Synthesis of polyacene by using a metal–organic framework

The acene series, an important class of linearly polycyclic aromatic hydrocarbons, are of interest owing to their unique physicochemical features. With an increase in the number of fused benzene rings, acenes display an evolution of electronic structure and properties. Thus, efforts have been devoted to the synthesis of longer acenes, with dodecacene being the longest acene (12 fused benzene rings) reported to date. However, the formation of polymeric acenes with numerous benzene rings, namely polyacene, has yet to be realized. Herein, we present a methodology for the synthesis of polyacene mediated by a metal–organic framework. Nanoconfined synthesis of precursor polymers in the channels of the metal–organic framework and the subsequent dehydro-aromatization reaction produced polyacene that was overwhelmingly longer than the previously reported acenes. The scalable synthesis of polyacene allowed us to unveil the stability and electronic properties of polyacene, paving the way for their widespread applications in optoelectronic and magnetic devices. Elongation of acenes with linearly fused benzene rings is a synthetic challenge. Now, the use of a metal–organic framework enables the synthesis of polymeric precursors, which undergo thermal transformation to form polyacenes.

Since the synthesis of pentacene was reported in 1912 (ref. 1), extension of linearly fused benzene rings has attracted research attention because of curiosity around the nature of aromatic molecules and their applications in optoelectronic nanodevices and biological imaging [2][3][4][5][6][7][8][9] . However, the synthesis of acenes longer than hexacene remains formidably challenging because of their low solubility and chemical instability, as explained by Clar's sextet theory 10 . More specifically, this theory explains how increases in the number of non-sextet rings along the acene series endows molecules with an unstable biradical character in their ground state 11 . Much effort has gone into the synthesis of acenes, and their length has gradually increased (one benzene ring at a time) for many decades. Functionalization with solubilizing and stabilizing substituents has allowed for the preparation of acenes of up to nine fused rings using solution synthesis methods 12 . The on-surface reaction has also emerged as a promising method to afford unsubstituted acenes under high-vacuum conditions 13,14 . This methodology has enabled the synthesis of the longest acene yet, one with 12 benzene rings 15 . However, despite these efforts, polymeric acenes consisting of numerous fused benzene rings, namely polyacene, have not been synthesized.
Using regular nanopores for ship-in-a-bottle synthesis has many advantages, including highly specific reactions in the pores and imposing nanoconfinement effects on reaction selectivity and kinetics [16][17][18][19][20][21][22][23][24][25] . Recently, metal-organic frameworks (MOFs), which comprise metal ions and organic ligands, have attracted much attention due to their applications in fields such as gas storage, separation, catalysts and drug delivery [26][27][28][29][30] . One of the characteristic features of MOFs is their structural diversity; their pore size and shape are controllable at the molecular level, providing an ideal compartment for encapsulating a variety of guest species and controlling their assembly structures 31,32 . The resulting guest molecules can be easily recovered by dissolution Article https://doi.org/10.1038/s44160-023-00310-w end-to-end fashion due to the geometrical constraint and host-guest interactions ( Supplementary Fig. 3b,c), which would facilitate the regulated reactions at the 3-and 7-position of the monomers.
Polymerization of the monomers in the nanochannels of 1 was performed as follows. First, the monomers were introduced into the pores of 1 by sublimation to obtain the host-monomer composites (1⊃BBMN and 1⊃BBMA). The monomers were exclusively adsorbed inside the nanochannels, as confirmed by X-ray powder diffraction (XRPD), thermogravimetric (TG) and N 2 adsorption measurements . Polymerization of BBMN and BBMA was then conducted by heating the composites at 250 °C (below the temperature of monomer release from the pores) for 24 h in a sealed glass tube, resulting in the composites of 1 with poly(naphthalene-2,3:6,7-tetrayl-6,7-dimethylene) (PNTD) and poly(anthracene-2,3:6,7-tetrayl-6,7-dimethylene) (PATD), respectively. We confirmed the formation of the composites (1⊃PNTD and 1⊃PATD) using a series of characterization techniques. XRPD measurements showed that the crystal structure of the host MOF was maintained during the heating process ( Supplementary Fig. 4). The morphology and size of the particles of 1 remained unchanged during the polymerization, as confirmed by scanning electron microscopy ( Supplementary Fig. 7). These results suggest that polymerization proceeded inside the channels of 1. Furthermore, a drastic decrease in the N 2 adsorption capacity of the composites compared with that of the pristine 1 was consistent with the accommodation of the precursor polymers in the nanochannels ( Supplementary Fig. 6).
The precursor polymers, PNTD and PATD, were released from the framework of 1 by digesting the host in an aqueous NaOH solution. The complete removal of 1 was confirmed by XRPD ( Supplementary  Fig. 4) and scanning electron microscopy-energy-dispersive X-ray spectroscopy ( Supplementary Fig. 8). We performed structural characterizations of the products using Fourier transform infrared (FTIR), solid-state 13 C nuclear magnetic resonance (NMR) and matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF-MS) measurements. The FTIR spectra of PNTD and PATD display peaks corresponding to the out-of-plane (opla) sp 2 C-H vibration mode, which contrast with the product obtained from thermal processing the bulk monomers under the same conditions ( Supplementary  Figs. 2 and 9), suggesting inhibition of the cross-linking reaction in of the host frameworks, affording well-defined nanomaterials with accurately controlled structures.
Here, we report the synthesis of polyacene by using a methodology, distinct from conventional methods, grounded on organic and/or surface chemistry. The proposed strategy involves two steps: controlled synthesis of precursor polymers within an MOF and subsequent conversion into polyacene (Fig. 1). The spatial constraints of the MOF allowed for the highly regulated polycoupling reaction of aromatic monomers within the one-dimensional (1-D) nanochannels, providing linearly extended polymeric precursors. Subsequently, the dehydro-aromatization reaction of the precursors provided the bulk quantity of polyacene without any peripheries, which was inaccessible using conventional methods.

