Violet-blue emitting 2-(N-alkylimino)pyrrolyl organoboranes: Synthesis, structure and luminescent properties

Article abstract of DOI:10.1016/j.dyepig.2016.11.017

The condensation reactions of 2-formylpyrrole (1) or 2-formylphenanthro[9,10-c]pyrrole (2) with various aliphatic amines afforded the corresponding 2-iminopyrrole ligand precursors 3–10, which, upon stoichiometric reaction with BPh₃, led to the

Full text of DOI:10.1016/j.dyepig.2016.11.017

Accepted Manuscript
Violet-blue emitting 2-(N-alkylimino)pyrrolyl organoboranes: Synthesis, structure and
luminescent properties
Paramasivam Krishnamoorthy, Bruno Ferreira, Clara S.B. Gomes, Diogo Vila-Viçosa,
Ana Charas, Jorge Morgado, Maria José Calhorda, António L. Maçanita, Pedro T.
Gomes
PII:
S0143-7208(16)30846-4
10.1016/j.dyepig.2016.11.017
DYPI 5586
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 27 September 2016
Revised Date: 10 November 2016
Accepted Date: 14 November 2016
Please cite this article as: Krishnamoorthy P, Ferreira B, Gomes CSB, Vila-Viçosa D, Charas A,
Morgado J, Calhorda MJ, Maçanita AL, Gomes PT, Violet-blue emitting 2-(N-alkylimino)pyrrolyl
organoboranes: Synthesis, structure and luminescent properties, Dyes and Pigments (2016), doi:
10.1016/j.dyepig.2016.11.017.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract sentence:
Violet-blue emitting 2-(N-Alkylimino)pyrrolyl organoboranes are synthesized, their structure
and luminescent properties being described
Graphical Abstract figure:

ACCEPTED MANUSCRIPT
Violet-Blue emitting 2-(N-Alkylimino)pyrrolyl Organoboranes:
Synthesis, Structure and Luminescent Properties
Paramasivam Krishnamoorthya, Bruno Ferreira a, Clara S.B. Gomes a, Diogo Vila-Viçosa d,
Ana Charas b, Jorge Morgado b,c, Maria José Calhorda d, António L. Maçanita a,
Pedro T. Gomesa,*
a Centro de Química Estrutural, Departamento de Engenharia Química, Instituto Superior
Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal.
b Instituto de Telecomunicações, Av. Rovisco Pais, 1049-001 Lisboa, Portugal.
c Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Av.
Rovisco Pais, 1049-001 Lisboa, Portugal.
d Centro de Química e Bioquímica, DQB, Faculdade de Ciências, Universidade de Lisboa,
Campo Grande, Ed. C8, 1749-016 Lisboa, Portugal.
* Corresponding author: Pedro T. Gomes; E-mail: pedro.t.gomes@tecnico.ulisboa.pt
1

ACCEPTED MANUSCRIPT
Abstract
The condensation reactions of 2-formylpyrrole (1) or 2-formylphenanthro[9,10-c]pyrrole (2)
with various aliphatic amines afforded the corresponding 2-iminopyrrole ligand precursors 3‒
10, which, upon stoichiometric reaction with BPh3, led to the new mononuclear boron chelate
compounds Ph2B[NC4H3C(H)=N-R] (R = Me 11; iPr 12; tBu 13; nOct 14; Cy 15; Adam 16),
and Ph2B(NC16H9C(H)=N-R) (R = Me 17; Adam 13), respectively. Boron complexes 11‒16,
containing a simple 2-(N-alkylformimino)pyrrolyl ligand, are violet emitters and showed
relatively modest fluorescence quantum efficiencies in solution (10% ‒ 16%), whereas
complexes 17 and 18, bearing the π-extended 2-(N-alkylformimino)phenanthro[9,10-
c]pyrrolyl ligand, are blue emitters presenting enhanced quantum efficiencies of 35% and
43%, respectively, in THF solution. DFT and TDDFT calculations were in good agreement
with experimental results, showing that π systems (pyrrolyl and phenanthropyrrolyl in this
case) have a strong influence on the observed optical properties by changing the nature of the
low energy transitions. Non-doped single-layer light-emitting diodes (OLEDs) were
fabricated with complexes 11‒18, deposited essentially by spin coating, those of complexes
17 and 18 revealing maximum luminances of 69 and 88 cd m-2, respectively.
Keywords
Iminopyrrolyl ligands; Boron; Fluorescence; DFT; OLED
2

ACCEPTED MANUSCRIPT
Graphical Abstract
Violet-blue emitting 2-(N-Alkylimino)pyrrolyl organoboranes are synthesized, their structure
and luminescent properties being described.
3

ACCEPTED MANUSCRIPT
Highlights
•
•
•
Several 2-(N-alkylimino)pyrrolyl diphenyl boron complexes were synthesized
Fluorescence properties of these blue-violet emitting compounds were studied
Non-radiative decays are much weaker than those of the N-2,6-R2-arylimino
analogues
•
•
DFT and TDDFT calculations support the experimental results
Simple OLED devices reveal maximum luminances of ca. 90 cd m-2
4

