2-Substituted vs 4-substituted-9,9'-spirobifluorene host materials for green and blue phosphorescent OLEDs: A structure-property relationship study

ACCEPTED MANUSCRIPT

2

-substituted vs 4-substituted-9,9'-spirobifluorene host materials

for green and blue phosphorescent OLEDs: A Structure-Property Relationship Study

a

c

c,d

Sébastien Thiery, Céline Declairieux, Denis Tondelier, Gijun Seo, Bernard Geffroy,

a

b

*a

Olivier Jeannin, Rémi Métivier, Joëlle Rault-Berthelot, Cyril Poriel

a. Université de Rennes 1-UMR CNRS 6226, "Institut des Sciences Chimiques de Rennes" MaCSE group. Bat

0C, Campus de Beaulieu, 35042 Rennes, France

1

b. PPSM, Institut d'Alembert, ENS Cachan, UMR CNRS 8531, 61 Avenue du Président Wilson, 94235 Cachan,

France

c. LPICM, Ecole Polytechnique, UMR CNRS 7647, Route de Saclay, 91128, Palaiseau

d. LICSEN, CEA Saclay, IRAMIS/NIMBE, 91191 Gif sur Yvette

Abstract

We report a structure-property relationship study of four 9,9'-spirobifluorene (SBF) derivatives (4-

5Pm-SBF, 2-5Pm-SBF, 4-Ph-SBF and 2-Ph-SBF), substituted with either phenyl or pyrimidine at

the C2 or C4 position of the SBF core. Structural, thermal, electrochemical and photophysical

properties have been examined and correlated to theoretical calculations in order to study the

influence of the nature and the position of the substituent. The emission properties of 4- vs 2-

substituted SBFs are noticeably different highlighting, in the excited state, the remarkable effect of

substitution in ortho position of SBF. All compounds have been used as host material for green

dopant in PhOLEDs with very high performances (2-5Pm-SBF: CE > 58 cd/A, PE > 35 lm/W, EQE

>

14 %). More importantly, the two 4-substituted SBFs have been used as host materials in blue

PhOLEDs, displaying high performance and a decrease of V_T_Hfor 4-5Pm-SBF due to the

incorporation of the electron-withdrawing pyrimidine.

Grpahical Abstract

N

Introduction

-

1 -

ACCEPTED MANUSCRIPT

In the past twenty years, the use of organic semiconducting materials for applications in “plastic

1

2,3

electronics”, such as field effect transistors (FETs), organic light emitting diodes (OLEDs), and organic

4,5

6-8

photovoltaic (OPVs), has gained fantastic interest. Particularly, OLEDs have attracted much scientific

and commercial interest because of their potential applications in organic flat-panel displays and solid-state

lighting. However, as full colour displays and white lighting require materials that emit over a broad range

of colour, the weak link of the technology remains the lack of efficient and stable blue light emitters.

Moreover, due to the triplet exciton/singlet exciton ratio of 3/1, the maximum achievable emission

quantum yield is limited to 25% for fluorescent emitters. Among the significant progress recently made,

phosphorescent organic emitting-diodes (PHOLEDs) which use heavy-metal complexes as emitters to

harvest both singlet and triplet excitons are considered as the most promising technology because,

theoretically the internal quantum efficiencies of these devices may reach 100 %. The fabrication of blue

phosphorescent devices is however still challenging because the prerequisites for designing organic host

materials are (i) a high triplet energy level (higher than E of the blue dopant), (ii) a high glass-transition

T

temperature (Tg) and (iii) matching HOMO/LUMO levels for hole and electron injection/transport. In this

context, most of host materials for blue PhOLEDs are based on carbazole, triphenylamines, phosphine

9-11

oxides or fluorene derivatives.

However, novel host materials with good thermal and morphological

stabilities to afford blue PhOLEDs with high efficiencies, stability and colour purity are still desirable. In this

context, 9,9'-spirobifluorene (SBF) based semiconductors have experienced a fantastic development during

1

2,13

14-17

the last thirty years, due to their use as chiral ligand,

as electropolymerizable building block,

as solid

18

19

20

state laser , in third-order nonlinear optics, in Metal Organic Frameworks but more importantly in

organic electronics, leading to fantastic breakthroughs in the field.

2

1-23

Thus, numbers of electron donating

and /or withdrawing functionalities have been introduced on the SBF scaffold in order to make the

21

electronic properties of the resulting dyes fitting with an application. For that purpose, electron-deficient

2

4

25

26

27

28

heterocycles such as 1,3,4-oxadiazoles, pyridine , quinoline, quinoxaline and pyridopyrazine

connected at C2 and C7 positions of SBF were reported in literature. Among those electron-deficient

heterocycles, only few examples describing the fluorene substitution by pyrimidine units (see chart 1) have

been reported to date. However, pyrimidine possesses many striking characteristics, as it is notably more

29

electron-deficient than pyridine, which has been widely studied. Thus Bryce's group has for example

30

demonstrated in 2003, that fluorene fluorophore F3 emits a pure blue emission in solution and thin films.

However the electroluminescent spectrum of F3-based OLED mainly presents the known parasite green

3

1-33

emission of dialkylfluorenes.

A similar green emission is observed on PL spectrum of the liquid crystal

34

F4. The introduction of a 9,9' Spirobifluorene unit (SBF) in TBPSF leads to improve the thermal, optical and

electronic properties and this new blue emitter has been used as host for fluorescent blue, red or green

organic dopants leading to efficient devices.

35,36

Chart 1. Literature report of fluorene substitution by pyrimidine units

-

2 -

ACCEPTED MANUSCRIPT

However, if 2-substituted SBFs have been impressively investigated for the last decades, it is far to be the

case of the three other positional isomers, that are 1-, 3- or 4-substituted SBFs. Thus, for example, 4-

37-42

substituted SBFs have only been reported in a few instances

but appears as a very appealing scaffold to

design rigid SBF-based organic host materials for PhOLEDs. In our quest for robust derivatives with high E

which can be used as host materials for green and blue phosphorescent dyes, we wish to report herein

T

43

the synthesis and the detailed study of the properties of two pyrimidine-substituted SBFs (4-5Pm-SBF and

2

-5Pm-SBF, see chart 2). Moreover, in order to give a detailed description of the influence of the pyrimidine

unit, the 2- and 4-phenyl-substituted SBFs (2-Ph-SBF and 4-Ph-SBF) have been also studied in detail (4-Ph-

44

SBF has been previously reported). Thus, after synthetic considerations, the electrochemical, thermal and

photo-physical properties will be discussed and coupled to a theoretical approach. The influence of the

nature of the substituent (pyrimidine vs phenyl) and of its position (2- vs 4-) will be notably carefully

investigated. Finally, the incorporation of these four semiconductors as host material for green and blue

PhOLEDs will be presented.

Chart 2. Molecules studied in this work

Results and discussion

The first step towards the synthesis of 4- and 2-substituted SBF was the access of the two key

platforms 4- and 2-bromo-9,9'-spirobifluorene (see structures of 4-Br-SBF and 2-Br-SBF in scheme 1). As

37,44

reported previously,

4-Br-SBF was synthesized by the mono lithium-halogen exchange of 2,2'-

dibromobiphenyl followed by the trapping of the corresponding lithiated intermediate with 9-fluorenone.

The intramolecular electrophilic ring closure of the corresponding fluorenol (not isolated), in acidic media

(

AcOH/HCl), further provides 4-Br-SBF with 75 % yield (2 steps). Finally, a subsequent Suzuki cross-coupling

between 4-Br-SBF and either 5-pyrimidine boronic acid or phenylboronic acid was performed

Pd(dppf)Cl /K CO /DMF) providing 4-5Pm-SBF or 4-Ph-SBF with significant good yields (94% and 91%

(

2

3

respectively). Similarly, access to 2-Br-SBF was performed by the lithium-halogen exchange of 2-

bromobiphenyl followed by the trapping of the corresponding lithiated intermediate with 2-bromo-9-

45

fluorenone. The intramolecular electrophilic ring closure of the corresponding fluorenol, in acidic media

AcOH/HCl), further provides 2-Br-SBF with 67% yield (2 steps), involved in a final Suzuki cross-coupling

(

with either 5-pyrimidine boronic acid or phenylboronic acid to lead the corresponding 2-5Pm-SBF or 2-Ph-

SBF with high yields (77% and 78% respectively).

-

3 -

ACCEPTED MANUSCRIPT

B(OH)2

(91%)

4-Ph-SBF

DMF 150°C

K2CO3 Pd(dppf)Cl2

N

Br

N

B(OH)₂

4

-Br-SBF

(75%)

N

-5Pm-SBF

4

(94%)

B(OH)2

2

-Ph-SBF

(

DMF (150°C) or

Xylenes (135°C)

7

8

%

)

Br

K2CO3 Pd(dppf)Cl2

N

2-Br-SBF

N

B(OH)₂

(67%)

N

2-5Pm-SBF

(77%)

Scheme 1. Synthesis of the four dyes, 4-Ph-SBF, 4-5Pm-SBF, 2-Ph-SBF and 2-5Pm-SBF

Structural Properties

Molecular structure of the four dyes was further confirmed by X-ray crystallography. The X-ray diffraction

data of single crystals of 2-Ph-SBF (Figure 1, left) and 4-Ph-SBF (Figure 1, right) reveal an asymmetric unit

44

containing two independent molecules 1 and 2. As previously reported, in 4-Ph-SBF, the angle between

the mean plane of the pendant phenyl ring and that of its attached phenyl ring of the fluorene is as large as

5

1.2° in molecule 1 and 56.6° in molecule 2.

In 2-Ph-SBF, the same angle appears to be noticeably smaller, ie 37.4° for molecule 1. Similar values have

46,47

been reported in literature for 2-substituted SBF (ca 30°)

and seems hence to be a characteristic of such

family of 2-substituted SBFs. This difference between the angle observed in 2-Ph-SBF and 4-Ph-SBF may be

at the origin of the restricted π-conjugation of the latter and will have remarkable consequences on the

electronic properties (see below). In addition, we note that the fluorene unit bearing the pendant phenyl

ring in the molecule 1 of 2-Ph-SBF is folded with a deformation of ca 7.1° between the mean planes of each

phenyl unit. However, in the case of molecule 2 of 2-Ph-SBF, we surprisingly note that the pendant phenyl

unit is almost coplanar with the fluorene unit with an angle between the mean plane of the pendant phenyl

ring and that of its attached phenyl ring of the fluorene unit as low as 4.6°. In addition and oppositely to

what was observed for molecule 1, the fluorene bearing the phenyl unit is now almost perfectly flat with

only a small deformation of 1.5° (7.1° measured in molecule 1, see above). Such structural feature appears

highly surprising and is assigned to the packing of the molecule. Indeed, numbers of short CH and CC

distances are detected in the packing diagram of 2-Ph-SBF and especially two short intermolecular CH

distances of ca 2.8 Å between the pendant phenyl ring and its close neighbours (see SI).

-

4 -

ACCEPTED MANUSCRIPT

2

-Ph-SBF

4-Ph-SBF

Molecule 1

Molecule 2

Molecule 1

Molecule 2

Figure 1. Molecular structure of the two molecules (molecule 1 and molecule 2) contained in 2-Ph-SBF (left) and 4-Ph-

44

SBF (right) single crystal asymmetric unit.

However, in the case pyrimidine based SBFs, ie 4-5Pm-SBF and 2-5Pm-SBF, we only managed to obtain

single crystals of 2-5Pm-SBF good enough for X-Ray diffraction and hence the direct comparison is not

possible. However, we are convinced that similar structural features structures occur in both 4-5Pm-SBF

and 2-5Pm-SBF. Indeed, in 2-5Pm-SBF, the angle between the mean plane of the pyrimidine plane and that

of its attached phenyl ring is recorded at 23.7° (See SI), lower than that recorded for its phenyl analogue 2-

Ph-SBF (37.4°).

Electrochemical Properties

Electrochemical properties have been investigated by cyclic voltammetry (CV) in CH Cl in oxidation

2

and reduction and are summarized in table 1. Pyrimidine-substituted SBF were studied in absence of Al O

2

3

as their rapid adsorption on Al O leads to a strong decrease of the intensity of the oxidation and reduction

2

3

waves.

