Dipyrrolonaphthyridinedione – (still) a mysterious cross-conjugated chromophore

Abstract

Dipyrrolonaphthyridinediones (DPNDs) entered the chemical world in 2016. This cross-conjugated donor–acceptor skeleton can be prepared in two steps from commercially available reagents in overall yield ≈15–20% (5 mmol scale). DPNDs can be easily and regioselectively halogenated which opens an avenue to numerous derivatives as well as to π-expansion. Although certain synthetic limitations exist, the current derivatization possibilities provided impetus for numerous explorations that use DPNDs. Structural modifications enable bathochromic shift of the emission to deep-red region and reaching the optical brightness 30 000 M⁻¹ cm⁻¹. Intense absorption and strong emission of greenish-yellow light attracted the interest which eventually led to the discovery of their strong two-photon absorption, singlet fission in the crystalline phase and triplet sensitization. Dipyrrolonaphthyridinedione-based twistacenes broadened our knowledge on the influence of twisting angle on the fate of the molecule in the excited state. Collectively, these findings highlight the compatibility of DPNDs with various applications within organic optoelectronics.

---

Bartłomiej Sadowski Bartłomiej Sadowski was educated in chemistry at the Warsaw University of Technology, Poland. He then obtained his PhD from the Institute of Organic Chemistry of the Polish Academy of Sciences in 2019 under the supervision of Prof. Daniel T. Gryko. After completing his post-doctoral work at Georg-August-Universität Göttingen in the group of Prof. Lutz Ackermann, he began his independent career at Centre of New Technologies, University of Warsaw. His research interests focus on the electrochemically-enabled synthesis of high energy intermediates as well as chemistry of functional aromatic molecules, notably dipyrrolonaphthyridinediones.

Daniel T. Gryko Daniel T. Gryko obtained his PhD from the Institute of Organic Chemistry of the Polish Academy of Sciences in 1997, under the supervision of Prof. J. Jurczak. After a post-doctoral stay with Prof. J. Lindsey at North Carolina State University (1998–2000), he started his independent career in Poland. He became Full Professor in 2008. The same year he received the Society of Porphyrins and Phthalocyanines Young Investigator Award and in 2017 Foundation for Polish Science Award. His current research interests are focused on the synthesis of functional dyes as well as on two-photon absorption, solvatofluorochromism, excited-state intramolecular proton transfer and fluorescence probes.

---

1. Introduction

A pyrrole ring is probably the most known small aromatic heterocycle. Many of its derivatives are found in a variety of natural products¹ as well as drugs.² In addition, this small heterocyclic motif is a key part of many functional dyes such as porphyrinoids,³ corroles,⁴ indolizines,⁵ 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY)⁶ or bis(difluoroboron)-1,2-bis((1H-pyrrol-2-yl)methylene)hydrazine (BOPHY).⁷

Due to its intrinsic electron-rich character, a pyrrole ring seems to be an ideal candidate for an electron-donating moiety in cross-conjugated chromophores. However, there is still a huge gap in the understanding of its nature when coupled with electron-accepting units. Some of the recent reports discussed an issue of aromaticity of cross-conjugated chromophores containing a pyrrole ring in a ground and excited singlet/triplet states.⁸

Donor–acceptor cross-conjugated dyes are well established in the literature.⁹ Compounds of this type, i.e. diketopyrrolopyrroles,¹⁰ isoindigos,¹¹ cibalakrot^8a,b,12 were tested as main components in organic electronics¹³ or biology.¹⁴ Donor–acceptor systems of cross-conjugated nature also played formidable role in human history. Indigo and Tyrian Purple (6,6′-dibromoindigo) were used as garment dyes for millennia.¹⁵ Moving to modern times, a diketopyrrolopyrrole molecule discovered by Farnum¹⁶ in 1974 and commercialized by Ciba Geigy,¹⁷ is a key structural motif of Pigment Red 254 – better known as ‘Ferrari Red’ as it was used to paint iconic Ferrari cars. Since then, there was no discovery of new type of donor–acceptor cross-conjugated dye. The unveiling of dipyrrolonaphthyridinediones (DPNDs) in our research group in 2016 has changed this situation (Fig. 1).¹⁸ The combination of very interesting and unique photophysical properties, straightforward synthesis as well as huge potential for functionalization are responsible for their career in the literature.

Fig. 1 The structure and numbering of key positions of the simplest dye belonging to the DPND family.

In this Perspective article, we will describe the beginnings as well as the history of DPNDs. First of all, we will briefly discuss the synthetic pathways leading to the core followed by describing already known synthetic modifications thereof. Finally, optoelectronic properties will be studied with special emphasis given to structure–property relationship followed by specific applications that has been tested until now. We believe that such structure of the Perspective will open up new avenues for this interesting and still undiscovered chromophore.

2. Synthesis of dipyrrolonaphthiridinediones

The DPND core (Fig. 1) features a 5,6,6,5-type architecture and is comprised of four linearly-fused cycles: two pyrrole rings at the periphery and two six-membered rings in the central part of a chromophore, each containing a C=O group.¹⁸ From the structural point of view, the DPND core closely resembles the BOPHY fluorophore,⁷ however the latter bears BF₂ moieties along with the bis-nitrogen bridge which results in different optoelectronic behavior.

Accordingly to X-ray analysis of 1f (see Table 1 and Fig. 2),¹⁸ the core is rigidly planar and deviations from the plane are no greater than 0.018 Å. Undoubtedly, the steric clash between C=O groups and alkyl chains at positions 6 and 12 has detrimental influence on the chromophore structure as bond angles between C=O groups and carbon atoms adjacent to the alkyl groups deviate significantly from their ideal trigonal values of 120°. The values are closer to 120° in the case of 1a that lacks alkyl chains at positions 6 and 12, as revealed by Wang and others¹⁹ (vide infra).

Table 1 Synthesis of various 6,12-disubstituted DPND derivatives^a

Carboxylic acid	Reaction time [h]	DPND	R	Yield [%]	Ref.
a Reagents proportions: 4 (0.5 mmol), carboxylic acid (3 mmol), TFAA (6 mmol), TFA (3 mmol). b 5 mmol scale. c Propionic anhydride was used instead of carboxylic acid. d Unpublished results.
CH3CO2H	5	1b	CH3	17	18
C7H15CO2H	3	1c	C7H15	29	18
				23^b	18
Et(Me)CHCO2H	6	1d	sec-Butyl	21	18
4-(MeO)C6H4CH2CO2H	6	1e	4-Methoxy-benzyl	10	18
(C2H5CO)2O^c	2	1f	C2H5	23	18
C3H7COOH	4	1g	C3H7	21	21
C6H13COOH	2	1h	C6H13	32	19
2-NO2C6H4COOH	3	1i	2-NO2C6H4	0	—^d
2-ThienylCH2COOH	3	1j	2-ThienylCH2	0	—^d
3,4-(MeO)2C6H4CH2CO2H	3	1k	3,4-Dimethoxy-benzyl	0	—^d
(E)-Cinnamic acid	3	1l	–C=C–Ph	0	—^d
PhCOOH	3	1m	Ph	0	—^d
2-ThienylCH2COOH	3	1n	2-ThienylCH2	0	—^d
HOOC(CH2)7COOH	3	1o	–(CH2)7–	3	20
HOOC(CH2)8COOH	3	1p	–(CH2)8–	15	20
HOOC(CH2)9COOH	3	1r	–(CH2)9–	3	20

---

Fig. 2 X-ray structure of compound 1f. Views perpendicular (top) and along (bottom) with respect to the chromophore plane. Adapted with permission from ref. 18. Copyright 2016 Royal Society of Chemistry.

Initial attempts to assemble the DPND core were based on the use of 2-formylpyrrole (2) and succinyl chloride (3) as simple and commercially available building blocks, however 1a could be isolated in only 3.4–6.4% yield along with unreacted 2 and a large amount of black precipitate in both cases (Scheme 1).¹⁸ The presence of the black precipitate was associated with base-mediated polymerization of 3 as its formation was also noted in the absence of the aldehyde.

Scheme 1 Initial attempts of the synthesis of DPND derivatives. Reaction conditions: (a) DMAP (20 mol%), Et₃N, CH₂Cl₂, rt, 2 h. Yield: 3.4%; (b) K₂CO₃, DMF, 0 °C, 2 h. Yield: 6.4%.

Another tested strategy towards 1a involved dipyrroyl derivative 4 as a source of pyrrole rings. Compound 4 can be easily synthesized from either 2,5-dimethoxytetrahydrofurane and succinamide^18,20 or pyrrole and 3.²¹ Indeed, subjecting 4 to the typical Vilsmeier–Haack reaction conditions (DMF/POCl₃) resulted in the formation of 1a, but still with low efficiency (Scheme 2). During the optimization process, intermediate of type 5 was not detected at all.

Scheme 2 Synthesis of DPNDs via acylation of a pyrrole ring.

