Tuning reductive and oxidative photoinduced electron transfer in amide-linked anthraquinone-porphyrin-ferrocene architectures

FULL PAPER

DOI:10.1002/ejic.201400118

Tuning Reductive and Oxidative Photoinduced Electron

Transfer in Amide-Linked Anthraquinone–Porphyrin–

Ferrocene Architectures

Jascha Melomedov,^[a]Julian Robert Ochsmann,^[b]Michael Meister,^[b]

Frédéric Laquai,^[b]and Katja Heinze*^[a]

Keywords: Electron transfer / Density functional calculations / Substituent effects / Porphyrinoids / Quinones

Porphyrin amino acids 3a–3h with meso substituents Ar of

tunable electron-donating power (Ar = 4-C₆H₄OnBu, 4-

C₆H₄OMe, 4-C₆H₄Me, Mes, C₆H₅, 4-C₆H₄F, 4-C₆H₄CF₃,

C₆F₅) have been linked at the N terminus to anthraquinone

Q as electron acceptor through amide bonds to give Q-P^Ar

dyads 4a–4h. These were conjugated to ferrocene Fc at the

C terminus as electron donor to give the acceptor-chromo-

ferrocene to give the tetrad Q-P^Ar1-P^Ar2-Fc 9. Almost all the

conjugates show strongly reduced fluorescence quantum

yields and excited-state lifetimes, which has been interpre-

ted as photoinduced electron transfer (PET) either from the

excited porphyrin to the quinone (oxidative PET) or from the

ferrocene to the excited porphyrin (reductive PET). Electro-

chemical data, absorption spectroscopy, steady-state emis-

phore-donor Q-P^Ar-Fc triads 6a–6h. To further modify the en- sion, time-resolved fluorescence, transient absorption pump-

ergies of the electronically excited and charge-separated

states, the triads 6a–6h were metallated with zinc(II) to give

the corresponding Q-(Zn)P^Ar-Fc triads Zn-6a–Zn-6h. The

Q-P^Ar1dyad (Ar¹= C₆H₅) was further extended with a second

porphyrin P^Ar2(Ar²= 4-C₆H₄Me) as well as appended to a

probe spectroscopy as well as DFT calculations have been

used to elaborate the preferred PET pathway (reductive vs.

oxidative PET) in these architectures with systematically var-

ied electron-donating substituents at the central chromo-

phore.

porphyrin^[3](Q-P) and porphyrin–ferrocene (P-Fc) sys-

tems^[5e–5i,7]have been investigated with respect to PET,

proving very useful for fundamental PET research and po-

tential applications. Furthermore, several other metal com-

plexes have been attached to porphyrins as electron-ac-

cepting units.^[9]Variation of Q has allowed the effect of exo-

thermicity on the rates of electron transfer to be studied in

considerable detail^[3c,3e]and variation of Fc substituents has

shown that energy- and electron-transfer pathways depend

on the oxidation potential of the ferrocene.^[10]Elaborated

C₆₀-P_n-Fc architectures feature long-lived charge-separated

states.^[5e–5h]In C₆₀-(Zn)P_n-Fc conjugates with C₆₀as the

electron acceptor, the first PET typically is the thermody-

namically favourable oxidative quenching of the excited

porphyrin by the fullerene followed by a thermal shift of

charge from the ferrocene to the central porphyrin.^[5e–5h]

The effect of the number of porphyrins sandwiched between

C₆₀and ferrocene has also been thoroughly investig-

ated.^[5e–5h]In systems containing both Fc and Q, the prefer-

ence of the initial photoinduced electron-transfer step is

ambiguous as the free energies for both steps are of a sim-

ilar magnitude.

Introduction

Photoinduced electron transfer (PET) is a key step in

natural photosynthesis.^[1]For a deeper understanding of

photosynthetic processes such as sunlight energy collection,

light-energy storage, light-to-chemical energy conversion,

electron-transport pathways and finally the efficient utilis-

ation of sunlight, a number of elegant functional artificial

reaction centres have been designed and intensively investi-

gated to gain an increasingly clear picture of the important

processes.^[2–7]The rational design of reaction centres to

mimic the multistep downhill electron transfer is theoretic-

ally based on the Marcus theory of electron transfer

(ET)^[8]and experimentally relies on the availability of suit-

able building blocks that can be assembled at will. Artificial

reaction centres consist of non-covalently or covalently

linked donor–acceptor units with quinones (natural ac-

ceptors),^[3]porphyrins^[4]or fullerenes^[5]acting as electron

acceptors and porphyrins,^[6]ferrocene^[5e–5i,7]or carotenoid

polyenes^[6]acting as electron donors. A variety of quinone–

[a] Institute of Inorganic and Analytical Chemistry, Johannes

Gutenberg University of Mainz,

Duesbergweg 10–14, 55128 Mainz, Germany

E-mail: katja.heinze@uni-mainz.de

To the best of our knowledge, systems combining the

electron-accepting features of quinones, the light-absorbing

properties of porphyrin chromophores and the electron-do-

nating properties of ferrocene derivatives have not yet been

studied with respect to either the preferred PET pathways

http://ww w .ak-heinze.chemie.uni-mainz.de/

[b] Max Planck Institute for Polymer Research,

Ackermannweg 10, 55128 Mainz, Germany

Supporting information for this article is available on the

WWW under http://dx.doi.org/10.1002/ejic.201400118.

Eur. J. Inorg. Chem. 2014, 1984–2001

1984

www.eurjic.org

FULL PAPER

or the tuning of PET pathways by modifying the excited- acceptor (Q) and donor (Fc) levels, namely oxidative (P^Ar

state energies of the central chromophore in a systematic to Q) versus reductive (Fc to P^Ar) PET with respect to the

manner with Fc donor and Q acceptor levels held at a fixed substituent Ar, by electrochemical methods, absorption

energy.

spectroscopy, steady-state emission, time-resolved fluores-

Linking reaction centre components through amide cence, transient absorption spectroscopy as well as by DFT

bonds has considerable advantages, namely a rather rigid calculations.

connectivity of the building blocks,^[11,12]a defined direc-

The results of this work will be of considerable value to

tional design in linear arrays (N terminusǞC terminus) the rational design of artificial photosynthetic systems that

and a simple orthogonal modification at the termini with feature multistep electron transfer by well-defined efficient

electron donors and acceptors. Furthermore, oligoamide single-electron transfer and charge-shift steps. Through our

synthesis can be conveniently accomplished by using solid- design strategy it is easy to tune the individual energy levels

phase synthesis methods.^[13]

of the chromophore radical ions relative to the acceptor

We have previously reported suitable building blocks for and donor levels. This versatile modular design will allow

directional oligoporphyrin amides, namely porphyrin the construction of even larger arrays with predefined se-

amino acids with a trans-AB₂C substitution pattern quences and excited-state energy levels.

(Scheme 1).^[11,14]These porphyrin amino acids feature elec-

tron-donating mesityl, phenyl and electron-withdrawing flu-

Results and Discussion

orinated aryl groups at the meso positions (Ar = Mes,

C₆H₅, 4-C₆H₄F, 4-C₆H₄CF₃, C₆F₅).^[6c,11,14]More strongly

electron-donating substituents such as Ar = 4-C₆H₄OR

have not been studied. The key building blocks used in this

study are porphyrin amino acids P^Arand their esters with

Synthesis and Characterisation of Building Blocks 3, Dyads

4, Triads 6 and Zn-6, and Tetrad 9

In addition to the recently reported trans-AB₂C por-

a full range of Ar substituents, anthraquinone-2-carboxylic phyrin amino acids P^Ar[11,14]with Ar = Mes (3c), C₆H₅(3e),

acid Q as the electron acceptor and aminoferrocene Fc- 4-C₆H₄F (3f), 4-C₆H₄CF₃(3g) and C₆F₅(3h), substituents

NH₂as the electron donor.^[15]

with a greater electron-donating power are required for a

comprehensive study. Hence, three novel trans-AB₂C-sub-

stituted porphyrin^[16]amino acid esters^[11,14]with electron-

releasing meso substituents [Ar = 4-C₆H₄OnBu (3a), 4-

C₆H₄OMe (3b), 4-C₆H₄Me (3d)] were synthesised following

a Lindsey-type 2+2 condensation procedure^[17]employing

nitro- and ester-substituted phenyldipyrromethanes.^[18]The

nitro functional group was reduced to the amine by re-

duction with tin(II) chloride in hydrochloric acid to obtain

the porphyrin amino acid esters 3a, 3b and 3d (see the Sup-

porting Information). As the optical properties of amines

are strongly pH-dependent and the oxidation of amines is

often irreversible, N-acetyl-protected reference compounds

Ac-3a, Ac-3b, Ac-3d and Ac-3e as well as their correspond-

ing Zn^IIcomplexes Zn-Ac-3a, Zn-Ac-3b, Zn-Ac-3d and Zn-

Scheme 1. Porphyrin amino acid building blocks.

Herein we report 1) the synthesis and properties of new Ac-3e were synthesised following established procedures

trans-AB₂C-substituted porphyrin amino acid esters with (see the Supporting Information).^[11,19]

electron-donating meso substituents Ar = 4-C₆H₄OnBu, 4-

The eight porphyrin amino acid esters 3a–3h, with Ar

[14]

C₆H₄OMe, 4-C₆H₄Me and C₆H₅

(3a, 3b, 3d, 3e) and exhibiting increasing electron-withdrawing power, were

their corresponding N-acetyl-protected derivatives (Ac-3a, coupled to anthraquinone-2-carboxylic acid (as electron ac-

Ac-3b, Ac-3d, Ac-3e) as building blocks and reference com- ceptor) after activating the latter as the acid chloride by

pounds, 2) the coupling of eight trans-AB₂C-substituted reaction with thionyl chloride to give the Q-P^Ardyads 4a–

porphyrin amino acid building blocks P^Arwith varying elec- 4h (Scheme 2). Cleavage of the methyl ester by using aque-

tron-withdrawing power (Ar = 4-C₆H₄OnBu, 4-C₆H₄OMe, ous KOH in methanol/THF delivered the quinone-por-

4-C₆H₄Me, Mes, C₆H₅, 4-C₆H₄F, 4-C₆H₄CF₃, C₆F₅) to phyrin acids 5a–5g, respectively (Scheme 2). The base-cata-

anthraquinone-2-carboxylic acid Q to give Q-P^Ardyads lysed cleavage of dyad 4h also led to the nucleophilic aro-

(4a–4h), 3) the amide coupling of Fc-NH₂to Q-P^Arto give matic substitution of fluoride in the electron-poor aryl sub-

Q-P^Ar-Fc triads (6a–6h) and metallation with zinc(II) to stituents by methoxide (Ar = C₆F₅ǞAr = 4-C₆F₄OMe) to

give Q-(Zn)P^Ar-Fc triads (Zn-6a–Zn-6h), 4) the (formal) ex- give acid 5i. Therefore to obtain 5h from 4h, an acid-cata-

tension of the triad Q-P^Ar1-Fc (Ar¹= C₆H₅) with an ad- lysed hydrolysis protocol needed to be employed

ditional porphyrin amino acid P^Ar2(Ar²= 4-C₆H₄Me) to (Scheme 2). Quinone-porphyrin-ferrocene triads Q-P^Ar-Fc

furnish the tetrad Q-P^Ar1-P^Ar2-Fc (9) and 5) the examina- (6a–6i) were assembled by activation of the acid groups in

tion of initial charge separation pathway preferences in dy- 5a–5i as acid chlorides by using thionyl chloride and cou-

ads 4, triads 6 and Zn-6 as well as in tetrad 9 with fixed pling with aminoferrocene^[15](as electron donor). The quin-

Eur. J. Inorg. Chem. 2014, 1984–2001

1985

www.eurjic.org

FULL PAPER

Scheme 2. Synthesis of dyads 4a–4h and triads 6a–6i and Zn-6a–Zn-6i.