ACCEPTED MANUSCRIPT
1. Introduction
Organic Light Emitting Diode (OLED) based flat panel displays for cell phones, digital
cameras, and TV sets are already commercially available. Despite this success, there is a
strong claim for a next generation of flexible, highly luminous and economically improved
red, green, blue and white OLEDs for displays and lighting applications, the two latter colors
being highly requested. Because of these needs, research involving luminescent organic and
organometallic complexes has received considerable attention [1].
Among them, the four-coordinate organoboron compounds are promising light emitting
materials owing to their intense luminescence and high carrier mobility. Various types of
tetracoordinate boron compounds containing different types of ligands, such as N,O-, N,N-,
N,C-, C,C-, C,O-, and O,O-chelates, have been analyzed, their photophysical properties being
strongly dependent upon the nature of the chelate ligand [2].
The 2-iminopyrrolyl ligands are an important class of anionic bidentate (N,N‒) chelates,
containing a pyrrolyl anionic ring and a neutral imine as donor moieties [3]. The first
homoleptic metal complexes of Co(II), Ni(II), Pd(II), Cu(II), and Zn(II) containing these
ligands were reported in the 1960s [4]. There has been a resurgence of interest in such systems
for various applications such as in catalytic organic transformations [3]. Another important
focus of these ligands has been the synthesis of coordination compounds with
photoluminescent and/or electroluminescent properties [5‒10].
In the last few years, our group has been involved in the synthesis of a variety of metal
complexes using 2-(N-arylimino)pyrrolyl derived ligands. In particular, we reported some
luminescent zinc complexes containing 2-(N-arylformimino)phenanthro[9,10-c]pyrrolyl
ligand [7], in which the π-conjugation was extended by fusing the phenanthrene ring on the
pyrrolyl C3-C4 bond. We also reported the luminescent properties of several new
5

ACCEPTED MANUSCRIPT
tetracoordinate mononuclear organoboron complexes containing 2-(N-arylformimino)pyrrolyl
ligands, having varied the electronic and steric nature of the N-aryl group (Chart 1, A) [8]. It
was found that the color of emission could be tuned from blue to bluish-green by increasing
the substituent’s electron-donating power. We extended our work to the synthesis of
polynuclear boron complexes [9], where the iminopyrrolyl ligands have different aromatic
bridging spacers (Chart 1, B), varying thereby the π-conjugation length and thus the color
tuning range from blue to yellow. We further used a third strategy for the color tuning by
extending the π-conjugation through the use of ligands containing fused aromatic fragments
onto the 2-iminopyrrolyl C4-C5 or C3-C4 bonds; in fact, the employment of 2-(N-
arylformimino)indolyl or the above mentioned 2-(N-arylformimino)phenanthro[9,10-
c]pyrrolyl ligands in the coordination to boron (Chart 1, C and D) provided colors in the
range blue to orange [10].
Chart 1.
6

ACCEPTED MANUSCRIPT
Among the family of reported mononuclear [BPh2(κ2N,N’-NC4H3C(H)=N-Ar)], the
complexes with bulky aryl groups, such as Ar = 2,6-Me2C6H3 or 2,6-iPr2C6H3 (Chart 1, A; R1
= Me or iPr; R2 = R3 = H) gave rise to violet emission [8c], In these cases, the hindered
rotation of the N-aryl fragment, which is unable to reach coplanarity with the 2-iminopyrrolyl
fragment, reduces the π-extension of the chromophore, thus blue-shifting the emission
spectra. However, the achievement of such a demanded violet emission occurs with very low
fluorescence quantum efficiencies (0.5% and 2.3%, respectively, in solution), because of the
extensive non-radiative quenching of the emission (knr = 27.6 and 7.52 ns-1, respectively)
operating in these molecules. This efficient non-radiative decay is essentially due to the
internal conversion (kic ≈ knr) [8c].
In the present work, we describe the synthesis of a family of violet-blue emitting 2-
iminopyrrolyl boron diphenyl complexes, in which significant non-radiative decay is avoided
by replacing the 2,6-disubstituted N-aryl fragments by N-alkyl groups. In both the 2-
iminopyrrolyl and 2-iminophenanthropyrrolyl boron chromophore frameworks, different N-
alkyl substituents with several stereochemical features were used, in order to vary the solid
state packing. The new mononuclear organoboron compounds of the types [BPh2(κ2N,N’-
NC4H3C(H)=N-Alkyl)] and [BPh2(κ2N,N’-NC16H9C(H)=N-Alkyl)], containing 2-(N-aliphatic
imino)pyrrolyl chelating ligands, were synthesized and characterized by multinuclear NMR,
single crystal X-ray diffraction, when possible, and cyclic voltammetry. Their photophysical
characterizations were performed using steady state photoluminescence (in solution and in the
solid state) and time-resolved fluorescence (in solution). Density-functional theory (DFT) and
time-dependent DFT (TDDFT) calculations were also carried out for these new boron
complexes to determine the geometry of the ground and first excited singlet state, and to
assign the nature of electronic transitions.
7

ACCEPTED MANUSCRIPT
2. Results and discussion
2.1. Synthesis and characterization of new 2-(N-aliphatic formimino)pyrrolyl boron
complexes 11-18
The 2-(N-aliphatic imino)pyrrole ligand precursors 3‒5, 7 and 8 (Scheme 1) used in this
study were synthesized and characterized according to the literature protocols [11]. The new
ligand precursors 6, 9 and 10 were also synthesized, their characterization data being
presented in the experimental section.
In general, the ligand precursors HNC4H3C(H)=N-CH3 (3), HNC4H3C(H)=N-CH(CH3)2
(4), HNC4H3C(H)=N-C(CH3)3 (5), HNC4H3C(H)=N-CH2(CH2)6CH3 (6), HNC4H3C(H)=N-
C6H11 (7), HNC4H3C(H)=N-C10H15 (8), HNC16H9C(H)=N-CH3 (9) and HNC16H9C(H)=N-
C10H15 (10) were synthesized by condensation of the appropriate formyl precursors
HNC4H3C(H)=O (1) or HNC16H9C(H)=O (2) with the respective aliphatic amines under
reflux. The 2-formylphenanthro[9,10-c]pyrrole 2 was prepared by a multistep procedure,
starting from phenanthrene [12].
All compounds are solids, although some with low melting points, and were characterized
1
13
by H and C NMR spectroscopies, their spectra being consistent with those reported in the
literature.
The reactions of the 2-(N-aliphatic imino)pyrrole chelate precursors 3‒10 with
triphenylboron (BPh3) in hot toluene, overnight, under a nitrogen atmosphere, afforded the
corresponding organoboron bidentate complexes 11‒18, which are depicted in Scheme 1.
8