Due to low efficiencies and relatively poor solubility of 1a in common organic solvents, we intended to introduce additional substituents that improve solubility of resulted chromophores. This idea was realized via an acylation reaction employing conditions previously developed for N-tosylpyrroles.²² Specifically, in the presence of excess of acetic acid, dipyrroyl derivative 4 undergoes double acylation followed by aldol-type condensation eventually affording 1b in 12% yield along with 9% of 6 (Scheme 2). The overall efficiency was further improved by the decrease in amounts of TFAA and TFA. In general, the developed method allows for assembling 6,12-difunctionalized DPND derivatives 1b–h as well as cyclophane analogues 1o–r (Table 1). During our ongoing adventure with these strongly fluorescent chromophores, it appeared that DPNDs are formed only when alkyl carboxylic acids are used, although with some exceptions (1j, 1k). The highest yield (32%) has been recently reported for enanthic acid (C₆H₁₃COOH)¹⁹ as a carboxylic partner. All attempts with carboxylic acids other than aliphatic ones failed.

3. Synthetic modifications of the DPND core

As the DPND core consists of two flanking pyrrole rings as well as the carbonyl-based central part, synthetic modifications of this chromophore should be a result of inherent reactivity (to a certain extent) of these two parts. It should be noticed here that all three available positions within the pyrrole ring are not equivalent due to unsymmetrical mode of fusion between 5- and 6-membered rings. As a proof of concept, we initially envisaged bromination of the pyrrole part (Scheme 3) as it is well known that aromatic halides are outstanding feedstocks in a wide range of transition-metal-catalyzed cross-coupling reactions.²³

Scheme 3 Derivatization of the DPND core based on halogenation reactions.

Indeed, 1 selectively undergoes double bromination reaction at the positions 3 and 9 (product 7) using NBS (N-bromosuccinimide) as a bromine atom source. Moreover, we found out that chloroform stabilized with amylene performed better in this reaction compared to that one stabilized with ethanol. Besides bromination, chlorination and iodination reactions were examined by Ayitou and others.²¹ While under the influence of CuCl₂·2H₂O double chlorination reaction occurs at the positions 3 and 9 (product 8), applying NIS (N-iodosuccinimide) as an iodination reagent results in completely different regioselectivity, giving rise to 2,8-diiodinated DPND (9). The unexpected regioselectivity of iodination was presumably ascribed to a large size of an iodine atom (potential steric clash with a carbonyl oxygen atom). However, the fact of decreased reactivity at the α position of a pyrrole ring caused by the presence of electron-withdrawing groups within this ring cannot be excluded.²⁴

The very presence of halogen atoms at positions 3 and 9 should enable further transformations. As an example, subjecting compound 7a to the Rosenmund-von Braun reaction conditions resulted in dicyano derivative 10 in 33% yield,¹⁸ proving that further derivatization of DPND-based halides is possible.

As mentioned above, halides constitute one of the best platforms for derivatization, especially towards enlarged architectures by employing acetylenes²⁵ or styrenes^6a,26 as coupling partners. As a matter of fact, dibrominated DPNDs of type 7 were successfully transformed into a variety of π-expanded platforms employing the main types of cross-coupling reactions. The Pd(OAc)₂/P(o-tol)₃ system proved to be effective in the construction of quadrupolar, centrosymmetric molecules 11a–f with double bond linkages via the Heck reaction.²⁷ Employing a typical catalytic system for the Sonogashira reaction led to molecules 12a–d bearing triple bond π-spacers.¹⁸ Finally, amine-decorated DPNDs 13a–f can be assembled via the Buchwald–Hartwig amination reaction using the Pd₂dba₃/SPhos system.²⁸ It should be mentioned here that some of π-expanded derivatives, especially those bearing strongly electron-donating groups, were not sufficiently stable during isolation or it was not possible to isolate them in a pure form due to similar affinity to a stationary phase and/or lower solubility.

Compared to cross-coupling reactions of heteroaryl halides, transition-metal-catalyzed C–H activation processes are characterized by higher atom and step economy.²⁹ In other words, employing C–H activation for the functionalization of organic chromophores helps avoiding often problematic derivatization (halogenation, borylation etc.).

The DPND core appeared to be an ideal platform for the study on direct arylation reaction (Scheme 4)³⁰ which was widely used as a common strategy toward organic materials.³¹ The catalytic system involves Pd₂dba₃ as a catalyst, PCy₃·HBF₄ as a ligand, PivOH as an additive, and allows for the functionalization of the DPND core at the positions 3 and 9 with differently decorated arene rings. From the view-point of industrial research, it was proven that Pd₂dba₃ can be replaced with cost-efficient Pd(OAc)₂.³² Variety of aryl halides are reactive towards the DPND core, nevertheless those bearing electron-withdrawing groups performed better. In a typical reaction, 2.2-3-fold excess of aryl halide was used to achieve doubly-arylated derivatives. Monoarylation is also achievable by applying 2.0 equivalents of 1cversus aryl halide, as proved for some nitroaryl-decorated DPNDs (Scheme 3).^30b Interestingly, only for 1-bromo-8-nitronaphthalene we were not able to obtain doubly arylated derivative due to high steric congestion within the reaction centre. This resulted in messy reaction outcome where the expected, doubly-functionalized product was not detected at all.

Scheme 4 Direct arylation of the DPND core.

The developed direct arylation methodology was further applied in the synthesis of N-doped analogues of polycyclic aromatic hydrocarbons (PAHs) (Scheme 5).³³ Arylation of the DPND core with sterically congested aryl halides bearing acetylamino groups led to a series of dyes 47 that can be smoothly transformed into polycyclic aromatics of type 48, under the influence of strong acid(s). Here, the reactivity of carbonyl moieties within the DPND core was tested in a condensation process, similarly to the reaction described for perylene bisimides.³⁴ Among tested derivatives, bis-arylated dye 47d failed as a precursor of 48d presumably due to low nucleophilicity of nitrogen caused by the presence of a strongly electron-withdrawing NO₂ group at the para position relative to NHAc.

Scheme 5 Functionalization of carbonyl moieties within the DPND core.

4. The optoelectronic properties of dipyrrolonaphthyridinedione and its applications

Unsubstituted compound 1a exhibits an intense structured absorption band with the longest maxima (λ^max_abs) at 509 nm in dichloromethane (CH₂Cl₂), and molar extinction coefficient of 26 600 M⁻¹ cm⁻¹ (Table 2). Alkyl-substituted derivatives 1b–h shows similar structural absorption bands that are hypsochromically-shifted compared to 1a (Fig. 3 and Table 2), probably due to additional steric interactions between alkyl chains at positions 6 and 12 with carbonyl groups. The origin of these “sub-bands”, although not investigated in detail for DPNDs, one may possibly attribute to vibrational progression in the excited state, as found previously for BOPHY dyes.⁷ In general, DPNDs 1a–h are strongly fluorescent in CH₂Cl₂ (Φ_fl up to 0.71). Compared to BOPHYs however, DPNDs emit less intensively as values of Φ_fl measured for simple BOPHY-type dyes approach ≈0.9.⁷

Table 2 Photophysical properties of DPNDs measured in dichloromethane¹⁸

Dye	λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν^a [cm^−1]
a Stokes shift i.e. difference between lowest energy absorption band and highest energy emission band expressed in cm^−1. b Reference: Rhodamine 6G in EtOH (Φfl = 0.94). c Reference: cresyl violet in MeOH (Φfl = 0.54).
1a	509	26 600	535	0.61^b	950
1b	499	29 200	523	0.66^b	900
1c	504	29 300	528	0.71^b	900
1d	505	23 100	543	0.58^b	1400
1e	510	28 300	536	0.26^b	950
1f	500	28 000	526	0.67^b	1000
6	503	24 600	601	0.46^b	3240
10	517	34 600	537	0.25^b	720
11a	621	60 200	699	0.016^c	1800
11b	610	56 300	671	0.022^c	1500
11c	603	55 100	662	0.041^c	1500
12a	599	57 300	633	0.51^c	900
12b	584	47 400	616	0.51^c	900
12c	601	52 400	643	0.59^c	1100
12d	645	56 600	736	0.17^c	1900

---

Fig. 3 Absorption and normalized emission spectra of dye 1c (red), 12a (blue), and 12d (green). Adapted with permission from ref. 18 Copyright 2016 Royal Society of Chemistry.

It should be mentioned that during the preparation of alkyl-substituted DPNDs (i.e.1c), a trifluoroacetyl derivative 6 is also obtained as a minor product (Scheme 2). Here, the emission band is red-shifted by ca. 80 nm as compared with its analog 1c (Table 2) and both absorption and emission bands become structureless.¹⁸ The most convenient way toward dyes that absorb and emit at significantly longer wavelengths is to introduce arylethynyl or arylethenyl moieties at the core's peripheries. The Sonogashira coupling of 7a with different arylacetylenes gave rise to a series of π-expanded DPNDs 12 which generally absorb at 584–601 nm and emit at 616–643 nm (Table 2 and Fig. 3).¹⁸ Importantly, the largest red-shift of both bands was observed for 12d bearing Et₂N auxochromes at peripheries which suggests that the DPND core as a whole behaves as an electron acceptor. Although the value of Φ_fl dropped down to 0.17 (compared with 12a–c), 12d absorbs in the red region of the spectrum and emits at 736 nm.

