jm/b1/b108520d/

808

J. Mater. Chem., 2002, 12, 808–828

This journal is # The Royal Society of Chemistry 2002

DOI: 10.1039/b108520d

View Article Online

employed to shift the redox potentials to less positive values.

Solution electrochemical studies revealed a variety of suitable

combinations of triple deckers that provide effective interleav-

ing of oxidation potentials. Several of these triple deckers were

then selected for thiol-derivatization and attachment to Au. In

general, the triple-decker SAMs were found to exhibit robust,

reversible electrochemical behavior.⁶These SAMs also exhib-

ited remarkably high charge-retention times (tens to hundreds

of seconds).⁷We then selected various combinations of the

thiol-derivatized triple deckers with interleaving oxidation

potentials and formed mixed SAMs of these complexes.⁷The

success of this approach depends on two criteria: (1) the

molecules partition equally onto the surface and (2) the redox

waves remain well-resolved and interleaved. Our initial studies

indicated that criterion 1 was met, but criterion 2 was not. In

particular, the mixed triple-decker SAMs exhibited very poorly

resolved redox waves. This behavior is most likely due to the

formation of very heterogeneous monolayers owing to the

different molecular shapes (because of the different macrocycle

substituents) of the two types of triple deckers.

In this paper, we describe an alternative approach aimed at

achieving criterion 2 above. In this approach two different

triple-deckers are joined in a covalent architecture bearing one

or two thiol linkers. The studies reported herein explore triple-

decker dyads in three different architectures. In one architec-

ture, the triple deckers are arranged vertically wherein the

planes of the macrocycles are anticipated to be (approximately)

perpendicular to the plane of the surface. This dyad contains a

single thiol linker and is designated V1. In the second and third

architectures, the triple deckers are arranged horizontally,

again with the planes of the macrocycles anticipated to be

(approximately) perpendicular to the plane of the surface.

One of these dyads contains a single thiol linker, and is

designated H1, whereas the other contains a linker on each of

the triple-decker building blocks and is designated H2. The

electrochemical properties of the several different dyads were

studied in both solution and in SAMs on a Au surface. The

studies reveal that the covalently linked dyad SAMs exhibit

well-resolved redox features (unlike the mixed triple-decker

SAMs) and are excellent candidates for preparing multistate

counters.¹²

Fig. 1 Schematic structures of the different types of triple deckers.

with double decker¹⁶(Por²)Eu(Pc) under reﬂux affords the

mixed type a triple decker (Por²)Eu(Pc)Ce(Por¹), which is

comprised of two different porphyrins (Por¹and Por²) and two

different metals (Ce and Eu). In the present work, these two

rational methods are employed for the synthesis of the type c

and type a triple-decker building blocks.

In our previous studies of triple deckers, we identiﬁed several

classes of complexes as promising candidates for constituents in

a multistate counter.⁷A collection of suitable triple deckers of

type a or type c, respectively, is presented in Chart 1. Type a

triple deckers (TD1–TD3) are of form (TTP)M¹(t-Bu₄Pc)M²-

(Por), and type c triple deckers (TD4–TD6) are of form

(t-Bu₄Pc)M¹(t-Bu₄Pc)M²(Por), where M¹/M²~ Ce or Eu;

t-Bu₄PcH₂~ tetra-tert-butylphthalocyanine; TTPH₂~ meso-

tetra-p-tolylporphyrin; and PorH₂is a porphyrin bearing

p-tolyl or n-pentyl groups at the meso-positions (except for

TD3 which is substituted with a p-iodophenyl group and a

trimethylsilyl (TMS)-ethynyl group as synthetic handles, and

two p-tolyl groups). Note that t-Bu₄PcH₂is composed of a

mixture of four regioisomers.¹⁸We have previously investi-

gated the Eu/Eu triple deckers.⁷The mixed Eu/Ce and Ce/Ce

triple deckers are new to this study. The Ce containing

complexes were synthesized because additional cationic states

are available due to the possibility of metal-centered oxida-

tion(s) (vide infra).¹⁹

The potentials for oxidation of TD1–TD6 in solution are

summarized in Table 1. The data for TD1, TD4, and TD6 were

taken from ref. 7; the data for TD2, TD3, and TD5 were

obtained in the present study. Examination of the electro-

chemical data shows that replacement of one or two Eu by Ce

does not lead to a large change in the potentials in most of the

cases (cf. TD1 and TD2; TD4 and TD5). The largest shift of

0.1 V for E_13/14is observed for Ce/Ce triple decker TD2 vs. Eu/

Eu triple decker TD1. However, Ce-containing triple deckers

generally exhibit more oxidation waves. Comparison of TD1

and TD3 shows that replacement of one Eu by Ce and

introduction of synthetic handles (iodophenyl and TMS-

ethynylphenyl) to one of the porphyrin rings results in minor

shifts of the redox potentials. Considering these observations,

we selected three pairs of triple deckers to be combined in dyads

(TD3/TD4, TD3/TD6, and TD4/TD6). These combinations

were not necessarily chosen to achieve the optimum interleav-

ing of potentials (i.e., to obtain the maximum number of

resolved oxidation waves). Instead, certain combinations were

Results and discussion

1 Overview

Choice of triple-decker constituents for the dyads. Lantha-

nide triple-decker sandwich complexes of porphyrins and

phthalocyanines can be prepared either by statistical^6,9or by

rational^13,14approaches. The statistical approach generally

results in a mixture containing the following three types of

triple deckers: (Por)Ln(Pc)Ln(Por) (type a), (Pc)Ln(Por)Ln(Pc)

(type b), and (Pc)Ln(Pc)Ln(Por) (type c) (Fig. 1).^7,9,15Isolation

of the desired triple decker from this mixture is typically

achieved by chromatography. Rational methods do not yet

exist for the synthesis of a wide variety of heteronuclear

heteroleptic members of the type a, b, and c complexes.

However, two rational methods are available for a limited

number of different central metals and porphyrin and

phthalocyanine ligands. Type c triple deckers are accessible

via a procedure analogous to that of Weiss,¹³wherein a double

decker¹⁶(Pc)Eu(Pc) is reacted with a half-sandwich complex

(Por)Eu(acac).¹⁴The latter is formed in situ starting from

the free base porphyrin and Eu(acac)₃?nH₂O in reﬂuxing 1,2,

4-trichlorobenzene (bp 214 uC).¹⁷We recently reported a

rational method for the preparation under milder conditions of

certain members of type a and type c triple deckers containing

Eu or Ce.¹⁴Treating CeI₃with LiN(SiMe₃)₂and a free base

porphyrin in bis(2-methoxyethyl) ether (bp 162 uC) and

reacting the resulting (Por¹)CeI half-sandwich complex the

J. Mater. Chem., 2002, 12, 808–828

809

View Article Online

Table 1 Potentials for the oxidation of the benchmark triple deckers in

solution^{a, b}

Potential/V

Triple

decker

E_0/11

E_11/12

E_12/13

E_13/14

E_14/15

E_15/16

Type a

TD1

TD2

0.26

0.24

0.62

0.67

0.64

0.98

0.93

1.00

1.27

1.17

1.24

1.42

1.48

1.70

1.68

TD3

Type c

TD4

TD5

0.13

0.06

0.09

0.56

0.57

0.45

1.01

0.89

1.25

1.26

1.60

TD6

^aData for TD1, TD4, TD6, from ref. 7; data for TD2, TD3, TD5, this

work. ^bObtained in BuCN (TD1, TD4, TD6) or CH₂Cl₂(TD2, TD3,

TD5) containing 0.1 M Bu₄NPF₆. E-values vs. Ag/Ag¹; FeCp₂/FeCp₂¹

0.19 V; scan rate ~ 0.1 V s²¹. Values are ¡0.03 V.

~

0.05–0.15 V more positive compared with those in solution.^2–4

The exact magnitude of the shift appears to depend on the

packing in the SAM. In addition, in large arrays, the potential

shift for a porphyrin closer to the surface appears to be slightly

larger than that for a porphyrin further from the surface. The

extent of these effects in the triple-decker dyads was unknown

at the outset of this study and was found to be dependent on the

exact composition of the dyad (vide infra).

The ﬁrst design consideration for a multistate counter

composed of two triple deckers is whether the dyad should be

constituted in a vertical architecture with one linker (V1) or a

horizontal architecture with one (H1) or two (H2) linkers, as

shown in Fig. 2. The vertical architecture affords the maximum

number of molecules per unit area (presuming the two triple

deckers are in registry) and could thereby increase functional

features such as information-storage density. However, the

diphenylethyne group that joins the two triple deckers permits

for torsional rotation about the –CMC– bond. Torsional

Chart 1

chosen to explore how dyad composition might affect SAM

quality. For example, the TD4/TD6 dyad is comprised of two

type c triple deckers, whereas the TD3/TD4 and TD3/TD6

dyads are each comprised of one type a and one type c triple

decker.

Dyad design strategy. The design of a viable multistate

counter comprised of two different triple deckers attached to an

electroactive surface requires the elucidation of the effects of

SAM formation on the redox characteristics of the dyad. In this

regard, prior studies of porphyrin arrays revealed that the

oxidation potentials of the SAMs are typically shifted

Fig. 2 Schematic structures of the vertical dyad with one linker (V1),

horizontal dyad with one linker (H1), and horizontal dyad with two

linkers (H2) in self-assembled monolayers (SAMs) attached via the

thiol linker.

810

J. Mater. Chem., 2002, 12, 808–828

View Article Online

rotation would signiﬁcantly increase the molecular area. In

addition, the two different triple deckers in the V1 architecture

are at intrinsically different distances from the electroactive

surface. This could be either be advantageous or deleterious for

features ranging from redox potential to charge-retention

characteristics. The horizontal architecture would occupy an

intrinsically larger molecular area on the surface. However, in

this design, the two triple deckers should be at similar distances

from the surface. Nonetheless, this feature of the H1 dyad

could be compromised by torsional rotation of the non-

attached triple decker around the –CMC– bond of the

diphenylethyne linker. In contrast, in the case of H2 dyads,

both triple deckers are attached to the surface and rotation is

suppressed. All three types of architectures were examined

herein to determine which design is best for achieving superior

SAM formation and redox characteristics.

The next design consideration is the placement of the

synthetic handles for joining the triple deckers and for

attachment of thiol linkers. These handles can be more easily

introduced via a suitably functionalized porphyrin rather than

a phthalocyanine because the synthetic chemistry of the former

class of molecules is better developed.²⁰To avoid possible

complications due to rotational isomers, we sought only one

derivatized porphyrin in each triple decker. In the V1 dyads,

the thiol linker is in the trans position with respect to the ethyne

linker between the two triple deckers, which requires a trans-

substituted porphyrin (trans-A₂BC) bearing the thiol linker. In

the H1 and H2 dyads, the thiol linker and the ethyne group are

in a cis position with respect to each other, which requires a

cis-substituted porphyrin (cis-A₂BC) bearing the thiol linker.

Iodophenyl and ethynylphenyl groups attached to the por-

phyrin constitute suitable handles for Sonogashira coupling of

thiol-derivatized linker groups at the triple-decker stage or

coupling of the two triple deckers to give the corresponding

dyad.

The incorporation of certain types of porphyrin building

blocks in triple-decker sandwich complexes can result in a

mixture of stereoisomers.²¹Porphyrins of the type A₄, A₃B,

cis-A₂B₂, trans-A₂B₂, and trans-A₂BC have at least one mirror

plane perpendicular to the plane of the macrocycle. As a result,

the formation of a corresponding triple decker affords a single

product with a mirror plane perpendicular to the porphyrin

and phthalocyanine rings.²²However, cis-A₂BC porphyrins

and ABCD porphyrins lack a perpendicular mirror plane and

are thus facially enantiotopic. The reaction with a lanthanide

metal occurs on one or the other of the two faces of the

porphyrin, affording the porphyrin half-sandwich complex as a

pair of enantiomers (Fig. 3). The reaction of an appropriate

double decker with the enantiomeric half-sandwich complexes

gives the corresponding triple decker as a pair of enantio-

mers.²²Note that the H1 and H2 dyads, but not the V1 dyads,

are derived from triple deckers that incorporate a cis-A₂BC

porphyrin.

