Experimental and computational exploration of photophysical and electroluminescent properties of modified 2,2′:6′,2″-terpyridine, 2,6-di(thiazol-2-yl)pyridine and 2,6-di(pyrazin-2-yl)pyridine ligands and their Re(I) complexes

Received: 25 July 2018

DOI: 10.1002/aoc.4611

Revised: 1 September 2018

Accepted: 3 September 2018

F U L L P A P E R

Experimental and computational exploration of

photophysical and electroluminescent properties of

modified 2,2′:6′,2″‐terpyridine, 2,6‐di(thiazol‐2‐yl)pyridine

and 2,6‐di(pyrazin‐2‐yl)pyridine ligands and their Re(I)

complexes

Tomasz Klemens¹| Anna Świtlicka¹| Agata Szlapa‐Kula¹| Stanisław Krompiec¹|

Piotr Lodowski¹| Anna Chrobok³| Magdalena Godlewska⁴| Sonia Kotowicz¹|

Mariola Siwy²| Katarzyna Bednarczyk¹| Marcin Libera¹| Sebastian Maćkowski⁵|

Tomasz Pędziński⁶| Ewa Schab‐Balcerzak^1,2| Barbara Machura¹

¹Institute of Chemistry, University of

The excited‐state characteristics of a series of 2,2′:6′,2″‐terpyridine (terpy),

2,6‐di(thiazol‐2‐yl)pyridine (dtpy) and 2,6‐di(pyrazin‐2‐yl)pyridine (dppy)

derivatives as well as their corresponding Re(I) complexes [ReCl(CO)₃(Lⁿ‐κ²N)]

were investigated both experimentally and theoretically, and the crucial effect

of pyrrolidine substituent and peripheral rings on the optical and electrochem-

ical properties was found evident. For Re(I) complexes bearing the ligands

with electron‐rich pyrrolidine substituent, different emission profiles were

found in polar and non‐polar solvents, indicating a change in the character

of the excited state. Dual‐emission effect of [ReCl(CO)₃(L⁴‐κ²N)] and

[ReCl(CO)₃(L⁵‐κ²N)] in chloroform was attributed to the presence of two

emitting states, identified as an ¹ILCT excited state deactivated at higher

energies and a longer‐lived red‐shifted phosphorescence assigned to the

³MLCT excited state. The triplet excited state was confirmed by recording the

nanosecond time‐resolved transient absorption spectra for the compound

[ReCl(CO)₃(L⁴‐κ²N)]. To verify the charge transfer problem of low‐lying excited

states of the free ligands, the Λ parameter was calculated. In addition, the

compounds were applied as emitting layers for both non‐doped and doped

single‐layer organic light‐emitting diodes fabricated by solution processing.

Silesia, 9 Szkolna Str., 40‐006 Katowice,

Poland

²Centre of Polymer and Carbon Materials,

Polish Academy of Sciences, 34 M. Curie‐

Sklodowska Str., 41‐819 Zabrze, Poland

³Faculty of Chemistry, Silesian University

of Technology, 9 Strzody Str., 44‐100

Gliwice, Poland

⁴Mass Spectrometry Group, Institute of

Organic Chemistry, Polish Academy of

Sciences, Kasprzaka 44/52, PO Box 58, 01‐

224 Warszawa, Poland

⁵Institute of Physics, Faculty of Physics,

Astronomy and Informatics, Nicolaus

Copernicus University, 5 Grudziądzka

Str., 87‐100 Torun, Poland

⁶Faculty of Chemistry, Adam Mickiewicz

University in Poznań, 89b Umultowska,

61‐614 Poznań, Poland

Correspondence

Schab‐Balcerzak and Barbara Machura,

Institute of Chemistry, University of

Silesia 9 Szkolna Str., 40‐006 Katowice,

Poland.

KEYWORDS

Email: ewa.schab ‐balcerzak@us.edu.pl;

barbara.machura@us.edu.pl

donor–acceptor ligands, electroluminescence, OLEDs, photoluminescence, Re(I) complexes

Funding information

Narodowe Centrum Nauki, Grant/Award

Number: DEC‐2017/25/B/ST5/01611

Appl Organometal Chem. 2018;e4611.

wileyonlinelibrary.com/journal/aoc

1 of 21

https://doi.org/10.1002/aoc.4611

2 of 21

KLEMENS ET AL.

functionalized by carbazole group^[11]and triphenylamine‐

substituted 1,10‐phenanthroline and 2‐pyridyl‐1,2,3‐tri-

azole.^[12]For fac‐[Re(L)(CO)₃(cbz₂phen)]^+/0(cbz₂phen =

4,7‐di(9H‐carbazol‐9‐yl)‐1,10‐phenanthroline and L = Cl

⁻, pyridine or 2‐aminomethylpyridine), solution‐phase

time‐resolved photoluminescence (PL) confirmed the pres-

1 | INTRODUCTION

The photophysical properties of Re(I) tricarbonyl com-

plexes with diimine ligands have been the focus of

numerous investigations^[1]since 1974, when unusual

excited‐state properties of [ReCl(CO)₃(4,7‐(Ph)₂phen)]

and [ReCl(CO)₃(5‐R‐phen)] were discovered by Wrighton

and Morse.^[2]The development of this area is driven not

only by intriguing photophysical features, but chiefly by

high potential of [ReL(CO)₃(N^∩N)]ⁿ⁺(n = 0 or 1) to serve

as catalysts,^[3]multimodal platforms for inorganic chemi-

cal biology,^[4]sensors,^[5]promising molecular switches^[6]

and organic light‐emitting device emitters.^[7]

1

ence of two emitting states, identified as ILCTc_bz2phen

excited state deactivation at higher energies and a long‐

lived phosphorescence attributed to the ³MLCT_{Re→cbz2phen}

excited state.^[5,11]Dual charge transfer has been also con-

firmed in Re(I) thioether substituted hexaazanaphthalene

complexes.^[13]It is assumed that photoexcitation of these

1

systems can optically populate both ligand‐centred ππ*

For the majority of these complexes, photoexcitation

involves metal‐to‐ligand charge transfer (MLCT) or

mixed MLCT/ligand‐to‐ligand charge transfer (frequently

denoted as MLLCT) type transitions, where the frontier

orbitals are considerably delocalized over both the

ligand(s) and the metal centre. It has been demonstrated

that the energies of the metal‐based d_πand diimine‐based

π*, which act as the electron donor and acceptor in

MLCT transitions, can be fine‐tuned by introduction of

electron‐withdrawing or electron‐donating groups into

diamine ligand skeleton (N^∩N) or changing the ancillary

ligand (L).^[1,8]

In recent years, special attention has been paid to strong

electron‐donating substituents, such as triarylamines and

thioethers, which – when introduced into a diamine skele-

ton – can give rise to intraligand charge transfer (ILCT)

transitions from donor orbital localized on the electron‐

rich group to diimine‐based π*acceptor orbital. Compared

to the conventional MLCT excitations, they have higher

extinction coefficients and they are characterized with

lower energies.^[9]

and ¹MLCT states, which can potentially relax via indepen-

dent pathways.^[14]Coexistence of the distinct excited

states offers additional opportunity for fine‐tuning the

luminescent properties of this class of compounds.

Most importantly, the use of highly emitting neutral

tricarbonyl Re(I) complexes with donor–acceptor character-

istics, such as [ReBr(CO)₃(PTIP)], [ReBr(CO)₃(TPIP)] and

[ReBr (CO)₃(TTIP)] (PTIP = N,N‐diphenyl‐ 4‐(1‐phenyl‐1H‐

imidazo[4,5‐f][1,10]phenanthrolin‐2‐yl)aniline, TPIP = N,

N‐diphenyl‐4‐(2‐phenyl‐1H‐imidazo[4,5‐f][1,10]phenanth-

rolin‐1‐yl)aniline, TTIP = 4,40‐(1H‐imidazo[4,5‐f][1,10]

phenanthroline‐1,2‐diyl)bis(N,N‐diphenylaniline)),asdoped

emitters in organic light‐emitting diodes (OLEDs) allowed

the achievement of remarkable device performances.^7e

In an attempt to further understand the effect of strong

electron‐releasing substituents, a series of Re(I) carbonyls

with modified 2,2′:6′,2″‐terpyridine (terpy), 2,6‐di(thiazol‐

2‐yl)pyridine (dtpy) and 2,6‐di(pyrazin‐2‐yl)pyridine

(dppy) ligands (Scheme 1) have been synthesized and the

nature and energies of the electronic transitions that occur

in these systems were investigated both experimentally

(with UV–visible absorption and emission spectroscopies)

and theoretically (using density functional theory (DFT)

calculations at the PBE1PBE/DEF2–TZVPD/DEF2‐TZVP

ILCT transitions were found to govern spectral proper-

ties of Re(I) carbonyls with sulfur‐ and triarylamine‐

substituted dipyridophenazine,^[10]1,10‐phenanthroline

SCHEME 1 Ligands employed in this

study.

KLEMENS ET AL.

3 of 21

level). Phenyl‐substituted triimine ligands (L¹–L³) and

their Re(I) complexes [ReCl(CO)₃(Lⁿ‐κ²N)] (1–3)

were synthesized as control compounds, to get a better

understanding of ILCT transitions in the systems with

electron‐rich pyrrolidine group (L⁴–L⁶and 4–6). To fur-

ther modulate both donor–acceptor and photopysical

properties of L¹–L⁶and 1–6, three different triimine skele-

tons (terpy, dtpy and dppy) were used.

In addition, the obtained compounds were character-

ized using electroluminescence (EL), and some of them

were applied as emitting layers for both non‐doped

and doped single‐layer OLEDs fabricated by solution

processing.

Consistent with the bidentate coordination mode of Lⁿ,

the resonances attributed to the outer piryd‐2‐yl/thiazol‐

1

2‐yl/pyrazin‐2‐yl protons are clearly differentiated in H

NMR spectra (Fig. S3 in supporting information). Distinc-

tive signals corresponding to the carbonyl groups appear

at 198 and 190 ppm in ¹³C NMR spectra.

Solubility tests of the synthesized compounds were

carried out in chloroform to determine whether the stud-

ied ligands and complexes reach a concentration of

10 mg ml⁻¹, used in the procedure of layer formation

(see Section 4). It was found that compounds with the

pyrrolidine substituent and without pyrazine ring were

fully soluble, whereas molecules without pyrrolidine unit

(L¹–L³and 1–3) or with pyrazine ring (L⁶and 6) were

only partially dissolved. It can be concluded that the pres-

ence of pyrrolidine substituent attached to the phenyl

group enhances the solubility, while the introduction of

pyrazine ring leads to its reduction.

2 | RESULTS AND DISCUSSION

2.1 | Synthesis and General

Characterization

Derivatives of terpy, dtpy and dppy were obtained by

condensation of the corresponding aldehyde with 2‐

acetylpyridine, 2‐acetylthiazole and 2‐acetylpyrazine,

respectively (Fig. S1 in supporting information). The

products of the base‐mediated Kröhnke condensation

were not isolated, but underwent in situ pyridine ring clo-

sure in the presence of an ammonia source.^[15]

The complexes 1–6 were isolated as orange (1, 4, 5),

yellow (2, 3) and red (6) solids by reaction of [Re(CO)₅Cl]

with one equivalent of the corresponding ligand in aceto-

nitrile at refluxing temperature.

2.2 | X‐ray Structures

The complexes 1–6 show a distorted octahedral geometry

around the central Re(1) atom, with three fac‐disposed

carbonyl ligands, chloride ion and two nitrogen atoms

of the organic ligand Lⁿ– one from the central pyridine

and one from the peripheral ring (pyridine in L¹and L⁴,

thiazole in L²and L⁵and pyrazine in L³and L⁶) (Fig. 1

and Fig. S9). The uncoordinated pyrid‐2‐yl in 1 and 4,

thiazol‐2‐yl in 2 and 5 and pyrazin‐2‐yl in 3 and 6 interact

sterically with the carbonyl group C(1)–O(1), which is

manifested in a significant increase of C(1)–Re(1)–N(2)

angle (101.70(2)° in 1, 102.2(3)° in 2, 100.27(19)° in 3,

102.70(17)° in 4, 100.40(18)° in 5, 103.36(16)° in 6), being

the largest one between any two cis‐arranged ligands in

[ReCl(CO)₃(Lⁿ‐κ²N)] (Table S1). Typically for Re(I) com-

plexes incorporating bidentate‐coordinated conjugated

triimine ligands, the Re(1)–N(2) bond length to the

The infrared (IR) spectra of 1–6 display two overlap-

ping lower energy bands (1932–1876 cm⁻¹) and sharp

intense bands at higher wavenumbers (2030–2018 cm⁻¹

)

attributed to ν(C≡O) of the fac‐[Re(CO)₃]⁺moiety. The

medium intensity stretching modes ν(CN) and ν(C¼C)

of the organic ligand Lⁿ(n = 1–6) occur in the range

1615–1526 cm⁻¹(Fig. S2 in supporting information).