In turn, vinylidene-linked systems 11a–c²⁷ absorb at longer wavelengths that is around 603–621 nm (Table 2) and this is due to better electronic conjugation³⁵ between the core and groups at the peripheries. These dyes are poorly emissive (λ^max_em = 662–669 nm, Φ_fl < 0.1), however, probably because of an additional energy dissipation mechanism, i.e. E–Z isomerization of a C=C double bond.³⁶

An “electron-accepting” character of the DPND core was further probed by studying a variety of weakly coupled, quadrupolar dyes prepared via direct arylation methodology.^30a As expected, the presence of biaryl-type connection between the DPND core and auxochromes at peripheries leads to weaker electronic conjugation between them. The dihedral angle between the DPND core and an aryl substituent was found to be ≈40°–45° or 50°–55° based on DFT methods^30a and X-ray analysis,^30b,30c respectively. Consequently, these dyes feature absorption and emission bands at shorter wavelengths compared to 11 and 12 (Table 3 and Fig. 4). In general, synthesized dyes are moderately fluorescent (Φ_fl ≈ 0.3–0.6) and the emission band is located in the red region of the spectrum (λ^max_em ≈ 600–620 nm).

Table 3 Photophysical properties of DPND dyes bearing a biaryl-type bridge

Dye	R	Solvent	λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν [cm^−1]
a Solutions were sonicated for 15–30 min directly before measurement. b Not measured due to low S/N ratio. c Sulforhodamine 101 was used as a reference (Φfl = 0.95 in EtOH). d Cresyl violet was used as a reference (Φfl = 0.54 in MeOH). e 4-Pyridyl groups are attached to the DPND core instead of R-C6H4.
15	4-CN	CH2Cl2	543	33 200	602	0.36^c	1800
		DMSO^a	549	31 000	614	0.38^c	1900
16	4-CHO	CH2Cl2	547	37 100	607	0.36^c	1800
		DMSO^a	555	36 600	619	0.36^c	1900
17	4-CO2Et	CH2Cl2	544	34 000	600	0.44^c	1700
		DMSO^a	550	27 200	614	0.40^c	1900
18	2-CO2Me	CH2Cl2	536	31 400	581	0.63^c	1400
		DMSO^a	542	32 100	591	0.57^c	1500
19	2-OMe	CH2Cl2	536	31 100	603	0.60^c	2100
		DMSO^a	541	30 200	607	0.54^c	2100
20	4-Pyridyl^e	CH2Cl2	535	31 600	582	0.46^c	1500
		DMSO^a	540	31 800	599	0.45^c	1900
21	4-CH2OMe	CH2Cl2	542	37 100	601	0.38^c	1800
		DMSO^a	555	27 100	620	0.42^c	1900
22	4-OMe	CH2Cl2	550	26 600	624	0.43^c	2100
		DMSO^a	557	34 800	643	0.36^c	2400
23	4-NEt2	CH2Cl2	615	39 500	735	0.10 ^d	2700
24	4-NPh2	CH2Cl2	583	38 000	—^b	—^b	—^b
25	4-NO2	CH2Cl2	548	29 300	611	0.46^c	1900
34	Pyrrolo[3,2-b]pyrrole core	CH2Cl2	399, 579	83 600, 47 300	—^b	—^b	—^b
43	2-NO2	CH2Cl2	538	26 500	—^b	—^b	—^b
		DMSO^a	545	26 800	—^b	—^b	—^b

---

Fig. 4 Top: absorption spectra of compounds 18 (red), 21 (green), 22 (blue), 23 (orange) and 25 (black) measured in dichloromethane. Bottom: normalized emission spectra of compounds 18 (red), 21 (green), 22 (blue), 23 (orange) and 25 (black) measured in dichloromethane. Adapted with permission from ref. 30a. Copyright 2018 Wiley.

The largest change in the optical behaviour was noted for DPNDs with NR₂ groups at peripheries: bathochromically-shifted both absorption and emission bands and lower value of Φ_fl by contrast with 15–21 (Table 3 and Fig. 4). According to theoretical calculations (Fig. 5), compounds bearing CN, H and Me groups at peripheries feature locally-excited transitions (S₁-LE(π–π*)), where HOMO and LUMO wavefunctions are located mostly within the DPND core. In contrast, for NMe₂ and OMe groups the electron density is shifted from outer groups toward the core upon the photoexcitation, with largest net change noted for the dye bearing NMe₂ group.

Fig. 5 Frontier molecular orbitals of compounds from DPND(C₆H₄R)₂ series calculated in C₂ symmetry at the DFT(B3-LYP) methods using the cc-pVDZ basis set. Adapted with permission from ref. 30a. Copyright 2018 Wiley.

The above-mentioned observations suggest that placing electron-donating groups at peripheries induce significant red shift of absorption and emission and the DPND core behaves like an “electron-acceptor”.

The electron-accepting character of the core was further deeply investigated²⁸ by studying a series of dyes 13a–f decorated with an amine function connected directly with the core or by a π-spacer (12d, 23 and 24). We found that these dyes feature evident solvatofluorochromic behavior (Fig. 6) that is the emission bands shift toward lower energies (usually λ^max_em > 700 nm) as well as the value of Φ_fl lowers as polarity of the medium increases (Table 4). Interestingly, compound 13e exhibits slightly different behavior compared to other dyes belonging to a series 13. The very presence of four CN groups withing amine moieties partially hampers electron density shift toward the DPND core and hence the efficient formation of dark charge-transfer (CT) states which results in less pronounced solvatofluorochromism. On the other hand, Φ_fl increases sharply in more polar environment reaching values of 0.4–0.6 (Table 4).

Fig. 6 Absorption (solid) and emission (dotted) spectra of compounds 13b (a), 13d (b), 13e (c) and 23 (d) measured in cyclohexane (black), toluene (green), THF (red), CH₂Cl₂ (yellow), MeCN (blue) and DMSO (orange). Adapted with permission from ref. 28 Copyright 2018 American Chemical Society.

Table 4 Photophysical properties of amine-decorated DPNDs in different solvents²⁸

Dye	Solvent	λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν^a [cm^−1]
a Stokes shift i.e. difference between lowest energy absorption band and highest energy emission band expressed in cm^−1. b Sulforhodamine 101 was used as a reference (Φfl = 0.95 in EtOH). c Cresyl violet was used as a reference (Φfl = 0.54 in MeOH).
13b	Cyclohexane	575	37 000	628	0.35^b	1500
	Toluene	582	38 000	645	0.32^b	1700
	THF	580	35 000	650	0.35^b	1900
	CH2Cl2	578	34 000	653	0.27^b	2000
	MeCN	576	29 000	659	0.21^b	2200
	DMSO	590	28 000	670	0.18^b	2000
13d	Cyclohexane	645	38 000	702	0.39^c	1300
	Toluene	654	35 000	718	0.26^c	1400
	THF	648	37 000	720	0.10^c	1500
	CH2Cl2	648	35 000	727	0.07^c	1700
	MeCN	642	34 000	732	0.01^c	1900
	DMSO	656	38 000	739	0.01^c	1700
13e	Cyclohexane	565	nd	639	0.37^b	2000
	Toluene	571	27 000	655	0.49^b	2200
	THF	560	28 000	658	0.62^b	2700
	CH2Cl2	558	27 000	659	0.61^b	2700
	MeCN	554	27 000	666	0.63^b	3000
	DMSO	564	26 000	677	0.47^b	3000
23	Cyclohexane	603	36 000	670	0.40^c	1700
	Toluene	613	37 000	702	0.25^c	2100
	THF	615	40 000	729	0.11^c	2500
	CH2Cl2	615	39 000	735	0.10^c	2700
	MeCN	607	34 000	750	0.02^c	3100
	DMSO	631	28 000	758	0.02^c	2700

---

Cyclic voltammetry (CV) appeared to be an excellent methodology to probe the electronic structure of dipyrrolonaphthiridinediones (DPNDs) (Fig. 7–9). A CV trace of unsubstituted dye 1a displays irreversible oxidation/reduction events and the shape of the CV curve changes with a rising number of redox cycles (Fig. 7).¹⁸ Upon an oxidation event a black, insoluble deposit appeared on an anode which did not dissolve upon reduction. Such a deposit did not form in the case of 1c, 10 and 12b (Fig. 8). For 1c, the results show a reversible reduction with E_1/2 = −1.125 V vs. SCE (E_LUMO = −3.2 eV) in the accessible potential range. The appearance of reducible CN groups in the chromophore structure (compound 10, Fig. 8) results in an additional reduction wave within the CV curve (E_1/2 = −0.675 V vs. SCE) that can be clearly connected with those groups, while the often reversible reduction event characteristic for the DPND core shifted toward more negative potentials (E_1/2 = −1.215 V vs. SCE) and the oxidation wave is not present in the accessible potential range. This constitutes a general feature for DPND derivatives bearing electron-withdrawing groups at the peripheries.^30a,b When it comes to electrochemically-driven oxidation events, most of DPND derivatives show irreversible oxidation wave (if any). The situation is changed when another, easily oxidizable groups are present in the structure (Fig. 9).²⁸ Cyclic voltammograms of electron-rich DPND derivatives containing NR₂ groups (i.e.13f, Fig. 9), beside one reversible reduction event associated with the core, frequently include two oxidation waves where the first event is connected with a reversible (often stepwise) oxidation of the –NR₂ auxochrome while the second one (irreversible) comes from an oxidation process within the core.

Fig. 7 Cyclovoltammetric curve registered for 1a in dichloromethane in the entire range of examined potentials: −1800 ÷ 2100 mV, v = 100 mV s⁻¹. Adapted with permission from ref. 18 Copyright 2016 Royal Society of Chemistry.

Fig. 8 Cyclovoltammogramms of the dyes 1c (dotted line), 10 (solid line) and 12b (dashed line) in dichloromethane measured using the saturated calomel electrode (SCE) as the reference. Adapted with permission from ref. 18 Copyright 2016 Royal Society of Chemistry.