Results and discussion
Hydroacenes, the partially saturated acenes, can be used as precursors of acenes via dehydro-aromatization 33 . Therefore, we envisioned that polymeric hydroacenes could serve as a precursor for the generation of polyacene. For this purpose, 2,6-bis(bromomethyl)naphthalene (BBMN) and 2,6-bis(bromomethyl)anthracene (BBMA) could be used as monomers if the selective coupling reaction would proceed at the 3-and 7-position. However, targeted precursor polymers were not obtained in the bulk reaction owing to the higher reactivity at zigzag positions than at the 3-and 7-position ( Fig. 1 and Supplementary  Fig. 1). Thermal treatment of the neat BBMN resulted in the formation of branched and graphitic structures ( Supplementary Fig. 2).
To initiate the site-selective polycoupling reaction, an MOF with 1-D nanochannels was used as a host in which the monomers could be aligned along the channel direction. We used [ZrO(L)] n (where L is the dicarboxylate ligand), with 1-D nanochannels along the c-axis, as the host because its pore size can be tuned at the molecular level by changing the dicarboxylate ligand 34 . Additionally, [ZrO(L)] n has high thermal stability because of the strong coordination bond and high coordination number of the zirconium nodes. On the basis of the molecular dimensions of BBMN (4.9 × 8.5 Å 2 ) and BBMA (4.9 × 11.2 Å 2 ), [ZrO(4,4-biphenyldicarboxylate)] n (1; pore size = 6.9 × 6.9 Å 2 ) was used as a host for the selective propagation of linear polymer chains (Supplementary Fig. 3a). Molecular dynamics (MD) simulations revealed that the accommodated monomers were densely assembled in an Article https://doi.org/10.1038/s44160-023-00310-w the MOF channels. Notably, the 13 C NMR spectra represent the characteristic signals of the polymeric hydroacenes ( Fig. 2a). Formation of the polymers resulted in a complete loss of the bromomethyl group; concomitantly, the appearance of resonance ascribable to the carbon of the methylene group was observed. Additionally, a cluster of peaks, assignable to aromatic carbons 35 , was observed around 120-140 ppm.
In the PNTD spectrum, the peak intensity for the aromatic carbons at the zigzag positions was higher than that of the BBMN heated without the MOF, supporting the progress of site-selective linear polymerization within the 1-D channels of 1 ( Supplementary Fig. 10). The peak at 20 ppm was attributed to the methyl terminus of the polymers via the debromination reaction 36 . The absence of other peaks indicates that the precursor polymers did not undergo any undesired side reactions, such as oxidation and dimerization, during their isolation process. The formation of polymeric compounds was corroborated by the MALDI-TOF-MS spectra, which showed a periodic pattern of signals in agreement with the molecular mass of the repeating unit ( Fig. 2b and Supplementary  Fig. 11). Remarkably, the precursor polymers consisted of up to several dozens of linearly fused rings, demonstrating the generation of highly extended polymeric hydroacenes mediated by the MOF nanochannels. The precursor polymers were transformed into polyacene with numerous benzene rings by heating at 300 °C under air atmosphere. The colour of the samples drastically changed from ochre to dark brown during the heating treatment, suggesting the formation of polyacene with an extended conjugated backbone. In contrast, the transformation reaction did not proceed at all under vacuum conditions, demonstrating that oxygen is essential for the dehydro-aromatization reaction ( Supplementary Fig. 12). The obtained polyacene was insoluble in all solvents due to the strong interchain π-π interaction, which may impede microscopic structural analysis; however, the bulk quantity of polyacene meant a wide range of techniques could be used to confirm the conversion of the precursors to polyacene. For instance, solidstate 13 C NMR spectroscopy of polyacene did not exhibit the signals for terminal methyl groups of the precursor polymers, presumably because of thermally induced elimination (Fig. 2a). Most strikingly, the methylene resonance peak completely disappeared. The presence of only a broad peak in the aromatic region confirms the formation of polyacene via quantitative dehydro-aromatization reaction 7,37 . This is also confirmed by the presence of characteristic polyacene peaks in FTIR spectroscopy measurements. The opla aromatic C-H vibration modes are classified as SOLO, DUO, TRIO and QUATRO, in reference to the number of adjacent C-H groups 38 . Only the SOLO and QUATRO mode were detected at 900 and 736 cm −1 , respectively, as was the case with the unsubstituted acene series (Fig. 3a) 39,40 . The disappearance of the DUO-mode band at approximately 800 cm −1 suggests the removal of the terminal methyl groups in the precursor polymers during the heating process 38,39,41 , which is consistent with NMR analysis (Fig. 2a).
