KCN/CuCN-mediated complete cyanation of hexabromotriphenylene

Abstract

Complete cyanation of highly brominated aromatics is often hindered by the challenge of converting multiple carbon–bromine (C–Br) bonds within a single molecular framework. 2,3,6,7,10,11-Hexabromotriphenylene (HBTP) exemplifies this challenge, as its conversion to triphenylene-2,3,6,7,10,11-hexacarbonitrile (HCNTP) requires the replacement of all six C–Br bonds with cyano groups. Herein, we report an efficient cyanation of HBTP using potassium cyanide (KCN) and copper(i) cyanide (CuCN) in dimethyl sulfoxide (DMSO). The combination of KCN and CuCN effectively promoted the complete cyanation of HBTP, whereas either cyanide reagent alone led to limited or negligible formation of HCNTP. Among the examined conditions, a KCN/CuCN molar ratio of 1 : 2 gave HCNTP in a 66.4% isolated yield after recrystallization. This result is consistent with the suppression of DMSO coordination to Cu(i) by free CN⁻ ions supplied by KCN, which maintains the active copper species required for the complete six-fold cyanation of HBTP. This approach offers a practical synthetic route to cyano-functionalized triphenylene derivatives from heavily brominated precursors, providing a promising π-acceptor building block for conductive MOFs and n-type organic materials.

A binary KCN/CuCN cyanide source enables complete six-fold cyanation of hexabromotriphenylene in DMSO by suppressing Cu(i) deactivation, providing access to electron-deficient hexacyanotriphenylene.

1. Introduction

Polyhalogenated aromatic compounds are valuable precursors to highly functionalized polyaromatic molecules because they contain multiple reactive carbon–halogen bonds. Their complete conversion into polynitrile derivatives is particularly attractive for accessing electron-deficient organic materials. The introduction of cyano groups enhances the electron-deficient character of π-conjugated systems and offers versatile sites for further chemical modifications.^1–4 Among such transformations, the complete cyanation of heavily brominated aromatics remains particularly demanding, and here we address this challenge using 2,3,6,7,10,11-hexabromotriphenylene (HBTP) as a representative substrate.

HBTP is a key precursor for the synthesis of highly substituted triphenylene derivatives.^5–7 However, converting HBTP to triphenylene-2,3,6,7,10,11-hexacarbonitrile (HCNTP) is considerably more challenging than the monocyanation of simple aryl halides. This transformation requires replacement of all six C–Br bonds on the rigid triphenylene framework with cyano groups, which substantially increases synthetic difficulty.⁸ Incomplete cyanation can generate partially substituted intermediates that share the same aromatic core, complicating isolation of the fully converted product. Notably, HCNTP has recently been employed as a π-acidic acceptor intercalated within an electrically conductive graphitic MOF, and electron-deficient triphenylenes have been explored as n-type organic semiconductors, highlighting the value of an efficient and selective route to this electron-deficient triphenylene derivative.^9,10 Although transition-metal-catalyzed methods, particularly those employing palladium catalysts, are widely used for the cyanation of aryl halides, their effectiveness in achieving complete substitution on heavily brominated aromatic substrates is not guaranteed, and the residual catalysts, ligands, and metal-containing byproducts introduce additional purification challenges.^11–15

To overcome these limitations, the selection of cyanide source and organic solvent was reexamined for the complete cyanation of HBTP. Here, we report that the combined use of potassium cyanide (KCN) and copper(i) cyanide (CuCN) in dimethyl sulfoxide (DMSO) enables the efficient conversion of HBTP to HCNTP. This study provides a simple and practical route to HCNTP through the appropriate combination of cyanide sources and organic solvent.

2. Experimental section

2.1. Materials

All solvents and reagents for synthesis were used without further purification, unless otherwise specified. Triphenylene and iron powder were purchased from Alfa Aesar. Bromine (Br₂) was purchased from JUNSEI Chemical Inc. KCN and CuCN powder were purchased from SAMCHUN Chemical. Anhydrous DMSO and N,N-dimethylformamide (DMF) were purchased from Daejung Chemical and used as received.

