Boron complexes of aromatic ring fused iminopyrrolyl ligands: Synthesis, structure, and luminescence properties

Article abstract of DOI:10.1039/c6dt02771g

The condensation reactions of 2-formylindole (1) or 2-formylphenanthro[9,10-c]pyrrole (2) with various aromatic amines afforded the corresponding phenyl or phenanthrene ring fused mono-/bis-iminopyrrole ligand precursors 3-8, which, upon reaction with BPh₃ in an appropriate molar ratio, led to the new mono- and diboron chelate compounds Ph₂B[NC₈H₅C(H)N-2,6-Ar] (Ar = 2,6-iPr₂C₆H₃9; C₆H₅10), Ph₂B[(NC₈H₅C(H)N)₂-1,4-C₆H₄]BPh₂11, Ph₂B(NC₁₆H₉C(H)N-Ar) (Ar = 2,6-iPr₂C₆H₃12; C₆H₅13), and Ph₂B[(NC₁₆H₉C(H)N)₂-1,4-C₆H₄]BPh₂14, respectively. Boron complexes 12-14, containing a phenanthrene fragment fused to the pyrrolyl C3-C4 bond, are highly fluorescent in solution, with quantum efficiencies of 37%, 61% and 58% (in THF), respectively, their emission colours ranging from blue to orange depending on the extension of π-conjugation. Complexes 9-11, containing a benzene fragment fused to the pyrrolyl C4-C5 bond, are much weaker emitters, exhibiting quantum efficiencies of 10%, 7% and 6%, respectively. DFT and TDDFT calculations showed that 2,6-iPr₂C₆H₃N-substituents or, to a smaller extent, the indolyl group prevent a planar geometry of the ligand in the excited state and reveal the existence of a low energy weak band in all the indolyl complexes, which is responsible for the different optical properties. Non-doped single-layer light-emitting diodes (OLEDs) were fabricated with complexes 9-14, deposited by spin coating, that of complex 13 revealing a maximum luminance of 198 cd m^-2.

Full text of DOI:10.1039/c6dt02771g

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. T. Gomes, D.
Suresh, B. Ferreira, P. S. Lopes, C. S. B. Gomes, K. paramasivam, A. Charas, D. Vila-Vicosa, J. Morgado, M.
J. Calhorda and A. Macanita, Dalton Trans., 2016, DOI: 10.1039/C6DT02771G.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/dalton

Page 1 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Boron Complexes of Aromatic Ring Fused Iminopyrrolyl
Ligands: Synthesis, Structure, and Luminescent Properties†
D. Suresh,a,e Bruno Ferreira,a Patrícia S. Lopes,a Clara S. B. Gomes,a
Paramasivam Krishnamoorthy,a Ana Charas,b Diogo Vila-Viçosa,d Jorge Morgado,b,c
Maria José Calhorda,d António L. Maçanita,a Pedro T. Gomes*,a
a Centro de Química Estrutural, Departamento de Engenharia Química, Instituto Superior Técnico, Universidade de
Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugal. E-mail: pedro.t.gomes@tecnico.ulisboa.pt
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.
e School of Chemical and Biotechnology, SASTRA University, Thanjavur – 613 401, India
†
Electronic Supplementary Information (ESI) available. CCDC 1485404 for 3, 1485405 for 7, 1485406 for 9,
1485407 for 10 and 1485408 for 13. For ESI and crystallographic data in CIF or other electronic format see DOI:
1
10.1039/

Dalton Transactions
Page 2 of 40
View Article Online
DOI: 10.1039/C6DT02771G
Abstract: The condensation reactions of 2-formylindole (1) or 2-formylphenanthro[9,10-
c]pyrrole (2) with various aromatic amines afforded the corresponding phenyl or phenanthrene
ring fused mono-/bis-iminopyrrole ligand precursors 3‒8, which, upon reaction with BPh3 in
appropriate molar ratio, afforded the new mono- and diboron chelate compounds
Ph2B[NC8H5C(H)=N-2,6-Ar] (Ar = 2,6-iPr2C6H3 9; C6H5 10), Ph2B[(NC8H5C(H)=N)2-1,4-
C6H4]BPh2 11, Ph2B(NC16H9C(H)=N-Ar) (Ar
=
2,6-iPr2C6H3 12; C6H5 13), and
Ph2B[(NC16H9C(H)=N)2-1,4-C6H4]BPh2 14, respectively. Boron complexes 12‒14, containing a
phenanthrene fragment fused on the pyrrolyl C3-C4 bond, are highly fluorescent in solution, with
quantum efficiencies of 37%, 61% and 58% (in THF), respectively, their emission colours
ranging from blue to orange depending on the extension of π-conjugation. Complexes 9‒11,
containing a benzene fragment fused on the pyrrolyl C4-C5 bond, are much weaker emitters,
exhibiting quantum efficiencies of 10%, 7% and 6%, respectively. DFT and TDDFT calculations
showed that 2,6-iPr2C6H3 N-substituents or, to a smaller extent, the indolyl group prevent a
planar geometry of the ligand in the excited state and reveal the existence of a low energy weak
band in all the indolyl complexes, which is responsible for the different optical properties. Non-
doped single-layer light-emitting diodes (OLEDs) were fabricated with complexes 9‒14,
deposited by spin coating, that of complex 13 revealing a maximum luminance of 198 cd m-2.
2

Page 3 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Introduction
The development of organic light emitting diodes (OLEDs),1 highly-tuned luminescent devices,
represents a further step towards progression in the display technology with much improved
clarity and resolution. Organoboron compounds2 are popular good candidates to be used in
emissive layers of OLEDs mainly because of the boron vacant pz orbital, which imparts Lewis
acidity to this centre and enables the formation of stable covalent dative bonds. This interaction
allows π-π* transitions through electronic conjugation of the π-electron rich organic ligand and
the empty pz orbital on boron. The ligands themselves come in a wide variety of designs, all
attempting to chelate the electrophile with the greatest stability possible. 3 Luminescent
tetracoordinate organoboron complexes have three main anionic chelating ligand scaffolds
containing one neutral N and one negative N-, or O- donor atoms, such as BODIPY (A),4 2-
pyridylphenolate (B)5 and quinolate (C)6 (Chart 1). The type of the ligands and the nature of the
substituted groups on either the ligands or the boron centres have a great influence on the
photophysical and electronic properties of the compounds due to the π-π* electronic transitions of
the chelate or charge transfer transitions from the substituted groups to the chelate in the exciting
7
process. The ligands with ring-fused structures having extended π-conjugation fetched
considerable attention for their applications in high performance optoelectronics.8 The rigid
structures in the ligands can effectively restrict molecular rotations, which usually cause serious
fluorescence quenching and the flat π-conjugated skeletons may facilitate charge transport due to
intermolecular π-electron delocalisation. Thus, the ligands with flat and rigid π-conjugated
skeletons are important to bring unique properties such as intense luminescence, good thermal
stability, and high charge carrier properties.
The 2-iminopyrrolyl ligands are also an important class of anionic bidentate chelating
ligands, containing a pyrrolyl anionic ring and a neutral imine as donor moieties.9 Several
functional groups, such as electron releasing or electron withdrawing substituents, or bulky
groups, can be introduced easily into the ligand systems. Thus, these chelate moieties are highly
flexible in ligand design, which makes them very interesting for various applications including
catalysis.9,10
3