one-(Zn)porphyrin-ferrocene triads Q-(Zn)P^Ar-Fc (Zn-6a– a second amide proton resonance at δ = 9.19–9.29 ppm in

1

Zn-6i) were obtained from the free base triads 6a–6i by me- the H NMR spectra of 6a–6h. The presence of three NH

tallation with zinc(II) acetate dihydrate (Scheme 2).

amide resonances at δ = 9.21, 10.21 and 10.33 ppm demon-

The construction of the Q-P^C6H5-P^C6H4Me-Fc tetrad 9 is strates the successful synthesis of tetrad 9. The resonances

depicted in Scheme 3. Q-P acid 5e was coupled to porphyrin corresponding to the quinone and ferrocene moieties in the

amino acid ester 3d (activated as the acid chloride by using dyads, triads and tetrads further prove the integrity of the

thionyl chloride) to give the Q-P^C6H5-P^C6H4Metriad 7. The oligoamides. Metallation of triads 6 with Zn^IIresulted in

ferrocene entity was attached to 7 through an amide bond the expected changes in the porphyrin resonances, such as

after base-catalysed methyl ester cleavage (to give free acid the disappearance of the pyrrole NH resonances. For the

8, see the Supporting Information). Subsequent formation Q-(Zn)P dyads featuring an azomethine linker, the quinone

of the acid chloride and coupling with aminoferrocene gave proton resonances are reported to be shifted to a higher

the target tetrad 9. The quinone reference N-ethylanthra- field. This has been rationalised by the coordination of the

quinone-2-carboxylic acid amide 10 as well as the bis(por- quinone carbonyl oxygen atom to the Zn^IIcentre.^[3m]The

phyrin) reference P^C6H5-P^C6H4Me11 were synthesised as de- Q proton resonances of the Q-Zn(P)-Fc triads Zn-6 are not

scribed in the Supporting Information.

shifted, and hence a Zn···O=C interaction in Zn-6 can be

The correct formation of all monoporphyrins 3, the dy- excluded. Possibly for the reported Q-Zn(P) dyads, a

ads 4, the triads 6, Zn-6 and 7, the tetrad 9, the quinone Zn···N_azomethineinteraction seems to be more likely than

reference compound 10 and the bis(porphyrin) reference 11 Zn···OC_quinonecoordination.^[3m]

was substantiated by multinuclear NMR spectroscopy (¹H,

¹³C, ¹⁹F), IR and UV/Vis spectroscopy as well as (high-

resolution) mass spectrometry (see the Supporting Infor-

mation). In all cases the mass spectra show the signals ex-

Absorption and Emission of Dyads 4, Triads 6 and Zn-6,

and Tetrad 9

pected for the intact porphyrin and porphyrin conjugates.

In the ¹H NMR spectra, successful amide formation is

The absorption data for the dyads, triads, tetrad and ref-

clearly indicated by the disappearance of the amine proton erence compounds are presented in Table 1. The absorption

resonances of 3a–3h at δ = 4.99 ppm and the appearance of spectra of 4a–4h, 6a–6h and Zn-6a–Zn-6h are essentially

the amide resonances of dyads 4a–4h at δ = 10.30– identical to those of the corresponding reference porphyrins

10.40 ppm. The covalent attachment of the ferrocene by Ac-3a–Ac-3h and Zn-Ac-3a–Zn-Ac-3h. As a result of the

amide coupling to the dyads 4 was confirmed by observing low extinction coefficients of N-ethylanthraquinone-2-carb-

Eur. J. Inorg. Chem. 2014, 1984–2001

1986

www.eurjic.org

FULL PAPER

Scheme 3. Synthesis of tetrad 9.

oxylic acid amide 10 (λ_max= 328 nm, ε = 5300 m^–1cm^–1)

Figure 2 shows the absorption spectra of triad Q-P^C6H5

-

and 1-(acetylamino)ferrocene (Fc-NHAc; λ_max= 441 nm, ε P^C6H4Me(7), tetrad Q-P^C6H5-P^C6H4Me-Fc (9) and reference

= 215 m^–1cm^–1),^[15]no absorption bands from these chro- dyad P^C6H5-P^C6H4Me(11). Appending a quinone or a ferro-

mophores are discernible in the absorption spectra of the cene to the bis(porphyrin) core does not alter the absorp-

dyads and triads. The shapes and intensities of the Soret tion spectrum, similarly to the monoporphyrin conjugates.

bands relative to their Q bands remain unaffected after the Furthermore, the absorption spectra of the bis(porphyrins)

linking of quinone and ferrocene to the porphyrin core 7, 9 and 11 are nearly identical to those of the constituent

(Figure 1 and the Supporting Information). A slight blue porphyrins Ac-3d and Ac-3e and exhibit no splitting of the

shift of the porphyrin Soret and porphyrin Q bands is ob- Soret or Q bands.^[11]In summary, the absorption spectra

served within the Q-P series 4aǞ4h and Q-P-Fc series are essentially a superposition of the individual unper-

6aǞ6h due to the increasing electron-withdrawing charac- turbed spectra of the building blocks.

ter of meso-aryl substituents Ar. Neither new charge-trans-

The fluorescence data, quantum yields and the exited-

fer bands^[20]nor perturbations of the porphyrin absorption state lifetimes of the new porphyrin amino acid esters 3a,

bands were detected, which indicates only a weak ground- 3b, 3d and 3e, and the reference porphyrins Ac-3a, Ac-3b,

state electronic interaction between the quinone or the fer- Ac-3d, Ac-3e, Zn-Ac-3a, Zn-Ac-3b, Zn-Ac-3d, Zn-Ac-3e

rocene with the central porphyrin. The Q-(Zn)P^Ar-Fc triads and bis(porphyrin) 11 are collected in Table 2. As discussed

Zn-6a–Zn-6h show a small bathochromic shift (1–4 nm) of previously^[11]for related porphyrins, a small hypsochromic

porphyrin absorbance bands in comparison with the corre- shift of the emission band is observed in the series

sponding reference porphyrins Zn-Ac-3a–Zn-Ac-3h (Fig- Ac-3aǞAc-3bǞAc-3dǞAc-3e and Zn-Ac-3aǞZn-Ac-

ure 1 and Table 1).

3bǞZn-Ac-3dǞZn-Ac-3e.

Eur. J. Inorg. Chem. 2014, 1984–2001

1987

www.eurjic.org

FULL PAPER

Table 1. Absorption data for Ac-3a–Ac-3h, Zn-Ac-3a–Zn-Ac-3h, 4a–4h, 6a–6i, Zn-6a–Zn-6i, 7, 9 and 11 in CH₂Cl₂at room temperature.

Ar

λ [nm] (ε [10^{–4 –1}cm^–1])

m

Soret

Q_y(1,0)

Q_y(0,0)

Q_x(1,0)

Q_x(0,0)

Ac-3a

4a

C₆H₄OnBu

422 (30.03)

422 (32.44)

422 (36.31)

424 (40.10)

425 (47.78)

517 (1.14)

518 (1.36)

518 (1.35)

552 (1.75)

552 (1.33)

555 (0.73)

555 (0.84)

555 (0.89)

594 (0.65)

594 (0.78)

593 (0.63)

593 (0.49)

593 (0.36)

649 (0.65)

649 (0.47)

650 (0.39)

6a

Zn-Ac-3a

Zn-6a

–

Ac-3b

4b

C₆H₄OMe

421 (35.40)

421 (39.59)

422 (39.19)

424 (48.62)

424 (51.07)

517 (1.38)

517 (1.40)

518 (1.67)

551 (1.90)

552 (1.96)

554 (0.87)

554 (0.89)

554 (1.11)

593 (0.63)

594 (0.73)

593 (0.45)

593 (0.33)

593 (0.69)

648 (0.42)

649 (0.35)

649 (0.53)

6b

Zn-Ac-3b

Zn-6b

–

Ac-3c

4c

Mes

420 (36.29)

421 (36.33)

421 (36.18)

423 (45.12)

424 (49.71)

516 (1.24)

516 (1.55)

516 (1.51)

551 (1.92)

552 (1.88)

550 (0.37)

551 (0.68)

552 (0.60)

592 (0.45)

594 (0.18)

593 (0.28)

592 (0.34)

592 (0.30)

652 (0.55)

651 (0.56)

648 (0.25)

–

6c

Zn-Ac-3c

Zn-6c

–

Ac-3d

4d

C₆H₄Me

420 (33.60)

420 (39.00)

420 (44.69)

421 (45.15)

423 (47.02)

516 (1.27)

517 (1.93)

517 (1.72)

549 (1.82)

551 (2.02)

552 (0.71)

552 (1.26)

553 (1.57)

588 (0.46)

593 (0.75)

592 (0.39)

591 (0.85)

591 (1.12)

647 (0.35)

647 (0.82)

647 (0.61)

6d

Zn-Ac-3d

Zn-6d

–

Ac-3e

4e

C₆H₅

419 (32.60)

419 (32.25)

419 (35.06)

420 (36.10)

423 (38.89)

515 (1.34)

516 (1.56)

549 (1.63)

551 (1.63)

551 (0.68)

551 (0.69)

551 (0.84)

589 (0.48)

592 (0.46)

591 (0.43)

591 (0.42)

591 (0.49)

646 (0.34)

646 (0.36)

646 (0.42)

6e

Zn-Ac-3e

Zn-6e

–

Ac-3f

4f

C₆H₄F

419 (50.42)

419 (46.25)

419 (41.79)

420 (41.45)

422 (47.69)

515 (2.16)

515 (1.96)

515 (2.22)

548 (1.87)

551 (2.05)

551 (1.07)

551 (0.98)

551 (1.38)

587 (0.50)

593 (0.63)

591 (0.72)

591 (0.59)

589 (1.00)

647 (0.61)

647 (0.45)

646 (0.86)

6f

Zn-Ac-3f

Zn-6f

–

Ac-3g

4g

C₆H₄CF₃

419 (41.40)

419 (31.57)

419 (32.42)

420 (28.30)

423 (34.13)

515 (1.78)

515 (1.45)

515 (1.54)

548 (1.17)

552 (1.59)

550 (0.79)

549 (0.69)

551 (0.78)

588 (0.21)

593 (0.48)

590 (0.57)

591 (0.47)

590 (0.53)

646 (0.40)

649 (0.41)

646 (0.41)

6g

Zn-Ac-3g

Zn-6g

–

Ac-3h

4h

C₆F₅

418 (39.31)

418 (34.69)

417 (41.02)

422 (51.09)

422 (47.51)

512 (2.38)

512 (2.15)

511 (2.20)

552 (2.01)

551 (1.68)

545 (0.84)

546 (0.70)

545 (0.42)

593 (0.36)

594 (0.65)

587 (0.93)

588 (0.79)

588 (0.49)

642 (0.56)

642 (0.44)

642 (0.48)

6h

Zn-Ac-3h

Zn-6h

–

6i

Zn-6i

C₆F₄OMe

419 (37.03)

423 (43.97)

512 (1.87)

552 (1.86)

546 (0.45)

592 (0.32)

589 (0.48)

–

644 (0.11)

–

11

7

9

C₆H₅/C₆H₄Me

422 (73.50)

422 (76.62)

422 (77.48)

516 (3.53)

516 (3.57)

517 (3.74)

552 (2.05)

552 (2.06)

553 (2.14)

591 (1.19)

591 (1.15)

591 (1.14)

647 (1.06)

647 (1.04)

647 (1.02)

The emission spectra of selected dyads, triads, tetrad 9 of the meso-aryl substituent Ar controls the efficiency of

and reference porphyrins are shown in Figures 1 and 2. fluorescence quenching by oxidative PET. Electron-rich

Linking anthraquinone to porphyrins 3 and Zn-3 as well as substituents induce stronger quenching than electron-with-

to (bis)porphyrin 11 leads to a significant decrease in the drawing substituents. Only dyad 4c with sterically de-

fluorescence intensity of quinone conjugates 4 and 7 (see manding mesityl groups (Ar = Mes) at the meso positions

Tables 2 and 3). The quenching efficiency varies between 10 deviates from the described trend. DFT geometry optimisa-

and 92%, which indicates the presence of a new non-radia- tions of 4c revealed a 90° dihedral angle between the mesityl

tive decay channel in the dyads, namely oxidative PET. The and porphyrin planes in contrast to the smaller angles ob-

strongest decrease in fluorescence intensity is found for served for all other substituents (Ar = C₆F₅: 79°; all other

dyad 4a with Ar = 4-C₆H₄OnBu (92%, Figure 1), whereas Ar: 63–66°; see the Supporting Information). Furthermore,

almost no quenching at all is seen for dyad 4h with Ar = full rotation of a mesityl group connected to a porphyrin

C₆F₅(10%, Table 2 and Table 3). As anticipated, the nature has an activation barrier of around 74 kJmol^–1in solu-

Eur. J. Inorg. Chem. 2014, 1984–2001

1988

www.eurjic.org

FULL PAPER

Table 2. Emission data for porphyrins 3 and bisporphyrin 11 in

CH₂Cl₂at room temperature.

