Chemical-Driven Reconfigurable Arithmetic Functionalities within a

J. Phys. Chem. C 2008, 112, 16973–16983

16973

Chemical-Driven Reconfigurable Arithmetic Functionalities within a Fluorescent Tetrathiafulvalene Derivative Wei Sun, Chun-Hu Xu, Zhi Zhu, Chen-Jie Fang, and Chun-Hua Yan* Beijing National Laboratory for Molecular Sciences, State Key Laboratory of Rare Earth Materials Chemistry and Applications, and Peking UniVersity-The UniVersity of Hong Kong Joint Laboratory in Rare Earth Materials and Bioinorganic Chemistry, Peking UniVersity, Beijing 100871, China ReceiVed: May 26, 2008; In Final Form: July 31, 2008

The charge transfer phenomena within a tetrathiafulvalene-derived fluorescent switch, 2-[4-(2,2′-bi-1,3-dithiol4-yl)-5-methoxy-1,3-thiazol-2-yl]-2-pyridine (2-MT) is investigated via various spectral and theoretical approaches at both the ground state and the excited state. At different oxidation states of 2-MT, the charge transfer behaviors between the redox switch, tetrathiafulvalene (TTF), and the fluorophore, 5-methoxy-2-(2pyridyl)thiazole (2-MPT), can be tuned differently. When a set of chemical inputs are introduced, the redox reactions coupled with coordination and protonation reactions trigger distinct charge-transfer processes within the ground state and the excited state, which are evidenced in both the absorption and fluorescent spectrum. The chemical-driven spectral changes thus provide parallel output channels to construct reconfigurable moleculebased binary algebra functionalities, that is, the half adder and the half subtractor. The charge-transfer induced fluorescence recovery of 2-MT is also sensitive to the input sequence, which implements a molecular keypad lock within the TTF derivative for the first time. Introduction

CHART 1: Molecular Structure of 2-MT

Exploring computing devices at nano- or subnanoscale is attracting great interest from chemists, physicists, and biologists during the past decades due to the increasing demand for stronger computing power in low dimensions.1 Since the first report of a molecular AND logic gate by de Silva in 1993,2a chemical reactions have also been encoded with computing task.2 Until now, the 2-input Boolean logic gates,3 moleculators,4 and even password-authentication devices5 have been constructed at the molecular level. However, unlike the homogeneous input and output signals in electronic devices, the chemical or irradiation input signals and the spectral output signals of molecular logic systems are heterogeneous, adding difficulties for integrating the chemical-driven independent fundamental logic gates into a functional computing circuit.1f As an example, more than 20 molecular or supramolecular systems have been reported with either half-adder or halfsubtractor functionality, but only a few examples among them can execute both addition and subtraction operations. Therefore, improving the integrity of logic functionalities inside a single molecule is still a challenge for chemists. One solution toward this challenge is to explore molecular switches with multiple spectral states, so that the logic functionalities can be assembled and reconfigured at the molecular level without the increase of processing unit dimension. The nonparallel spectral changes triggered by the same input set, which are utilizing absorption difference at distinct wavelengths, the sum or ratio of fluorescent intensity at different emission wavelengths, or different changing trends between absorption and fluorescence spectrum, provide various output channels for multiple logic functionalities, and enable the reconfigurable capacity for molecular logic circuits; whereas the switch of chemical-encoded input signals can also produce a distinct * To whom correspondence should be addressed. E-mail: [email protected]. Fax: +86-10-62754179. Tel: +86-10-62754179.

spectral response in a preset output channel, which can also reconfigure the logic functionalities within a unique molecular platform.2e,3d,e,i For the moleculators, the half adder and the half subtractor are constructed within a three-state switch, whereas a reconfigurable arithmetic calculator must possess at least four spectral states in response to different input signals. The binarylogic-encoded spectral changes, either in absorption or in fluorescence, arise from the charge transfer among molecular orbitals. When chemical input is introduced to bind with the photoactive substrate, several modified molecular orbitals will afford distinct charge transfer processes at either the ground state or the excited state compared with those in the initial state. Notably, the charge transfer in the ground state, as shown in the absorption spectrum, may not be synchronous with that in the excited state, as evidenced in the emission spectrum, thus endowing the molecular switch with multiple output states.3b,4a Following this strategy, a molecular fluorescent switch with rich charge-transfer phenomena is a potential building block for the construction of reconfigurable logic functionalities. Tetrathiafulvalene (TTF) and its derivatives have been widely researched in the field of organic conductors, superconductors, sensors, and motors because of their novel redox-controlled charge-transfer behaviors.6 Neutral TTF can be reversibly oxidized to the cationic radical and dicationic state by chemical or electrochemical oxidation, resulting in the change of frontier orbital occupancy. This feature promotes the characteristic intramolecular charge transfer (ICT) phenomena of the oxidized TTF derivatives at the ground state. In the TTF derivatives that

10.1021/jp804628u CCC: $40.75  2008 American Chemical Society Published on Web 10/04/2008

16974 J. Phys. Chem. C, Vol. 112, No. 43, 2008

Sun et al.

Figure 1. The redox processes for 2-MT (left) and cyclic voltammogram (CV) and differential voltammogram (DPV) of 2-MT in acetonitrile (0.1 mM, potential is recorded versus Ag/AgCl, right).

TABLE 1: Two-Step Oxidation Potentials (Versus Ag/AgCl) of 2-MT in Different Solvents E1/2ox1/V E1/2ox2/V ∆E/V

dichloromethane

acetonitrile

acetone

tetrahydrofuran

0.317 0.877 0.560

0.383 0.791 0.408

0.462 0.821 0.359

0.551 0.858 0.307

bear a fluorophore, the TTF-centered oxidation also inhibits the photoinduced electron transfer (PET) from TTF skeleton to fluorophore.7 Thus, using either an ICT or PET pathway, Zhang and our group have successfully utilized a TTF derivative as building block for different logic operations.8 Furthermore, the rich charge transfer properties within TTF derivatives also enable us to explore potential for the reconfigurable arithmetic functionalities. We here present the implementation of reconfigurable arithmetic functionalities, including both a half adder and a half subtractor, inside a simple TTF-derived fluorescent switch, 2-[4(2,2′-bi-1,3-dithiol-4-yl)-5-methoxy-1,3-thiazol-2-yl]-2-pyridine (2-MT) bearing a fluorophore 5-methoxy-2-(2-pyridyl)thiazole (2-MPT, Chart 1). The integration of the arithmetic functionalities is achieved through the utilization of different responses to the oxidation coupled with coordination reactions in the parallel ICT and PET pathways. The charge-transfer sensitive fluorescent recovery of 2-MT also composes a molecular keypad lock to authenticate different password entries. Experiment Section 2-MT was prepared following the published Stille coupling procedure.8c Other chemicals were all purchased from commercial availability and used without further purification. The solvents used for spectral characterization are HPLC grade. No impurity was found in the interested wavelength region in either