Dalton Transactions
Page 4 of 40
View Article Online
DOI: 10.1039/C6DT02771G
Chart 1.
Recently, there has been considerable interest in the synthesis of luminescent metal complexes
containing these ligands. Ma and co-workers11 reported weakly fluorescent zinc complexes
containing the bis(pyrrol-2-ylmethyleneamine) ligands showing an emission maximum around
400 nm. Copper complexes of phosphine or bis(phosphine) ligands and 2-(N-
phenylformimino)pyrrolyl as ancillary ligands were reported as phosphorescent complexes.12
Both the experimental and the theoretical results proved that the photophysical properties of these
complexes largely depend on the anionic ancillary ligands. Our group earlier reported blue or
green colour emitting luminescent zinc complexes containing 2-formylphenanthro[9,10-
c]pyrrolyl ligands,13 where the π-conjugation was extended by fusing the phenanthrene ring onto
the C3-C4 bond (c bond). We also reported the synthesis and luminescent properties of several
tetracoordinate organoboron complexes containing 2-(N-arylformimino)pyrrolyl ligands (Chart 1,
D) having varied the electronic and steric nature of the aryl group.14 It was found that the colour
of emission could be tuned from blue to bluish-green by increasing the electron-donating power
of the substituent. We further developed our work towards the synthesis of polynuclear boron
complexes, where the iminopyrrolyl ligands had different aromatic bridging spacers thereby
varying the π-conjugation length (Chart 1, E).14a,b,15 In this second approach, the emission could
be tuned from blue to yellow.
In the light of these interesting results, we describe herein the use of a third approach to tune
the emission in this type of compounds: the synthesis of new boron complexes containing 2-
iminopyrrolyl scaffolds having fused aromatic rings, leading to a further increased π-conjugated
4

Page 5 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
length in the ligand. For this purpose, the 2-formylindole (1) and 2-formylphenanthro[9,10-
c]pyrrole (2) (Scheme 1) were chosen as starting materials. In compounds 1 and 2, either benzene
or phenanthrene rings are fused onto the C4-C5 or C3-C4 bonds (i.e. d or c bonds) of 2-
formylpyrrole fragment, respectively, the corresponding 2-iminopyrrole ligand precursors being
obtained by condensation reactions with aromatic amines. The reactions of these ligand
precursors with triphenylboron led to the synthesis of mono- and binuclear organoboron
compounds, which were characterised by multinuclear NMR, X-ray diffraction and cyclic
voltammetry. The photophysical characterisation of the resulting new boron complexes was
performed using steady state and time-resolved luminescence techniques in solution. Density-
functional theory (DFT) and time-dependent DFT (TDDFT) calculations (ADF program) were
also carried out for these new boron complexes to assign the electronic transitions, to determine
the geometry of the first excited state, and to try to rationalise the luminescence behaviour
exhibited. Further, these compounds were utilized as emissive layers in single-layer light-
emitting diodes (OLED), being the corresponding films prepared by spin coating.
Results and Discussion
Syntheses and characterisation of boron complexes
The condensation reactions of 2-formylindole (HNC8H5C(H)=O) (1) with 2,6-diisopropyl aniline
and aniline, under appropriate reaction conditions, i.e. reflux in absolute ethanol with MgSO4 and
in the presence of a catalytic amount of p-toluenesulfonic acid,9d,e,g-j,13,14a,c,15 afforded the
corresponding mono 2-(N-arylformimino)indole ligand precursors, HNC8H5C(H)=N-2,6-
iPr2C6H3 (3) and HNC8H5C(H)=N-C6H5 (4), respectively, in good yields (Scheme S1 in ESI).
The 2-formylphenanthro[9,10-c]pyrrole (HNC16H9C(H)=O) (2) was obtained by a multistep
procedure, starting from phenanthrene, as reported in the literature.16 Similarly, condensation
reactions of 2 with 2,6-diisopropyl aniline and aniline afforded the corresponding formimines
HNC16H9C(H)=N-2,6-iPr2C6H3 (6) and HNC16H9C(H)=N-C6H5 (7), respectively, in moderate
yields, under standard reaction conditions,9d,e,g-j,13,14a,c,15 as shown in Scheme 1. The bis(2-(N-
arylformimino)indole)
(HNC8H5C(H)=N)2-1,4-C6H4
(5)
and
the
bis(2-(N-
arylformimino)phenanthro[9,10-c]pyrrole) (HNC16H9C(H)=N)2-1,4-C6H4 (8) ligand precursors
were synthesised by the condensation reaction of two equivalents of 1 or 2, respectively, with one
equivalent of 1,4-phenylene diamine in absolute ethanol, without addition of MgSO4 to the
5

Dalton Transactions
Page 6 of 40
View Article Online
DOI: 10.1039/C6DT02771G
reaction media. The ligand precursors 3‒8 were characterised by various techniques including
multinuclear NMR, the molecular structures of 3 and 7 being confirmed by single crystal X-ray
diffraction (see ESI).
The stoichiometric reactions of ligand precursors 3, or 4, 6 and 7 with triphenyl boron, in
toluene under refluxing conditions, afforded the corresponding mononuclear organoboron
complexes
Ph2B[NC8H5C(H)=N-2,6-iPr2C6H3] (9),
Ph2B[NC8H5C(H)=N-C6H5]
(10),
Ph2B[NC16H9C(H)=N-2,6-iPr2C6H3] (12), Ph2B[NC16H9C(H)=N-C6H5] (13), respectively, in
good yields. The binuclear boron complexes Ph2B[(NC8H5C(H)=N)2-1,4-C6H4]BPh2 (11) and
Ph2B[(NC16H9C(H)=N)2-1,4-C6H4]BPh2 (14) were synthesised by reacting 5 and 8, respectively,
with triphenyl boron in a 1:2 molar ratio, under nitrogen atmosphere, in refluxing toluene over 16
h (Scheme 1).
1
The mononuclear organoboron compounds 9, 10, 12 and 13 were characterised by H, 13C
and 11B NMR, while the poor solubility of the binuclear 14 in CD2Cl2 prevented the recording of
13
its C NMR spectrum. The imine proton (HC=N) 1H NMR resonance of the complexes appears
as a singlet in the range δ 8.43 to 9.14. The 11B NMR resonance of these compounds is in the
range of δ 5.08 to 6.99, confirming the formation of tetracoordinate boron species. The elemental
compositions of all the complexes also supported the formulation proposed for the products. Both
the mono- and binuclear organoboron complexes were moderately air and moisture stable,
however exposure to atmosphere for prolonged periods led to decomposition. Thus, these
compounds were best stored as solids under inert atmosphere.
The molecular structures of mononuclear boron complexes 9, 10 and 13 were determined by
single-crystal X-ray diffraction studies. Perspective views of the molecular structures of 9, 10 and
13 are shown in Fig. 1 and 2. Selected bond lengths and bond angles are given as caption in the
corresponding figures. Crystals suitable for these studies were obtained by cooling double-
layered dichloromethane and n-hexane or toluene and n-hexane solutions of 9 or 10 or 13 to
‒20°C, for 2 days. All the boron complexes crystallized in the monoclinic crystal system, with
space groups Pn for 9, and P21/n for 10 and 13.
6

Page 7 of 40
Dalton Transactions
Scheme 1. Synthesis of organoboron complexes 9–14.
7