ACCEPTED MANUSCRIPT
Scheme 1. Syntheses of the 2-(N-aliphatic imino)pyrrole ligand precursors 3‒10 and the new
corresponding 2-(N-aliphatic formimino)pyrrolyl diphenylboron complexes 11‒18.
The new organoboron compounds 11‒18 are moderately air stable, and were fully
1
13
11
characterized by H, C and B NMR spectroscopies (see Experimental section and Figs.
S1‒S30 in Supporting Information). The absence of the typical broad NH resonance in the 1H
NMR spectra of the complexes confirms the anionic coordination of the pyrrolyl-N to the
boron center. The imine proton (HC=N) resonances of the complexes appear as singlets in the
range δ 8.04 to 8.64, downfield-shifted in relation to the respective ligand precursor,
evidencing the neutral imine coordination to the boron atom. Furthermore, the 11B NMR
resonance of these compounds is in the range of δ 3.61 to 7.17, confirming the formation of
tetracoordinate boron compounds, being also consistent with our earlier reports [8a,c,9].
Moreover, the elemental compositions of all the complexes were determined and the results
confirmed the formation of the desired products (see Experimental section). Finally, the
9

ACCEPTED MANUSCRIPT
molecular structures of the boron complexes 14, 15 and 18 were determined by single-crystal
X-ray diffraction studies.
2.2. X-ray diffraction studies
Perspective views of the molecular structures of 14, 15 and 18 are shown in Fig. 2–4. Selected
bond lengths and bond angles are given as captions in the corresponding figures. Crystals
suitable for single crystal X-ray diffraction studies were obtained by cooling ethereal
solutions of 14, 15 and 18 double-layered with n-hexane, at -20 °C, for 2 days. Boron
complexes 14 and 15 crystallized with two molecules in the unit cell, in triclinic and
orthorhombic crystal systems, with P-1 and Pna21 space groups, respectively. Compound 16
also crystallized in the triclinic system within P-1 space group, showing, however, an
asymmetric unit composed by one boron complex and one Et2O molecule. In the three
molecular structures, the boron centers are tetracoordinate adopting typical distorted
tetrahedral geometries. The N-aliphatic 2-iminopyrrolyl ligands chelate each boron center via
the N1, N2 atoms to form a virtually planar five membered chelate ring in all the compounds,
the remaining coordination sites being occupied by the quaternary (ipso) carbon atoms of two
phenyl groups. The latter groups are located above and below the planes of the chelating
ligand and the boron center. In addition, the B-CPh distance in all complexes was found to be
in the range of 1.595(10)‒1.629(6) Å. The bond distances between the boron center and the
two chemically different nitrogen atoms, i.e., B1-N1pyrrolyl of 1.565(9) and 1.571(9) (14),
1.561(6) and 1.566(6) (15) and 1.580(4) (18) and B1-N2imine of 1.614(8) and 1.619(9) (14),
1.605(6) and 1.615(5) (15), and 1.615(4) (18) Å, respectively, are quite comparable to the
previously reported organoboron compounds [8a,c,9]. The chelating N,N' ligands exhibit bite
angles N–B–N of 95.6(4)° and 94.8(5)º (14), 95.6(3)° and 94.9(3)º (15), and 95.4(2)º (18),
10

ACCEPTED MANUSCRIPT
Fig. 2. Perspective view of the molecular structure of complex 14. The ellipsoids were drawn
at 50% probability level. All the calculated hydrogen atoms and the second molecule of the
asymmetric unit were omitted for clarity. Molecule A – Selected Bond lengths (Å): B1A-
N1A, 1.565(9); B1A-N2A, 1.614(8); B1A-C15A, 1.595(10); B1A-C21A, 1.607(9); C5A-
N1A, 1.346(8); C2A-N1A, 1.382(8); C2A-C6A, 1.405(9); N2A-C6A, 1.302(8); N2A-C7A,
1.458(8). Selected Bond angles (°): N1A-B1A-N2A, 95.6(4); N1A-B1A-C21A, 111.0(5);
N1A-B1A-C15A, 111.1(5); N2A-B1A-C15A, 107.6(5); B1A-N1A-C2A, 112.0(5); B1A-
N1A-C5A, 139.9(5); B1A-N2A-C6A, 110.7(5); B1A-N2A-C7A, 123.8(5); C7A-N2A-C6A,
125.4(5); C15A-B1A-C21A, 117.7(5). Molecule B – Selected Bond lengths (Å): B1B-N1B,
1.571(9); B1B-N2B, 1.619(9); B1B-C15B, 1.615(9); B1B-C21B, 1.602(9); C5B-N1B,
1.346(8); C2B-N1B, 1.371(8); C2B-C6B, 1.418(9); N2B-C6B, 1.318(9); N2B-C7B, 1.472(8).
Selected Bond angles (°): N1B-B1B-N2B, 94.8(5); N1B-B1B-C21B, 111.8(5); N1B-B1B-
C15B, 112.9(5); N2B-B1B-C15B, 107.5(5); B1B-N1B-C2B, 112.9(5); B1B-N1B-C5B,
139.6(6); B1B-N2B-C6B, 111.5(5); B1B-N2B-C7B, 122.5(5); C7B-N2B-C6B, 125.9(5);
C15B-B1B-C21B, 115.8(5).
11