FIGURE 1 Perspective views demonstrating the molecular structures of 4, 5 and 6 with atom numbering. Atomic displacement ellipsoids

are drawn at the 50% probability level.

4 of 21

KLEMENS ET AL.

central pyridine (2.223(5) Å in 1, 2.244(6) Å in 2,

2.198(4) Å in 3, 2.204(4) Å in 4, 2.236(4) Å in 5,

2.213(3) Å in 6) is longer than the corresponding Re(1)–

N(1) distance to the pyrid‐2‐yl/thiazol‐2‐yl/pyrazin‐2‐yl

ring (2.179(5) Å in 1, 2.157(5) Å in 2, 2.160(4) Å in 3,

2.170(3) Å in 4, 2.151(4) Å in 5, 2.159(4) Å in 6). To a

large extent, the angular distortion of 1–6 from the ideal

octahedral geometry is attributed to the occurrence of

five‐member chelate ring of Lⁿresulting in N(2)–Re(1)–

N(1) angle of 74.58(17)° for 1, 74.9(2)° for 2, 74.42(14)°

for 3, 74.25(13)° for 4, 74.11(14) for 5, 74.22(12)° for 6

(Table S1), significantly smaller than the ideal 90° angle

of an octahedron. The coordinated peripheral ring is

almost coplanar with the central pyridine ring, with a

dihedral angle of 13.43° in 1, 2.37° in 2, 11.35° in 3,

8.71° in 4, 15.09° in 5, 10.56° in 6, while the uncoordi-

nated piryd‐2‐yl/thiazol‐2‐yl/pyrazin‐2‐yl rings are

inclined to the central pyridine at 50.18° in 1, 59.28° in

2, 48.21° in 3, 64.00° in 4, 33.36° in 5, 61.48° in 6. The

dihedral angles between the least squares planes of the

pyridine and phenyl rings are 11.15° in 1, 6.63° in 2,

17.71° in 3, 13.87° in 4, 17.67° in 5, 13.99° in 6.

To explore the impact of the triimine modification on

the local geometry, the sum of bond angles around the

Re(I) centre has also been calculated. For structures 1–

6, the calculated values of 1604.11°, 1596.78°, 1600.37°,

1602.67°, 1599.47° and 1603.56°, slightly lower than the

1620° for an ideal octahedral ligand arrangement,^[16]are

comparable to each other and indicate a negligible effect

of the pyrrolidine substituent and peripheral rings of Lⁿ

on the local geometry around the rhenium centre.

weak hydrogen bonds C―H⋅⋅⋅O, C―H⋅⋅⋅Cl or C―H⋅⋅⋅S

are dominant in the examined crystal structures. For 1

and 3, also contribution of π‐stacking was evident (see

also Table S3).

2.3 | Differential Scanning Calorimetry

and Thermogravimetric Analysis

Temperature stability of compounds incorporated in

OLED structures is required to sustain continuous Joule

heating when the voltage is applied and the resulting pos-

sible thermal breakdown. It was found that local heating

in a current‐driven organic device may lead to localized

electrical shortcuts, which raised the temperature to

180–200°C within a device.^[17]The thermal properties of

the free ligands L¹–L⁶and complexes 1–6 were measured

using differential scanning calorimetry and thermogravi-

metric analysis under nitrogen atmosphere and the data

are collected in Table S5, while representative differential

scanning calorimetry curves of this family of compounds

can be found in the supporting information (Fig. S11).

Considering the effect of the pyrrolidine substituent

attached to the phenyl group, it can be concluded that

introduction of such a unit slightly affects the melting

temperature (T_m). However, the presence of pyrrolidine

gives the possibility of generating an amorphous phase,

and, in the second heating scan, after rapid cooling, the

glass transition (T_g) was observed. The presence of pyr-

azine rings in ligands L³and L⁶increases both T_mand

T_gvalues. All of the complexes exhibited higher T_mwith

respect to the corresponding ligands. The complexes

melted with decomposition except for 3 and 5. Thermal

decomposition of the ligands, in contrast to the com-

plexes, proceeded in a single step. The temperature of

5% weight loss (T₅), taken as the beginning of decomposi-

tion, was also higher for the Re(I) complexes. It can be

concluded that these compounds have good thermal sta-

bility with T₅in the range 221–350°C.

A summary of the intermolecular contacts in the

crystal structures of 1–6 is provided in spots of Hirshfeld

surfaces mapped with d_normand in two‐dimensional fin-

gerprint plots (Fig. S10 in supporting information). The

relative contributions of various intermolecular interac-

tions to the Hirshfeld surfaces are illustrated in Fig. 2.

As shown in Fig. 2 and Fig. S10 (supporting information),

FIGURE 2 Relative contributions of various intermolecular interactions to the Hirshfeld surfaces for complexes 4–6.

KLEMENS ET AL.

5 of 21

2.4 | Electrochemistry

The redox potentials of the free ligands L¹–L⁶and com-

plexes 1–6 were measured using cyclic voltammetry and

differential pulse voltammetry in MeCN with the aid of

0.1 M Bu₄NPF₆as the supporting electrolyte. The results

are summarized in Table S6. The onsets of the first oxida-

tion and reduction peak were used to estimate the ioniza-

tion potentials (IP) and the electron affinities (EA),

assuming IP of ferrocene to be equal to −5.1 eV.^[18]

The voltammograms of the free ligands display one

or two reduction processes in the negative potential

range. The first reversible reduction potentials were

found to be unaffected by introduction of the pyrrolidine

substituent into triimine ligand. The reduction of L⁴–L⁶

occurs at potentials similar to those for the correspond-

ing L¹–L³, which implies that the LUMOs of the ligands

are dominated by the contributions of π* orbitals of

terpy, dtpy and dppy skeleton. This conclusion is sup-

ported by the fact that the reduction potentials of L¹–L⁶

are sensitive to the peripheral rings of the heterocyclic

systems. As shown by the data summarized in Table

S6, the more negative reduction potentials are reported

for terpy derivatives (L¹and L⁴). Thus, π‐acceptor capa-

bility of the studied triimine ligands is improved by the

inclusion of thiazole and pyrazine rings. For the Re(I)

complexes, the reduction peaks were found at more

positive values compared to the corresponding ligands,

and the first reversible reduction potentials of 1–6 are

influenced by both peripheral rings and pyrrolidine

substituent of the ligand Lⁿ. The complexes bearing the

pyrrolidine moiety display more negative reduction

potentials in relation to the corresponding 1, 2 and 3,

which can be attributed to electron‐donor effect induced

by the pyrrolidine group. For both series of compounds,

L¹–L³and L⁴–L⁶, the first reduction potentials have

become more positive in the order pyridine < thiazole

< pyrazine, consistent with the trend observed for the

free ligands. This corresponds to lowering the LUMO

levels in both series 1–3 and 4–6 (Fig. 3).

FIGURE 3 HOMO and LUMO energy levels for PVK (as 7), PBD

(as 8) and IP and EA of synthesized compounds with respect to the

vacuum level.

importantly, all the examined Re(I) complexes exhibited

values of E_gthat make them potentially useful for opto-

electronic applications.

2.5 | UV–Visible Absorption Spectroscopy

The electronic absorption spectra of L¹–L⁶show a collec-

tion of bands in the range 374–191 nm. The position of

the lowest‐energy band (at 374–309 nm) is unaffected by

changes in solvent polarity (Table S7; Figs S12 and S13),

indicating that the ground and excited states have similar

dipole moments. Of note, the introduction of electron‐

donating pyrrolidine substituent results in significant red

shift and intensity increase of the lowest energy band.

The longest wavelength absorption peak of L⁴–L⁶

appeared at wavelengths more than 20 nm longer with

respect to that recorded for the ligands L¹–L³, and its

absorption coefficient is more than twice larger. This

enhanced and red‐shifted absorption of L⁴–L⁶can be

attributed to excitation leading to the formation of an

intramolecular charge transfer (ICT) state, which is a

result of charge delocalization from the pyrrolidine donor

unit to π‐conjugated triimine acceptor moiety. Due to

strong electron‐donating properties of pyrrolidine substit-

uent, the HOMO orbital of L⁴–L⁶is expected to rise higher

in energy compared to L¹–L³, which leads to the shrinking

of the HOMO–LUMO energy gap and a red shift in absorp-

tion spectra. The location of the lowest energy absorption

band of the free ligands is also sensitive to the donor–

acceptor properties of the peripheral rings: pyridine in L¹

and L⁴, thiazole in L²and L⁵and pyrazine in L³and L⁶.

For both series, L¹–L³and L⁴–L⁶, the red shift was in the

order thiazole > pyrazine > pyridine.

The first oxidation potentials of the Re(I) complexes

bearing the pyrrolidine moiety (4–6) were found at very

low values. These values are insensitive to the peripheral

rings and they all fall in the range observed for the free

ligands L⁴–L⁶. Most probably, the oxidation process in

L⁴–L⁶and 4–6 is associated with oxidation of the pyrrol-

idine moiety itself, and the HOMO of these systems is

chiefly influenced by the N‐donor substituent. With refer-

ence to previous studies,^[19]the redox waves at more pos-

itive potentials, for both 1–3 and 4–6, can be assigned to

the metal‐centred Re(I/II) oxidation process.

The electrochemical band gaps E_gare in satisfactory

Upon formation of 1–6, the spin‐allowed intraligand

transitions of the substituted terpy, dtpy and dppy ligands

agreement with the optical band gaps E_g^opt, but, more

6 of 21

KLEMENS ET AL.

are slightly red‐shifted (falling in the range 200–375 nm)

and an additional absorption emerges in the visible part

of each spectrum (Table S7 and Fig. 4a). Of note, the low-

est‐energy band of 4, 5 and 6 has a significantly higher

extinction coefficient and it is red‐shifted by 24–50 nm

in comparison with 1, 2 and 3, respectively (Fig. 4b).

These variations in absorption characteristics may indi-

cate changes in the character of the electronic transitions:

¹MLCT arising from the d_π(Re) → π*_Ltransitions for 1–

3^{[8b,c,e,h–k]}and ¹ILCT originating from charge transfer

from the pyrrolidine donor to π‐conjugated triimine

acceptor for 4–6.^[20]Such an assignment is consistent

with the electrochemical data indicating that the first oxi-

dation for 4–6 is centred at the pyrrolidine unit as well as

being supported by theoretical calculations (see

Section 2.9). Nevertheless, due to some overlap of MLCT

and ILCT transitions, it is not possible to completely

exclude the participation of Re_d→ π*_triimineexcitations

in the lowest energy band of 4–6.

As evident from Table S7 and shown in Fig. 4, the

energies of MLCT bands for 1–3 and ILCT bands for 4–6

are influenced by acceptor properties of the peripheral

rings of the triimine skeleton (pyridine in 1 and 4, thiazole

in 2 and 5 and pyrazine in 3 and 6). For both 1–3 and 4–6,

the red shift was in the order pyrazine > thiazole > pyri-

dine, which implies the strongest electron‐withdrawing

properties for compounds bearing dppy ligands.

respect to the solution phase (cf. Fig. S14). Such a shift

of λ_maxis noted for various compounds and it can be

explained by ‘J’ type aggregation, which takes place when

molecules stack in a head‐to‐tail arrangement.^[21]The

first absorption band in UV–visible spectra of Re(I) com-

plexes in film is slightly shifted to higher energy spectral

range compared to the chloroform solution (cf. Fig. S14

and Table S7).

2.6 | Photoluminescence

A summary of the photophysical data for L¹–L⁶and 1–6 is

given in Tables S8 and S9, respectively, and the normal-

ized excitation spectra, emission spectra and decay curves

are shown in Fig. S15 in supporting information.

In solution, the free ligands L¹–L³exhibited weak or

moderate emission in the near‐UV region of 340–400 nm

(Fig. S16). Fluorescence of the ligands bearing electron‐

rich pyrrolidine substituent (L⁴–L⁶) appeared in a signifi-

cantly lower energy region (462–645 nm). There is also a

marked difference in the fluorescence maximum wave-

length depending on the electronic properties of the

peripheral rings of the triimine ligands. The wavelength

of the ligand emission is significantly red‐shifted in the

orders L³> L²> L¹and L⁶> L⁵> L⁴.