The number of isomers grows for dyads comprised of

two different triple deckers when one or both incorporates a

cis-A₂BC porphyrin. For an H1 dyad wherein the triple decker

bearing the thiol linker is built around a cis-A₂BC porphyrin,

the triple decker dyad exists as a pair of enantiomers. For an

H2 dyad, both triple deckers incorporate one cis-A₂BC

porphyrin (Fig. 4). In the latter case, the reaction of the

enantiomeric mixture of one triple decker (derived from a

cis-A¹B¹C¹porphyrin) and the enantiomeric mixture of a

2

second triple decker (derived from a cis-A²B²C²porphyrin) at

2

the B sites yields the covalently linked triple-decker dyad as

four isomers.²²There are two pairs of syn/anti stereoisomers,

each of which consists of a pair of enantiomers. The terms syn

and anti refer to the positions of the C¹/C²substituents with

respect to each other. The thiol linker is coupled to the C¹/C²

sites for attachment of the resulting dyads to an electroactive

surface. The syn and anti isomers have different spatial

demands on the surface (Fig. 4), which could potentially

compromise the quality of the SAM.²³

2 Synthesis of triple-decker building blocks and dyads

Porphyrin building blocks. The rational synthesis of por-

phyrins bearing a distinct pattern of meso-substituents begins

with dipyrromethanes.²⁴Dipyrromethanes bearing aryl sub-

stituents at the 5-position are readily available via a one-ﬂask

synthesis.²⁵The 5-aryl dipyrromethanes (1–3) employed in this

work differ only in the nature of the p-aryl substituent (1 ~

methyl;²⁵2 ~ trimethylsilylethynyl;²⁶3 ~ iodo²⁵). Following a

general procedure,²⁷reaction of dipyrromethane 1 with

pyridyl thioester 4²⁷at 278 uC gave the monoacylated dipyr-

romethane 5 (Scheme 1). Treatment of 5 with EtMgBr followed

by acid chloride 6 in tetrahydrofuran (THF) at room

temperature afforded 8 in 65% yield. Similar treatment of 5

with 4-iodobenzoyl chloride (7) gave 9 in 68% yield. To

minimize the amount of free EtMgBr present at any time,

EtMgBr (2, 2, 1 equiv) and acid chloride (1, 1, 0.5 equiv) were

added sequentially and repeatedly at 5–10 min intervals rather

than the entire amount all at once.²⁷

The synthesis of the porphyrins proceeds by acid-catalyzed

condensation of a dipyrromethane and a dipyrromethane–

dicarbinol, followed by oxidation upon addition of 2,3-

dichloro-5,6-dicyano-1,4-benzoquinone (DDQ).²⁴This method

affords the corresponding substituted porphyrin. Using this

method, reduction of 8 with NaBH₄followed by condensation

of the resulting dicarbinol with dipyrromethane 2 in the

presence of triﬂuoroacetic acid (TFA) (30 mM in acetonitrile)

and subsequent oxidation with DDQ at room temperature

afforded cis-A₂BC porphyrin 11 in 20% yield (Scheme 2). In the

same manner, diacyldipyrromethane 9 and dipyrromethane 2

afforded cis-A₂BC porphyrin 12, and diacyldipyrromethane

10²⁸and dipyrromethane 3 gave the A₃B porphyrin 13, each in

21% yield. Porphyrins 11–13 are key precursors to the cor-

responding triple-decker building blocks. The porphyrins 5-(4-

iodophenyl)-10,20-bis(4-methylphenyl)-15-[4-(2-(trimethylsilyl)

ethynyl)phenyl]porphyrin (14)²⁴and 5-(4-iodophenyl)-10,15,

20-tri-n-pentylporphyrin (15)⁶have been synthesized previ-

ously using the same rational method.

Triple-decker building blocks. The type a and type c triple

deckers were prepared following rational synthetic pathways.

The synthesis of the type c triple-decker building blocks is

a

standard method,^13,14

Fig. 3 A meso-substituted porphyrin with substituents in a cis-A₂BC

(or ABCD) pattern is facially enantiotopic. The formation of a half-

sandwich complex with a lanthanide metal (M) affords a pair of

enantiomers (the ligand on the trivalent metal is omitted for clarity).

Subsequent reaction with a lanthanide double decker affords the

corresponding triple decker as a pair of enantiomers.²²

shown in Scheme 3. Following

porphyrin 14 was treated with Eu(acac)₃?nH₂O and double

decker¹⁶(t-Bu₄Pc)₂Eu^14,29in reﬂuxing 1,2,4-trichlorobenzene

(1,2,4-TCB) to afford triple decker TD7 in 74% yield, bearing

one iodophenyl and one TMS-ethyne group. Similarly, triple

J. Mater. Chem., 2002, 12, 808–828

811

View Article Online

Fig. 4 syn/anti Stereoisomers of an H2 dyad in a SAM. The coupling of two triple-decker building blocks, each of which comprises a pair of

enantiomers, at sites B¹1 B²(e.g., iodo 1 ethyne) affords four triple-decker dyads. The four dyads include a pair of syn enantiomers and a pair of

anti enantiomers. Such syn and anti isomers are expected to pack and orient differently upon attachment to a surface.²³

deckers TD8, TD9 and TD10 were obtained in 79%, 62% and

25% yield, respectively, starting from porphyrins 11, 13 and 15.

Note that in TD7 the two functional handles are in trans

conﬁguration, whereas they are in cis conﬁguration in TD8.

To provide for a free ethyne group in a subsequent

Sonogashira-coupling step, the TMS group of triple decker

TD8 was selectively cleaved in the presence of the TIPS group.

The cleavage was performed by treating TD8 with ﬁnely

powdered K₂CO₃in a mixture of chloroform, THF, and

methanol (1 : 5 : 2) at room temperature, affording ethyne

triple decker TD8’ in 90% yield after chromatography via a

single silica column (Scheme 4).

Compounds TD8’ and TD9 represent the desired functio-

nalized analogues to TD4 (Chart 1), whereas TD10 corre-

sponds to TD6. The two functional handles (TIPS-ethyne and

ethyne) in TD8’ are cis to each other, thereby qualifying TD8’

as a building block for horizontal dyads (see Fig. 2). The

iodophenyl unit in triple deckers TD9 and TD10 provides a site

for a Sonogashira coupling. Thus, TD9 and TD10 will

constitute the terminal, non-attached triple-decker units (TD²)

in V1 and H1 dyads.

The synthesis of the type a triple-decker building blocks is

shown in Scheme 5. Applying a standard procedure,¹⁴reaction

of porphyrin 12 with CeI₃and LiN(SiMe₃)₂in reﬂuxing bis(2-

methoxyethyl) ether followed by treatment with double

decker¹⁶(t-Bu₄Pc)Eu(TTP)¹⁴afforded TD11 in 54% yield. In

the same manner, TD3 was resynthesized in 46% yield by

reaction of porphyrin 14 with (t-Bu₄Pc)Eu(TTP). However,

implementation of the reported synthesis¹⁴of TD3 failed in

three out of ﬁve attempts, which prompted us to develop a

modiﬁed procedure employing stringent avoidance of air and

moisture. This reﬁned procedure was applied to the synthesis of

TD11 on three occasions without fail.

Subsequent Sonogashira coupling of TD11 with 1-[4-(S-

acetylthiomethyl)phenyl]acetylene (16)²in a mixture of dry

THF and triethylamine (TEA) containing CuI afforded

S-acetylthiomethyl-substituted triple decker AcS-TD11 in

57% yield (Scheme 6). In the same manner but using

812

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Scheme 2

derivatives).³⁰For packing in SAMs, the isomers of H1 and H2

triple-decker dyads caused by the cis-conﬁguration of the

porphyrin were anticipated to impose far greater spatial

constraints than those caused by different substitution patterns

of the tert-butyl groups about the perimeter of the phthalo-

cyanine ligands. We ﬁrst focused on the V1 dyads and decided

to prepare a series of three different dyads comprised of

monomers with a good interleaving of the oxidation potentials.

Scheme 1

N,N-diisopropylethylamine (DIEA) instead of TEA, AcS-TD3

and AcS-TD7 were obtained in 64% and 59% yield after

chromatography, starting from TD3 and TD7, respectively

(Schemes 7, 8).

Deprotection of the TMS-ethyne groups of AcS-TD3, AcS-

TD7 and AcS-TD11 with (n-Bu)₄NF in THF at 0 uC afforded

the desired triple-decker building blocks AcS-TD3’ (Scheme 7),

AcS-TD7’ (Scheme 8), and AcS-TD11’ (Scheme 6) in 97%, 89%

and 66% yield, respectively. Note that the S-acetylthio group

and the ethyne group in AcS-TD11’ are cis to each other,

whereas they are trans in AcS-TD3’ and AcS-TD7’. Thus, AcS-

TD3’ and AcS-TD7’ will allow the construction of V1 dyads,

whereas AcS-TD11’ is a building block for an H1 dyad (see

Fig. 2).

In summary, the triple-decker building blocks were prepared

via rational synthetic routes. Each triple decker prepared herein

exists as a mixture of isomers due to one or both of the

following types of substitution of the ligands. (1) The presence

of a cis-A₂BC porphyrin. A cis-A₂BC porphyrin is facially

enantiotopic and gives rise to enantiomers of triple deckers

(TD8, TD11, and their derivatives). (2) The presence of one or

two tetra-tert-butylphthalocyanine ligands. The four regio-

isomers of tetra-tert-butylphthalocyanine¹⁸result in a mixture

of diastereomers of the triple decker (TD1–TD11 and their

Vertical dyads with one linker (V1). The synthesis of Dyad1 is

shown in Scheme 9. S-Acetylthio-derivatized ethynylphenyl

triple decker AcS-TD7’ was reacted with iodophenyl triple

decker TD10 in a Sonogashira coupling, applying copper-free

reaction conditions [tris(dibenzylideneacetone)dipalladium(0)

(Pd₂(dba)₃) and tri-p-tolylphosphine (P(o-tol)₃) in a 5 : 1

mixture of toluene–DIEA at 35 uC] that we have developed for

use with porphyrins.³¹A copper-free Pd-catalyzed coupling

procedure was employed to avoid homo-coupling of monomer

AcS-TD7’. The progress of the coupling was monitored by

analytical size exclusion chromatography (SEC). Because the

reaction proceeded quite slowly, a further batch of catalyst

[Pd₂(dba)₃and P(o-tol)₃] was added twice during the course of

the reaction. After a total elapsed reaction time of 48 h, the

SEC trace still showed starting triple-decker monomers

(y32%), but also already a considerable amount of oligomeric

side products (y26%).³²Therefore, the reaction was stopped

by evaporating the solvent. The desired Dyad1 was then

obtained in 25% yield after a three-column procedure. The

J. Mater. Chem., 2002, 12, 808–828

813

View Article Online

Scheme 3

laser-desorption mass spectrometry (LD-MS) spectrum with

the matrix 1,4-bis(5-phenyloxazol-2-yl)benzene (POPOP)

showed the molecule ion peak at m/z ~ 5001.2 together with

several fragmentation peaks at lower mass.

m/z ~ 4979.4 and a peak at m/z ~ 4902.4 due to cleavage

of the S-acetylthio linker.

Note that Dyad2 and Dyad3 are each comprised of one type

a triple decker and one type c triple decker, whereas Dyad1

consists of two type c triple deckers. In addition, in Dyad2 and

Dyad3 one europium is replaced by cerium, which provides for

additional oxidation states in the information-storage mole-

cule. For the preparation of Dyad1–3, we initially chose the

hindered base DIEA as recommended in Pd-coupling reactions

involving the S-acetylthiophenyl group.³³However, we found

that in all three cases upon use of DIEA, the formation of the

triple-decker dyad proceeded very slowly and was accompanied

by the formation of considerable amounts of oligomeric side-

products as indicated by analytical SEC. The formation of

signiﬁcant amounts of higher molecular weight material is also

observed in Pd-coupling reactions with porphyrins.^2,31To

overcome the problems of long reaction times, low yields, and

the loss of starting material due to the formation of oligomers,

we decided to employ TEA in place of DIEA in the following

Pd-mediated coupling reactions. Coupling reactions involving

TEA have proved successful for the (S-acetylthiomethyl)aryl

group,^3,34which is more stable than the S-acetylthiophenyl

group, and is used as the linker in our case.

Analogously, S-acetylthio-derivatized triple decker AcS-

TD3’ was treated with TD10 in a Sonogashira reaction to

give Dyad2 (Scheme 10). Again, further catalyst was added

twice, because the reaction had slowed down. Analytical SEC

after 21 h revealed that oligomeric side-products had formed

(y39%) while unreacted triple-decker monomers (y26%) were

still present in the reaction mixture.³²The reaction was stopped

after 24 h, affording Dyad2 in 25% yield after chromatography.