ACCEPTED MANUSCRIPT
Fig. 3. Perspective view of the molecular structure of complex 15. The ellipsoids were drawn
at 50% probability level. All the calculated hydrogen atoms and the second molecule of the
asymmetric unit were omitted for clarity. Molecule A – Selected Bond lengths (Å): B1A-
N1A, 1.561(6); B1A-N2A, 1.605(6); B1A-C19A, 1.620(6); B1A-C13A, 1.605(6); N1A-C5A,
1.342(5); N1A-C2A, 1.378(5); N2A-C6A, 1.301(5); N2A-C7A, 1.476(5); C2A-C6A,
1.409(6). Selected Bond angles (°): N1A-B1A-N2A, 95.6(3); N1A-B1A-C13A, 110.8(3);
N1A-B1A-C19A, 112.7(4); N2A-B1A-C13A, 112.7(4); N2A-B1A-C19A, 110.4(3); C5A-
N1A-C2A, 107.8(3); C5A-N1A-B1A, 140.3(4); C6A-N2A-B1A, 111.5(3); C6A-N2A-C7A,
122.4(3); C13A-B1A-C19A, 113.4(3); B1A-N2A-C7A, 125.9(3). Molecule B – Selected
Bond lengths (Å): B1B-N1B, 1.566(6); B1B-N2B, 1.615(5); B1B-C19B, 1.620(6); B1B-
C13B, 1.629(6); N1B-C5B, 1.369(5); N1B-C2B, 1.347(5); N2B-C6B, 1.303(5); N2B-C7B,
1.471(5); C5B-C6B, 1.412(6). Selected Bond angles (°): N1B-B1B-N2B, 94.9(3); N1B-B1B-
C13B, 111.7(4); N1B-B1B-C19B, 111.1(3); N2B-B1B-C13B, 108.5(3); N2B-B1B-C19B,
111.9(3); C5B-N1B-C2B, 107.4(3); C5B-N1B-B1B, 112.3(3); C6B-N2B-B1B, 111.6(3);
C6B-N2B-C7B, 122.8(3); C13B-B1B-C19B, 116.7(3); B1B-N2B-C7B, 125.6(3).
12

ACCEPTED MANUSCRIPT
Fig. 4. Perspective view of the molecular structure of complex 18. The ellipsoids were drawn
at 50% probability level. All the calculated hydrogen atoms and one Et2O molecule were
omitted for clarity. Selected Bond lengths (Å): B1-N1, 1.580(4); B1-N2, 1.615(4); B1-C29,
1.617(4); B1-C35, 1.613(4); N1-C5, 1.338(3); N1-C2, 1.384(3); N2-C6, 1.310(3); N2-C19,
1.503(3); C2-C6, 1.406(4). Selected Bond angles (°): N1-B1-N2, 95.4(2); N1-B1-C29,
110.7(2); N1-B1-C35, 107.2(2); N2-B1-C29, 109.8(2); N2-B1-C35, 114.0(2); C5-N1-C2,
108.8(2); C5-N1-B1, 139.1(2); C6-N2-B1, 110.6(2); C6-N2-C19, 119.7(2); C29-B1-C35,
117.5(2); B1-N2-C19, 129.2(2).
which are much smaller than the ideal bond angle (109°) for a regular tetrahedral geometry,
strongly supporting the observed distortion.
The supramolecular packing of complex 14, 2-(N-octylimino)pyrrolyl diphenylboron,
shows, along the a axis, a 3D multilayer structure (see Fig. S31 in Supporting Information),
formed by the aligned nonpolar aliphatic n-octyl chains in a bilayer type tail-to-tail
arrangement. The layers are held together by C-HꢀꢀꢀπC(Ar) weak hydrogen bonds and π-π
stacking of phenyl rings (see Fig. S32 and Table S2 in Supporting Information). Compound
15, 2-(N-cyclohexylimino)pyrrolyl diphenylboron, displays weak C‒HꢀꢀꢀπC(Ar) short contacts
13

ACCEPTED MANUSCRIPT
(see Fig. S33 and Table S2 in Supporting Information) with no special orientation or motifs
being observed. The crystal packing of 18, 2-(N-adamantylimino)phenanthro[9,10-c]pyrrolyl
diphenylboron, shows head-to-tail dimers formed by C-HꢀꢀꢀπC(Ar) short contacts and 2D-sheets
containing the phenanthro[9,10-c]pyrrolyl-boron moieties (see Fig. S34 and Table S2 in
Supporting Information), which also exhibit π-π stacking with neighboring dimers through the
phenanthrenic rings of the ligands (πC(Ar)⋯πC(Ar), ca. 3.51 Å).
2.3. Photoluminescence studies
The UV-Vis absorption and fluorescence spectra of complexes 11‒18 in THF are shown in
Fig. 5.
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
11
12
13
14
15
16
17
18
(a)
11
12
13
14
15
16
17
18
b)
250
300
350
400
450
350
400
450
500
550
600
Wavelength (nm)
Wavelength (nm)
(a)
(b)
Fig. 5. Normalized absorption (a) and emission (b) spectra of complexes 11‒18 in THF.
The absorption spectra of 2-(N-alkyl formimino)pyrrolyl boron chelates 11‒16 in THF
virtually superimpose (Fig. 5a), showing very similar features, namely wavelength maxima in
the range 346‒352 nm, increasing with the polarizability of the aliphatic N-alkyl group (Table
1). The same occurs with the absorption spectra of the 2-(N-alkyl formimino)phenanthro-
14