Ar

λ [nm]

Φ

τ [ns]

[a]

Q(0,0)

Q(0,1)

3a

C₆H₄OnBu

660

657

606

723

720

651

0.1631 n.d.^[b]

Ac-3a

0.1594

0.1003

9.28

1.71

Zn-Ac-3a C₆H₄OnBu

3b

C₆H₄OMe

659

657

606

721

719

651

0.1448 n.d.^[b]

Ac-3b

0.1348

0.0940

9.49

1.81

Zn-Ac-3b C₆H₄OMe

[b]

3d

C₆H₄Me

657

654

600

721

719

646

0.0878 n.d.

0.1311

0.0884

Ac-3d

9.69

1.91

Zn-Ac-3d C₆H₄Me

Ac-3e

C₆H₅

Zn-Ac-3e C₆H₅

652

597

717

645

0.1167

0.1037

9.88

1.99

11 C₆H₅/C₆H₄Me

654

718

0.1284

9.70

[a] All decays are monoexponential; excitation wavelength λ_exc

=

400 or 550 nm. [b] n.d.: not determined.

coefficients of the constituent porphyrins Ac-3d and Ac-3e

(Table 1), it is reasonable to assume that both porphyrins in

7 are excited with nearly equal probability [ε(Ac-3d)/ε(Ac-

3e) ≈ 1.03:1.00]. If energy transfer between the two por-

phyrins was absent, approximately 80%(4e)/2 = 40%

quenching would be expected. The larger quenching ob-

served suggests rapid energy transfer from the Ar =

C₆H₄Me-substituted porphyrin to the Ar = C₆H₅-substi-

tuted porphyrin in 7. The latter excited porphyrin un-

dergoes oxidative PET to the appended quinone.

Figure 1. Normalised absorption and emission spectra of a) Ac-3a,

4a, 6a, b) Zn-Ac-3a, Zn-6a, c) Ac-3e, 4e, 6e, d) Zn-Ac-3e, Zn-6e,

e) Ac-3h, 4h, 6h and f) Zn-Ac-3h, Zn-6h in CH₂Cl₂at room tem-

perature.

The fluorescence studies of the ferrocene conjugates 6

and Zn-6 revealed an even stronger reduction of the emis-

sion quantum yield compared with the corresponding refer-

ence compounds Ac-3 and Zn-Ac-3 (Figure 1, Tables 2 and

4). The quenching of triads 6a–6h and metallated triads Zn-

6a–Zn-6h is almost quantitative and most of them are es-

sentially non-fluorescent. Therefore a further efficient non-

radiative decay pathway is enabled, namely reductive PET

with ferrocene acting as electron donor and/or triplet en-

ergy transfer to the ferrocene.^[7a,10]

To distinguish between electron and energy transfer, the

emission spectra of Ac-3a, 4a and 6a (Figure 3) as well as

Ac-3h, 4h and 6h were exemplarily recorded at low tempera-

tures (300Ǟ77 K). Indeed, upon cooling, the integrated

emission intensity of 6a (ϫ5.0) increases much more

strongly than that of Ac-3a (ϫ1.3) or 4a (ϫ1.7) (Figure 3).

A similar trend is observed for the Ac-3h (ϫ1.8), 4h (ϫ1.4),

6h (ϫ3.3) series. Even more pronounced is the gain in fluo-

rescence intensity in the zinc triads Zn-6a (ϫ17.5) and Zn-

6h (ϫ16.4) at low temperature. This supports the existence

Figure 2. Normalised absorption and emission spectra of the refer-

ence compound P^C6H5-P^C6H4Me(11), triad Q-P^C6H5-P^C6H4Me(7)

and tetrad Q-P^C6H5-P^C6H4Me-Fc (9) in CH₂Cl₂at room tempera-

ture.

tion,^[21]which also prevents the population of intermediate of PET pathways in triads 6 and Zn-6 that are impeded at

more co-planar conformations. Therefore the rather fixed low temperatures.^[3i]However, some residual quenching by

orthogonal arrangement of the Mes substituent precludes energy transfer to ferrocene cannot be fully excluded on the

the strong electronic influence expected from the trialkyl- basis of these data alone.

ated aryl group.

In quinone-bis(porphyrin) 7, the emission is quenched by the triads 6. The quantum yield of the Q-P^C6H5-P^C6H4Me

only 64% due to the attached quinone. From the extinction Fc conjugate 9 is decreased by 89% with respect to the

Tetrad 9 essentially shows similar emission behaviour to

-

Eur. J. Inorg. Chem. 2014, 1984–2001

1989

www.eurjic.org

FULL PAPER

Table 3. Emission data for Q-P^Ardyads 4a–4h and triad 7 in CH₂Cl₂at room temperature.

τ [ns]^[a]

Quenching [%]

k_ET[10⁹s^–1]^[d]

Φ_ET

[e]

Ar

λ [nm]

Φ

Q(0,0)

Q(0,1)

4a

4b

4c

4d

4e

4f

4g

4h

7

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

657

656

653

654

652

647

653

720

719

717

712

718

0.0125

0.0140

0.0532

0.0161

0.0230

0.0281

0.0548

0.0628

0.0458

0.613

0.700

1.820

0.942

1.370

2.970

4.810

9.550

4.070

92^[b]

90^[b]

32^[b]

78^[b]

80^[b]

72^[b]

49^[b]

10^[b]

64^[c]

1.524

1.323

0.458

0.958

0.629

0.233

0.108

0.003

0.143

0.93

0.83

0.90

0.86

0.69

0.52

0.03

0.58

C₆H₅/C₆H₄Me

[a] All decays are monoexponential; excitation wavelength λ_exc= 400 or 550 nm. [b] Relative to reference compound Ac-3. [c] Relative to

reference compound 11. [d] k_ET= 1/τ – 1/τ_ref. [e] Φ = k_ETτ.

Table 4. Emission data for triads 6a–6i, Zn-6a–Zn-6i and tetrad 9 in CH₂Cl₂at room temperature.

Quenching [%] k_ET[10⁹s^–1]^[d]

Φ_ET

[e]

Ar

λ [nm]

Φ

τ₁[ps] (A₁[%]);

Q(0,0)

Q(0,1)

τ₂[ps] (A₂[%])^[a,b]

6a

6b

6c

6d

6e

6f

6g

6h

6i

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

656

657

652

655

651

652

648

650

720

718

720

717

718

717

712

714

0.0042

0.0034

0.0041

0.0036

0.0030

0.0025

0.0063

0.0029

0.0006

214.99 (100)

240.54 (100)

272.83 (100)

229.37 (100)

164.02 (86); 644.23 (14)

148.07 (94); 586.62 (6)

94.96 (86); 467.65 (14)

26.39 (96); 274.57 (4)

39.30 (93); 398.03 (7)

97

94

97

98

94

4.544

4.052

3.574

4.247

5.996

6.650

10.43

37.79

n.d.^[c]

0.98

0.97

0.98

0.99

1.00

n.d.^[c]

96

C₆F₄OMe

n.d.^[c]

Zn-6a

Zn-6b

Zn-6c

Zn-6d

Zn-6e

Zn-6f

Zn-6g

Zn-6h

Zn-6i

9

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

607

606

604

607

601

602

605

602

654

652

651

653

647

649

653

654

651

718

0.0011

0.0005

0.0014

0.0012

0.0009

0.0049

0.0028

0.0025

0.0020

0.0145

23.04 (87); 139.15 (13)

33.34 (92); 182.80 (8)

25.92 (77); 184.51 (13)

27.13 (83); 189.38 (17)

29.53 (88); 107.63 (12)

133.53 (73); 281.39 (27)

35.20 (75); 285.80 (25)

20.65 (94); 215.88 (6)

15.33 (90); 142.73 (10)

361.77 (67); 2152.55 (33)

99

98

99

97

42.82

29.39

38.12

36.34

33.36

6.948

27.93

0.99

0.98

0.99

0.93

0.98

0.99

n.d.^[c]

96

47.91

C₆F₄OMe

C₆H₅/C₆H₄Me

n.d.^[c]

89

n.d.^[c]

2.661; 0.363

[a] Decays are mono- or biexponential; excitation wavelength λ_exc= 400 or 550 nm. [b] The relative amplitudes A were calculated from

fitting the data by the equations Y = A₁e^–k/τ1+ A₂e^–k/τ2and A₁+ A₂= 100%. [c] n.d.: not determined. [d] k_ET= 1/τ – 1/τ_ref. [e] Φ_ET

_ETτ; τ = τ₁with the largest amplitude used in the equations.

=

k

bis(porphyrin) reference 11. When we neglect interpor- cited states of the reference porphyrins Ac-3, Zn-Ac-3, ref-

phyrin energy transfer, approximately 50% of the decay erence dyad 11 and Q-P dyads 4a–4h decay with a single

should occur oxidatively through Q-P^C6H5and 50% re- rate constant. The PET from the porphyrin to the anthra-

ductively through P^C6H4Me-Fc. The observed intermediate quinone in the Q-P dyads 4 opens a new deactivation path-

quenching of 9 (89%) between that of 4e (80%) and 6d way for the excited singlet state and reduces its lifetime gen-

(97%) is in accord with the concept of two decay pathways. erating the Q^·–-P^·+charge-separated state (CS state). Other

Therefore both PET pathways appear possible in 9.

pathways (e.g., originating from adventitiously reduced

anthraquinone to the hydroquinone^[3l]) are not detected in

dyads 4. The fluorescence lifetimes of 4 are in the range of

τ = 0.613 ns (4a, Ar = 4-C₆H₄OnBu) to τ = 9.55 ns (4h, Ar

= C₆F₅). A clear trend (with the exception of 4c, Ar = Mes,

see above) is observed with the more electron-rich por-

phyrins featuring shorter lifetimes (Table 3). From these

data, rate constants k_ETfor oxidative PET were calculated

(Table 3). The rate constants of the electron-rich porphyrins

4a, 4b and 4d (1.5ϫ10⁹–0.9ϫ10⁹s^–1) are fully compatible

with those of the reported Q-P dyads consisting of electron-

rich porphyrins and quinones.^[3]The much lower rates of

Kinetics of Photoinduced Electron Transfer by Time-

Resolved Fluorescence Decay

The fluorescence lifetimes τ of dyads, triads, tetrad 9 and

reference compounds are presented in Tables 2–4 and the

fluorescence decay curves of selected conjugates and refer-

ence porphyrins are shown in Figure 4.