absorption or fluorescence spectra. The concentrations of the substrate and chemical inputs are listed in the footnotes of corresponding figures. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) were performed on a CHI-840 electrochemical work station in a three-electrode cell equipped with a platinum disk work electrode, a platinum wire counter-electrode, and a Ag/ AgCl reference electrode. Solvents are dry and oxygen-free prior to use. Bu4NPF6 (0.1 M) is used as the support electrolyte. All spectral characterizations were executed at room temperature within a 10 mm quartz cell. UV-vis absorption spectra were measured with a Shimadzu UV-3100 spectrometer, and the fluorescence spectra were recorded upon the excitation at 350 nm on a Hitachi F-4500 fluorescence spectrometer. The fluorescence lifetime was obtained from an Edinburgh LifeSpecRed ps fluorescence lifetime spectrometer. The fluorescent quantum yields were determined following eq 19 with Rhodamine B ethanol solution as the reference. Φfr and Φ are the quantum yield of the reference and the test sample, respectively; Ar and A are the absorbance at the excitation wavelength of the reference and the test sample, respectively; Lr and L are the light path length in the absorption cells of the reference and the test sample, respectively; Nr and N are the indices of refraction of the solvents in the reference sample and the test sample, respectively; and Dr and D are the integrated areas of the emission peaks of the reference and the test sample, respectively. The sample and reference were prepared with the absorbance of 0.1 at the excitation wavelength.

Φ ) Φfr ×

1 - 10-ArLr N2 D × × 1 - 10-AL N2r Dr

(1)

Geometry optimization for 2-MT in the neutral and oxidized forms were carried out with Gaussian 03 software10 (Gaussian Inc.) at the B3LYP/6-31G(d,p) level by DFT theory. The adiabatic electronic transitions for the different oxidation states of 2-MT in solvents were performed at the MPW1PW91/ 6-31G+(d,p) level for the neutral and dicationic state and spin unrestricted MPW1PW91/6-31G+(d,p) level for the cationic radical state with TD-DFT theory using Gaussian 03 software. The solvent effect has been considered using the polarizable continuum model (PCM). Results and Discussion

Figure 2. Absorption spectra of neutral, cationic radical, and dicationic 2-MT (0.05 mM) in acetonitrile solution.

Electrochemical Characterization. Similar to other TTF derivatives, the acetonitrile solution of 2-MT exhibits two reversible one-electron oxidation processes in cyclic voltam-

TTF-Based Reconfigurable Arithmetic Functionalities

J. Phys. Chem. C, Vol. 112, No. 43, 2008 16975

Figure 3. Absorption spectra for cationic radical 2-MT in different solution (0.05 mM, left) and absorption spectra for cationic radical 2-MT in dichloromethane solution at different concentration (right).

Figure 4. Optimized molecular structure, the atom labeling, and the molecular orbital electron densities of neutral 2-MT.

metry and differential pulse voltammetry characterization (Figure 1). The first oxidation processes at 0.38 V corresponding to the generation of a cationic radical state 2-MT. Compared with that of TTF in acetonitrile, the oxidation potential is shifted cathodically by 0.04 V. This is attributed to the electron-donating effect of the electron-rich 2-MPT moiety. Further oxidation processes at 0.79 V further oxidize the cationic radical state 2-MT to the dicationic form. The difference between the two oxidation potentials (∆E) is decreased when the solvent polarity increases (Table 1). In dichloromethane, the oxidation potential difference is 0.560 V, whereas in tetrahydrofuran (THF) it is decreased to 0.307 V (Figures S1 and S2, Supporting Information). Kato has reported that ∆E value can be a measure of the on-site Coulomb energy in metal dithiolene complexes.11 The on-site Coulomb energy is also responsible for the solvent-dependent ∆E value in the unsymmetrical TTF derivative 2-MT. In polar solvents, charge distribution tends to localize to produce a larger molecular dipole

TABLE 2: Geometrical Parameters for the Ground State of Neutral, Cationic Radical, and Dicationic State 2-MT in Acetonitrile parametera neutral 2-MT cationic radical 2-MT dicationic 2-MT R(C1-C2) R(C1-S1) R(C5-S1) R(C5-C6) R(C2-S4) R(C4-S4) R(C3-C4) θ ∆E0-90

1.350 1.782 1.753 1.348 1.788 1.764 1.337 10.1 2.287

1.388 1.744 1.721 1.369 1.756 1.748 1.343 0.0 1.092

1.414 1.722 1.684 1.393 1.737 1.735 1.350 90.0 0.331

a Bond distances are reported in angstroms, dihedral angles are in degrees, and rotation barrier energy is in eV. Atom labeling is in accordance with that in Figure 4.

moment to increase the solute-solvent stabilization energy. Thus, when the solvent polarity increases, the electron density


Sun et al.

Figure 5. Optimized molecular structure and the molecular orbital electron densities of cationic radical 2-MT.