ACCEPTED MANUSCRIPT
λmax
Table 1 Wavelength maximum (
) and molar extinction coefficient (εmax) of the first
abs
λmax
absorption band, wavelength maximum (
) and wavelength of the first vibronic transition
em
λ0−0
(
) of the emission band, fluorescence quantum yield (φf), lifetime (τf), rate constants (kf),
em
sum of non-radiative rate constants (knr) of the boron complexes 11‒18, in THF, at 293 K.
max
abs
0−0
em
max
b
c
a
0−0
em
max
λ
λ
em
Compound
No.
λ
kf
knr
εmax
φ
f
τf
λ
− λ
abs
Complex
(ns-1) (ns-1)
(nm)
(ns)
(nm) (nm)
377 395
(nm)
346
1.31
1.82
49
0.15 0.90 0.17
0.10 0.59 0.17
0.94
1.52
11
12
349
378 397
48
350
350
1.72
1.51
379 398
379 398
48
48
0.15 0.95 0.16
0.10 0.71 0.14
0.89
1.27
13
14
15
351
352
1.41
1.51
379 397
380 399
46
47
0.10 0.61 0.16
0.16 0.89 0.18
1.48
0.94
16
17
410
418
2.15
2.32
428 431
21
14
0.35 2.60 0.13
0.43 2.34 0.18
0.25
0.24
432 432
18
a 104 L mol-1 cm-1; b kf = φf f; c knr = (1-
/τ φf)/τf.
15

ACCEPTED MANUSCRIPT
pyrrolyl boron derivatives 17 and 18, which also exhibit maxima at very close wavelengths
(410 and 417 nm, respectively), though significantly red-shifted in relation to those of 11‒16,
owing to their higher π-conjugation length. These close similarities clearly indicate that the
chromophores within each of these two families of molecules are basically the same,
coinciding with the 2-iminopyrrolyl or π-extended 2-iminophenanthropyrrolyl fragments,
respectively.
For the same reasons, the fluorescence spectra of each type of complexes nearly
superimpose (Fig. 5b), showing wavelength maxima within 395‒399 nm for the simple 2-
iminopyrrolyl complexes, and 431‒432 nm for the 2-iminophenanthropyrrolyl derivatives, in
dilute solutions of THF. In this solvent, complexes 11‒16 emit in the violet region (Fig. 6)
with fluorescence quantum yields (ϕf) varying between 0.10 and 0.16, which can be
considered much higher than the efficiencies reported previously for the violet emitters
[BPh2(κ2N,N’-NC4H3C(H)=N-Aryl)] containing bulky aryl groups (2,6-iPr2C6H3 and 2,6-
Me2C6H3) [8c].
Fig. 6. Colors of complexes 11–18 in THF under UV-irradiation at 365 nm.
The fluorescence decays of all compounds 11‒18 are single-exponentials with shorter
lifetimes (τf) for 11‒16 (ca.0.6-0.9 ns) than for 17 and 18 (2.6 and 2.34 ns, respectively;
column 9 in Table 1). The radiative rate constant (kf =
φ
/
f τf) values (column 10, Table 1) of
16

ACCEPTED MANUSCRIPT
11‒16 are, in fact, similar to those of the bulky N-aryl substituted derivatives (and others)
[8c], but the corresponding non-radiative decay constants (knr) are 5 to 30 times lower because
the N-alkyl groups are not intrinsically involved in the electronic transition as the N-aryl
groups. On the other hand, the knr values of 11‒16, and thus their fluorescence quantum
yields, are similar to those measured for [BPh2(κ2N,N’-NC4H3C(H)=N-Aryl)] complexes
containing non-bulky aryl groups substituted by electron-releasing substituents in positions 3
or 4 [8c].
Within the family 11‒16 two subgroups may be differentiated: the one with compounds
bearing more rigid N-alkyl groups, such as the methyl (11), and the tertiary t-butyl (13) and
adamantyl (16), and the other one containing less rigid primary or secondary N-substituents,
such as the n-octyl (14), i-propyl (12) and cyclohexyl (15). The latter subgroup shows higher
non-radiative constants (1.27‒1.52 ns-1) than the former (0.89‒0.94 ns-1), and thus lower
φ
f
values (0.10 vs. 0.15-0.16).
Complexes 17 and 18, bearing the rigid phenanthrene moiety fused on the 2-iminopyrrolyl
ring, exhibit good fluorescence quantum yields (0.35 for 17 and 0.43 for 18) and are blue
emitters (Fig. 6). They show sizeable bathochromic shifts of 64 and 66 nm in relation to their
corresponding simple 2-formiminopyrrolyl derivatives 11 and 16, respectively, due to their
extended π-conjugation. Their kf values are in the same order of magnitude of those of 11‒16
but their non-radiative rate constants knr are considerably lower (0.25 ns-1 vs. 0.89‒1.52 ns-1),
which is related to the higher rigidity of these extended π-conjugated fluorophores.
In summary, it is possible to set the fluorescence emission of the 2-iminopyrrolyl boron
diphenyl chromophore to the violet region of the spectrum by using N-alkyl substituents, by
limiting the π-system of the chromophore to the pyrrolyl ring and the 2-imino arm, still
reaching quantum efficiencies of 10 to 15%. The fusion of a phenanthryl fragment on the C3-
17