In contrast to L¹–L³, the fluorescence maximum

wavelength of the ligands containing pendant amine

group (L⁴–L⁶) is strongly dependent on the solvent polar-

ity. On passing from chloroform to acetonitrile, the emis-

sion of L⁴, L⁵and L⁶shifted by 74, 105 and 123 nm

towards longer wavelengths, respectively. Positive

solvatochromism, which indicates better stabilization of

the excited state in more polar solvents, is expected for

ICT transitions. To further evaluate the ICT character of

the excited state in the case of L⁴–L⁶, the difference

between the excited and ground state dipole moments

(Δμ = μ_e− μ_g) was estimated using the Lippert–Mataga

equation (equation (1)):

Expectedly for charge transfer bands, the position of

the lowest energy absorption of 1–6 is affected by changes

in solvent polarity. On passing from acetonitrile to chlo-

roform solution, the longest wavelength absorption band

of the complexes is red‐shifted by 22.5 nm for 1, 28.8 nm

for 2, 26.4 nm for 3, 6.8 nm for 4, 10.7 nm for 5 and

15.9 nm for 6. In addition, the UV–visible spectra for

selected compounds (L¹, L⁴, L⁵, 1, 4 and 5) as thin films

on a glass substrate were recorded. In the case of the

ligands, the position of maximum absorption band (λ_max

)

in the solid state was bathochromically shifted with

FIGURE 4 (a) UV–visible spectra of ligands and Re(I) complexes in MeCN demonstrating the impact of pyrrolidine substituent. (b) UV–

visible spectra of Re(I) complexes in MeCN demonstrating the impact of peripheral rings.

KLEMENS ET AL.

7 of 21

ꢀ

ꢁ

2

the ground state, and thus small fluorescence quantum

yield.^[24]The most considerable decrease was observed

for compound L⁶, and its quantum efficiency in MeCN is

ca 20 times lower compared to chloroform. Expectedly,

fluorescence quantum yields of L¹–L⁶are strongly depen-

dent on their own molecular structure. The attachment of

pyrrolidine to the phenyl ring of terpy and dppy deriva-

tives resulted in a significant increase in the fluorescence

quantum yield. On the contrary, a marked drop in fluores-

cence quantum yields is seen upon replacing the pyridine

peripheral ring (L¹and L⁴) by pyrazine (L³and L⁶). All the

ligands reported here were also found to be emissive in

the solid state, and stronger emissive properties were

found for the ligands bearing the electron‐rich pyrrolidine

moiety, except for dtpy derivatives. Investigation of effect

of excitation wavelength (λ_ex) on PL of the ligands in film

showed that the highest PL intensity was recorded under

λ_ex= 310 nm for L¹–L³and L⁶or 330 nm for L⁴and L⁵

(cf. Fig. S17). The compounds with pyrrolidine substituent

(L⁴–L⁶) as powder and in the form of a layer on a glass

substrate exhibited green fluorescence, while the fluores-

cence of L¹–L³was generally located in near‐UV or blue

spectral regions (Fig. S16). The presence of two bands in

the case of L²and L³may be attributed to coexistence of

monomer and excimer‐like emissions.^[25]Except for L⁶,

the maximum of the emission band (λ_em) of the ligands

in the form of a layer on a glass substrate was

bathochromically shifted with respect to the solution

phase (in CHCl₃) (cf. Fig. S18a,b). In the case of the

ligands without pyrrolidine substituent (L¹–L³), the red

shift is rather small and falls in the range 10–19 nm. More

pronounced shift, of about 70 nm, was observed in the PL

spectra of the ligands bearing pyrrolidine unit (L⁴and L⁵).

Such a shift may indicate, among others, the contribution

2 μ_e−μ_g

ΔE_exc‐em

¼

Δf þ const:

(1)

(2)

hca³

ε − 1

2ε þ 1 2n²þ 1

n²− 1

Δf ¼

−

where ΔE_exc‐emis the Stokes shift (cm⁻¹), h is Planck's

constant, c is the speed of light in vacuum, a is the Onsa-

ger cavity radius and Δ f is the orientation polarizability

of the solvent, which measures both electron mobility

and dipole moment of the solvent molecule. Δ f was

calculated using equation (2), where ε is the dielectric

constant of the solvent and n the optical refractive index

of the solvent. Onsager cavity radii (a) were calculated

theoretically with use of Gaussian09^[22]and are collected

in Table S10 in supporting information. The structured

profile of the emission of the ligands bearing electron‐rich

pyrrolidine substituent (Fig. 5) may be attributed to

locally excited terpy‐, dtpy‐ and dppy‐centred π → π*

transitions, which lack the charge transfer character.

With increased solvent polarity, the vibronic structure of

emission spectra turned to broad unstructured profile,

suggesting a conversion from locally excited to ICT state.

The large values of Δμ (21.4, 23.1 and 21.8 D for L⁴, L⁵

and L⁶, respectively) estimated by fitting the emission

data to the Lippert–Mataga equation imply a dominant

contribution of ICT character in the excited state.^[23]

Measurements of the PL emission quantum yields

revealed a marked decrease in the fluorescence yield of

L⁴–L⁶with increasing polarity, which is supportive for

acceleration of ICT → TICT. The formation of a TICT

state leads to orbitally decoupled π systems of the donor

and acceptor units with forbidden radiative transition to

FIGURE 5 Emission spectra of ligands L⁴–L⁶in various solvents with Lippert–Mataga plots.

8 of 21

KLEMENS ET AL.

of ILCT transitions and the presence of strong excited‐

state interactions between the neighbouring molecules.

In PL spectra of the ligands dispersed in a poly(9‐

vinylcarbazole)

(PVK):2‐(4‐tert‐butylphenyl)‐5‐(4‐

biphenylyl)‐1,3,4‐oxadiazole (PBD) matrix, an emission

band with λ_emlocated in the range from 395 to 442 nm

was observed. Noteworthy, the emission of ligands L¹–L⁴

corresponded to matrix emission (cf. Fig. S18a,b). In the

case of compounds L⁵and L⁶in matrix, λ_emwas

bathochromically shifted with respect to λ_emof PVK:

PBD (cf. Fig. S18a,b).

Excitation of 1–6 in solution at the lowest absorption

band gave rise to an emission band with a maximum

located in the red spectral region. The decrease in

luminescence intensity compared to the free ligands can

be explained by the smaller energy gap and faster

non‐radiative decay according to the energy gap law.

According to the data summarized in Table S9, different

emission properties were found for series 1–3 and 4–6.

Most importantly, the complexes bearing the triimine

ligands with pyrrolidine substituent exhibited different

emission profiles in polar acetonitrile and non‐polar

chloroform, which may indicate changes in the nature

of the excited states depending on solvent polarity. For

complexes 1–6 being MLCT emitters, the incorporation

of the pyrolidine group, which act as electron‐donating

substituent, is expected to destabilize the MLCT excited

state leading to hypsochromic shifts of emission in

relation to the parent Re(I) complexes – not having

electron‐rich moiety attached to the ligand L. This trend

was seen in non‐polar chloroform for 4–6 in relation to

complexes 1–3 being pure MLCT emitters. Of note, the

emission spectra of 4 and 5 in CHCl₃exhibited two

well‐separated bands, at 507 and 677 nm for 4 and

536 and 686 nm for 5. For both compounds, the intensity

of the short‐wavelength emission decreased with

increasing excitation wavelengths (Fig. 6). In analogy to

the previously reported Re(I) polypyridyl compounds

fac‐[Re(L)(CO)₃(cbz₂phen)]^+/0, where cbz₂phen = 4,7‐

di(9H‐carbazol‐9‐yl)‐1,10‐phenanthroline and L = Cl⁻,

pyridine or 2‐aminomethylpyridine,^8hit can be specu-

lated that the two emission bands of 4 and 5 originate

from different excited states. The higher energy band of

4 and 5 is suggested to be of singlet origin and ILCT char-

acter. The strong electron‐donating property of the pyr-

rolidine substituent gives rise to an ILCT transition

from the pyrrolidine donating group to the terpy

accepting moiety (see Section 2.9). Compared to the free

ligands, these bands are red‐shifted by 45 nm for 4 and

61 nm for 5, which is attributed to the impact of the metal

centre. The broader lower energy emission peak of 4 and

5, associated with a longer lifetime, is proposed to be of

triplet MLCT origin.

FIGURE 6 Excitation, emission and absorption of complexes 4

and 5 in CHCl₃.

For complex 6, excitation at the lowest absorption

band gave rise to one structureless emission band with a

maximum at 690 nm. Similar to 4 and 5, the emission

band is blue‐shifted in relation to the corresponding

unsubstituted complex 3, supporting MLCT character of

the excited state. The striking difference between complex

6 and complexes 4 and 5 concerning lack of higher energy

band of singlet origin and ILCT character in the case of 6

can be explained by the smaller energy gap and faster

non‐radiative decay for 6 in comparison with 4 and 5.

In more polar acetonitrile, the emission spectra of 4–6

displayed only a long‐wavelength band, with a maximum

at 702 nm for 4, 741 nm for 5 and 781 nm for 6. Com-

pared to the parent complexes without electron‐rich pyr-

rolidine group 1–3, the emission maximum of 4–6 is

bathochromically shifted. This behaviour is inconsistent

with that expected for pure MLCT emitters. In this case

the nature of the excited state seems to be ILCT. For both

series 1–3 and 4–6, bathochromic shift of the emission

band was in the order pyridine < thiazole < pyrazine,

consistent with the increase of π‐acceptor capability

(Table S9).

Upon cooling to low temperature (at 77 K in EtOH–

MeOH (4:1 v/v) glassy medium), the emission of the com-

plexes occurred at higher energy (543–632 nm), which is

attributed to the rigidochromic effect. This effect is

KLEMENS ET AL.

9 of 21

responsible for raising the energy of the emissive MLLCT

states due to the lack of solvent reorganization following

excitation.^[26]The emission bands of 1–6 are structure-

less, which is consistent with the charge transfer charac-

ter (Fig. S19 in supporting information). On the other

hand, the microsecond excited state lifetimes in concert

with large Stokes shifts are supportive of the phosphores-

cence assignment.

Except for 5 showing emission in two regions (Fig.

S21), the solid‐state emission spectra of the Re(I) com-

plexes exhibited a single structureless band with a maxi-

mum located in the orange‐red spectral region. Distinct

differences in lifetimes were observed between 1–3 and

4–6. The complexes bearing triimine ligand functional-

ized with pyrrolidine substituent exhibit significantly

slower decay rate. For all the complexes, lifetimes fall

into microsecond or sub‐microsecond range and quan-

tum yields, except for 3, are higher than those in solution.

The Re(I) complexes, likewise the ligands, were also

photoluminescent in the form of a thin film. In this case,

of L⁵from 1 and 2 to 15 wt% was noted, probably due to

aggregation of compound with increasing content.

A higher PL quantum yield was found for the blend

with 1 wt% of L⁵. In the case of complex 5, the increase

of its content from 1 to 2 to 15 wt% resulted in the

appearance of the second emission band at 598 nm.

Blend with 2 wt% of 5 showed higher PL quantum yield

with respect to matrix containing lower concentration of

5 (cf. Table S8).

2.7 | Identification of Excited Triplet State

using Laser Flash Photolysis

To confirm the triplet excited state, compound 4 was

studied by recording the nanosecond time‐resolved tran-

sient absorption spectra in acetonitrile and chloroform,

and the results are shown in Fig. 8. The positive band,

spanning from ca 480 to ca 700 nm, is separated from

the ground‐state bleaching with well‐defined isosbestic

points, and the bleaching recovery is synchronous

with the positive absorption decay. This indicates that

the ground state is repopulated (T₁→ S₀) as the positive

T₁–T_nabsorption decays. The decay of T₁absorption is

mono‐exponential, and transient triplet state is trans-

formed to S₀within a submicrosecond scale. The triplet

lifetime of 4 (774 ns) in argon‐saturated acetonitrile is

sufficiently long for photosensitizing the production of

the effect of λ_exnot only on PL intensity, but also on λ_em

,

is pronounced (cf. Fig. S17b). PL of all the films was

dependent on excitation wavelength. Together with a

decrease of λ_exenergy from 310 nm (4 eV) to 340 nm

(3.6 eV), a bathochromic shift of λ_emfrom 375 to

413 nm was seen. In the PL spectra of complexes 1–4,

recorded using the same λ_ex,two emission bands with

λ_emat 370–380 and 572–609 nm can be distinguished.

The emission spectra of all the films and blends are very

similar and they are dominated by the band with λ_em

around 380–410 nm.

3

1

singlet oxygen (T₁+ O₂→ S₀+ O₂). In air‐saturated

solution, there is a significant decrease of the τ_Tvalue to

162 ns. To determine the singlet oxygen quantum yields

(Φ_Δ) of 4, the luminescence method was used.

In addition, λ_emof the complexes in the form of film

was hypsochromically shifted with respect to that of the

complexes in solution. This not so typical phenomenon,

that is, the blue shift of the solid‐state PL emission in

comparison to solution, was also reported for other

compounds.^[27]The effect of luminophore content in

matrix on PL was evaluated for compounds bearing dtpy

(L⁵and 5) (cf. Fig. 7). The shift of the peak emission to

longer wavelengths with increasing doping concentration

2.8 | Electroluminescence

The EL abilities of both the ligands and complexes were

investigated in diodes with guest–host configuration. As

active layers in the devices, the layers of the same compo-

sition as those used for the PL measurements, that is,

neat compounds and blends with PVK:PBD, were

applied. Operation of

a diode with a guest–host

FIGURE 7 PL spectra of ligand L⁵and complex 5 in film form and blends with various compound contents.

10 of 21

KLEMENS ET AL.