The fluorescence lifetimes of the free-base reference por-

phyrins and P^Ph-P^PhMe(11) are τ ≈ 10 ns and those of the

zinc(II) porphyrins are τ ≈ 2 ns, as expected.^[11]All the ex-

Eur. J. Inorg. Chem. 2014, 1984–2001

1990

www.eurjic.org

FULL PAPER

Figure 4. Fluorescence decay profiles of a) Ac-3a, 4a, 6a, b) Zn-Ac-

3a, Zn-6a, c) Ac-3e, 4e, 6e, d) Zn-Ac-3e, Zn-6e, e) Ac-3h, 4h, 6h,

f) Zn-Ac-3h, Zn-6h and g) 11, 7, 9 in CH₂Cl₂(λ_exc= 400 and

550 nm).

Figure 3. Normalised emission spectra of Ac-3a, 4a and 6a in 2-

methyltetrahydrofuran at T = 300Ǟ77 K.

stituents (Table 3), the reverse trend is observed for Q-P^Ar-

ET observed for the electron-withdrawing porphyrin dyads Fc triads 6 (Table 4). The latter observation would be ex-

4f–4h are related to the lower thermodynamic driving force pected for reductive PET from ferrocene to the excited por-

for the ET (see Electrochemical Studies below).

phyrin as the dominant pathway. Indeed, k_ETis minimal for

Triads 6a–6d feature a monoexponential decay whereas 6b–6d, increases slightly for electron-rich 6a favouring PET

the decays of all other dyads have to be fitted by biex- to the quinone and increases dramatically for electron-poor

ponential decay curves (Table 4). The minor component (τ₂, 6e–6h favouring PET from ferrocene. Hence reductive PET

A₂) could be due to additional quenching by the heavy- seems to be favourable in 6e–6h whereas oxidative and re-

atom effect of the ferrocene,^[7c]population of the ferrocene ductive pathways are both accessible in 6a–6d.

triplet state,^[10]the presence of ferrocene/amide conforma-

In the zinc(II) porphyrin triads Zn-6, the rate constants

tions that are less suitable for ET or intermolecular k_ETare large (30ϫ10⁹–40ϫ10⁹s^–1; except for Zn-6f, which

Zn···O=C(quinone) interactions in Zn-6.^[3m]The latter is unexplained at the moment) and ET is basically quantita-

path, however, seems to be less pronounced based on NMR tive. No clear correlation between k_ETand the nature of

spectroscopic data for Zn-6 (see above). Rate constants the porphyrin substituents can be discerned (Table 4). This

were estimated from the fluorescence lifetimes with the might be explained by the larger driving force [Δ(ΔG_ET) ≈

largest amplitude (τ₁, A₁; Table 4). These rate constants for 0.17 eV] of both PET pathways in zinc(II) porphyrin donor/

6a–6h are larger than those of the corresponding 4a–4h acceptor assemblies due to the higher energy of the singlet

lacking ferrocene (Table 3 and Table 4). Whereas k_ETde- excited state of zinc(II) porphyrins and the lower oxidation

creases for Q-P^Ardyads 4 with electron-withdrawing sub- potential of zinc(II) porphyrins relative to free-base por-

Eur. J. Inorg. Chem. 2014, 1984–2001

1991

www.eurjic.org

FULL PAPER

phyrins (Table 4; see Electrochemical Studies below). Other

In Q-P dyads 4a–4h an additional redox event at a rather

3

scenarios would be accelerated quenching via ZnP states constant potential of E_1/2= –1.20 V corresponding to the

due to the heavy-atom effect of ferrocene^[7c]or via ferrocene first reversible reduction of the quinone to the semiquinon-

triplet states.^[10]

ate is observed. For dyad 4c even the second reduction of

As suggested from the quantum yield data, the singlet anthraquinone at E_1/2= –1.72 V has been detected. The in-

excited states of tetrad 9 can be quenched by PET from the dividual redox potentials of the porphyrins Ac-3 are essen-

C₆H₅-substituted porphyrin to its adjacent quinone and/or tially unperturbed by the appended quinone in 4 (Table 6

by PET from ferrocene to its adjacent C₆H₄Me-substituted and Figure 5).

porphyrin. The two fluorescence lifetimes of 9 [τ₁= 362 ps

(67%) and τ₂= 2153 ps (33%)] roughly correspond to the

lifetimes of Q-P^C6H4Me-Fc 6d (τ = 229 ps) and Q-P^C6H54e

(τ = 1370 ps). This suggests that indeed both initial PET

pathways are conceivable in 9 with the reductive PET being

more probable than the oxidative PET. On the basis of this

interpretation, the rate constants were estimated to be

2.66ϫ10⁹s^–1(reductive PET) and 0.36ϫ10⁹s^–1(oxidative

PET). The thermodynamic feasibility of reductive and oxi-

dative PET will be discussed in the next sections.

Electrochemical Studies

Cyclic voltammograms of porphyrins typically show two

reversible oxidation and two reversible reduction waves.

Figure 5. Cyclic voltammograms of quinone reference 12, por-

Porphyrins 3a–3h with amine substituents are only quasi-

reversibly or even irreversibly oxidised.^[11,14]As the amino

porphyrins do not qualify as references, their N-acetylated

phyrin Ac-3e, dyad 4e and triad 6e in (nBu₄N)(PF₆)/CH₂Cl₂.

Attachment of a ferrocene unit (6a–6h, Zn-6a–Zn-6h)

analogues Ac-3a–Ac-3h were used instead. The redox po- yields a further oxidation process at a rather constant po-

tentials of the new amino porphyrins 3a, 3b, 3e and 3d and tential of E_1/2= –0.08 V corresponding to the oxidation of

the porphyrin reference compounds Ac-3a, Ac-3b, Ac-3d ferrocene to ferrocenium (Table 6 and Figure 5). Only

and Ac-3e are presented in Table 5. As described previously, minor effects of ferrocene on the porphyrin and anthra-

the meso-aryl substituents cause a stepwise anodic shift of quinone redox potentials are observed. Likewise only minor

porphyrin oxidation and reduction.^[11]The new N-acetyl- effects of the Ar substituent on the Fc and Q potentials are

ated porphyrins Ac-3a, Ac-3b, Ac-3d and Ac-3e confirm noted, which indicates a weak interchromophore interac-

and complete the trend in the full series Ar = 4- tion in the ground state. The quinone in the zinc porphyrins

C₆H₄OnBuǞ4-C₆H₄OMeǞ4-C₆H₄Me ǞMesǞC₆H₅Ǟ Zn-6 is slightly easier to reduce than in the free-base por-

4-C₆H₄FǞ4-C₆H₄CF₃ǞC₆F₅. The shifts observed phyrins 6 (Table 6).

amount to 0.34 V for the first oxidation and 0.25 V for the

first reduction from Ac-3a (Ar = C₆H₄OnBu) to Ac-3h (Ar sented in Table 6. The additional porphyrin superimposes

= C₆F₅^[11]).

its redox waves onto those of the respective partner mono-

The redox data for bis(porphyrins) 11, 7 and 9 are pre-

Table 5. Redox potentials of monoporphyrins 3, Ac-3, Zn-Ac-3 and (bis)porphyrin 11 vs. Fc/Fc⁺in (nBu₄N)(PF₆)/CH₂Cl₂at room

temperature.

Ar

C₆H₄–RЈ

E_1/2(ox¹) [V]

E_1/2(ox²) [V]

E_1/2(red¹) [V]

E_1/2(red²) [V]

3a

C₆H₄OnBu

NH₂

NHAc

0.420

0.480

0.300

0.590

0.730

0.720

–1.660^[a]

–1.640

–1.790

–2.010^[a]

–1.980

–2.160

Ac-3a

Zn-Ac-3a

3b

C₆H₄OMe

NH₂

NHAc

0.540^[a]

0.510

0.310

n.o.^[b]

0.750

0.710

–1.660

–1.630

–1.780

–1.990

–1.970

n.o.^[b]

Ac-3b

Zn-Ac-3b

3d

C₆H₄Me

NH₂

NHAc

0.440

0.520

0.361

0.550

0.830

0.800

–1.640^[a]

–1.610

–1.720

–1.980^[a]

–1.950

–2.090

Ac-3d

Zn-Ac-3d

Ac-3e

Zn-Ac-3e

C₆H₅

NHAc

0.550

0.360

0.840

0.780

–1.630

–1.750

–2.000

–2.110

11

C₆H₅/C₆H₄Me

NHAc

0.540 (2e)

0.860

–1.660 (2e)

–2.010

[a] Irreversible. [b] n.o.: not observed.

Eur. J. Inorg. Chem. 2014, 1984–2001

1992

www.eurjic.org

FULL PAPER

Table 6. Redox potentials of 4, 6, Zn-6, 11, 7 and 9 in V vs. Fc/Fc⁺in (nBu₄N)(PF₆)/CH₂Cl₂at room temperature.

Ar

E(P)_1/2(ox¹) [V] E(P)_1/2(ox²) [V]

E(Fc)_1/2(ox) [V] E(Q)_1/2(red) [V] E(P/Q)_1/2(red¹) [V] E(P)_1/2(red²) [V]

4a

4b

4c

4d

4e

4f

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

0.500

0.560

0.540

0.580

0.600

0.670

0.830

0.750

0.760

0.960

0.840

0.870

0.880

0.960

1.050

–

–1.230

–1.210

–1.190

–1.210

–1.200

–1.240

–1.640

–1.660

–1.720 (2e)

–1.600

–1.560

–1.580

–2.000

–1.990

–2.070

–1.980

–1.940

–1.930

–1.870

–1.850

4g

4h

–1.530

–1.420

6a

6b

6c

6d

6e

6f

6g

6h

6i

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

0.510

0.500

0.560

0.550

0.560

0.585

0.640

0.800

0.770

0.990

0.855

0.925

0.945^[a]

0.990

1.040

1.075

–0.070

–0.079

–0.070

–0.075

–0.090

–0.085

–0.080

–0.100

–0.050

–1.210

–1.220

–1.230

–1.190

–1.260

–1.235

–1.230

–1.270

–1.240

–1.650

–1.710 (2e)

–1.605

–1.670

–1.625

–1.570

–1.430

–1.420

–2.010^[a]

–2.070

–1.965

–2.060

–2.015^[a]

–1.930

–1.900

–1.890

C₆F₄OMe

Zn-6a C₆H₄OnBu

Zn-6b C₆H₄OMe

Zn-6c Mes

Zn-6d C₆H₄Me

Zn-6e C₆H₅

0.370

0.360

0.410

0.390

0.430

0.440

0.480

0.630

0.600

0.670

0.750

0.710

0.740

0.750

0.770

0.910

0.880

–0.070

–0.080

–0.060

–0.089

–0.080

–0.100

–0.090

–1.170

–1.150

–1.620

–1.590^[a]

–1.620^[a]

–1.630^[a]

–1.580^[a]

–1.590

–2.200^[a]

n.o.^[b]

–1.180^[c]

–1.190^[c]

–1.120

–2.220^[a]

–2.230^[a]

–2.190^[a]

–2.180^[a]

n.o.^[b]

Zn-6f C₆H₄F

Zn-6g C₆H₄CF₃

Zn-6h C₆F₅

–1.120

–1.240^[a]

–1.180

–1.610^[a]

n.o.^[b]

–2.000^[a]

–2.040^[a]

Zn-6i C₆F₄OMe

–1.160

–1.610^[a]

11

7

9

C₆H₅/C₆H₄Me

0.540 (2e)

0.570 (2e)

0.590 (2e)

0.860 (2e)

0.890

0.870

–

–1.650 (2e)

–1.600 (2e)

–2.010 (2e)

–1.970

–1.980

–1.200

–1.190

–0.060

[a] Irreversible, E_pgiven. [b] n.o.: not observed. [c] Quasi-reversible.

porphyrin, and two-electron reversible redox waves for the Table 7) with k_ETincreasing the more negative ΔG_ETis (ex-

first oxidation and first reduction of both porphyrins at cept for the mesityl derivative 4c, see above). For 4h, the

around –0.59 and –1.60 V are detected. Tetrad 9 addition- ET is even calculated to be slightly endergonic, which ac-

ally shows the one-electron reduction of the quinone at counts for the observed weak fluorescence quenching.