All compounds present at least two successive oxidation waves (with maxima E and E ) before 2.2 V (figure

). Except for 2-5Pm-SBF, the oxidation is accompanied by an electropolymerization process signed on the

1

2

present CVs by a reduction wave at the back sweep. This polymerization process is more clearly observed

along recurrent sweeps (see CVs in SI). Compared one to another, polymerization process is highly efficient

for 2-Ph-SBF and 4-Ph-SBF and largely less efficient for 4-5Pm-SBF. Moreover, the electroactivity of the

corresponding polymers, ie Poly(2-Ph-SBF) and Poly(4-Ph-SBF) are similar with onset oxidation potentials

at ca 1.1 V vs SCE (polymers HOMO: -5.5 eV). In addition, the p-doping process reversibility of the two

polymers occurs in a large potential range between 1 and 2 V. Oppositely, Poly(4-5Pm-SBF) presents a p-

-

5 -

ACCEPTED MANUSCRIPT

doping process only occurring since 1.35 V (polymer HOMO : -5.75 eV) and appearing less intense and

reversible.

Table1. Selected electrochemical, theoretical and optical data

Electrochemical studies

Theoretical calculations

Optical

studies

1

2

b

c

el

d

theo

opt

e

E , E oxidation

peaks/

E

Reduction peak/

HOMO

(eV)

LUMO

(eV)

∆E (eV)

HOMO LUMO

∆E (eV) / E

T

∆E (eV)

red

a

E

onset (V)

(eV)

-6.16

-5.97

-6.13

-5.87

(eV)

-1.65

-1.27

-1.85

ox

a

onset (V)

4

2

-5Pm-SBF

-Ph-SBF

1.82, 2.23/

.57

1.68, 1.88/

.55

1.75, 2.00/

.56

1.60, 1.93/

.48

+ Bu NPF

-2.38/

-2.17

-2.75/

-2.45

-2.33/

-2.10

-5.97

-5.95

-5.96

-5.88

-2.23

-1.95

-2.30

-2.10

3.74

4.00

3.66

3.78

4.51/2.80

4.70/2.82

4.29/2.63

4.39/2.62

3.72

3.82

3.66

3.70

1

-5Pm-SBF

-Ph-SBF

1

No peak in CH

-2.3

2 2

Cl /

-1.48

1

ox

a. Measured in CH

2

Cl

2

4

6

0.2 M at 100 mV/s, b. Calculated from the onset oxidation potential Eonset , c. Calculated from the onset reduction

red

el

potential Eonset , d: ∆E = |HOMO-LUMO|, e: Calculated from the onset of the absorption spectrum in solution in cyclohexane (see spectra figure 5

left) by using ∆E = hc/λ (∆E (eV)=1237.5/λ(nm).

opt

2

E

40

30

20

10

0

4 5- Pm-S BF

2 5- Pm-S BF

2

E

3

0

E

1

25

E

1

2

0

5

1

10

5

0

-

10

0

-5

.0

0.5

1.0

1.5

2.0

2.5

0.5

1.0

1.5

2.0

2.5

E(V) vs SCE

4-Ph-SBF

40

30

20

8

6

4

2

0

²E

2

-Ph-SBF

2

E

0

E

1

E

1

10

0

-

10

20

-

20

40

-

0.0

0.5

1.0

1.5

2.0

2.5

0.5

1.0

1.5

2.0

2.5

E(V) vs SCE

E(V) vs ECS

-

3

-3

Figure 2. Cyclic voltammetry of 4-5Pm-SBF (5 × 10 M ) and 2-5Pm-SBF (2 × 10 M) (top) and of 4-Ph-SBF (5 × 10 M)

-

3

-1

and 2-Ph-SBF (5 10 M) (bottom) solutions in CH Cl + Bu NPF 0.2 M, sweep-rate 100 mV.s . Platinum disk (Ø: 1mm)

2

4

6

working electrode.

48

From the onset oxidation of the CVs, we determined the HOMO levels of 4-5Pm-SBF at -6.0 V

49

close to that of 4-Ph-SBF (HOMO: -5.95 eV) and of its constituting building block SBF (HOMO: -5.94 eV)

indicating that the substitution in ortho position leads to a π-conjugation interruption surely due to the

large angle observed in the X-ray structure between the pendant unit and the fluorene core. In addition, we

note that the substitution by a 5-pyrimidine or a phenyl group have both very weak effects on the HOMO

level of the 4-substituted-SBFs. Indeed, as seen in theoretical calculations, the pendant substituent is not

involved in the HOMO character of 4-substituted SBFs (Figure 3) being hence in accordance with our

electrochemical conclusions. On the other hand, HOMO levels of 2-5Pm-SBF and 2-Ph-SBF calculated from

their respective onset oxidation potentials are estimated at -5.96 eV and -5.88 eV. The highest HOMO of 2-

-

6 -

ACCEPTED MANUSCRIPT

Ph-SBF compared to that of SBF signs an extension of the π-conjugation between the fluorene and the

phenyl group in 2-Ph-SBF. This effect is however neutralized in 2-5Pm-SBF by the electron withdrawing

effect of the pyrimidine group leading to an HOMO level identical to that of SBF. Thus, in the case of 2-Ph-

SBF, the substituent is involved in the character of the HOMO (Figure 3), which is not the case for the

HOMO of 2-5Pm-SBF. Cathodic exploration in dichloromethane, reveal that all compounds are reduced at

negative potential in a non-reversible process close to the electrolytic medium reduction (see CVs in SI).

However, when studied in THF, allowing reaching a more cathodic potential range, one or two reduction

waves are observed before the reduction of the electrolytic solution (see CVs in SI). LUMO levels were

determined from the onset reduction potential in CH Cl at -2.23 ; -1.95 ; -2.30 and -2.10 eV for 4-5Pm-SBF ;

2

4

-Ph-SBF : 2-5Pm-SBF and 2-Ph-SBF respectively (these values are close to that measured in THF, see in SI).

The lowest LUMO level is recorded for 2-5Pm-SBF as low as -2.3 eV, 0.41 eV lower than the LUMO of SBF (-

4

1

.89 eV) signing the double effect of the extension of conjugation due to the substitution at the C2 of SBF

and to the electron-withdrawing effect of the electron-deficient pyrimidinium group. Thus, we note that

the pyrimidine unit is strongly involved in the LUMO character of 2-5Pm-SBF and even more strongly

involved in the case of 4-5Pm-SBF (Figure 3). Thus, the LUMO level of 4-5Pm-SBF (-2.23 eV) is shifted by

0

.07 eV compared to that of 2-5Pm-SBF. In addition, the LUMO level of 4-5Pm-SBF is shifted by 0.27 eV

compared to that of SBF, (-1.96 eV) due to the effect of the incorporation of the electron-withdrawing

pyrimidine fragment. Finally, due to the absence of π-conjugation between the phenyl and the C4-

44

fluorene, LUMO levels of 4-Ph-SBF is the highest of the present series, ie -1.96 eV, and very close to that

of SBF.

el

Finally, HOMO-LUMO gaps (∆E ) extend between 3.66 eV for 2-5Pm-SBF to 4.00 eV for 4-Ph-SBF. This trend

in HOMO-LUMO gap changes is well reproduced by theoretical calculations (figure 3) with nevertheless a

deviation of ca 0.7 eV.

Figure 3. Calculated frontier molecular orbitals by DFT of 4-5Pm-SBF ; 4-Ph-SBF : 2-5Pm-SBF and 2-Ph-SBF, after

geometry optimization at the B3LYP/6-311G+(d,p) level of theory, shown with an isovalue of 0.04 .

Geometry optimization of the four substituted dyes in the singlet and triplet states was performed using

Density Functional Theory (DFT) at the Gaussian09 B3LYP/6-311+G(d,p) level of theory. There is a clear

resemblance between the HOMO values determined by CVs and by theoretical calculations. 2-Ph-SBF is the

easiest oxidizable molecule due to the extension of conjugation between the phenyl and the 2-substituted

fluorene unit and its HOMO presents a phenyl-fluorene character, perfectly translating this extension of

conjugation. 4-5Pm-SBF is the more difficult oxidizible molecule due to the withdrawing effect of the

pyrimidine group on the fluorenyl core. However, it should be noted that 2-5Pm-SBF and 4-5Pm-SBF

presents almost identical HOMO character and HOMO energy level (-6.13 eV vs -6.16 eV resp) clearly

meaning that the position of the pyrimidine group only has a weak influence on the HOMO energy levels

and distribution. As exposed above, the more easily reduced molecule is 2-5Pm-SBF due to the double

effect of the extension of conjugation and of the electron-accepting properties of the pyrimidine unit, its

LUMO has hence a clear pyrimidine-fluorene character. In 4-5Pm-SBF, the LUMO exclusively has a

pyrimidine character and its energy level is 0.2 eV higher than that of 2-5Pm-SBF due to the absence of

-

7 -

ACCEPTED MANUSCRIPT

conjugation between the fluorene and the pyrimidine unit. This is a very important feature of the present

chemical design and clearly highlights its efficiency. Thus, in 4-substituted SBFs, the HOMO character is

almost identical with no (or weak) contribution of the substituent in accordance with our electrochemical

conclusions (see above). In addition, it must be pointed out that the LUMO+1 of 4-5Pm-SBF is very close to

its LUMO (0.01 eV) with a 4-5Pm-fluorenyl and 4,5Pm character respectively. Thus, both 2-5Pm-SBF and 4-

5

Pm-SBF have a clear fluorene/pyrimidine character in their LUMO and LUMO+1 (almost degenerated with

the LUMO which exclusively possess a pyrimidine character) respectively but present different energy

levels. This highlights the importance of the torsion angle between the fluorene and the pyrimidine, which

is surely at the origin of these different energy values. The computed triplet adiabatic S to T excitation

0

1

energies, defined as the energy difference between total energy of the molecule in their respective singlet

and triplet states, is of 2.80 and 2.82 eV for 4-5Pm-SBF and 4-Ph-SBF and of 2.63 and 2.62 eV for 2-5Pm-

44

SBF and 2-Ph-SBF respectively. These calculations predict that 4-5Pm-SBF will be, as 4-Ph-SBF, a good

candidate as host for blue and green phosphorescent dyes and both 2-5Pm-SBF and 2-Ph-SBF will be both

only adapted to green dopants.

The thermal properties of all SBF derivatives were investigated by thermogravimetric analysis (TGA) and

differential scanning calorimetry (DSC) and the results are summarized in table 2. The Td is defined as the

temperature at which the sample has lost the first 5% of mass during heating.