FIGURE 8 Nanosecond transient absorption spectra of 4 measured in MeCN and CHCl₃at room temperature (excitation wavelength λ_ex

of 355 nm). Transient decays were monitored at 570 nm for 4 in MeCN and at 600 nm for 4 in MeCN.

configuration is based on two mechanisms, that is, energy

transfer and charge trapping.^[28]In the energy transfer

mechanism, singlet excitons are formed in the host

matrix due to electrical excitation and they can then be

transferred to the guest molecule via Förster resonance

energy transfer (FRET). FRET efficiency significantly

depends on the spectral overlap between the matrix emis-

sion and the luminophore absorption. In the charge trap-

ping mechanism, the excitons are formed directly on the

luminophore without the necessity of energy transfer

from host to guest. For effective charge trapping, the

HOMO and LUMO of the guest need to lie within the

TABLE 1 Position of λ_ELof light emitted by diodes and turn‐on voltage together with the thickness of selected active layers

ITO/PEDOT:PSS/compound/Al ITO/PEDOT:PSS/PVK:PBD:compound/Al

Code

V_on(V)

λ_EL(nm) (d, nm)

V_on(V)

λ_EL(nm) (d, nm)

L¹

L²

L³

L⁴

L⁵

1.6

1.1

1.5

2.6

—

3.4

1.7

2.8

1.8

614

— (76)

—

590

533, 580 (90)

590 (125)

659

654

2.5

500

2.0^a

5.4^b

473^a

547^b(72)

L⁶

1

1.4

1.2

1.1

2.2

2.8

2.0

—

2.5

2.8

2.1

545

600

625

600

583

—

2

651

660

638

657

3

4

5

1.7

606 (109)

602^a

623^b(106)

2.0^a

6.8^b

6

1.1

655

1.0

609

^a1 wt% content of compound in PVK:PBD matrix.

^b15 wt% content of compound in PVK:PBD matrix.

KLEMENS ET AL.

11 of 21

bandgap of the host. Considering the UV–visible spectra

of ligands, partial spectral overlap between guest absorp-

tion and PVK:PBD emission only in the case of L⁴–L⁶was

seen (cf. Figs S12 and S14). The Re(I) complexes exhibited

better spectral overlap. Based on IP and EA values, esti-

mated from cyclic voltammetry measurements (corre-

sponding the HOMO and LUMO energies, respectively),

it can be noted that all of the compounds (except for L¹

and L²) have their HOMO and LUMO levels appropri-

ately aligned with respect to the corresponding levels of

PVK and PBD, to ensure trapping electrons from PBD

and holes from PVK (cf. Fig. S17c).

Diodes with the structures ITO/PEDOT:PSS/com-

pound/Al and ITO/PEDOT:PSS/PVK:PBD:compound/Al

were fabricated. For the prepared devices, current

density–voltage (J–V) characteristics were recorded up

to 5 V, and for selected active layers, thickness was

measured by atomic force microscopy. The obtained

turn‐on voltage (V_on), position of EL band maximum

(λ_EL) and thickness of active layer (d) are summarized in

Table 1. The representative J–V characteristics and EL

spectra of diodes based on L⁵and complexes with pyrrol-

idine unit (4–6) together with intensity dependencies on

the applied external voltage are depicted in Fig. 9. EL

spectra of other diodes are presented in Fig. S20.

The turn on‐voltages of all the diodes were found to

be in a similar range of 1.0–2.8 V, except for device with

L¹blend, L⁵(15 wt% in matrix) and 5 (1 and 15 wt% in

PVK:PBD), which were equal to 3.4, 5.4 and 4.6 and

6.8 V, respectively. Most of the fabricated diodes emitted

light under external voltage differing in intensity and

position of λ_EL, depending on both the chemical structure

of the compound and the composition of the active layer.

Diodes based on neat ligands without pyrrolidine

FIGURE 9 EL spectra of diodes based on (a–d) L⁵with various ligand contents in active layer together with (e) J–V characteristics and (f–i)

neat complexes 4, 5 and 6 together with device with 6 dispersed in PVK:PBD matrix.

12 of 21

KLEMENS ET AL.

substituent (L¹–L³) and with dppy (L⁶) did not show EL

response. Other ligand‐containing diodes emitted light

with λ_ELranging from 500 to 659 nm. The highest inten-

sity of emitted light was found for the device containing

as active layer the neat ligand with pyrrolidine and dtpy

units (L⁵). Considering the devices based on ligands

molecularly dispersed in a matrix, it was found that the

lowest EL intensity was observed in diodes with ligand

bearing terpy (L¹). Increase of the ligand (L⁵) content in

blend resulted in bathochromic shift of λ_ELfrom blue to

green spectral range. The position of λ_ELof light emitted

by diodes containing the neat ligands and complexes

was hypsochromically shifted with respect to devices with

the compounds blended with PVK:PBD. Focusing on the

diodes based on Re(I) complexes, it was found that only

in the case of the device with the compound containing

terpy without pyrrolidine unit (1) was no light emission

observed. Utilization of the complexes blended with

PVK:PBD as active layers resulted in an increase of

emitted light intensity. Similar to the diode with L⁵, a

higher content of the complex 5 in blend shifted the λ_EL

position to lower energy, but to a smaller extent. Consid-

ering the intensity of emitted light from diodes differing

in L⁵and 5 content, it can be noted that the optimal

amount of compound in blend is 2 wt%, which gives the

highest intensity. All devices based on complexes emitted

red light, except for 4 dispersed in a matrix (cf. Table 1).

Replacement of free ligand L⁴with analogous complex 4

resulted in a slight shift of λ_ELfrom 659 to 638 nm.

Generally, application as active layer of both ligands

and complexes in a matrix caused an increase of emitted

light intensity.

2.9 | Theoretical Calculations

DFT and time‐dependent DFT (TD‐DFT) methods were

applied to describe the electronic structure of excited

states of ligands L¹–L⁶and complexes 1–6. The excitation

FIGURE 10 Energy diagram of frontier molecular orbitals for L¹and L⁴ligands.

KLEMENS ET AL.

13 of 21

energies were calculated in the wavelength range from ca

200 to ca 500 nm. Obtained wavelengths, oscillator

strengths and experimental absorption spectra are pre-

sented in Fig. 10 and Figs S22 and S23. The energies

and characteristics of selected singlet electronic transi-

tions, together with assignment to the experimental

absorption bands are presented in Tables S11–S22. For

L¹–L⁶, the electronic excitations with the largest oscillator

strengths come mainly from π → π* transitions and gen-

erally can be characterized as π_L→ π*_L, π_R→ π*_Rand

π_L/π_R→ π*_L/π*_Rexcitations, where L and R denote

triimine skeleton and substituent group, respectively. In

the low‐energy part of the calculated spectrum, most of

the mentioned electronic transitions have rather charge

transfer character. Occupied and unoccupied π orbitals,

involved in charge transfer electronic transitions, are

localized on various parts of the molecule, namely on

terpyridine/dithiazolylpyridine/dipyrazinylpyridine moi-

ety or R substituent (Fig. 10).

Analysing the results of calculations, it is possible to

notice that the energy order of the lowest excited states

of π_R→ π*_L, π_L→ π*_Rand π_L→ π*_Ltype is different

for L¹–L³and L⁴–L⁶groups of ligands. For L¹the lowest

electronic transition to S₁state with medium oscillator

strength (0.2131) has π_terpy→ π*_terpycharacter. The next

S₄, S₅and S₆electronic transitions of larger oscillator

strength (0.3599, 0.1751, 0.3754) can be characterized in

sequence as π_terpy→ π*_terpy/π*_R, π_terpy/π_R→ π*_terpy/π*_R

and π_terpy/π_R→ π* excitations. For L⁴the first S₁elec-

tronic transition has very small oscillator strength

(0.0002) and is of π_R→ π*_terpy/π*_Rtype, while the second

S₂transition of π_R→ π*_terpycharacter has large value of

oscillator strength (0.5734). Similar π_R→ π*_terpycharacter

of excitation can be also assigned to S₃electronic state.

The charge transfer π_terpy→ π*_terpy/π*_Rand local πterpy-

π*_terpytransitions correspond to S⁴and S⁸excited states,

for which the energy of excitation is very close to the exci-

tation energy of similar electronic transitions for L¹.

Overall, for L¹the lowest intense electronic transition

occurs to the S¹state and it is related to the partial charge

transfer from terpyridine to phenyl ring, whereas for L⁴

the first intense transition involves S²excited state and

it is an electronic excitation with the charge transfer in

the opposite direction, namely from phenyl ring on

terpyridine fragment.

For all the ligands with electron‐donor pyrrolidine

substituent (L⁴–L⁶), in the range from about 370 to about

390 nm, calculated electronic spectra have characteristic

low‐lying π_R→ π*_L/π*_Rtransition with large oscillator

strength. This intense electronic excitation can be

assigned to the first band in experimental spectrum.

However, as can be seen in Fig. S22, it is clearly shifted

in the direction of lower energies in relation to the maxi-

mum of the experimental band and the average differ-

ence between the calculated excitation energy and

maximum band is about 0.36, 0.18 and 0.22 eV for L⁴,

L⁵and L⁶, respectively. It is widely held that charge trans-

fer excitation energies are often significantly

underestimated using standard local xc‐functionals. For

L⁴and S₂low‐lying excited states, test calculations were

performed with the use generalized gradient approxima-

tion (GGA), hybrid exchange correlation, and long‐range

separated exchange correlation functionals. The test cal-

culations were performed in order to estimate the effect

of the functional on the excitation energy of S₂electronic

state. Obtained results are presented in Fig. S24. Pure

GGA functionals significantly underestimate the excita-

tion energy of S₂state. The average difference in energy

between the lowest experimental band and the calculated

S₂transitions in the case of GGA functionals is equal to

1.18 eV. Hybrid functionals also underestimate energy

of this excitation, but with smaller average difference,

which amounts to 0.50 eV. Amongst tested hybrid func-

tionals, two of them, M062x and BHandHLYP, are excep-

tions, because both overestimate the excitation energy

by about 0.25 and 0.38 eV, respectively. Long‐range

separated functionals also overestimate or underestimate

excitation energy of S₂state. In the case of most of the

tested long‐range separated functionals, excitation energy

to S₂state is overestimated, by an average of around

0.55 eV; however, the two functionals OHSE1PBE and

OHSE2PBE give lower excitation energies as compared

to the energy of maximum of the lowest experimental

band. Based on the obtained results, it is possible to sug-

gest that most of the used hybrid exchange correlation

functionals give a good approximation of the excitation

energy for the S₂electronic state.

The differences in the electronic structure of lowest

excited states between L¹–L³and L⁴–L⁶are visible very

well in the molecular orbital picture (Fig. 10). The

HOMO (H) in L¹is mainly localized on terpyridine frag-

ment; simultaneously, the occupied H − 1 orbital is local-

ized on phenyl ring and partly on the central ring of

terpyridine. In L⁴, two HOMOs (H and H − 1) have very

similar character, but their energy order is reversed. The

lowest unoccupied orbitals, L and L + 1, in L¹are π* anti-

bonding orbitals. The first of them (L) is a π* orbital

partially delocalized on the entire structure; the second

The computational results show that the electronic

structure and remaining properties of lowest excited state

of the L², L³and L⁵, L⁶species are very similar to those of

L¹and L⁴, respectively. It can be concluded that differ-

ences in lowest excited electronic states between L¹–L³

and L⁴–L⁶are related to changes in the electronic struc-

ture resulting from the presence of pyrrolidine substitu-

ent attached to the phenyl group.

14 of 21

KLEMENS ET AL.

is localized solely on terpyridine fragment. For L⁴two

LUMOs have the same character, but the reversed order.

The L and L + 1 are π*_terpyand π*_terpy/π*_Rorbitals,

respectively. The HOMO orbital in L⁴has π_Ph–p_Ncharac-

ter and is antibonding combination of π orbital of phenyl

ring with p_Norbital of nitrogen in pyrrolidine substituent.

This antibonding combination effectively increases the

energy of the HOMO orbital compared to the similar

H − 1 orbital in L¹, causing the lowest electronic excita-

tions to be dominated by transitions of π_R→ π*_terpytype.

To verify the charge transfer problem of low‐lying

excited states of L¹and L⁴ligands, the Λ parameter was

calculated. Parameter Λ is a simple diagnostic test based

on orbital overlap and may be used to help judge the reli-

ability of excitation energies in TD‐DFT.^[29]It was clearly

demonstrated that parameter Λ correlates with error in

excitation energies, i.e. excitation energies are signifi-

cantly underestimated when Λ is very small, whereas

the errors tend to be smaller when Λ is large. It has also

been observed that for hybrid functional and Λ > 0.3,

the estimated excitation energies for charge transfer tran-

sitions have rather acceptable values. The calculated Λ

parameters for ten lowest singlet excited states of L¹and

L⁴ligands are in the range from 0.2 to 0.8 (Table S23).