Dalton Transactions
Page 8 of 40
View Article Online
DOI: 10.1039/C6DT02771G
(a)
(b)
Fig. 1 Perspective views of molecular structures of (a) 9 and (b) 10. The ellipsoids were drawn at
50% probability level. All the calculated hydrogen atoms were omitted for clarity. Compound 9:
Selected bond lengths (Å): N1-C2, 1.384(10); N1-C5, 1.348(10); N1-B1, 1.567(10); N2-C6,
1.302(10); N2-C11, 1.423(9); N2-B1, 1.657(10); C23-B1, 1.618(11); C29-B1, 1.597(11).
Selected bond angles (°): C2-N1-C5, 107.2(6); C2-N1-B1, 111.6(6); C5-N1-B1, 139.2(6); C6-
N2-C11, 122.2(6); C6-N2-B1, 111.1(6); C11-N2-B1, 126.0(5); N2-B1-C29, 111.8(6); C23-B1-
C29, 115.3(6); N1-B1-N2, 94.9(5); N1-B1-C23, 109.3(6); N1-B1-C29, 115.6(6); N2-B1-C23,
109.3(6). Compound 10: Selected bond lengths (Å): N1-C2, 1.3822(17); N1-C5, 1.3616(17); N1-
B1, 1.5495(18); N2-C6, 1.2971(17); N2-C7, 1.4335(17); N2-B1, 1.6458(18); C17-B1,
1.6458(18); C23-B1, 1.6147(19). Selected bond angles (°): C2-N1-C5, 107.03(11); C2-N1-B1,
113.01(11); C5-N1-B1, 139.09(11); C6-N2-C7, 121.37(11); C6-N2-B1, 111.20(11); C7-N2-B1,
127.36(10); N1-C2-C3, 111.49(12); N1-C2-C6, 108.30(11); N1-B1-C17, 113.54(11); N1-B1-
C23, 110.86(11); N1-B1-N2, 94.90(9); N2-B1-C23, 109.67(10); C17-B1-C23, 115.47(11); N2-
B1-C17, 110.48(10).
8

Page 9 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Fig. 2 Perspective view of the molecular structure of 13. The ellipsoids were drawn at 50%
probability level. All the calculated hydrogen atoms were omitted for clarity. Selected bond
lengths (Å): N1-C2, 1.387(3); N1-C5, 1.330(3); N1-B1, 1.574(3); N2-C6, 1.324(3); N2-C7,
1.426(3); N2-B1, 1.626(3); C25-B1, 1.610(3); C31-B11.617(3). Selected bond angles (°): N2-B1-
C31, 112.13(18); C25-B1-C31, 116.13(19); N2-B1-C25, 110.21(18); N1-B1-N2, 94.74(17); C2-
N1-C5, 109.23(19); C2-N1-B1, 112.60(18); C5-N1-B1, 137.9(2); N2-B1-C31, 112.13(18); C25-
B1-C31, 116.13(19); N2-B1-C25, 110.21(18); N1-B1-N2, 94.74(17); N1-B1-C25, 112.45(18);
N1-B1-C31, 109.19(18).
In all complexes, the boron centre is tetracoordinate and adopts a typical distorted tetrahedral
geometry. Each boron centre is chelated by the corresponding N,N' ligands via N1, N2 atoms to
form a five membered ring, the remaining two coordination sites being occupied by the
quaternary carbon atoms of the phenyl groups. Bond distances and angles are similar to those of
other related organoboron compounds previously reported.14a,c,15 The dihedral angles between the
2-indolyl or phenanthro[9,10-c]pyrrolyl and the aromatic imine substituent planes (C6–N2–C7–
C12 in 10, and C6–N2–C7–C8 in 13) are ‒43.35(17)° and ‒36.9(3)°, respectively, suggesting no
severe restricted rotation around the C–Nimine bond, whereas in 9 the restricted rotation is
reflected in the C6–N2–C11–C16 torsion angle of 76.7(9)°.
The supramolecular structure of complex 9, 2-(N-2,6-diisopropylphenylimino)indolyl boron
diphenyl reveals four C‒H∙∙∙πC(Ar) short contacts and no special motifs are observed (see Fig. S2
and Table S2 in ESI). On the other hand, the head-to-tail dimers of the 2-(N-phenylimino)indolyl
9

Dalton Transactions
Page 10 of 40
View Article Online
DOI: 10.1039/C6DT02771G
boron diphenyl derivative 10 results from weak C–H⋯πC(Pyrr) short contacts (Fig. S3 and Table
S2 in ESI). Additionally, the occurrence of other C‒H∙∙∙πC(Ar) short contacts completes the 3D
crystal packing of this complex. The crystal packing of compound 13, 2-(N-
phenylimino)phenanthro[9,10-c]pyrrolyl boron diphenyl, is particularly complex, involving
πC(Ar)⋯πC(Ar) stacking interactions leading to the formation of a 1D chain of head-to-tail dimers
(Fig. S4 in ESI), and weak C‒H∙∙∙πC(Ar) hydrogen bonds, responsible for the 2D-sheets containing
the phenanthro[9,10-c]pyrrolyl-boron moieties (Fig. S5 and Table S2 in ESI).
Molecular geometries and electronic structures
All geometries of boron complexes 9–14 were optimised without any symmetry constraints, with
the same methodology that was used to optimise the geometry of the previously reported
complexes 15−1714c,15 (Chart 2). DFT17 calculations were performed with the ADF18 program
(BP86 functional and TZ2P basis set, as detailed in Computational details).
Complexes 15−17 were based on the initial 2-iminopyrrolyl scaffold14c,15 and provide a
reference for comparison with the new boron derivatives having fused aromatic rings, and
obtained from HNC8H5C(H)=N-2,6-iPr2C6H3 (3) and HNC8H5C(H)=N-C6H5 (4) as described
above.
Chart 2. Previously reported reference compounds in this work.
The optimised geometries of complexes 9–14 in the ground state (modelled after the single
crystal X-ray structures of 9, 10, and 13) and in the first singlet excited state (obtained by
promoting one electron from the HOMO to the LUMO) are shown in Fig. 3. The B‒C and B‒N
distances do not vary significantly and the calculated values match the experimental ones for
complexes 9, 10, and 13. Despite some small differences, the most relevant geometrical change
10

Page 11 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
upon excitation to the first excited state is the Cimino‒Nimino‒Caryl-ipso‒Caryl-ortho dihedral angle (α)
(Chart 3), which tends to become closer to zero, as the 2-(N-aryl formimino)pyrrolyl ligand
approaches planarity.
Chart 3. Definition of dihedral angle α. Front (a) and side (b) views.
In the ground state, the calculated dihedral angles are similar in the 2-N-phenylimino
mononuclear derivatives 16, 10, 13, where the pyrrolyl fragment (29.2°) is replaced by the
indolyl (33.3°), and then by the phenanthropyrrolyl (28.6°). These values are lower than the
experimental ones (-43.35° for 10 and -36.90° for 13), probably owing to packing effects in the
latter. In the singlet excited state, the ligand becomes almost planar in 16 (1.7°), remaining at 9.2°
and 3.9° for 10 and 13, respectively, the indolyl preventing the flattening of the ligand.
In the three mononuclear complexes bearing the two isopropyl substituents on the phenyl (15,
9 and 12), the dihedral angle remains very high, even in the excited state, as expected from the
steric repulsions at work. The highest value is observed for 15 (80.4°), while 9 and 12 have
similar values (69.4°, 66.9°). Interestingly, 15 exhibits the largest change, the dihedral angle
reaching 41.7° in the excited state. The less planar excited state geometry, however, is found
again for the indolyl derivative (59.4°), where the smallest change is observed (Δ = 10.0°).
Finally, the binuclear complexes 17, 11 and 14 display symmetric structures, with dihedral angles
in the range 25‒29°. They remain symmetric in the excited state, 17 and 14 being almost flat (4‒
5°), while the indolyl complex shows the widest angles (~9°).
The first singlet excited state was also optimized with TDDFT.19 Although the dihedral
angles are slightly different, the trends are the same. For 9 and 12 they are even higher, 71.4° and
65.4°, respectively.
11

Dalton Transactions
Page 12 of 40
View Article Online
DOI: 10.1039/C6DT02771G
Fig. 3 Optimised geometries of the iminopyrrolyl boron complexes 9–14 in the ground state and
in the first excited singlet state, showing the dihedral angles α (°), and their comparison with
reference complexes 15‒17.
12