The LD-MS spectrum showed the molecule ion peak at m/z ~

4922.1, and also a peak at m/z ~ 4876.2 corresponding to

fragmentation owing to the cleavage of the S-acetylthio group.

The same conditions [Pd₂(dba)₃, P(o-tol)₃in a 5 : 1 mixture

of toluene–DIEA at 35 uC] were applied to the reaction of

AcS-TD3’ and iodophenyl triple decker TD9 (Scheme 11). The

coupling reaction was monitored by analytical SEC, and a

further amount of catalyst was added after 20 h. After a total

reaction time of 24 h, removal of the solvent and chromato-

graphic work-up afforded Dyad3 in 51% yield. The LD-MS

spectrum (POPOP) showed the molecule ion peak at

814

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Scheme 4

Horizontal dyads with one linker (H1). An H1 dyad

analogous to Dyad3 was prepared to investigate the inﬂuence

of the different architectures of triple-decker dyads on the

electrochemical properties in solution as well as in SAMs. For

this purpose, ethynylphenyl triple decker AcS-TD11’ was

treated with iodophenyl triple decker TD9, applying the

reaction conditions described above with use of TEA instead

of DIEA (Scheme 12). The reaction was monitored by

analytical SEC and by LD-MS. The reaction proceeded

faster than those employing DIEA and was stopped after

8.5 h. However, further catalyst had to be added twice. The

desired Dyad4 was obtained in 38% yield after chromatography

on silica, SEC (THF), and again on silica. LD-MS (POPOP)

showed the molecule ion peak at m/z ~ 4975.8. Note that

Dyad4 is isomeric to Dyad3 with the S-acetylthiomethyl linker

cis instead of trans to the inter-triple decker axis.

Horizontal dyads with two linkers (H2). In the H2 dyads,

each triple decker bears one S-acetylthiomethyl linker for

attachment to the gold surface. As stated above, this

arrangement is anticipated to prevent rotation of the triple-

decker moieties around the bridging ethyne group upon

binding to a surface, but a mixture of syn/anti isomers will

also be present. For purposes of comparing the H1 and H2

architectures, we prepared the H2 dyad analogous to Dyad4,

substituted with a thiol-attachment unit at both triple deckers.

Applying the same reaction conditions as for the synthesis of

Dyad4, iodophenyl triple decker TD11 was coupled with

ethynylphenyl triple decker TD8’ (Scheme 13). Note that both

TD11 and TD8’ each incorporate one cis-A₂BC porphyrin.

After 5.5 h, analytical SEC indicated y67% of product TMS/

TIPS-Dyad5, and the reaction was stopped.³²In contrast to the

coupling reactions involving DIEA, almost no oligomeric

species were present in the reaction mixture. The desired dyad

was isolated in 59% yield after a two-column procedure. The

TIPS and TMS groups of TMS/TIPS-Dyad5 were both cleaved

in one step by treatment with (n-Bu₄)NF in THF at room

Scheme 5

temperature. LD-MS showed that deprotection was complete

after 1 h, affording Dyad5’ in 96% yield. The S-acetylthio-

methyl-linker groups were then introduced by Sonogashira

coupling [Pd₂(dba)₃, P(o-tol)₃, toluene–TEA (5 : 1), 35 uC] of

Dyad5’ with excess 1-(S-acetylthiomethyl)-4-iodobenzene (17),²

affording the desired Dyad5 in 42% yield. LD-MS showed the

molecule ion peak at m/z ~ 5155.7 together with peaks at

m/z ~ 5080.8 and 4980.6 owing to cleavage of the linker

groups. Dyad5 consists of a large number of isomers due to the

four syn/anti architectures and the presence of the four patterns

of tert-butyl groups on each phthalocyanine ligand. The four

syn/anti architectures are expected to exhibit distinct spatial

demands as shown in Fig. 4.

J. Mater. Chem., 2002, 12, 808–828

815

View Article Online

Scheme 6

3 General characterization of triple-decker building blocks and

dyads

TD9, TD10), which are attributed to the B-band of the

phthalocyanine moieties. Absorptions in the region at 421–427

nm for type a and at 416–418 nm for type c are assigned to the

porphyrin B-band. The remaining absorption in the visible

region is attributed to the Q-bands of the macrocycles, in which

the lower-energy Q-bands are contributed mainly by the

phthalocyanine.³⁵The data are in good agreement with those

reported for similar heteroleptic triple-decker (phthalocyani-

nato)(porphyrinato)Eu(^III)^6,9,36,37and Ce(III) complexes.^38,39

Interestingly, the different substituents at the aryl groups of the

porphyrins within two series (type a triple deckers TD3, AcS-

TD3, AcS-TD3’, TD11, AcS-TD11, AcS-TD11’; type c triple

deckers TD8, TD8’, TD9) do not induce a substantial shift of

the absorption maxima, an observation noted by Ng for some

Eu(^III) triple-decker complexes.⁹In the case of type a triple

deckers, the porphyrin B-band is broadened or split owing to

the presence of two different porphyrins. Due to the decrease in

ratio of porphyrin : phthalocyanine when going from type a

(2 : 1) to type c (1 : 2); this band is the most intense in

the spectra for type a, whereas the phthalocyanine B-band is

dominant for type c.^9,36,38,39Similar features are observed for

the triple-decker dyads, and their absorption spectra are

essentially the sum of the spectra of the corresponding

monomers. The phthalocyanine B-band of Dyad2–5 (com-

prised of type a and type c) is located at 350–353 nm (between

that of type a and type c triple deckers). The porphyrin B-band

is broadened or split and shows the highest intensity. In the

The triple-decker building blocks and dyads were assessed for

homogeneity by thin layer chromatography (TLC) and by

analytical SEC. Characterization was generally performed by

absorption spectroscopy, LD-MS, ¹H NMR spectroscopy, and

fast-atom bombardment mass spectrometry (FAB-MS). Many

of the triple deckers and all of the S-acetylthiomethyl-

derivatized dyads were characterized electrochemically in

solution, and all of the latter were also characterized elec-

trochemically in SAMs.

Mass spectra. LD-MS has proved particularly effective for

identifying the desired lanthanide porphyrin/phthalocyanine

complexes. The triple-decker monomers each display the

molecule ion peak together with characteristic fragmentation

peaks upon analysis without a matrix. For the dyads, the

best spectra were obtained employing POPOP as a matrix. In

addition, a FAB-MS spectrum was obtained for each of

Dyad1–4. Samples of TMS/TIPS-Dyad5, Dyad5’, and Dyad5

did not ionize.

Absorption spectra. The UV–Vis spectra of the triple-decker

monomers exhibit absorptions at 362–364 nm for type a

(AcS-TD3, AcS-TD3’, TD11, AcS-TD11, AcS-TD11’) and at

345–348 nm for type c (TD7, AcS-TD7, AcS-TD7’, TD8, TD8’,

816

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Scheme 7

case of the only type c/type c dyad (Dyad1), the phthalocyanine

B-band is the most intense absorption band. Thus, a simple

inspection of the UV/Vis spectra allows distinction between

type a/type c and type c/type c triple-decker dyads.

with the b-pyrrole protons of the porphyrin ligand give rise to

a broad multiplet in the range of 3.6–2.5 ppm. Speciﬁc

assignments can be made for some signiﬁcant signals. A singlet

at d ~ 0.84 ppm in the spectrum of TD8 is ascribed to the TMS

group. This signal is absent in the spectrum of TD8’, showing

complete deprotection. Instead, a singlet is observed at d ~

3.86 ppm due to the free ethyne group. The TIPS group in TD8

or TD8’ gives rise to a sharp peak at d ~ 1.77 or 1.75 ppm,

respectively.

Triple deckers of type a (AcS-TD3, AcS-TD3’, TD11, AcS-

TD11, AcS-TD11’) contain two different porphyrins (TTP and

a porphyrin bearing 3 or 4 nonequivalent phenyl rings). The

signals are generally shifted to higher ﬁelds owing to the

presence of one Ce instead of Eu in the triple-decker com-

plex.^14,41The TMS groups in TD11, AcS-TD11, and AcS-TD3

result in a singlet between d ~ 20.05 and 20.08 ppm. After

deprotection, a singlet is found at d ~ 2.56 ppm for the ethyne

proton in AcS-TD11’ or AcS-TD3’. The S-acetylthiomethyl

linker groups in each of AcS-TD3, AcS-TD3’, AcS-TD11, and

AcS-TD11’ give rise to singlets at d ~ 2.26–2.27 and 3.96–

3.98 ppm.

¹H NMR spectra. Heteroleptic triple-decker complexes with

1

paramagnetic metal centers such as Ce(^III) and Eu(III) give H

NMR spectra wherein the signals are generally broad and

spread over a wide region.³⁵In addition, the presence of one or

more tetra-tert-butylphthalocyanine ligand (comprised of four

regioisomers) in each triple decker caused the ¹H NMR spectra

of the triple-decker monomers and dyads to be very complex

and difﬁcult to interpret.³⁰In particular, the aromatic protons

result in broad (eventually overlapping) multiplet signals.

Restricted rotation along the C(meso)–C(aryl) bonds in

porphyrins of symmetrical type a and type c Eu(^III) triple-

decker complexes has been attributed as the source of 4 or 5

signals for the meso-phenyl protons.^9,36The triple deckers of

type c (TD8, TD8’, TD9, TD10) are comprised of two

nonequivalent phthalocyanine rings (inner and outer in the

sandwich complex) and a porphyrin bearing 3 or 4 non-

equivalent meso-aryl rings, which further complicates the

spectra. Nevertheless, by comparison with data for similar

symmetrical triple deckers,^6,9,36,40multiplets at d ~ 13.6–12.6,

11.26–11.24, 11.0–8.6, 7.4–7.1, 6.8, and 5.4–5.0 ppm can be

attributed to the a and b protons of the phthalocyanines and

to the meso-phenyl protons of the porphyrin moiety. The

tert-butyl groups on the phthalocyanine macrocycle together

The ¹H NMR spectra of the triple-decker dyads are even

more complex owing to the superimposed (complex) spectra of

the corresponding monomers. However, as in the case for the

monomers, singlet signals at d ~ 2.61 and 4.44 ppm for Dyad1

and at d ~ 2.31–2.32 and 4.01–4.06 ppm for Dyad2–5 are

attributable to the S-acetylthiomethyl linker(s) attached to one

or both triple-decker moieties.

J. Mater. Chem., 2002, 12, 808–828

817

View Article Online

Scheme 8

Scheme 9

818

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Scheme 10

Scheme 11

J. Mater. Chem., 2002, 12, 808–828

819

View Article Online

Scheme 12

4 Electrochemical characteristics of the dyads in solution and in

SAMs

data in Fig. 5–9 (bottom panels). The oxidation potentials for

the SAMs of all ﬁve dyads are summarized in Table 2. In

general, the voltammograms of the SAMs of all ﬁve dyads are

of excellent quality and are far superior to those obtained for

the mixed monolayers of triple-decker monomers that we

previously examined.⁷These results indicate that the strategy

of covalently linking the triple deckers is key to obtaining a

viable multistate counter. In addition, all three architectures

appear to be robust. Closer inspection of the data for the dyad

SAMs reveals a number of other features. These features are

discussed in more detail below.

(1) The quality of the voltammograms for the dyad SAMs

appears to be dependent on the composition of constituent

triple deckers. In particular, the V1 dyads Dyad1 and Dyad2,

both of which contain a meso-pentyl-substituted porphyrin in

the upper triple decker, give superior voltammetry to Dyad3,

which contains a meso-aryl-substituted porphyrin in this unit.

This is evidenced by the fact that Dyad3 appears to exhibit a

subset of oxidation waves that are shifted to lower potential

than the principal peaks (see Fig. 7, bottom panel). For

example, additional waves are evident at y0.93 V and y0.02 V.

These lower potential waves could correspond to a minority

domain wherein the dyads are more poorly packed (hence the

lower potential, vide infra) than the majority domain. The

additional oxidation waves are absent in the SAMs of the

horizontal Dyad4 and Dyad5 (although there is a hint of

another wave at y0.82 V for the former dyad). Thus, it is

possible that certain dyads are better suited for a horizontal

architecture than a vertical one (and vice versa). Additional

studies on dyads of like composition in horizontal vs. vertical

architectures will be required to fully elucidate this issue.