Table 2. Thermal properties (melting point: m.p., decomposition temperature: T , glass transition: T , crystallization

d

g

temperature: T all obtained from the tangent to the transition peak) of the four SBF dyes.

c,

4

-5Pm-SBF 4-Ph-SBF 2-5PmSBF 2-Ph-SBF

b

a

1

97-198

42

m.p.(°C)

T_d(°C)

T_g(°C)

T_c(°C)

173

277

85

-

213

100-108

2

254

76

238

78

-

9

4

-

115

a: This transition appears to be not very clear, occurring on a large range. b. from reference 44

4

2

-Ph-SBF

-5Pm-SBF

-Ph-SBF

4

2

Ph-SBF

1

00

-5Pm-SBF

-Ph-SBF

-5Pm-SBF

3

2

1

0

.0

.5

.0

.5

.0

.5

.0

-5Pm-SBF

80

60

40

20

0

-

0.5

0

50

100

150

200

250

0

100

200

300

400

500

600

700

Temperature (°C)

Figure 4. DSC curves from second heating cycle (Left) and TGA curves (right) of 4-Ph-SBF (black line), 4-5Pm-SBF (red

line), 2-Ph-SBF (blue line), 2-5Pm-SBF (green line)

5

% decomposition (T ) of 4-Ph-SBF (Figure 4 right, black line) and 4-5Pm-SBF (Figure 4 right, red line) respectively

d

44

occurs at 254°C and 277°C, higher than the T of SBF (234°C) showing the interest of the incorporation of

d

a pendant phenyl or pyrimidine unit at the C4 of SBF. Decomposition temperature of the 2-substituted-SBF

occurs at 242°C (2-5Pm-SBF) and 238°C (2-Ph-SBF) close to the T of SBF, clearly highlighting that the

d

substitution in position 2 only has a weak effect on the thermal stability, whereas the substitution in

position 4 leads to a more stiff structure being hence more thermally stable. Thus, we note that the

position of the substitution has a clear influence on the T . On the other hand, 4-5Pm-SBF and 4-Ph-SBF

d

respectively present melting transitions at 173 and 206°C, whereas that of 2-5Pm-SBF is detected at ca

1

97/198° (see SI). It should be noted that the fusion of 2-Ph-SBF was very surprising appearing on a wide

-

8 -

ACCEPTED MANUSCRIPT

range of temperature. The substituent also has a clear influence on the whole thermal properties as the

replacement of phenyl by pyrimidine leads to different glass transitions (T ). Thus, we note that lower T_g

g

(

Figure 4, left) are detected (recorded on the second heating cycle of DSC, Figure 4 left) for phenyl-

substituted SBF (76 and 78°C for 4-Ph-SBF and 2-Ph-SBF respectively) compared to pyrimidine-substituted

SBF (85 and 94°C for 4-5Pm-SBF and 2-5Pm-SBF respectively). The slightly higher Tg of pyrimidine-

substituted compounds compared to phenyl-substituted compounds may be the consequence of stronger

…

50

intermolecular interactions due to CH N hydrogens bonding interactions. It is important to stress that all

51

compounds present T higher than those of classical host materials for PhOLEDs such as CBP (62°C) or m-

CP (55°C). Finally, it should be mentioned that 4-5Pm-SBF, 2-5Pm-SBF and 2-Ph-SBF do no present any

g

10

cristallization phenomena,T , being hence highly promising for devices incorporation. A cristallization

c

transition was only recorded for 4-Ph-SBF but at high temperature, 115°C, during the second heating cycle

in DSC.

Photophysical studies

The UV-Vis absorption and emission spectra recorded in cyclohexane are presented figure 5. All compounds

present similar absorption bands in the range 260-315 with the same maxima at 308 nm similar to the main

37,38,44

absorption band of SBF attributed to π−π∗ transition of the fluorene unit in SBF.

2

4

2

-Ph-SBF

-5Pm-SBF

-Ph-SBF

4-Ph-SBF

4-5Pm-SBF

1

0

,0

,8

,6

,4

,2

,0

2

-Ph-SBF

-5Pm-SBF

1

0

2

60 280 300 320 340 360 380 400

3

00

350

400

450

500

Wavelength (nm)

Figure 5. Normalized absorption (left) and emission (right, λ_e_x_c=300 nm for 4-Ph-SBF, 4-5Pm-SBF and 2-Ph-SBF and 305

-

6

nm for 2-5Pm-SBF) spectra of the four dyes recorded in cyclohexane, C=10 M.

At higher wavelength, each compound possesses its own signature with a clear supplemental absorption

peak at 320 or 319 nm for 2-5Pm-SBF or 2-Ph-SBF signing an extension of the conjugation between the

fluorene and the pyrimidine or the phenyl rings when the fluorene is substituted at C2 position. Literature

reports a λ_m_a_xof 328 nm (CH Cl ) for the disubstituted analogue 2,7-di-5Pm-9,9-dihexylfluorene (compound

2

F2 in chart 1) clearly showing an even more important extension of π-conjugation due to the presence of

30

two pyrimidine units. It means that efficient π-conjugation occurs between fluorene and pyrimidine unit

leading to a red shift of ca 8/10 nm per pyrimidine unit. In the case of 4-substituted SBF, we note the

presence of wavelength tail between 315 and 340 nm, which appears to be larger in the case of 4-5Pm-SBF

than in the case of 4-Ph-SBF. This tail may be due to a more limited π-conjugation between the pyrimidine

or the phenyl ring and the fluorene when substituted at the C4 position. It could also reflect a minor

proportion of conformers of the molecules with a more planar structure, allowing a certain degree of π-

conjugation between the pyrimidine or phenyl rings and the fluorene moiety. As these bands at low energy

seem to be more intense for pyrimidine-based dyes than for phenyl-based dyes, one may conclude that

more efficient π-conjugation seems to occur between fluorene and pyrimidine units.

-

9 -

ACCEPTED MANUSCRIPT

opt

From the onset absorption wavelength, optical energy gap (∆E ) varying from 3.66 eV for 2-5Pm-SBF to

3

.82 eV for 4-Ph-SBF were determined (see table 1) with a variation following the same trend as that

obtained from electrochemical measurement and theoretical calculations.

The experimental UV-vis spectra of the four dyes have been compared with predicted electronic absorption

spectrum from TD-DFT calculations (see figures in SI) and compared to that of its constituting building block

SBF. From those calculations one may observe that except 4-Ph-SBF, all compounds present an additional

absorption band at a lower energy than that the main band observed in SBF (see Figures in SI). This

additional band presents an oscillator strength value as high as 0.3552 for 2-Ph-SBF and 0.1677 for 2-5Pm-

SBF, similar to that calculated for the neighbouring band (0.3346 for 2-Ph-SBF and 0.1553 for 2-5Pm-SBF),

itself calculated close to the lowest energy band of SBF. This is in accordance with the experimental

conclusions from absorption spectroscopy presented above. Looking at the detailed transitions, for 2-Ph-

SBF and 2-5Pm-SBF, the lowest energy absorption band corresponds to an HOMO-LUMO transition

involving the 2-substituted-fluorene unit (see figure 3). For 4-5Pm-SBF, the two calculated lowest energy

bands present strongly less intense oscillator strength (0.0198 and 0.0303 for excited state 1 and 2

respectively) and the more important transitions involved in these two bands correspond to (i) HOMO vs

LUMO transition (65 % in excited state 1) involving two different molecular fragments of the fluorophore

(

HOMO possessing a spirobifluorenyl character and LUMO possessing a pyrimidine character) presenting

very low orbital overlap and hence a very strong charge transfer character and (ii) HOMO vs LUMO+1

transition (58.7% in excited excited state 2) involving two fluorenyl units and hence an orbital overlap

between them (HOMO possessing

a spirobifluorenyl character and LUMO+1 possessing a

pyrimidine/fluorenyl character). Finally, calculated UV-spectrum of 4-Ph-SBF is very similar to that

calculated for SBF (see SI) and the lowest energy absorption bands with very weak oscillator strength (0.06

and 0.059 for calculated excited states 1 and 2) correspond to the HOMO vs LUMO and HOMO vs LUMO+1

transitions, corresponding mainly to fluorene-fluorene transitions.

Fluorescence spectra of the four dyes (figure 5, right) clearly show drastically different emission spectra

depending on the substitution. The first type of emission spectra recorded for the 2-substituted-SBFs

presents a well resolved emission spectra with maxima at 338/355 and 335/351 nm for 2-5Pm-SBF and 2-

49

Ph-SBF respectively, signing the extension of π-conjugation compared to SBF (λ_m_a_x= 311/323 nm).

Structured emission spectrum was also reported for 2,7-di-5Pm-9,9-dihexylfluorene (compound F2 in chart

) with maxima at 361 and 378 nm due to an even more important extension of conjugation in the

1

30

disubstituted-fluorenyl core. 4-substituted SBFs presents structureless and large emission spectra centred

at 365 nm for 4-5Pm-SBF and 358 nm for 4-Ph-SBF. This surprising loss of resolution was also observed for

the rare examples of 4-substituted SBF found in literature, ie 4-diphenylphosphine oxide-SBF (λ_m_a_x: 346

3

9

41

42

nm) , 4-dibenzofuran-SBF (λ_m_a_x: 369 nm) and 4-dibenzothiophene-SBF (λ_m_a_x: 355 nm). This feature

appears hence to be a unique characteristic of 4-substituted SBF derivatives as 2-substituted SBF analogues

(

and 9,9'dialkylfluorene as well) always present well resolved emission bands assigned to the double bond

52-55

character of the C-C bond linking the pendant substituent and the fluorene core in the excited state.

In addition, there is also an intriguing red shift observed between 4-substituted-SBF and 2-substituted-SBF

derivatives (shift of 27 nm for 5Pm-SBFs and of 23 nm for Ph-SBFs). It is clear that the position of the

substitution (C2 vs C4) is at the origin of this spectacular effect. We could assume that the C4 substitution

of phenyl or pyrimidine fragments on the fluorene do not affect dramatically the π-conjugation in the

ground state (absorption spectra are only slightly altered with a low energy tail, as described above), but

contribute to a strong planarization in the excited state leading to an impressive Stokes shift. More detailed

theoretical and spectroscopic investigations need to be conducted to unravel this critical issue.

The emission quantum yields (Φsol) were determined in cyclohexane by using standard procedures with

quinine sulphate as reference. On one hand, phenyl-substituted SBFs appear to be efficiently fluorescent

with Φsol of 0.42 and 0.87 and lifetimes of 4.20 ns and 1.56 ns for 4-Ph-SBF and 2-Ph-SBF, respectively. On

the other hand, a quenching of fluorescence is observed for pyrimidine-substituted SBFs with Φsol of 0.04

and 0.02 and lifetimes of 0.61 ns and 0.44 ns for 4-5Pm-SBF and 2-5Pm-SBF, respectively. It should be

-

10 -

ACCEPTED MANUSCRIPT

noted that in other solvents, such as THF, the quantum yields of 4-5Pm-SBF and 2-5Pm-SBF remain very

low and the solvent can hence not be responsible of these low values. It must be noted that such a

decrease of the fluorescence was not observed for previously reported compounds with pyrimidine units

35

30

such as TBPSF (Φsol=1 in solution) and F3 (Φsol=0.85 in solution) (see structures in chart 1) bearing

respectively either a 5 pyrimidine or a 2 pyrimidine unit, both being also substituted with a phenyl ring. The

presence of this phenyl ring on the pyrimidine unit appears hence to be very important and we believe

that, in our case, the absence of this phenyl ring and hence the presence of the aromatic NCHN fragment in

the 5-pyrimidine unit has a strong impact on the deactivation rates of the excited state. Indeed, pyrimidine

substituted derivatives 4-5Pm-SBF and 2-5Pm-SBF show a lower radiative rate constant k and a much

r

higher non-radiative rate constant k compared to phenyl-substituted derivatives 4-Ph-SBF and 2-Ph-SBF,

nr

as mentioned in Table 3.

Table 3. Optical properties of the four SBF dyes.

4

-5Pm-SBF

4-Ph-SBF

2-5Pm-SBF

2-Ph-SBF

λabs, solution (nm)

ε (10 , L.mol .m )

297 (0.5), 308

(0.6), 318 (sh)

297 (0.7), 308 (1.4), 296 (2.5), 308 (2.5), 297 (2.4), 308 (2.2),

4

-1

-2 a

318 (sh)

298, 311

320 (1.8)

319 (1.6)

b

λabs, thin-film (nm)

298, 311

301, 312, 326

301, 312, 329

a

λem, solution (nm)

λem, thin-film (nm)

365

372

358

363

0.42

4.20

338, 355

355, 366

0.02

335, 351

343, 359

0.87

b

a,c

Φsol

0.04

0.61

a

τ (ns)

0.44

1.56

-

1 a

7

8

7

8

k (s )

6.6×10

1.6×10

1.0×10

1.4×10

4.6×10

5.6×10

r

-

1 a

9

8

9

7

k (s )

2.2×10

8.3×10

nr

a. in cyclohexane, b. Spin-coated thin film from a solution in THF, [solution] = 10 mg/mL), c. Quantum yields in solution (Φsol) were

calculated relative to quinine sulfate (Φsol = 0.546 in H SO 1N)

2

4

From a careful inspection of the dependence of the absorption behaviour of the four dyes on solvent

polarity, we note that the absorption maximum is almost insensitive to the dielectric constant of the

environment (see Figures in SI). This indicates that the electronic and structural characteristics of the

56

ground and Franck Condon excited states do not differ much with a change in solvent polarity.

