Revisiting phosphorus analogues of phthalimides and naphthalimides: Syntheses and comparative studies

Article abstract of DOI:10.1002/chem.201405494

A series of phosphorus analogues of aromaticfused monoimides (phthalimides and naphthalimides) bearing a mesityl group on the P center have been synthesized. In a comparison of their photophysical, electrochemical, and thermal properties with those of the corresponding imides, the impact of P incorporation was revealed. Furthermore, theoretical studies using DFT methods were conducted to understand their properties.

Full text of DOI:10.1002/chem.201405494

DOI: 10.1002/chem.201405494
Full Paper
&
Phosphorus Heterocycles
Revisiting Phosphorus Analogues of Phthalimides and
Naphthalimides: Syntheses and Comparative Studies
Youhei Takeda,*[a, b] Takuya Nishida,[b] Kota Hatanaka,[b] and Satoshi Minakata*[b]
Abstract: A series of phosphorus analogues of aromatic-
fused monoimides (phthalimides and naphthalimides) bear-
ing a mesityl group on the P center have been synthesized.
In a comparison of their photophysical, electrochemical, and
thermal properties with those of the corresponding imides,
the impact of P incorporation was revealed. Furthermore,
theoretical studies using DFT methods were conducted to
understand their properties.
Introduction
The exploration of new scaffolds of p-conjugated organic com-
pounds offers great opportunities for creating novel organic
functional materials. In this regard, recently, organophosphorus
p-conjugated systems, exemplified with phospholes,[1] have in-
creasingly been attracting attention[2] because of their unique
physicochemical properties, which are induced by the incorpo-
ration of phosphorus functionalities into carbonic p-conjugat-
ed frameworks. Specifically, the capability of s*–p* hyperconju-
gation to stabilize the LUMO and upper vacant orbital ener-
gies, and the multivalency of P centers, which allows fine-
tuning of their electronic structures, endow organophosphorus
conjugated molecules with notable features such as prominent
electron-accepting and luminescent properties. In light of the
huge array of organophosphorus compounds that have been
diligently created by organic chemists over the past 170 years,
revisiting the poorly explored organophosphorus p-conjugated
scaffolds would be a fascinating and valuable research task
from the viewpoint of materials science.
Scheme 1. A representative array of reported diketophosphanyl compounds.
their physicochemical properties oriented toward organic elec-
tronics applications had not been reported until 2013, when
Baumgartner and co-workers shed light on the unique optoe-
lectronic properties of bithiophene-fused seven-membered di-
ketophosphanyl compounds dithieno[3,2-c:2’,3’-e]-2,7-diketo-
phosphepin (DTDKPs) (Scheme 1).[4] Notably, DTDKPs feature
high electron affinities (EAs) that are comparable to that of
a privileged fullerene-based material PC61BM[5] and also show
unique dual-emission properties. More recently, we have dis-
closed the synthesis, dynamic molecular behavior, and physico-
chemical properties of phosphorus analogues of pyromellitic
diimides for the first time (Scheme 1),[6,7] which represent an n-
channel (electron-transporting) semiconducting material with
the minimal p-conjugated core in the rylene diimides family.[8]
Importantly, the novel bis(diketophosphanyl) compounds were
found to possess excellent electron-accepting abilities that
were superior to that of the corresponding diimide. Further-
more, theoretical calculations revealed that s*–p* hyperconju-
gation between the exocyclic CꢀP s* fragment and the p*
moiety developing over the benzene core and two carbonyl
groups works effectively in the LUMO+1 molecular orbitals of
Aromatic-fused diketophosphanyl compounds, which can be
regarded as phosphorus analogues of aromatic-fused imide
compounds such as phthalimides and perylene imides, consti-
tute a poorly explored p-conjugated organophosphorus scaf-
fold. Although several reports on the synthesis of such com-
pounds are available (see Scheme 1),[3] surprisingly, studies on
[a] Dr. Y. Takeda
Frontier Research Base for Global Young Researchers
Graduate School of Engineering, Osaka University
Yamadaoka 2-1, Suita, Osaka 565-0871 (Japan)
E-mail: takeda@chem.eng.osaka-u.ac.jp
[b] Dr. Y. Takeda, T. Nishida, K. Hatanaka, Prof. Dr. S. Minakata
Department of Applied Chemistry
Graduate School of Engineering
Osaka University
Yamadaoka 2-1, Suita
Osaka 565-0871 (Japan)
E-mail: minakata@chem.eng.osaka-u.ac.jp
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405494.