Fig. 9 A cyclovoltammogramm of dye 13f in dichloromethane measured using Ag/AgCl/NaCl as the reference. Adapted with permission from ref. 28 Copyright 2018 American Chemical Society.

Although highly-emissive in a solution, simple DPNDs i.e.1c^30c or 1h^8d do exhibit weak fluorescence in the solid state, as these molecules mostly form H-aggregates in a crystalline state (Fig. 16, vide infra). The tetraphenylethylene moiety (TPE) seemed to be an ideal platform for inducing emission in the solid state,³⁷ thus we investigated emission properties both in the solid and aggregated states for the quadrupolar dye 46 bearing two TPE units.^30c Beside red-shifted absorption and emission in a CH₂Cl₂ solution as compared with 1c (Table 2) and 14 (Table 5), 46 emits weakly in the solid state with λ^max_em = 659 nm and Φ_fl = 0.12 (Table 5). Fluorescence properties in the aggregated state were studied in THF/water mixtures (Fig. 10). DPND 1c showed aggregation-caused quenching effect (ACQ) as fluorescence intensity measured at 523 nm lowered at water fractions equal or higher than 80%. In turn, 46 undergoes aggregation at lower water proportions probably due to increased hydrophobicity, and then modest jump in the emission intensity was observed at water contents from 70 to 80%, possibly related to aggregation effects (aggregation-induced emission (AIE)). It means that for 46 both ACQ (dominant) and AIE (weak) effects can be observed. Direct comparison between 46 and its simple analogue 14 reveals almost no enhancement in solid state emission (Table 5). Moreover, we performed DFT calculations for some pairs of DPND molecules extracted from respective X-ray crystal structures. For most of pairs, very low or zero oscillator strengths were determined for S₁ → S₀ transitions.^30c Additionally, some of them manifested CT character. In this work we proved somehow that the TPE moiety is not a magic group that always induces emission in the solid/aggregated state, but appearance of such feature depends only on which mode of crystal lattice arrangement is dominant (H- or J-aggregates).

Table 5 Photophysical properties of DPND dyes 14 and 46 in CH₂Cl₂ solution and in a solid state

Dye		λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν [cm^−1]
a Determined using a spectrofluorimeter equipped with a calibrated integrating sphere.
14	CH2Cl2	536	24 000	599	0.71	2000
	Solid	562	—	601	0.15	1200
46	CH2Cl2	553	29 000	644	0.42	2600
	Solid	574	—	659	0.12	2200

---

Fig. 10 Fluorescence spectra of 1c (a) and 46 (c) in THF–water mixtures of different relative proportions. Plots of maximum intensity vs. % water fraction (f_w) for dyes 1c (b) and 46 (d). Insets: photographs of 1c (b) and 46 (d) in THF–water mixtures with different water fractions under UV illumination (0% to 90% water fraction from left to right). Dye concentration: ∼10 mM. Adapted with permission from ref. 30c. Copyright 2018 Royal Society of Chemistry.

Probing reactivity of C=O moieties, we successfully designed and synthesized N-doped analogues of polycyclic aromatic hydrocarbons (PAHs) (48) starting from the DPND core (Fig. 11a and Table 6).³³ These propeller-shaped dyes are intensively blue in a solution while having extremely weak emission (Φ_fl < 0.001, Table 6 and Fig. 11b). Based on transient-absorption measurements we concluded that the first singlet excited state of these dyes tend to deexcite via internal conversion rather than fluorescence or triplet state formation. A moderate red-shift of both bands can be observed when OMe group is present in the chromophore structure (48avs.48b, Table 6). Due to the presence of basic nitrogen atoms, 48a was found to be acid-responsive (Fig. 11c–e). Namely, adding an increasing amount of trifluoroacetic acid (TFA) to a solution of 48a in CH₂Cl₂ results in two new absorption bands centered at 467 nm and 776 nm, while the main absorption band of pure 48a vanished substantially. Broad nature of the longer-wavelength absorption band suggests that mainly monoprotonation took place and the resulted molecule features internal donor–acceptor (D–A) character. Double protonation can be achieved by employing stronger acid – methanesulphonic acid (MsOH) (Fig. 11d). First of all, significantly smaller amount of acid is needed for monoprotonation to occur as compared with TFA (10 eq. vs. 416 eq., respectively). Similarly to TFA, upon adding 10 eq. of MsOH, a new, broad absorption band appeared around 776 nm. Higher excess of MsOH (10 eq. → 250 eq.) led to complete disappearance of a broad band above 700 nm and a new absorption band emerged at 675 nm confirming stepwise double protonation of 48a. Both protonation processes are completely reversible as by adding excess of triethylamine (Et₃N) the absorption spectrum as well as the color of 48a can be fully recovered (Fig. 11e).

Fig. 11 (a) General structure of the QIIQ skeleton; (b) absorption (solid) and emission (dotted) spectra of 48a (black), 48b (red), and 48c (blue) in toluene. Emission spectra from 570 nm excitation; (c) changes in the absorption spectra and color of 48a in CH₂Cl₂ (2.8 × 10⁻⁵ M) upon addition of MsOH (0–10 equiv.); (d) changes in the absorption spectra and color of 48a in CH₂Cl₂ (2.8 × 10⁻⁵ M) upon addition of MsOH (10–250 equiv.); (e) Photo of cuvettes containing 48a in CH₂Cl₂ solution before and after addition of a large excess of TFA (left) and MsOH (right). Adapted with permission from ref. 33. Copyright 2020 American Chemical Society.

Table 6 Photophysical properties of N-doped analogues of PAHs derived from the DPND core³³

Dye	λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν [cm^−1]
a The fluorescence quantum yield (Φfl) of 48c in toluene was obtained using Nile Blue (Φfl = 0.271 in ethanol) as a standard and corrected for refractive index differences of the solvents. Compound 48c was chosen as it is the most emissive, and Φfl of 48a and 48b were then referenced to 48c.
48a	596	29 000	679	0.00083^a	2050
48b	608	26 000	704	0.00018^a	2240
48c	597	27 000	678	0.0023^a	2000

---

In all of the above-described chromophores, a geometry of the core was influenced only by the presence of substituents at positions 6 and 12 as well as vicinity of substituents at positions 3 and 9 (an α position within a pyrrole ring). DFT calculations for some arylated DPNDs revealed that the expected deviation from the core plane mainly caused by the presence of arene rings at position 3 and 9 should be no greater that 8°.^30a In order to test how high distortion influences of the photophysical aspects, we assembled three analogues of cyclophanes based on the DPND core (1o–r)²⁰ that differ in a length of an alkane bridge. X-ray analysis of 1p revealed that the distortion from the planarity reaches 28° (Fig. 12).

Fig. 12 Single crystal structure of 1p (CCDC 2125168). Adapted with permission from ref. 20. Copyright 2022 Royal Society of Chemistry.

DFT calculation supported the obtained degree of distortion (Table 7). Interestingly, chromophore 1o (C₇ bridge) tends to increase its distortion degree upon photoexcitation while other cyclophane analogues as well as its open analogue 1o^open feature lower value of a distortion angle (Table 7). In general, 1o–r absorb at shorter wavelengths than 1c, but the most significant difference between an “open dye” and cyclophanes lies in their luminescence (Table 8). Namely, while emission bands determined for 1p–r are red-shifted by 30–40 nm as compared with 1c, in the case of 1o the emission band is located above 600 nm that is in the red region of the spectrum and particularly low values of Φ_fl were determined. In terms of emission intensity, 1p–r behave similarly to 1c (Φ_fl = 0.40–0.55), however emission bands are located at lower energies. Transient-absorption studies performed for cyclophane analogues suggested that presumably efficient formation of triplet state is responsible for distinct emission properties of 1p.

Table 7 Deviation from planarity for DPND dyes determined theoretically in the ground (S₀) and first excited state (S₁)^20 a

Dye	ϕ (S0)^a	ϕ (S1)^a	SE/kcal mol^−1
a ϕ – deviation from planarity. b Determined based on X-ray crystallography of 1p. c Determined using DFT/B3-LYP level of theory. SE – strain energy computed at MP2/cc-pVDZ level of theory.
1o^open	7.4°	5.6°	0
1o	32.8°	43.6°	31.2 (27.8^c)
1p	27.8° (28.0°^b)	25.6°	13.6 (18.0^c)
1r	15.4°	12.8°	8.2 (12.2^c)

---

Table 8 Photophysical properties of DPNDs 1c and 1o–r measured in different solvents²⁰

Dye	Solvent	λ abs [nm]	λ em [nm]	Φ fl	Δν [cm^−1]
a Absorption maximum (bold) and shoulder. b Fluorescence maximum (bold) and shoulder. c Relative Φfl were obtained using Rhodamine 6G in ethanol (ΦPL = 0.95, λexc = 480 nm) as a reference.
1c	PhMe	472, 506	526, 560	0.66	750
	THF	470, 502	524, 558	0.57	800
	C6H5CN	476, 508	531, 564	0.65	850
	MeCN	469, 499	527, 557	0.54	1100
1o	PhMe	492	618	0.06	4100
	THF	488	616	0.04	4300
	C6H5CN	493	625	0.04	4300
	MeCN	487	623	0.02	4500
1p	PhMe	483, 507	552, 577	0.55	2600
	THF	482, 503	573, 553	0.41	3300
	C6H5CN	489, 507	562, 581	0.40	3200
	MeCN	481	577	0.32	3500
1r	PhMe	479, 507	543, 572	0.57	2500
	THF	478, 505	542, 572	0.48	2500
	C6H5CN	483, 509	551, 576	0.52	2600
	MeCN	477, 500	572, 550	0.40	3500

---

Dipyrrolonaphthyridinediones versus other cross-conjugated chromophores

Although cross-conjugated, the DPND core is different from well-known dyes of this type such as isoindigo (II),³⁸ diketopyrrolopyrrole,³⁹ or cibalakrot¹² (Fig. 13) as it contains a “pure” pyrrole ring as an electron-donating moiety. This factor directly translates into particularly distinct one-photon optoelectronic behavior compared with other cross-conjugated structures (Table 9 and Fig. 13). According to the data collected in Table 9, among all dyes B-II-B is weakly- or non-fluorescent,⁴⁰ features the lowest molar absorption coefficient (ε_max) and absorbs at shorter wavelength (λ^max_abs = 517 nm) than 14. Compound 14 emits light with Φ_fl = 0.71 with a maximum noted at λ^max_em = 599 nm that is most red-shifted compared with other cross-conjugated chromophores, although Ph-DPP-Ph has a larger value of Stokes' shift.