Page 13 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
The HOMOs and LUMOs of the complexes are shown in Fig. 4, with their relative energies (gas
phase). They are organised in groups of three, mononuclear with N-phenyl, mononuclear with N-
2,6-diisopropylphenyl substituents, and binuclear. In each group, from left to right the pyrrolyl
changes to indolyl and to phenanthropyrrolyl. The smallest HOMO-LUMO energy gap is always
observed for the indolyl complexes 10, 9, and 11. It results from the lowest energy LUMO in the
three complexes, and from the highest HOMO of the mononuclear species. The HOMO of the
binuclear is destabilised in 14 relative to 11, because it is mostly localised in the π system of the
iminopyrrolyl ligand and the extension from indolyl to phenanthropyrrolyl introduces more
antibonding interactions. The symmetry of the frontier orbitals of the binuclear species leads to
the symmetric geometry of the singlet excited state. The introduction of the isopropyl substituents
has negligible effect on the energy of HOMO and LUMO.
Fig. 4 DFT calculated energies (gas phase, eV) and three-dimensional representations of the
HOMOs and LUMOs of the complexes 9–17: blue, N-C6H5; black, N-2,6-iPr2C6H3; turquoise,
binuclear derivatives.
Electrochemical studies
Electrochemical measurements, using cyclic voltammetry (CV), were performed to estimate the
ionisation potential (IP) and the electron affinity (EA) of compounds 9‒14. These measurements
were performed in dichloromethane with tetrabutylammonium tetrafluoroborate or
tetrabutylammonium perchlorate as the electrolyte, at room temperature and under inert (N2)
13

Dalton Transactions
Page 14 of 40
View Article Online
DOI: 10.1039/C6DT02771G
atmosphere. The IP and EA values were determined from the reduction and oxidation onset
potentials, respectively, after being converted to the absolute scale, using Fc/Fc+
(ferrocene/ferrocenium ion redox couple) as external reference.14a The values obtained are
summarised in Table 1, along with the energies of the HOMOs and LUMOs of the corresponding
complexes calculated by DFT with solvent correction (THF). The same data for reference
compounds 15‒17 are also listed for comparison. As expected, the values of ‒IP correlate
relatively well with those of the energies of the HOMOs, with the IP values differing between 0.1
and 0.3 eV from the calculated ones (see Fig. S6 in ESI). Instead, the ‒EA values show a larger
difference from the calculated LUMO energies, which can be explained by the fact that the
calculated LUMO refers to the neutral species whereas EA refers to the stability of the anionic
species. The values of the HOMOs and LUMOs calculated by the SOPERT (SO) method do not
fit so well as those with the THF correction.
Table 1 Ionisation potentials (IP), electron affinities (EA) of complexes 9‒17, estimated from
cyclic voltammetry measurements, and corresponding energies of HOMOs and LUMOs,
determined by DFT (THF).
Cyclic Voltammetry
DFT (THF)
EHOMO
IP
EA
ELUMO
(eV)
Complexes
(eV)
(eV)
(eV)
15 a
9
5.83
5.58
5.58
5.64
5.43
5.43
5.50
5.45
5.02
2.56
2.76
2.67
2.82
3.14
2.90
3.44
3.24
3.17
-5.548
-2.699
-5.280
-5.355
-5.414
-5.278
-5.266
-5.201
-5.277
-5.082
-3.105
-2.988
-2.877
-3.208
-3.094
-3.123
-3.501
-3.317
12
16a,b
10
13
17 c
11
14
a Ref. 14c. b Ref. 14a. c Ref. 15.
14

Page 15 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Photophysical characterisation of new boron complexes 9–14
The UV–Vis absorption and fluorescence spectra of complexes 9−14 in THF are shown in Fig. 5.
The ligand precursors 3–8 are non-emissive while most of their organoboron complexes show
intense fluorescence emission. The organoboron compounds BPh2(NC4H3C(H)=N-2,6-iPr2C6H3)
(15), BPh2(NC4H3C(H)=N-C6H5) (16) and Ph2B[(NC4H3C(H)=N)2-1,4-C6H4]BPh2 (17), already
reported,14,15 are included here for comparison with the new related compounds.
The absorption spectra of indolyl (9–11) and phenanthropyrrolyl (12–14) derivatives in THF
show wavelength maxima within the 370–390 nm and 420–493 nm ranges, respectively, while
the simple non-fused pyrrolyl derivatives (15–17) show absorption maxima in the range between
353–419 nm (Table 2). This indicates that the fusion of rigid aromatic rings on the pyrrolyl
moiety onto the C3-C4 bond (phenanthropyrrolyl) has significant impact on the absorption
spectra when compared to the C4-C5 fusion (indolyl), reflecting the much more π-extended fused
fragment in the first case. On the other hand, the absorption spectra of 9, 10 and 11 (blue spectra
in Fig. 5) show abnormal tails, extending to longer wavelengths, which could be due to the
presence of less allowed S1←S0 transitions at wavelengths longer than those observed for the
absorption maxima. This tail is particularly evident in the case of complex 9 (Fig. 5a).
The absorption spectra were calculated using TDDFT calculations in gas phase (GP, BP86
functional and TZ2P basis set, as in geometry optimisation), in THF (THF, COSMO approach in
ADF), and with the SOPERT method (SO, PBE0 functional, all electron basis set, see
Computational details). Although the spin-orbit coupling treatment available in this approach is
not needed for boron derivatives, it allows the calculation of the excited state lifetimes, but the
same calculation (functional, basis set) without the spin-orbit coupling yields the same results.
For some of the compounds being studied, the SO methodology affords the closest
agreement with the experimental values (Table 2).
The GP results are closer to the experimental values for the pyrrolyl derivatives 15 and 16, while
the solvent correction approaches the calculated values to the experimental ones for the indolyl
and phenanthropyrrolyl species 12 and 13. The SO energy is significantly better for the three
binuclear compounds 17, 11, and 14. For compound 14, for instance, it increases from 2.05 (GP)
to 2.09 (THF) and to 2.52 eV (SO), much closer to the observed 2.51 eV. The three compounds
9, 10 and 11 are different, since all the calculation methods lead to the appearance of a low
energy band with small oscillator strength, followed by a more intense one at higher energies,
respectively at 2.96 and 3.50 eV (9), at 2.76 and 3.20 eV (10), and at 2.57 and 2.72 eV (11) in the
SO approach. Notice that the higher energy band agrees rather well with the experimental
15

Dalton Transactions
Page 16 of 40
View Article Online
DOI: 10.1039/C6DT02771G
Fig. 5 Normalised absorption and emission spectra of boron complexes 9–17 in THF: (a) N-2,6-
iPr2C6H3; (b) N-C6H5; and (c) binuclear derivatives. The colour code is: red, 2-iminopyrrolyl;
blue, 2-iminoindolyl; black, 2-iminophenanthropyrrolyl scaffolds.
16

Page 17 of 40
Dalton Transactions
View Article Online
max
abs
Table 2 Experimental wavelength maximum (
λ
) and molar extinction coefficientDO(εI:m10ax.1)03o9f/Ct6hDeT02771G
first absorption band, in THF, at 293 K, and the transition energy (in eV) of the boron complexes
max
abs
max
abs
max
abs
9–17, calculated in the gas-phase,
E
(GP) , in THF,
E
(THF) and SOPERT,
E
(SO)
method.
max
abs
a
λ
Complex
Structure
εmax
(nm)
(eV)
3.51
(eV)
3.76
(eV)
3.79
(eV)
3.92
15 14a,c
353
1.7
N
2.45
3.20
2.56
3.24
2.96
3.50
N
B
370
422
2.51
2.11
3.35
2.94
9
N
B
2.88
3.29
2.90
3.34
3.26
3.48
12
N
16 14a,c
383
391
1.73
2.54
3.24
3.17
N
2.25
3.00
2.44
3.07
2.76
3.20
N
B
10
N
B
441
428
438
2.26
3.0
2.81
2.90
2.90
2.71
2.47
2.24
2.77
2.52
3.04
2.93
13
N
B
17 14a,15
N
N
N
N
B
1.81
2.35
2.57
2.72
B
N
N
N
N
3.67
11
B
B
N
N
N
N
B
493
3.9
2.51
2.05
2.09
2.52
14
a 104 L mol-1 cm-1
absorption maxima and the second, weaker one, explains the tail. In all the other compounds the
lowest energy transition corresponds to the larger oscillator strength. Considering this
17