2.2. Characterization

Fourier-transform infrared (FT-IR) spectra were recorded on a Spectrum 100 (PerkinElmer, USA) using KBr disk pellets. ¹H and ¹³C NMR spectra were obtained on a 400 MHz FT-NMR AVANCE III HD spectrometer (Bruker, USA). Solid-state ¹H-NMR measurements were performed on an Agilent VNMRS (600 MHz) spectrometer (Agilent, USA). Polarizing microscopic images were obtained on DM2500 P (LEICA, Germany). The field emission scanning electron microscope (FE-SEM) images were obtained using a Nanonova 230 (FEI, USA). Elemental analysis (EA) was conducted using a Flash 2000 instrument (Thermo Scientific, Netherlands). Electrospray ionization (ESI) mass spectra were recorded on a Xevo G2 QTof instrument (Waters, USA). UV-Vis absorption spectra were recorded on a Cary 5000 spectrophotometer (Agilent, USA). Thermogravimetric analysis (TGA) was carried out on an STA 8000 simultaneous thermal analyzer (PerkinElmer, USA) under a nitrogen atmosphere. The residual copper content was determined by inductively coupled plasma mass spectrometry (ICP-MS) using a NexION 5000 instrument (PerkinElmer, USA).

2.3. Preparation of 2,3,6,7,10,11-hexabromotriphenylene (HBTP)

2,3,6,7,10,11-Hexabromotriphenylene (HBTP) was synthesized from triphenylene according to a previously reported procedure.⁵ Triphenylene (1.00 g, 4.73 mmol) and iron powder (0.095 g, 1.70 mmol) were placed in a Schlenk flask with nitrobenzene (40 mL). After the flask was equipped with a reflux condenser, the mixture was purged with N₂ for 30 min. Bromine (4.36 mL, 85.1 mmol) was slowly added, and the reaction mixture was heated at 200 °C for 2 h. After cooling to room temperature, 1 M aqueous sodium bisulfite solution (100 mL) was added to neutralize residual bromine, and the mixture was stirred for 30 min. The mixture was then washed with water and diethyl ether until a white solid was obtained. The insoluble white solid was collected by filtration and recrystallized from 1,3,5-trichlorobenzene (300 mL) at 200 °C to give white crystals (2.80 g, 91%). M.p. > 300 °C;²⁵ solid-state ¹H-NMR (600 MHz) δ = 7.90 ppm (Fig. S1); DART-HRMS: m/z calcd for C₁₈H₆Br₆, 701.67; found, 701.54 (Fig. S2). Elemental analysis: anal. calcd (%) for C₁₈H₆Br₆: C, 30.81; H, 0.86; Br, 68.33. Found C, 30.84; H, 0.85; Br, 68.31. IR (KBr): ν_max (cm⁻¹) = 1583, 1549, 1452, 1363, 1118, 871, 861.

2.4. Preparation of triphenylene-2,3,6,7,10,11-hexacarbonitrile (HCNTP, entry 5)

HBTP (1.00 g, 1.42 mmol), CuCN (2.03 g, 22.7 mmol), and KCN (0.74 g, 11.4 mmol) were placed in a Schlenk flask containing DMSO (50 mL). After the flask was equipped with a reflux condenser, the mixture was purged with N₂ for 30 min. The reaction mixture was heated to 160 °C and maintained at this temperature for 24 h. After cooling to room temperature, the reaction mixture was poured into distilled water and stirred in an ice bath for 2 h to precipitate the product. The cold suspension was filtered, and the collected solid was washed with an aqueous ethylenediamine solution (ethylenediamine, 3 mL in water, 500 mL) at least three times. The resulting pale-yellow powder was dissolved in acetonitrile (50 mL), and the solution was filtered to remove insoluble impurities. The filtrate was heated to boiling and then stored in a refrigerator for slow recrystallization over 24 h, giving pale needle-like crystals (0.36 g, 66.4%, Fig. S3). All other entries were performed under the same reaction conditions, except for the cyanide source, solvent, and KCN/CuCN molar ratio. ¹H-NMR (acetone-d₆, 400 MHz) δ = 9.86 ppm; ¹³C-NMR (acetone-d₆, 400 MHz) δ = 133.01, 132.78, 116.48, 116.28 ppm; ESI-MS: m/z calcd for C₂₄H₆N₆, 378.07; found, 377.04 [M–H]⁻. Elemental analysis: anal. calcd (%) for C₂₄H₆N₆: C, 76.19; H, 1.60; N, 22.21. Found: C, 75.84; H, 2.09; N, 23.04. IR (KBr pellet): ν_max (cm⁻¹) = 1562, 1500 (aromatic C–C), 2236 (C N), 3060 (C–H in aromatic).