Solution. The solution electrochemical properties of Dyad1–

5 were examined and compared with those of the most

structurally similar triple-decker benchmark complexes. These

benchmarks are as follows: Dyad1, TD4/TD6; Dyad2, TD3/

TD6; Dyad3–5, TD3/TD4. The square-wave voltammograms

of Dyad1–5 are shown in Fig. 5–9 (top panels). The potentials

for all ﬁve dyads are summarized in Table 2. In the table, the

potentials are listed in order of increasing value. Accordingly,

each successive pair of values for a dyad corresponds to the

lower potential for each of the two triple deckers in the dyad

(i.e., dyad (E_0/11/E_11/12) corresponds to TD^a(E_0/11)/TD^b(E_0/11);

dyad(E_12/13/E_13/14) corresponds to TD^a(E_11/12)/TD^b(E_11/12),

etc., where TD^aand TD^bdenote the two triple deckers in

the dyad). Comparison of the data in Tables 1 and 2 shows that

the solution redox potentials of the dyads are quite close

to those of the benchmark triple deckers. In all cases, the

potentials for the analogous states of the dyads and the

benchmark triple deckers differ by 0.05 V or less. The largest

differences in potential (albeit small) occur for Dyad1 and

Dyad2 where one of the meso-pentyl substituents of the

porphyrin in the benchmark triple decker is replaced by a

diphenylethyne linker. The potential differences are negligible

for the other dyads because the diphenylethyne linker and the

(S-acetylthiomethyl)diarylethyne linker(s) have very similar

electron-donating properties to that of the aryl group of the

benchmark triple deckers.

SAMs. Representative fast-scan (100 V s²¹) voltammo-

grams of the SAMs of Dyad1–5 are compared with the solution

820

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Scheme 13

(2) For each of Dyad1–5, the oxidation potential for a

particular wave in the SAM is generally more positive than the

corresponding wave in solution by 0.05–0.20 V. This behavior

is consistent with our previous studies of porphyrin SAMs.^2–7

However, Dyad5 does not follow this trend. For this dyad, the

oxidation potentials of the SAM are generally lower than those

observed for the solution. We have no clear explanation for this

phenomenon. The positive shift in redox potentials observed

upon SAM formation has generally been attributed to poorer

access of counterions to the redox center due to packing of the

molecules on the surface.⁴²If so, the absence of a positive

potential-shift would suggest relatively poor packing. In this

regard, the area per molecule occupied by Dyad5 is much larger

than that of any of the other dyads (vide infra). This could be

due to the fact that Dyad5 is in an H2 architecture, which

renders the molecule more rigid and perhaps compromises

packing on the surface.

similar in solution and in the SAMs. For example, in the case of

Dyad2, the splitting between E_0/11and E_11/12is 0.22 V in

solution and 0.18 V in the SAM; the splitting between E_12/13

and E_13/14is y0.17 V in solution and 0.15 V in the SAM (see

Table 2). Likewise for Dyad1, the waves for E_0/11and E_11/12

are overlapped in both the solution and the SAM. This

behavior shows that for the design of an optimum multistate

counter, it is not necessary to consider differential surface

effects (at least for the redox potential) on the upper and lower

members of vertical dyads.

(4) The surface area measurements indicate that dyads in the

vertical architecture generally occupy less surface area than

those in the horizontal architectures. In particular, the average

2

˚

area for the three V1 structures (Dyad1–3) is y760 A

molecule²¹, whereas the areas for the H1 (Dyad4) and H2

2

(Dyad5) structures are y1640 A molecule²¹and 2700 A

˚

molecule²¹, respectively. All of these areas are relatively large

compared with those that might be expected for a closest-

packed structure. Based on the molecular dimensions, closest

(3) For each of the V1 dyads Dyad1–3, the potentials of the

lower vs. upper triple deckers do not appear to be differentially

shifted upon SAM formation. This is evident from the fact that

the splittings between the pairs of E-values that correspond to

oxidation of each triple decker in the dyad are qualitatively

2

˚

packing of the V1 dyads would yield areas of y300 A

molecule²¹; closest packing of the H1 or H2 dyads would yield

areas of y700 A molecule²¹(assuming (1) no torsional

2

˚

J. Mater. Chem., 2002, 12, 808–828

821

View Article Online

Fig. 5 Voltammetry of Dyad1 in solution (top panel) and in a SAM

(bottom panel). The solvent was CH₂Cl₂containing 0.1 M (solution) or

1.0 M (SAM) Bu₄NPF₆; the scan rate was 0.1 V s²¹(solution) or 100 V s²¹

(SAM).

Fig. 7 Voltammetry of Dyad3 in solution (top panel) and in a SAM

(bottom panel). The solvent was CH₂Cl₂containing 0.1 M (solution) or

1.0 M (SAM) Bu₄NPF₆; the scan rate was 0.1 V s²¹(solution) or 100 V s²¹

(SAM).

rotation for the V1 and H1 structures about the –CMC– bond of

the linker and (2) a perpendicular orientation of the macrocycle

planes with respect to the surface). The larger areas for the V1

and H1 architectures could be due to torsional rotation. The

very large area for the H2 structure could result from the added

rigidity of the molecule upon binding to the surface (as noted

above); the presence of syn versus anti isomers of this dyad

could also contribute. Regardless, the fact that the vertical

architecture generally results in more dense packing of the

dyads is an important design consideration for a viable mul-

tistate counter. On nanoplatforms, more dense packing would

afford more facile detection of electrical signals from the

molecules in the SAM.

Conclusions

We have described the design of covalently linked dyads of

lanthanide porphyrinic triple-decker complexes with interleav-

ing cationic oxidation potentials that could be used in a

molecular multistate counter. Five dyads were prepared by Pd-

catalyzed coupling (Sonogashira reaction) of appropriately

functionalized triple-decker building blocks (type a or type c).

Fig. 6 Voltammetry of Dyad2 in solution (top panel) and in a SAM

(bottom panel). The solvent was CH₂Cl₂containing 0.1 M (solution) or

1.0 M (SAM) Bu₄NPF₆; the scan rate was 0.1 V s²¹(solution) or 100 V s²¹

(SAM).

Fig. 8 Voltammetry of Dyad4 in solution (top panel) and in a SAM

(bottom panel). The solvent was CH₂Cl₂containing 0.1 M (solution) or

1.0 M (SAM) Bu₄NPF₆; the scan rate was 0.1 V s²¹(solution) or 100 V s²¹

(SAM).

822

J. Mater. Chem., 2002, 12, 808–828

View Article Online

Experimental

General

¹H NMR spectra were collected in CDCl₃(300 MHz) unless

noted otherwise. Absorption spectra (HP 8451A, Cary 3) were

collected in toluene. Porphyrins and triple-decker sandwich

complexes were analyzed by LD-MS (Bruker Proﬂex II) and

FAB-MS (JEOL HX 110HF), either in high (HR) or low

resolution (LR). High-resolution mass spectrometry was

carried out at greater than unit resolution. LD-MS analysis

was done without a matrix or with the matrix POPOP.⁴³

All operations involving organometallic compounds and Pd-

catalyzed couplings were carried out under an inert atmosphere

of argon using standard Schlenk techniques. Eu(acac)₃?nH₂O

was obtained from Alfa Aesar. Unless otherwise indicated, all

other reagents were obtained from Aldrich Chemical Com-

pany, and all solvents were obtained from Fisher Scientiﬁc.

Non-commercial compounds

The following compounds were prepared as described in

the literature: dipyrromethanes 1,²⁵2,²⁴and 3;²⁵S-2-pyridyl

(4-methyl)thiobenzoate (4);²⁷acid chlorides 6 and 7;²⁴diacyl

dipyrromethane 10;²⁸porphyrins 14²⁴and 15;⁶double-decker

complexes¹⁶(TTP)Eu(t-Bu₄Pc) and (t-Bu₄Pc)₂Eu;^14,29and

thiol-protected linkers 16 and 17.²Triple deckers TD1,⁷

TD2,¹⁴TD3,¹⁴TD4,⁷TD5,¹⁴and TD6⁷have been prepared

previously.

Fig. 9 Voltammetry of Dyad5 in solution (top panel) and in a SAM

(bottom panel). The solvent was CH₂Cl₂containing 0.1 M (solution) or

1.0 M (SAM) Bu₄NPF₆; the scan rate was 0.1 V s²¹(solution) or 100 V s²¹

(SAM).

The porphyrins in the triple deckers and the triple deckers

themselves were prepared by rational synthetic methods. Three

of the ﬁve dyads were constituted for a vertical architecture

(V1), bearing one S-acetylthiomethyl group for attachment to

an electroactive gold surface, whereas the other two dyads were

designed for a horizontal arrangement with one (H1) or two

(H2) S-acetylthiomethyl linkers.

The electrochemical characteristics of the ﬁve thiol-

derivatized dyads were investigated both in solution and in

SAMs. The dyad SAMs generally yield robust, reversible

voltammograms with features comparable to those observed in

solution (albeit with some shifts in potential). This behavior

can be contrasted with that observed for mixtures of non-

covalently linked triple-decker complexes. These complexes

exhibit poorly resolved redox waves indicative of a highly

heterogeneous, poorly packed SAM. The covalent linkage

essentially overcomes this problem. The high quality voltam-

metric characteristics of the dyad SAMs indicate that these

constructs are excellent candidates for multistate counters that

could be used for storage of information at the molecular level.

Chromatography

Adsorption column chromatography was performed using

ﬂash silica (Baker, 60–200 mesh). Preparative-scale SEC was

performed using BioRad Bio-beads SX-1 in THF and eluted

with gravity ﬂow.⁴⁴Analytical-scale SEC was performed with a

˚

Hewlett-Packard 1090 HPLC using a 1000 A column (5 mL,

styrene–divinylbenzene copolymer) with THF as eluent

(0.8 mL min²¹).⁴⁴Preparative-scale TLC was performed on

silica gel plates (Merck, 60 F₂₅₄s, layer thickness 0.5 mm).

1-(4-Methylbenzoyl)-5-(4-methylphenyl)dipyrromethane (5)

Following a standard procedure,²⁷a solution of EtMgBr

(43 mL, 43 mmol, 1.0 M in THF) was slowly added via syringe

to a stirred solution of dipyrromethane 1 (4.06 g, 17.2 mmol) in

THF (17 mL) under Ar. The mixture was stirred at room

temperature for 10 min and then cooled to 278 uC. S-2-Pyridyl

(4-methyl)thiobenzoate (4) (3.94 g, 17.2 mmol) was then added,

and the mixture was maintained at 278 uC for 30 min. The

cooling bath was removed, and the mixture was maintained at

Table 2 Potentials for the oxidation of the dyads in solution^aand in SAMs^b

Potential/V^c

Dyad

E_0/11

E_11/12

E_12/13

E_13/14

E_14/15

E_15/16

E_16/17

E_17/18

Solution

Dyad1

Dyad2

Dyad3

Dyad4

Dyad5

y0.07

0.01

0.07

0.06

0.07

y0.07

0.23

0.24

0.25

0.48

0.44

y0.58

y0.60

0.55

0.61

y0.58

y0.60

0.94

y0.94

y0.99

y1.01

1.05

y0.94

y0.99

y1.01

y1.28

y1.25

y1.24

y1.25

y1.28

y1.25

y1.24

y1.25

SAM

Dyad1

Dyad2

Dyad3

Dyad4

Dyad5

y0.15

0.20

0.26

0.16

0.05

y0.15

0.38

0.46

0.35

0.20

0.52

0.57

y0.78

y0.67

y0.53

0.61

0.72

y0.78

y0.67

y0.53

0.98

y1.05

y1.14

y1.04

y0.91

1.08

y1.05

y1.14

y1.04

y0.91

y1.31

y1.35

y1.40

y1.29

y1.16

y1.31

y1.35

y1.40

y1.29

y1.16

^aObtained in CH₂Cl₂containing 0.1 M Bu₄NPF₆. E-values vs. Ag/Ag¹; FeCp₂/FeCp₂¹~ 0.19 V; scan rate ~ 0.1 V s²¹. Values are ¡0.03 V.