Figure 6. Theoretical predicted molecular dipole moment with changes of rotation angles at different oxidation states.

at TTF moiety, the electron donor, will decrease, whereas the electron density at MPT moiety, the electron acceptor, increases in consequence. The polarity-triggered electron density change hence reduces the electron repulsion in TTF moiety, which in turn reduces the ∆E value in polar solvents. Absorption Characterization. Neutral 2-MT acetonitrile solution exhibits a weak overlapped absorption peak centered at 465 nm in the visible region, whereas in the ultraviolet region, a much stronger absorption peak emerges at 382 nm (Figure 2 and Figure S3 in the Supporting Information). The above two peaks are absent in the absorption spectra of neither neutral TTF8a nor 2-MPT,12 which indicates that the 2-MPT and TTF moieties are strongly coupled at the ground state in 2-MT. When 2-MT is oxidized to the cationic radical state, original absorbance at 385 nm is gradually decreased. The characteristic absorption peaks of cationic radical state TTF emerge at 417,

466, and 522 nm, similar to previous reports of TTF derivatives.8a A new absorption band centered at 730 nm appears in the near-infra (NIR) region. The absorption peak in the NIR region is hypsochromic shifted from 780 nm in dichloromethane to 728 nm in acetonitrile (Figure 3, left). The solvatochromic behavior of the NIR peak reveals that the transition dipole is opposite to the molecular dipole moment. The absorbance of the NIR region peak is linear to the solute concentration (Figure 3, right). When the solute concentration is decreased from 1 × 10-4 to 1 × 10-5 M, the absorbance ratio at 780 nm compared to that at 425 nm maintains a value of 0.45, which confirms the unique molecular nature of the ICT from 2-MPT to cationic radical TTF moiety and excludes the possibility that the NIR region absorption arises from the formation of intermolecular π dimmers.13 The absorption peaks in the visible and NIR regions corresponding to the cationic radical 2-MT are shrinking when the TTF moiety is further oxidized to the dicationic state, and a new solvent-sensitive absorption peak appears in the visible region. In dichloromethane, the absorption maximum is at 570 nm (Figure S4, Supporting Information), whereas in acetonitrile and acetone, the lowest-energy transition is blue-shifted to 565 and 550 nm (Figure S5, Supporting Information), respectively. The hypsochromical shifted electronic transition suggests that the TTF moiety receives electron density from electron-donor MPT moiety in its dicationic oxidation state. Theoretical Calculation. To explore the origin of each ICT induced absorption band, theoretical computation is employed to optimize the structure and calculate the adiabatic electronic transitions of different valence 2-MT through DFT and TD-



Figure 7. Twisted molecular structure and the molecular orbital electron densities of dicationic 2-MT.

TABLE 3: Excitations in Neutral 2-MT that Contribute to the Transitions in the 350-700 nm Range along with Their Relative Contributions Given by the Expansion Coefficients in Acetonitrile and Dichloromethane solvent

λex (nm)

Eex (eV)

oscillator strength

acetonitrile acetonitrile acetonitrile dichloromethane dichloromethane dichloromethane

466.6 440.6 377.3 472.9 443.1 379.3

2.657 2.814 3.286 2.622 2.798 3.269

0.0201 0.0001 0.1394 0.0195 0.0001 0.1438

excited state HOMO f LUMO HOMO f LUMO+2 HOMO f LUMO+1 HOMO f LUMO HOMO f LUMO+2 HOMO f LUMO+1

expansion coefficient 0.659681 0.67386 0.59340 0.69710 0.68470 0.56938

TABLE 4: Excitations in Cationic Radical 2-MT that Contribute in the NIR and Visible Region along with Their Relative Contributions Given by the Expansion Coefficients in Acetonitrile and Dichloromethane solvent

λex (nm)

Eex (eV)

oscillator strength

excited state

expansion coefficient

acetonitrile acetonitrile acetonitrile acetonitrile acetonitrile dichloromethane dichloromethane dichloromethane dichloromethane dichloromethane

846.6 506.3 496.4 461.7 447.6 877.1 510.8 493.6 459.5 449.4

1.464 2.449 2.498 2.686 2.770 1.414 2.427 2.512 2.698 2.759

0.2180 0.0463 0.0053 0.0210 0.0001 0.2243 0.0532 0.0042 0.0219 0.0001

SOMO-1 β f SOMO β SOMO-2 β f SOMO β SOMO-1 β fSOMO+1 β SOMO R f SOMO+3 R SOMO R f SOMO+2 R SOMO-1 β f SOMO β SOMO-2 β f SOMO β SOMO-1 β f SOMO+1 β SOMO R f SOMO+3 R SOMO R f SOMO+2 R

0.97201 0.76035 0.53296 0.69641 0.98060 0.97403 0.87071 0.57351 0.67926 0.97725

DFT calculation with B3LYP and MPWPW91 methods, respectively. As for the neutral form of 2-MT, geometry optimization in the gas phase produced a slightly distorted structure of TTF moiety (Figure 4). The ending ethenylene group of the TTF moiety is slightly folded away from the central plane composed of two carbon atoms and four sulfur atoms in the TTF moiety. The dihedral angle (θ) between C1C5C6S1S2 plane and C1C2S1S2 plane is 10.1°. The same folding of the TTF skeleton has also been found in the TTF molecule;14 whereas for N1C7C8S5 plane and C1C5C6S1S2 plane, the dihedral angle is found as 0.0°, indicating that the 2-MPT and TTF moieties

keeps a coplanar structure. DFT calculations also find that the rotation around the C1-C2 double bond will encounter an approximate rotation barrier (∆E0-90) of 2.287 eV in acetonitrile (Table 2). The HOMO of 2-MT locates mainly on the TTF moiety, and LUMO of 2-MT locates mainly on the 2-MPT moiety (Figure 7), similar to their mother compound TTF and 2-MPT. As for HOMO-1 and LUMO+1, they exhibit an expanding π-nature orbital covering both TTF and 2-MPT moieties. TD-DFT calculations for neutral 2-MT in acetonitrile (Table 3) suggest a weak electronic transition from TTF-based HOMO to MPT-based LUMO with an excitation energy of 2.657 eV,


Sun et al.