While phenyl-substituted SBFs showed minor differences of the PL spectra in various solvents (Figure 6,

top), distinct fluorescence solvatochromism was observed in pyrimidine-substituted SBF when the solvent

polarity increases from cyclohexane to ethanol (Figure 6, bottom). Indeed, both dyes are highly sensitive to

the solvent polarity with fluorescence maxima ranging from 366 and 338 nm in cyclohexane to 390 and 375

nm in ethanol for 4-5Pm-SBF and 2-5Pm-SBF respectively. This dependence of the emission wavelength on

the solvent polarity is indicative of dipole-dipole interactions between Pm-substituted SBFs and polar

solvent molecules and hence of a photoinduced intramolecular charge transfer between the donor

57

(

fluorene) and the acceptor part (pyrimidine) of the dye.

-

11 -

ACCEPTED MANUSCRIPT

Cyclohexane

Toluene

THF

Cyclohexane

Toluene

THF

1,0

1

0

,0

,8

,6

,4

,2

,0

CH Cl

DCM

EtOH

ACN

2

Ethanol

0

,8

,6

,4

,2

,0

CH CN

3

00

350

400

450

500

350

375

400

450

Wavelength (nm)

Cyclohexane

Toluene

THF

Cyclohexane

Toluene

THF

338

1,0

CH Cl₂

2

3

66

390

1

0

,0

,8

,6

,4

,2

,0

Ethanol

CH Cl

2

0,8

0,6

0,4

0,2

0,0

CH CN

Ethanol

3

CH CN

3

25 350 375 400 425 450 475 500

3

50

400

450

500

Wavelength (nm)

Figure 6. Normalized PL spectra of solutions of 4-Ph-SBF (top left) and 2-Ph-SBF (top right) and of 4-5Pm-SBF (bottom

left) and 2-5Pm-SBF (bottom right) in cyclohexane, toluene, THF, CH Cl , ethanol and CH CN

2

3

The solid state absorption spectra of all dyes (figure 7, left) appear very similar to those in solution (figure 5

left) with only a small red shift of less than 8 nm detected and a slight increase of the lowest energy band

intensity (compared to the band at 310 nm) for the 2-substituted-SBFs. Such similarity between absorption

spectra in solution and in solid state clearly signs that there are minimal intermolecular interactions in the

ground state in thin films.

2

4

2

-Ph-SBF

-5Pm-SBF

-Ph-SBF

4

2

-Ph-SBF

-5Pm-SBF

-Ph-SBF

1,0

0

,8

,6

,4

,2

,0

2-5Pm-SBF

-5Pm-SBF

1

0

3

00

350

400

450

500

260 280 300 320 340 360 380 400

Wavelength (nm)

Figure 7. Absorption (left) and Emission (right) spectra of the four dyes recorded in solid state (spin-coated thin film

from a solution in THF, [solution] = 10 mg/mL).

Regarding the solid-state fluorescent properties (figure 7, right), except for 2-5Pm-SBF, they are almost

identical to those recorded in solution (figure 5, right) with only a small red shift ranging from 3 nm for 4-

Ph-SBF to 8 nm for 4-5Pm-SBF and 2-Ph-SBF. Solid-state fluorescent spectrum of 2-5Pm-SBF (figure 7, right,

pink line) nevertheless appears poorly resolved with two maxima recorded at 355/366 nm, red shifted by

1

7/11 nm compared to that in solution. Finally, to examine the suitability of the four dyes as host materials

for phosphorescent green and/or blue dopant, the phosphorescence of the molecules was recorded at 77 K

in a methylcyclohexane/2-methylpentane mixture (1:1), Figure 8. From those spectra, E values of the

T

molecules were calculated from the lowest phosphorescent peak of 2.75 eV for 4-5Pm-SBF and 2.56 eV for

-

12 -

ACCEPTED MANUSCRIPT

2

-Ph-SBF. Compared to the E value of 4-Ph-SBF (2.77 eV) and of SBF (2.87 eV) reported by our group

T

44

recently, we note that the E level of 4-5-Pm-SBF is sufficiently high for its use as host material for both

T

green and blue phosphorescent dyes. The E of 2-Ph-SBF (2.56 eV) and that of 2-5-Pm-SBF (2.58 eV) is

T

58

lower than that of the blue dopant FIrpic (E : 2.62 eV ) and these molecules may be used as host material

T

59

for green dopant Irppy (E : 2.42 eV ).

3

T

4

-Ph-SBF

4-5Pm-SBF

450

500

550

600

400

450

500

550

Wavelength (nm)

2-Ph-SBF

2-5Pm-SBF

4

50

500

550

600

450

500

550

Wavelength (nm)

Figure 8. Emission spectra at 77 K in a methylcyclohexane/2-methylpentane mixture (1:1) of 4-Ph-SBF ( _e_x_c

Pm-SBF (λ_e_x_c= 300 nm) and 2-Ph-SBF (λ_e_x_c= 250 nm) and 2-5PmSBF (λ_e_x_c= 250 nm). The determination of the Et value

has been done at from the lowest phosphorescent peak (black arrow)

λ = 300 nm, 4-

5

In conclusion of the above presented photophysical and physico-chemical analysis, the position (C2 vs C4)

and the nature of the substituent (phenyl vs pyrimidine) on the SBF core leads to drastic alterations of the

resulting properties. Substitution at the C2 leads to an increase of the conjugation length and therefore to a

decrease of the singlet and triplet state energy level, while the substitution at the C4 only has a weak

influence on these two key parameters, keeping them almost identical to those of their constituting

building block SBF. This has been assigned to π-conjugation breaking induced by the large torsion angle

made by the substituents in C4. This chemical design allows hence obtaining organic semiconductors based

on

a C4-substituted SBF with high singlet and triplet state energy level with excellent

thermal/morphological properties. On the other hand, the substitution by the pyrimidine group leads to a

decrease of the LUMO level due to the pyrimidine electron affinity, which should lead in principle to a

decrease of the threshold voltage of the resulting devices. Thus, the four dyes possess the prerequisites for

their uses as host in PhOLEDs that is, a high T and Td, HOMO/LUMO levels adapted for hole and electron

g

injection/transport in the devices and E higher that of the green phosphorescent emitter Ir(ppy) (2.42

T

3

58

59

eV ) and in the case of 4-Ph-SBF and 4-5Pm-SBF for the blue phosphorescent emitter FIrpic (2.62 eV ).

-

13 -

ACCEPTED MANUSCRIPT

EML

-1.95

-2.1

-2.3

-2.4

LiF/Al

2.9 eV

-2.7

-

2.9

-

3.0

-

3.5

CuPc

NPB

TCTA

5.7 eV

TPBI

Ir(ppy)₃

ITO

4.7 eV -5.2

FIrpic

-

5.4 eV

-

5.6

-

-5.8

-5.88

-5.95

-

5.96

-5.97

-

6.2 eV

Scheme 2. Energy level diagram of the devices.

In order to explore the potential of the present semi-conductors as host for triplet emitters, green (Ir(ppy)₃)

and blue (FIrpic) PhOLEDs were fabricated. The device configuration was ITO/CuPc(10 nm)/NPB (40

nm)/TCTA (10 nm)/ EML:dopant (20 nm)/TPBi (40 nm)/LiF (1.2 nm)/Al (100 nm). ITO is used as the anode,

CuPc (copper phtalocyanine) is the hole injecting layer, NPB (N,N’-di(1-naphtyl)-N,N’-diphenyl-[1,1’-biphe-

nyl]-4,4’-diamine) is the hole-transporting layer, TCTA (4,4',4''-Tris(carbazol-9-yl)-triphenylamine) is the

electron/exciton blocking layer, TPBI (1,3,5-Tris(1-phenyl-1H-benzimidazol-2-yl)benzene) is both the

electron transporting layer and the hole blocking layer and a thin film of lithium fluoride covered with

aluminum is the cathode. Ir(PPy) and FIrpic are used as dopant for green and blue OLEDs respectively. The

3

relative energy levels of the successive layers of the devices are reported in scheme 2. Table 4 gathers

selected data obtained from the average of two experiments.

Due to their E values higher than that of Ir(ppy) , the four substituted-SBFs were used as host for Ir(ppy)

3

T

3

green dopant. Performance of the best green PhOLEDs is presented in figure 9 (left) and average

parameters are gathered in table 4. The best green PhOLED performance was obtained with 2-5Pm-SBF and

4

-5Pm-SBF host materials doped with 10% Ir(ppy) . As shown in figure 9 (left), maximum current and power

3

efficiencies of 58 cd/A and 35 lm/W are obtained respectively for the pyrimidine substituted SBF. Maximum

2

luminance as high as 30 000 cd/m is reached for these devices as shown in table 4. Phenyl substituted

dyes, 4-Ph-SBF and 2-Ph-SBF presents lower performance with EQE of 10.4% and 11.3 %, highlighting the

great interest of the incorporation of the pyrimidine unit in SBF scaffold. The threshold voltage for all

devices appears to be low and as low as 2.5 eV for 2-5Pm-SBF.

The EL spectra of the four devices based on the different host materials are identical, exclusively showing

the emission of the green dopant at 516/540 nm (figure 9 right) close to the PL emission of pure Ir(ppy)

film (509/540 nm) with no parasite emission in the range of the undoped device. This result demonstrates

3

57

an efficient energy transfer from the four SBF-derivatives to Ir(ppy) , due to their high triplet energy levels.

3

The device performance obtained with small pure hydrocarbon host molecules 4-Ph-SBF and 2-Ph-SBF are

comparable to those obtained with more complicated pure-hydrocarbon-based host materials recently

38,37,58

reported,

Performances obtained with pyrimidine-substituted SBF are even more interesting with

lowest turn-on voltages and highest EQE (close to 14%) and clearly demonstrate the interest of the

pyrimidine substitution in notably favouring the injection of electron.

-

14 -

ACCEPTED MANUSCRIPT

Table 4. Selected EL data of green and bue devices (average values calculated from two experiments).

at J= 10 mA/cm²

max

-2

Host

Guest (%)

V_o_n(V) EQE (%)

Current

Power

efficiency

(lm/W)

L

(cd.m ) at

CIE (x ; y) at

10 mA.cm

-

2

/

efficiency

current density

-

1

-2

(

cd.A )

(mA.cm )

Green Phosphorescent OLEDs

2

-Ph-SBF

2.6

2.5

3.5

3.3

11.3

13.3

10.4

13.7

42.7

49.2

40.3

49.9

17.8

21.5

16.7

17.6

27 990 (210)

30 000 (160)

25 700 (230)

28 000 (150)

0.33 ; 0.62

0.33 ; 0.61

0.32 ; 0.62

0.34 ; 0.61

/

Ir(ppy) (10)

3

2

-5Pm-SBF

/

Ir(ppy) (9)

3

4

-Ph-SBF

/

Ir(ppy) (10)

3

4

-5Pm-SBF

/

Ir(ppy) (10)

3

Blue Phosphorescent OLEDs

4

/

-Ph-SBF

FIrPic (19)

-5Pm-SBF

FIrPic (19)

4.0

3.3

5.7

4.8

16.2

6.0

3800 (70)

2030 (50)

0.20 ; 0.44

0.20 ; 0.45

4

/

14.1

5.5

8

7

6

5

4

3

2

1

0

4

-5Pm-SBF

4-Ph-SBF

4-5Pm-SBF

2-Ph-SBF

2-5Pm-SBF

at 10mA/cm²

0

4-Ph-SBF

1,00

2-Ph-SBF

2-5Pm-SBF

0

,75

,50

0,25

0

,00

4

00

500

600

700

800

0

10

20

30

40

2

50

wavelength (nm)

Current density (mA/cm )

Figure 9. Current (empty symbol) and power efficiencies (filled symbol) versus current density of the green devices

using 4-Ph-SBF, 4-5Pm-SBF, 2-Ph-SBF and 2-5Pm-SBF doped with Ir(ppy) (10% in mass) as emitting layer (left).

3

2

Corresponding EL spectra recorded at 10 mA/cm (right).