Chem. Eur. J. 2014, 20, 1 – 8
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
those phosphorus compounds rather than the LUMOs to stabi-
lize the orbital energies. Through our previous investigation,
we became intrigued by much simpler aromatic-fused diketo-
phosphanyl systems such as P surrogates of phthalimides and
naphthalimides as models for understanding the structure–
property relationships that underlie such compounds. Al-
though the synthesis and characterization data (MS, IR, and
NMR) of phosphorus analogues of phthalimide and 1,8-naph-
thalimide were reported by Cowley in 1987 (Scheme 1),[3a] to
our knowledge, their physicochemical properties have not
been investigated.
Herein, we disclose the syntheses, characterization, structure,
and physicochemical properties (UV/Vis absorption and lumi-
nescence profiles, thermostabilities, and electrochemical be-
haviors) of phosphorus analogues of phthalimide and naphtha-
limides P1–P4 (Scheme 2) as well as their P-functionalized de-
rivatives P4-O and P1-AuCl, in comparison with the corre-
sponding imide compounds N1–N4. Furthermore, theoretical
Scheme 3. Synthetic routes to aromatic-fused diketophosphanyl com-
pounds.
sulting diketophosphanyl products against air and
moisture. Compound P1 in CDCl3 showed a 31P{1H}
NMR signal as a sharp singlet at d=ꢀ25.1 ppm
against the resonance corresponding to H3PO4 as an
external standard (d=0 ppm), which is in good
agreement with the value for P phenyl derivative
(d=ꢀ28.0 ppm) reported by Cowley.[3a] Other spec-
troscopic analyses (1H and 13C NMR, IR, and MS) as
well as elemental analysis also confirmed the produc-
tion of P1 (for detailed data, see the Supporting In-
formation). With the aim of enhancing the electron-
accepting ability of P1, tetrafluorinated diketophos-
phanyl compound P2 was prepared in a similar fash-
ion (Scheme 1). In contrast, attempts to prepare P3
through the condensation of naphthalene 2,3-dicar-
bonyl dichloride with MesP(SiMe3)2 in Et2O failed,
probably due to low solubility of the dichloride in
ether. After modification studies, we found that the
target molecule P3 was obtained when condensation was per-
formed in N,N-dimethylformamide (DMF) heated to reflux,
albeit in an extremely low yield (Scheme 1). However, six-mem-
bered cyclic diketophosphanyl compound P4, which is a consti-
tutional isomer of P3, was prepared through condensation of
1,8-naphthalic anhydride with MesP(SiMe3)2 in a similar proce-
dure to that reported by Cowley.[3a] Importantly, single crystals
of P4 that were suitable for X-ray crystallographic analysis
were obtained, thereby the molecular structure of six-mem-
bered diketophosphanyl compound was clarified for the first
time (see below). The condensation of MesP(SiMe3)2 with 1,8-
dicarbonyl dichloride instead of the corresponding anhydride
did not produce P4 at all. The five-membered diketophosphan-
yl compounds P1–P3 were not obtained from the correspond-
ing five-membered dicarboxylic anhydrides, and only gave
complex mixtures; the reasons why these combinations did
not work are not clear at this point. To investigate the impact
of the P functionalization of PIII compounds P1–P4 on their
physicochemical properties, chemical modifications were con-
Scheme 2. Diketophosphanyl and imide compounds used in this study.
calculations were also conducted to understand their
properties.
Results and Discussion
Synthesis and characterization
The synthetic routes to P1–P4 and the chemical modification
of the P center leading to P4-O and P1-AuCl are illustrated in
Scheme 3 and Equation (1). According to a similar experimen-
tal procedure reported by Cowley[3a] and Baumgartner,[4] the
phosphorus analogue of phthalimide P1 was synthesized in
70% yield through the condensation of an equimolar amount
of phthaloyl dichloride and P,P-bis(trimethylsilyl)mesitylphos-
phane [MesP(SiMe3)2] in Et2O with the reaction temperature
being elevated slowly from ꢀ788C to room temperature over
18 h (Scheme 1). The choice of mesityl group on the P center
was made by taking into account the kinetic and thermody-
namic stabilities of both of disilylphosphane reagent and re-
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
ducted. Oxidation of P4 with mCPBA gave pentavalent phos-
phorus compound P4-O in good yield (Scheme 1), whereas
similar treatment of P1–P3 resulted in complex mixtures with
the corresponding phosphane oxides not being formed. The
Baumgartner group demonstrated that complexation of a bi-
thiophene-fused diketophosphanyl compound with AuI drasti-
cally stabilizes the LUMO orbital energy, and thereby the elec-
tron-accepting ability was enhanced.[4a] Upon mixing P1 with
an equimolar amount of Au(tht)Cl [tht=tetrahydrothiophene]
in CH2Cl2 at room temperature, yellow solids started to precipi-
tate. The collected solids were washed with Et2O and dried in
vacuo to give P1-AuCl in 52% yield [Eq. (1)]. The complexation
1
resulted in significant downfield-shift of the H NMR resonan-
ces corresponding to the CH3 (on the 2,6-positions of Mes,
Dd= +0.30 ppm) and Ar-H (on the 3,5-positions of Mes, Dd=
+0.11 ppm), indicating a marked decrease in electron density
on the P center. Reference compounds N1–N4 (Scheme 2)
were readily prepared by classical condensation of the corre-
sponding aromatic carboxylic anhydrides with MesNH2 (for de-
tails, see the Supporting Information).