Dalton Transactions
Page 18 of 40
View Article Online
DOI: 10.1039/C6DT02771G
interpretation, it is clear that the calculated general trend is the same as observed experimentally,
namely a shift to lower energy as the π system increases from pyrrolyl, to indolyl and
phenanthropyrrolyl. Fig. 6 shows the simulated absorption spectra for the two complexes 10 and
13 in the low energy range.
Fig. 6 Calculated (DFT, SO) low energy absorption spectra of complexes 10 (top) and 13
(bottom), and the orbital contributions to each transition (see below).
In the simplest iminopyrrolyl complex 16, only one band is observed in the visible, which
consists of a HOMO to LUMO transition (90%). This is an intra-ligand (IL) transition within the
iminopyrrolyl ligand.14b The nature and the composition of the transitions responsible for the low
energy absorptions of the new compounds 9‒14 are listed in Table 3. We shall start by analysing
the indolyl (10) and phenanthropyrrolyl (13) derivatives.
18

Page 19 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Table 3 Composition, energy, wavelength and oscillator strength of the most intense TDDFT
electronic transitions calculated for complexes 9–14 (SO).
Transition λ (nm) E (eV) Composition
O.S. a
Complex 9
1
418
361
357
344
2.96
3.44
3.47
3.60
H→L (92%)
0.076
0.106
0.222
0.138
2
H-2→L (61%), H-1→L (20%), H-3→L (11%)
H-1→L (53%), H-2→L (34%)
H-3→L (78%), H-1→L (12%)
3
4
Complex 10
1
448
386
2.77
3.20
H→L (84%), H-1→L (15%)
H-1→L (79%), H→L (14%)
0.060
0.584
2
Complex 11
1
482
456
2.57
2.72
H→L (83%), H-1→L (11%)
H-2→L (78%), H→L (14%)
0.469
1.024
2
Complex 12
1
381
374
3.25
3.31
H→L (95%)
0.476
0.076
2
H-1→L (95%)
Complex 13
1
410
391
3.02
3.16
H→L (97%)
0.629
0.113
2
H-2→L (97%)
Complex 14
1
493
2.52
H→L (97%)
1.593
a Oscillator strength
There are two excitations calculated for compound 10. The one at lower energy results from a
mixed HOMO→LUMO (84%) and HOMO-1→LUMO (15%) transition and the oscillator
strength is only 0.060, while the second one consists of analogous transitions (HOMO→LUMO
and HOMO-1→LUMO), but the HOMO-1→LUMO component becomes the predominant. Also,
the oscillator strength (0.584) is almost 10 times larger (Fig. 6). Both transitions can be assigned
as IL (within the iminoindolyl ligand) and start from different π levels of the indolyl.
These two excitations account for the absorption maximum and the low energy tail. On the
other hand, complex 13 presents the strongest component, a HOMO→LUMO excitation at lower
energy, with an oscillator strength of 0.629, and a higher energy tail with an oscillator strength of
0.113 and HOMO-2→LUMO composition (see Fig. 6). These calculated features reflect the
experimental findings and can be also observed when comparing 9 and 12 or 11 and 14 (see Fig.
S7, S8, S9, and S10 in ESI, respectively).
19

Dalton Transactions
Page 20 of 40
View Article Online
DOI: 10.1039/C6DT02771G
The experimental emission spectra for the newly synthesised complexes (Fig. 5) show a wide
range of wavelength maxima from 438–585 nm in THF depending on the N-aryl substituents, the
relevant fluorescence data being collected in Table 4. The boron compounds 9 and 12, having
sterically hindered phenyl groups at positions 2 and 6, emit in green (λem = 519 nm) and blue
regions (λem = 438 nm), respectively, while 10 and 13 (without any substituents at positions 2 and
6) emit in the yellow (λem = 571 nm) and green regions (λem = 496 nm), respectively. The diboron
complexes, which have extended π-conjugation through the bridging phenyl group, emit in the
green-yellow (λem = 543 nm) and orange-red regions (λem = 584 nm), respectively for 11 and 14.
0−0
em
max
Finally, the large apparent Stokes shifts of indolyl compounds 9‒11 (
λ
−
λ
, column 5
abs
max
in Table 4) result from the fact that their
λ
correspond to the stronger S2←S0 transitions and
abs
not to the S1←S0 transition responsible for the S1→S0 emission.
The emission energies calculated by the difference between the energy of the first singlet
state and the energy of the ground state with the same geometry, obtained in a gas-phase
calculation with ADF (BP86, TZ2P), or using the same procedure based on energies determined
in THF (COSMO implementation in ADF) are too far away from the experimental values (see
Table S3 in ESI). On the other hand, the emission energies (and wavelengths) obtained by the
TDDFT optimisation of the first singlet excited state give a better estimation (Table 4),
reproducing the main trends, which are associated with the singlet excited state geometry
described above. In particular, compounds 9 and 12, with large Cimino‒Nimino‒Caryl-ipso‒Caryl-ortho
dihedral angle in the ground and singlet states emit at lower wavelengths.
The rigid phenanthrene fused iminopyrrolyl complexes showed good fluorescence quantum
yields (φf = 0.37 for 12, 0.61 for 13 and 0.58 for 14), while the phenyl ring fused 2-iminopyrrolyl
(2-iminoindolyl) derivatives showed quite low quantum yields (0.10 for 9, 0.7 for 10 and 0.6 for
11).
Fluorescence decays of complexes 9–14 were measured at three emission wavelengths (onset,
maximum, and tail of the fluorescence spectrum). The decays were single exponential functions
with fluorescence lifetimes (τf in Table 4) appreciably longer for the 2-iminoindolyl derivatives
9, 10 and 11 (4.91, 4.51 and 2.64 ns, respectively). The fluorescence rate constant
(kf (exp) =
φ
/
f τf) values of 9‒11 were found ca. one order of magnitude lower than those of the
other compounds, while the values of the non-radiative rate constant (knr = (1‒
φ
f)/ f) did not
τ
change more than a factor of two. This was the origin of the low fluorescence quantum yields of
9‒11.
20

Page 21 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Table 4 Experimental and calculated fluorescence data of complexes 9–17: experimental wavelengths
max
em
0−0
of the fluorescence maximum (
λ
) and the first vibronic S1→S0 transition (
λ
); fluorescence
em
quantum yield (φf), lifetime (τf), rate constants of fluorescence kf (exp), and sum of non-radiative rate
constants (knr), in THF, at 293 K; calculated (TDDFT) wavelengths of the fluorescence maximum
max
(
λ
(TDDFT)) and calculated fluorescence rate constants (kf (SO)).
em
a
0−0
em
max
em
max
em
0−0
em
max
b
kf
kf(SO)
λ
− λ
abs
λ
λ
λ
(TDDFT)
Complex
15 14a,c
9
Structure
knr
φf
τf
1
1
(ns-1)
(ns) (ns-
)
(ns-
)
(nm)
(nm) (nm)
(nm, eV)
404,
3.07
386 412
33
0.023 0.13 0.18 7.52 0.01
0.10 4.91 0.02 0.18 0.03
N
589,
2.10
N
B
477 519
438 438
107
16
420,
2.95
N
B
0.37 1.9 0.19 0.33 0.22
0.34 1.90 0.18 0.35 0.22
12
N
422,
2.94
16 14a‒c
451 479
527 571
68
N
B
609
2.04
N
136
0.07 4.51 0.016 0.21
0.02
10
446
2.78
N
B
495 495
54
0.61 2.47 0.24 0.17 0.25
13
N
B
413
3.02
17 14a,b,15
N
512 512
537 571
584 584
84
34
91
0.69 2.2 0.31 0.14 0.40
0.06 2.64 0.023 0.36 0.14
0.58 1.96 0.30 0.21 0.43
N
N
N
B
B
649,
1.91
N
N
N
11
N
B
B
474,
2.62
N
N
N
N
14
B
a kf =
φ /τ φf) /τf.
f; b knr = (1-
f
21