For all π → π* transitions with large or medium oscillator

strength, Λ parameters are greater than 0.3. Only in the

case of L⁴ligand, S₁, S₃and S₉electronic excitations with

small values of oscillator strength have Λ below 0.3.

These parameters are 0.27, 0.21 and 0.2 for S₁, S₃and S₉

transitions, respectively. It should be emphasized that

problematic S₂transition for L⁴has an acceptable Λ value

for the applied functional and obtained excitation energy

probably should not be considerably underestimated. In

summary, the charge transfer problem rather is not

appreciable in the theoretical description of low‐lying sin-

glet electronic states of the investigated ligands.

The CASSCF/MS‐CASPT2 level of theory was applied

for additional verification of the TD‐DFT results. Multi-

state version of CASSCF and CASPT2 was used to deter-

mine the excitation energy and the description of the

electronic structure of the four lowest π → π* excited

states for L¹and L⁴. The calculations were done based

on the CAS(12,12) active space. Active space contains 12

electrons as well as 6 occupied and 6 unoccupied π

orbitals (Figs S25 and S26). The active space was selected

to contain π orbitals, similar to those that are involved in

the TD‐DFT excitations, and in this way, calculated

excited states should correspond to the lowest π → π*

excited states at TD‐DFT level of theory. The MS‐CASPT2

results, collected in Tables S17 and S18, correlate well

with the TD‐DFT calculations, but some differences are

also apparent. For L¹the electronic structure of S₁and

S₂states can be characterized as π_terpy/π_R→ π*_terpy/π*_R

and π_terpy/π_R→ π*_terpyelectronic excitations and well

correspond to S₁and S₂states at TD‐DFT level of theory.

Both states come from excitations between orbitals local-

ized mainly on the terpyridine fragment, but due to the

fact that the unoccupied orbital in active space is partially

delocalized on the phenyl ring (82 L, Fig. S25), some con-

tribution of the π_terpy→ π*_Rcharge transfer excitation is

also visible. The S₁and S₂electronic states for L⁴have

π_R/π_terpy→ π*_terpy/π*_Rand π_R/π_terpy→ π*_terpycharacter

and can be assigned to the first two excitations in TD‐

DFT calculations. Occupied orbital involved in excitation

to the S₁and S₂state is mainly localized on the phenyl

ring (98 H − 2, Fig. S26), so in contrast to the two lowest

transitions for L¹, low lying electronic states for L⁴are

rather associated with the dominating participation of

charge transfer between R fragment and terpyridine.

Excitation to S₂state, which corresponds to a large oscil-

lator strength (0.70), is practically a clean charge transfer

transition of a π_R→ π*_terpytype, which remains in satis-

factory accordance with TD‐DFT results. The S₁state

has a more mixed character and electronic excitation for

this state has a much lower value of the oscillator

strength (0.02). The characteristic difference of the oscil-

lator strength between S₁and S₂is very similar to that

which can be observed in the TD‐DFT calculations. The

MS‐CASPT2 results once again show the essential differ-

ence between characters of the lowest charge transfer

states for ligands with the substituted and unsubstituted

phenyl ring. For L¹ligand, in which the phenyl ring is

unsubstituted, intense low‐lying π → π* electronic transi-

tions occur mainly within terpyridine motif but with

some participation of terpyridine → R charge transfer

transition, while in the case of substituted ring as in L⁴,

low‐lying transition is an electron excitation mainly of a

R → terpyridine charge transfer character. From an

energetics point of view, the calculated MS‐CASPT2

excitation energies of the two lowest transitions for L¹

and L⁴ligands are always higher in relation to appropri-

ate values for TD‐DFT states. Average difference of the

calculated excitation energies, between the relation to

appropriate values for TD‐DFT and MS‐CASPT2 results,

is about 0.51 eV. Compared to the energy of maximum

of the lowest experimental band, MS‐CASPT2 excitation

energies are also higher in energy with the exception of

S₁excitation for L⁴. The calculated MS‐CASPT2 S₁and

S₂excitations energies for L¹and L⁴, respectively, are

higher in energy by about 0.56 and 0.22 eV compared to

the position of lowest experimental band. Lastly, it is

crucial to point out that the S₂electronic state, described

as ‘problematic’ in the TD‐DFT calculation, is also a sec-

ond lowest excited state at the MS‐CASPT2 level and has

explicitly the π_R→ π*_terpycharge transfer character.

Based on this result, it is possible to suggest that for L⁴,

KLEMENS ET AL.

15 of 21

FIGURE 11 Experimental and theoretical UV–visible spectra of complexes 1 and 4 in MeCN as well as composition of selected frontier

molecular orbitals (light blue, terpy; violet, R; green, Cl; red, 3CO; dark blue, Re).

excitation energy of intense S₂transition estimated using

PBE1PBE or other hybrid exchange correlation func-

tionals well correlates with the lowest experimental band.

In the case of both ligands L¹and L⁴, the remaining two

electronic states (S₃and S₄) at MS‐CASPT2 level have a

more complex electronic structure, and assigning them

to the relevant π → π* states at the TD‐DFT level of the-

ory becomes ambiguous and difficult.

energy in relation to the maximum of first experimental

band. The difference between the calculated excitation

energy and maximum band is 0.29 eV for 4, 0.20 eV for

5 and 0.33 eV for 6. For 1–3, the high‐wavelength range

of theoretical absorption spectra comprises electronic

transitions with smaller oscillator strengths (0.0034–

0.1094) of MLLCT character (d_Re→ π*_L/π_CO→ π*_L/

p_Cl→ π*_L) and energy falling within the first experimen-

tal band (Fig. 11).

It can be supposed that for Re(I) complexes with the

ligand containing pyrrolidine substituent attached to the

phenyl group, the first low‐lying electronic excitation will

not engage the orbitals localized on the metal, but it will

be determined by the electronic transition localized only

on the ligand (Fig. S27). DFT and TD‐DFT studies

showed that the HOMO, HOMO − 1 and HOMO − 2

orbitals of 1–3 contain significant 5d rhenium character

(45, 42 and 56%, respectively) along with ca 25% contribu-

tions from CO component, whereas HOMOs of 4–6 are π

orbitals of the trimine ligand, predominately centred on

the R substituent – 80.19% for 4, 81.42% for 5 and

80.66% for 6. For 4–6, the orbitals 5d_π(Re) and π*(CO)

contribute to HOMO − 1 and HOMO − 2 and HOMO − 3.

For all of 1–6_,however, the LUMOs have similar distribu-

tion and they are largely participated by π* orbitals

localized on the terpy in 1 and 4, dtpy in 2 and 5 and

dippy in 3 and 6. The high‐wavelength range of theoreti-

cal absorption spectra of 4–6 is dominated by intense

transitions, mainly of ILCT (π_R→ π*_L) character. This

intense electronic excitation can be assigned to first

band in experimental spectrum. Similar to the free

ligands L⁴–L⁶, it is clearly shifted in the direction of lower

The energies of phosphorescence emissions for 1–6

were calculated from the optimization of the first triplet

excited state and from the energy difference between

the ground singlet and the triplet state ΔE_{T −S}. As shown

1

0

in Table S26, the lowest lying triplet state of Re(I) com-

3

plexes 1–3 has mainly MLLCT character, while ILCT

character dominates in T₁for remaining complexes with

the triimine ligand decorated with the pyrrolidine substit-

uent (4–6).

3 | CONCLUSIONS

To summarize, a series of triimine ligands, varied in the

chemical structure of peripheral rings (pyridine, thiazole

and pyrazine) and the presence or absence of pyrrolidine

substituent, as well as their corresponding Re(I)

carbonyls were prepared and investigated. The spectral

and electrochemical properties of the ligands and Re(I)

complexes bearing electron‐rich pyrrolidine group were

strongly influenced by ILCT transitions. For compounds

4–6, different emission profiles were found in polar and

16 of 21

KLEMENS ET AL.

non‐polar solvents, indicating change in the character of

available and they were used without further purification.

The films and blends with PVK:PBD (50:50 w/w) on a

glass substrate as well as all the devices with configura-

tions ITO:PEDOT:PSS/compound/Al and ITO:PEDOT:

PSS/PVK:PBD:compound/Al with complex content of

15, 2 and 1 wt% in blend were prepared according to

the method reported in our previous work.^1l

the excited state. In chloroform, presence of two emitting

1

states, identified as an ILCT excited state deactivated at

higher energies and a longer‐lived red‐shifted phospho-

3

rescence attributed to the MLCT excited state, was evi-

denced in the case of 4 and 5. On the contrary, the

emission of 4–6 in acetonitrile, bathochromically shifted

compared to that for the corresponding 1–3, seems to be

of ILCT nature. Consistent with increase of π‐acceptor

capability in the order pyridine < thiazole < pyrazine,

the emission of the free ligand and Re(I) complexes bear-

ing dppy unit was significantly red‐shifted in relation to

the corresponding analogues incorporating terpy moiety.

All of the synthesized compounds showed high melt-

ing points ranging from 201 to 349°C, higher in the case

of Re(I) complexes. The beginning of decomposition was

noted above T_mand Re(I) complexes showed higher T₅

(221–354°C) value than the corresponding ligands. All

the ligands with pyrrolidine substituent and complex

bearing dtpy (5) exhibited behaviour of molecular glasses

with T_gfrom 74 to 93 and 205°C. The presence of dppy

unit in the ligand and complex increases both T_mand

T_gvalues, but slightly reduced their solubility. The lack

of pyrrolidine substituent attached to the phenyl group

lowers their solubility. The Re(I) complexes showed a

lower EA value, about 0.61–0.84 eV, than the analogous

ligands. Pyrrolidine moiety in the ligands increases the

IP value. Diodes with active layer consisting of neat

ligands and complex 1 did not show an EL response con-

trary to others. However, in most cases, better device per-

formance was found in diodes with guest–host

configuration. Generally, diodes based on ligands emitted

green light with λ_ELlocated between 500 and 590 nm. In

all devices with complexes (except for 4), red emission

with λ_ELin the range 600–657 nm was seen. Considering

the results of the preliminary EL investigations, it can be

pointed out that the most promising for further OLED

parameter estimation, including also modification of

diode architecture, seems to be the compounds bearing

pyrrolidine substituent, that is, ligands without terpy (L⁵

and L⁶) and the corresponding complexes (4–6).

4.2 | General Synthesis Route of Ligands

(L¹–L⁶)

Ketone (20 mmol; 2‐acetylpyridine, 2‐acetylthiazole or 2‐

acetylpyrazine) was added to a solution of aldehyde

(10 mmol; benzaldehyde or 4‐pyrrolidinobenzaldehyde)

in ethanol (75 ml). KOH (1.54 g, 27.5 mmol) and aqueous

NH₃(35 ml) were then added. The solution was stirred at

room temperature for 24 h. The solid was collected by fil-

tration and washed with water. Recrystallization from

ethanol (L¹and L⁴) or toluene (L², L³, L⁵and L⁶) afforded

a crystalline solid.

1

L¹: Yield: 45%. H NMR (400 MHz, CDCl₃, δ, ppm):

8.75 (s, 2H), 8.74–8.71 (m, 2H), 8.66 (d, J = 8.0 Hz, 2H),

7.90 (d, J = 7.1 Hz, 2H), 7.85 (dt, J = 7.7, 1.8 Hz, 2H),

7.53–7.42 (m, 3H), 7.34–7.31 (m, 2H). ¹³C NMR

(100 MHz, CDCl₃, δ, ppm): 156.37, 156.03, 150.41,

149.23, 138.61, 136.92, 129.10, 129.01, 127.44, 123.89,

121.44, 119.02. HRMS (ESI): calcd for C₂₁H₁₆N₃

[M + H]⁺310.1344; found 310.1353.

1

L²: Yield: 52%. H NMR (400 MHz, CDCl₃, δ, ppm):

8.47 (s, 2H), 7.97 (d, J = 3.2 Hz, 2H), 7.87–7.83 (m, 2H),

7.57–7.46 (m, 5H). ¹³C NMR (100 MHz, CDCl₃, δ, ppm):

168.93, 151.71, 150.96, 144.26, 137.43, 129.76, 129.27,

127.35, 122.00, 118.00. HRMS (ESI): calcd for

C₁₇H₁₂N₃S₂[M + H]⁺322.0473; found 322.0473.

1

L³: Yield: 57%. H NMR (400 MHz, CDCl₃, δ, ppm):

9.87 (s, 2H), 8.73 (s, 2H), 8.68–8.65 (m, 4H), 7.90–7.84

(m, 2H), 7.58–7.45 (m, 3H). ¹³C NMR (100 MHz, CDCl₃,

δ, ppm): 154.65, 151.00, 144.93, 143.88, 143.86, 143.69,

129.60, 129.29, 127.39, 120.05. HRMS (ESI): calcd for

C₁₉H₁₄N₅[M + H]⁺312.1249; found 312.1248.