E_1/2= –1.190 V and the one-electron oxidation of the ferro-

Table 7. Energies of the porphyrin S₁and CS states for dyads 4a–

cene at E_1/2= –0.060 V.

4h and the driving forces ΔG_ET

.

Ar

E(S₀ǞS₁)

E(Q^·–-P^·+)^[a]

ΔG_ET[eV]

[eV]

Thermodynamic Driving Force for Photoinduced Electron

Transfer

4a

4b

4c

4d

4e

4f

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

1.89

1.90

1.91

1.93

1.63

1.69

1.65

1.67

1.71

1.77

1.97

–0.26

–0.21

–0.25

–0.24

–0.20

–0.14

+0.04

The energies of the porphyrin first excited singlet states

E(S₀ǞS₁) of 4a–4h were obtained from the average values

of the Q_x(0,0) absorption and Q(0,0) emission bands

(Tables 1–4). The energies of the CS states Q^·–-P^·+were

evaluated from the first oxidation potential of the por-

phyrin E_1/2(P/P^·+), the first reduction potential of the quin-

4g

4h

[a] Including Coulomb term of 0.104 eV.

2

one E_1/2(Q/Q^·–) and the Coulomb term e₀/[4πε₀ε-

(CH₂Cl₂)r_AD] (Table 6). From these data the thermo-

As suggested above, two different PET pathways exist in

dynamic driving forces for PET were calculated according triads 6a–6h and Zn-6a–Zn-6h. The first one is the oxidative

to the Rehm–Weller equation:^[22]ΔG_ET= E_1/2(D/D^·+) – pathway, as is also found in Q-P dyads 4, leading to the Q^·–-

2

E_1/2(A/A^·–) – e₀/[4πε₀ε(CH₂Cl₂)r_AD] – E(S₀ǞS₁) with ε₀= P^·+-Fc and Q^·–-(Zn)P^·+-Fc CS states, respectively (Table 8,

8.85519ϫ10^–12Fm^–1, ε(CH₂Cl₂) = 8.93 and r_AD= 15.5 Å, Figures 7 and 8). The second one is the reductive pathway

as determined by DFT models (see below), to give a Cou- leading to the Q-P^·–-Fc^·+and Q-(Zn)P^·–-Fc^·+CS states,

lomb term of 0.104 eV in CH₂Cl₂. The data are presented respectively (Table 8, Figures 7 and 8). The final CS states

in Table 7. Figure 6 shows the energy level diagram for all are described by Q^·–-P-Fc^·+and Q^·–-(Zn)P-Fc^·+, respec-

Q-P dyads 4a–4h. This series of compounds displays a k_ET

/

tively. The free energies of the initial and final CS states

ΔG_ETcorrelation in the normal Marcus region (Table 3 and were estimated from the Rehm–Weller equation^[22]and are

Eur. J. Inorg. Chem. 2014, 1984–2001

1993

www.eurjic.org

FULL PAPER

Figure 7. Energy level diagram for Q-P-Fc triads 6a–6i (the por-

phyrin triplet state T₁is estimated to be 1.43 eV^[3f]).

Figure 6. Energy level diagram for Q-P dyads 4a–4h (IC = internal

conversion; the porphyrin triplet state T₁is estimated to be

1.43 eV^[3f]).

given in Table 8 (quinone-porphyrin centre-to-centre dis-

tance r_AD= 15.5 Å; Coulomb term for Q^·–-P^·+-Fc 0.104 eV;

porphyrin centre to iron distance r_AD= 13.0 Å; Coulomb

term for Q-P^·–-Fc^·+0.124 eV; quinone-centre to iron dis-

tance r_AD

=

28.3 Å; Coulomb term for Q^·–-P-Fc^·+

0.057 eV).

In the free-base porphyrin triads 6a–6i, the reductive

PET to give Q-P^·–-Fc^·+as the initial CS state is clearly fa-

voured over the oxidative PET that gives the Q^·–-P^·+-Fc ini-

tial CS state (Figure 7). This is especially pronounced in 6g–

6i with strongly electron-withdrawing substituents. In the

zinc(II) porphyrin triads Zn-6, oxidative PET becomes

competitive with reductive PET with respect to the driving

Figure 8. Energy level diagram for Q-(Zn)P-Fc triads Zn-6a–Zn-6i

(the zinc porphyrin triplet state T₁is estimated to be 1.53 eV.^[5b]).

force ΔG_ET(Table 8 and Figure 8). The final Q^·–-(Zn)P-Fc^·+centre distance 19.3 Å) above the excited singlet state S₁and

CS states of Zn-6 are slightly lower than the final Q^·–-P- are thus less relevant for excited-state decay (Figure 9 and

Fc^·+CS states of 6 due to the slightly preferred reduction Table 9). This is also reflected in the unchanged lifetime of

of Q in the zinc porphyrins Zn-6 (see above, Table 6).

11 (τ = 9.70 ns) with respect to reference porphyrins Ac-3d

(τ = 9.69 ns) and Ac-3e (τ = 9.88 ns). The thermodynamic

·–

In bis(porphyrin) 11, the conceivable P₁^·+-P₂and P₁

-

·+

P₂CS states are located 0.15 and 0.16 eV (including the driving forces for ET in tetrad 9 were again calculated from

Coulomb term of 0.08 eV; porphyrin-porphyrin centre-to- the Rehm–Weller equation^[22](see above) and are given in

Table 8. Energies of porphyrin S₁and CS states for 6 and Zn-6 and driving forces ΔG_ET

.

Ar

E(S₀ǞS₁) E(Q^·–-P^·+-Fc)^[a]E(Q-P^·–-Fc^·+)^[b]E(Q^·–-P-Fc^·+)^[c]ΔG_ET(Q^·–-P^·+-Fc) ΔG_ET(Q-P^·–-Fc^·+) ΔG_ET(Q^·–-P-Fc^·+)

[eV]

6a

6b

6c

6d

6e

6f

6g

6h

6i

C₆H₄OnBu

C₆H₄OMe

Mes

C₆H₄Me

C₆H₅

C₆H₄F

C₆H₄CF₃

C₆F₅

1.90

1.91

1.92

1.62

1.69

1.64

1.72

1.77

1.97

1.91

1.46

1.45

1.39

1.41

1.46

1.42

1.37

1.21

1.25

1.08

1.10

1.06

1.11

1.09

1.11

1.13

–0.28

–0.22

–0.27

–0.19

–0.04

+0.05

–0.01

–0.44

–0.45

–0.52

–0.50

–0.45

–0.49

–0.54

–0.71

–0.67

–0.82

–0.81

–0.84

–0.79

–0.82

–0.81

–0.79

C₆F₄OMe

Zn-6a C₆H₄OnBu

Zn-6b C₆H₄OMe

Zn-6c Mes

Zn-6d C₆H₄Me

Zn-6e C₆H₅

Zn-6f C₆H₄F

Zn-6g C₆H₄CF₃

Zn-6h C₆F₅

2.07

2.08

2.07

2.08

1.44

1.41

1.49

1.48

1.45

1.46

1.62

1.71

1.66

1.43

1.39

1.44

1.42

1.38

1.39

1.37^[d]

1.40

1.04

1.01

1.06

1.04

0.98

1.08

1.03

1.01

–0.63

–0.66

–0.58

–0.59

–0.63

–0.62

–0.45

–0.36

–0.42

–0.64

–0.66

–0.63

–0.65

–0.70

–0.69

–0.68

–0.70

–0.68

–1.03

–1.06

–1.01

–1.03

–1.10

–0.99

–1.04

–1.07

Zn-6i C₆F₄OMe

[a] Including Coulomb term of 0.104 eV. [b] Including Coulomb term of 0.124 eV. [c] Including Coulomb term of 0.057 eV. [d] Calculated

with E(P)_1/2(red¹) = –1.58 eV from reference porphyrin Ac-3h, see ref.^[10]

.

Eur. J. Inorg. Chem. 2014, 1984–2001

1994

www.eurjic.org

FULL PAPER

Table 9. Reductive PET (ΔG_ET= –0.48 eV) to give Q–P₁– scribe the node structure of the orbitals^[24]) and the lowest

P₂^·–-Fc^·+is thermodynamically favoured in 9 over oxidative π* orbital of the quinone located between the HOMOs and

PET (ΔG_ET= –0.26 eV) similarly to 6d and 6e (Tables 8, 9 LUMOs of the porphyrin. Two points are immediately evi-

and Figure 9). This excellently supports the kinetic analysis dent: 1) electron-withdrawing Ar substituents reduce the

of ET in 9 (see above). The shift of charge from P₂to P₁to energy gap between the LUMO of the porphyrin and the

give Q-P₁^·–-P₂-Fc^·+is slightly endergonic (0.09 eV) whereas π* orbital of the quinone (Figure 11), which disfavours oxi-

the following thermal ET to give Q^·–-P₁-P₂-Fc^·+is again dative PET, and 2) the symmetry of the local porphyrin

exergonic by 0.42 eV (Figure 9).

LUMO in 4h (Figure 10, b; LUMO+1; Ar = C₆F₅) is dif-

ferent to that in all the other dyads and features a node at

the meso carbon atoms in the amide direction. This orbital

inversion is typically observed in porphyrins with strongly

electron-withdrawing substituents.^[11,25]Naturally, small or-

bital coefficients at the bridging unit between redox centres

Figure 9. Energy level diagram for Q–P₁–P₂-Fc tetrad 9.

Table 9. Energies of porphyrin S₁and CS states for 9 and driving

forces ΔG_ET

.

State

E [eV]^[a]

r_AD

[Å]

Coulomb term

[eV]

E_P1(S₀ǞS₁)

E_P2(S₀ǞS₁)

1.91

–

E(Q-P₁^·+-P₂^·–-Fc) or

E(Q-P₁^·–-P₂^·+-Fc)^[b,c,d,e]

E(Q^·–-P₁^·+- P₂-Fc)^[b]

E(Q^·–-P₁-P₂^·+-Fc)^[c]

E(Q-P₁-P₂^·–-Fc^·+)^[d]

E(Q-P₁^·–-P₂-Fc^·+)^[e]

E(Q^·–-P₁-P₂-Fc^·+)

2.07/2.08

19.3

0.084

1.65

1.67

1.43

1.52

1.10

15.5

34.8

13.0

32.3

47.8

0.104

0.046

0.124

0.050

0.033

Driving force

ΔG_ET[eV]

ΔG_ET(Q^·–-P₁^·+-P₂-Fc)

ΔG_ET(Q^·–-P₁-P₂^·+-Fc)

ΔG_ET(Q–P₁-P₂^·–-Fc^·+)

ΔG_ET(Q-P₁^·–-P₂-Fc^·+)

ΔG_ET(Q^·–-P₁-P₂-Fc^·+)

–0.26

–0.24

–0.48

–0.39

–0.81

[a] Including Coulomb terms. [b] Calculated with E(P₁)_1/2(ox¹) =

0.550 V from reference porphyrin Ac-3e. [c] Calculated with

E(P₂)_1/2(ox¹) = 0.520 V from reference porphyrin Ac-3d. [d] Calcu-

lated with E(P₁)_1/2(red¹) = –1.630 V from reference porphyrin Ac-

3e. [e] Calculated with E(P₂)_1/2(red¹) = –1.610 V from reference por-

phyrin Ac-3d.