^bObtained in CH₂Cl₂containing 1.0 M Bu₄NPF₆. E-values vs. Ag/Ag¹; FeCp₂/FeCp₂¹~ 0.19 V; scan rate ~ 100 V s²¹. E-values are listed as

approximate because the oxidation waves for the two triple deckers in the dyad overlap.

c

J. Mater. Chem., 2002, 12, 808–828

823

View Article Online

room temperature for 30 min. The reaction was quenched with

saturated aqueous NH₄Cl. The mixture was poured into ethyl

acetate. The organic layer was washed with water, dried

(Na₂SO₄), and concentrated, affording a dark foam. Puriﬁca-

tion by ﬂash column chromatography [silica, neat CH₂Cl₂to

CH₂Cl₂–ethyl acetate (100 : 3)] afforded a golden amorphous

solid (4.56 g, 75%): mp 77–81 uC; ¹H NMR d 2.48 (s, 3H), 2.58

(s, 3H), 5.67 (s, 1H), 6.14 (m, 1H), 6.24 (m, 1H), 6.30 (m, 1H),

6.83 (m, 1H), 6.96 (m, 1H), 7.20–7.30 (m, 4H), 7.41 (d, J ~

7.8 Hz, 2H), 7.89 (d, J ~ 7.8 Hz, 2H), 8.35 (br s, 1H), 9.88 (br s,

1H); ¹³C NMR d 18.7, 21.3, 21.8, 44.0, 107.8, 108.7, 110.5,

110.6, 117.8, 120.5, 128.4, 129.2, 129.3, 129.7, 130.8, 131.4,

135.9, 137.2, 137.9, 141.7, 142.5, 184.5; FAB-MS obsd (HR)

354.1754, calcd exact mass 354.1732 (C₂₄H₂₂N₂O); Anal. Calcd

for C₂₄H₂₂N₂O: C, 81.33; H, 6.26; N, 7.90. Found: C, 80.69; H,

6.34; N, 7.73%.

5-{4-[2-(Triisopropylsilyl)ethynyl]phenyl}-15,20-bis(4-

methylphenyl)-10-{4-[2-(trimethylsilyl)ethynyl]phenyl}porphyrin

(11)

Following a standard procedure,²⁴a solution of 8 (420 mg,

0.65 mmol) in a mixture of dry THF–methanol (10 : 1, 29 mL)

under a stream of Ar at room temperature was treated with

NaBH₄(497 mg, 13.1 mmol, 20 mol equiv) in small portions

(y0.1 g every 2 min) with rapid stirring. The progress of the

reaction was monitored by TLC [alumina, CH₂Cl₂–ethyl

acetate (3 : 2) and silica, CH₂Cl₂–ethyl acetate (5 : 1)].

After the reaction was complete (about 45 min), the reaction

mixture was poured into a stirred mixture of saturated aqueous

NH₄Cl (40 mL) and CH₂Cl₂(80 mL). The organic phase was

washed with water, dried (Na₂SO₄), and placed in a 500 mL

round-bottomed ﬂask. Removal of the solvent under reduced

pressure yielded the dicarbinol as a solid (y0.65 mmol). A

sample of dipyrromethane 2 (208 mg, 0.65 mmol) was added

followed by acetonitrile (265 mL). TFA (613 mL, 7.96 mmol)

was then added, and the reaction was monitored by UV/Vis

spectroscopy. After 3 min, DDQ (438 mg, 1.93 mmol) was

added, and the mixture was stirred at room temperature for 1 h.

Then triethylamine (1.11 mL, 7.96 mmol) was added, and the

entire reaction mixture was ﬁltered through a pad of alumina

(eluted with CH₂Cl₂) until the eluent was no longer dark.

After removal of the solvent, column chromatography [silica,

CH₂Cl₂–hexanes (1 : 1)] afforded a purple solid (121 mg, 20%):

¹H NMR d 22.80 (s, 2H), 0.38 (s, 9H), 1.26 (s, 21H), 2.71 (s,

6H), 7.56 (d, J ~ 7.8 Hz, 4H), 7.85–7.89 (m, 4H), 8.09 (d, J ~

7.2 Hz, 4H), 8.15 (d, J ~ 7.5 Hz, 4H), 8.78–8.88 (m, 8H); LD-

MS obsd 921.3 [M¹], 878.0, 848.5, 837.9, 821.4, 807.4, 792.3;

FAB-MS obsd (HR) 918.4527, calcd exact mass 918.4513

(C₆₂H₆₂N₄Si₂); l_abs423, 517, 552, 594, 649 nm.

1-{4-[2-(Triisopropylsilyl)ethynyl]benzoyl}-9-(4-methylbenzoyl)-

5-(4-methylphenyl)dipyrromethane (8)

Following a standard procedure,²⁴to a stirred solution of

monoacyl dipyrromethane 5 (1.45 g, 4.09 mmol) in dry toluene

(16 mL) at room temperature was slowly added EtMgBr (8 mL,

8 mmol, 1.0 M in THF) under Ar, and the mixture was stirred

for 5 min. To the resulting mixture was added 6 (1.31 g,

4.08 mmol). After 10 min, the same process was repeated once

[8 mL, 8 mmol, 1.0 M in THF of EtMgBr, and 1.31 g,

4.08 mmol of 6]. After stirring for 10 min, the mixture was

treated with additional EtMgBr (4 mL, 4 mmol, 1.0 M in THF)

followed by 6 (0.64 g, 2.0 mmol). After stirring the contents for

20 min, saturated aqueous NH₄Cl (50 mL) and ethyl acetate

(120 mL) were added. The organic layer was separated, dried

(Na₂SO₄), and concentrated. Puriﬁcation by column chroma-

tography [silica, neat CH₂Cl₂, then CH₂Cl₂–ethyl acetate (98 : 2,

then 98 : 3, then 90 : 10)] followed by slow precipitation from

propan-2-ol afforded a pink amorphous solid (1.69 g, 65%): mp

142–143 uC; ¹H NMR d 1.13 (s, 21H), 2.38 (s, 3H), 2.40 (s, 3H),

5.61 (s, 1H), 5.95–5.99 (m, 2H), 6.51–6.56 (m, 2H), 7.20 (d, J ~

8.1 Hz, 4H), 7.41 (d, J ~ 7.8 Hz, 2H), 7.49 (d, J ~ 8.1 Hz, 2H),

7.68 (d, J ~ 8.4 Hz, 2H), 7.71 (d, J ~ 8.4 Hz, 2H), 11.28

(s, 1H), 11.32 (s, 1H); ¹³C NMR d 11.5, 18.9, 21.4, 21.8, 44.8,

93.9, 106.6, 111.3, 111.6, 120.7, 121.1, 127.0, 128.9, 129.6,

129.8, 129.9, 131.1, 131.3, 131.8, 135.7, 137.4, 137.8, 140.9,

141.8, 142.4, 183.6, 184.5; FAB-MS obsd (HR) 638.3262, calcd

exact mass 638.3329 (C₄₂H₄₆N₂O₂Si); Anal. Calcd for

C₄₂H₄₆N₂O₂Si: C, 78.95; H, 7.26; N, 4.38. Found: C, 78.74;

H, 7.29; N, 4.26%.

5-(4-Iodophenyl)-15,20-bis(4-methylphenyl)-10-{4-[2-

(trimethylsilyl)ethynyl]phenyl}porphyrin (12)

Analogous to the preparation of 11,²⁴a sample of diacyl

dipyrromethane 9 (1.35 g, 2.31 mmol) was reduced with

NaBH₄(1.75 g, 46.3 mmol) in dry THF–methanol (10 : 1,

110 mL), affording the dicarbinol as a foam-like solid

(y2.31 mmol). The resulting dicarbinol was reacted with

dipyrromethane 2 (736 mg, 2.31 mmol). Column chromato-

graphy [silica, CH₂Cl₂–hexanes (1 : 1)] afforded a purple solid

(420 mg, 21%): ¹H NMR d 22.81 (s, 2H), 0.38 (s, 9H), 2.71 (s,

6H), 7.56 (d, J ~ 7.8 Hz, 4H), 7.87 (d, J ~ 8.1 Hz, 2H), 7.94 (d,

J ~ 8.1 Hz, 2H), 8.09 (d, J ~ 6.9 Hz, 6H), 8.16 (d, J ~ 8.1 Hz,

2H), 8.79–8.89 (m, 8H); LD-MS obsd 864.1 [M¹], 782.1, 767.2,

739.1 [M¹2 I]; FAB-MS obsd (HR) 864.2137, calcd exact

mass 864.2145 (C₅₁H₄₁IN₄Si); l_abs422, 516, 552, 593, 649 nm.

1-(4-Iodobenzoyl)-9-(4-methylbenzoyl)-5-(4-

methylphenyl)dipyrromethane (9)

5-(4-Iodophenyl)-10,15,20-tris(4-methylphenyl)porphyrin (13)

Following the procedure described for 8,²⁴monoacyl dipyrro-

methane 5 (1.77 g, 4.99 mmol) in dry toluene (20 mL) was

treated with EtMgBr (25 mL, 25 mmol, 1.0 M in THF) and

4-iodobenzoyl chloride (7) (3.33 g, 12.5 mmol) under Ar at

room temperature. Puriﬁcation by column chromatography

[silica, CH₂Cl₂–ethyl acetate (98 : 2, then 97 : 3)] followed by

slow precipitation from propan-2-ol afforded a pink amor-

phous solid (1.98 g, 68%): mp 139–141 uC; ¹H NMR d 2.38 (s,

3H), 2.40 (s, 3H), 5.61 (s, 1H), 5.95–6.00 (m, 2H), 6.52–6.60 (m,

2H), 7.20 (d, J ~ 8.1 Hz, 4H), 7.39 (d, J ~ 8.1 Hz, 2H), 7.48 (d,

J ~ 8.4 Hz, 2H), 7.68 (d, J ~ 8.1 Hz, 2H), 7.76 (d, J ~ 8.4 Hz,

2H), 11.22 (s, 1H), 11.25 (s, 1H); ¹³C NMR d 14.3, 21.4, 21.8,

22.7, 31.8, 44.9, 99.2, 111.3, 111.6, 120.7, 121.1, 128.2, 128.88,

128.91, 129.8, 130.0, 130.8, 131.26, 131.30, 135.7, 137.36,

137.42, 137.44, 137.8, 141.0, 142.1, 142.4, 183.5, 184.5;

FAB-MS obsd (HR) 584.0944, calcd exact mass 584.0961

(C₃₁H₂₅IN₂O₂); Anal. Calcd for C₃₁H₂₅IN₂O₂: C, 63.71; H,

4.31; N, 4.79. Found: C, 65.24; H, 4.57; N, 4.68%.

Analogous to the preparation of 11,²⁴a sample of diacyl

dipyrromethane 10 (220 mg, 0.47 mmol) was reduced with

NaBH₄(351 mg, 9.31 mmol) in dry THF–methanol (10 : 1,

11 mL), affording the dicarbinol as an orange oil

(y0.47 mmol). The resulting dicarbinol was reacted with

dipyrromethane 3 (258 mg, 0.47 mmol). Column chromato-

graphy [silica, hexanes–CH₂Cl₂(3 : 2)] afforded a purple solid

(77 mg, 21%): ¹H NMR d 22.80 (s, 2H), 2.71 (s, 9H), 7.56 (d,

J ~ 7.5 Hz, 6H), 7.95 (d, J ~ 8.1 Hz, 2H), 8.09 (d, J ~ 8.1 Hz,

8H), 8.80–8.97 (m, 8H); LD-MS obsd 782.8 [M¹], 656.5 [M¹–

I]; FAB-MS obsd (HR) 783.2005, calcd exact mass 783.1985

(C₄₇H₃₅IN₄); l_abs376, 421, 486, 516, 551, 593, 649 nm.

TD7

Following a standard procedure,^13,14a mixture of Eu(acac)₃?nH₂O

(319 mg) and porphyrin 14 (87 mg, 0.10 mmol) in 1,2,4-

trichlorobenzene (11 mL) was stirred at 220 uC under a slow

824

J. Mater. Chem., 2002, 12, 808–828

View Article Online

stream of Ar for 8 h. After cooling to room temperature, a

sample of (t-Bu₄Pc)₂Eu (163 mg, 0.10 mmol) was added, and

the mixture was reﬂuxed under Ar for a further 14 h. After

removal of the solvent, puriﬁcation by column chromato-

graphy (silica, CHCl₃), then SEC (THF) and ﬁnally by a silica

column (CHCl₃) afforded a dark-green solid (195 mg, 74%): ¹H

NMR d 0.84 (s, 9H), 2.03 (s, 6H), 2.2–4.1 (m, 80H), 4.99–5.22

(m, 4H), 6.81 (m, 2H), 7.13 (m, 1H), 7.35 (m, 1H), 8.4–11.0 (m,

24H), 11.28 (m, 4H), 12.4–13.8 (m, 12H); LD-MS obsd 2640.3

[M¹], 2514.1 [M¹2 I]; FAB-MS obsd (HR) 2640.84; calcd

exact mass 2640.84 (C₁₄₇H₁₃₅Eu₂IN₂₀Si); l_abs346, 418, 526,

622, 728 nm.