1

L⁴: Yield: 39%. H NMR (400 MHz, CDCl₃, δ, ppm):

8.73 (d, J = 4.7 Hz, 2H), 8.71 (s, 2H), 8.66 (d,

J = 8.4 Hz, 2H), 7.90–7.84 (m, 4H), 7.36–7.31 (m, 2H),

6.66 (d, J = 8.4 Hz, 2H), 3.37 (t, J = 5.9 Hz, 4H), 2.09–

1.99 (m, 4H). ¹³C NMR (100 MHz, CDCl₃, δ, ppm):

156.84, 155.72, 150.28, 149.14, 148.70, 136.89, 128.22,

124.60, 123.68, 121.48, 117.44, 111.92, 47.69, 25.59. HRMS

(ESI): calcd for C₂₅H₂₃N₄[M + H]⁺379.1923; found

379.1924.

4 | EXPERIMENTAL

4.1 | Materials

Re(CO)₅Cl (Sigma Aldrich), poly(9‐vinylcarbazole) (PVK;

M_n

= 25 000–50 000 g ; Sigma Aldrich),

mol⁻¹

poly(3,4(ethylenedioxy)thiophene):polystyrenesulfonate

(PEDOT:PSS; 0.1–1.0 S cm⁻¹) and substrates with

pixilated ITO anodes (supplied by Ossila) as well as all

solvents for synthesis (of reagent grade) and for spectro-

scopic studies (of HPLC grade) were commercially

1

L⁵: Yield: 43%. H NMR (400 MHz, CDCl₃, δ, ppm):

8.42 (s, 2H), 7.96 (d, J = 3.2 Hz, 2H), 7.79 (d, J = 8.8 Hz,

2H), 7.47 (d, J = 3.2 Hz, 2H), 6.63 (d, J = 8.8 Hz, 2H), 3.35

(t, J = 6.5 Hz, 4H), 2.05–2.01 (m, 4H). ¹³C NMR

KLEMENS ET AL.

17 of 21

(100 MHz, CDCl₃, δ, ppm): 169.59, 151.35, 150.75, 149.07,

144.07, 128.19, 123.23, 121.67, 116.22, 112.02, 47.70, 25.61.

HRMS (ESI): calcd for C₂₁H₁₉N₄S₂[M + H]⁺391.1051;

found 391.1053.

¹H NMR (400 MHz, DMSO‐d₆, δ, ppm): 9.09 (d,

J = 8.2 Hz, 1H), 9.04 (d, J = 5.3 Hz, 1H), 8.94 (s, 1H),

8.78 (d, J = 4.7 Hz, 1H), 8.37 (t, J = 7.9 Hz, 1H), 8.12

(d, J = 8.8 Hz, 2H), 8.04 (d, J = 9.5 Hz, 2H), 7.86 (d,

J = 7.8 Hz, 1H), 7.78–7.73 (m, 1H), 7.61 (dd, J = 7.4,

4.9 Hz, 1H), 6.71 (d, J = 8.8 Hz, 2H), 3.38 (t, J = 5.6 Hz,

4H), 2.01 (t, J = 6.3 Hz, 4H).¹³C NMR (100 MHz,

DMSO‐d₆, δ, ppm): 198.33, 195.03, 191.65 (3CO), 161.55,

158.58, 157.19, 157.02, 153.30, 153.07, 150.67, 149.64,

140.32, 137.32, 129.38, 127.69, 125.68, 125.54, 125.25,

122.59, 119.11, 114.94, 48.48, 25.39, 24.40. HRMS (ESI):

calcd for C₂₈H₂₂ClN₄O₃ReNa [M + Na]⁺707.14 g mol

1

L⁶: Yield: 54%. H NMR (400 MHz, CDCl₃, δ, ppm):

9.87 (s, 2H), 8.69 (s, 2H), 8.68–8.64 (m, 4H), 7.85 (d,

J = 8.8 Hz, 2H), 6.78 (d, J = 8.4 Hz, 2H), 3.42 (t,

J = 6.5 Hz, 4H), 2.13–2.05 (m, 4H). ¹³C NMR (100 MHz,

CDCl₃, δ, ppm): 154.31, 151.51, 150.84, 149.11, 144.64,

143.91, 143.60, 128.22, 124.03, 118.40, 112.17, 47.79,

25.64. HRMS (ESI): calcd for C₂₃H₂₁N₆[M + H]⁺

381.1828; found 381.1828.

⁻¹; found 707.08 g mol⁻¹

.

5: Yield: 80%. IR (KBr, cm⁻¹): 2019(vs), 1912(s) and

1886(vs) ν(C≡O); 1589(s), 1534(m) ν(C¼N) and ν(C¼C).

¹H NMR (400 MHz, acetone‐d₆, δ, ppm): 8.67 (s, 1H),

8.33 (d, J = 3.3 Hz, 1H), 8.21 (d, J = 3.3 Hz, 1H), 8.12

(s, 2H), 8.06 (s, 1H), 8.04 (d, J = 2.7 Hz, 2H), 6.75 (d,

J = 8.8 Hz, 2H), 3.43 (t, J = 6.5 Hz, 4H), 2.10 (s,

4H).¹³C NMR not recorded due to insufficient complex

solubility. HRMS (ESI): calcd for C₂₄H₁₈ClN₄O₃S₂ReNa

4.3 | [ReCl(CO)₃(Lⁿ‐κ²N)] Complexes (1–6)

Re(I) carbonyl complexes were prepared using Re(CO)₅Cl

(0.10 g, 0.27 mmol) and appropriate ligand Lⁿ(0.27 mmol)

according to the procedure given in our previous paper.^1l

1: Yield: 75%. IR (KBr, cm⁻¹): 2018(vs), 1916(vs) and

1876(vs) ν(C≡O); 1614(m), 1583(w) and 1539(w) ν(C¼N)

and ν(C¼C). ¹H NMR (400 MHz, DMSO‐d₆, δ, ppm):

9.08 (t, J = 7.3 Hz, 3H), 8.80 (d, J = 3.8 Hz, 1H), 8.38 (t,

J = 7.8 Hz, 1H), 8.22–8.15 (m, 3H), 8.06 (t, J = 7.5 Hz,

1H), 7.91 (d, J = 7.7 Hz, 1H), 7.78 (t, J = 6.2 Hz, 1H),

7.65–7.60 (m, 4H). ¹³C NMR (100 MHz, DMSO‐d₆, δ,

ppm): 198.22, 194.90, 191.42 (3CO), 161.91, 158.25,

157.61, 156.70, 153.17, 151.21, 149.70, 140.46, 137.43,

135.23, 131.43, 129.87, 128.31, 127.95, 125.94, 125.63,

125.45, 124.86, 121.12. HRMS (ESI): calcd for

C₂₄H₁₅ClN₃O₃Re [M − Cl]⁺579.55 g mol⁻¹; found

[M + Na]⁺719.18 g mol⁻¹; found 719.0 g mol⁻¹

.

6: Yield: 70%. IR (KBr, cm⁻¹): 2023(vs), 1932(s) and

1890(vs) ν(C≡O); 1594(s), 1535(m) ν(C¼N) and ν(C¼C).

¹H NMR (400 MHz, DMSO‐d₆, δ, ppm): 10.34 (s, 1H),

9.17 (d, J = 1.8 Hz, 1H), 9.11 (d, J = 1.2 Hz, 2H), 8.96

(d, J = 3.0 Hz, 1H), 8.90 (t, J = 2.2 Hz, 2H), 8.25 (d,

J = 1.8 Hz, 1H), 8.18 (d, J = 9.0 Hz, 2H), 6.72 (d,

J = 9.0 Hz, 2H), 3.38 (d, J = 6.5 Hz, 4H), 2.01 (t,

J = 6.4 Hz, 4H).13C NMR not recorded due to insufficient

580.07 g mol⁻¹

.

complex

C₂₆H₂₀N₆O₃ReCl [M + Na]⁺709.13 g mol⁻¹; found

709.07 g mol⁻¹

solubility.

HRMS

(ESI):

calcd

for

2: Yield: 75%. IR (KBr, cm⁻¹): 2021(vs), 1913(vs) and

1887(vs) ν(C≡O); 1611(s) and 1542(w) ν(C¼N) and

.

1

ν(C¼C). H NMR (400 MHz, acetone‐d₆, δ, ppm): 8.84

(s, 1H), 8.37 (d, J = 3.2 Hz, 1H), 8.27 (s, 2H), 8.14 (dd,

J = 5.7, 2.6 Hz, 3H), 8.09 (d, J = 3.0 Hz, 1H), 7.64 (d,

J = 1.1 Hz, 1H), 7.63 (d, J = 2.4 Hz, 2H). ¹³C NMR not

recorded due to insufficient complex solubility. HRMS

(ESI): calcd for C₂₀H₁₁ClN₃O₃ReS₂Na [M + Na]⁺

4.4 | Crystal Structure Determination and

Refinement

The X‐ray diffraction data of 1–6 were collected with a

Gemini A Ultra diffractometer (Mo Kα radiation;

λ = 0.71073 Å) using a procedure reported previously.^1l

Details of the crystallographic data collection, structural

determination and refinement^[30,31]for 1–6 are given in

Table S4.

650.10 g mol⁻¹; found 649.93 g mol⁻¹

.

3: Yield: 70%. IR (KBr, cm⁻¹): 2030(vs), 1932(s) and

1

1880(vs) ν(C≡O); 1615(s) ν(C¼N) and ν(C¼C). H NMR

(400 MHz, DMSO‐d₆,:δ, ppm): 10.36 (s, 1H), 9.35 (s,

1H), 9.16 (s, 1H), 9.14 (d, J = 2.1 Hz, 1H), 8.99 (d,

J = 3.0 Hz, 1H), 8.93 (s, 2H), 8.44 (s, 1H), 8.27–8.22 (m,

2H), 7.65 (d, J = 3.4 Hz, 3H).¹³C NMR (125 MHz,

DMSO‐d₆, δ, ppm): 196.99, 195.05, 190.02 (3CO), 159.24,

155.98, 153.96, 151.51, 151.43, 148.39, 147.57, 146.35,

145.78, 144.58, 134.81, 131.77, 129.94, 128.47, 126.00,

122.19. HRMS (ESI): calcd for C₂₂H₁₃ClN₅O₃Re

4.5 | Physical Measurements

Infrared (IR) spectra (KBr pellets) were measured with a

Nicolet iS5 FTIR spectrophotometer (400–4000 cm⁻¹).

Absorption spectral measurements were carried out using

a Thermo Scientific Evolution 220 UV/Vis spectrometer

(in solution) and a Jasco V570 UV–Vis–NIR spectrometer

(in solid state as film deposited on a glass substrate and as

blends with PVK:PBD on a glass substrate). ¹H NMR and

[M − Cl]⁺581.53 g mol⁻¹; found 582.06 g mol⁻¹

.

4: Yield: 75%. IR (KBr, cm⁻¹): 2019(vs), 1912(s) and

1886(vs) ν(C≡O); 1597(s), 1526(m) ν(C¼N) and ν(C¼C).

18 of 21

KLEMENS ET AL.

¹³C NMR spectra were collected (295 K) with a Bruker

Avance 400 NMR spectrometer or Bruker Avance 500

NMR spectrometer.

Electrospray ionization mass spectrometry (ESI‐MS)

was performed with a Varian 500‐MS IT mass spectrome-

ter ion trap apparatus using a procedure reported

previously.^1l

Steady‐state emission measurements of solid‐state and

solution samples were carried out with an FLS‐980

fluorescence spectrophotometer. To determine the PL

lifetimes, time‐correlated single‐photon counting or

multi‐channel scaling method was used. PL spectra in

solid state as film deposited on a glass substrate and as

blends with PVK:PBD on a glass substrate were collected

with a Hitachi F‐2500 spectrometer. A precise voltage

supply (Gw Instek PSP‐405) with the sample fixed to an

XYZ stage was applied to collect EL spectra, and all the

The ANO‐L basis was used with the contractions 3s2p1d

for nitrogen and carbon, and 2s for hydrogen. Several

active spaces were tested and finally a CAS(12,12) space

was used to describe the lowest π → π*electronic excita-

tions. The selected space contains 12 electrons and 6

occupied and 6 unoccupied π orbitals, the character of

which is similar to those involved in TD‐DFT transitions.

PCM/acetonitrile solvation model was used to incorpo-

rate the effects of the solvent.

4.7 | Nanosecond Laser Flash Photolysis

Samples for time‐resolved photolysis were excited at

355 nm (third harmonic of a Nd:YAG laser). Nd:YAG

laser pulses (Spectra Physics, Mountain View, CA, USA,

model INDI 40‐10) of 6–8 ns duration were employed.