Densitiy Functional Calculations on Ground and CS States

To estimate distances between redox sites (Q–P centre-

to-centre; P centre to Fe, Q centre to Fe, P–P centre-to-

centre) and to visualise frontier molecular orbitals, the

ground states of dyads and triads with electron-donating

and -withdrawing substituents were optimised by DFT

(B3LYP/LANL2DZ, PCM, CH₂Cl₂)^[23]methods (see the

Supporting Information). Figure 10 exemplarily depicts the

frontier molecular orbitals of dyads 4b and 4h showing the

typical Gouterman four orbital scheme for porphyrins (no-

Figure 10. B3LYP/LANL2DZ, PCM-calculated frontier orbitals

menclature adopted from D_4hlocal point symmetry to de- for a) 4b and b) 4h (isosurface value 0.05 a.u.).

Eur. J. Inorg. Chem. 2014, 1984–2001

1995

www.eurjic.org

FULL PAPER

disfavours ET. Hence, PET in 4h is not only thermodynami- (Table 4; for the molecular orbitals of Zn-6, see the Sup-

cally slightly uphill (Figure 6), but probably also kinetically porting Information).

hindered due to the unfavourable orbital symmetry.

Figure 11. B3LYP/LANL2DZ, PCM-calculated molecular orbital

energy diagram for dyads 4.

A similar picture with respect to the quinone π* orbital

arises for the triads 6 (Figure 12) and Zn-6. The porphyrin

LUMO inversion is also observed for 6g (Figure 13) and

6h. In addition, the occupied ferrocene-based δ orbitals

(d_xy, d_x₂_–y₂) are close in energy to the local porphyrin

HOMOs (Figure 12). For electron-rich 6a–6f, the δ(Fc) or-

bitals are below the porphyrin a_2uorbital, in 6g the a_2u

orbital is lowered in energy and mixes with a δ(Fc) orbital

(Figure 13, HOMO) and in 6h the porphyrin a_2uorbital is

located even below the δ(Fc) orbitals. Naturally, this simple

orbital picture cannot account for quantitative relation-

ships, but it clearly demonstrates the lowering of the por-

phyrin HOMOs relative to the essentially constant ferro-

cene HOMOs (Figure 12), which should favour reductive

PET in triads with electron-withdrawing substituents (see

Tables 4 and 8). The near degeneracy of porphyrin and fer-

rocene HOMOs in terms of energy also allows for orbital

mixing (Figure 13) and mixing of states, which should also

allow energy-transfer processes between porphyrin and fer-

rocene. These energy-transfer processes might be assigned

to the second decay (τ₂, A₂) in 6e–6i and Zn-6a–Zn-6i

Figure 13. B3LYP/LANL2DZ, PCM-calculated frontier orbitals

for triad 6g (isosurface value 0.05 a.u.).

Tetrad 9 features a double nearly degenerate set of Gout-

erman orbitals as well as the quinone π* and the occupied

ferrocene d orbitals, as expected from the constituent build-

ing blocks (see the Supporting Information). The porphyrin

HOMOs are of local a_2usymmetry with large coefficients

at the connecting meso carbon atoms. This should allow for

energy and electron transfer between the two porphyrins^[11]

based on orbital symmetry arguments, as suggested in Fig-

ure 9.

For dyad 4b (Ar = 4-C₆H₄OMe) we succeeded in op-

timising the Q^·–-P^·+CS triplet state by DFT methods (Fig-

ure 14, a). The calculated spin density is clearly distributed

over the porphyrin radical cation and quinone radical

anion, as expected from the molecular orbital scheme (Fig-

ure 10). The quinone CO bond lengths were calculated to

be 1.26 Å in the ground state and 1.30 Å in the CS state,

clearly reflecting the semiquinone character. The analogous

DFT optimisation of 4d (Ar = 4-C₆H₄Me) resulted in the

porphyrin triplet state T₁(Figures 6 and 14, b), which sug-

gests a close proximity of the CS (1.65 eV, Table 7) and lo-

cal triplet states in terms of energy. Indeed, the porphyrin

triplet state of H₂TPP has been reported to have an energy

of 1.43 eV.^[3f]

To gain an impression of the spin density and charge

Figure 12. B3LYP/LANL2DZ, PCM-calculated molecular orbital

energy diagram for triads 6.

distribution in the final Q^·–-P-Fc^·+and Q^·–-(Zn)P-Fc^·+CS

Eur. J. Inorg. Chem. 2014, 1984–2001

1996

www.eurjic.org

FULL PAPER

are increased significantly from 1.73 to 1.91 Å. The energy

of the triplet state of unsubstituted ferrocene has been re-

ported to be 1.16 eV,^[10]which is clearly close to the energies

of the Q^·–-P-Fc^·+and Q^·–-(Zn)P-Fc^·+CS states (cf. Table 8).

To artificially stabilise the Q^·–-P-Fc^·+and Q^·–-(Zn)P-Fc^·+

CS states relative to the ferrocene triplet we placed two

Lewis acidic^[26]BF₃molecules at the quinone oxygen atoms,

which allowed optimisation of the CS states of 6e(·2BF₃)

and Zn-6e(·2BF₃) as the lowest-energy triplet states (Fig-

ure 15, b,c). The calculated spin densities are clearly at both

the ferrocenium and semiquinonato radicals. Again, the

quinone CO bond lengths are elongated from 1.26 to

1.30 Å. Furthermore, the Fe···Cp(centroid) distances are

elongated from 1.73 to 1.80 Å, which indicates the ferrocen-

ium state.

Transient Absorption Spectroscopy of Selected Compounds

Compared with the porphyrin bands, the absorptions of

the semiquinone radical anion Q^·–and ferrocenium cation

Fc^·+are rather weak. Chemical reduction of the reference

N-ethylanthraquinone-2-carboxylic acid amide with deca-

methylcobaltocene [CoCp*₂]^[27](see the Supporting Infor-

mation) generated the semiquinone radical anion Q^·–with

an absorption maximum at 618 nm and ε₆₁₈= 750 m^–1cm^–1

(THF). The ferrocenium reference [Fc-NHAc]⁺absorbs at

759 nm with ε₇₅₉= 350 m^–1cm^–1(CH₂Cl₂).^[28]Owing to the

low extinction coefficients, that is, small absorption cross-

sections, these bands are difficult to detect in transient ab-

sorption (TA) pump-probe experiments as they are over-

whelmed by the excited-state absorptions of the strongly ab-

sorbing porphyrin excited states. Hence all transient absorp-

tion spectra are largely dominated by features of the por-

phyrin radical(s) and porphyrin triplet states.

Figure 14. DFT-calculated spin density of a) the Q^·–-P^·+CS triplet

state of 4b and b) the porphyrin triplet state of 4d (isosurface value

0.01 a.u.).

states of the triads we tried to optimise this (triplet) state,

but all attempts yielded only the optimised ferrocene triplet

state with a Mulliken spin density of 2.03 at the iron centre

(Figure 15, a). The calculated Fe···Cp(centroid) distances

The pico- to nanosecond TA spectra of compounds 4a

and 6a in THF are presented in Figure 16. The spectra

show multiple peaked features that correspond to ground-

Figure 15. DFT-calculated spin densities of a) the ferrocene triplet

state of 6e, b) the Q^·–-P-Fc^·+triplet CS state of 6e(·2BF₃) and c) the

Q^·–-(Zn)P-Fc^·+triplet CS state of Zn-6e(·2BF₃) (isosurface value

0.01 a.u.).

Figure 16. ps–ns transient absorption spectra of a) 4a and b) 6a in

THF after excitation at λ_exc= 420 nm. Note the significantly faster

decay of the TA signal for compound 6a.

Eur. J. Inorg. Chem. 2014, 1984–2001

1997

www.eurjic.org

FULL PAPER

state bleaching and stimulated emission overlapped by the of the reference porphyrin Ac-3a, we obtained inverse rate

excited-state absorption of the porphyrin. The spectra do constants of 12 ns and 7 μs.

not change much during the observed period of time, except

for compound 4a, for which there is an increase in the in-

tensity of the photoinduced absorption in the spectral re-

gion between 750–900 nm. This spectral evolution may be

related to electron transfer from the porphyrin to the quin-

one, but a clear assignment is difficult.

As a consequence of the superposition of several bands

corresponding to porphyrin radical ions, namely the ex-

cited-state absorption, the ground-state bleaching and

stimulated emission bands, and the low extinction coeffi-

cients of the Q^·–and Fc^·+radicals, the transient spectra of

intermediate CS states are not observed for dyad 4a and

triad 6a. However, very clearly a much more rapid decay of

the entire TA spectrum is observed for the ferrocenyl deriv-

ative 6a. The decay is well fitted by a single exponential

with an inverse rate of 220 ps, which is essentially the same

as the 215 ps determined for the fluorescence lifetime of 6a.

This fast decay is likely due to the fast reductive quenching

of the porphyrin excited state by the ferrocene, parallel oxi-

dative quenching by the quinone (Figure 7) and/or heavy

atom quenching by the ferrocene to give the porphyrin trip-

let state. For the two PET pathways, porphyrin radical

anions and radical cations could be simultaneously present

and would create superimposed excited-state absorption

spectra that are difficult to analyse. After charge transfer

Figure 17. ns–μs transient absorption dynamics of a) Ac-3a, b) 4a,

c) 4a plus 100 equiv. B(C₆F₅)₃and d) 6a in THF solution after exci-

from P^·–to Q or from Fc to P^·+, the final CS state should

tation at λ_exc= 532 nm. The dynamics were monitored in the spec-

tral region 550–620 nm. The solid red lines represent exponential

fits to the data and the inverse decay rates are also shown. Note the

significantly slower TA signal decay for compound 4a plus additive

indicating the stabilisation of the CS state by the Lewis acid.

show no features arising from porphyrin, but only weak

bands corresponding to Q^·–and Fc^·+radicals. These latter

would be difficult to detect, especially if the final CS state

is only formed in low yields. However, the significant de-

crease in the TA signal intensity on the nanosecond times-

cale observed for 6a in comparison with 4a indicates that

The nanosecond lifetime τ₁corresponds approximately

in the former a large fraction of the porphyrin excited states to the measured fluorescence lifetime of around 10 ns

are rapidly quenched and that the ground state of the por- within the error that can be expected between the two dif-

phyrin in 6a is quickly recovered after photoexcitation. This ferent experimental techniques. The microsecond lifetime τ₂

is in line with the experimentally observed faster photolu- can be assigned to the decay of the porphyrin triplet T₁

minescence decay of 6a, the much lower fluorescence quan- state generated by intersystem crossing from the porphyrin

tum yield and the predicted higher driving force for photo- singlet S₁state due to its significantly longer lifetime in the

induced electron transfer from the ferrocene to the por- range typical for the decay of triplets. Fitting the signal dy-

phyrin in comparison with electron transfer from the por- namics of 4a yields inverse rate constants of 5.7 ns and

phyrin to the quinone moiety, all of which indicate that the 1.7 μs. The former is shorter than the lifetime observed for

charge-transfer process from ferrocenium to the porphyrin the reference porphyrin Ac-3a, indicating additional decay

is efficient.

channels, but longer than the observed fluorescence lifetime

As the pico- to nanosecond TA spectra are rather diffi- of compound 4a (in CH₂Cl₂), which might be a solvent ef-

cult to interpret due to the multitude of features and poten- fect. However, in this case we most probably approached

tially parallel photophysical processes, we had a closer look the temporal resolution of our ns–μs TA set-up, which is

at the nano- to microsecond spectra and dynamics. On the about several nanoseconds. The second longer-lived compo-

nanosecond timescale, the shapes of the transient absorp- nent has a considerably shorter lifetime than the porphyrin

tion spectra of Ac-3a, 4a and 6a are rather similar (see the triplet-state lifetime observed for compound Ac-3a and thus