3. Results and discussion

The cyanation of HBTP was first investigated under conventional CuCN-mediated conditions, then optimized with respect to three parameters: cyanide source, solvent, and the equivalent ratio of KCN to CuCN. Scheme 1 compares the conventional and optimized conditions, with detailed results summarized in Table 1. The CuCN/DMF conditions (entry 1) yielded a yellow crystalline solid (Fig. 1a), and TLC analysis under UV irradiation (365 nm, DCM/hexane = 1 : 2) revealed multiple spots, indicating the formation of partially cyanated triphenylene derivatives alongside HCNTP (Fig. 1b). In contrast, the optimized KCN/CuCN/DMSO conditions (entry 5) afforded HCNTP as pale, needle-like crystals after recrystallization (Fig. 1c), with TLC showing a single spot (Fig. 1d).

Scheme 1. Cyanation of HBTP under different reaction conditions. ^a Isolated yield after recrystallization.

Table 1. Optimization of reaction conditions for the synthesis of HCNTP from HBTP.

Entry	Cyanide source	Solvent	KCN (equiv.)	CuCN (equiv.)	Yielda (%)
1	CuCN	DMF	—	24	7.0
2	CuCN	DMSO	—	24	N.D.b
3	KCN	DMSO	24	—	N.D.b
4	KCN/CuCN	DMSO	12	12	16.4
5	KCN/CuCN	DMSO	8	16	66.4
6	KCN/CuCN	DMSO	16	8	N.D.b

^aIsolated yield. The entry 1 product was purified by recrystallization, followed by column chromatography (silica gel, DCM/hexane = 1 : 2) and a final recrystallization; all other entries were purified by recrystallization from acetonitrile.

^bNot detected. No formation of HCNTP was observed by TLC (DCM/hexane = 1 : 2, UV 365 nm). All reactions were carried out at 160 °C for 24 h.

Fig. 1. Photographs of isolated products and TLC images under UV irradiation (365 nm, DCM/hexane = 1 : 2) for entry 1 (CuCN, DMF) (a and b) and entry 5 (KCN/CuCN (1 : 2), DMSO) (c and d). HCNTP isolated yields: 7% (entry 1) and 66.4% (entry 5).