Fig. 13 Correlation diagram of the HOMOs and LUMOs of three representative cross-conjugated units and a DPND derivative 14 determined by cyclic voltammetry (CV) measurements in CH₂Cl₂ (the values in parentheses were determined using DFT calculations due to lack of the experimental data).

Table 9 Comparison of absorption and emission properties of selected dyes bearing cross-conjugated structure

Dye	Solvent	λ ^maxabs [nm]	ε max [M^−1 cm^−1]	λ ^maxem [nm]	Φ fl	δν [cm^−1]	Ref.
a Estimated from Fig. 3 in ref. 12.
Ph-DPP-Ph	PhMe	480	12 600	535, 575	0.85	3400	39
Ph-II-Ph	PhMe	517	10 700	—	<0.1	—	38
Ph-BAI-Ph	PhMe	552	∼33 000^a	575	0.77	720	12
14	CH2Cl2	536	24 000	599	0.71	2000	30c

---

Direct comparison of electrochemically-derived energetic levels may be useful from the view-point of optoelectronics (Fig. 13). DPND 14 is the most susceptible to oxidation as its E_HOMO level lies higher in energy that those noted for other cross-conjugated chromophores. When it comes to E_LUMO levels, 14 features a similar value to Ph-DPP-Ph and at even lower energies than that of Ph-II-Ph which are both considered as electron-acceptors in dyes tested toward organic electronics.

DPNDs as singlet-fission (SF) materials

The singlet fission (SF) is a process in which a higher-energy singlet exciton, typically generated by the absorption of a photon, is converted into two lower-energy triplet excitons.⁴¹ This phenomenon features the potential to enhance the efficiency of solar cells⁴² and other optoelectronic devices as it seems to be an attractive solution to overcome Shoeckley-Queisser limit.⁴³ Taking a lesson from cibalakrot analogues, Wang and co-workers have investigated^8d the possibility of using 1h as a potential SF material. First of all, the authors found that 1h exhibits phosphorescence at 1019 and 1034 nm both in a solution and aggregated state, respectively (Fig. 14), while fluoresces weakly in the aggregated state. This result suggests that under slightly endothermic energetic conditions, a SF process can occur in polycrystalline film of 1h that outcompetes other energy dissipation channels suggesting strong intermolecular coupling between molecules in the solid state. The authors proved that the SF process takes place in 1h with up to 173% triplet yield and its film exhibits excellent stability upon exposure to air and light compared with pentalene⁴⁴ or cibalakrot analogues (Fig. 15).⁴⁵

Fig. 14 Normalized absorption (black) and fluorescence (red) spectra of DPND in CH₂Cl₂ solution (10⁻⁵ M) and solid thin film (100 nm), as well as sensitized phosphorescence spectra (blue) of 1h in polystyrene (PS) matrix (DPND : PtTPBP : PS = 1 : 5 : 94) and doped thin film (DPND : PtTPBP = 95 : 5). Inset: sample photographs of solution and thin film under 365 nm UV light. Adapted with permission from ref. 8d. Copyright 2020 American Chemical Society.

Fig. 15 Stability testing of 1h molecule: (a) steady-state absorption spectra and (b) normalized intensity of the λ_max of thin films exposed to air and light. Adapted with permission from ref. 8d. Copyright 2020 American Chemical Society.

In their follow-up work, the authors investigated¹⁹ in detail effect of molecular aggregation on SF dynamics. Specifically, the DPND molecule described above contains two longer alkyl chains (1h – DPND6 in Fig. 16) which directly influences molecular arrangement in the crystal lattice with specific values of π–π distance, transverse and longitudinal offsets as well as a slipping angle (Fig. 16). In turn, compound 1a (DPND) crystallizes in a nearly cofacial pattern of packing arrangement with the smaller values of transverse and longitudinal offsets, π–π distance, and a slipping angle of 66°. Here, the formation of face-to-face dimers results in stronger coupling of molecules in the crystal lattice thus an accelerated singlet fission (SF) process was eventually observed. This result may help in the future to design more efficient SF materials.

Fig. 16 Molecular arrangement of 1a and 1h in the solid state. Adapted with permission from ref. 19. Copyright 2021 American Chemical Society.

Recently, Wu and others have presented new insights into the SF mechanism by studying time-resolved spectroscopy as a function of temperature for 10.⁴⁶ Here, the applied kinetic model developed based on the measurements and calculations includes initial conversion of S₁ state to ¹(T₁T₁) state followed by thermally-activated at room temperature dissociation of ¹(T₁T₁) into two T₁ states (Fig. 17). The yield of T₁ states formation was determined as 154%.

Fig. 17 Kinetic model of the SF process determined for 10. Adapted with permission from ref. 46. Copyright 2021 American Chemical Society.

Fluorescent nitroaromatics

Nitroaromatics fluoresce occasionally as numerous possible energy dissipation channels are activated after the excitation.⁴⁷ The NO₂ auxochrome is one of the strongest electron-withdrawing moieties which ideally should improve stability and performance of NO₂-containing materials. However, the introduction of this group into the chromophore structure often results in weak emission. Although the literature outlines some strategies for enhancing emission properties of nitroaromatics,^47a there are still no general rules how to make nitroaromatics glow. To gain more insight into the processes that are responsible for energy dissipation in nitroaromatics, we have studied a number of these molecules where the DPND core was chosen as a reference chromophore (Table 10, Fig. 18–20).^30b,48 The introduction of the nitro groups at the peripheries undoubtedly implies emission intensity loss (from 0.71 for 14 to 0.41 for 25 in CH₂Cl₂), however 25 still fluoresce in the red region of the spectrum featuring red-shifted both absorption and emission bands. This initial result set the stage for the comprehensive study on fluorescence of nitroaromatics bearing the DPND core.

Table 10 Optical properties of nitro-decorated DPNDs

Dye	Solvent	λ abs [nm]	λ em [nm]	Φ fl	Δν [cm^−1]
25	o-C6H4Cl2	569	605	0.45	1050
	CH2Cl2	562	601	0.41	1150
	MeCN	559	599	0.07	1190
25^mono	o-C6H4Cl2	539	571	0.76	1050
	CH2Cl2	534	568	0.36	1110
	MeCN	529	570	0.006	1370
26	o-C6H4Cl2	536	580	0.61	1410
	CH2Cl2	533	578	0.49	1460
	MeCN	529	579	0.013	1630
27	o-C6H4Cl2	547	612	0.46	1940
	CH2Cl2	543	610	0.21	2040
	MeCN	542	604	0.005	1910
28	o-C6H4Cl2	524	560	0.96	1230
	CH2Cl2	521	560	0.34	1340
	MeCN	518	564	0.005	1570
29	o-C6H4Cl2	557	607	0.58	1480
	CH2Cl2	549	605	0.45	1690
	MeCN	545	608	0.24	1900
30	o-C6H4Cl2	556	606	0.58	1480
	CH2Cl2	549	604	0.44	1660
	MeCN	546	603	0.31	1890
31	o-C6H4Cl2	558	606	0.45	1400
	CH2Cl2	548	602	0.31	1640
	MeCN	543	606	0.005	1940
32	o-C6H4Cl2	555	602	0.006	1390
	CH2Cl2	549	608	0.003	1760
	MeCN	542	584	0.003	1330
33	o-C6H4Cl2	563	618	0.38	1580
	CH2Cl2	555	615	0.28	1760
	MeCN	550	608	0.05	1730
35	o-C6H4Cl2	551	582	0.49	970
	CH2Cl2	544	577	0.28	1050
	MeCN	541	575	0.04	1090
35^mono	o-C6H4Cl2	529	555	0.62	870
	CH2Cl2	524	551	0.09	940
	MeCN	520	548	0.02	1000
36	o-C6H4Cl2	527	558	0.95	1050
	CH2Cl2	523	555	0.92	1100
	MeCN	521	553	0.43	1110
37	o-C6H4Cl2	532	583	0.61	1640
	CH2Cl2	525	584	0.52	1920
	MeCN	524	582	0.12	1900
38	o-C6H4Cl2	525	568	0.76	1440
	CH2Cl2	522	566	0.58	1490
	MeCN	520	569	0.075	1660
39	o-C6H4Cl2	535	586	0.63	1630
	CH2Cl2	531	582	0.59	1650
	MeCN	528	590	0.20	1990
40	o-C6H4Cl2	547	593	0.69	1420
	CH2Cl2	541	593	0.46	1620
	MeCN	537	592	0.17	1730
41	o-C6H4Cl2	553	604	0.64	1530
	CH2Cl2	546	604	0.43	1760
	MeCN	543	609	0.18	2000
42	o-C6H4Cl2	575	669	0.12	2440
	CH2Cl2	565	670	0.02	2770
	MeCN	557	688	0.04	3420
43	o-C6H4Cl2	555	598	598	980
	CH2Cl2	548	592	592	1160
	MeCN	544	592	592	1490
43^mono	o-C6H4Cl2	532	575	—	1420
	CH2Cl2	527	552	—	1020
	MeCN	522	561	—	1360
44	o-C6H4Cl2	551	595	0.06	1360
	CH2Cl2	545	594	—	1500
	MeCN	542	589	—	1480
44^mono	o-C6H4Cl2	527	569	0.02	1400
	CH2Cl2	524	551	—	940
	MeCN	521	565	—	1500
45^mono	o-C6H4Cl2	527	569	—	1400
	CH2Cl2	524	562	—	1290
	MeCN	522	559	—	1270