TABLE 5: Excitations in Dicationic 2-MT that Contribute in the NIR and Visible Region along with Their Relative Contributions Given by the Expansion Coefficients in Acetonitrile and Dichloromethane solvent

rotation angle (degrees)

λex (nm)

Eex (eV)

oscillator strength

acetonitrile acetonitrile acetonitrile acetonitrile acetonitrile dichloromethane dichloromethane dichloromethane dichloromethane dichloromethane dichloromethane

0.0 0.0 0.0 90.0 90.0 0.0 0.0 0.0 0.0 90.0 90.0

843.1 468.6 439.8 739.0 612.1 888.0 486.9 473.7 402.7 812.2 647.2

1.470 2.646 2.819 1.678 2.026 1.396 2.547 2.617 3.079 1.526 1.916

0.5007 0.1759 0.0000 0.0000 0.2710 0.5143 0.1475 0.0000 0.1123 0.0000 0.2732

which is approximate to the experimental spectral data of 2.632 eV. Thus for the lowest-energy electronic transition in the neutral state of 2-MT, the TTF moiety performs as an electron donor at the ground state, whereas the 2-MPT moiety is the electron acceptor at the ground state. Because of a relatively small transition dipole, the same ICT transition in dichloromethane has an excitation energy value of 2.634 eV. The intense absorption peaks at 3.246 eV in acetonitrile and 3.224 eV in dichloromethane are found with the same electronic transition origin from HOMO to LUMO+1, and the excitation energy is calculated as 3.286 eV in acetonitrile and 3.269 eV in dichloromethane. Upon the one-electron TTF-centered oxidation to form cationic radical 2-MT, the original distorted TTF moiety turns

Figure 8. Fluorescent spectra for neutral, cationic radical, and dicationic 2-MT (0.05 mM) in acetonitrile solution. The insect figure shows the fluorescence of neutral and cationic radical 2-MT.

SCHEME 1: Charge Transfer Phenomena between MPT and TTF Moieties at Different Oxidation States

excited state HOMO f LUMO HOMO-1 f LUMO HOMO-2 f LUMO HOMOf LUMO HOMO f LUMO+1 HOMO f LUMO HOMO-1f LUMO HOMO-2 f LUMO HOMO-4 f LUMO HOMOf LUMO HOMO f LUMO+1

expansion coefficient 0.59409 0.65715 0.70446 0.69849 0.64605 0.59010 0.66096 0.70462 0.62922 0.69814 0.64335

planar with the dihedral angle decreased to 0° (Figure 5). The central C1-C2 distance is lengthened from 1.350 Å in the neutral form to 1.388 Å in the oxidized form, whereas the C1-S1 distance is shortened from 1.782 to 1.744 Å. The bond distance changes are attributed to the occupancy of frontier orbitals. In the original HOMO orbital, the central C1-C2 bond exhibits a bonding character, whereas the outer C1-S1 bond exhibits an antibonding character. Thus the decrease of the HOMO occupancy will decrease the former bond but increase the latter one. The occupancy change of frontier orbitals also decreases the rotation barrier around the central C1-C2 bond to 1.092 eV in acetonitrile. Spin-unrestricted TD-DFT calculation in the cationic radical 2-MT predicts a low-energy electronic transition from the charge delocalized SOMO-1 (original HOMO-1) to the charge localized TTF-based SOMO (original HOMO) at 1.464 eV in acetonitrile and 1.414 eV in dichloromethane. Notably theoretical calculation also indicates that the transition dipole is just opposite to the ground-state molecular dipole moment, resulting in a higher energy for solvent recombination in polar solvents, which are in accordance with the hypsochromic shift of NIR region absorption in polar solvents. A similar-origin ICT band has also been identified by Decurtins and co-workers in a conjugated TTF derivative.15 The visible-region absorption peaks are also found as the electronic transition from inner molecular orbitals to SOMO through theoretical calculations. When the cationic radical 2-MT is oxidized to the dicationic state, the decrease of frontier orbital occupancy results in the increase of the central C1-C2 distance from 1.388 to 1.414 Å, and the decrease of the C1-S1 distance from 1.744 to 1.722 Å. The oxidation results in the formation of single bond between the central carbon atoms, C1 and C2, which further decreases the rotation barrier around the bond to 0.331 eV in acetonitrile. Meanwhile, the rotation around the central carbon atoms, C1 and C2, produces different effects on the molecular dipole moment dependent on the oxidation state (Figure 6). In the dicationic state, the rotation increases the dipole moment, with a maximum value at the rotation angle of 90°. While in the neutral and cationic radical state, the dipole moment exhibits the minimized value at the rotation value of 90°. In solution, larger molecular polar moment generally produces more stabilization energy through the solvent-solute interactions, which helps to stabilize the twisted conformation of the dicationic form in solution (Table 4). TD-DFT calculations for the dicationic 2-MT in acetonitrile (Table 5) find that calculated lowest excitation energy in the planar conformation is 1.470 eV, far less than the experimental value of 2.214 eV. However, the calculated lowest excitation energy value 2.026 eV in the conformation with a rotation angle of 90° fits the experimental data with less error, which further



Figure 9. The absorption (left) and fluorescence (right) spectrum for 2-MT (0.1 mM) in response to different chemical inputs (a) neutral 2-MT solution, (b) NOBF4 (0.4 mM), (c) CuCl2 (0.6 mM), (d) (NH4)2Ce(NO3)6 (0.3 mM), (e) TFA (1 mM).

Figure 10. Molecular logic gates for the half subtractor. Top left: absorption spectrum for 2-MT (0.1 mM) in acetonitrile in the presence of different chemical inputs. Top right: normalized fluorescence spectrum (λex ) 350 nm) for 2-MT (0.1 mM) in acetonitrile in the presence of different chemical inputs. Bottom: truth table for the half subtractor. Inputs: 0.4 mM NOBF4, 0.6 mM CuCl2. Outputs: absorbance at 630 nm (O1), and fluorescent intensity at 450 nm (O2). The dotted line shows the threshold value for different logic-0 and logic-1 states.