We attempted to analyse the polyacene chain length using MALDI-TOF-MS measurements; however, no mass peak corresponding to polyacene was detectable, probably because of its high molecular weight Article https://doi.org/10.1038/s44160-023-00310-w and strong intermolecular interactions which reduced the detection efficiency of polyacene. Therefore, because vibrational spectroscopy is a powerful method for quantitative structural analysis, we analysed polyacene length using IR 42,43 . The peaks corresponding to the opla sp 2 C-H vibration modes (SOLO and QUATRO) of the previously reported acene series were analysed and we found a correlation between those peaks and the number of the benzene rings, demonstrating that the IR analysis is capable of evaluating polyacene length ( Supplementary Fig. 13) 39 . Therefore, several acene molecules with defined structures were synthesized and simulated for IR analysis ( Fig. 3 and Supplementary  Fig. 14), revealing a linear correlation between the relative peak area of SOLO to QUATRO modes and the benzene ring number. According to the line of best fit, the mean numbers (±S.D.) of benzene rings in polyacene from PNTD and PATD were estimated to be 17.8 ± 3.3 and 18.6 ± 3.5, respectively (Fig. 3b). Additionally, the absorption spectra of polyacene corroborated the presence of acenes with numerous fused benzene rings. The S 0 to S 1 transition band of acenes (p band) shifts bathochromically with increasing size of the acene system 44,45 . The polyacene exhibited substantial absorption bands in the near-IR region (Fig. 3c). The bandgaps of polyacene from PNTD and PATD were estimated to be 1.30 and 1.28 eV, respectively, on the basis of the absorbance onsets, which were notably small enough to be comparable with the theoretical limit value (1.23 eV) for acenes with infinite chain length 44 . Also, given that the precursor polymers consisted of up to several dozen rings ( Fig. 2b and Supplementary Fig. 11), these results indicate that, to our knowledge, the polyacene obtained in this study is the longest among the acene series reported so far.
The unprecedented length of the generated polyacene incentivized our efforts to unveil its physicochemical properties. The study on the structural stability of polyacene produced remarkable results. Article https://doi.org/10.1038/s44160-023-00310-w It is well known that longer acenes are more susceptible to oxidation and dimerization reactions because of their inherent singlet biradical character 46,47 . However, the 13 C NMR spectra of polyacene did not show the peaks for sp 3 -bridgehead carbons 37 and carbonyl groups generated by these reactions (Fig. 2a). Therefore, we concluded that such unfavourable reactions did not take place in the bulk polyacene; therefore, we evaluated the biradical character of polyacene using electron spin resonance (ESR) and superconducting quantum interfering device (SQUID). In the ESR spectrum of polyacene, we observed a signal with a g value of 2.003, ascribable to a carbon-centred radical ( Fig. 4a and Supplementary Fig. 15a) 48 . SQUID polyacene data revealed a fitting component with a steep decrease in the magnetic susceptibility upon cooling from 70 to 20 K, in accordance with the Bleaney-Bowers equation ( Supplementary Fig. 15b) 49 . This magnetic behaviour is typical for open-shell singlet biradical molecules 50 ; therefore, these results suggested that polyacene did have a biradical nature, which is in agreement with theoretical calculations [51][52][53] . This is also supported by the NMR polyacene analysis: we detected NMR signals that were broader than those of the precursors which were ascribed to a thermally populated paramagnetic triplet biradical (Fig. 2a) 48 .
To identify the mechanisms underlying the unexpectedly high stability of polyacene, the biradical species were quantitatively analysed by ESR. The spin concentration of polyacene was calculated to be approximately 1 × 10 18 spins g −1 (Fig. 4a). Notably, this value is several orders of magnitude smaller than that predicted from the length of polyacene 54 . A similar reduction in radical nature has been reported for several π-conjugated radical molecules that form π-dimers in the solid state. The intermolecular antiferromagnetic interactions result in these molecules being undetectable via ESR 55 . The insolubility of polyacene in all solvents implied strong interchain interactions; therefore, the aggregation of polyacene was studied using MD simulations, and the π-stacked structure was energetically most stable ( Supplementary  Fig. 16a). The aggregation structure was also revealed by XRPD of the polyacene sample, showing a peak corresponding to π-π stacking (Supplementary Fig. 16b). The observed broad diffraction peak suggested the presence of the stacked structures with non-uniform interchain distance, as shown in the MD structure of polyacene. Therefore, it is probable that interchain antiferromagnetic coupling occurs when in close proximity, giving rise to the decrease in the inherent biradical character of polyacene 56,57 .
We evaluated the composition of polyacene using X-ray photoelectron spectroscopy by initially focusing on the surface of the polyacene particles. The C 1s core level region displays peaks corresponding to the oxidized carbon ( Supplementary Fig. 17). The surface of the polyacene particles was etched using argon plasma to examine the inner composition, which led to a drastic decrease in the oxidized carbon peaks. We observed a small NMR peak, assignable to the oxidized carbons ( Supplementary Fig. 18), when polyacene was left in air for a long period (>30 days). These results suggest that the oxidation reaction took place only near the surface of the polyacene particles owing to their biradical character. The high stability of the bulk polyacene was also confirmed by TG analysis, which showed no weight loss up to 350 °C ( Supplementary Fig. 19a). The FTIR spectra of polyacene before and after heating up to 350 °C displayed no obvious changes, and the opla sp 2 C-H vibration modes were clearly observable, indicating its high thermal stability ( Supplementary Fig. 19b). Again, this behaviour was in sharp contrast to that of acenes in solution, which readily decompose via oxidation and/or dimerization reactions. Therefore, the unexpectedly high stability of polyacene in the solid state can be attributed to the chain aggregation that decreased its biradical character, limited oxygen access to the polymer chains and restricted the chain motion required for undesirable oxidation and dimerization reactions 7,11,37,39,58,59 .