Due to their high E values (2.77 eV for 4-Ph-SBF and 2.75 eV for 4-5Pm-SBF), higher than that of FIrpic (E :

T

2

.62 eV), the two 4-substituted-SBFs were used as host for this blue dopant. Characteristics of the two blue

devices are presented in figure 10. The best performance is obtained with 4-Ph-SBF with an EQE of 5.7 %, a

LE of 16.2 cd/A and a threshold voltage of 4 V while blue PhOLEDs with 4-5Pm-SBF as host reach an EQE of

2

4

.8 %, a LE of 14.1 cd/A and a threshold voltage of 3.3 V reported at a current density value of 10 mA/cm

(

Figure 10 left and Table 4). The threshold voltage of 4-5Pm-SBF is hence decreased by ca 0.7 V compared

to that of 4-Ph-SBF due to the incorporation of the pyrimidinium group in the former (which leads to a

decrease of the LUMO level and hence an easier injection of the electron in the device). This highlights the

efficiency of the chemical design investigated in this work. Both devices, reach brightness of 3800 / 2030

2

cd/m for 4-Ph-SBF / 4-5Pm-SBF based devices respectively.

EL spectra of the devices (Figure 10 right) point a main emission of the blue dopant with maxima at 473 and

5

9

5

00 nm close to the PL emission of pure FIrpic film (475/500 nm) but also, in the case of 4-PhSBF of an

additional tiny emission in the range of the undoped device (360/440 nm). This emission may be due to a

small amount of recombination in the other organic layers and especially in TCTA. CIE coordinates of 4-Ph-

SBF-based device (0.20 ; 0.44) and of 4-5Pm-SBF-based device (0.20 ; 0.45) appear nevertheless almost

identical as shown in Table 4.

-

15 -

ACCEPTED MANUSCRIPT

30

25

20

15

10

5

0

4

-5Pm-SBF

4-Ph-SBF

4-5Pm-SBF

at 10mA/cm²

4-Ph-SBF

1

0

,00

,75

,50

,25

0,00

0

10

20

30

40

2

50

400

500

600

700

800

wavelength (nm)

Current density (mA/cm )

Figure 10. Current (empty symbol) and power efficiencies (filled symbol) versus current density of the blue devices

using 4-Ph-SBF and 4-5Pm-SBF doped with FIrPic (19% in mass) as emitting layer (left). Corresponding EL spectra

2

recorded at a current density of 10 mA/cm (right).

Conclusion

In summary, four substituted SBF dyes have been synthesized and their thermal, electrochemical and

photophysical properties have been scrutinized and compared to the properties of their constituting SBF

unit. From these studies it has been demonstrated that, compare to SBF, the substitution by a phenyl unit

at the C4 position of SBF leads to weaker modification of the HOMO and LUMO levels compared to the

phenyl substitution at the C2 position, which leads to a conjugation extension and hence a decrease of the

HOMO-LUMO gap. On the other hand, the substitution by the electro-deficient pyrimidine unit compared

to the substitution by the phenyl unit leads to a significant lowering of the LUMO level. Overall, the

combination of the two effects (lowering of the HOMO through the extension of conjugation, lowering of

the LUMO through pyrimidine electro deficient effect) in 2-5Pm-SBF leads to the highest modification of

the optical energy gap (from 4.05 eV in SBF to 3.66 eV in 2-5Pm-SBF). Thus, the optical energy gap of the

four compounds is finely tuned from 4 eV for 4-Ph-SBF to 3.66 eV for 2-5Pm-SBF. In addition, the

substitution in C2 or in C4 has an impressive influence on the E . Indeed, the C2-substituted dyes have E

T

values at ca 2.42 eV, whereas the C4 substituted dyes have an E as high as 2.7 eV due to the π-conjugation

T

disruption. This structure properties relationship study clearly highlights how a different substitution may

lead to very different properties keeping others intact. As all compounds possess a triplet energy level

higher than 2.42 eV, they have been used as host for green dopant in PhOLEDs. Remarkable performance

has been obtained with 2-5Pm-SBF as host (CE > 58 cd/A, PE > 35 lm/W, EQE > 14 % and threshold voltage

of 2.5 V). Furthermore, the two 4-substituted SBF with triplet energy levels as high as 2.8 eV have also been

successfully used as host materials for blue PhOLEDs with high performance. The effect of the pyrimidine

substituent leads to a lowering of the threshold working voltage of the device from 4 V for 4-Ph-SBF based

blue device to 3.3 V for 4-5Pm-SBF based blue device.

Acknowledgements

S.T. and G. S. thank ANR (HOME-OLED Project n°ANR-11-BS07-020-01) for a studentship. C. D thanks ANR

(

HOME-OLED Project n°ANR-11-BS07-020-01) for a financial support as postdoctoral position. We wish to

thank the C.R.M.P.O (Centre Régional de Mesures Physiques de l'Ouest, Rennes) for high resolution mass

measurements, Service de Microanalyse-CNRS (Gif sur Yvette) for CHN analyses, CINES (Montpellier) for

computing time, the CDIFX (Rennes) for X-Ray data collection, the "Institut des Sciences Analytiques" (UMR

CNRS 5280, Villeurbanne) for TGA. Arnaud Brosseau (Cachan) is thanked for his precious assistance in

photophysical measurements. Dr Franck Camerel (Rennes) is warmly acknowledged for his help and advice

-

16 -

ACCEPTED MANUSCRIPT

in the understanding of thermal properties. Prof. Abdou Boucekkine is also warmly thanked for his help in

DFT and TD-DFT calculations.

Experimental Section

Synthesis:

All manipulations of oxygen- and moisture-sensitive materials were conducted with a standard

Schlenk technique. Commercially available reagents and solvents were used without further purification

other than those detailed below. THF was distilled from sodium/benzophenone prior to use. Light

petroleum refers to the fraction with bp 40-60°C. 2.5 M solutions of n-BuLi in hexanes or THF were

purchased from Sigma Aldrich. 2,2’-dibromobiphenyl was purchased from Fluorochem. Reactions were

stirred magnetically, unless otherwise indicated. Analytical thin layer chromatography was carried out using

aluminum backed plates coated with Merck Kieselgel 60 GF254 and visualized under UV light (at 254 and

3

2

60 nm). Chromatography was carried out using Teledyne Isco CombiFlash® Rf 400 (UV detection

00-360nm), over standard silica cartridges (Redisep® Isco, GraceResolv™ Grace or Puriflash® columns

1

13

1

Interchim). H and C NMR spectra were recorded using Bruker 300 MHz instruments ( H frequency,

corresponding C frequency: 75 MHz); chemical shifts were recorded in ppm and J values in Hz. In the

13

C

NMR spectra, signals corresponding to C, CH, CH or Me groups, assigned from DEPT, are noted. The

2

residual signals for the NMR solvents are: CDCl ; 7.26 ppm for the proton and 77.00 ppm for the carbon,

3

CD Cl ; 5.32 ppm for the proton and 53.80 ppm for the carbon. The following abbreviations have been used

2

for the NMR assignment: s for singlet, d for doublet, t for triplet and m for multiplet. High resolution mass

spectra were recorded at the Centre Régional de Mesures Physiques de l'Ouest (Rennes) on (i) Bruker

MicrO-Tof-Q II (Source: Atmospheric Pressure Chemical Ionization (APCI - direct introduction) (ASAP–

Atmospheric Solids Analysis Probe) at a temperature of 30°C - positive mode) or on (ii) Waters Q-Tof II. The

44

synthesis of 4-bromo-9,9'-spirobi[fluorene] (1) and of 4-Ph-SBF have been previously described

6

3

2

-bromo-9,9'-spirobi[fluorene] (2) 2-bromo-1,1’-biphenyl (2.16 g, 9.27 mmol, 1.15 eq) was dissolved

in dry THF (40 mL) under argon atmosphere, cooled at -78°C, and stirred during 10 minutes at this

temperature. A 2.5 M n-BuLi solution in THF (4.31 mL, 10.78 mmol, 1.34 eq) was then slowly injected via a

seringue, at -78°C. The resulting yellow mixture was stirred at the same temperature for 30 min. 2-bromo-

9

H-fluoren-9-one (2.08 g, 8.03 mmol) in dry THF (35 mL) was then added dropwise, and the mixture was

stirred for another 30 min at -78°C, and allows warming up to room temperature gradually and stirred

overnight. Saturated brine solution (10 mL) was added, and the mixture was extracted twice with ethyl

acetate (2x30 mL). The combined organic extracts were dried over anhydrous magnesium sulfate, filtered,

and concentrated under reduced pressure. Without other purification, the crude was dissolved into a

mixture of acetic acid/hydrochloric acid (50 mL/5 mL) and warmed at 70°C during 2 h under stirring. Then,

the mixture was poured into water/ice (200 mL), and the solution was neutralized with solid sodium

hydroxide until pH 7. Then the organic layer was extracted three times with dichloromethane (3x50 mL).

The combined organic extracts were dried over magnesium sulfate, filtered, and concentrated under

reduced pressure. The residue was purified by flash chromatography on silica gel (light petro-

leum/dichloromethane 95/5) to afford a colorless solid (2.14 g). [column conditions: Silica cartridge 40 g

(

2

Serlabo); solid deposit on Celite®; λ_d_e_t_e_c_t_i_o_n: (254 nm, 280 nm); light petroleum/dichloromethane (95/5) at

5 mL/min; collected fraction: 18-42 min]. Yield: 67%. mp: 168-170°C (lit. 160 °C); H NMR (300 MHz,

64 1

CD Cl ): δ 7.89 (m, 3H, ArH), 7.76 (d, J= 8.1 Hz, 1H, ArH), 7.52 (dd, J = 8.1, 1.8 Hz; 1H, ArH), 7.42 (m, 3H,

2

1

3

ArH), 7.15 (m, 3H, ArH), 6.87 (m, 1H, ArH), 6.71 (m, 3H, ArH); C NMR (75 MHz, CD Cl ): δ 151.5 (C), 149.1

2

(

1

C), 148.5 (C), 142.4 (C), 141.4 (C), 141.3 (C), 131.4 (CH), 128.8 (CH), 128.6 (CH), 128.54 (CH), 128.48 (CH),

27.6 (CH), 124.37 (CH), 124.35 (CH), 122.1 (CH), 121.8 (C), 120.8 (2xCH seen from HSQC), 66.4 (Spiro C).

General procedure for Suzuki coupling

,9'-spirobi[fluorene] brominated compound, boronic acid compound, K CO and Pd(dppf)Cl were

*

2

9

2

3

dissolved in dry DMF or xylenes. The mixture was warmed to 150°C (DMF) and 135°C (xylenes) stirred

overnight. After cooling to room temperature, saturated solution of ammonium chloride (50 mL) was ad-

*

Pd(dppf)Cl : [1,1'-Bis(diphenylphosphino)ferrocene]palladium(II) dichloride complex with dichloromethane

2

-

17 -

ACCEPTED MANUSCRIPT

ded, and organic layer was extracted three times with dichloromethane (3x30 mL) and washed with brine

(

3x30 mL). The combined organic extracts were dried over magnesium sulfate, filtered, and concentrated

under reduced pressure. The residue was purified by flash chromatography on silica gel using Teledyne Isco

CombiFlash® Rf 400. Column conditions can be found in each compounds description below.

5

-(9,9'-spirobi[fluoren]-4-yl)pyrimidine (4-5Pm-SBF). 5-(9,9'-spirobi[fluoren]-4-yl)pyrimidine was synthesi-

44

zed according to the general coupling procedure between 4-bromo-9,9'-spirobi[fluorene] (1) (0.808 g,

.045 mmol), pyrimidin-5-yl-boronic acid (0.333 g, 2.688 mmol, 1.31 eq), K CO (2.800 g, 20.260 mmol, 9.91

2

3

eq) and Pd(dppf)Cl (0.088 g, 0.108 mmol, 0.05 eq) in dry DMF (30 mL). The residue was purified by flash

2

chromatography on silica gel (light petroleum/ethyl acetate 85/15) to afford a colorless solid (0.759 mg,

1

2

.924 mmol) [column conditions: Silica cartridge 12 g (Serlabo); solid deposit on Celite®; λ_d_e_t_e_c_t_i_o_n: (254 nm,

80 nm); ethyl acetate in ligth petroleum (85/15) at 20 mL/min; collected fraction: 8-17 min]. Yield: 94%.