Figure 1. a) Molecular structure of P4 in the solid state. Thermal ellipsoids
are set at the 50% probability level, and hydrogen atoms are omitted for
clarity. Selected bond lengths (ꢁ) and angles (degree): P1ꢀC1, 1.820(2); P1ꢀ
C10, 1.835(4); P1ꢀC20, 1.839(3); C10ꢀC11, 1.487(3); C19ꢀC20, 1.487(4); O1ꢀ
C10, 1.228(4); O2ꢀC20, 1.221(4); C11ꢀC12, 1.389(5); C12ꢀC13; 1.395(5); C13ꢀ
C14, 1.374(5); C14ꢀC15, 1.413(5); C15ꢀC21, 1.429(3); C15ꢀC16, 1.415(5);
C16ꢀC17, 1.368(6); C17ꢀC18, 1.404(5); C18ꢀC19, 1.385(4); C19ꢀC21,
1.427(5); C10-P1-C20, 99.8(1); C1-P1-C10, 107.0(1); C1-P1-C20, 105.5(1); C19-
C20-P1, 117.1(2); C11-C10-P1, 117.4(2); C10-C11-C21, 122.8(3); C20-C19-C21,
123.2(3); C11-C21-C19, 124.7(3). b) A dimeric structure of P4 found in the
packing structure.
Crystal structure of P4
A yellow single crystal of P4 that was suitable for X-ray diffrac-
tion analysis was obtained by recrystallization from its CH2Cl2/
hexane biphasic solution. Although the molecular structures of
five-,[6,9] seven-,[4a] and twelve-membered diketophosphanyl
compounds[3e] clarified by X-ray crystallographic analysis, have
been reported so far, to our knowledge, this is the first report
on the structural analysis of a six-membered cyclic diketophos-
phanyl compound (Figure 1a).[10] The phosphorus center (P1 in
Figure 1a) adopts a pyramidal structure as expected, with the
deviation from the mean plane of the naphthalene unit being
approximately 0.71 ꢁ and the interplane angle between C10-
P1-C20 and C10-C11-C21-C19-C20 (mean plane) being approxi-
mately 398. In comparison with reported five-[6,9] and seven-
membered[4a] diketophosphanyl compounds, the deviations of
the PIII-center from the mean plane of dicarbonyl-fused aro-
matic ring and the interplane angles (calculated in a similar
way to that for P4) were found to decrease in the order of ring
size as follows: seven- (ca. 0.79 ꢁ, ca. 488)[4a] > six- (ca. 0.71 ꢁ,
ca. 398, P4)> five-membered compound (ca. 0.38 ꢁ, ca. 178).[9]
In conjunction with the fact that bond lengths of the exocyclic
PꢀC bond (P1ꢀC1 in Figure 1a, ca. 1.82 ꢁ) and two endocyclic
PꢀC bonds (P1ꢀC10 and P1ꢀC20 in Figure 1a, ca. 1.84 ꢁ for
both) have almost the same values as those previously repor-
ted,[4a,6,9] the order of P deviations and interplane angles men-
tioned above would be derived from the variation in ring
strain rather than from the degree of p conjugation. Further-
more, a unique intermolecular dimeric motif that is comprised
of the adjacent two molecules was found in the packing struc-
ture, with the two naphthalene planes being stacked face-to-
face (Figure 1b).
Photophysical properties
Physicochemical properties of diketophosphanyl compounds
P1–P4, P1-AuCl, and P4-O, as well as those of reference imide
compounds N1–N4 for comparison, are summarized in Table 1.