Conclusions
Unsubstituted acenes have thus far been fabricated on surfaces under ultrahigh vacuum conditions because of their unstable zigzag edges 15,60 , presenting a barrier to their scalability. Here, we used an MOF to demonstrate the bulk-scale synthesis of polyacenes with exceptional length. The physicochemical properties of the acene series depending on benzene ring number led to the development of theoretical arguments 45,61,62 . Our findings represent an important step toward not only unveiling the unique topological properties of the acene series 63,64 but also its future applications in various areas, including molecular electronics, optoelectronics and spintronics. For example, polyacene, being a mixture of polymeric chains with different lengths, is capable of absorbing light over a wide wavelength, ranging from visible to near IR (Fig. 3c). Along with its remarkably high stability, this light-absorbing feature would be beneficial for applications in photoenergy conversion systems. Article https://doi.org/10.1038/s44160-023-00310-w

Synthesis of precursor polymers in 1
The general procedure for synthesis was as follows: Degassed 1 was prepared by heating it at 160 °C for 12 h in a vacuum. BBMN (298 mg) and degassed 1 (1,000 mg) were mixed in a round-bottom flask (20 ml) and heated at 150 °C for 1 h under reduced pressure, leading to vapour adsorption of BBMN throughout the internal and external surface of 1.
The externally absorbed monomer was selectively removed by heating at 150 °C for 12 h under vacuum, affording a composite of 1 including BBMN (1⊃BBMN, 1,237 mg). The amount of BBMN adsorbed in the composite was calculated to be 18.7%, as determined by TG measurements. 1⊃BBMN was then heated at 250 °C in a sealed reaction container for 24 h to perform polymerization, resulting in 1 and PNTD nanocomposite (1⊃PNTD, 1,124 mg). PATD was synthesized identically within the nanochannels from BBMA (loading amount of BBMA was 33.5%).