-

1

mp: 167°C; IR (ATR, cm ) ν = 593, 646, 723, 752, 908, 1030, 1157, 1282 1405, 1444, 1545, 3020, 3072; H

NMR (300Mhz, CD Cl ): δ 9.36 (s, 1H, ArH), 9.04 (s, 2H, ArH), 7.89 (m, 2H, ArH), 7.41 (td, J = 7.5, 1.2 Hz, 2H,

2

13

ArH), 7.22-7.01 (m, 7H, ArH), 6.76 (m, 3H, ArH), 6.68 (s, 1H, ArH); C NMR (75MHz, CD Cl ): δ 158.6 (CH),

2

1

57.4 (CH), 150.7 (C), 149.9 (C), 149.0 (C), 142.5 (C), 141.3 (C), 139.7 (C), 135.1 (C), 130.8 (C), 130.7 (CH),

28.7 (CH), 128.53 (CH), 128.48 (CH), 128.3 (CH), 128.2 (CH), 124.9 (CH), 124.7 (CH), 124.3 (CH), 123.0 (CH),

+

20.8 (CH), 66.24 (Spiro C) ; HRMS calculated for C H N 395.1542, found 395.1542 [M+H] ; Elemental

29

19

2

analysis calculated for C H N : C, 88.30 %; H, 4.60 %; N, 7.10 %. Found: C, 87.81 %; H, 4.81 %; N, 6.81 %;

2

9

18

2

4

-1

λ_a_b_s[nm] (ε[10 .L.mol .cm ]) = 297 (0.5), 308 (0.6).

2

-phenyl-9,9'-spirobi[fluorene] (2-Ph-SBF). 2-phenyl-9,9'-spirobi[fluorene] was synthesized according to

the general procedure between 2-bromo-9,9’-spirobi[fluorene] (2) (0.751 g, 1.901 mmol), phenylboronic

acid (0.304 g, 2.488 mmol, 1.31 eq), K CO (1.227 g, 8.883 mmol, 4.67 eq) and Pd(dppf)Cl (0.061 g, 0.076

2

3

2

mmol, 0.04 eq) in xylenes (20 mL). The residue was purified by flash chromatography on silica gel (light

petroleum/dichloromethane 9/1) to afford a colorless solid (0.581 g, 1.482 mmol), [column conditions:

Silica cartridge 24 g (Serlabo); solid deposit on Celite®; λ_d_e_t_e_c_t_i_o_n: (254 nm, 280 nm); dichloromethane in light

*

-1

petroleum (9/1) at 20 mL/min; collected fraction: 35-45 min]. Yield: 78 %.mp:100-108 IR (ATR, cm ) ν =

1

4

76, 494, 617, 634, 650, 692, 727, 748, 829, 1020, 1155, 1280, 1444, 1597, 3014, 3059; H NMR (300 MHz,

CD Cl ): δ 7.96 (dd, J = 8.0, 0.6 Hz, 1H, ArH), 7.93 – 7.88 (m, 3H, ArH), 7.66 (dd, J = 8.0, 1.7 Hz, 1H, ArH), 7.45

2

–

7.36 (m, 5H, ArH), 7.35 – 7.20 (m, 3H, ArH), 7.13 (td, J = 7.5, 1.1 Hz, 3H, ArH), 6.94 (dd, J = 1.7, 0.6 Hz, 1H,

13

ArH), 6.78 – 6.72 (m, 2H, ArH), 6.72 – 6.67 (m, 1H, ArH); C NMR (75 MHz, CD Cl ): δ 150.1 (C), 149.7 (C),

1

2

†

49.2 (C), 142.4 (C), 142.0 (C), 141.7 (C), 141.4 (C), 141.3 (C), 129.1 (CH), 128.40 (2xCH ), 128.38 (CH),

28.37 (CH), 127.8 (CH), 127.43 (CH), 127.38 (CH), 124.4 (CH), 124.3 (CH), 123.0 (CH), 121.0 (CH), 120.8

+

(

CH), 120.7 (CH), 66.6 (Spiro C); HRMS calculated for C H 393.1643, found 393.1646 [M+H] ; Elemental

31 21

4

-

analysis calculated for C H : C, 94.86 %; H, 5.14 %. Found: C, 94.93 %; H, 5.24 %; λ [nm] (ε[10 .L.mol

31

20

abs

1

.

-1

cm ]) = 297 (2.4), 308 (2.2), 319 (1.6).

5

-(9,9'-spirobi[fluoren]-2-yl)pyrimidine (2-5Pm-SBF). 5-(9,9'-spirobi[fluoren]-2-yl)pyrimidine was synthesi-

zed according to the general procedure between 2-bromo-9,9’-spirobi[fluorene] (2) (0.749 g, 1.897 mmol),

pyrimidin-5-yl-boronic acid (0.312 g, 2.519 mmol, 1.33 eq), K CO (1.226 g, 8.872 mmol, 4.68 eq) and

2

3

Pd(dppf)Cl (0.059 g, 0.073 mmol, 0.04 eq) in dry DMF (20 mL). The residue was purified by flash

2

chromatography on silica gel (light petroleum/ethyl acetate 8/2) to afford a colorless solid (0.5743 g, 1.456

mmol) [column conditions: Silica cartridge 24 g (Serlabo); solid deposit on Celite®; λ_d_e_t_e_c_t_i_o_n: (254 nm, 280

nm); ethyl acetate in light petroleum (8/2) at 20 mL/min; collected fraction: 34-50 min]. Yield (77 %). mp:

-

1

7

97-198°C; IR (ATR, cm ) ν = 491, 604, 619, 636, 721, 731, 750, 833, 862, 1018, 1105, 1153, 1186, 1267,

1

400, 1414, 1446, 1554, 3016, 3039, 3053; H NMR (300 MHz, CD Cl ): δ 8.94 (s, 1H, ArH), 8.67 (s, 2H, ArH),

.93 (dd, J = 7.9, 0.6 Hz, 1H, ArH), 7.87 – 7.83 (m, 1H, ArH), 7.83 – 7.79 (m, 2H, ArH), 7.56 (dd, J = 7.9, 1.7

2

Hz, 1H, ArH), 7.37 – 7.26 (m, 3H, ArH), 7.06 (dtd, J = 10.2, 7.5, 1.1 Hz, 3H, ArH), 6.86 (dd, J = 1.7, 0.6 Hz, 1H,

*

This transition appears to be not very clear, occurring on a large range.

In the light of the protons assignment of 2-5Pm-SBF, we believe that 2 carbons are present at 128.40 ppm.

†

-

18 -

ACCEPTED MANUSCRIPT

1

3

ArH), 6.68 – 6.60 (m, 3H, ArH); C NMR (75 MHz, CD Cl ): δ 157.8 (CH), 155.2 (CH), 150.7 (C), 149.7 (C),

2

1

48.8 (C), 143.2 (C), 142.4 (C), 141.4 (C), 134.53 (C), 134.45 (C), 129.0 (CH), 128.6 (CH), 128.5 (2xCH seen

from HSQC), 127.3 (CH), 124.42 (CH), 124.35 (CH), 122.9 (CH), 121.6 (CH), 121.1 (CH), 120.8 (CH), 66.6

+

4

-1

(

Spiro C); HRMS calculated for C H N 395.1548, found 395.1548 [M+H] ; λ [nm] (ε[10 .L.mol .cm ]) =

29 19 2 abs

2

96 (2.5), 308 (2.5), 320 (1.8).

X-Ray :

Crystal was picked up with a cryoloop and then frozen at 150 K under a stream of dry N on a APEX

2

II Brucker AXS diffractometer for X-ray data collection (Mo Kα radiation, λ = 0.71073 Å). 4-Ph-SBF structure

44

was published previously.

-Ph-SBF (C H )); M = 784.94. APEXII, Bruker-AXS diffractometer, Mo-Kα radiation (λ = 0.71073

2

62

40

Å), T = 150(2) K; triclinic P-1 (I.T.2), a = 10.9027(4), b = 13.7902(6), c = 15.7771(5) Å, α = 102.1400(10), β =

3

-3

-1

9

7.4710(10), γ = 111.5900(10) °, V = 2099.47(14) Å .Z = 2, d = 1.242 g.cm , μ = 0.07 mm . The structure was

65

solved by direct methods using the SIR97 program, and then refined with full-matrix least-square

methods based on F (SHELXL-97) with the aid of the WINGX program. All non-hydrogen atoms were

2

66

67

refined with anisotropic atomic displacement parameters. H atoms were finally included in their calculated

2

positions. A final refinement on F with 9543 unique intensities and 559 parameters converged at ωR(F ) =

0

.1128 (R(F) = 0.0447) for 7649 observed reflections with I > 2σ(I).

-5Pm-SBF (C H N , C H O); M = 426.5. APEXII, Bruker-AXS diffractometer, Mo-Kα radiation (λ =

2

29

18

2

4

0

4

.71073 Å), T = 150(2) K; orthorhombic P c a b (I.T.#61), a = 8.4342(3), b = 13.6588(5), c = 38.5506(13) Å, V =

3 -3 -1

441.1(3) Å .Z = 8, d = 1.276 g.cm , μ = 0.078 mm . The structure was solved by direct methods using

2

the SIR97 program [1], and then refined with full-matrix least-square methods based on F (SHELXL-97) [2]

with the aid of the WINGX [3] program. All non-hydrogen atoms were refined with anisotropic atomic

displacement parameters. H atoms were finally included in their calculated positions. A final refinement

2

on F with 5055 unique intensities and 300 parameters converged at ωR(F ) = 0.1246 (R(F) = 0.0461) for

820 observed reflections with I > 2σ(I).The figures were generated with Mercury software 3.3

3

Spectroscopic studies:

Cyclohexane (analytical grade, VWR) was used without further purification. Standard 1N solution of

sulfuric acid was purcharsed from Alfa Aesar. UV-visible spectra were recorded using an UV-Visible

spectrophotometer SHIMADZU UV-1605. The energy gap was calculated from the absorption edge of the

opt

UV-vis absorption spectra in solution in cyclohexane, using the formula ΔE (eV) = hc/λ, λ being the

-

34

-19

8

-1

absorption edge (in meter). With h = 6.6x10 J.s (1eV = 1.6x10 J) and c = 3.0x10 m.s , this equation may

opt

be simplified as: ΔE (eV) = 1237.5/λ (in nm). Triplet energy level E was calculated from the maximum of

T

the first phosphorescence emission peak, and conversion in electron-volt was obtained with the previous

formula. Emission spectra were recorded with a PTI spectrofluorimeter (PTI-814 PDS, MD 5020, LPS 220B)

using a Xenon lamp. Quantum yields in solution (ø ) were calculated relative to quinine sulfate (ø = 0.546

sol

in H SO 1N). ø was determined according to the following equation,

2

4

sol

where, subscripts s and r refer respectively to the sample and reference. The integrated area of the

emission peak in arbitrary units is given as T, n is the refracting index of the solvent (n = 1.426 for cyclohe-

s

xane) and A is the absorbance. Three solutions of different concentration of the substrate (A<0.1) and three

solutions of the reference (quinine sulfate) were prepared. The quinine sulfate concentration was chosen

so as the absorption of the reference and the substrate were the same at the excitation wavelength. Three

quantum yields were then calculated at this wavelength and the average value is reported. IR spectra were

recorded on a Bruker Vertex 70 using a diamond crystal MIRacle ATR (Pike). Thin films were prepared by

spin-coating 300 μL of solution of THF (10 g/L) on a sapphire plate (10 mm x 10 mm) at 2500 tr/min on a

Süss MicroTech Labspin 6.