UV/Vis absorption and photoluminescence (PL) spectra of di-
luted CH2Cl2 solutions (10ꢀ5 m order) of all compounds are illus-
trated in Figure 2. Compared with the simplest imide N1, the
absorption spectrum of its P analogue P1 exhibited a redshift
throughout the entire region, featuring a strong absorption at
labs =239 nm (e 4.32ꢂ104 mꢀ1 cmꢀ1), which is assignable to the
HOMO-6!LUMO+1 transition by TD-DFT calculation [B3LYP/
6-31G(d)//B3LYP/6-31G(d) level of theory (for details, see the
Supporting Information)], a shouldered absorption at around
labs =253 nm (e 2.25ꢂ104 mꢀ1 cmꢀ1, HOMO!LUMO+2 transi-
tion), and a weak absorption at labs =296 nm (e 0.49ꢂ
104 mꢀ1 cmꢀ1, HOMO-3!LUMO transition) (Figure 2a). In addi-
tion, a very weak and broadened absorption was observed in
the lower energy region (350–420 nm) (see the enlarged spec-
trum in the Supporting Information), which can be ascribed by
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
terestingly, a frozen solution of P3 in a mixed EET sol-
vent [Et2O/EtOH/toluene (v/v 2:1:1)] at 77 K emitted
bright-yellow light that showed vibrational fine struc-
tures in the PL spectrum [lem =518, 536, and 563 nm,
t1 =40 ns (48%), t2 =7.4 ns (52%), Figure 2b], where-
as no luminescence was observed from P3 at room
temperature (298 K). In contrast, a CH2Cl2 solution of
reference compound N3 showed faintly blue emis-
sion [lem =368 and 386 nm, Ff =0.03, t1 =1.4 ns
(100%), Figure 2b] at room temperature (298 K),
which is known as “short-wavelength (SW) emission”
from its excited S1 to the S0 state of 2,3-naphthali-
mides.[11] Judging from the short lifetime of P3 at
77 K, the observed yellow luminescence would be
fluorescence, although the Stokes shift of P3 is signif-
icantly larger (Dl =3700 cmꢀ1) compared with that
of N3 (Dl =700 cmꢀ1). The reason for this discrepan-
cy between these Stokes shifts requires further inves-
tigation. A constitutional isomer of P3, that is P4, ex-
hibited a sharp absorption ascribed to p–p* transi-
tion (HOMO-6!LUMO+1, TD-DFT) at labs =235 nm
in the higher-energy regime (200–250 nm) and
a broad absorption in the lower-energy region (250–
400 nm, Figure 2c). The oxidation of the PIII-center of
P4 leading to P4-O caused a slight decrease in the
extinction coefficient (e) in the higher-energy region
(250–400 nm) as well as a raise of a shouldered peak
in the lower-energy regime (380–400 nm, Figure 2c).
A similar behavior in the UV/Vis absorption spectra
has been reported with a seven-membered diketo-
phosphanyl compound by Baumgartner and co-work-
ers.[4a] Notably, both of P4 and P4-O exhibited yellow
emission at 77 K, as observed for P3, featuring short
lifetimes [P4: lem =532 and 577, t1 =0.63 ns (12%),
t2 =5.1 ns (88%); P4-O: lem =523, 538, and 566, t1 =
12 ns (78%), t2 =3.3 ns (22%), Figure 2c], which were
not observed at room temperature (298 K).[12]
Table 1. Summary of physicochemical properties of phosphorus analogues of aromat-
ic monoimides and reference imide compounds.