ACCEPTED MANUSCRIPT
C4 pyrrolyl bond of the 2-(N-alkyl formimino)pyrrolyl ring shifts the emission to the red,
largely increasing the quantum efficiencies to 35‒43%.
2.4. Computational studies
The geometries of all boron complexes 11‒18 were optimized, based on the crystal structures
of 14, 15 and 18 described above [8,9]. The DFT [13] methodology was the same as reported
before for analogous compounds, using the ADF program [14] with a BP86 functional and
TZ2P basis sets for all atoms, without symmetry constraints (see also Computational details).
We also optimized the geometry of the first singlet states by promoting one electron from the
HOMO to the LUMO (same spin) and the geometry optimization was carried out as described
for the ground state.
In the previous studies, all the boron derivatives had a phenyl substituent on the 2-
iminopyrrolyl scaffold and it was shown that the dihedral angle α (Chart 2), associated with
the planarity of the bidentate ligand, changed upon excitation to the singlet excited state,
being therefore of major relevance in the definition of the photophysical properties. Although
the phenyl group is absent in all complexes of this work, this dihedral angle seems a suitable
parameter to compare different electronic states.
Chart 2.
Our analysis will address mostly four compounds (11, 16-18), allowing the comparison of
two ligands (methyl and adamantyl) and two types of π systems (2-formiminopyrrolyl and 2-
18

ACCEPTED MANUSCRIPT
formiminophenanthropyrrolyl). In the two complexes with methyl substituents (11 and 17), α
is 3.5° and 4.0°, respectively, in the ground state and increases to 50.1° in the singlet state in
11, suggesting that the pyrrolyl group does not constrain rotation, while it only reaches 35.5°
in 17. In the adamantyl complexes, the ground state α angles are 48.9° and 47.0°, increasing
only to 54.8° and 52.5° in the singlet state, for 16 and 18, respectively. The bulkiness of the
adamantyl prevents a different orientation in the ground state, so that the angle barely change
upon reaching the singlet state. The same angles in the crystal structures described above are
53.87° (18) and 3.05° for 14 (n-octyl, a good model for methyl 11). This contrasting behavior
parallels the different photophysical properties of the two groups of complexes.
The absorption spectra of all the complexes were obtained from TDDFT calculations [15],
using the implementation in ADF [14], with the BP86 functional and a TZ2P basis set, as in
the geometry optimization, both in gas phase and in solvent (COSMO in ADF), and also with
the SOPERT [16] method (spin-orbit coupling, SO, PBE0 functional and an all electron TZP
basis set), which allows the calculations of excited states lifetimes (see subsection
Computational Studies in the Experimental Section). Though the results should be the same,
since all atoms are light, the approach with spin-orbit coupling often provides the best
agreement with experimental data (maintaining the nature of the transitions) due to the use of
different functional and basis set.
The lower energy absorption of the four complexes (11, 16-18) is a broad band involving
transitions from the HOMO, HOMO-1 and HOMO-2 to the LUMO, for the pyrrolyl
derivatives 11 and 16, and from the HOMO and HOMO-1 to the LUMO in the case of the
phenanthropyrrolyl ones 17 and 18 (Fig. 7, top and bottom, respectively; Table 2). In the
pyrrolyl derivatives 11 and 16, both the HOMO and LUMO are localized in the 2-
iminopyrrolyl ligand, while the HOMO-1 and HOMO-2 have their greatest contribution from
the phenyl groups (one in each orbital). Therefore, the transition can be described as ILCT +
19

ACCEPTED MANUSCRIPT
LLCT (intraligand charge transfer + ligand-to-ligand charge transfer). On the other hand, in
complexes 17 and 18, the HOMO-1 is completely localized in the phenanthropyrrolyl moiety,
while the HOMO extends to the imino nitrogen. The electronic transition can therefore be
assigned as IL.
Fig. 7. The frontier orbitals of complexes 11, 16 (top) and 17, 18 (bottom) involved in the two
transitions leading to the low energy absorption.
The conclusion to be taken is that the extension of the π system modifies the nature of the
frontier orbitals and the transition. This has already been observed before in the boron
complexes with aryl substituents at the imino nitrogen, though the effect is not exactly the
same [10]. Moreover, despite the differences between methyl and adamantyl, none of their
fragment orbitals contribute to the frontier orbitals and the excitations.
20

ACCEPTED MANUSCRIPT
Table 2 Wavelength (
λ
), energy (E), composition and oscillator strength (O.S.) of the most
intense TDDFT electronic transitions calculated for complexes 11, 16–18 (SO).
Transition λ (nm) E (eV) Composition
O.S.
Complex 11
1
316
308
303
3.92
4.02
4.08
H→L (64%), H-1→L (25%)
0.151
0.056
0.099
2
H-1→L (66%), H-2→L (18%), H→L (12%)
H-2→L (68%), H→L (20%)
3
Complex 16
1
324
308
301
3.83
4.02
4.11
H→L (71%), H-1→L (24%)
H-1→L (70%), H→L (17%), H-2→L (11%),
H-2→L (82%)
0.196
0.109
0.058
2
3
Complex 17
1
376
368
3.30
3.37
H→L (90%)
0.260
0.221
2
H-1→L (90%)
Complex 18
1
2
380
370
3.26
3.35
H→L (93%)
0.364
0.192
H-1→L (93%)
Although the difference is not very relevant, two subgroups were identified with the
complexes 11-16. Since we analyzed 11 and 16, and saw that their frontier orbitals and
electronic transitions showed no difference, we also looked at one of the others, the isopropyl
derivative 12. The dihedral angle α defined in Chart 2 surprisingly decreases a little, from
69.4° to 64.9°. The lowest energy absorption is calculated in relatively good agreement with
experimental values (Fig. S35, Table S3), as for the complexes 11 and 16-18 (Table 3 and
Fig. 7). Only two transitions are responsible for the broad band calculated at 3.97 eV (312
nm) and both start either at the HOMO, HOMO-1, or HOMO-2, ending in the LUMO as in
complexes 11 and 16, being also assigned as mixed ILCT and LLCT. The HOMO-2 is similar
to that of 11 and 16 (Fig. 7), with a stronger contribution from the two phenyl groups.
21