Fluorescence intensity decays were obtained by the time-correlated single-photon counting

(

TCSPC) method with femtosecond laser excitation using a Spectra-Physics set-up composed of a Titanium

Sapphire Tsunami laser pumped by a doubled YAG laser Millennia. Light pulses at 990 nm (resp. 900 nm)

were selected by optoacoustic crystals at a repetition rate of 4 MHz, and the third harmonics at 330 nm

-

19 -

ACCEPTED MANUSCRIPT

(

resp. 300 nm) was obtained through non-linear crystals. Fluorescence photons were detected through a

monochromator by means of a Hamamatsu MCP R3809U photomultiplier. The time-to-amplitude converter

was purchased from Tennelec. The fluorescence data were analyzed by a nonlinear least-squares global

method using the Globals software package developed at the Laboratory for Fluorescence Dynamics at the

University of Illinois, Urbana-Champaign.

Electrochemical studies. Electrochemical experiments were performed under argon atmosphere using a Pt

disk electrode (diameter 1 mm), the counter electrode was a vitreous carbon rod and the reference

electrode was a silver wire in a 0.1M AgNO solution in CH CN. Ferrocene was added to the solution at the

3

+

end of a series of experiments. The ferrocene/ferrocenium (Fc/Fc ) couple served as internal standard. The

three electrodes cell was connected to a PAR Model 273 potentiostat/galvanostat (PAR, EG&G, USA)

monitored with the ECHEM Software. Activated Al O was added in the electrolytic solution to remove

2

3

excess moisture. For a further comparison of the electrochemical and optical properties, all potentials are

+

referred to the SCE electrode that was calibrated at –0.405 V vs. Fc/Fc system. Following the work of

48

Jenekhe, we estimated the electron affinity (EA) or lowest unoccupied molecular orbital (LUMO) and the

ionisation potential (IP) or highest occupied molecular orbital (HOMO) from the redox data (tangent to the

red

oxidation or reduction wave). The LUMO level was calculated from: LUMO (eV)= -[E

(vs SCE) +4.4] and

onset

ox

the HOMO level from: HOMO (eV) = -[E_o_n_s_e_t(vs SCE) + 4.4], based on an SCE energy level of 4.4 eV relative

to the vacuum. The electrochemical gap was calculated from: ΔE =|HOMO-LUMO| (in eV).

el

68 69

Theoretical modeling. Full geometry optimization with Density Functional Theory (DFT)

and Time-

Dependent Density Functional Theory (TD-DFT) calculations were performed with the hybrid Becke-3

7

0-72

73

parameter exchange

functional and the Lee-Yang-Parr non-local correlation functional (B3LYP)

74

implemented in the Gaussian 09 (Revision B.01) program suite using the 6-311G+(d,p) basis set and the

default convergence criterion implemented in the program. The figures were generated with GaussView

5

.0. The computed triplet adiabatic S to T excitation energies were calculated from the difference

0 1

between the total energy of the molecule in their respective singlet and triplet states.

Thermal analysis:

Thermal Gravimetric Analysis (TGA) was carried out by using TA SDT Q600 instrument, at the "Ecole

Nationale Supérieure de Chimie de Rennes" or at the "Institut des Sciences Analytiques" (UMR CNRS 5280)

of Villeurbanne. TGA curves were measured at 10°C/min from 0 to 600°C under an azote atmosphere.

Differential Scanning Calorimetry (DSC) was carried out by using NETZSCH DSC 200 F3 instrument equipped

with an intracooler. DSC traces were measured at 10°C/min. Two heating/cooling cycles were succesively

nd

carried out and the glass transition Tg was determined from the 2 heating cycle. Temperatures of the

different transitions given are obtained with the tangent to the transition peak.

Device fabrication and characterisation:

OLEDs based on a multilayer structure have been fabricated onto patterned ITO coated glass substrates

from XinYan Tech (thickness: 100 nm and sheet resistance: less of 20 W/m). The organic materials (from

-

6

Aldrich and Lumtec) are deposited onto the ITO anode by sublimation under high vacuum (< 10 Torr) at a

rate of 0.2 – 0.3 nm/s. The structure of the device is the following: ITO/CuPc(10 nm)/NPB (40 nm)/TCTA (10

nm)/Hosts:dopant (20 nm)/TPBi (40 nm)/LiF (1.2 nm)/Al (100 nm). In this device, ITO is used as the anode,

CuPc (copper phtalocyanine) is the hole injecting layer, NPB (N,N’-di(1-naphtyl)-N,N’-diphenyl-[1,1’-biphe-

nyl]-4,4’-diamine) is the hole-transporting layer, TCTA (4,4',4''-Tris(carbazol-9-yl)-triphenylamine) is the

electron/exciton blocking layer, TPBI (1,3,5-Tris(1-phenyl-1H-benzimidazol-2-yl)benzene) is both the

electron transporting layer and the hole blocking layer and a thin film of lithium fluoride over aluminum is

the cathode. The entire device is fabricated in the same run without breaking the vacuum. In this study, the

thicknesses of the different organic layers were kept constant for all the devices. The active area of the

devices defined by the overlap of the ITO anode and the metallic cathode was 0.3 cm². The

current-voltage-luminance (I-V-L) characteristics of the devices were measured with a regulated power

2

supply (Laboratory Power Supply EA-PS 3032-10B) combined with a multimeter and a 1 cm area silicon

calibrated photodiode (Hamamatsu). The spectral emission was recorded with a SpectraScan PR650

spectrophotometer. All the measurements were performed at room temperature and at ambient

atmosphere with no further encapsulation of devices.

-

20 -

ACCEPTED MANUSCRIPT

REFERENCES

(

1) Meng, Q.; Hu, W. Phys. Chem. Chem. Phys. 2012, 14, 14152–14164.

2) Müllen, K.; Scherf, U. Organic Light-Emitting Devices: Synthesis, Properties and Applications; Wiley-VCH Verlag GMbH & Co.

KGaA: Weinheim, 2006.

(

3) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913-915.

4) Mishra, A.; Bäuerle, P. Angew. Chem. Int. Ed. 2012, 51, 2020–2067.

5) Shaheen, S. E.; Ginley, D. S.; Jabbour, G. E. Mat. Res. Soc. Bull. 2005, 30, 10-19.

6) Forrest, S. R.; Thompson, M. E. Chem. Rev. 2007, 107, 923-925.

7) Grimsdale, A. C.; Müllen, K. Macromol. Rapid Commun. 2007, 28, 1676-1702.

8) Special issue: π-Functional Materials; Bredas, J.-L.; Marder, S. R.; Reichmanis, E. Chem. Mater. 2011, 23.

9) Shih, P.-I.; Chien, C.-H.; Chuang, C.-Y.; Shu, C.-F.; Yang, C.-H.; Chen, J.-H.; Chi, Y. J. Mater. Chem. 2007, 17, 1692-1698.

10) Yeh, S.-J.; Wu, M.-F.; Chen, C.-T.; Song, Y.-H.; Chi, Y.; Ho, M.-H.; Hsu, S.-F.; Chen, C. H. Adv. Mater. 2005, 17, 285-289.

11) Yook, K. S.; Lee, J. Y. L. Adv. Mater. 2012, 24, 3169–3190.

12) Xie, J.-H.; Zhou, Q.-L. Acc. Chem. Res. 2008, 41, 581-594.

13) Poriel, C.; Ferrand, Y.; Le Maux, P.; Rault-Berthelot, J.; Simonneaux, G. Synth. Met. 2008, 158, 796-801.

14) Poriel, C.; Ferrand, Y.; Le Maux, P.; Rault-Berthelot, J.; Simonneaux, G. Chem. Commun. 2003, 1104-1105.

15) Poriel, C.; Ferrand, Y.; Le Maux, P.; Rault-Berthelot, J.; Simonneaux, G. Inorg. Chem. 2004, 43, 5086-5095.

16) Ferrand, Y.; Poriel, C.; Le Maux, P.; Rault-Berthelot, J.; Simonneaux, G. Tet. Asym. 2005, 16, 1463-1472.

17) Poriel, C.; Ferrand, Y.; Le Maux, P.; Paul-Roth, C.; Simonneaux, G.; Rault-Berthelot, J. J. Electroanal. Chem. 2005, 583, 92-103.

18) Schneider, D.; Rabe, T.; Riedl, T.; Dobbertin, T.; Kröger, M.; Becker, E.; H.-H., J.; Kowalski, W.; Weimann, T.; Wang, J.; Hinze, P.;

Gerhard, A.; Stössel, P.; Vestweber, H. Adv. Mater. 2005, 17, 31-34.

(

19) Kim, Y. S.; Lee, M.; Boo, B. H. J. Chem. Phys. 1998, 109, 2593.

20) Moreau, F.; Audebrand, N.; Poriel, C.; Moizan-Baslé, V.; Ouvry, J. J. Mater. Chem. 2011, 21, 18715-18722.

21) Saragi, T. P. I.; Spehr, T.; Siebert, A.; Fuhrmann-Lieker, T.; Salbeck, J. Chem. Rev. 2007, 107, 1011-1065.

22) Romain, M.; Tondelier, D.; Vanel, J.-C.; Geffroy, B.; Jeannin, O.; Rault-Berthelot, J.; Métivier, R.; Poriel, C. Angew. Chem. Int. Ed.

2

013, 52, 14147–14151.

(

23) Thirion, D.; Romain, M.; Rault-Berthelot, J.; Poriel, C. J. Mater. Chem. 2012, 22, 7149-7157.

24) Tao, Y.; Ao, L.; Wang, Q.; Zhong, C.; Yang, C.; Qin, J.; Ma, D. Chem. Asian J. 2010, 5, 278-284.

25) Jeon, S. O.; Yook, K. S.; Lee, J. Y. Thin Solid Films 2010, 519, 890-893.

26) Jahng, Y.; Rahman, A. F. M. M. Bull. Chem. Soc. Jpn. 2010, 83, 672-677.

27) Ying, L.; Zou, J.; Yang, W.; Zhang, A.; Wu, Z.; Zhao, W.; Cao, Y. Dyes and Pigments 2009, 82, 251-257.

28) Wu, P.-T.; Kim, F. S.; Champion, R. D.; Jenekhe, S. A. Macromol. 2008, 41, 7021 - 7028.

29) Hughes, G.; Bryce, M. R. J. Mater. Chem. 2005, 15, 94-107.

30) Hughes, G.; Wang, C.; Batsanov, A. S.; Fern, M.; Frank, S.; Bryce, M. R.; Perepichka, I. F.; Monkman, A. P.; Lyons, B. P. Org.

Biomol. Chem. 2003, 1, 3069-3077.

(

31) Montilla, F.; Mallavia, R. Adv. Funct. Mater. 2007, 17, 71-78.

32) Poriel, C.; Rault-Berthelot, J.; Thirion, D.; Barrière, F.; Vignau, L. Chem. -Eur. J. 2011, 17, 14031-14046.

33) Grimsdale, A. C. Curr. Org. Chem. 2010, 14, 2196-2217.

34) Farrar, S. R.; Contoret, A. E. A.; O'Neill, M.; Nicholls, J. E.; Eastwood, A. J.; Richards, G. J.; Vlachos, P.; Kelly, S. M. Appl. Surf. Sc.

2

(

8

002, 186, 435-440.

35) Wu, C.-C.; Lin, Y.-T.; Chiang, H.-H.; Cho, T. Y.; Chen, C. W.; Wong, K.-T.; Liao, Y.-L.; Lee, G. H.; Peng, S. M. App. Phys. Lett. 2002,

1, 577-579.

(

36) Lee, R.-H.; Huang, Y.-W.; Wang, Y.-Y.; Chang, H.-Y. Thin Solid Films 2008, 516, 5062-5068.

37) Jiang, Z.; Yao, H.; Zhang, Z.; Yang, C.; Liu, Z.; Tao, Y.; Qin, J.; Ma, D. Org. Lett. 2009, 11, 2607–2610.

38) Fan, C.; Chen, Y.; Gan, P.; Yang, C.; Zhong, C.; Qin, J.; Ma, D. Org. Lett. 2010, 12, 5648–5651.

39) Jang, S. E.; Joo, C. W.; Ok, J. S.; Soo, Y. K.; Lee, J. Y. Org. Elec. 2010, 11, 1059-1065.

