Synlett 2024; 35(04): 437-440
DOI: 10.1055/a-2112-9605
cluster
11th Singapore International Chemistry Conference (SICC-11)

Synthesis and Coordination Behavior of 9,10-Diarsatriptycene

Ryoto Inaba
a   Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
,
Takashi Kikuchi
b   Rigaku Corporation, 3-9-12 Matsubara-cho, Akishima-shi, Tokyo 196-8666, Japan
,
Hiroaki Imoto
a   Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
c   FOREST, JST, Honcho 4-1-8, Kawaguchi, Saitama 332-0012, Japan
,
Kensuke Naka
a   Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
d   Materials Innovation Lab, Kyoto Institute of Technology, Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
› Author Affiliations

This work was supported by the Japan Society for the Promotion of Science KAKENHI [20H02812 (Grant-in-Aid for Scientific Research (B) to HI) and 22H02131 (Grant-in-Aid for Scientific Research (B) to KN)].
 


Abstract

Herein, 9,10-diheterotriptycenes (DHTs) containing heavy pnictogens (Pn, Pn = As, Sb, and Bi) are synthesized without using the dangerous chemicals used in conventional synthetic methods: tert-butyllithium, organomercury reagents, or trichloroarsine. In particular, 9,10-diarsatriptycene is obtained in relatively high yield and is stable under oxidation and coordination reactions. Additionally, the gold chloride complex 9,10-diarsatriptycene forms a one-dimensional supramolecular polymer constructed through coordination and aurophilic interactions.


#

Triptycene (9,10-dihydro-9,10-[1,2]benzenoanthracene) is a rigid, propeller-shaped molecule with D 3h symmetry (Figure [1a]).[1] It was first synthesized in 1942 by Bartlett via the Diels–Alder reaction of anthracene and benzyne.[2] Since then, various triptycene derivatives have been developed and it has been applied in numerous fields owing to its unique structure, which comprises, host molecules,[3] molecular gears,[4] and highly ordered organic materials.[5] Triptycenes with heteroatoms at the bridgehead positions, called heterotriptycenes,[6] have attracted considerable attention, and various elements, such as phosphorus[7] and boron,[8] have been incorporated. Heterotriptycenes are used as ligands,[9] Lewis acids, and bases.[8] [10]

Zoom Image
Figure 1 Chemical structures of (a) triptycene, (b) 9-phospha-10-heterotriptycenes, and (c) heavy Pn-containing DHTs (this work).

Diheterotriptycenes (DHT), in which two bridgehead carbon atoms are replaced by heteroatoms (Figure [1b]), are an important class of heterotriptycenes. Phosphorus-containing DHTs are one of the most extensively studied because the tightly bound phosphor atoms exhibit unique reactivities and coordination abilities. For example, the Lewis acidity of 9-phospha-10-boratriptycene is enhanced by a reaction around the phosphorus atom.[10e] 9-Phospha-10-silatriptycene and 9-phosphatriptycene-10-phenylborate anions have been synthesized and their transition-metal complexes have been reported.[9d] A series of 9-phospha-10-heterotriptycenes was used as reagents for the Wittig reaction, and the effects of the bridging heteroatom on the isomerization of the intermediates were investigated.[11]

9,10-Diarsatriptycence (1), 9,10-distibatriptycene (2), and 9,10-dibismatriptycene (3), being heavier pnictogen (Pn)-containing DHT analogues, were already synthesized in the 1980s.[12] [13] In particular, trivalent arsenic atoms have higher oxidative resistance than phosphorus; thus, 1 is a promising building block for constructing coordination architectures, such as one-dimensional polymers and metal-organic frameworks. However, their functionality has rarely been studied, and only the bromination of Pn atoms has been examined. A serious drawback of conventional synthetic procedures is the use of dangerous reagents and/or precursors (Scheme [1]). Highly toxic organomercury compounds[12] and the explosive pyrophoric tert-butyllithium (t-BuLi)[13] are used as nucleophiles and/or bases. Additionally, in the case of 1, trichloroarsine (AsCl3), which is necessary as a precursor, is highly volatile (vapor pressure, 10 mmHg at 23.5 °C) and toxic.[14] Owing to concerns regarding the dangers of these chemicals, experimental study on 13 have been avoided. Recently, we developed synthetic methods for organoarsenic compounds using nonvolatile arsenic precursors.[15] Tribromoarsine (AsBr3), which can be readily prepared from arsenic trioxide (As2O3) and hydrobromic acid (HBr), was proposed as an alternative electrophile to AsCl3 because of its relatively low volatility (vapor pressure, 1 mmHg at 41.8 °C).[16] Organoarsenic compounds were synthesized from AsBr3. These results motivated us to develop a synthetic route for 13 without using dangerous precursors or reagents that are conventionally used.