Supporting Information). The decay of the TA signals on appears to originate from a state other than the T₁triplet.

this timescale should largely reflect the recombination dy- Given that the efficiency of ET obtained by analysis of the

namics of the triplet and CS states in these compounds. In fluorescence quenching is Φ_ET= 0.93, we assign this com-

fact, the dynamics of the TA signals of all three compounds ponent to the decay of the Q^·–-P^·+CS state, which is formed

Ac-3a, 4a and 6a can be well described by the sum of two in high yield in the case of compound 4a. The dynamics of

exponentials with a nano- and microsecond component in- the TA signals of compound 6a can be fitted with inverse

dicating two processes, as shown in Figure 17. In the case rates of τ₁= 3 ns and τ₂= 6.8 μs. Although the former is

Eur. J. Inorg. Chem. 2014, 1984–2001

1998

www.eurjic.org

FULL PAPER

certainly limited by the temporal resolution of our set-up by electron-withdrawing meso substituents Ar [k_ET

=

and represents the sub-nanosecond dynamics of the PET 4ϫ10⁹to 38ϫ10⁹s^–1; Δ(ΔG_ET) = –0.2 eV]. Comparable

processes, the latter component is close to the triplet life- rates and driving forces for reductive and oxidative quench-

time observed for compound Ac-3a. This appears reason- ing are found for Ar = 4-C₆H₄OnBu, 4-C₆H₄OMe and 4-

able as in this compound no features of porphyrin radicals C₆H₄Me in triads 6. The zinc porphyrins Zn-6 are basically

should be observed from the final Q^·–-P-Fc^·+CS state after non-fluorescent. If the major deactivation in zinc conju-

electron transfer to Q and from Fc are complete. Therefore gates indeed occurs by PET, both pathways feature similar

we observe the decay of a fraction of porphyrin triplet states driving forces for Ar = 4-C₆H₄OnBu, 4-C₆H₄OMe, Mes, 4-

created either by ISC from the porphyrin singlet state di- C₆H₄Me and C₆H₅. In tetrad 9, both pathways are also

rectly after photoexcitation or, which is more likely, by a feasible, although reductive PET (ΔG_ET= –0.48 eV; k_ET

=

transition of the initial Q-P^·–-Fc^·+CS state (1.46 eV, Table 8 2.66ϫ10⁹s^–1) is preferred over oxidative PET (ΔG_ET

and Figure 7) into the T₁state of the porphyrin (1.43 eV^[3f]). –0.26 eV; k_ET= 0.363ϫ10⁹s^–1). With these preparative,

As suggested by the DFT calculations, a CS state involv- spectroscopic, electrochemical and theoretical results in

ing the quinone can be stabilised by coordinating a Lewis hand, more complex architectures with well-defined ex-

acid to the quinone carbonyl groups. Excess tris(penta- cited-state energies in a structurally precisely organised

fluorophenyl)boron was added to 4a and 6a in THF and manner can be envisaged.

the ps–ns as well as ns–μs transient absorption spectra and

Supporting Information (see footnote on the first page of this arti-

dynamics were studied (see Figure 17 and the Supporting

cle): Experimental and analytical details, normalised absorption

Information). Some spectral but mostly temporal differ-

and emission spectra of selected Q-P dyads, Q-P-Fc triads and ref-

ences were observed with respect to the spectra and dynam-

ics of 4a and 6a in the absence of the Lewis acid. The most

striking observation is a pronounced decrease in the decay

rate, in other words, increase in the lifetime of the Q^·–-P^·+

CS state observed for compound 4a. In fact, the ns–μs TA

experiments showed that the lifetime of the CS state is sig-

nificantly increased from τ₂= 1.7 μs to τ₂= 80 μs upon

addition of B(C₆F₅)₃(cf. Figure 17, b,c). This suggests that

the Q^·–-P^·+CS state is significantly stabilised by the Lewis

acid, as suggested by the DFT calculations, and further

supports our assignment of the dynamics to the recombina-

tion of the charge-separated state. Although the effect is

pronounced for 4a, it is much less obvious for 6a. This ap-

pears reasonable as we assigned the dynamics in the latter

to the recombination of the porphyrin triplet state, which

should not be affected by the presence of the Lewis acid.

erence porphyrins in CH₂Cl₂n DFT (B3LYP/LANL2DZ, PCM)

calculated frontier orbitals of Zn-6b, Zn-6g, Zn-6h and 9, absorp-

tion spectra of N-ethylanthraquinone-2-carboxamide treated with

CoCp*₂in THF; ns to μs transient absorption spectra of Ac-3a,

4a, 4a + B(C₆F₅)₃, 6a and 6a + B(C₆F₅)₃in THF; after excitation

at λ_exc= 532 nm, energies (eV) of relevant frontier molecular orbit-

als of the Q-P dyads 4a–4h determined by DFT (B3LYP/

LANL2DZ, PCM), energies (eV) of relevant frontier molecular or-

bitals of the Q-P-Fc triads 6a–6i and Q-P-Fc triads Zn-6a–Zn-6i

determined by DFT (B3LYP/LANL2DZ, PCM), Cartesian coordi-

nates of DFT-optimised geometries.

Acknowledgments

This work was supported by the Deutsche Forschungsgemeinschaft

(DFG) (He 2778-6/1). J. R. O. and M. M. thank the Max Planck

Graduate Center for support and F. L. (all Max Planck Institute

for Polymer Research of Mainz, Germany) thanks the Max Planck

Society for funding the Max Planck Research group.

Conclusions

[1] a) M. G. Rockley, M. W. Windsor, R. J. Cogdellt, W. W. Par-

sont, H. Neurath, Proc. Natl. Acad. Sci. USA 1975, 72, 2251–

2255; b) K. J. Kaufmann, P. L. Dutton, T. L. Netzel, J. S.

Leigh, P. M. Rentzepis, Science 1975, 188, 1301–1304; c) J. J.

Katz, J. R. Norris, L. L. Shipman, M. C. Thurnauer, M. R.

Wasielewski, Ann. Rev. Biophys. Bioeng. 1978, 7, 393–434; d) J.

Deisenhofer, O. Epp, K. Miki, R. Huber, H. Michel, J. Mol.

Biol. 1984, 180, 385–398.

[2] a) M. R. Wasielewski, Chem. Rev. 1992, 92, 435–461; b) A.

Harriman, J.-P. Sauvage, Chem. Soc. Rev. 1996, 25, 41–48; c)

D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2001, 34,

40–48; d) L. Sun, L. Hammarström, B. Åkermark, S. Styring,

Chem. Soc. Rev. 2001, 30, 36–49; e) S. Fukuzumi, Org. Biomol.

Chem. 2003, 1, 609–620.

[3] a) A. R. McIntosh, A. Siemiarczuk, J. R. Bolton, M. J. Still-

man, T.-F. Ho, A. C. Weedon, J. Am. Chem. Soc. 1983, 105,

7215–7223; b) A. Siemiarczuk, A. R. McIntosh, T.-F. Ho, M. J.

Stillman, K. J. Roach, A. C. Weedon, J. R. Bolton, J. S. Con-

nolly, J. Am. Chem. Soc. 1983, 105, 7224–7230; c) M. R. Wasie-

lewski, M. P. Niemczyk, W. A. Svec, E. B. Pewitt, J. Am. Chem.

Soc. 1985, 107, 1080–1082; d) D. Gust, T. A. Moore, P. A. Lid-

dell, G. A. Nemeth, L. R. Makings, A. L. Moore, D. Barrett,

P. J. Pessiki, R. V. Bemasson, M. Rougée, C. Chachaty, F. C.

Anthraquinone–porphyrin dyads (Q-P, 4), anthraqui-

none–porphyrin–ferrocene triads (Q-P-Fc, 6) and zincated

triads [Q-(Zn)P-Fc, Zn-6] with meso substituents Ar of in-

creasing electron-withdrawing character in the porphyrin

component have been designed and synthesised by amide

coupling (Ar

C₆H₄Me, C₆H₅, 4-C₆H₄F, 4-C₆H₄CF₃and C₆F₅). Based on

this modular amide-coupling approach,

Q-P^C6H5

= 4-C₆H₄OnBu, 4-C₆H₄OMe, Mes, 4-

a

-

P^C6H4Me-Fc tetrad (9) was constructed. In the ground states

of 4, 6, Zn-6 and 9, only marginal interactions between the

individual building blocks Q, P and Fc are observed, and

energy transfer between the two porphyrins in 9 is possible.

The fluorescence of dyads 4 is quenched by oxidative pho-

toinduced electron transfer (PET). The rate of ET increases

by several orders of magnitude with increasing electron-do-

nating power of the meso substituent Ar (k_ET= 0.003ϫ10⁹

to 1.5ϫ10⁹s^–1) due to the increased driving force Δ(ΔG_ET

)

= –0.3 eV. In triads 6 an additional reductive PET pathway

is thermodynamically feasible. This ET pathway is favoured

Eur. J. Inorg. Chem. 2014, 1984–2001

1999

www.eurjic.org

FULL PAPER

De Schryver, M. Van der Auweraer, A. R. Holzwarth, J. S.

Connolly, J. Am. Chem. Soc. 1987, 109, 846–856; e) A. D.

Joran, B. A. Leland, P. M. Felker, A. H. Zewail, J. J. Hopfield,

P. B. Dervan, Nature 1987, 327, 508–511; f) J. A. Schmidt,

A. R. McIntosh, A. C. Weedon, J. R. Bolton, J. S. Connolly,

J. K. Hurley, M. R. Wasielewski, J. Am. Chem. Soc. 1988, 110,

1733–1740; g) J. von Gersdorff, M. Huber, H. Schubert, D. Ni-

ethammer, B. Kirste, M. Plato, K. Möbius, H. Kurreck, R.

Eichberger, R. Kietzmann, F. Willig, Angew. Chem. Int. Ed.

Engl. 1990, 29, 670–672; Angew. Chem. 1990, 102, 690–692; h)

J. Zimmermann, J. Von Gersdorff, H. Kurreck, B. Ridder, J.

Photochem. Photobiol. B: Biology 1997, 40, 209–217; i) J. P. Su-

mida, P. A. Liddell, S. Lin, A. N. Macpherson, G. R. Seely,

A. L. Moore, T. A. Moore, D. Gust, J. Phys. Chem. A 1998,

102, 5512–5519; j) O. Korth, A. Wiehe, H. Kurreck, B. Röder,

Chem. Phys. 1999, 246, 363–372; k) Y. K. Kang, I. V. Rubtsov,

P. M. Iovine, J. Chen, M. J. Therien, J. Am. Chem. Soc. 2002,

124, 8275–8279; l) J. Springer, G. Kodis, L. De Garza, A. L.

Moore, T. A. Moore, D. Gust, J. Phys. Chem. A 2003, 107,

3567–3575; m) P. P. Kumar, G. Premaladha, B. G. Maiya, J.

Chem. Sci. 2005, 117, 193–201.

a) A. M. Brun, A. Harriman, V. Heitz, J.-P. Sauvage, J. Am.

Chem. Soc. 1991, 113, 8657–8663; b) V. Heitz, S. Chardon-

Noblat, J.-P. Sauvage, Tetrahedron Lett. 1991, 32, 197–198; c)

E. K. L. Yeow, P. J. Sintic, N. M. Cabral, J. N. H. Reek, M. J.

Crossley, K. P. Ghiggino, Phys. Chem. Chem. Phys. 2000, 2,

4281–4291; d) K. Kilså, J. Kajanus, A. N. Macpherson, J.

Mårtensson, B. Albinsson, J. Am. Chem. Soc. 2001, 123, 3069–

3080; e) S. Fukuzumi, K. Ohkubo, W. E. Z. Ou, J. Shao, K. M.

Kadish, J. A. Hutchison, K. P. Ghiggino, P. J. Sintic, M. J.