---

Fig. 18 Comparison of Φ_fl for chosen nitroaromatics bearing the DPND core in three different solvents: 1,2-dichlorobenzene (green), CH₂Cl₂ (blue) and MeCN (orange). Adapted with permission from ref. 48. Copyright 2023 Royal Society of Chemistry.

Fig. 19 Schematic diagram showing the scenario of temporal evolution of the photoexcited ortho-nitrophenyl substituted DPND. The electrons are initially photoexcited to locally excited (LE) state (¹π–π*) and is non-adiabatically transferred to the CT state along the reaction coordinate. The photoexcited system recombines radiatively from S₁ to S₀ or nonradiatively through ¹π–π* → ¹CT → S₀ transitions. CI – conical intersection. Adapted with permission from ref. 30b. Copyright 2021 Royal Society of Chemistry.

Fig. 20 Schematic diagram showing the general scenario of temporal evolution of the photoexcited meta- and para-nitrophenyl substituted DPNDs. The system is initially photoexcited to the ¹π–π* state and is non-adiabatically transferred to the ¹n–π* state along the reaction coordinate. The photoexcited system recombines radiatively from S₁ to S₀ or nonradiatively through ¹π–π* → ¹n–π* → S₀ transitions. CI – conical intersection. Adapted with permission from ref. 30b. Copyright 2021 Royal Society of Chemistry.

In general, two structural factors play a decisive role in governing emission intensity: (1) the position of NO₂ group relative to the core and (2) the substitution pattern of flanking aryl rings (Table 10, Fig. 18). Regarding the first factor, both para- and meta-NO₂-subsituted derivatives are fluorescent and emission intensity lowers with the increase in solvent polarity, suggesting that a dark, charge-transfer-type (CT-type) state may be involved in energy dissipation. In contrast, close proximity of the nitro group (43–45, Scheme 4) results in weak fluorescence response (ortho- or peri-NO₂-substituted molecules) in all tested solvents.

Secondly, the introduction of additional substituents (Me, OMe, NEt₂) within flanking arene rings also contribute significantly to fluorescence modulation. As an example, placing either Me or OMe group at the ortho position relative to the DPND core increases emission intensity (i.e.35vs.36) by hindering rotation around C_aryl–C_DPND bond. Those groups may also affect electronic distribution via both inductive and mesomeric effects. As a matter of fact, careful structure optimization by taking into account all of these factors allowed us to describe nitroaromatics featuring enormous value of Φ_fl (up to 0.96), even in relatively polar dichloromethane (28 or 36).

The origin of the specific dependence of emission intensity on solvent polarity was investigated^30b by means of theoretical calculations which revealed surprising, aborted photochemical reactivity after photoexcitation (Fig. 19 and 20). The fluorescence in the series of DPND-based nitroaromatics clearly comes from deexcitation of the lowest in energy S₁ (¹π–π*) state. However, the S₁ state may also adiabatically relaxed via a transition to ¹CT (ortho- or peri-NO₂ isomers) or ¹np* (para- or meta-NO₂ isomers) states. Then, these states efficiently undergo non-radiative depopulation via a conical intersection (CI) with the S₀ state.

Theoretical results also suggest that the nitro group twists upon photoexcitation around the C_DPND–N_nitro bond with subsequent formation of a new, covalent C_DPND–O_nitro bond in ¹CT and ¹np* states. The formation of a new bond with more polar character tends to occur much easier in polar environment. Then, this bond breaks when transitioning from the CI to the more favourable ground-state configuration during relaxation. This phenomenon is called the aborted photochemistry (Fig. 19 and 20).⁴⁹

Two-photon absorption (TPA) in DPNDs

Two-photon absorption (TPA) phenomenon⁵⁰ have found multiple, real-life applications such as 3D imaging of materials,⁵¹ microfabrication and photopolymerization,⁵² photopharmacology⁵³ or imaging of biological tissues.⁵⁴ An efficient TPA absorber should feature a long system of conjugated multiple bonds and contain donor and acceptor groups that potentially would participate in the formation of strongly polarized charge-transfer states.⁵⁵ As a completely new chromophore, the DPND core was investigated as a TPA absorber for the first time already in 2018.²⁷ We have found out that dyes bearing double- or triple-bond linkages between the DPND core and flanking aryl rings exhibit superior values of two-photon absorption cross-sections (σ^max₂) (Table 11), where the highest value of σ^max₂ was noted for compound 11a bearing a double-bond linkage between the DPND core and nitroaryl flanking moieties (σ^max₂ = 5180 GM). Notably, while compounds from both series (11 and 12) possessed a beneficial ratio of σ^max₂/MW (1.6–9.8 GM g⁻¹), dyes of type 12 show high values of two-photon brightness σ^max₂·Φ_fl, which is an important feature from the viewpoint of bioimaging applications.^55b

Table 11 Two-photon absorption data for chosen, nitro-decorated DPNDs in CH₂Cl₂

Dye	λ ^max2PA/nm	σ ^max2/GM	σ ^max2 φ fl/GM	σ ^max2/MW^c
a Determined using the two-photon excited fluorescence (TPEF) technique.^56 b Determined using Z-scan setup.^57 c The molecular weights were calculated by replacing alkyl groups at positions 6 and 12 by a methyl group (CH3).
1c	750	44	32	<1
11a	820	5180	83	9.3
11b	740	5100	112	9.8
11c	740	1710	70	3.2
12a	720	2840	1450	5.1
12b	720	850	500	1.6
12c	860	1990	340	3.6
25	<690	>3718	>1524	7.3
26	<680	312	153	<1
27	<690	>835	>175	1.5
28	<680	>197	>67	<1
29	<690	>1177	>530	2.2
30	<690	>1326	>583	2.3
31	<690	>1134	>352	1.8
32	<690	>9	0	0
33	720	1384	388	2.1
35	<680	>233	>65	<1
36	<685	>175	>161	<1
37	<685	>167	>87	<1
38	<685	>96	>56	<1
39	<685	≥161	≥95	<1
40	<680	>340	>156	<1
41	<700	≥513	≥221	<1
42	755	104	2	<1

---

Later on, we carefully evaluated weakly-coupled bis-arylated DPND derivatives containing nitroaryl moieties as potential TPA absorbers (Scheme 4 and Table 11).⁴⁸ Weak electronic communication between the nitro group auxochrome and the DPND core resulted in weaker non-linear response compared with 11a or 12a. Among compounds tested, 25 showed a reasonable TPA response (σ^max₂ > 1524 GM) whereas other derivatives performed weaker (σ^max₂ < 600 GM) in the accessible spectral window size.⁴⁸

DPNDs as triplet sensitizers

Halogenation is the most established strategy to intensify population of a triplet excited state⁵⁸ due to enhanced spin–orbit coupling (SOC).⁵⁹ Such halogenated derivatives have been widely studied and as emerging agents for photodynamic therapy (PDT). Along this strategy, Ayitou and others tested halogenated DPNDs (7–9) as potential (PDT) agents (Fig. 21 and Table 12).^21,60 The presence of heavy halogen atoms within the DPND core was believed to enhance spin–orbit coupling (SOC) dynamics, thus leading to efficient singlet oxygen (¹O₂) production. They found that all halogen-substituted dyes do produce ¹O₂ with admirable efficiencies and the highest value of Φ_Δ was determined for 9. Surprisingly, values of fluorescence quantum yields measured for 7c and 8 (75 and 76%, respectively) are comparable to the value measured for 1c, although they bear chlorine and bromine atoms within the core.