The monitoring system consisted of a 150 W pulsed

Xe lamp with a lamp pulser (Applied Photophysics,

Surrey, UK), a monochromator (Princeton Instruments,

Spectra Pro SP‐2357, Acton, MA, USA) and an R955

photomultiplier (Hamamatsu, Japan), powered by a PS‐

310 power supply (Stanford Research System, Sunnyvale,

CA, USA). The data processing system consisted of

measurements were performed using

reported in our previous work.^1l

a procedure

Differential scanning calorimetry studies were carried

out with the use of a TA‐DSC 2010 apparatus under nitro-

gen atmosphere, with a heating rate of 20°C min⁻¹. An

ATLAS 0531 electrochemical unit and impedance

analyser potentiostat was used to perform electrochemi-

cal measurements.^1lThickness of active layers was deter-

mined with atomic force microscopy (Topometrix

Explorer TMX 2000).

real‐time acquisition using

a

digital oscilloscope

(WaveRunner 6100A, LeCroy, Chestnut Ridge, NY,

USA) which was triggered by a fast photodiode (Thorlabs,

DET10M, ca 1 ns rise time) and transferred to a computer

equipped with software based on LabView 8.0 (National

Instruments, Austin, TX, USA) which controlled the

timing and acquisition functions of the system. The

experiments were carried out in rectangular 1 cm × 1 cm

quartz cells at room temperature.

4.6 | Computational Details

The calculations (geometry optimization, absorption and

emission spectra) were performed using the Gaussian‐09

program package^[22]at the DFT level with the PBE1PBE

hybrid exchange correlation functional, and the def2‐

TZVPD basis set for rhenium and def2‐TZVP basis set

for other elements.^[32–34]

ACKNOWLEDGEMENTS

The research was co‐financed by Narodowe Centrum

Nauki grant no. DEC‐2017/25/B/ST5/01611. The

calculations were carried out in Wroclaw Centre for Net-

working and Supercomputing (http://www.wcss.wroc.pl).

The authors thank Dr H. Janeczek for differential scan-

ning calorimetry measurements.

To verify the charge transfer problem at the TD‐DFT

level of theory, the Λ parameter^[29]was calculated for

ten lowest singlet excited states of L¹and L⁴ligands.

For optimized geometries of L¹and L⁴ligands, ten lowest

electronic transitions were recalculated using the Gamess

program^[35]and the same method described above. In cal-

culations, ligand environment was taken into account by

using the polarizable continuum model (PCM) and aceto-

nitrile as solvent. The solvation model density^[36]imple-

mented in the Gamess package was applied for PCM

electrostatics. To examine the properties of the four low‐

lying singlet excited states of L¹and L⁴, the CASSCF^[37]

and MS‐CASPT2^[38]level calculations were carried out

with MOLCAS^[39]quantum chemical package. The

CASSCF/MS‐CASPT2 calculations were performed using

the ANO‐L^[40]basis set, assuming the PBE1PBE/def2‐

TZVPD/def2‐TZVP optimized geometry of L¹and L⁴.

ORCID

Barbara Machura http://orcid.org/0000-0001-7688-6491

REFERENCES

[1] a) D. J. Stufkens, A. Vlček Jr., Coord. Chem. Rev. 1998, 177, 127.

b) R. A. Kirgan, B. P. Sullivan, D. P. Rillema, Top. Curr. Chem.

2007, 281, 45. c) A. Cannizzo, A. M. Blanco‐Rodríguez, A. El

Nahhas, J. Šebera, S. Záliš, A. Vlček Jr., M. Chergui, J. Am.

Chem. Soc. 2008, 130(28), 8967. d) A. Kumar, A. S. Sun, A. J.

KLEMENS ET AL.

19 of 21

Lees, Top. Organometal. Chem. 2010, 29, 1. e) A. Vlček Jr., Top.

Malicka, M. Cocchi, K. C. Gordon, S. Stagni, M. Massi, Dalton

Trans. 2015, 8379. c) G.‐W. Zhao, J.‐H. Zhao, Y.‐X. Hu, D.‐Y.

Zhang, X. Li, Synth. Met. 2016, 212, 131. d) Y.‐X. Hu, G.‐W.

Zhao, Y. Dong, Y.‐L. Lü, X. Li, D.‐Y. Zhang, Dyes Pigm 2017,

137, 569. e) X. Li, G.‐W. Zhao, Y.‐X. Hu, J.‐H. Zhao, Y. Dong,

L. Zhou, Y.‐L. Lv, H.‐J. Chi, Z.‐J. Su, Mater. Chem. C 2017, 5,

7629.

Organometal. Chem. 2010, 29, 73. f) R. Baková, M. Chergui, C.

Daniel, A. Vlček Jr., S. Záliš, Coord. Chem. Rev. 2011, 255, 975.

gS. Záliš, C. J. Milne, A. El Nahhas, A. M. Blanco‐Rodríguez, R.

M. van der Veen, A. Vlček Jr., Inorg. Chem. 2013, 52(10), 5775.

h)A. El Nahhas, R. M. van der Veen, T. J. Penfold, V. T. Pham,

˘

F. A. Lima, R. Abela, A. M. Blanco‐Rodriguez, S. Zális, A.

˘

Vlcek, I. Tavernelli, U. Rothlisberger, C. J. Milne, M. Chergui,

[8] a) A. I. Baba, J. R. Shaw, J. A. Simon, R. P. Thummel, R. H.

Schmehl, Coord. Chem. Rev. 1998, 171, 43. b) M. P. Coogan,

V. Fernández‐Moreira, B. M. Kariuki, S. J. A. Pope, F. L.

Thorp‐Greenwood, Angew. Chem. Int. Ed. 2009, 48, 4965. c)

A. J. Amoroso, A. Banu, M. P. Coogan, P. G. Edwards,

G. Hossain, K. M. A. Malik, Dalton Trans. 2010, 39, 6993. d)

F. Zhao, W. Q. Liu, H. Ying Xia, Y. Bo Wang, Comput. Theor.

Chem. 2012, 997, 49. e) D. Wang, Q.‐L. Xu, S. Zhang, H.‐Y.

Li, C.‐C. Wang, T.‐Y. Li, Y. M. Jing, W. Huang, Y. X. Zheng,

G. Accorsi, Dalton Trans. 2013, 42, 2716. f) H. van der Salm,

C. B. Larsen, J. R. W. McLay, M. G. Fraser, N. T. Lucas, K. C.

Gordon, Dalton Trans. 2014, 43, 17775. g) B. Laramee‐Milette,

C. Lachance‐Brais, G. S. Hanan, Dalton Trans. 2015, 44, 41.

h) T. Klemens, A. Switlicka‐Olszewska, B. Machura, M.

Grucela, E. Schab‐Balcerzak, K. Smolarek, A. Mackowski, A.

Szlapa, S. Kula, S. Krompiec, P. Lodowski, A. Chrobok, Dalton

Trans. 2016, 45, 1746. i) T. Klemens, A. Świtlicka‐Olszewska, B.

Machura, M. Grucela, H. Janeczek, E. Schab‐Balcerzak, A.

Szlapa, S. Kula, S. Krompiec, K. Smolarek, D. Kowalska, S.

Mackowski, K. Erfurt, P. Lodowski, RSC Adv. 2016, 6, 56335.

j) L. D. Ramos, R. N. Sampaio, F. F. de Assis, K. T. de Oliveira,

P. Homem‐de‐Mello, A. O. T. Patrocinio, K. P. M. Frin, Dalton

Trans. 2016, 45, 11688.

J. Phys. Chem. A 2013, 117(2), 361. i) C. Daniel, Coord. Chem.

Rev. 2015, 282–283, 19. j) K. M.‐C. Wong, C. Wang, H.‐C.

Lam, N. Zhu, Polyhedron 2015, 86, 133. k) A. Świtlicka‐

Olszewska, T. Klemens, I. Nawrot, B. Machura, R. Kruszynski,

J. Lumin. 2016, 171, 166. l) A. Świtlicka, T. Klemens, B.

Machura, E. Schab‐Balcerzak, K. Laba, M. Lapkowski, M.

Grucela, J. Nycz, M. Szala, M. Kania, RSC Adv. 2016, 6,

112908. m) S. Mai, H. Gattuso, M. Fumanal, A. Munoz‐Losa,

A. Monari, C. Daniel, L. González, Phys. Chem. Chem. Phys.

2017, 19, 27240. n) C.‐C. Ko, V. Wing‐Wah Yam, Acc. Chem.

Res. 2018, 51, 149. o) D. A. Kurtz, K. R. Brereton, K. P. Ruoff,

H. M. Tang, G. A. N. Felton, A. J. M. Miller, J. L. Dempsey,

Inorg. Chem. 2018, 57, 5389.

[2] M. Wrighton, D. L. Morse, J. Am. Chem. Soc. 1974, 96, 998.

[3] a) T. Morimoto, J. Tanabe, K. Sakamoto, K. Koike, O. Ishitani,

Res. Chem. Intermed. 2013, 39, 437. b) E. Portenkirchner, D.

Apaydin, G. Aufischer, M. Havlicek, M. White, M. C. Scharber,

N. Serdar Sariciftci, ChemPhysChem 2014, 15, 3634. c) A.

Zarkadoulas, E. Koutsouri, C. Kefalidi, C. A. Mitsopoulou,

Coord. Chem. Rev. 2015, 304–305, 55. d) J. K. Nganga, C. R.

Samanamu, J. M. Tanski, C. Pacheco, C. Saucedo, V. S. Batista,

K. A. Grice, M. Z. Ertem, A. M. Angeles‐Boza, Inorg. Chem.

2017, 56, 3214. e) J. F. Martinez, N. T. La Porte, M. R.

Wasielewski, J. Phys. Chem. C 2018, 122, 2608.

[9] a) Y. Yue, T. Grusenmeyer, Z. Ma, P. Zhang, T. Tat Pham, J. T.

Mague, J. P. Donahue, R. H. Schmehl, D. N. Beratan, I. V.

Rubtsov, J. Phys. Chem. B 2013, 117, 15903. b) Y.‐X. Peng, D.

Xu, N. Wang, T. Tao, B. Hua, W. Huang, Tetrahedron 2016,

72, 3443.

[4] a) R.‐R. Ye, C.‐P. Tan, M.‐H. Chen, L. Hao, L.‐N. Ji, Z.‐W. Mao,

Chem. Eur. J. 2016, 22, 7800. b) S. Hostachy, C. Policar, N.

Delsuc, Coord. Chem. Rev. 2017, 351, 172. c) L. C.‐C. Lee, K.‐

K. Leung, K. Kam‐Wing Lo, Dalton Trans. 2017, 46, 16357. d)

M. T. Gabr, F. C. Pigge, Dalton Trans. 2017, 46, 15040.

[10] a) C. Jia, S.‐X. Liu, C. Tanner, C. Leiggener, A. Neels, L.

Sanguinet, E. Levillain, S. Leutwyler, A. Hauser, S. Decurtins,

Chem. Eur. J. 2007, 13, 3804. b) M. G. Fraser, A. G. Blackman,

G. I. S. Irwin, C. P. Easton, K. C. Gordon, Inorg. Chem. 2010,

49, 5180. c) C. B. Larsen, H. van der Salm, C. A. Clark, A. B.

S. Elliott, M. G. Fraser, R. Horvath, N. T. Lucas, X.‐Z. Sun,

M. W. George, K. C. Gordon, Inorg. Chem. 2014, 53, 1339. d)

H. van der Salm, C. B. Larsen, J. R. W. McLay, G. S. Huff, K.

C. Gordon, Inorg. Chim. Acta 2015, 428, 1. e) C. B. Larsen, H.

van der Salm, G. E. Shillito, N. T. Lucas, K. C. Gordon, Inorg.

Chem. 2016, 55, 8446. f) B. S. Adams, G. E. Shillito, H. van

der Salm, R. Horvath, C. B. Larsen, X.‐Z. Sun, N. T. Lucas,

M. W. George, K. C. Gordon, Inorg. Chem. 2017, 56, 12967.

[5] a) C.‐O. Ng, S.‐W. Lai, H. Feng, S.‐M. Yiu, C.‐C. Ko, Dalton

Trans. 2011, 40, 10020. b) K. K.‐W. Lo, M.‐W. Louie, K. Y.

Zhang, 18th Int. Symp. Photochem. Photophys. Coord. Compd.

Sapporo 2009 2010, 254, 2603. c) G. Xu, M. Lu, C. Huang, Y.

Wang, S. Ge, Spectrochim. Acta. A 2014, 123, 369. d) A.

Ramdass, V. Sathish, E. Babu, M. Velayudham, P.

Thanasekaran, S. Rajagopal, Coord. Chem. Rev. 2017, 343,

278. e) A. Ramdass, V. Sathish, M. Velayudham, P.

Thanasekaran, S. Umapathy, S. Rajagopal, Sens. Actuators B

2017, 240, 1216.

[6] a) A. Fernández‐Acebes, J.‐M. Lehn, Chem. Eur. J. 1999, 5,

3285. b) S.‐S. Sun, E. Robson, N. Dunwoody, A. S. Silva, I. M.

Brinn, A. J. Lees, Chem. Commun. 2000, 201. c) J. D. Lewis,

R. N. Perutz, J. N. Moore, Chem. Commun. 2000, 1865. d) A.

O. T. Patrocinio, K. P. M. Frin, N. Y. Murakami Iha, Inorg.

Chem. 2013, 52, 5889. e) V. Sathish, A. Ramdass, P.