Rosenmund–von Braun cyanation is the most widely employed method for the conversion of aryl halides into aryl nitriles, typically using CuCN in a polar aprotic solvent such as DMF at elevated temperature.¹⁶ When applied to HBTP under these conditions, the reaction yielded mainly partially cyanated triphenylene derivatives, with HCNTP isolated in only 7% yield (entry 1), indicating that the conventional system is insufficient to drive complete six-fold cyanation of the HBTP framework. To identify the origin of this incomplete conversion, both the solvent and the cyanide source were examined. DMSO, which has a higher boiling point and greater thermal stability than DMF, was first tested as the solvent.¹⁷ However, when CuCN was used alone in DMSO, no formation of HCNTP was detected (entry 2). This result is consistent with the known interaction of DMSO with copper cyanide species. DMSO can act as a bridging bidentate ligand, coordinating to two Cu(i) centers through both its oxygen and sulfur atoms to form (CuCN·DMSO)_n frameworks in which the reactive copper centers are locked within an extended coordination network. In addition, Cu(i) in such DMSO-rich environments is susceptible to aerial oxidation, generating Cu(ii) species that further stabilize as [Cu(ii)(DMSO)₆]²⁺ octahedral complexes and are thereby removed from the catalytic cycle.^18,19 Both processes are reported to occur at temperatures of 70 °C or below. The elevated temperature of 160 °C employed here therefore makes these deactivation pathways more dominant. As a result, the reactive Cu(i)–cyanide species required for aryl C–Br activation is progressively converted into DMSO-coordinated or oxidatively deactivated copper complexes, suppressing the cyanation. Replacement of CuCN with KCN as the sole cyanide source likewise failed to produce HCNTP (entry 3). Although KCN is widely employed for the cyanation of alkyl halides through a classical S_N2 pathway, this mechanism cannot operate on aryl C–Br bonds because the sp²-hybridized aromatic carbon is shielded by the π-system from backside nucleophilic attack.²⁰ These results confirm that neither component alone is sufficient: CuCN is required to activate the aryl C–Br bond, while DMSO simultaneously deactivates the active copper species. To address this issue, the combined use of KCN and CuCN in DMSO was investigated. KCN, which is appreciably soluble in DMSO, provides a continuous source of free CN⁻ ions, whereas CuCN has limited solubility owing to its polymeric Cu–CN network structure.²¹ We hypothesized that the free CN⁻ supplied by KCN modulates the copper coordination environment and competes with DMSO for the copper centers, thereby preserving the reactivity of the Cu(i)–cyanide species.^22,23 To probe this hypothesis, a control heating experiment was conducted. As shown in Fig. S4, heating CuCN alone in DMSO at 160 °C for 1 h induced a pronounced dark yellow-green discoloration and its UV-Vis spectrum showed a broad Cu(ii) d–d transition in the 650–900 nm region (Fig. S5), consistent with the formation of DMSO-coordinated or oxidatively deactivated copper species.²⁴ In contrast, no significant color change occurred when KCN and CuCN were heated together at a 1 : 1 molar ratio under identical conditions. This observation provides direct evidence that CN⁻ ions supplied by KCN suppress the deactivation pathway observed for CuCN/DMSO.

Based on this result, the cyanation of HBTP was carried out with mixed KCN/CuCN systems while varying the molar ratio. At a 1 : 1 ratio, HCNTP was obtained in 16.4% yield—more than twice that of CuCN/DMF—confirming that the combined system enhances cyanation efficiency (entry 4). Increasing the CuCN proportion to a 1 : 2 ratio further improved the yield to 66.4% (entry 5), corresponding to a 9.4-fold increase relative to the CuCN/DMF baseline and a four-fold increase relative to the 1 : 1 condition. Conversely, reducing the CuCN proportion to a 2 : 1 ratio gave no detectable HCNTP (entry 6), indicating that excess KCN without sufficient CuCN cannot drive the reaction. These results demonstrate that complete cyanation of HBTP is not governed solely by the concentration of free CN⁻; rather, an optimal balance between the CN⁻ supply from KCN and the activation of aryl C–Br bonds by CuCN is essential. Among the conditions screened, the 1 : 2 KCN/CuCN ratio in DMSO was identified as the optimum for the complete six-fold cyanation of HBTP.

The products from entry 1 and entry 5 were characterized by FT-IR spectroscopy and nuclear magnetic resonance (NMR) spectroscopy (Fig. 2). The FT-IR spectra of both products exhibited a characteristic C N stretching band near 2240 cm⁻¹, confirming the incorporation of cyano groups. In the spectrum of entry 5, the Ar–Br-associated absorption band at approximately 1118 cm⁻¹ was absent, whereas a residual band persisted in the spectrum of entry 1, indicating the presence of unreacted aryl bromide moieties from partially cyanated species (Fig. 2a).^25,26 This distinction was further supported by ¹H NMR spectroscopy. Entry 5 displayed a sharp singlet at 9.86 ppm corresponding to the aromatic protons, consistent with the high molecular symmetry of fully cyanated HCNTP (see Fig. S6 for the full ¹H NMR spectrum). In contrast, entry 1 showed multiple split aromatic signals, indicative of a mixture of partially cyanated triphenylene derivatives (Fig. 2b). The persistence of these byproducts after recrystallization likely reflects their structural similarity to HCNTP and their comparable solubility.