Dalton Transactions
Page 22 of 40
View Article Online
DOI: 10.1039/C6DT02771G
The 10-fold lower kf values of 9‒11 indicates a 10-fold lower dipole moment for the S1→S0
transition, which is, apparently, inconsistent with the similar molar extinction coefficient at the
absorption maximum (εmax) measured for all compounds (Table 2), but agrees with the presence
of a weaker band (tail) at lower energy. The experimental and calculated kf values are in general
in very good agreement (Table 4, columns 9 and 11).
In summary, the fusion of aromatic fragments onto the C3-C4 or C4-C5 bonds of the 2-
iminopyrrolyl ligand is an approach that enables the colour tuning in the range of blue to orange,
in the case of the more extended 2-(N-arylformimino)phenanthro[9,10-c]pyrrolyl boron diphenyl
compounds 12‒14, and green to yellow, in the case of the less extended 2-(N-
arylformimino)indolyl boron diphenyl compounds 9‒11 (Figure 7). For the latter compounds, the
emission efficiencies are considerably reduced in relation to the corresponding simple 2-(N-
arylformimino) pyrrolyl boron derivatives (15‒17), whereas in the former complexes an
important enhancement of the fluorescence efficiency is observed.
Fig. 7 Colours of complexes 9–14 in THF under UV-irradiation at 365 nm.
Electroluminescent properties
All complexes 9‒17 were tested in non-doped, single-layer light-emitting diodes (LEDs) with the
structure ITO/PEDOT:PSS/complex/cathode, where the complex films were prepared by spin
coating from a THF solution, and either LiF, calcium or barium, protected with an overlayer of
aluminum, were used as cathodes.
Results for 15‒17 were previously reported.14a,c,15 Among the new complexes, we found that
those bearing the 2-iminoindolyl ligands (9‒11) show the poorest performance. As observed in
the series 15‒17, where the binuclear complex (17) showed the best performance, devices based
on 11 performed better than those based on either 9 or 10. Yet, the performance of the OLED
based on 11 (maximum luminance of 1.2 cd m-2 and an external quantum efficiency (EQE) of
22

Page 23 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
-3
1.9×10 %) is very poor, and worse than that of a similar device based on the binuclear complex
17 (Lmax of 948 cd m-2 and EQE of 8.4×10-2 %) (see Table 5).
OLEDs based on the mononuclear complexes containing the 2-iminophenanthropyrrolyl
ligands, 12 and 13, performed better than the corresponding ones based on the simple 2-
iminopyrrolyl and the ones on 2-iminoindolyl, although exhibiting significant red-shifts in the
solid state emission, likely due to the formation of aggregates (see discussion of the X-ray
molecular structure of 13). In fact, for OLEDs based on 13, a maximum luminance of 95 cd m-2
and EQE of 0.011% was obtained with Ca cathode, increasing up to 198 cd m-2 and 0.022%,
respectively, when Ba was used instead. At variance with the performance of the binuclear boron
complexes with 2-iminopyrrolyl and 2-iminoindolyl ligands, the devices based on 14 performed
worse than the corresponding mononuclear ones (12 and 13). In summary, among the new
compounds 9‒11, and 12‒14, the one that leads to good performing OLEDs is the mononuclear
complex 13. Table 5 compares the performance of the best devices among the different cathodes
for the various compounds. Fig. 8 shows the output characteristics of a device based on 13.
When considering the solution photoluminescence (PL) quantum yields (φf), none of the
compounds with values below 0.1 give rise to reasonably performing devices. It should be noted
that the solid state PL quantum yields are usually lower than those in solution, due to
concentration quenching effects, and that the electroluminescence (EL) performance is dependent
not only on the PL quantum yields but also on the balanced charge (electrons and holes transport).
Among the complexes with higher PL quantum yield (13, 14 and 17 – see Table 4) only
devices based on 14 showed a really poor performance. As shown in Table 5, for the latter
compound, the film emission shows the highest red-shifted emission when compared to solution.
This observation points to a strong degree of aggregation of this species in the solid state, which,
combined with the rough surface films obtained by spin coating and used in the OLEDs, is likely
at the origin of that poor performance.
23

Dalton Transactions
Page 24 of 40
View Article Online
DOI: 10.1039/C6DT02771G
Table 5 Comparison of the performances of the OLEDs based on the various boron complexes
9‒17. Also included is the solution fluorescence (PL) quantum yield ( f), solution and solid state
φ
emission (PL) spectra for the most relevant compounds and, when available, the corresponding
EL spectrum.
2-Iminopyrrolyl a,b,c
2-Iminoindolyl
Complex 9 (N-2,6-iPr2C6H3)
2-Iminophenanthropyrrolyl
Complex 12 (N-2,6-iPr2C6H3) Complex 15 (N-2,6-iPr2C6H3) a
L negligible
EQE negligible
Lmax = 2 cd m-2
Lmax < 0.1 cd m-2
EQEmax = 3.4×10-3%
EQEmax = 1.92×10-6%
φf (soln.) = 0.10
φf (soln.) = 0.37
φf (soln.) = 0.023
Complex 10 (N-C6H5)
Complex 13 (N-C6H5)
Complex 16 (N-C6H5) a,b
L negligible
EQE negligible
Lmax = 198 cd m-2
Lmax= 0.35 cd m-2
EQEmax = 0.022%
EQEmax = 1.5×10-4%
φf (soln.) = 0.07
φf (soln.) = 0.61
φf (soln.) = 0.34
Complex 17 (binuclear) c
Complex 11 (binuclear)
Complex 14 (binuclear)
Lmax = 2 cd m-2
Lmax = 1 cd m-2
Lmax = 958 cd m-2
EQEmax = 1.67×10-3%
EQEmax = 5.4×10-4%
EQEmax = 0.084%
φf (soln.) = 0.06
φf (soln.) = 0.58
φf (soln.) = 0.69
a Ref. 14c. b Ref. 14a. c Ref. 15.
24

Page 25 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
Fig. 8 Current (I) and luminance (L) as a function of the bias applied to a
ITO/PEDOT:PSS/13/Ba/Al OLED. The inset shows the corresponding EQE.
Conclusion
In this work, a third approach for the colour tuning of 2-(N-arylformimino)pyrrolyl boron
diphenyl fluorophores was used, consisting in the extension of the chromophore π-conjugation
through the fusion of aromatic fragments onto the 2-iminopyrrolyl C3-C4 or C4-C5 bonds (i.e. c
or d bonds). This strategy, applied together with previously described methods, i.e., (a) the
variation of the electronic and steric nature of the N-phenyl substituents (structures D in Chart
1),14c and (b) the use of variable length aromatic spacers bridging binuclear 2-iminopyrrolyl
boron fragments (structures E in Chart 1),15 provided a wide range of emission colours, from blue
to orange.
Brightly fluorescent materials were obtained in the case of the more extended new 2-(N-
arylformimino)phenanthro[9,10-c]pyrrolyl boron derivatives 12‒14, which are considerably more
luminescent than their parent simple 2-iminopyrrolyl derivatives (15‒17). The phenanthrene
fused iminopyrrolyl complexes showed good fluorescence quantum yields (φf = 0.37 for 12, 0.61
for 13 and 0.58 for 14), which were based on the high S1→S0 (LUMO–HOMO) transition dipole
moment. On the contrary, the new benzene ring d-fused 2-iminopyrrolyl (2-(N-
arylformimino)indolyl) derivatives 9‒11, despite exhibiting emission colours between green and
yellow, displayed poor quantum yields (0.10 for 9, 0.07 for 10 and 0.06 for 11), much lower than
those of the corresponding simple derivatives 15‒17. This can be explained by the 10-fold lower
25