reveals the twisted conformation in solution (Figure 7). The lowest-energy electronic transition arises from the ICT from the delocalized HOMO (HOMO-1 in the neutral form) to the TTF-centered LUMO (HOMO in the neutral form). The transition dipole is just opposite to the dipole moment of the cationic 2-MT, which is in accordance with the hypsochromic shift of lowest-energy absorption peak in polar solvents. Fluorescence Characterization. Analogous to previously reported TTF-derived fluorescent switches, the emission of 2-MPT fluorophore in 2-MT is quenched over 99.9% compared with that of the free 2-MPT, indicating that the two moieties are strongly coupled at the excited state. The emission maximum wavelength is also red-shifted from 400 nm in 2-MPT to 450 nm in 2-MT (Figure 8). However, when the same concentration of neutral TTF and 2-MPT were mixed in the solution, no significant fluorescent quenching was observed, suggesting that the oxidative PET (oPET) process from the electron donor TTF to the excited fluorophore 2-MPT is an intramolecular behavior.7b The Gibbs free energy (∆G) change for the oPET process is -0.569 eV, as calculated from the Rehm-Weller equation,16 indicating that the oPET process is spontaneous. 2-MT remains nonfluorescent when it is oxidized to the cationic radical form. Compared with the emission properties

of neutral 2-MT, cationic radical state 2-MT exhibits an even weaker fluorescent intensity as well as the eclipsed band shape. The emission intensity reaches a high value at 400 nm and drops significantly at 425 nm. The emission band further recovers to its maximum at 450 nm. The wavelengths with high and low fluorescent intensities are in accordance with the different absorbances at the corresponding wavelengths, suggesting that the energy transfer quenching between excited MPT and cationic radical TTF moieties is a key factor to affect the fluorescence of 2-MT. TTF-centered oxidation also influences the PET processes. The oPET process7b is hindered, due to the formation of vacancy in TTF moiety, with the Gibbs free energy for the oPET decreased to -0.161 eV in the mono-oxidized form. The reductive PET (rPET) process,7b which resulted from the electron transfer from excited MPT moiety to the vacancy in TTF moiety, is elevated to -1.503 eV. Both energy transfer and PET processes are responsible for the nonfluorescence of the cationic radical 2-MT. The emission intensity is increased alongside with the second TTF-centered oxidation step. The quantum yield finally increases to 10% after the solution standing in dark over a whole night when 2 equiv of oxidants are added.7g The spectral overlap between the emission band and the absorption bands is largely


Sun et al.

Figure 11. Molecular logic gates for the half adder. Top left: absorption spectrum for 2-MT (0.05 mM) in acetonitrile in respond to input set 2. Top right: normalized fluorescence spectrum (λex)350 nm) for 2-MT (0.1 mM) in acetonitrile in respond to input set 2. Bottom: truth table for the half adder. Inputs: 1 mM TFA, 0.3 mM (NH4)2Ce(NO3)6. Outputs: absorbance at 630 nm (O1), and fluorescent intensity at 450 nm (O2). The dotted line shows the threshold value for different logic-0 and logic-1 states.

Figure 12. Chemical conversion processes (top) and electronic logic circuits (bottom) for the half adder and the half subtractor.

reduced in the dicationic state compared with that in the cationic radical state, which dramatically reduced the energy transfer quenching efficiency. The oxidative PET pathway is thoroughly blocked because the TTF-based π frontier orbital is unoccupied, and the reductive PET pathway is also deactivated with a Gibbs free energy value of -1.007 eV. Although the Gibbs free energy value is slightly lower than that in the neutral form, the restored fluorescence of 2-MT still indicates that the reductive PET pathway is less-efficient in dicationic 2-MT to quench the fluorescence compared with the oxidative PET pathway in neutral 2-MT. Further characterization on the fluorescent dicationic 2-MT suggests the fluorescent lifetime is 2.8 ns, which is slightly decreased from that of the original 2-MPT and suggests a charge-separation rate constant (Kcs) of 4.5 × 107 s-1.17 Compared with those with the same Gibbs free energy value, the lower Kcs value17 suggests the twisted geometry of dicationic 2-MT may be responsible for the intramolecular charge transfer kinetic energy barrier between the excited MPT and dicationic TTF moieties and thus in turn restoring the fluorescence. Charge Transfer Phenomena in 2-MT. As evidenced in the absorption and fluorescent spectra, the electronic transitions in 2-MT can be tuned by the oxidation states (Scheme 1). Neutral TTF has a fully occupied active frontier π orbital and can perform as an electron donor to give electron to the vacancy in MPT frontier orbitals at both the ground state and the excited state. In cationic radical TTF, the active frontier π orbital is semioccupied and thus can accept electron from the occupied

MPT orbital at the ground state and give electron to the vacancy in MPT frontier orbital at the excited state. In dicationic TTF, the active frontier π orbital is unoccupied, and can only perform as electron acceptor to receive electrons from the occupied frontier orbital centered at MPT moiety at the ground state. However, the geometry twisting hampers the TTF moiety from accepting electron from 2-MPT at the excited state. Binding Feature of 2-MT. In 2-MT, the 2-MPT fluorophore can bind with proton or metal ions, whereas the TTF moiety is sensitive to the chemical oxidant. The chemical recognition features of different structural moieties demonstrate that the charge transfers triggered by different electron-donating and -withdrawing effects of TTF moieties can be tuned by the external chemical inputs (Figure 9). When trifluoroacetic acid (TFA) is introduced, the proton binds with the pyridine ring just as in other MPT derivatives. The protonation enhances the electron-withdrawing capacity of the pyridine ring and lowers the transition energy from TTF-centered HOMO to MPTcentered LUMO, resulting in the red-shift of the CT band from 390 nm in the UV region to 530 nm in the visible region. The protonation also promotes the ICT process from the methoxy group to the pyridine ring to some extent, which slightly recovers the fluorescence at 450 nm. When oxidative metal ions, such as cupric ion or ceric (IV) ion, are introduced, the coordination at the MPT moiety occurs as well as the oxidation at the TTF moiety, leading to distinct spectral changes in both absorption and emission. However, the resulting spectra are dependent on the metal ions due to their different oxidation potential. As for the cupric ion, it oxidizes the TTF moiety to the cationic radical state,18 resulting in the appearance of its characteristic absorption bands in the visible and NIR regions. The coordination between MPT and the cupric ion also decreases the electron density at MPT ring, which increases the transition energy from MPT moiety to TTF moiety and slightly blue-shifts the absorption peak in the near-infra region from 730 to 715 nm. As for the ceric(IV) ion, it directly oxidizes the TTF moiety to the dicationic state with the emergence of a CT band centered at 600 nm, owing to its high oxidation potential. For both of the two metal ions, their



Figure 13. Molecular key-pad lock within 2-MT (0.05 mM in acetonitrile). Inputs: 0.1 mM NOBF4 (O), 0.1 mM NaBH4 (R), and 2 min light irradiation at 350 nm (L). Outputs: fluorescent intensity at 450 nm. The solid line shows the threshold value.