Fig. 21 Schematic illustration of energy levels and triplet sensitization in 7–9.^21,60

Table 12 Summary of the lifetimes, rate constants of these processes extracted from the ns-TA and fs-TA experiments

Dye	τ fl/ns	k ISC/s^−1	τ T/μs	τ Phos/ms	Φ Δ/%
7a	5.1	—	5.1	8.2	31
8	5.7	1.6 × 10^8	16	7.9	28
9	0.34	2.9 × 10^9	0.31	9.0	52

---

DPND-based polymer as a potential material for OFETs

Direct arylation methodology appeared to be useful in the preparation of polymer 49 in one step from the DPND core and dibrominated derivative bearing the diketopyrrolopyrrole (DPP) unit (Fig. 22).³² The weight-average molecular weight (M_w) and polydispersity index (PD) determined by HT-GPC were 77.7 kDa and 2.64, respectively, while the UV-VIS spectrum recorded in toluene showed a strong absorption band in the NIR region with λ_max = 866 nm. Polymer 49 was examined by BASF as a potential electron transport material in OFETs, nevertheless a particularly low value of electron mobility (μ = 8.85 × 10⁻¹² cm² V⁻¹ m⁻¹) was measured.

Fig. 22 The structure of polymer 49.

5. Summary and outlook

Although the history of cross-conjugated dyes dates back to the ancient times, still there are numerous possible structures that have never been either assembled or investigated. In seven years which elapsed since their discovery, dipyrrolonaphthyridinediones made a notable impact on chemistry. Their straightforward synthesis combined with functionalization possibilities are responsible for the quick initial progress in terms of structural exploitation. In the second phase the deciding factor turned out to be their intriguing photophysical characteristics and in particular: (1) strong and easily tunable emission; (2) low energy of T₁ excited state which enables singlet fission in the crystalline state; (3) large two-photon absorption cross-sections after derivatization. Studies on dipyrrolonaphthyridinediones: (1) were the deciding factor enabling formulation of strategies towards strongly fluorescent nitroaromatics; (2) provided the shortest pathway to twistacenes; (3) delivered dyes with large optical brightness and exceptionally high values of two-photon absorption cross-section divided by molecular mass. Notwithstanding their success, there are still numerous synthetic limitations which have to be overcome in the nearest future to ensure continuous progress. The main envisioned areas for synthetic explorations are as follows: (1) preparation of DPNDs directly from aromatic acids; (2) extension of the DPNDs' synthesis to embrace such substrates as indole, isoindole etc.; (3) transformation of DPNDs into polar, water-soluble probes. In 2016 it would be impossible to imagine that only seven years later the chemistry of DPNDs would expand to create independent field of study within the chemistry of functional dyes. One can venture hypothesis that once these synthetic limitations are overcome these new π-expanded DPNDs could witness rapid development. The analysis of research performed within the last few years suggest that the number of known as well as possible structures of cross-conjugated nature is limited only by our imagination.

Author contributions

All authors contributed to the conceptualization. B. S. performed literature search, wrote the first draft and made all schemes/figures. All authors were involved in revising, editing, and proofreading.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The work was financially supported by the Polish National Science Centre (Sonata 2021/43/D/ST4/02267 and OPUS 2020/37/B/ST4/00017). This project has received funding from EU's Horizon 2020 research and innovation programme under Grant Agreement No. 860762. Authors thank Dr Marek Grzybowski for discovering DPNDs.