TD10

Following a standard procedure,^13,14a mixture of Eu(acac)₃?nH₂O

(221 mg) and porphyrin 15 (50.0 mg, 69.2 mmol) in 1,2,4-

trichlorobenzene (7.6 mL) was stirred at 220 uC under a slow

stream of Ar for 5 h. After cooling to room temperature, a

sample of (t-Bu₄Pc)₂Eu (113 mg, 69.2 mmol) was added, and the

mixture was reﬂuxed under Ar for a further 14 h. After removal

of the solvent, the residue was chromatographed (silica,

CHCl₃). Further puriﬁcation by SEC (THF), a silica column

(CHCl₃), and ﬁnally again by SEC (THF) afforded 43 mg

(25%) of a dark-green solid. A ¹H NMR spectrum was

recorded, but the complexity due to the mixture of regioisomers

prevented a complete interpretation. LD-MS obsd 2496.7

[M¹], 2482.0, 2439.9, 2371.0 [M¹2 I]; FAB-MS obsd (LR)

2498.71, calcd exact mass 2498.91 (C₁₃₇H₁₄₁Eu₂IN₂₀); l_abs345,

416, 530, 581, 627, 733 nm.

TD8

Following a standard procedure,^13,14a mixture of Eu(acac)₃?nH₂O

(136 mg) and porphyrin 11 (42.0 mg, 45.7 mmol) in 1,2,4-

trichlorobenzene (5 mL) was stirred at 220 uC under a slow

stream of Ar for 6 h to form the corresponding europium

porphyrin half-sandwich complex. After cooling to room

temperature, a sample of (t-Bu₄Pc)₂Eu (74.8 mg, 46.0 mmol)

was added, and the mixture was again reﬂuxed under Ar for a

further 5 h. After removal of the solvent, puriﬁcation by

column chromatography [silica, CH₂Cl₂] afforded a dark-green

solid, which was further puriﬁed by SEC (THF) and ﬁnally a

silica column (CH₂Cl₂) (97 mg, 79%): ¹H NMR d 0.84 (s, 9H),

1.38 (s, 3H), 1.77 (s, 21H), 1.98 (s, 3H), 2.5–3.6 (m, 80 H), 5.12–

5.36 (m, 4H), 6.81 (m, 2H), 7.22 (m, 2H), 8.6–11.0 (m, 16H),

11.26 (m, 4H), 12.6–13.6 (m, 12H); LD-MS obsd 2693.2 [M¹],

2678.5, 2663.9; FAB-MS obsd (LR) 2695.26, calcd exact mass

2695.08 (C₁₅₈H₁₅₆Eu₂N₂₀Si₂); l_abs347, 419, 527, 621, 729 nm.

TD11

The following is a reﬁned version of the standard procedure¹⁴

designed to minimize exposure of the reaction mixture to

moisture and air. A Schlenk ﬂask was equipped with a reﬂux

condenser and all essential inlets and outlets at the beginning of

the procedure. A solution of CeI[N(SiMe₃)₂]₂was prepared

in situ by reaction of CeI₃(75.0 mg, 0.144 mmol) and

LiN(SiMe₃)₂(288 mL, 0.288 mmol, 1.0 M in THF) in bis(2-

methoxyethyl) ether (1.5 mL). This ﬂask was not opened. A

deaerated solution of porphyrin 12 (25.0 mg, 28.9 mmol) in

bis(2-methoxyethyl) ether (1 mL) was added by syringe. The

mixture was reﬂuxed for 1 h, affording the metalated porphyrin

as determined by UV/Vis. Then the double-decker complex

(TTP)Eu(t-Bu₄Pc) (45.2 mg, 28.9 mmol) was added. The mix-

ture was reﬂuxed for 6 h. After cooling to room temperature,

chromatography [silica, toluene–hexanes (1 : 1), then CH₂Cl₂]

gave three bands (green). The ﬁrst and second bands were

combined and concentrated, affording a solid, which was

further puriﬁed by SEC (THF), affording a brownish-green

solid (40 mg, 54%): A ¹H NMR spectrum was recorded, but the

complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd 2557.5 [M¹], 2542.9,

2459.8, 2431.7 [M¹2 I]; FAB-MS obsd (LR) 2559.86, calcd

exact mass 2559.72 (C₁₄₇H₁₂₃CeEuIN₁₆Si); l_abs363, 421, 425,

492, 560, 609 nm.

TD8’

A sample of TD8 (54.5 mg, 20.2 mmol) in a mixture of CHCl₃–

THF–methanol (1 : 5 : 2, 8 mL) was treated with powdered

K₂CO₃(71.8 mg, 0.52 mmol) and stirred under Ar at room

temperature for 5 h. The progress of the reaction was

monitored by LD-MS. Then the reaction mixture was ﬁltered.

The ﬁltrate was washed with 10% aqueous NaHCO₃(100 mL)

and water (150 mL). The organic layer was dried (Na₂SO₄),

concentrated and chromatographed (silica, CH₂Cl₂), affording

a dark-green solid (47.8 mg, 90%): ¹H NMR d 1.36 (s, 3H), 1.75

(s, 21H), 1.90 (s, 3H), 2.5–3.6 (m, 80H), 3.86 (br s, 1H), 5.10–

5.28 (m, 4H), 6.79 (br s, 2H), 7.16 (br s, 2H), 8.6–11.0 (m, 16H),

11.23 (m, 4H), 12.4–13.6 (m, 12H); LD-MS obsd 2627.7 [M¹],

2613.1, 2518.7, 2461.2; FAB-MS obsd (LR) 2623.20, calcd

exact mass 2623.04 (C₁₅₅H₁₄₈Eu₂N₂₀Si); l_abs347, 418, 527, 621,

729 nm.

AcS-TD3

A 25 mL Schlenk ﬂask containing samples of TD3 (26.1 mg,

10.2 mmol), 1-[4-(S-acetylthiomethyl)phenyl]acetylene (16)

(39.2 mg, 205 mmol), Pd(PPh₃)₂Cl₂(2.4 mg, 3.4 mmol), and

CuI (0.3 mg, 1.6 mmol) was evacuated and purged with Ar three

times. Then deaerated THF (2.5 mL) and deaerated DIEA

(0.4 mL) were added, and the reaction mixture was stirred

under Ar at 35 uC for 7 h. After removal of the solvent, the

residue was redissolved in toluene and chromatographed on

silica (toluene). Further puriﬁcation by SEC (THF) and ﬁnally

a silica column (toluene) afforded an olive-green solid (17 mg,

64%). A ¹H NMR spectrum was recorded, but the complexity

due to the mixture of regioisomers prevented a complete

interpretation. LD-MS obsd 2631.5 [M¹], 2553.0 [M¹2 SAc];

FAB-MS obsd (HR) 2621.82, calcd exact mass 2621.85

(C₁₅₈H₁₃₂CeEuN₁₆OSSi); l_abs363, 421, 427, 492, 560, 609 nm.

TD9

Following a standard procedure,^13,14a mixture of Eu(acac)₃?nH₂O

(117 mg) and porphyrin 13 (28.7 mg, 36.7 mmol) in 1,2,4-

trichlorobenzene (4 mL) was stirred at 220 uC under a slow

stream of Ar for 6 h. After cooling to room temperature, a

sample of (t-Bu₄Pc)₂Eu (60.1 mg, 36.7 mmol) was added, and

the mixture was again reﬂuxed under Ar for a further 5 h. After

removal of the solvent, puriﬁcation by column chromato-

graphy [silica, CHCl₃(containing 0, 1 and 5% ethyl acetate,

respectively)] afforded a dark-green solid, which was further

puriﬁed by SEC (THF) and ﬁnally a silica column (CHCl₃)

(58 mg, 62%): ¹H NMR d 1.35 (s, 3H), 1.94 (s, 6H), 2.5–3.6 (m,

80H), 5.00–5.17 (m, 4H), 6.80 (br s, 3H), 7.34 (m, 1H), 8.6–9.3

(m, 4H), 9.40 (br s, 3H), 9.7–11.0 (m, 9H), 11.25 (m, 4H), 12.6–

13.5 (m, 12H); LD-MS obsd 2556.1 [M¹], 2429.5 [M¹2 I];

FAB-MS obsd (HR) 2558.75, calcd exact mass 2558.82

(C₁₄₃H₁₂₉Eu₂IN₂₀); l_abs346, 417, 526, 620, 727 nm.

AcS-TD7

A 50 mL Schlenk ﬂask containing samples of TD7 (174 mg,

65.9 mmol), 16 (126 mg, 659 mmol), Pd(PPh₃)₂Cl₂(14.1 mg,

19.8 mmol), and CuI (1.9 mg, 9.9 mmol) was evacuated and

purged with Ar three times. THF (25 mL) and DIEA (3.4 mL)

were added, and the reaction mixture was stirred under Ar at

35 uC for 6 h. The progress of the reaction was monitored by

J. Mater. Chem., 2002, 12, 808–828

825

View Article Online

LD-MS. After removal of the solvent, the residue was puriﬁed

by column chromatography [silica, CHCl₃(containing 0, then

2, then 5% ethyl acetate)] to afford a dark-green solid, which

was further puriﬁed by SEC (toluene) and ﬁnally preparative

water, dried (Na₂SO₄), concentrated, and puriﬁed by pre-

parative TLC (silica, toluene), affording an olive-green solid

(5.1 mg, 66%). A ¹H NMR spectrum was recorded, but the

complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd 2549.5 [M¹], 2475.1

[M¹2 SAc]; FAB-MS obsd (HR) 2549.77, calcd exact mass

2549.81 (C₁₅₅H₁₂₄CeEuN₁₆OS); l_abs363, 421, 492, 560, 609,

676 nm.

1

TLC [silica, toluene–diethyl ether (20 : 1)] (105 mg, 59%): H

NMR d 0.80 (s, 9H), 1.84 (s, 6H), 2.0–3.6 (m, 80H), 2.62 (s,

3H), 4.45 (s, 2H), 5.08–5.21 (m, 4H), 6.78 (br s, 2H), 7.05–7.16

(m, 2H), 7.71 (d, J ~ 8.1 Hz, 2H), 8.08 (d, J ~ 7.5 Hz, 2H),

8.5–11.0 (m, 16H), 11.22 (m, 4H), 12.5–13.7 (m, 12H); LD-MS

obsd 2707.0 [M¹], 2635.4 [M¹2 SAc]; FAB-MS obsd (LR)

2702.98, calcd exact mass 2702.97 (C₁₅₈H₁₄₄Eu₂N₂₀OSSi); l_abs

346, 419, 527, 621, 728 nm.

Dyad1

A 25 mL Schlenk ﬂask containing samples of TD10 (10.0 mg,

4.00 mmol), AcS-TD7’ (10.5 mg, 4.12 mmol), Pd₂(dba)₃(0.61 mg,

0.67 mmol), and P(o-tol)₃(1.63 mg, 5.36 mmol) was evacuated

and purged with Ar three times. Then deaerated toluene

(3.5 mL) and deaerated DIEA (0.7 mL) were added, and the

reaction mixture was stirred under Ar at 35 uC. The progress of

the reaction was monitored by analytical SEC. Because the

reaction had stopped, further catalyst [Pd₂(dba)₃, 0.61 mg

(0.67 mmol); P(o-tol)₃, 1.22 mg (4.02 mmol)] was added after 22 h

and again after 31 h. After a total reaction time of 48 h, the

solvent was removed. The residue was redissolved in toluene–

diethyl ether (100 : 1) and loaded onto a silica column packed

with the same solvent. Three bands were collected (ﬁrst band:

blue; second band: green). The desired product was collected in

a third band (green), employing toluene–diethyl ether (20 : 1)

as the eluent. Further puriﬁcation by SEC (THF) gave ﬁve

bands, from which the third band (green) contained the

product. Impurities collected from the SEC column were ﬁnally

removed by a silica column [toluene–diethyl ether (100 : 1,

then 50 : 1)], affording a dark-green solid (5.0 mg, 25%). A ¹H

NMR spectrum was recorded, but the complexity due to the

mixture of regioisomers prevented a complete interpretation.

LD-MS obsd (POPOP) 5001.2 [M¹], 4985.5, 4972.5, 4957.4,

4938.5, 4924.4 [M¹2 SAc]; FAB-MS obsd (LR) 5002.01, calcd

exact mass 5001.93 (C₂₉₂H₂₇₆Eu₄N₄₀OS); l_abs350, 421, 427,

493, 530, 625 nm.