coordination at MPT moiety increases the nonradioactive relaxation of excited MPT, which further quenches the fluorescence of MPT moiety in 2-MT. Hence, the same chemical introduction may trigger different effects between ICT at the ground state and PET at the excited state. Reconfigurable Binary Arithmetic. On the basis of the cooperation effects of different chemical inputs, the charge transfer at the ground state and at the excited state may respond differently toward the same chemical input, which enables the reconfigurable binary arithmetic functionalities to be constructed within 2-MT. Two sets of chemical input signals are selected to execute the chemical-driven arithmetic operations. The input set 1 is constituted of the introduction of 0.4 mM NOBF4 (I1) and 0.6 mM CuCl2 (I2), whereas the input set 2 is selected as 1 mM TFA (I3) and 0.3 mM (NH4)2Ce(NO3)6 (I4). The output channels in response to the input sets 1 and 2 are selected as the absorbance at 630 nm (O1) and the emission intensity at 450 nm (O2) under the excitation at 350 nm. In each output channel, the logic-0 state is defined as signal intensity below the threshold. When input set 1 is applied to the 2-MT acetonitrile solution, the initial molecular system executes half-subtractor functionality. As shown in Figure 10, the absorption and fluorescence signals are low (i.e., O1 ) O2 ) 0) when no input signal is introduced (i.e., I1 ) I2 ) 0). When I1 is present alone (i.e., I1 ) 1, I2 ) 0), the introduced NOBF4 oxidizes TTF moiety to the dicationic state, promoting the ICT from MPT to TTF moiety at the ground state but blocking the PET from TTF to MPT moiety at the excited state. Thus, the output signals in both channels are high (i.e., O1 ) O2 ) 1). When I2 is present alone (i.e., I1 ) 0, I2 ) 1), the cupric ion oxidizes the TTF moiety to the cationic radical state, and also promotes the ICT from the MPT to the TTF moiety at the ground state, producing a high-output state at O1 (i.e., O1 ) 1). However, the nonradioactive relaxation of excited-state MPT is further enhanced by the coordination of cupric and thus produces a low-output state at O2 (i.e., O2 ) 0). When both inputs are present (i.e., I1 ) I2 ) 1), the coordination of the cupric ion with dicationic 2-MT decreases the absorption assigned to the ICT from MPT to TTF, and quenches the fluorescence of the dicationic 2-MT. Thus both the two output channels exhibit a low-output state (i.e., O1 ) O2 ) 0). Therefore, an XOR gate is then constructed at O1, whereas an INHIBIT gate is constructed at O2. A halfsubtractor is then accessible with input set 1. When input set 2 is applied, the arithmetic functionality of the molecular system is reconfigured to a half adder. As shown

in Figure 11, the output signals are low in both channels (i.e., O1 ) O2 ) 0) when no input signal is introduced (i.e., I3 ) I4 ) 0). When I3 is present alone (i.e., I3 ) 1, I4 ) 0), protonation enhances the ICT from TTF to MPT at the ground state and slightly blocks the PET from TTF to MPT at the excited state. Thus the signal at the absorption output channel is high, whereas the fluorescence output channel exhibits low-output signal (i.e., O1 )1, O2 ) 0). When I4 is present alone (i.e., I3 ) 0, I4 ) 1), the ceria(IV) ion oxidizes TTF moiety to the dicationic state and promotes the ICT from the MPT to the TTF moiety at the ground state, producing a high-output state at O1 (i.e., O1 ) 1). However, the nonradioactive relaxation of the excited-state MPT moiety is also enhanced by the coordination of ceria(IV) ion and thus produces a low output state at O2 (i.e., O2 ) 0). When both input signals are present (i.e., I3 ) I4 ) 1), the coordination of ceria (IV) ion with dicationic 2-MT is hampered because of the proton existence and recovers the original fluorescence of protonated dicationic 2-MT to produce the highoutput state at O2 (i.e., O2 ) 1). However, the protonated dicationic 2-MT also exhibits weak absorbance at 630 nm, thus producing low output state at O1 (i.e., O2 ) 1). An XOR gate is then constructed at O1, whereas an AND gate is implemented at O2. A half adder is then accessible with input set 2 (Figure 12). Molecular Keypad Lock. The stepwise oxidation of 2-MT can also be utilized to distinguish different input sequences and thus serves as a molecular keypad lock to authenticate 3-digit password entries. The input signals are introduced through a keypad, in which keys are linked with different chemical or photoirradiation, such as NOBF4 (1 eq., denoted as O), NaBH4 (1 eq., denoted as R) and 350 nm excitation (2 min impulse, denoted as L), and the output requirement to pass the password authentication is the fluorescent intensity at 450 nm being above the threshold. As the neutral and the cationic radical 2-MT are both nonfluorescent, the fluorescence is activated after the introduction of two times NOBF4, which sequentially oxidizes 2-MT to its dicationic state. To trigger the fluorescence, a light source must be introduced after the oxidation reactions. Thus among different combinations of 3-digit input sequences, the input string OOL is the only one to produce the required fluorescent signal above the threshold, as shown in Figure 13. The reversible redox reactions of 2-MT is then encoded with the keypad lock functionality to authenticate user’s password entries at the molecular level.