ACCEPTED MANUSCRIPT
λmax
Table 3. Experimental wavelength maxima (
) of the first absorption band, absorption
abs
λmax
) of the emission bands, and fluorescence
em
energies (
), wavelength maxima (
rate constants (kf). Calculated (TDDFT) absorption energies (
,
), wavelength
max
maxima (
λ
(THF)) of the emission band, and fluorescence rate constants (kf (SO)) of the
em
boron complexes 11‒18.
a
max
abs
max
em
max
kf
λ
λ
(THF)
em
λ
k
(ns-
f (SO)
Compound
Complex
1
(ns-1)
No.
)
(nm) (nm, eV)
(nm)
346
(eV)
3.58
(eV)
3.76
(eV)
4.02
522
395
0.17
0.10
0.11
11
2.38
516
349
3.55
3.74
3.97
397
0.17
12
2.40
503
350
350
3.54
3.54
3.76
3.77
3.97
4.01
398
0.16
0.14
0.09
0.16
13
14
15
16
2.46
521
398
2.38
538
351
352
3.53
3.52
3.73
3.73
3.94
3.94
397
0.16
0.18
0.14
0.13
2.31
527
399
2.35
595
410
418
3.02
2.97
2.93
2.98
3.34
3.93
431
0.13
0.18
0.12
0.17
17
2.08
604
432
18
2.05
a kf = φf
/τf
22

ACCEPTED MANUSCRIPT
The emissions directly calculated from the difference between the energy of the first singlet
excited state, obtained by promoting one electron from the HOMO to the LUMO and
optimizing, and the energy of the ground state with the same geometry, reproduce the
experimental trends. The emissions from 17 and 18 occur at higher wavelengths than
emissions from 11 and 16, though the absolute values are shifted.
2.5. Electrochemical Studies
The electrochemical properties, namely the ionization potential (IP) and the electron affinity
(EA) of the complexes 11‒18 were investigated by cyclic voltammetry. These measurements
were performed in dichloromethane solutions with tetrabutylammonium tetrafluoroborate as
electrolyte salt, at room temperature and under inert (N2) atmosphere. The IP and EA values
(Table 2) were determined from the measured reduction and oxidation onset potentials, after
being converted to the absolute scale, using Fc/Fc+ (ferrocene/ferrocenium ion redox couple)
as external reference [8a]. As the energy level of Fc/Fc+ (ferrocene/ferrocenium ion redox
couple) is at 4.80 eV below the vacuum level we calculate IP (-HOMO) (eV) = Eonset,ox (eV) +
(4.80-EFc/Fc+) and EA (eV) = Eonset,red (eV) + (4.80- EFc/Fc+) where EFc/Fc+ represents the half-
wave potential of Fc/Fc+ measured in the investigated solvent media under our setup. The
values obtained are summarized in Table 4 (see also cyclic voltammograms in Figs. S36‒S43
in Supporting Information), along with the energies of the HOMOs and LUMOs of the
corresponding complexes calculated by DFT with solvent correction (THF).
As expected, the values of ‒IP correlate relatively well with the energies of the HOMOs,
with the IP values differing between 0.18 and 0.27 eV from the calculated ones (see Fig. S44a
in ESI). The ‒EA values also correlate well with the calculated LUMO energies, with
differences varying between 0.03 and 0.27 eV (see Fig. S44b in ESI).
23

ACCEPTED MANUSCRIPT
Table 4. Ionization potentials (IP), electron affinities (EA) of complexes 11‒18, estimated
from cyclic voltammetry measurements, and corresponding energies of HOMOs and LUMOs,
determined by DFT (THF).
Cyclic Voltammetry
DFT (THF)
IP
EA
EHOMO
ELUMO
(eV)
Complexes
(eV)
(eV)
(eV)
-5.406
-2.493
5.67
5.63
5.62
5.65
5.60
5.60
5.40
5.39
2.53
2.34
2.34
2.35
2.32
2.36
2.59
2.52
11
12
13
14
15
16
17
18
-5.409
-5.388
-5.381
-5.372
-5.346
-5.215
-5.171
-2.487
-2.513
-2.512
-2.511
-2.500
-2.777
-2.791
2.6. Electroluminescent properties
Non-doped
single-layer
light-emitting
diodes
(LEDs)
with
the
structure
ITO/PEDOT:PSS/complex/LiF/Al were prepared and characterized for all complexes except
for 12. The films of the complexes were prepared by spin coating. In the series 11‒16, we
could not obtain LEDs with reasonable performances. The best device with 11 gave a
maximum luminance of 1 cd m-2, and that based on 14 gave 18 cd m-2. The remaining ones
(13, 15 and 16) showed negligible light emission. Conversely, OLEDs based on 17 and 18
showed reasonable performances (as shown in Fig. 8), with a maximum luminance of 69 and
88 cd m-2, respectively, and with a maximum luminous efficiency of 0.041 and 0.019 cd A-1,
24