Fig. 2. (a) FT-IR spectra (KBr) of HBTP (brown), entry 1 (pink), and entry 5 (blue) products. (b) Partial ¹H NMR spectra (acetone-d₆, 400 MHz) of entry 1 (pink) and entry 5 (blue) products in the aromatic region.

To further verify the molecular symmetry of HCNTP, ¹³C NMR analysis was performed on entry 5. The spectrum displayed only four resonances at 133.01, 132.78, 116.48, and 116.28 ppm, corresponding to the aromatic C–H, ring-junction, nitrile, and aromatic C–CN carbons, respectively, in agreement with the four chemically distinct carbon environments expected for the idealized D_3h molecular symmetry of HCNTP (Fig. 3a). The ESI mass spectrum of entry 5 showed a peak at m/z 377.04 corresponding to [M–H]⁻, consistent with the calculated molecular mass of HCNTP (Fig. 3b). Together, the NMR and MS data establish the molecular identity of HCNTP as the fully cyanated product.

Fig. 3. (a) ¹³C NMR spectrum (acetone-d₆, 400 MHz) of HCNTP (entry 5) with an expanded inset of the 112–136 ppm region. (b) ESI mass spectrum of HCNTP (entry 5) in negative mode.

The purity and thermal stability of the isolated product were then examined. ICP-MS analysis showed that entry 5 contained only 2.03 mg kg⁻¹ of residual copper, ∼22-fold lower than that of the conventional CuCN/DMF product (entry 1, 45.02 mg kg⁻¹; Fig. S7). Elemental analysis and SEM-EDS further confirmed the purity of the isolated product, with the measured carbon, hydrogen, and nitrogen contents in close agreement with the theoretical values calculated for C₂₄H₆N₆ (Fig. S8 and Table S1). Finally, thermogravimetric analysis (TGA) under N₂ showed that HCNTP has no distinct melting point and remains thermally stable up to an onset of 522 °C, with a substantial residue remaining at 900 °C (Fig. S9). Collectively, these data confirm that the optimized KCN/CuCN ratio of 1 : 2 in DMSO effectively achieves complete substitution of all six-aryl C–Br bonds in HBTP with cyano groups.

4. Conclusions

In summary, a method for the complete cyanation of HBTP to HCNTP was developed by combining KCN and CuCN in DMSO. The conventional CuCN/DMF conditions gave HCNTP in only 7% yield, and neither KCN/DMSO nor CuCN/DMSO conditions produced detectable amounts of the product, indicating that a single cyanide source is insufficient for the six-fold substitution of the HBTP framework. The poor reactivity in CuCN/DMSO was attributed to the deactivation of Cu(i) species through DMSO coordination, which converts the active copper intermediates into unreactive forms. The addition of KCN to the CuCN/DMSO conditions suppressed this deactivation pathway by providing free CN⁻ ions that compete with DMSO for the copper centers, thereby preserving the reactivity of the Cu(i)–cyanide species. The optimized KCN/CuCN (1 : 2) ratio in DMSO gave HCNTP in 66.4% isolated yield after recrystallization, corresponding to a 9.4-fold improvement over the conventional CuCN/DMF conditions. These results demonstrate that the cooperative use of two cyanide sources is essential for the complete cyanation of highly brominated triphenylene frameworks. The straightforward access to HCNTP enabled by this method is expected to facilitate its application as an electron-deficient building block for organic semiconductors and porous organic frameworks.

Author contributions

Boo-Jae Jang: conceptualization, methodology, investigation, formal analysis, visualization, and writing – original draft. Jin-Seok Lee: investigation, validation, and writing – review and editing. Jeong-Gu Yeo: supervision, project administration, funding acquisition, and writing – review and editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting the findings of this study are included in the article and its supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d6ra06600c.