Isolation of precursor polymers from 1
1⊃PNTD (1,680 mg) was stirred in a 1 M aqueous solution of NaOH (120 ml) for 24 h, followed by washing three times with a 5% v/v aqueous solution of HF (20 ml) for complete decomposition of the host framework. The collected PNTD was washed with CHCl 3 and then dried under reduced pressure to obtain PNTD (123 mg, 70% yield based on the loading amount of BBMN in 1⊃BBMN). PATD was isolated from 1 in a similar manner (39% yield).

Conversion from PNTD and PATD to polyacene
The precursor polymers, PNTD (48 mg) and PATD (24 mg), were heated at 300 °C under air for 24 h to obtain polyacene of 38 and 18 mg, respectively.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability
The data supporting the findings of this study are available within the article and its Supplementary Information. Source data are provided with this paper.
Corresponding author(s): Takahi Uemura Last updated by author(s): Mar 17, 2023 Reporting Summary Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist.

Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, A description of all covariates tested A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals) For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted

Data analysis
Provide a description of all commercial, open source and custom code used to analyse the data in this study, specifying the version used OR state that no software was used.
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information.

Data
Policy information about availability of data All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: -Accession codes, unique identifiers, or web links for publicly available datasets -A description of any restrictions on data availability -For clinical datasets or third party data, please ensure that the statement adheres to our policy The data supporting the findings of this study are available within the Article and its Supplementary Information.

Human research participants
Policy information about studies involving human research participants and Sex and Gender in Research.

Ethics oversight
Identify the organization(s) that approved the study protocol.
Note that full information on the approval of the study protocol must also be provided in the manuscript.

Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf

Life sciences study design
All studies must disclose on these points even when the disclosure is negative.

Sample size
Describe how sample size was determined, detailing any statistical methods used to predetermine sample size OR if no sample-size calculation was performed, describe how sample sizes were chosen and provide a rationale for why these sample sizes are sufficient.

Blinding
Describe whether the investigators were blinded to group allocation during data collection and/or analysis. If blinding was not possible, describe why OR explain why blinding was not relevant to your study.

Behavioural & social sciences study design
All studies must disclose on these points even when the disclosure is negative.

Study description
Briefly describe the study type including whether data are quantitative, qualitative, or mixed-methods (e.g. qualitative cross-sectional, quantitative experimental, mixed-methods case study).

Research sample
State the research sample (e.g. Harvard university undergraduates, villagers in rural India) and provide relevant demographic information (e.g. age, sex) and indicate whether the sample is representative. Provide a rationale for the study sample chosen. For studies involving existing datasets, please describe the dataset and source.

Sampling strategy
Describe the sampling procedure (e.g. random, snowball, stratified, convenience). Describe the statistical methods that were used to predetermine sample size OR if no sample-size calculation was performed, describe how sample sizes were chosen and provide a rationale for why these sample sizes are sufficient. For qualitative data, please indicate whether data saturation was considered, and what criteria were used to decide that no further sampling was needed.