References

1. (a) N. Singh, S. Singh, S. Kohli, A. Singh, H. Asiki, G. Rathee, R. Chandra and E. A. Anderson, Org. Chem. Front., 2021, 8, 5550–5573; (b) J. T. Gupton, in Heterocyclic Antitumor Antibiotics, ed. M. Lee, Springer Berlin Heidelberg, Berlin, Heidelberg, 2006, pp. 53–92.
2. (a) V. Bhardwaj, D. Gumber, V. Abbot, S. Dhiman and P. Sharma, RSC Adv., 2015, 5, 15233–15266; (b) G. Li Petri, V. Spanò, R. Spatola, R. Holl, M. V. Raimondi, P. Barraja and A. Montalbano, Eur. J. Med. Chem., 2020, 208, 112783.
3. (a) J. Kim, J. Oh, A. Osuka and D. Kim, Chem. Soc. Rev., 2022, 51, 268–292; (b) B. Szyszko, M. J. Białek, E. Pacholska-Dudziak and L. Latos-Grażyński, Chem. Rev., 2017, 117, 2839–2909.
4. R. Orłowski, D. Gryko and D. T. Gryko, Chem. Rev., 2017, 117, 3102–3137.
5. B. Sadowski, J. Klajn and D. T. Gryko, Org. Biomol. Chem., 2016, 14, 7804–7828.
6. (a) C. Bellomo, M. Chaari, J. Cabrera-González, M. Blangetti, C. Lombardi, A. Deagostino, C. Viñas, N. Gaztelumendi, C. Nogués, R. Nuñez and C. Prandi, Chem.–Eur. J., 2018, 24, 15622–15630; (b) A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932.
7. (a) A. N. Bismillah and I. Aprahamian, Chem. Soc. Rev., 2021, 50, 5631–5649; (b) S. Boodts, E. Fron, J. Hofkens and W. Dehaen, Coord. Chem. Rev., 2018, 371, 1–10; (c) I.-S. Tamgho, A. Hasheminasab, J. T. Engle, V. N. Nemykin and C. J. Ziegler, J. Am. Chem. Soc., 2014, 136, 5623–5626.
8. (a) W. Zeng, O. El Bakouri, D. W. Szczepanik, H. Bronstein and H. Ottosson, Chem. Sci., 2021, 12, 6159–6171; (b) W. Zeng, D. W. Szczepanik and H. Bronstein, J. Phys. Org. Chem., 2023, 36, e4441; (c) M. Swart, Theor. Chem. Acc., 2020, 139; (d) L. Wang, L. Lin, J. Yang, Y. Wu, H. Wang, J. Zhu, J. Yao and H. Fu, J. Am. Chem. Soc., 2020, 142, 10235–10239.
9. (a) A. Borissov, Y. K. Maurya, L. Moshniaha, W.-S. Wong, M. Żyła-Karwowska and M. Stępień, Chem. Rev., 2022, 122, 565–788; (b) M. Stępień, E. Gońka, M. Żyła and N. Sprutta, Chem. Rev., 2017, 117, 3479–3716.
10. M. Grzybowski and D. T. Gryko, Adv. Opt. Mater., 2015, 3, 280–320.
11. R. Stalder, J. Mei, K. R. Graham, L. A. Estrada and J. R. Reynolds, Chem. Mater., 2014, 26, 664–678.
12. J. Kaleta, M. Dudič, L. Ludvíková, A. Liška, A. Zaykov, I. Rončević, M. Mašát, L. Bednárová, P. I. Dron, S. J. Teat and J. Michl, J. Org. Chem., 2023, 88, 6573–6587.
13. (a) A. D. Hendsbee, J.-P. Sun, L. R. Rutledge, I. G. Hill and G. C. Welch, J. Mater. Chem. A, 2014, 2, 4198–4207; (b) M. Vatanparast and Z. Shariatinia, Sol. Energy, 2021, 230, 260–268; (c) B. He, A. B. Pun, D. Zherebetskyy, Y. Liu, F. Liu, L. M. Klivansky, A. M. McGough, B. A. Zhang, K. Lo, T. P. Russell, L. Wang and Y. Liu, J. Am. Chem. Soc., 2014, 136, 15093–15101.
14. C. Du, S. Fu, X. Ren, X. Wang, Z. Wang, J. Zhou and H. Wang, New J. Chem., 2018, 42, 3493–3502.
15. (a) E. D. Głowacki, G. Voss, L. Leonat, M. Irimia-Vladu, S. Bauer and N. S. Sariciftci, Isr. J. Chem., 2012, 52, 540–551; (b) R. J. H. Clark, C. J. Cooksey, M. A. M. Daniels and R. Withnall, Endeavour, 1993, 17, 191–199.
16. D. G. Farnum, G. Mehta, G. G. I. Moore and F. P. Siegal, Tetrahedron Lett., 1974, 15, 2549–2552.
17. (a) J. Pfenninger, A. Iqbal, A. C. Rochat, O. Wallquist and A. G. Ciba-Geigy, Eur. pat.Appl. 184982, 1986; (b) A. Iqbal and L. Cassar, Ciba-Geigy Corporation, US Pat.4415685, 1983; (c) L. Cassar, A. Iqbal, A. C. Rochat and A. G. Ciba-Geigy, Eur. Pat.Appl. 98808, 1983.
18. M. Grzybowski, I. Deperasińska, M. Chotkowski, M. Banasiewicz, A. Makarewicz, B. Kozankiewicz and D. T. Gryko, Chem. Commun., 2016, 52, 5108–5111.
19. L. Wang, W. Cai, J. Sun, Y. Wu, B. Zhang, X. Tian, S. Guo, W. Liang, H. Fu and J. Yao, J. Phys. Chem. Lett., 2021, 12, 12276–12282.
20. B. Sadowski, D. Mierzwa, S. Kang, M. Grzybowski, Y. M. Poronik, A. L. Sobolewski, D. Kim and D. T. Gryko, Chem. Commun., 2022, 58, 3697–3700.
21. J. Morgan, Y. J. Yun, A. M. Jamhawi, S. M. Islam and A. J.-L. Ayitou, Photochem. Photobiol., 2023, 99, 761–768.
22. C. Song, D. W. Knight and M. A. Whatton, Tetrahedron Lett., 2004, 45, 9573–9576.
23. (a) J. I. Ayogu and E. A. Onoabedje, Catal. Sci. Tech., 2019, 9, 5233–5255; (b) K. H. Shaughnessy, in Metal-Catalyzed Reactions in Water, Wiley, 2013, pp. 1–46; (c) R. Chinchilla and C. Nájera, Chem. Rev., 2007, 107, 874–922; (d) S. E. Hooshmand, B. Heidari, R. Sedghi and R. S. Varma, Green Chem., 2019, 21, 381–405.
24. (a) L. I. Belen’kii, T. G. Kim, I. A. Suslov and N. D. Chuvylkin, Arkivoc, 2003, 59–67; (b) S. Nomiyama, T. Ogura, H. Ishida, K. Aoki and T. Tsuchimoto, J. Org. Chem., 2017, 82, 5178–5197; (c) D. Ghorai and G. Mani, Inorg. Chem., 2014, 53, 4117–4129; (d) T. Buchała, A. Chudoba and S. Roszak, Struct. Chem., 2016, 27, 185–189.
25. (a) Y. Gao, C. Feng, T. Seo, K. Kubota and H. Ito, Chem. Sci., 2022, 13, 430–438; (b) B. Godlewski, D. Baran, M. De Robichon, A. Ferry, S. Ostrowski and M. Malinowski, Org. Chem. Front., 2022, 9, 2396–2404.
26. T. Rohand, W. Qin, N. Boens and W. Dehaen, Eur. J. Org Chem., 2006, 4658–4663.
27. B. Sadowski, H. Kita, M. Grzybowski, K. Kamada and D. T. Gryko, J. Org. Chem., 2017, 82, 7254–7264.
28. B. Sadowski, M. Loebnitz, D. R. Dombrowski, D. H. Friese and D. T. Gryko, J. Org. Chem., 2018, 83, 11645–11653.
29. P. G. Chirila and C. J. Whiteoak, Dalton Trans., 2017, 46, 9721–9739.
30. (a) B. Sadowski, M. F. Rode and D. T. Gryko, Chem.–Eur. J., 2018, 24, 855–864; (b) B. Sadowski, M. Kaliszewska, Y. M. Poronik, M. Czichy, P. Janasik, M. Banasiewicz, D. Mierzwa, W. Gadomski, T. D. Lohrey, J. A. Clark, M. Lapkowski, B. Kozankiewicz, V. I. Vullev, A. L. Sobolewski, P. Piatkowski and D. T. Gryko, Chem. Sci., 2021, 12, 14039–14049; (c) B. Sadowski, S.-H. Su, T.-C. Lin, T. D. Lohrey, I. Deperasińska, P.-T. Chou and D. T. Gryko, J. Mater. Chem. C, 2018, 6, 12306–12313.
31. (a) H. Bohra and M. Wang, J. Mater. Chem. A, 2017, 5, 11550–11571; (b) X. Wang, Y. Li, J. Li, Y. Zhang, J. Shao and Y. Li, Molecules, 2023, 28, 3515; (c) I. A. Stepek and K. Itami, ACS Mater. Lett., 2020, 2, 951–974.
32. M. Grzybowski, D. T. Gryko, B. Sadowski, K. Strassel, D. Kaelblein and P. Hayoz, Polymers and compounds based on dipyrrolo[1,2-b:1’,2’-g][2,6]naphthyridine-5,11-dione, International Pat no. WO2017068009, 2018.
33. B. Sadowski, D. J. Stewart, A. T. Phillips, T. A. Grusenmeyer, J. E. Haley, T. M. Cooper and D. T. Gryko, J. Org. Chem., 2020, 85, 284–290.
34. S. Ito, S. Hiroto and H. Shinokubo, Org. Lett., 2013, 15, 3110–3113.
35. J. O. Morley, Int. J. Quantum Chem., 1993, 46, 19–26.
36. P. J. McCartin, J. Chem. Phys., 2004, 42, 2980–2981.
37. (a) Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Commun., 2009, 4332–4353; (b) Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Soc. Rev., 2011, 40, 5361.
38. L. A. Estrada, R. Stalder, K. A. Abboud, C. Risko, J.-L. Brédas and J. R. Reynolds, Macromolecules, 2013, 46, 8832–8844.
39. J. Dhar, N. Venkatramaiah and S. Patil, J. Mater. Chem. C, 2014, 2, 3457–3466.
40. (a) D. Bialas, S.-L. Suraru, R. Schmidt and F. Würthner, Org. Biomol. Chem., 2011, 9, 6127–6132; (b) S. Luňák, P. Horáková and A. Lyčka, Dyes Pigm., 2010, 85, 171–176.
41. (a) M. B. Smith and J. Michl, Chem. Rev., 2010, 110, 6891–6936; (b) M. B. Smith and J. Michl, Annu. Rev. Physiol., 2013, 64, 361–386.
42. J. Xia, S. N. Sanders, W. Cheng, J. Z. Low, J. Liu, L. M. Campos and T. Sun, Adv. Mater., 2017, 29, 1601652.
43. A. Rao and R. H. Friend, Nat. Rev. Mater., 2017, 2, 17063.
44. L. Wang, Y. Wu, Y. Liu, L. Wang, J. Yao and H. Fu, J. Chem. Phys., 2019, 151.
45. K. J. Fallon, P. Budden, E. Salvadori, A. M. Ganose, C. N. Savory, L. Eyre, S. Dowland, Q. Ai, S. Goodlett, C. Risko, D. O. Scanlon, C. W. M. Kay, A. Rao, R. H. Friend, A. J. Musser and H. Bronstein, J. Am. Chem. Soc., 2019, 141, 13867–13876.
46. Y. Wu, L. Lu, B. Yu, S. Zhang, P. Luo, M. Chen, J. He, Y. Li, C. Zhang, J. Zhu, J. Yao and H. Fu, J. Phys. Chem. Lett., 2023, 14, 4233–4240.
47. (a) Y. M. Poronik, B. Sadowski, K. Szychta, F. H. Quina, V. I. Vullev and D. T. Gryko, J. Mater. Chem. C, 2022, 10, 2870–2904; (b) M.-C. Chen, D.-G. Chen and P.-T. Chou, ChemPlusChem, 2021, 86, 11–27; (c) W. Rodriguez-Cordoba, L. Gutierrez-Arzaluz, F. Cortes-Guzman and J. Peon, Chem. Commun., 2021, 57, 12218–12235.
48. B. Sadowski, M. Kaliszewska, G. Clermont, Y. M. Poronik, M. Blanchard-Desce, P. Piatkowski and D. T. Gryko, Chem. Commun., 2023, 59, 11708–11711.
49. (a) A. Sinicropi, W. M. Nau and M. Olivucci, Photochem. Photobiol., 2002, 1, 537–546; (b) D. Shemesh, A. L. Sobolewski and W. Domcke, J. Am. Chem. Soc., 2009, 131, 1374–1375; (c) D. Tuna, A. L. Sobolewski and W. Domcke, Phys. Chem. Chem. Phys., 2014, 16, 38–47.
50. M. Göppert-Mayer, Ann. Phys., 1931, 401, 273–294.
51. C. Dorfer, D. Hits, L. Kasmi, G. Kramberger, M. Lucchini, M. Mikuž and R. Wallny, Appl. Phys. Lett., 2019, 114.
52. (a) S. Maruo, O. Nakamura and S. Kawata, Opt. Lett., 1997, 22, 132–134; (b) Z. Faraji Rad, P. D. Prewett and G. J. Davies, Microsyst. Nanoeng., 2021, 7, 71; (c) S. O'Halloran, A. Pandit, A. Heise and A. Kellett, Adv. Sci., 2023, 10, 2204072.
53. M. Izquierdo-Serra, M. Gascón-Moya, J. J. Hirtz, S. Pittolo, K. E. Poskanzer, È. Ferrer, R. Alibés, F. Busqué, R. Yuste, J. Hernando and P. Gorostiza, J. Am. Chem. Soc., 2014, 136, 8693–8701.
54. (a) F. Helmchen and W. Denk, Nat. Methods, 2005, 2, 932–940; (b) G. Sancataldo, O. Barrera and V. Vetri, in Principles of Light Microscopy: From Basic to Advanced, ed. V. Nechyporuk-Zloy, Springer International Publishing, Cham, 2022, pp. 215–241.
55. (a) M. Pawlicki, H. A. Collins, R. G. Denning and H. L. Anderson, Angew. Chem., Int. Ed., 2009, 48, 3244–3266; (b) H. Myung Kim and B. R. Cho, Chem. Commun., 2009, 153–164.
56. (a) M. A. Albota, C. Xu and W. W. Webb, Appl. Opt., 1998, 37, 7352–7356; (b) C. Xu and W. W. Webb, J. Opt. Soc. Am. B, 1996, 13, 481–491.
57. K. Kamada, K. Matsunaga, A. Yoshino and K. Ohta, J. Opt. Soc. Am. B, 2003, 20, 529–537.
58. (a) J. Grüne, G. Londi, A. J. Gillett, B. Stähly, S. Lulei, M. Kotova, Y. Olivier, V. Dyakonov and A. Sperlich, Adv. Funct. Mater., 2023, 33, 2212640; (b) W. Hu, R. Zhang, X.-F. Zhang, J. Liu and L. Luo, Part, 2022, 272, 120965.
59. (a) M. Fagnoni, Angew. Chem., Int. Ed., 2010, 49, 6709–6710; (b) C. M. Marian, Wiley Interdiscip. Rev. Comput. Mol. Sci., 2012, 2, 187–203.
60. J. Morgan, Y. J. Yun and A. J.-L. Ayitou, Photochem. Photobiol., 2022, 98, 57–61.

---