Dalton Transactions
Page 26 of 40
View Article Online
DOI: 10.1039/C6DT02771G
kf values observed for 9‒11, indicating a much smaller S1→S0 (LUMO–HOMO) transition dipole
moment for these compounds. Therefore, the type of fusion (on the c or d bonds of 2-
iminopyrrolyl) influences very much the emissivity of each type of compounds, the fusion on the
C3-C4 (c-fusion) being highly beneficial for the luminescence properties of 2-iminopyrrolyl
chromophores, whereas that on the C4-C5 (d-fusion) is highly detrimental.
Non-doped single-layer electroluminescent devices were fabricated using these compounds
as both emitter and ambipolar charge-transporting materials. In general, the complexes with the
highest solution fluorescence quantum yield, gave rise to OLEDs with reasonable performances.
Among the new complexes, OLEDs based on 13 (
cd m-2, with an efficiency (EQE) of 0.022%, although exhibiting an emission in the solid state
significantly shifted to the red. Conversely, devices based on complex 14 ( f = 0.58) show a
φf = 0.61) show a maximum luminance of 198
φ
disappointing performance, which may attributed to strong aggregation. This suggests that the
use of these complexes in doped OLEDs is likely to lead to strong performance improvements.
Experimental Section
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 Air Liquide, and purified by passage through 4 Ǻ molecular sieves. Unless
otherwise stated, all reagents were purchased from commercial suppliers (e.g., Acrös, Aldrich,
Fluka, Alfa Aesar) 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 dichloromethane). Solvents and solutions were
transferred using a positive pressure of nitrogen through stainless steel cannulae and mixtures
were filtered in a similar way using modified cannulae that could be fitted with glass fibre filter
disks.
Nuclear magnetic resonance (NMR) spectra were recorded on Bruker Avance III 300 (1H,
13C, 11B, and 19F) spectrometer. Deuterated solvents were dried by storage over 4 Ǻ molecular
sieves and degassed by the freeze-pump-thaw method. Spectra were referenced internally using
the residual protio solvent resonance relative to tetramethylsilane (δ=0). All chemical shifts are
26

Page 27 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
quoted in δ (ppm) and coupling constants given in hertz. Multiplicities were abbreviated as
follows: broad (br), singlet (s), doublet (d), triplet (t), quartet (q), heptet (h), and multiplet (m).
For air- and/or moisture sensitive materials, samples were prepared in J. Young NMR tubes in a
glovebox. Elemental analyses were obtained from the IST elemental analysis services.
The ligand precursors, HNC16H9C(H)=O (2),16 HNC8H5C(H)=N-C6H3-2,6-iPr2 (3),12
HNC8H5C(H)=N-C6H5 (4),12 HNC16H9C(H)=N-C6H3-2,6-iPr2 (6),13 and HNC16H9C(H)=N-C6H5
(7),13 were synthesised according to literature procedures.
Synthesis of (HNC8H5C(H)=N)2-1,4-C6H4 (5): Two equivalents of 2-formylindole (0.500 g,
3.40 mmol) and one equivalent of 1,4-phenylenediamine (0.186 g, 1.70 mmol) and a catalytic
amount of p-toluenesulfonic acid were suspended in 20 mL of absolute ethanol in a round-bottom
flask, fitted with a condenser and a CaCl2 guard tube. The mixture was heated to reflux overnight
turning to a brown precipitated solution. The mixture was allowed to cool and filtered off,
1
washed several times with dichloromethane and dried under vacuum. Yield: 0.418 g (67%). H
NMR (300 MHz, CDCl3): δ 10.74 (s, 2H, NH), 7.76 (s, 2H, CH=N), 6.77 (d, JHH = 8.1 Hz, 2H,
Ind-benz), 6.65 (d, JHH = 8.1 Hz, 2H, Ind-benz), 6.52 (s, 4H, aryl), 6.37 (t, JHH = 7.2 Hz, 2H, Ind-
benz), 6.20 (t, JHH = 7.5 Hz, 2H, Ind-benz), 6.16 (s, 2H, Ind-pyrr). Anal. Calcd (%) for
C24H18N4 0.125CH2Cl2: C, 77.67; H, 4.93; N, 15.01. Found: C, 77.71; H, 5.14; N, 15.05.
Synthesis of (HNC16H9C(H)=N)2-1,4-C6H4 (8): In the same manner as described above, a
mixture of phenanthro[9,10-c]pyrrole-1-carboxaldehyde (0.600 g, 2.41 mmol) and 1,4-phenylene
diamine (0.132 g, 1.20 mmol) afforded 8 as a dark red solid. Yield: 0.425 g (62%). 1H NMR (300
MHz, DMSO-d6): δ 9.25 (s, 2H, CH=N), 8.90 (d, JHH = 7.8 Hz, 2H, Phen), 8.70 (d, JHH = 7.8 Hz,
2H, Phen), 8.63 (d, JHH = 7.5 Hz, 2H, Phen), 8.33 (d, JHH = 7.2 Hz, 2H, Phen), 8.24 (s, 2H, Phen-
pyrr), 7.66-7.47 (m, 7H, Phen + Aryl + NH). 13C{1H} NMR (75 MHz, CD2Cl2): δ 149.4, 149.1,
137.6, 129.8, 128.2, 128.1, 127.7, 127.4, 126.4, 126.1, 125.5, 124.9, 123.9, 123.6, 123.4, 122.1,
121.4, 121.0, 117.8, 114.4. Anal. Calcd (%) for C40H26N4 1.25CH2Cl2: C, 74.07; H, 4.29; N, 8.37.
Found: C, 74.41; H, 4.41; N, 8.45.
General Procedure for the Syntheses of Mono- and Binuclear Organoboron Complexes
(9−14): In a typical experiment, one or two equivalents of triphenyl boron and one equivalent of
the desired mono- or bis-iminopyrrolyl derived ligand precursors in 15−25 mL of toluene were
heated to reflux overnight, under nitrogen atmosphere. The reaction mixture was brought to room
27