red
Compd labs [nm][a]
(e/104
lem [nm]
E
[V][b]
ELUMO
[eV][c]
Td (5 wt%)
under N2 and air
[8C]
1/2
[mꢀ1 cmꢀ1])
P1
239 (4.32),
253 (br, 2.25),
296 (0.49)
242 (3.08),
318 (0.21),
370 (0.12)
236 (4.05),
249 (br, 2.24),
297 (0.62)
279 (6.64),
331 (0.76)
235 (6.24),
321 (1.43)
237 (5.29),
335 (1.12)
226 (3.51),
237 (1.73),
294 (0.20)
224 (2.34),
240 (1.41),
300 (0.38)
261 (6.54),
292 (0.84),
358 (0.46)
234 (5.17),
334 (1.50),
349 (1.28)
ND
ND
ND
ꢀ1.79
ꢀ3.01
216, 198
168, 161
199, 195
red1
P1-
AuCl
E
E
E
ꢀ1.40, ꢀ3.40
ꢀ1.64,
1/2
red2
1/2
red3
ꢀ1.85
1/2
P2
ꢀ1.50
ꢀ3.30
red1
P3
518, 536,
563[d]
532, 577[d]
E
–
176, 175
226, 225
253, 245
199,193
onset
ꢀ1.81[e]
P4
ꢀ1.84
ꢀ2.96
ꢀ3.31
ꢀ2.84
P4-O
N1
523, 538,
566[d]
ND
ꢀ1.49
ꢀ1.96
N2
N3
N4
ND
ꢀ1.57
ꢀ3.23
–
166, 164
249, 248
246, 231
368, 386[e]
ND
E
red1
onset
ꢀ1.99[f]
ꢀ1.85
ꢀ2.95
[a] Absorption maxima of dilute CH2Cl2 solutions (10ꢀ5 m order). [b] red
E1/2 =(Eap +Ecp)/2
versus the Fc/Fc+ redox couple measured as CH2Cl2 solutions using 0.1m nBu4N·PF6 as
red1
the supporting electrolyte. [c] ELUMO =ꢀ(4.8+
E ) [eV]. [d] PL spectra were mea-
1/2
sured at 77 K with EET [Et2O/EtOH/toluene (v/v 2:2:1)] solutions (10ꢀ5 m); Excited at
310 nm. [e] PL spectra were measured at room temperature with a dilute CH2Cl2 solu-
tion (10ꢀ5 m); A quantum yield was determined to be 0.03 by an integrating sphere.
red
[f] Observed as an electrochemically irreversible process;
E
indicates the onset
onset
potential versus the Fc/Fc+ couple that corresponds to the first reduction process.
ND=not detected.
TD-DFT to n-p* transition from the lone pair of the P center
(HOMO-1) to the largely extended p* moiety over the ben-
Electrochemical behavior
zene–diketophosphanyl core (LUMO) (see the Supporting Infor-
mation). The incorporation of four fluorine atoms onto the
diketo-fused phenyl ring (P2) resulted in a slight blueshift of
absorptions in the higher-energy regime (230–280 nm) when
compared with P1, whereas the spectrum resembles that of P1
in appearance (Figure 2a). Fluorinated compound P2 also ex-
hibited an absorption tail ascribed to n-p* (HOMO-2!LUMO)
transition in the energy-low region (350–420 nm, see the en-
larged spectrum in the Supporting Information) as observed
for P1. Compared with P1 and P2, the complexation of P1
with Au(tht)Cl at the P center gave a more clearly recognizable
absorption in the region of 350–400 nm, which is assignable to
n-p* transition from the lone pair of Cl (HOMO-1) to the ben-
zene-diketophosphanyl p* moiety. In contrast, 2,3-naphthalene
diketophosphanyl compound P3 showed a significant redshift
To access the impact of the P substitution of imide compounds
on their electron-accepting abilities, the electrochemical be-
havior of CH2Cl2 solutions (10ꢀ4 m order) of the synthesized
compounds was investigated by cyclic voltammetry (CV,
Table 1). Cyclic voltammograms of N1, P1, P2, and P1-AuCl are
illustrated in Figure 3 as representative examples (for full de-
tails, see the Supporting Information). Phthalimide N1 showed
a reversible redox couple at redE1/2 =ꢀ1.74 V (vs Ag/Ag+; the
values in Table 1 are corrected using the Fc/Fc+ redox pair that
was added to the sample solutions as an internal standard), in-
dicating the good electron-accepting nature of N1 (Figure 3).
Notably, substitution of the N atom of N1 with a phosphorus
(P1) nicely demonstrated a striking positive-shift of the half-
wave potential redE1/2 =ꢀ1.58 V (vs Ag/Ag+), supporting the ef-
fectiveness of the P-substitution strategy for designing diketo-
phosphany-based electron-acceptors. Moreover, the introduc-
tion of fluorine atoms onto the benzene core of P1 (P2) result-
of a sharp absorption corresponding to p–p* transition (labs
=
279 nm, HOMO-1!LUMO+2 transition), compared with that
of reference compound N3 (labs =261 nm, Figure 2b). More in-
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 3. Cyclic voltammograms of N1, P1, P2, and P1-AuCl measured as
CH2Cl2 solutions (10ꢀ4 m order) with 0.1m nBu4N·PF6 as the supporting elec-
trolyte. Scan rate=100 mV for P1, P2, and P1-AuCl; 50 mV for N1. Ferrocene
was added as an internal standard.