AcS-TD11

A 25 mL Schlenk ﬂask containing samples of TD11 (15.0 mg,

5.86 mmol), 16 (38.1 mg, 200 mmol), Pd(PPh₃)₂Cl₂(1.4 mg,

2.0 mmol), and CuI (0.2 mg, 1.1 mmol) was evacuated and

purged with Ar three times. Then deaerated THF (1.5 mL) and

deaerated TEA (0.3 mL) were added, and the reaction mixture

was stirred under Ar at 35 uC for 2 h. After removal of the

solvent, the residue was redissolved in CH₂Cl₂–hexanes (1 : 2)

and chromatographed on silica [CH₂Cl₂–hexanes (1 : 2)],

affording an olive-green solid (8.8 mg, 57%). A ¹H NMR

spectrum was recorded, but the complexity due to the mixture

of regioisomers prevented a complete interpretation. LD-MS

obsd 2622.4 [M¹], 2546.0 [M¹2 SAc]; FAB-MS obsd (HR)

2621.82, calcd exact mass 2621.85 (C₁₅₈H₁₃₂CeEuN₁₆OSSi);

l_abs363, 421, 426, 492, 560, 609, 676 nm.

AcS-TD3’

To a stirred solution of AcS-TD3 (17 mg, 6.5 mmol) in dry

THF (2 mL) at 0 uC was added (n-Bu)₄NF (7.1 mL, 1.0 M in

THF). After stirring for 5 min at 0 uC, the reaction was

quenched by addition of water (1 mL). The mixture was

extracted with CH₂Cl₂. The organic layer was washed with

water, dried (Na₂SO₄), concentrated, and puriﬁed by pre-

parative TLC (silica, toluene), affording an olive-green solid

(16 mg, 97%). A ¹H NMR spectrum was recorded, but the

complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd 2548.9 [M¹], 2473.6

[M¹2 SAc]; FAB-MS obsd (HR) 2549.82, calcd exact mass

2549.81 (C₁₅₅H₁₂₄CeEuN₁₆OS); l_abs363, 421, 427, 492, 560,

609 nm.

Dyad2

A 25 mL Schlenk ﬂask containing samples of TD10 (10.0 mg,

3.91 mmol), AcS-TD3’ (10.2 mg, 4.00 mmol), Pd₂(dba)₃(0.55 mg,

0.60 mmol), and P(o-tol)₃(1.34 mg, 4.40 mmol) was evacuated

and purged with Ar three times. Then deaerated toluene

(3.5 mL) and deaerated DIEA (0.7 mL) were added, and the

reaction mixture was stirred under Ar at 35 uC. The progress of

the reaction was monitored by analytical SEC. Because the

reaction had stopped, further catalyst [Pd₂(dba)₃0.55 mg

(0.60 mmol) and P(o-tol)₃1.01 mg (3.32 mmol)] was added after

7 h and again after 18 h. After a total reaction time of 24 h, the

solvent was removed. The residue was redissolved in toluene–

diethyl ether (100 : 1) and loaded onto a silica column packed

with the same solvent. The desired product was collected in the

second band (green), employing toluene–diethyl ether (50 : 1)

as the eluent. Further puriﬁcation by SEC (THF) gave three

bands, from which the second band (green) contained the

product. Impurities collected from the SEC column were ﬁnally

removed by a silica column [toluene–diethyl ether (50 : 1)],

affording a dark-green solid (5.0 mg, 25%). A ¹H NMR

spectrum was recorded, but the complexity due to the mixture

of regioisomers prevented a complete interpretation. LD-MS

obsd (POPOP) 4922.1 [M¹], 4876.2, 4848.2 [M¹2 SAc];

FAB-MS obsd (LR) 4921.77, calcd exact mass 4920.81

(C₂₉₂H₂₆₄CeEu₃N₃₆OS); l_abs352, 420, 493, 523, 616, 676,

726 nm.

AcS-TD7’

To a stirred solution of AcS-TD7 (90 mg, 33.3 mmol) in THF

(5 mL) at 0 uC was added (n-Bu)₄NF (37 mL, 1.0 M in THF).

After stirring for 5 min, the reaction was quenched by addition

of water (3 mL), and the solution was extracted with CHCl₃,

washed with water, and dried (Na₂SO₄). After removal of the

solvent, puriﬁcation by preparative TLC [silica, toluene–

diethyl ether (30 : 1)] afforded a dark-green solid (78 mg,

1

89%): H NMR d 1.88 (s, 6H), 2.0–3.6 (m, 80H), 2.63 (s, 3H),

3.85 (s, 1H), 4.46 (s, 2H), 5.09–5.21 (m, 4H), 6.79 (br s, 2H),

7.16 (br s, 2H), 7.71 (d, J ~ 7.2 Hz, 2H), 8.09 (d, J ~ 7.5 Hz,

2H), 8.5–11.0 (m, 16H), 11.25 (m, 4H), 12.5–13.7 (m, 12H);

LD-MS obsd 2630.8 [M¹], 2555.2 [M¹2 SAc], 1618.1

[(t-Bu₄Pc)₂Eu¹]; FAB-MS obsd (LR) 2631.21, calcd exact

mass 2630.94 (C₁₅₅H₁₃₆Eu₂N₂₀OS); l_abs346, 419, 526, 620,

728 nm.

AcS-TD11’

To a stirred solution of AcS-TD11 (8.0 mg, 3.0 mmol) in dry

THF (1 mL) at 0 uC was added (n-Bu)₄NF (3.5 mL, 1.0 M in

THF). After stirring for 6 min at 0 uC, the reaction was

quenched by addition of water (1 mL). The mixture was

extracted with CH₂Cl₂. The organic layer was washed with

Dyad3

A 25 mL Schlenk ﬂask containing samples of TD9 (6.0 mg,

2.4 mmol), AcS-TD3’ (6.0 mg, 2.4 mmol), Pd₂(dba)₃(0.34 mg,

0.37 mmol), and P(o-tol)₃(0.84, 2.79 mmol) was evacuated and

826

J. Mater. Chem., 2002, 12, 808–828

View Article Online

purged with Ar three times. Then deaerated toluene (2 mL) and

deaerated DIEA (0.4 mL) were added, and the reaction mixture

was stirred under Ar at 35 uC. The progress of the reaction was

monitored by analytical SEC. Because the reaction had

stopped after 20 h, the same amount of catalyst was added

again, and the mixture was stirred for a further 4 h. After

removal of the solvent, the residue was redissolved in toluene

and loaded onto a silica column packed with the same solvent.

Elution with toluene gave two bands (ﬁrst band: blue; second

band: olive-green, containing starting triple decker). The

desired product was collected in a third band, employing

toluene–ethyl acetate (20 : 1) as the eluent. Further puriﬁca-

tion by SEC (THF) gave four bands. The ﬁrst two bands

(green) contained higher molecular weight material. The third

band (green) contained the product and a fourth band

contained triple-decker starting material. Impurities collected

from the SEC column were ﬁnally removed by a silica column

[toluene, then toluene–diethyl ether (50 : 1)], affording a green

5054.85 (C₃₀₂H₂₇₀CeEu₃N₃₆Si₂); l_abs353, 421, 492, 616,

729 nm.

Dyad5’

To a solution of TMS/TIPS-Dyad5 (12.0 mg, 2.37 mmol) in

THF (1 mL) at room temperature was added (n-Bu)₄NF

(5.7 mL, 1.0 M in THF). After stirring for 1 h, the reaction was

quenched by addition of water, and the mixture was extracted

with CH₂Cl₂, washed with water, and dried (Na₂SO₄). After

removal of the solvent, puriﬁcation by column chromato-

graphy (silica, toluene) afforded a dark-green solid (11 mg,

96%). A ¹H NMR spectrum was recorded, but the complexity

due to the mixture of regioisomers prevented a complete

interpretation. LD-MS obsd (POPOP) 4825.8 [M¹], calcd exact

mass 4826.67 (C₂₉₀H₂₄₂CeEu₃N₃₆); l_abs353, 421, 493, 616,

729 nm.

1

Dyad5

solid (6.0 mg, 51%). A H NMR spectrum was recorded, but

the complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd (POPOP) 4979.4 [M¹],

4965.1, 4902.4 [M¹2 SAc]; FAB-MS obsd (LR) 4982.13, calcd

exact mass 4980.72 (C₂₉₈H₂₅₂CeEu₃N₃₆OS); l_abs346, 420, 528,

625, 730 nm.

A 25 mL Schlenk ﬂask containing samples of Dyad5’ (11.0 mg,

2.28 mmol), 1-(S-acetylthiomethyl)-4-iodobenzene (17) (13.3 mg,

45.5 mmol), Pd₂(dba)₃(1.0 mg, 1.1 mmol), and P(o-tol)₃(2.8 mg,

9.2 mmol) was evacuated and purged with Ar three times. Then

deaerated toluene (3 mL) and deaerated TEA (0.6 mL) were

added, and the reaction mixture was stirred at 35 uC. The

progress of the reaction was monitored by analytical SEC and

by LD-MS. Further batches of Pd₂(dba)₃(1.0 mg, 1.1 mmol)

and P(o-tol)₃(2.8 mg, 9.2 mmol) were added after 1 h 30 min

and again after 4 h. After a total reaction time of 6 h, the

solvent was removed. Puriﬁcation by preparative TLC [silica,

toluene–diethyl ether, (40 : 1)] afforded the desired product

(5.0 mg, 42%). A ¹H NMR spectrum was recorded, but the

complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd (POPOP) 5155.7 [M¹],

5080.8 [M¹2 SAc], 4980.6, 4267.7 [M¹2 (t-Bu₄Pc)Eu]; calcd

exact mass 5154.73 (C₃₀₈H₂₅₈CeEu₃N₃₆O₂S₂); l_abs348, 421,

493, 525, 617, 729 nm.

Dyad4

A 25 mL Schlenk ﬂask containing samples of TD9 (5.0 mg,

2.0 mmol), AcS-TD11’ (4.0 mg, 1.6 mmol), Pd₂(dba)₃(0.42 mg,

0.46 mmol), and P(o-tol)₃(1.17 mg, 3.84 mmol) was evacuated

and purged with Ar three times. Then deaerated toluene

(1.0 mL) and deaerated TEA (250 mL) were added, and the

reaction mixture was stirred at 35 uC. The progress of the

reaction was monitored by analytical SEC and by LD-

MS. Further batches of Pd₂(dba)₃(0.42 mg, 0.46 mmol) and

P(o-tol)₃(1.17 mg, 3.84 mmol) were added after 2 h and again

after 6 h. After a total reaction time of 8 h 30 min, the solvent

was removed. The residue was dissolved in CH₂Cl₂and loaded

onto a silica column packed with the same solvent. Elution with

CH₂Cl₂gave two bands. The desired product was collected in

the second band. Further puriﬁcation by SEC (THF) and

ﬁnally a silica column (CH₂Cl₂) afforded the desired product

(3.0 mg, 38%). A ¹H NMR spectrum was recorded, but the

complexity due to the mixture of regioisomers prevented a

complete interpretation. LD-MS obsd (POPOP) 4975.8 [M¹],

4085.8 [M¹2 (t-Bu₄Pc)Eu]; FAB-MS obsd 4981.76 (LR),

calcd exact mass 4980.72 (C₂₉₈H₂₅₂CeEu₃N₃₆OS); l_abs352,

420, 427, 493, 523, 615, 727 nm.

Electrochemical studies

The solution electrochemical studies were performed using

techniques and instrumentation previously described.⁴⁵For all

triple deckers and dyads, the solvent was dried distilled CH₂Cl₂

and 0.1 M Bu₄NPF₆(Aldrich, recrystallized three times

from methanol and dried under vacuum at 110 uC) served as

supporting electrolyte. The potentials were measured vs. Ag/

Ag¹; E_1/2(FeCp₂/FeCp₂¹) ~ 0.19 V. The scan rate was

0.1 V s²¹. All CH₂Cl₂used in the following procedure also

was dried and distilled.

Cyclic voltammograms of the SAMs were recorded with an

Ensman Instruments 400 potentiostat. These electrochemical

studies utilized a 75 mm Au microband working electrode that

was formed on a borosilicate glass microscope slide (Fisher,

Fair Lawn, NJ) by vapor deposition of 99.99% gold (1 Oz.