Crossley, J. Am. Chem. Soc. 2003, 125, 14984–14985; f) J. H.

Jang, H. J. Kim, H.-J. Kim, C. H. Kim, T. Joo, D. Won Cho,

M. Yoon, Bull. Korean Chem. Soc. 2007, 28, 1967–1972; g) J.

Fortage, J. Boixel, E. Blart, L. Hammarström, H. C. Becker, F.

Odobel, Chem. Eur. J. 2008, 14, 3467–3480; h) J. Fortage, A.

Scarpaci, L. Viau, Y. Pellegrin, E. Blart, M. Falkenstrçm, L.

Hammarström, I. Asselberghs, R. Kellens, W. Libaers, K.

Clays, M. P. Eng, F. Odobel, Chem. Eur. J. 2009, 15, 9058–

9067; i) J. Fortage, J. Boixel, E. Blart, H. C. Becker, F. Odobel,

Inorg. Chem. 2009, 48, 518–526.

a) D. Kuciauskas, S. Lin, G. R. Seely, A. L. Moore, T. A.

Moore, T. Drovetskaya, C. A. Reed, P. D. W. Boyd, J. Phys.

Chem. 1996, 100, 15926–15932; b) C. Luo, D. M. Guldi, H.

Imahori, K. Tamaki, Y. Sakata, J. Am. Chem. Soc. 2000, 122,

6535–6551; c) H. Imahori, N. V. Tkachenko, V. Vehmanen, K.

Tamaki, H. Lemmetyinen, Y. Sakata, S. Fukuzumi, J. Phys.

Chem. A 2001, 105, 1750–1756; d) V. Vehmanen, N. V. Tkach-

enko, H. Imahori, S. Fukuzumi, H. Lemmetyinen, Spectro-

chim. Acta Part A 2001, 57, 2229–2244; e) H. Imahori, D. M.

Guldi, K. Tamaki, Y. Yoshida, C. Luo, Y. Sakata, S. Fuku-

zumi, J. Am. Chem. Soc. 2001, 123, 6617–6628; f) H. Imahori,

K. Tamaki, Y. Araki, Y. Sekiguchi, O. Ito, Y. Sakata, S. Fuku-

zumi, J. Am. Chem. Soc. 2002, 124, 5165–5174; g) H. Imahori,

Y. Sekiguchi, Y. Kashiwagi, T. Sato, Y. Araki, O. Ito, H. Yam-

ada, S. Fukuzumi, Chem. Eur. J. 2004, 10, 3184–3196; h) M. U.

Winters, E. Dahlstedt, H. E. Blades, C. J. Wilson, M. J.

Frampton, H. L. Anderson, B. Albinsson, J. Am. Chem. Soc.

2007, 129, 4291–4297; i) C. Wijesinghe, M. E. El-Khouly, M. E.

Zandler, S. Fukuzumi, F. DЈSouza, Chem. Eur. J. 2013, 19,

9629–9638.

a) D. Gust, T. A. Moore, A. L. Moore, F. Cao, D. Luttrul,

J. M. Degraziano, X. C. Ma, L. R. Makings, S. Lee, T. T. Trier,

E. Bittersmann, G. R. Seely, S. Woodward, R. V. Bensasson,

M. Roug, F. C. De Schryver, M. Van der Auweraer, J. Am.

Chem. Soc. 1991, 113, 3638–3649; b) D. Gust, T. A. Moore,

A. L. Moore, A. N. Macpherson, A. Lopez, J. M. DeGraziano,

I. Gouni, E. Bittersmann, G. R. Seely, F. Gao, R. A. Nieman,

X. C. Ma, L. J. Demanche, S. Hung, D. K. Luttrull, S. Lee,

P. K. Kerrigan, J. Am. Chem. Soc. 1993, 115, 11141–11152; c)

S. L. Gould, G. Kodis, R. E. Palacios, L. De La Garza, A.

Brune, D. Gust, T. A. Moore, A. L. Moore, J. Phys. Chem. B

2004, 108, 10566–10580.

[7] a) R. Ciasson, E. J. Lee, X. Zbao, M. S. Wrighton, J. Phys.

Chem. 1993, 97, 2596–2601; b) M. Kubo, Y. Mori, M. Otani,

M. Murakami, Y. Ishibashi, M. Yasuda, K. Hosomizu, H. Mi-

yasaka, H. Imahori, S. Nakashima, Chem. Phys. Lett. 2006,

429, 91–96; c) H. Mansour, M. E. El-khouly, S. Y. Shaban, O.

Ito, N. Jux, J. Porphyrins Phthalocyanines 2007, 11, 719–728;

d) V. S. Shetti, M. Ravikanth, Eur. J. Org. Chem. 2010, 494–

508.

[8]

a) R. A. Marcus, J. Chem. Phys. 1956, 24, 979–989; b) R. A.

Marcus, Annu. Rev. Phys. Chem. 1964, 15, 155–196; c) R. A.

Marcus, Angew. Chem. Int. Ed. Engl. 1993, 32, 1111–1121; An-

gew. Chem. 1993, 105, 1161–1172.

[9]

a) S. Fukuzumi, K. Ohkubo, W. E. Z. Ou, J. Shao, K. M. Kad-

ish, J. Am. Chem. Soc. 2003, 125, 14984–14985; b) M. Linke-

Schaetzel, C. E. Anson, A. K. Powell, G. Buth, E. Palomares,

J. D. Durrant, T. Silviu Balaban, J.-M. Lehn, Chem. Eur. J.

2006, 12, 1931–1940; c) B. M. J. M. Suijkerbuijk, R. J. M.

Klein Gebbink, Angew. Chem. 2008, 120, 7506–7532; d) J. For-

tage, J. Boixel, E. Blart, L. Hammarström, Chem. Eur. J. 2008,

14, 3467–3480; e) K. Kiyosawa, N. Shiraishi, T. Shimada, D.

Masui, H. Tachibana, J. Phys. Chem. C 2009, 113, 11667–

11673; f) J. Fortage, A. Scarpaci, L. Viau, Y. Pellegrin, E. Blart,

M. Falkenström, L. Hammarström, I. Asselberghs, R. Kellens,

W. Libaers, K. Clays, M. P. Eng, F. Odobel, Chem. Eur. J. 2009,

15, 9058–9067; g) A. C. Benniston, A. Harriman, P. Li, J. Am.

Chem. Soc. 2010, 132, 26–27; h) H. Kon, T. Nagata, Chem.

Eur. J. 2012, 18, 1781–1788.

[4]

[10]

[11]

[12]

Y. Araki, Y. Yasumura, O. Ito, J. Phys. Chem. B 2005, 109,

9843–9848.

J. Melomedov, A. Wünsche Von Leupoldt, M. Meister, F. La-

quai, K. Heinze, Dalton Trans. 2013, 42, 9727–9739.

a) A. Breivogel, K. Hempel, K. Heinze, Inorg. Chim. Acta 2011,

374, 152–162; b) D. Siebler, M. Linseis, T. Gasi, L. M. Carrella,

R. F. Winter, C. Förster, K. Heinze, Chem. Eur. J. 2011, 17,

4540–4551; c) J. Dietrich, U. Thorenz, C. Förster, K. Heinze,

Inorg. Chem. 2013, 52, 1248–1264; d) J. Dietrich, A.

Wünsche von Leupoldt, M. Grabolle, U. Resch-Genger, K.

Heinze, Eur. J. Inorg. Chem. 2013, 3009–3019.

a) K. Heinze, U. Wild, M. Beckmann, Eur. J. Inorg. Chem.

2007, 617–623; b) K. Heinze, M. Beckmann, K. Hempel,

Chem. Eur. J. 2008, 14, 9468–9480; c) K. Heinze, K. Hempel,

Chem. Eur. J. 2009, 15, 1346–1358; d) G. Dirscherl, B. König,

Eur. J. Org. Chem. 2008, 597–634; e) D. Siebler, C. Förster, K.

Heinze, Dalton Trans. 2011, 40, 3558–3575.

K. Heinze, A. Reinhart, Dalton Trans. 2008, 469–480.

a) B. Bildstein, M. Malaun, H. Kopacka, K. Wurst, M. Mit-

terböck, K. Ongania, G. Opromolla, P. Zanello, Organometal-

lics 1999, 18, 4325–4336; b) K. Heinze, M. Schlenker, Eur. J.

Inorg. Chem. 2004, 2974–2988.

a) P. D. Rao, S. Dhanalekshmi, B. J. Littler, J. S. Lindsey, J.

Org. Chem. 2000, 65, 7323–7344; b) A. K. Burrell, D. L. Offi-

cer, P. G. Plieger, D. C. W. Reid, Chem. Rev. 2001, 101, 2751–

2796; c) J. S. Lindsey, Acc. Chem. Res. 2010, 43, 300–311; d)

M. O. Senge, Chem. Commun. 2011, 47, 1943–1960.

C. M. Carcel, J. K. Laha, R. S. Loewe, P. Thamyongkit, K.

Schweikart, V. Misra, D. F. Bocian, J. S. Lindsey, N. Carolina,

J. Org. Chem. 2004, 69, 6739–6750.

[13]

[5]

[14]

[15]

[16]

[17]

[18]

a) B. J. Littler, Y. Ciringh, J. S. Lindsey, J. Org. Chem. 1999, 64,

2864–2872; b) K. Heinze, A. Reinhart, Z. Naturforsch. B 2005,

60, 758–762; c) T. Rohand, E. Dolusic, T. H. Ngo, W. Maes,

W. Dehaen, ARKIVOC 2007, x, 307–324.

K. Phukan, M. Ganguly, N. Devi, Synth. Commun. 2009, 39,

2694–2701.

S. B. Colbran, S. T. Lee, D. G. Lonnon, F. J. D. Maharaj, A. M.

McDonagh, K. A. Walker, R. D. Young, Organometallics 2006,

25, 2216–2224.

P. Heier, C. Förster, D. Schollmeyer, N. Boscher, P. Choquet,

K. Heinze, Dalton Trans. 2013, 42, 906–917.

[6]

[19]

[20]

[21]

Eur. J. Inorg. Chem. 2014, 1984–2001

2000

www.eurjic.org

FULL PAPER

[22] A. Weller, Z. Phys. Chem. (Muenchen Ger.) 1982, 133, 93–98.

[23] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,

M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.

Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,

H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-

nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-

gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,

T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.

Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-

erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,

J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.

Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-

amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,

A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-

tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,

J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.

Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, re-

vision A.02, Gaussian, Inc., Wallingford, CT, 2009.

M. Gouterman, J. Chem. Phys. 1959, 30, 1139–1161.

D. Holten, D. F. Bocian, J. S. Lindsey, Acc. Chem. Res. 2002,

35, 57–69.

For the use of Lewis acidic metal ions to stabilise a CS state,

see: a) S. Fukuzumi, K. Okamoto, H. Imahori, Angew. Chem.

Int. Ed. 2002, 41, 620–622; Angew. Chem. 2002, 114, 642–644;

b) S. Fukuzumi, K. Okamoto, Y. Yoshida, H. Imahori, Y.

Araki, O. Ito, J. Am. Chem. Soc. 2003, 125, 1007–1013.

[24]

[25]

[26]

[27] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.

[28] D. Siebler, C. Förster, K. Heinze, Eur. J. Inorg. Chem. 2010,

3986–3992.

Received: January 28, 2014

Published Online: February 27, 2014

Eur. J. Inorg. Chem. 2014, 1984–2001

2001

Article Doi

Tuning reductive and oxidative photoinduced electron transfer in amide-linked anthraquinone-porphyrin-ferrocene architectures

DOI: 10.1002/ejic.201400118

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.201400118

Full text of DOI:10.1002/ejic.201400118

Products guided by the article

R&D Labs maybe for 1586038-68-7

Relevant to this article

Hot Product