presumably derived from the complexation with gold. Accord-
ing to the DFT calculations, the complexation with Au (P1-
AuCl) significantly stabilizes the LUMO and upper vacant orbi-
tals, even with the LUMO+1 being much lower than the
LUMOs of other compounds (see below), probably explaining
this unique ability of P1-AuCl to accept multiple electrons. It
should be noted that the LUMO energy levels [ELUMO =ꢀ(4.8+
red1
E
) eV; red1E1/2 =(Eap1 +Ecp1)/2 versus the Fc/Fc+ redox
1/2
couple] of P2 and P1-AuCl (ꢀ3.30 and ꢀ3.40 eV, respectively)
are close to those of privileged electron-acceptors, pyromellitic
diimides (ca. 3.5 eV).[8] Regarding the 2,3-fused naphthalimide
N3 and its P analogue P3, a similar trend that the reduction
potential of phosphorus compound (P3) is slightly more posi-
tive than that of imide (N3) was observed, albeit with irreversi-
ble reduction processes for both compounds (for the detailed
voltammograms, see the Supporting Information). In contrast,
the phosphorus analogue of 1,8-naphthalimide (P4) showed
almost the same redox potential corresponding to electron-in-
Figure 2. UV/Vis absorption and PL spectra of synthesized compounds.
a) UV spectra were recorded at room temperature as dilute CH2Cl2 solutions
(10ꢀ5 m order). b) and c) PL spectra were recorded either at room tempera-
ture as dilute CH2Cl2 solutions (10ꢀ5 m order) of N3 and N4, or at 77 K as
dilute EET [Et2O/EtOH/toluene (v/v 2:1:1)] frozen solutions (10ꢀ5 m) of P3, P4,
and P4-O with excitation wavelength of 310 nm. The inset photograph in
c) shows a frozen EET solution of P4 upon excitation with a 365 nm UV lamp
at 77 K.
jection/rejection processes (redE1/2 ꢀ1.84 V vs. Fc/Fc+) as that of
red
N4 (
E
ꢀ1.85 V vs. Fc/Fc+). The P oxidation (P4-O) gave
1/2
red
a large positive shift in
E
(ꢀ1.49 V vs. Fc/Fc+) compared
1/2
with P4 by 0.35 eV, which was supported by DFT calculations
(see below).
Thermal stabilities
ed in further positive-shift of the redox potential (redE1/2
=
ꢀ1.36 V vs. Ag/Ag+), compared with both N1 and P1 [redE1/2
The thermal stability of organic functional compounds is a cru-
cial factor from the viewpoints of processibility and durability.
For evaluation of the thermal stabilities of the compounds
used in this study, thermogravimetric analysis (TGA) under
a flow of air and of N2 gas was performed (Table 1, for the de-
tailed TGA-profiles, see the Supporting Information). Whereas
the reference compound N1 started to degrade at 1998C [Td
(5 wt% loss)] under a stream of N2 gas, which was almost the
same as that measured under the air [Td (5 wt%)=1938C], its
phosphorus analogue P1 exhibited slightly higher thermosta-
bility [Td (5 wt%, N2)=2168C, Td (5 wt%, air)=1988C]. Overall,
red
red
red
(P2)ꢀ
E
(N1)= +0.38 V;
1/2
E
(P2)ꢀ
E (P1)= + 0.22 V],
1/2
1/2
indicating the marked stabilization of the LUMO energy by flu-
orine incorporation. In fact, the energy-stabilization effect by P
and F introduction were rationalized by DFT calculations (see
below). More interestingly, the complexation of P1 with gold
(P1-AuCl) drastically enhanced the electron-accepting ability of
P1, as evidenced by three reversible reduction processes with
positively shifted half-wave potentials (Figure 3). Such multi-
electron-accepting behavior was not observed with other dike-
tophosphanyl compounds, therefore this unusual feature is
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
all the compounds show good to high thermostabilities
(Table 1).