16982 J. Phys. Chem. C, Vol. 112, No. 43, 2008 Conclusion In this article, we report a new way of processing information within a TTF-based fluorescent switch. The differences between ICT and PET processes within 2-MT provides various spectral output states for the binary arithmetic computation, whereas the recognition of 2-MT with protons, metal ions, and oxidants are employed to switch the spectral states to execute the arithmetic operations. The major conclusions are as follows: (1) The charge transfer phenomena within different oxidation states of 2-MT are investigated via spectral and theoretical approaches. As in the neutral state, the TTF moiety performs as electron donor at both the ground state and the excited state. While in the oxidized state, TTF moiety mainly performs as an electron acceptor. The structural twisting between the TTF and MPT moieties may also block the PET process. (2) Both the half adder and the half subtractor are assembled in the same TTF-based fluorescent switch. The arithmetic functions are accomplished by the nonsynchronous ICT and PET behaviors in response to the two-input chemical introduction. (3) On the basis of the sequential activation processes for the fluorescence of 2-MT, the redox-controlled fluorescent switch can serve as a molecular keypad lock to differ 3-digit input sequences. As a chemical-driven arithmetic system, the execution of algebraic functionalities in the current molecule is still facing the challenge from signal resetting and chemical waste accumulation, which is a common problem for moleculators. Although previous reports by Shanzer,19 Zhang,8a Tian,5b and us4c solved the resetting problem through the introduction of reset agents, the chemical waste accumulation after multicycle execution still needs further investigation. Recently, development in assembling molecular logic units into the microfluidic chip may provide an alternative based on the low chemical consumption and refreshable feature in device.20 Although 2-MT is only a prototypical model, the reconfigurable arithmetic functions integrated in a single fluorescent switch 2-MT through the utilization of differences between ICT and PET processes may also provide a way to increase the computing power for TTF derivatives in future molecular devices. Acknowledgment. The authors acknowledge the support for this research through the NSFC under contract numbers 20490213, 20221101, and 20423005, and through Peking University. Supporting Information Available: Detailed electrochemical, spectral and computation characterization of 2-MT in different solvents. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) (a) Packan, P. A. Science 1999, 285, 2079–2081. (b) Schulz, M. Nature 1999, 399, 729–730. (c) Hu, C. M. Nanotechnology 1999, 10, 113– 116. (d) Green, J. E.; Choi, J. W.; Boukai, A.; Bunimovich, Y.; JohnstonHalperin, E.; DeIonno, E.; Luo, Y.; Sheriff, B. A.; Xu, K.; Shin, Y. S.; Tseng, H.-R.; Stoddart, J. F.; Heath, J. R Nature 2007, 445, 414–417. (e) Saha, S.; Flood, A. H.; Stoddart, J. F.; Impellizzeri, S.; Silvi, S.; Venturi, M.; Credi, A. J. Am. Chem. Soc. 2007, 129, 12159–12171. (f) Credi, A. Angew. Chem., Int. Ed. 2007, 46, 5472–5475. (2) (a) de Silva, A. P.; Gunnlaugsson, H. Q. N.; McCoy, C. P. Nature 1993, 364, 42–44. (b) de Silva, A. P.; McClenaghan, N. D.; McCoy, C. P. Molecular-LeVel Electronics, Imaging and Information, Energy and EnVironment, in Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: Weinheim, Germany, 2001; Vol. 5. (c) Balzani, V.; Venturi, M.; Credi, A. Molecular DeVices and Machines A Journey into the Nano World; WileyVCH: Weinheim, Germany, 2003. (d) de Silva, A. P. Nat. Mater. 2005, 4,

Sun et al. 15–16. (e) de Silva, A. P.; Uchiyama, S. Nat. Nanotechnol. 2007, 2, 399– 410. (f) Ballardin, R.; Ceroni, P.; Credi, A.; Gandolfi, M. T.; Maestri, M.; Semararo, M.; Venturi, M.; Balzani, V. AdV. Funct. Mater. 2007, 17, 740– 750. (g) Balzani, V.; Credi, A.; Venturi, M. Chem.sEur. J. 2008, 14, 26– 39. (3) (a) Brown, G. J.; de Silva, A. P.; Pagliari, S. Chem. Commun. 2002, 2461–2463. (b) de Silva, A. P.; McClenaghan, N. D. Chem.sEur. J. 2002, 8, 4935–4945. (c) Raymo, F. M.; Giordani, S. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 4941–2944. (d) de Silva, A. P.; McCaughan, B.; McKinney, B. O. F.; Querol, M. Dalton Trans. 2003, 1902–1913. (e) de Silva, A. P.; McClenaghan, N. D. Chem.sEur. J. 2004, 10, 574–586. (f) Guo, X. F.; Zhang, D. Q.; Zhu, D. B AdV. Mater. 2004, 16, 125–130. (g) Qu, D. H.; Ji, F. Y.; Wang, Q. C.; Tian, H. AdV. Mater. 2006, 18, 2035–2038. (h) Zhou, W. D.; Li, Y. J.; Li, Y. L.; Liu, H. B.; Wang, S.; Li, C. H.; Yuan, M. J.; Liu, X. F.; Zhu, D. B. Chem. Asian J. 2006, 1-2, 224–230. (i) Li, Z. X.; Liao, L. Y.; Sun, W.; Xu, C. H.; Zhang, C.; Fang, C. J.; Yan, C. H. J. Phys. Chem. C 2008, 112, 5190–5196. (4) (a) Coskun, A.; Deniz, E.; Akkaya, E. U. Org. Lett. 2005, 7, 5187– 5189. (b) For a recent review, please see: Pishel, U. Angew. Chem., Int. Ed. 2007, 46, 4026–4040, and references therein. (c) Perez-Inestrosa, E.; Montenegro, J.-M.; Collado, D.; Suau, R.; Casado, J. J. Phys. Chem. C 2007, 111, 6904–6909. (d) Sun, W.; Zheng, Y. R.; Xu, C. H.; Fang, C. J.; Yan, C. H. J. Phys. Chem. C 2007, 111, 11706–11711. (e) Pishel, U.; Heller, B. New J. Chem. 2008, 32, 395–400. (5) (a) Margulies, D.; Felder, C. E.; Melman, G.; Shanzer, A. J. Am. Chem. Soc. 2007, 129, 347–354. (b) Guo, Z. Q.; Zhu, W. H.; Shen, L. J.; Tian, H. Angew. Chem., Int. Ed. 2007, 46, 5549–5553. (c) Strack, G.; Ornatska, M.; Pita, M.; Katz, E. J. Am. Chem. Soc. 2008, 130, 4234–4235. (6) (a) Bryce, M. R. AdV. Mater. 1999, 11, 11–23. (b) Yamada, J.; Sugimoto, T. TTF Chemistry: Fundamentals and Applications of TetrathiafulValene; Kodansha-Springer: Tokyo, 2004. (c) Segura, J. L.; Martı´n, N Angew. Chem., Int. Ed. 2001, 40, 1372–1409, and references therein. (d) For recent reviews, please see: Special Issue: Molecular Conductors. (e) Batail, P. Guest Ed. Chem. ReV. 2004, 104, 4887–5782. (f) Balzani, V.; Clemente-Leon, M.; Credi, A.; Ferrer, B.; Venturi, M.; Flood, A. H.; Stoddart, J. F. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1178–1183. (7) (a) Wang, C.; Bryce, M. R.; Batsanov, A. S.; Stanley, C. F.; Beeby, A.; Howard, J. A. K. J. Chem. Soc., Perkin Trans. 2 1997, 1671–1678. (b) Farren, C.; Christensen, C. A.; Gerald, S. F.; Bryce, M. R.; Beeby, A. J. Org. Chem. 2002, 67, 9130–9139. (c) Li, H.; Jeppesen, J. O.; Levillain, E.; Becher, J. Chem. Commun. 2003, 846–847. (d) Guo, X. F.; Zhang, D. Q.; Zhang, H. J.; Fan, Q. H.; Xu, W.; Ai, X. C.; Fan, L. Z.; Zhu, D. B. Tetrahedron 2003, 59, 4843–4850. (e) Zhang, G. X.; Zhang, D. Q.; Guo, X. F.; Zhu, D. B. Org. Lett. 2004, 6, 1209–1212. (f) Xiao, X. W.; Xu, W.; Zhang, D. Q.; Xu, H.; Lu, H. Y.; Zhu, D. B. J. Mater. Chem. 2005, 15, 2557–2561. (g) Leroy-Lhez, S.; Baffreau, J.; Perrin, L.; Levillain, E.; Allain, M.; Blesa, M. J.; Hudhomme, P. J. Org. Chem. 2005, 70, 6313–6320. (h) Xiao, X. W.; Xu, W.; Zhang, D. Q.; Xu, H.; Liu, L.; Zhu, D. B. New J. Chem. 2005, 29, 1291–1294. (i) Zhang, G. X.; Zhang, D. Q.; Zhao, X. H.; Ai, X. C.; Zhang, J. P.; Zhu, D. B. Chem.sEur. J. 2006, 12, 1067–1073. (j) Tian, W.; Zhang, D. Q.; Wu, H.; Zhu, D. B. Tetrahedron Lett. 2008, 49, 1361–1364. (8) (a) Zhou, Y.; Wu, H.; Qu, L.; Zhang, D. Q.; Zhu, D. B. J. Phys. Chem. B 2006, 110, 15676–15679. (b) Zhang, G.; Zhang, D. Q.; Zhou, Y.; Zhu, D. B. J. Org. Chem. 2006, 71, 3970–3972. (c) Fang, C. J.; Zhu, Z.; Sun, W.; Xu, C. H.; Yan, C. H. New J. Chem. 2007, 31, 580–586. (9) Demas, J. N.; Grosby, G. A. J. Phys. Chem. 1971, 75, 991–1024. (10) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian, Inc.; Wallingford, CT, 2004. (b) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270–283. (c) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299– 310. (d) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284–298. (e) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785–789. (f) Adamo, C.; Barone, V. Chem. Phys. Lett. 2000, 330, 152–160. (g) Jodicke, C. J.; Luthi, H. P. J. Chem. Phys. 2002, 117, 4146–4156. (h) Scholz, R.; Kobitski, A. Y.; Zahn, D. R. T.; Schreiber, M. Phys. ReV. B 2005, 72, 245208–245225. (i) Neiss, C.; Saalfrank, P.; Parac, M.; Grimme, S. J. Phys. Chem. A 2003, 107, 140–147. (11) Kato, R. Chem. ReV. 2004, 104, 5319–5346.