Dalton Transactions
Page 28 of 40
View Article Online
DOI: 10.1039/C6DT02771G
temperature and all the volatiles were removed under vacuum. The resulting solid was extracted
with 5−10 mL of dichloromethane and double layered with n-hexane and the solution was kept at
-20 °C to afford the corresponding boron complexes. In the case of binuclear boron complexes 11
and 14, due to their poor solubility, after removing all the volatiles, the compounds were washed
with dichloromethane, dried and used as such for further analysis. Except for the compounds 9,
10 and 13, all other compounds were obtained as amorphous powders. Further crystallisations
were attempted, including double layering with THF/n-hexane and toluene/n-hexane.
Synthesis of Ph2B[NC8H5C(H)=N-C6H3-2,6-iPr2] (9): In the same manner as described above,
a mixture of 2 (0.123 g, 0.50 mmol) and BPh3 (0.121 g, 0.50 mmol) afforded 9 as a yellow solid.
Crystals suitable for single crystal X-ray diffraction studies were obtained by double layering of
1
the compound with dichloromethane and n-hexane. Yield: 0.183 g (78%). H NMR (300 MHz,
CD2Cl2): δ 8.43 (s, 1H, CH=N), 7.87 (d, JHH = 7.3 Hz, 1H, Ind-benz), 7.42 (s, 1H, Ind-pyrr), 7.32
(t, JHH = 7.7 Hz, 1H, Ph) 7.19 - 7.04 (m, 14H, BPh2 + Ind-benz + Ph), 6.96 (d, JHH = 8.0 Hz, 1H,
Ind-benz), 2.32 (sep, JHH = 6.8 Hz, 2H, CH-(CH3)2), 0.89 (d, JHH = 6.9 Hz, 6H, CH-(CH3)2), 0.32
(d, JHH = 6.9 Hz, 6H, CH-(CH3)2). 13C{1H} NMR (75 MHz, CD2Cl2): δ 159.8, 144.9, 142.1,
138.7, 138.4, 138.3, 134.1, 133.2, 129.2, 127.6, 127.0, 126.1, 124.3, 123.9, 120.9, 115.4, 106.5,
29.4, 26.5, 21.4. 11B NMR (96.29 MHz, CD2Cl2): δ 6.99. Anal. Calcd (%) for C33H33BN2: C,
84.61; H, 7.10; N, 5.98. Found: C, 84.22; H, 7.20; N, 5.60.
Synthesis of Ph2B[NC8H5C(H)=N-C6H5] (10): In the same manner as described above, a
mixture of 3 (0.110 g, 0.50 mmol) and BPh3 (0.121 g (0.50 mmol) afforded 10 as a yellow solid.
Crystals suitable for single crystal X-ray diffraction studies were obtained by double layering of
1
the compound with toluene and hexane. Yield: 0.158 g (82%). H NMR (300 MHz, CD2Cl2): δ
8.77 (s, 1H, CH=N), 7.79 (d, JHH = 7.9 Hz, 1H, Ind-benz), 7.36-7.28 (m, 10H, Ph + BPh2 + Ind-
pyrr), 7.22-7.20 (m, 6H, BPh2), 7.14- 7.04 (m, 3H, Ind-benz). 13C{1H} NMR (75 MHz, CD2Cl2):
δ 154.8, 142.3, 141.5, 138.7, 138.3, 134.0, 133.0, 129.8, 129.1, 127.9, 127.2, 126.2, 124.0, 123.4,
120.8, 114.5, 107.0. 11B NMR (96.29 MHz, CD2Cl2): δ 5.83. Anal. Calcd (%) for
C27H21BN2 0.25CH2Cl2: C, 80.71; H, 5.34; N, 6.91. Found: C, 80.52; H, 5.40; N, 7.42.
Synthesis of Ph2B[(NC8H5C(H)=N)2-1,4-C6H4]BPh2 (11): In the same manner as described
above, a mixture of 4 (0.154 g, 0.42 mmol) and BPh3 (0.205 g, 0.85 mmol) afforded 11 as a
1
reddish brown solid. Yield: 0.170 (58%). H NMR (300 MHz, CD2Cl2): δ 8.71 (s, 2H, CH=N),
28

Page 29 of 40
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT02771G
7.77 (d, JHH = 9.0 Hz, 1H, Ind-benz), 7.37 (s, 1H, Ind-pyrr), 7.26-7.23 (m, 6H, BPh2), 7.20-7.17
(m, 6H, BPh2 + Ph), 7.15-7.02 (m, 3H, Ind-benz), 13C{1H} NMR (75 MHz, CD2Cl2): δ 154.6,
144.3 (br), 142.0, 141.9, 138.2, 133.9, 133.2, 128.0, 127.3, 126.7, 124.2, 124.1, 121.1, 114.6,
107.9. 11B NMR (96.29 MHz, CD2Cl2): δ 5.64. Anal. Calcd (%) for C48H36B2N4: C, 83.50; H,
5.26; N, 8.11. Found: C, 83.72; H, 5.80; N, 8.28.
Synthesis of Ph2B[NC16H9C(H)=N-C6H3-2,6-iPr2] (12): In the same manner as described
above, a mixture of 6 (0.200 g, 0.50 mmol) and BPh3 (0.119 g, 0.50 mmol) afforded 12 as a
1
yellow solid. Yield: 0.176 g (62%). H NMR (300 MHz, CD2Cl2): δ 8.73-8.66 (m, 2H, Phen),
8.62 (s, 1H, CH=N), 8.26-8.24 (m, 2H, Phen), 8.20 (d, JHH = 0.9 Hz, 1H, Phen-pyrr), 7.69-7.58
(m, 4H, Ph + Phen), 7.36-7.30 (m, 1H, Phen), 7.17-7.12 (m, 12H, BPh2 + Phen), 2.46 (h, JHH
=
6.9 Hz, 2H, CH-(CH3)2), 0.95 (d, JHH = 6.9 Hz, 6H, CH-(CH3)2), 0.36 (d, JHH = 6.6 Hz, 6H, CH-
(CH3)2). 13C{1H} NMR (75 MHz, CD2Cl2): δ 154.9, 146.1, 145.2 (br), 138.9, 133.5, 130.6, 129.6,
128.9, 128.8, 128.5, 128.1, 128.0, 127.8, 127.7, 127.1, 127.0, 126.1, 125.1, 124.6, 124.3, 124.0,
123.9, 123.6, 29.3, 26.6, 21.6 (two carbons missing or superimposed on other resonances). 11B
NMR (96.29 MHz, CD2Cl2): δ 5.47. Anal. Calcd (%) for C41H37BN2 0.5CH2Cl2 C, 81.58; H,
6.27; N, 4.58. Found: C, 81.72; H, 6.22; N, 4.54.
Synthesis of Ph2B[NC16H9C(H)=N-C6H5] (13): In the same manner as described above, a
mixture of 7 (0.264 g, 0.82 mmol) and BPh3 (0.200 g, 0.82 mmol) afforded 13 as a yellow solid.
Crystals suitable for single crystal X-ray diffraction studies were obtained by double layering of
1
the compound with toluene and hexane. Yield: 0.291 g (71%). H NMR (300 MHz, CD2Cl2): δ
9.11 (s, 1H, CH=N), 8.70-8.60 (m, 2H, Phen), 8.41-8.38 (m, 1H, Phen), 8.14 (d, JHH = 0.9 Hz,
1H, Phen-pyrr), 8.13-8.09 (m, 1H, Phen), 7.70-7.66 (m, 2H, Ph), 7.58-7.55 (m, 2H, Phen), 7.44-
7.20 (m, 15H, Phen, Ph, BPh2). 13C{1H} NMR (75 MHz, CD2Cl2): δ 149.0, 142.8, 133.7, 130.6,
129.8, 129.7, 128.9, 128.3, 128.2, 128.1, 127.9, 127.7, 127.6, 127.4, 127.3, 127.26, 127.2, 126.2,
125.2, 124.5, 124.4, 124.0, 123.7, 122.3 (one carbon missing or superimposed on other
resonances). 11B NMR (96.29 MHz, CD2Cl2): δ 5.08. Anal. Calcd (%) for C35H25BN2∙0.5CH2Cl2
C, 80.92; H, 4.97; N, 5.31. Found: C, 81.02; H, 5.00; N, 5.29.
Synthesis of Ph2B[(NC16H9C(H)=N)2-1,4-C6H4]BPh2 (14): In the same manner as described
above, a mixture of 8 (0.200 g, 0.35 mmol) and BPh3 (0.172 g, 0.71 mmol) afforded 14 as a
reddish brown solid. Yield: 0.167 g (53%). Owing to the low solubility of this compound in
29