40) Lee, K. H.; Kim, S. O.; Yook, K. S.; Jeon, S. O.; Lee, J. Y.; Yoon, S. S. Synt. Met. 2011, 161, 2024-2030.

41) Dong, S.-C.; Gao, C.-H.; Zhang, Z.-H.; Jiang, Z.-Q.; Lee, S.-T.; Liao, L.-S. Phys. Chem. Chem. Phys. 2012, 14, 14224-14228.

42) Dong, S.-C.; Gao, C.-H.; Yuan, X.-D.; Cui, L.-S.; Jiang, Z.-Q.; Lee, S.-T.; Liao, L.-S. Org. Elec. 2013, 14, 902-908.

43) Poriel, C.; Métivier, R.; Rault-Berthelot, J.; Thirion, D.; Barrière, F.; Jeannin, O. Chem. Commun. 2011, 47, 11703-11705.

44) Thiery, S.; Tondelier, D.; Declairieux, C.; Seo, G.; Geffroy, B.; Jeannin, O.; Rault-Berthelot, J.; Métivier, R.; Poriel, C. J. Mater.

Chem. C 2014, 2014, 2, 4156-4166

(

45) Pei, J.; Ni, J.; Zhou, X.-H.; Cao, X.-Y.; Lai, Y.-H. J. Org. Chem. 2002, 67, 4924-4936.

46) Demers, E.; Maris, T.; Wiest, J. D. W. Cryst. Growth Des. 2005, 5, 1227-1235.

47) Wong, K.-T.; Liao, Y.-L.; Peng, Y.-C.; Wang, C.-C.; Lin, S.-Y.; Yang, C.-H.; Tseng, S.-M.; Lee, G.-H.; Peng, S.-M. Cryst. Growth Des.

2

005, 5, 667–671.

(

48) Kulkarni, A. P.; Tonzola, C. J.; Babel, A.; Jenekhe, S. A. Chem. Mater. 2004, 16, 4556–4573.

49) Poriel, C.; Liang, J.-J.; Rault-Berthelot, J.; Barrière, F.; Cocherel, N.; Slawin, A. M. Z.; Horhant, D.; Virboul, M.; Alcaraz, G.;

Audebrand, N.; Vignau, L.; Huby, N.; Wantz, G.; Hirsch, L. Chem. -Eur. J. 2007, 13, 10055-10069.

50) Aizawa, N.; Pu, Y.-J.; Sasabe, H.; Kido, J. Org. Electron. 2012, 13, 2235-2242.

(

-

21 -

ACCEPTED MANUSCRIPT

(

51) Tsai, M.-H.; Hong, Y.-H.; Chang, C.-H.; Su, H.-C.; Wu, C.-C.; Matoliukstyte, A.; Simokaitiene, J.; Grigalevicius, S.; Grazulevicius, J.

V.; Hsu, C.-P. Adv. Mater. 2007, 19, 862-866.

(

52) Belletête, M.; Ranger, M.; Beaupré, S.; Leclerc, M.; Durocher, G. Chem. Phys. Lett. 2000, 316, 101-107.

53) Wang, J.-F.; Feng, J.-K.; Ren, A.-M.; Yang, L. Chin. J. Chem. 2005, 23, 1618-1624.

54) Thirion, D.; Poriel, C.; Métivier, R.; Rault-Berthelot, J.; Barrière, F.; Jeannin, O. Chem. -Eur. J. 2011, 17, 10272-10287.

55) Thirion, D.; Poriel, C.; Barrière, F.; Métivier, R.; Jeannin, O.; Rault-Berthelot, J. Org. Lett. 2009, 11, 4794-4797.

56) Ooyama, Y.; Ito, G.; Kushimoto, K.; Komaguchi, K.; Imae, I.; Harima, Y. Org. Biomol. Chem. 2010, 8, 2756-2770.

57) Chi, C.-C.; Chiang, C.-L.; Liu, S.-W.; Yueh, H.; Chen, C.-T.; Chen, C.-T. J. Mater. Chem. 2009, 19, 5561-5571.

58) Adachi, C.; Kwong, R. C.; Djurovich, P.; Adamovich, V.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79,

2

082.

(

59) Wong, K.-T.; Chen, Y.-M.; Lin, Y.-T.; Su, H.-C.; Wu, C.-C. Org.Lett. 2005, 7, 5361-5364.

60) Seo, J. H.; Han, N. S.; Shim, H. S.; Kwon, J. H.; Song, J. K. Bull. Korean Chem. Soc. 2011, 32, 1415.

61) Chi, L.-C.; Hung, W.-Y.; Chiu, H.-C.; Wong, K.-T. Chem. Commun. 2009, 3892-3894.

62) Han, N. S.; Sohn, S. H.; Park, S. M.; Song, J. K. Bull. Korean Chem. Soc. 2013, 34, 1547-1550.

63) Winter-Werner, B.; Diederich, F.; Gramlich, V. Helv. Chim. Acta 1996, 79, 1338–1360.

64) Jeon, S. O.; Yook, K. S.; Joo, C. W.; Son, H. S.; Lee, J. Y. Thin Solid Films 2010, 518, 3716-3720.

65) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, C.; Giacovazzo, A.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R.

J. Appl. Cryst. 1999, 32, 115-119.

(

66) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112-122.

67) Farrugia, L. J. J. Appl. Cryst. 2012, 45, 849-854.

68) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864-B871.

69) Calais, J.-L. Int. J. Quantum Chem. 1993, 47, 101.

70) Becke, A. D. Phys. Rev. A 1988, 38, 3098-3100.

71) Becke, A. D. J. Chem. Phys. 1993, 98, 5648 –5652.

72) Becke, A. D. J. Chem. Phys. 1993, 98, 1372-1377.

73) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.

74) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,

B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada,

M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;

Montgomery, J. A. J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.;

Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.;

Klene, M.; Knox, J. E.; 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.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;

Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J., Gaussian 09, Revision B.01, Gaussian,

Inc., Wallingford CT, 2010.

-

22 -

ACCEPTED MANUSCRIPT

Supplementary Materials for

2-substituted vs 4-substituted-9,9'-spirobifluorene host materials

for green and blue phosphorescent OLEDs: A Structure-Property Relationship

Study

a

c

Sébastien Thiery, Céline Declairieux, Denis Tondelier, Gijun Seo, Bernard

*a

c,d

a

b

*a

Geffroy, Olivier Jeannin, Rémi Métivier, Joëlle Rault-Berthelot, Cyril Poriel

a. Université de Rennes 1-UMR CNRS 6226, "Institut des Sciences Chimiques de Rennes"

MaCSE group. Bat 10C, Campus de Beaulieu, 35042 Rennes, France

b. PPSM, Institut d'Alembert, ENS Cachan, UMR CNRS 8531, 61 Avenue du Président

Wilson, 94235 Cachan, France

c. LPICM, Ecole Polytechnique, UMR CNRS 7647, Route de Saclay, 91128, Palaiseau

d. LICSEN, CEA Saclay, IRAMIS/NIMBE, 91191 Gif sur Yvette

ꢀ

Summary

X-RAY DIFFRACTION............................................................................................................ 2

PHOTOPHYSICAL PROPERTIES .......................................................................................... 8

THEORETICAL MODELING................................................................................................ 16

THERMAL PROPERTIES...................................................................................................... 22

ELECTROCHEMICAL PROPERTIES .................................................................................. 25

2D NMR STUDIES ................................................................................................................. 29

COPY OF NMR SPECTRA .................................................................................................... 47

COPY OF MASS SPECTRA................................................................................................... 59

1

ACCEPTED MANUSCRIPT

X-RAY DIFFRACTION

Table 1 Crystal data and structure refinement for 2-Ph-SBF

Identification code

2-Ph-SBF

Empirical formula

Formula weight

Temperature

62 40

C H

784.94

150(2) K

Wavelength

0.71073 Å

Crystal system

Space group

Triclinic

P-1

Unit cell dimensions

a = 10.9027(4) Å = 102.1400(10)°.

b = 13.7902(6) Å = 97.4710(10)°.

c = 15.7771(5) Å  = 111.5900(10)°.

Volume

2099.47(14) Å³

Z

2

Density (calculated)

Absorption coefficient

F(000)

1.242 g/cm³

0.070 mm^-¹

824

Crystal size

0.26 x 0.24 x 0.09 mm³

Theta range for data collection 1.36 to 27.50°.

Index ranges

-14<=h<=13, -17<=k<=17, -20<=l<=20

Reflections collected

Independent reflections

20625

9543 [R(int) = 0.0307]

Completeness to theta = 27.50° 98.9 %

Absorption correction

Semi-empirical from equivalents

Max. and min. transmission

Refinement method

0.994 and 0.982

Full-matrix least-squares on F²

9543 / 0 / 559

Data / restraints / parameters

Goodness-of-fit on F²

1.074

Final R indices [I>2sigma(I)]

R indices (all data)

R1 = 0.0447, wR2 = 0.1128

R1 = 0.0583, wR2 = 0.1272

0.261 and -0.269 e.Å^-³

Largest diff. peak and hole

2

ACCEPTED MANUSCRIPT

Table 2 Crystal data and structure refinement for 2-5Pm-SBF

Identification code

Empirical formula

Formula weight

Temperature

2-5Pm-SBF

30 22 2

C H N O

426.5

150(2) K

Wavelength

0.71073 Å

orthorhombic

P c a b

Crystal system

space group

Unit cell dimensions

a = 8.4342(3) Å, α = 90 °

b = 13.6588(5) Å, β = 90 °

c = 38.5506(13) Å, γ = 90 °

3

Volume

Z

4441.1(3) Å

8

-

3

Calculated density

Absorption coefficient

F(000)

1.276 (g.cm )

-

1

0.078 mm

1792

Crystal size

0.36 x 0.31 x 0.05 mm

1.06 to 27.46 °

-10, 9

Theta range for data collection

h_min, h_max

k_min, k_max

-17, 16

l_min, l_max

-45, 50

a

Reflections collected / unique

Reflections [I>2σ]

Completeness to theta_max

Absorption correction type

Refinement method

Data / restraints / parameters

Goodness-of-fit

21000 / 5055 [R(int) = 0.0327]

3820

0.994

multi-scan

2

Full-matrix least-squares on F

5055 / 0 / 300

1.124

c

d

Final R indices [I>2σ]

R indices (all data)

Largest diff. peak and hole

R1 = 0.0461, wR2 = 0.1246

c

d

R1 = 0.0711, wR2 = 0.1511

-

-3

0.36 and -0.315 e .Å

3

ACCEPTED MANUSCRIPT

2-Ph-SBF

2

-Ph-SBF MOLECULE 1

S 1 Top: Angle between the mean plane of the pendant phenyl ring and that of its attached phenyl

ring of the fluorene : 37.37° ; Bottom: Angle between the two mean planes of each fluorene: 7.08°

4

ACCEPTED MANUSCRIPT

2

-Ph-SBF MOLECULE 2

S 2 Angle between the mean plane of the pendant phenyl ring and that of its attached phenyl ring of

the fluorene: 4.58°

5

ACCEPTED MANUSCRIPT

S 3 Short contacts (CH and C-C, distance < (sum of r_v_d_w-0.1) detected in crystal packing of 2-Ph-

SBF.

S 4 Short contacts (CH and C-C, distance < (sum of r_v_d_w) detected for the 2-Ph-SBF molecule 2 (in

red above)

6

ACCEPTED MANUSCRIPT

S 5 Short contacts (CH and C-C, distance < (sum of r_v_d_w) detected for the pendant phenyl unit of 2-

Ph-SBFmolecule 2 (in red above)

S 6 Short contacts (CH and C-C, distance < (sum of r_v_d_w) detected for the pendant phenyl unit of 2-

Ph-SBF molecule 1 (in blue above)

7

Article Doi

2-Substituted vs 4-substituted-9,9'-spirobifluorene host materials for green and blue phosphorescent OLEDs: A structure-property relationship study

DOI: 10.1016/j.tet.2014.05.081

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.tet.2014.05.081

Full text of DOI:10.1016/j.tet.2014.05.081

Products guided by the article

R&D Labs maybe for 1610503-97-3

Relevant to this article

Hot Product