Thanasekaran, K.‐L. Lu, S. Rajagopal, J. Photochem, C 2015,

23, 25.

[11] a) L. D. Ramos, R. N. Sampaio, F. F. de Assis, K. T. de Oliveira,

P. Homem‐de‐Mello, A. O. T. Patrocinio, K. P. M. Frin, Dalton

Trans. 2016, 45, 11688.

[12] a) G. S. Huff, W. K. C. Lo, R. Horvath, J. O. Turner, X.‐Z. Sun,

G. R. Weal, H. J. Davidson, A. D. W. Kennedy, C. J. McAdam,

J. D. Crowley, M. W. George, K. C. Gordon, Inorg. Chem. 2016,

55, 12238. b) G. E. Shillito, T. B. J. Hall, D. Preston, P. Traber,

L. Wu, K. E. A. Reynolds, R. Horvath, X. Z. Sun, N. T. Lucas, J.

D. Crowley, M. W. George, S. Kupfer, K. C. Gordon, J. Am.

Chem. Soc. 2018, 140, 4534.

[7] a) R. C. Evans, P. Douglas, C. J. Winscom, Coord. Chem. Rev.

2006, 250, 2093. b) M. V. Werrett, G. S. Huff, S. Muzzioli, V.

Fiorini, S. Zacchini, B. W. Skeleton, A. Maggiore, J. M.

20 of 21

KLEMENS ET AL.

[13] H. van der Salm, M. G. Fraser, R. Horvath, J. O. Turner, G. M.

Greetham, I. P. Clark, M. Towrie, N. T. Lucas, M. W. George,

K. C. Gordon, Inorg. Chem. 2014, 53, 13049.

[24] a G. Rajib, D. K. Palit, J. Phys. Chem. A 2015, 119, 11128. b G.

Haberhauer, R. Gleiter, C. Burkhart, Chem. Eur. J. 2016, 22,

971. c G. Haberhauer, Chem. Eur. J. 2017, 23, 9288.

[25] M. Cocchi, J. Kalinowski, L. Murphy, J. A. G. Williams, V.

[14] K. S. Kisel, T. Eskelinen, W. Zafar, A. I. Solomatina, P. Hirva,

E. V. Grachova, S. P. Tunik, I. O. Koshevoy, Inorg. Chem.

2018, 57(11), 6349.

Fattori, Org. Electron. 2010, 11, 388.

[26] a A. J. Lees, Chem Rev. 1987, 87, 711. b D. Wang, Q.‐L. Xu, S.

Zhang, H.‐Y. Li, C.‐C. Wang, T.‐Y. Li, C. C. Wang, T. Y. Li,

Y. M. Jing, W. Huang, Y. X. Zheng, G. Accorsi, Dalton Trans.

2013, 42, 2716.

[15] a) J. Wang, G. Hanan, Synlett 2005, 2005, 1251. b) Z. Yin, G.

Zhang, T. Phoenix, S. Zheng, J. C. Fettinger, RSC Adv. 2015,

5, 36156. c) A. Maroń, A. Szlapa, T. Klemens, S. Kula, B.

Machura, S. Krompiec, J. G. Małecki, A. Świtlicka‐Olszewska,

K. Erfurt, A. Chrobok, Org. Biomol. Chem. 2016, 14, 3793. d)

A. Maroń, S. Kula, A. Szlapa‐Kula, A. Świtlicka, B. Machura,

S. Krompiec, Eur. J. Org. Chem. 2017, 2017, 2730.

[27] B. N. Ghosh, F. Topic, P. K. Sahoo, P. Mal, J. Linnera, E.

Kalenius, H. M. Tuononen, K. Rissanen, Dalton Trans. 2015,

44, 254.

[28] a Q. Wu, T. Zhang, Q. Peng, D. Wang, Z. Shuai, Phys. Chem.

Chem. Phys. 2014, 16, 5545. b T. Zhu, T. Van Voorhis, J. Phys.

Chem. C 2016, 120, 19987. c K. Kotwica, P. Bujak, D. Wamil,

A. Pieczonka, G. Wiosna‐Salyga, P. A. Gunka, T. Jaroch, R.

Nowakowski, B. Luszczynska, E. Witkowska, I. Glowacki, J.

Ulanski, M. Zagorska, A. Pron, J. Phys. Chem. C 2015, 119,

10700. d S. Jenatsch, L. Wang, N. Leclaire, E. Hack, R. Steim,

S. B. Anantharaman, J. Heier, B. Ruhstaller, L. Penninck, F.

Nüesch, Org. Electron. 2017, 48, 77.

[16] D. G. Vargas‐Pineda, T. Guardado, F. Cervantes‐Lee, A. J.

Metta‐Magana, K. H. Pannell, Inorg Chem. 2010, 49, 960.

[17] I. R. de Moraes, S. Scholz, M. Hermenau, M. L. Tietze, T.

Schwab, S. Hofmann, M. C. Gather, K. Leo, Org. Electron.

2015, 26, 158.

[18] P. Bujak, I. Kulszewicz‐Bajer, M. Zagorska, V. Maurel, I.

Wielgus, A. Pron, Chem. Soc. Rev. 2013, 42, 8895.

[29] a M. J. G. Peach, P. Benfield, T. Helgaker, D. J. Tozer, J. Chem.

Phys. 2008, 128, 044118. b M. J. G. Peach, C. R. L. Sueur, K.

Ruud, M. Guillaume, D. J. Tozer, Phys. Chem. Chem. Phys.

2009, 11, 4465. c J. Plötner, D. J. Tozer, A. Dreuw, J. Chem,

Theory Comput. 2010, 6, 2315.

[19] a) B. Laramee‐Milette, C. Lachance‐Brais, G. S. Hanan, Dalton

Trans. 2015, 44, 41. b) C.‐C. Ko, L. T.‐L. Lo, C.‐O. Ng, S.‐M. Yiu,

Chem. Eur J. 2010, 16, 13773. c) Q.‐H. Wei, F.‐N. Xiao, L.‐J.

Han, S.‐L. Zeng, Y.‐N. Duan, G.‐N. Chen, Dalton Trans. 2011,

40, 5078. d) D. Wang, Q.‐L. Xu, S. Zhang, H.‐Y. Li, C.‐C. Wang,

T.‐Y. Li, Y. M. Jing, W. Huang, Y. X. Zheng, G. Accorsi, Dalton

Trans. 2013, 42, 2716. e) P. Gomez‐Iglesias, F. Guyon, A.

Khatyr, G. Ulrich, M. Knorr, J. M. Martin‐Alvarez, D. Miguel,

F. Villafane, Dalton Trans. 2015, 44, 17516.

[30] Agilent Technologies, CrysAlis CCD and CrysAlis RED. Oxford

Diffraction Ltd, Yarnton, Oxfordshire, UK 2012.

[31] G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112.

[32] a) D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß,

Theor. Chim. Acta 1990, 77, 123. b) F. Weigend, R. Ahlrichs,

Phys. Chem. Chem. Phys. 2005, 7, 3297. c) D. Rappoport, F.

Furche, J. Chem. Phys. 2010, 133, 134105.

[20] L. D. Ramos, R. N. Sampaio, F. F. de Assis, K. T. de Oliveira, P.

Homem‐de‐Mello, A. O. T. Patrocinio, K. P. M. Frin, Dalton

Trans. 2016, 45, 11688.

[21] a) S. Kim, T. K. An, J. Chen, I. Kang, S. H. Kang, D. S. Chung,

C. E. Park, Y.‐H. Kim, S.‐K. Kwon, Adv. Funct. Mater. 2011, 21,

1616. b) Y. Zhang, L. Zhang, H. Liu, D. Sun, X. Li,

CrystEngComm 2015, 17, 1453.

[33] a) C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158. b) J. P.

Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.

[34] a) B. Mennucci, J. Tomasi, J. Chem. Phys. 1997, 106, 5151. b) E.

Cancès, B. Mennucci, J. Tomasi, J. Chem. Phys. 1997, 107, 3032.

c) M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys.

Lett. 1998, 286, 253.

[22] 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.

Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.

Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.

Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F.

Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin,

V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,

A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.

Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V.

Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,

O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,

R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.

Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö.

Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox,

Gaussian 09, Gaussian Inc., Wallingford, CT, USA 2009.

[35] a) M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M.

S. Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A.

Nguyen, S. Su, T. L. Windus, M. Dupuis, J. A. Montgomery,

J. Comput. Chem. 1993, 14, 1347. b) M. S. Gordon, M. W.

Schmidt, Theory and Applications of Computational Chemistry,

Elsevier 2005 1167.

[36] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B

2009, 113, 6378.

[37] a) B. O. Roos, in Lecture Notes in Quantum Chemistry, (Ed: B.

O. Roos) Vol. 58, Springer, Berlin 1992 177. b) B. O. Roos, in

Advances in Chemical Physics, (Ed: K. P. Lawley) Vol. 69, John

Wiley & Sons, Inc., Hoboken, NJ, USA 1987 399.

[38] a) K. Andersson, P. A. Malmqvist, B. O. Roos, A. J. Sadlej, K.

Wolinski, J. Phys. Chem. 1990, 94, 5483. b) J. Finley, P.‐Å.

Malmqvist, B. O. Roos, L. Serrano‐Andrés, Chem. Phys. Lett.

1998, 288, 299.

[23] T. Mutai, J. D. Cheon, S. Arita, K. Araki, J. Chem. Soc. Perkin

Trans. 2 2001, 1045.

KLEMENS ET AL.

21 of 21

[39] a)G. Karlström, R. Lindh, P.‐Å. Malmqvist, B. O. Roos, U.

Ryde, V. Veryazov, P.‐O. Widmark, M. Cossi, B.

Schimmelpfennig, P. Neogrady, L. Seijo, Comput. Mater. Sci.

2003, 28, 222. b)V. Veryazov, P. O. Widmark, L. Serrano‐

Andrés, R. Lindh, B. O. Roos, Int. J. Quantum Chem. 2004,

100, 626. c)F. Aquilante, L. De Vico, N. Ferré, G. Ghigo, P.‐Å.

Malmqvist, P. Neogrády, T. B. Pedersen, M. Pitonak, M. Reiher,

B. O. Roos, L. Serrano‐Andrés, M. Urban, V. Veryazov, R.

Lindh, J. Comput. Chem. 2010, 31, 224. d) The following per-

sons have contributed to the development of the Molcas

software: K. Andersson, F. Aquilante, M. Barysz, A.

Bernhardsson, M. R.A. Blomberg, J. Boström, Y. Carissan, L.

Chibotaru, D. L. Cooper, M. Cossi, M. Delcey, A. Devarajan,

L. De Vico, I. F. Galván, N. Ferré, M. P. Fülscher, A. Gaenko,

L. Gagliardi, G. Ghigo, C. de Graaf, S. Gusarov, B. A. Hess,

D. Hagberg, J. M. Hermida‐Ramón, A. Holt, G. Karlström, J.

W. Krogh, R. Lindh, P.‐Å. Malmqvist, T. Nakajima, P.

Neogrády, J. Olsen, T. B. Pedersen, D. Peng, M. Pitonak, J.

Raab, M. Reiher, B. O. Roos, U. Ryde, B. Schimmelpfennig,

M. Schütz, L. Seijo, L. Serrano‐Andrés, I. Schapiro, P. E. M.

Siegbahn, J. Stålring, B. Suo, P. Sushko, T. Thorsteinsson, T.

Tsuchiya, L. Ungur, S. Vancoillie, V. Veryazov, V. Vysotskiy,

P.‐O. Widmark.

[40] a) P.‐O. Widmark, P.‐Å. Malmqvist, B. O. Roos, Theor. Chim.

Acta 1990, 77, 291. b) P.‐O. Widmark, B. J. Persson, B. O. Roos,

Theor. Chim. Acta 1991, 79, 419.

SUPPORTING INFORMATION

Additional supporting information may be found online

in the Supporting Information section at the end of the

article.

How to cite this article: Klemens T, Świtlicka A,

Szlapa‐Kula A, et al. Experimental and

computational exploration of photophysical and

electroluminescent properties of modified

2,2′:6′,2″‐terpyridine, 2,6‐di(thiazol‐2‐yl)pyridine

and 2,6‐di(pyrazin‐2‐yl)pyridine ligands and their

Re(I) complexes. Appl Organometal Chem. 2018;

e4611. https://doi.org/10.1002/aoc.4611

Article Doi

Experimental and computational exploration of photophysical and electroluminescent properties of modified 2,2′:6′,2″-terpyridine, 2,6-di(thiazol-2-yl)pyridine and 2,6-di(pyrazin-2-yl)pyridine ligands and their Re(I) complexes

DOI: 10.1002/aoc.4611

Source and publish data:

Authors:

Article abstract of DOI:10.1002/aoc.4611

Full text of DOI:10.1002/aoc.4611

Products guided by the article

R&D Labs maybe for 515125-29-8

Relevant to this article

Hot Product