carbonyls-P compared with that of N1, thereby lowering its
energy compared with that of N1 [ELUMO+1 (N1)=ꢀ1.11 eV,
ELUMO+1 (P1)=ꢀ1.64 eV] (Figure 4). The introduction of fluorine
atoms onto the benzene core of P1 drastically stabilizes both
HOMO and LUMO energies by 0.44 and 0.31 eV, respectively
Theoretical calculations
(Figure 4), which is consistent with the more positive potential
To further understand the effect of P incorporation into dicar-
bonyl-conjugated systems on the molecular orbitals and physi-
cochemical properties, theoretical calculations using the DFT
method (B3LYP/6-31G(d) level of theory) was performed.[13] An
illustrative summary of the molecular orbitals and an energy
diagram of the phthalimide family (N1, P1, P2, and P1-AuCl) is
depicted in Figure 4 (for details of the results, see the Support-
red
E
(P2) compared with that of P1 (Figure 3 and Table 1). In-
1/2
terestingly, this fluorine introduction resulted in a switch be-
tween the orbital symmetries of the HOMO-1 and HOMO-2 of
P1. Furthermore, the installation of a gold atom onto the P
center (P1-AuCl) significantly stabilizes all the orbital energies,
even with the LUMO+1 being much lower than the LUMOs of
other compounds. This result would explain the large positive
red1
shift in
E
and the multiple electron accepting behavior ob-
1/2
served in the cyclic voltammetry experiments (Figure 3). Re-
garding 2,3-naphthalimide series (N3 and P3), the HOMO of P3
is developed on the Mes-(CO)2 moiety (see Figure S13 in the
Supporting Information), which is in good agreement with the
shape of HOMO of P analogues of naphthalimide P1 and P2
and could be a common feature in aromatic-fused five-mem-
bered diketophosphanyl compounds. In contrast to the phtha-
limide family (P1 and P2), the LUMO of P3 is largely developed
over the p moiety including the P center, which lowers its
energy (see the Figure S13 in the Supporting Information). The
replacement of the N atom of 1,8-naphthalimide N4 with a P
atom (P4) caused degeneracy of the HOMO/HOMO-1 and
HOMO-2/HOMO-3 orbitals (see the Figure S14 in the Support-
ing Information), with their energies being significantly in-
creased. The P oxidation (P4-O) certainly stabilized the LUMO
energy compared with P4, showing good agreement with
electrochemical experiments.
Conclusion
We have synthesized and characterized a series of diketophos-
phanyl compounds as P analogues of aromatic diimides. Com-
prehensive studies highlight the effectiveness of P substitution
of N atom of phthalimides and naphthalimides in terms of the
enhancement of lowering the LUMO energies as well as the
LUMO+1 levels, leading to increased electron-accepting abili-
ties except for the 2,3-naphthalene-fused diketophosphanyl
compound. Furthermore, unique luminescent properties of
naphthalene-fused diketophosphanyl compounds at low tem-
perature were uncovered.
Figure 4. Energy diagram and molecular orbitals of N1, P1, P2, and P1-AuCl
calculated by the DFT method at the B3LYP/6-31G(d) level of theory. For the
calculation of P1-AuCl, a basis set of LanL2DZ was used for the Au atom.
ing Information). Substitution of the N atom of N1 with a P
atom resulted in a negligible decrease in the LUMO [ELUMO
(N1)=ꢀ2.18 eV, ELUMO (P1)=ꢀ2.22 eV] but a significant in-
crease in the HOMO [EHOMO (N1)=ꢀ6.32 eV, EHOMO (P1)=
ꢀ6.16 eV], leading to a narrower band gap [DEHOMO–LUMO (N1)=
4.14 eV, DEHOMO–LUMO (P1)=3.94 eV, Figure 4]. Whereas the
HOMO orbital of imide N1 is mainly comprised of the p moiety Acknowledgements
on the Mes group, that of P1 lies on a large conjugated
system formed from the mesityl plane and two carbonyl
groups, leading to the increase in the HOMO energy (Figure 4).
A significant contribution of the lone pair on the P atom was
found in the HOMO-1 of P1, whereas almost no contribution
of the N lone pair was recognized with N1. Both of the LUMO
orbitals of N1 and P1 are almost the same shape, consisting of
(benzene) p* and (CO) p* moieties (Figure 4). Notably, the
LUMO+1 orbital of P1 is largely developed on the benzene-
The authors deeply thank Dr. Gaku Fukuhara and Dr. Kei
Okubo of Osaka University for the assistance of the measure-
ments of luminescence spectra and lifetime at low tempera-
ture. This research was partly supported by the Shorai Founda-
tion for Science and Technology (to Y.T.). One of the authors
(Y.T.) would like to acknowledge support from the Frontier Re-
search Base for Global Young Researchers, Osaka University, in
the Program of MEXT, Japan.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
this paper. These data are available free of charge from the Cambridge
Crystallographic Data Center via www.ccdc.cam.ac.uk/data_request/cif.
[11] In the absence of 2,6-dimethyl substituents on the N-aryl moiety, N-aryl
2,3-naphthalimides are known to emit dual emission (SW and LW) from
two different Frank–Condon excited states with the configurations
being coplanar (LW) and nonplanar (SW), due to free rotation around
the NꢀC(Ar) single bond, see: a) A. Demeter, T. Bꢃrces, L. Biczꢄk, V. Wint-
gens, P. Valat, J. Kossanyi, J. Phys. Chem. 1996, 100, 2001; b) P. Valat, V.