Canadian Maple Leaf, obtained locally) in a CHA SE-600

4-pocket E-beam evaporator (CHA Industries, Menlo Park,

CA). The microscope slides are ﬁrst cleaned by boiling in

piranha solution (30% H₂O₂–10% solution, 70% H₂SO₄) for

20 min, rinsed with deionized water and dried in a vacuum oven

TMS/TIPS-Dyad5

A 25 mL Schlenk ﬂask containing samples of TD11 (12.1 mg,

4.73 mmol), TD8’ (12.2 mg, 4.65 mmol), Pd₂(dba)₃(1.0 mg,

1.1 mmol), and P(o-tol)₃(2.8 mg, 9.2 mmol) was evacuated and

purged with Ar three times. Then deaerated toluene (2.5 mL)

and deaerated TEA (0.6 mL) were added, and the reaction

mixture was stirred at 35 uC. The progress of the reaction was

monitored by analytical SEC and by LD-MS. Further batches

of Pd₂(dba)₃(1.0 mg, 1.1 mmol) and P(o-tol)₃(2.8 mg, 9.2 mmol)

were added after 2 h and again after 3 h 40 min. After a total

reaction time of 5 h 30 min, the solvent was removed. The

residue was dissolved in toluene and loaded onto a silica

column packed with the same solvent. Elution with toluene

gave two bands. The desired product was collected in the

second band. Further puriﬁcation by SEC (THF) gave the

desired product (14 mg, 59%). A ¹H NMR spectrum was

recorded, but the complexity due to the mixture of regioisomers

prevented a complete interpretation. LD-MS obsd (POPOP)

5050.5 [M¹], 4165.3 [M¹2 (t-Bu₄Pc)Eu]; calcd exact mass

˚

at 100 uC. A shadow mask is used to allow deposition of 2000 A

˚

of Au (utilizing a 100 A Cr underlayer to promote adhesion) to

form the electrode (four Au microbands per slide, each 75 mm

wide) with a contact pad at the edge of the glass slide.

The electrochemical cell was deﬁned by placing a patterned

2

mm thick sheet of polymerized polydimethylsiloxane

(PDMS) to frame a y12 mm²circular area over the Au

microband electrode. PDMS adheres well to glass surfaces and

prevents leakage of solution, thereby deﬁning the area of

electrode that will be exposed to electrolyte solution. This

deﬁnes a y3000 mm²Au microband electrode. The PDMS well

J. Mater. Chem., 2002, 12, 808–828

827

View Article Online

was ﬁlled with a 2 mg mL²¹solution of the dyad in CH₂Cl₂

containing 1 drop of ethanol and sonicated for 10 min. After

sonication, the solution along with the well was removed from

the slide. The slide was copiously rinsed with CH₂Cl₂and dried

in a stream of N₂. A fresh PDMS well was placed around the

working electrode, ﬁlled with electrolyte (CH₂Cl₂containing

1.0 M Bu₄NPF₆), and the Ag counter and Pt reference

electrodes were inserted into the solution. The potentials were

measured vs. Ag/Ag¹; E_1/2(FeCp₂/FeCp₂¹) ~ 0.19 V. The scan

K. M. Smith and R. Guilard, Academic Press, San Diego, CA,

2000, Vol. 1, pp. 45–118.

21 The meso-substituted metalloporphyrins have the following

symmetry point groups: A₄porphyrin (D_4h); A₃B porphyrin

(C_2v); cis-A₂B₂porphyrin (C_2v); trans-A₂B₂porphyrin (D_2h); trans-

A₂BC porphyrin (C_2v); cis-A₂BC porphyrin (C_s); ABCD

porphyrin (C_s). Only the latter two types of substituted porphyrins

have no mirror plane perpendicular to the plane of the

macrocycle.

22 This analysis assumes that (a) the Por and/or Pc components that

complete the triple decker themselves have mirror planes

perpendicular to the plane of the macrocycle, and (b) rotation

about the cylindrical axis of the triple decker is unhindered.

23 This analysis describes the minimum number of structural isomers

that can result for each triple decker and for each type of dyad. In

practice, each triple decker described herein incorporates one or

two tetra-tert-butylphthalocyanine units; accordingly, each triple

decker exists as a mixture of diastereomers. The presence of the

diastereomers stems from the fact that the tetra-tert-butylphtha-

locyanine employed in these reactions is comprised of four

regioisomers. Thus, those triple deckers that incorporate a cis-

A₂BC porphyrin afford a mixture of diastereomers rather than a

mere pair of enantiomers due to the presence of one or two tetra-

tert-butylphthalocyanine units in the triple decker. It follows that

each dyad that incorporates one or more tetra-tert-butylphthalo-

cyanine ligands also exists as a mixture of diastereomers. However,

the presence of syn versus anti isomers is expected to pose more

distinct spatial requirements upon surface binding of a horizontal

dyad than the existence of diastereomers due to the presence of

multiple tetra-tert-butylphthalocyanine units.

rate was 100 V s²¹

.

The molecular area of each dyad SAM was determined by

integrating the charge under an oxidation wave that corre-

sponded to a one-electron event. The electrochemical surface

area was evaluated by measuring the peak current resulting

from the oxidation of 1 mM ferrocene and applying the

Randle–Sevcik equation.⁴⁶In this measurement, the SAM was

ﬁrst stripped from the surface by scanning to high potential; the

solution and electrolyte were removed from the cell and the

cell was rinsed. A solution containing the ferrocene was then

added and a voltammogram recorded.

Acknowledgements

This work was supported by the DARPA Moletronics

Programs, administered by the ONR (MDA-972-01-C-0072),

and by ZettaCore, Inc. 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 NSF.

24 P. D. Rao, S. Dhanalekshmi, B. J. Littler and J. S. Lindsey, J. Org.

Chem., 2000, 65, 7323.

25 B. J. Littler, M. A. Miller, C.-H. Hung, R. W. Wagner,

D. F. O’Shea, P. D. Boyle and J. S. Lindsey, J. Org. Chem.,

1999, 64, 1391.

26 W.-S. Cho, H.-J. Kim, B. J. Littler, M. A. Miller, C.-H. Lee and

J. S. Lindsey, J. Org. Chem., 1999, 64, 7890.

References

27 P. D. Rao, B. J. Littler, G. R. Geier III and J. S. Lindsey, J. Org.

Chem., 2000, 65, 1084.

28 D. Gryko and J. S. Lindsey, J. Org. Chem., 2000, 65, 2249.

29 D. Battisti, L. Tomilova and R. Aroca, Chem. Mater., 1992, 4,

1323.

30 The relative distribution of the four isomers of each tetra-tert-

butylphthalocyanine ligand in the triple deckers is not known.

However, the complexity of the ¹H NMR spectra is consistent with

the presence of tetra-tert-butylphthalocyanine isomers in the triple

deckers.

31 R. W. Wagner, Y. Ciringh, C. Clausen and J. S. Lindsey, Chem.

Mater., 1999, 11, 2974.

32 Analytical SEC data are uncorrected.

1

K. M. Roth, N. Dontha, R. B. Dabke, D. T. Gryko, C. Clausen,

J. S. Lindsey, D. F. Bocian and W. G. Kuhr, J. Vac. Sci. Technol.

B, 2000, 18, 2359.

2

3

4

5

6

7

8

D. T. Gryko, C. Clausen, K. M. Roth, N. Dontha, D. F. Bocian,

W. G. Kuhr and J. S. Lindsey, J. Org. Chem., 2000, 65, 7345.

C. Clausen, D. T. Gryko, R. B. Dabke, N. Dontha, D. F. Bocian,

W. G. Kuhr and J. S. Lindsey, J. Org. Chem., 2000, 65, 7363.

C. Clausen, D. T. Gryko, A. A. Yasseri, J. R. Diers, D. F. Bocian,

W. G. Kuhr and J. S. Lindsey, J. Org. Chem., 2000, 65, 7371.

D. T. Gryko, F. Zhao, A. A. Yasseri, K. M. Roth, D. F. Bocian,

W. G. Kuhr and J. S. Lindsey, J. Org. Chem., 2000, 65, 7356.

J. Li, D. Gryko, R. B. Dabke, J. R. Diers, D. F. Bocian,

W. G. Kuhr and J. S. Lindsey, J. Org. Chem., 2000, 65, 7379.

D. Gryko, J. Li, J. R. Diers, K. M. Roth, D. F. Bocian, W. G. Kuhr

and J. S. Lindsey, J. Mater. Chem., 2001, 11, 1162.

33 R. P. Hsung, J. R. Babcock, C. E. D. Chidsey and L. R. Sita,

Tetrahedron Lett., 1995, 26, 4525.

34 A. V. Rukavishnikov, A. Phadke, M. D. Lee, D. H. LaMunyon,

P. A. Petukhov and J. F. W. Keana, Tetrahedron Lett., 1999, 40,

6353.

J. W. Buchler and D. K. P. Ng, in The Porphyrin Handbook, eds.

K. M. Kadish, K. M. Smith and R. Guilard, Academic Press, San

Diego, CA, 2000, Vol. 3, pp. 245–294.

35 D. K. P. Ng and J. Jiang, Chem. Soc. Rev., 1997, 26, 433.

36 J. Jiang, R. L. C. Lau, T. W. D. Chan, T. C. W. Mak and

D. K. P. Ng, Inorg. Chim. Acta, 1997, 255, 59.

37 M. Moussavi, A. De Cian, J. Fischer and R. Weiss, Inorg. Chem.,

1986, 25, 2107.

9

J. Jiang, W. Liu, W.-F. Law and D. K. P. Ng, Inorg. Chim. Acta,

1998, 268, 49.

10 M. Ruben, E. Breuning, J.-P. Gisselbrecht and J.-M. Lehn, Angew.

Chem., Int. Ed., 2000, 39, 4139.

11 C. B. Gorman, Adv. Mater., 1997, 9, 1117.

38 D. Chabach, M. Lachkar, A. De Cian, J. Fischer and R. Weiss,

New J. Chem., 1992, 16, 431.

39 T.-H. Tran-Thi, T. A. Mattioli, D. Chabach, A. De Cian and

R. Weiss, J. Phys. Chem., 1994, 98, 8279.

40 W. Liu, J. Jiang, N. Pan and D. P. Arnold, Inorg. Chim. Acta,

2000, 310, 140.

41 J. W. Buchler, A. De Cian, J. Fischer, M. Kihn-Botulinski,

H. Paulus and R. Weiss, J. Am. Chem. Soc., 1986, 108, 3652.

42 S. E. Creager and G. K. Rowe, J. Electroanal. Chem., 1994, 370,

203.

43 (a) D. Fenyo, B. T. Chait, T. E. Johnson and J. S. Lindsey,

J. Porphyrins Phthalocyanines, 1997, 1, 93; (b) N. Srinivasan,

C. A. Haney, J. S. Lindsey, W. Zhang and B. T. Chait,

J. Porphyrins Phthalocyanines, 1999, 3, 283.

44 R. W. Wagner, T. E. Johnson and J. S. Lindsey, J. Am. Chem.

Soc., 1996, 118, 11166.

12 In the electrical engineering literature, a ‘multistate counter’ of the

type we describe would more appropriately be referred to as an

integrator owing to the different potential that accompanies each

distinct state. By contrast, in a true counter, successive digits are

accessed by application of the same potential. We have elected to

use the term ‘multistate counter’ herein owing to the simple

correspondence of distinct oxidation states and counting level.

13 D. Chabach, A. De Cian, J. Fischer, R. Weiss and M. E. M. Bibout,

Angew. Chem., Int. Ed. Engl., 1996, 35, 898.

14 T. Gross, F. Chevalier and J. S. Lindsey, Inorg. Chem., 2001, 40,

4762.

15 The term Por or Pc refers generically to the dianion of a free base

porphyrin or free base phthalocyanine, respectively.

16 Each europium double decker is assumed to be a radical species.

17 C.-P. Wong, R. F. Venteicher and W. D. Horrocks Jr., J. Am.

Chem. Soc., 1974, 96, 7149.

45 J. Seth, V. Palaniappan, R. W. Wagner, T. E. Johnson,

J. S. Lindsey and D. F. Bocian, J. Am. Chem. Soc., 1996, 118,

11194.

46 A. J. Bard and L. R. Faulkner, Electrochemical Methods:

Fundamentals and Applications, Wiley, New York, 1980.

18 M. Sommerauer, C. Rager and M. Hanack, J. Am. Chem. Soc.,

1996, 118, 10085.

19 J. K. Duchowski and D. F. Bocian, J. Am. Chem. Soc., 1990, 112,

8807.

20 J. S. Lindsey, in The Porphyrin Handbook, eds. K. M. Kadish,

828

J. Mater. Chem., 2002, 12, 808–828

Article Doi

Design, synthesis, and characterization of prototypical multistate counters in three distinct architectures

DOI: 10.1039/b108520d

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b108520d

Full text of DOI:10.1039/b108520d

Products guided by the article

R&D Labs maybe for 441287-73-6

Relevant to this article

Hot Product