ACCEPTED MANUSCRIPT
respectively. Nevertheless, their emission is strongly controlled by aggregation, which is
consistent with the observation of a dimeric supramolecular structure, involving π- π stacking,
in the X-ray packing of complex 18 (see above in subsection “X-ray diffraction studies”). The
recorded electroluminescence (EL) spectra are compared with the solution and film
photoluminescence spectra in Fig. 9. The EL maximum occurs at 630 nm for the device based
on 17, and at 625 nm for that based on 18, while their solution emission maxima occur at 431
nm and 432 nm, respectively. The EL spectra appear to be extended to shorter wavelengths,
showing higher energy tails, which are attributed to the emission of the residual “isolated”
complexes. The photoluminescence (PL) spectra recorded for the spin coated films are similar
to the EL ones. In case of complex 17, we also recorded the PL spectrum of a sublimed film,
where two main emission bands are observed, with intensity maxima at 527 and 621 nm,
which may be attributed to different types of aggregates and a residual emission from
disordered complexes. This behavior shows that the film preparation details have a significant
effect on the films photoluminescence, which is consistent with aggregation. It is worth
mentioning that, upon naked eye inspection, while the sublimed film of 17 was clear and
homogeneous, that prepared by spin coating was somewhat heterogeneous with macroscopic
aggregates, having the spin coated films of 18 a similar appearance.
Fig. 8. Current (I) and luminance (L) of the devices based on 17 and 18 as function of the
applied voltage.
25

ACCEPTED MANUSCRIPT
Fig. 9. Electroluminescence (EL) spectra of (a) 17 and (b) 18 compared to their photoluminescence
(PL) spectra in THF solutions and in films prepared by spin-coating and, for the case of complex 17,
also by sublimation (sublim).
Compounds 17 and 18 exhibit similar fluorescence quantum yields and similar HOMO and
LUMO energies, all determined in solution, yet device performance is significantly different,
as shown in Fig. 8. In particular, the LED based on 17 shows a higher light-onset voltage and
a much lower current. This, we believe, is due to a stronger detrimental effect of the
aggregation on the charge transport as a minimum balanced current (electrons and holes) is
required to lead to measurable light emission. Aggregation will definitely modify also the
HOMO and LUMO energies with respect to the solution situation, and therefore the charge
26

ACCEPTED MANUSCRIPT
injection barriers. However, based on the available data, we cannot compare such effects for
both compounds.
This behavior shows that the film preparation details have a significant effect on the films
photoluminescence and also on the LEDs performance, which is consistent with aggregation.
In conclusion, the aggregation of these boron complexes has a strong effect on their EL
emission spectrum and efficiency. Their blending, for instance with conjugated polymers with
adequate frontier levels energy, is expected to prevent or strongly suppress their aggregation
and thereby improve the performance of the LEDs.
3. Conclusion
The synthesis of new mononuclear violet or blue emitting 2-iminopyrrolyl boron diphenyl
complexes, containing either the 2-(N-alkylformimino)pyrrolyl (11‒16) or the more π-
extended 2-(N-alkylformimino)phenanthro[9,10-c]pyrrolyl ligands (17 and 18), respectively,
with N-alkyl groups possessing different stereochemical features, was achieved in good
yields. The simple 2-iminopyrrolyl boron violet emitters 11‒16 showed fluorescence quantum
yield values in the range
φ = 0.10 ‒ 0.16, though significantly larger than those previously
f
reported for related violet emitters [BPh2(κ2N,N’-NC4H3C(H)=N-Ar)] containing bulky aryl
groups, such as Ar = 2,6-Me2C6H3 or 2,6-iPr2C6H3 (structure A in Chart 1, with R1 = Me or
iPr, and R2 = R3 = H) (
φ = 0.005 and 0.023, respectively, in THF solution), which presented
f
extensive non-radiative processes (knr = 27.6 and 7.52 ns-1, respectively) [8c]. Conversely, the
corresponding π-extended 2-iminophenanthropyrrolyl blue emitters 17 and 18, showed
enhanced quantum yields (
φ = 0.35 and 0.43, respectively, in THF solution). This different
f
behavior is associated with the lowest energy absorption, being ILCT + LLCT (intraligand
27

ACCEPTED MANUSCRIPT
charge transfer + ligand-to-ligand charge transfer) in complexes 11‒16, involving also the
phenyl groups, and simply ILCT in complexes 17 and 18.
Non-doped single-layer electroluminescent devices were fabricated using these
compounds as both emitter and ambipolar charge-transporting materials. In general,
complexes 11‒16 led to OLEDs with poor or negligible performances, complex 14 giving rise
to the best device with a maximum luminance of 18 cd m-2. Conversely, OLEDs based on 17
and 18 showed reasonable performances, with a maximum luminance of 69 and 88 cd m-2,
respectively, and maximum luminous efficiencies of 0.041 and 0.019 cd A-1, respectively.
Nevertheless, their emission is strongly controlled by aggregation, which is clearly observed
by the presence of highly red-shifted emission bands in the electroluminescence (EL) spectra
in relation to those of the “isolated” complexes.
4. Experimental section
4.1. General Procedures
All experiments dealing with air- and/or moisture-sensitive materials were carried out under
inert atmosphere using a dual vacuum/nitrogen line and standard Schlenk techniques.
Nitrogen gas was supplied in cylinders by Air Liquide and purified by passage through 4 Ǻ
molecular sieves. Unless otherwise stated, all reagents were purchased from commercial
suppliers (e.g., Acrös, Alfa Aesar, Aldrich, Fluka) and used without further purification. All
solvents to be used under inert atmosphere were thoroughly deoxygenated and dehydrated
before use. They were dried and purified by refluxing over a suitable drying agent followed
by distillation under nitrogen. The following drying agents were used: sodium (for toluene,
diethyl ether, and tetrahydrofuran (THF)), calcium hydride (for n-hexane and
28