Wintgens, J. Kossanyi, L. Biczꢄk, A. Demeter, T. Bꢃrces, J. Am. Chem. Soc.
1992, 114, 946.
[12] No luminescence was detected with N4, probably because the PL in-
tensity was below measurable limits of the fluorometer. This is reasona-
ble because the reported quantum yield of N-2,6-dimethylphenyl-1,8-
naphthalimide is as low as approximately 0.00002 (in n-hexane). See,
Ref. [11a].
[13] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
Keywords: electron-deficient compounds · gold · phosphorus
heterocycles · photophysics · pi interactions
[1] For reviews on phosphole-based functional molecules, see: a) C.
Romero-Nieto, T. Baumgartner, Synlett 2013, 24, 920; b) Y. Ren, T. Baum-
gartner, Dalton Trans. 2012, 41, 7792; c) A. Fukazawa, S. Yamaguchi,
Chem. Asian J. 2009, 4, 1386; d) Y. Matano, H. Imahori, Org. Biomol.
Chem. 2009, 7, 1258; e) J. Crassous, R. Rꢃau, Dalton Trans. 2008, 6865;
f) M. G. Hobbs, T. Baumgartner, Eur. J. Inorg. Chem. 2007, 3611.
[2] For reviews on organophosphorus conjugated materials, see: a) T.
Baumgartner, Acc. Chem. Res. 2014, 47, 1613; b) M. Stolar, T. Baumgart-
ner, Chem. Asian J. 2014, 9, 1212; c) T. Baumgartner, R. Rꢃau, Chem. Rev.
2006, 106, 4681; Chem. Rev. 2006, 107, 303.
[3] a) A. R. Barron, S. W. Hall, A. H. Cowley, J. Chem. Soc. Chem. Commun.
1987, 1753; b) C. L. Liotta, M. L. McLaughlin, D. G. V. Derveer, B. A.
O’Brien, Tetrahedron Lett. 1984, 25, 1665; c) D. Fenske, E. Langer, M.
Heymann, H. J. Becher, Chem. Ber. 1976, 109, 359; d) K. Issleib, K. Mohr,
H. Sonnenschein, Z. Anorg. Allg. Chem. 1974, 408, 266; e) A. J. Saunders,
I. R. Crossley, M. P. Coles, S. M. Roe, Chem. Commun. 2012, 48, 5766.
[4] a) X. He, J. Borau-Garcia, A. Y. Y. Woo, S. Trudel, T. Baumgartner, J. Am.
Chem. Soc. 2013, 135, 1137; b) X. He, T. Baumgartner, Organometallics
2013, 32, 7625.
[5] C.-Z. Li, H.-L. Yip, A. K.-Y. Jen, J. Mater. Chem. 2012, 22, 4161.
[6] Y. Takeda, T. Nishida, S. Minakata, Chem. Eur. J. 2014, 20, 10266.
[7] The synthesis and self-assembly behavior of a 12-membered cyclic bis-
(diketophosphanyl)compound has been reported by the Crossley
group, see Ref. [3e].
[8] a) Q. Zheng, J. Huang, A. Sarjeant, H. E. Katz, J. Am. Chem. Soc. 2008,
130, 14410; b) S. Kola, N. J. Tremblay, M.-L. Yeh, H. E. Katz, S. B. Kirschner,
D. H. Reich, ACS Macro Lett. 2012, 1, 136; c) S. Kola, J. H. Kim, R. Ireland,
M.-L. Yeh, K. Smith, W. Guo, H. E. Katz, ACS Macro Lett. 2013, 2, 664.
[9] A. Decken, E. D. Gill, F. Bottomley, Acta Crystallogr. Sect. E: Struct. Rep.
Online 2004, E60, o1456.
Received: October 2, 2014
Revised: October 27, 2014
[10] For a summary of crystal data of P4, see the Supporting Information.
CCDC-1022838 contains the supplementary crystallographic data for
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Can P do it? A series of phosphorus an-
alogues of N-mesityl phthalimides and
naphthalimides have been synthesized
and characterized (see figure). In a com-
parison of the physicochemical proper-
ties of these diketophosphanyl com-
pounds with those of the corresponding
reference imides, the effect of P incor-
poration into the dicarbonyl-fused aro-
matics on their properties has been
revealed.
&
Phosphorus Heterocycles
Y. Takeda,* T. Nishida, K. Hatanaka,
S. Minakata*
&& – &&
Revisiting Phosphorus Analogues of
Phthalimides and Naphthalimides:
Syntheses and Comparative Studies
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!