TTF-Based Reconfigurable Arithmetic Functionalities (12) Zheng, M. H.; Jin, J. Y.; Sun, W.; Yan, C. H. New J. Chem. 2006, 30, 1192–1196. (13) (a) Spanggaard, H.; Prehn, J.; Nielsen, M. B.; Levillain, E.; Allain, M.; Becher, J. J. Am. Chem. Soc. 2000, 122, 9486–9494. (b) Khodorkovsky, V.; Shapiro, L.; Krief, P.; Shames, A.; Mabon, G.; Gorgues, A.; Giffard, M. Chem. Commun. 2001, 2736–2737. (14) Pou-Amerigo, R.; Orti, E.; Merchhan, M.; Rubio, M.; Viruela, P. M. J. Phys. Chem. A 2002, 106, 631–640. (15) Jia, C. Y.; Liu, S. X.; Tanner, C.; Leiggener, C.; Neels, A.; Sanguinet, L.; Levillain, E.; Leutwyler, S.; Hauser, A.; Decurtins, S. Chem.sEur. J. 2007, 133, 804–3812. (16) (a) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259–271. (b) For 2-MT: ∆G ) E(ox)- E(red)- E0-0 -e2/4πr, E0-0 ) 3.155 eV, ) 36.64, r ) 9.223 Å. For oxidative PET in neutral 2-MT, E(ox)- E(red) ) 2.632 eV; For oxidative PET in cationic radical 2-MT, E(ox)- E(red) ) 3.040

J. Phys. Chem. C, Vol. 112, No. 43, 2008 16983 eV; For reductive PET in cationic radical 2-MT, E(ox)- E(red) ) 1.698 eV; for reductive PET in dicationic 2-MT, E(ox)- E(red) ) 2.194 eV. For computation details, please see the supporting information. (17) Guo, X. F.; Gan, Z. H.; Luo, H. X.; Araki, Y.; Zhang, D. Q.; Zhu, D. B.; Ito, O. J. Phys. Chem. A 2003, 107, 9747–9753. (18) Martinez, R.; Ratera, I.; Tarraga, A.; Molina, P.; Veciana, J. Chem. Commun. 2006, 3809–3811. (19) Margulies, D.; Melman, G.; Shanzer, A. Nat. Mater. 2005, 4, 768– 771. (20) Kou, S. Z.; Lee, H. N.; van Noort, D.; Swamy, K. M. K.; Kim, S. H.; Soh, J. H.; Lee, K. M.; Nam, S. W.; Yoon, J. Y.; Park, S. S. Angew. Chem., Int. Ed. 2008, 47, 872–876.

JP804628U

Chemical-Driven Reconfigurable Arithmetic Functionalities within a

Recommend Documents