Zoom Image
Scheme 1 Conventional synthetic routes of DHTs using (a) organomercury reagent and (b) t-BuLi

In this study, we synthesized 13 using magnesium (Mg) and 1,2-dibromobenzene instead of an organomeric reagent or t-BuLi. Herein, AsBr3 was used as the precursor of 1 instead of AsCl3. The structures of 13 were analyzed using density functional theory (DFT) calculations.[17] In addition, the oxidation and coordination of 1 were examined.

First, to examine the synthetic routes of heavy Pn-containing DHTs, antimony analogue 2 was selected. Trichlorostibine (SbCl3), which is a starting material, is commercially available, and tris(2-bromophenyl)stibine can be easily prepared from SbCl3. Tris(2-bromophenyl)stibine without t-BuLi was also used; however, 2 was not obtained (Table S1). This is probably because trimetalation is difficult to attain using n-butyllithium, sec-buthyllithium, or a turbo-Grignard reagent. Therefore, we decided to react SbCl3, 1,2-dibromobenzene, and Mg all at once in tetrahydrofuran (THF) to generate nucleophilic species in situ (Scheme [2]). A solution of SbCl3 and 1,2-dibromobenzene was then added to Mg dispersion. Evidently, the reaction solvent and temperature produced compound 2; however, the isolated yield was low (1%). The present synthetic protocol was applied to arsenic (1) and bismuth (3) derivatives in isolated yields of 6% (1) and 0.6% (3), respectively.[18] This synthetic route has the advantage of avoiding dangerous chemicals, which were used in conventional synthetic routes of 9,10-diarsatriptycene, 9,10-distibatriptycene, and 9,10-dibismatriptycene derivatives, though the isolated yields of 2 and 3 were particularly low.

Zoom Image
Scheme 2 Synthesis of 13

The structures of 13, 9,10-diphosphatriptycene (DPT), and triphenylpnictines (PnPh3, Pn = P, As, Sb, Bi) were analyzed using DFT calculations to understand the structural effects of the bridgehead element. Optimization of the full geometries and natural bond orbital (NBO) analysis were conducted using B3LYP/def2tzvp (Table [1]). Evidently, negligible differences were noted between the DHTs and PnPh3 in the Pn–C bonds. However, the distortion of the C–Pn–C angles was highly affected by the Pn atom, judging from the difference between the DHTs and PnPh3; essentially, the heavier the Pn atom, the smaller the distortion. According to the distortion around the Pn atom, the s-character of the lone pair (LP) increased with the distortion around the Pn atom. Presumably, the tightly bound structure of DHT narrows the C–Pn–C angles, thereby leading to an increase in the s-character. The heavier Pn atom had a smaller C–Pn–C angle and higher s-character of the LP in nonbridged PnPh3. Eventually, heavier atoms fit better into the DHT framework. However, from the viewpoint of synthesis, this result is interesting considering that the isolated yields were in the order of 1 > 2 > 3, as shown in Scheme [2]. The antimony and bismuth species possibly decomposed during the reaction, even though the strains in 2 and 3 were lower than that in 1.

Table 1 Average Bond Length, Angle, WBI, and s-Character of Pn Lone Pair (LP)a

Pn–C (Å)b

C–Pn–C (°)c

s-Character of LP (Pn) (%)d

DPT

1.86

 97.0

50.2

1

1.98

 95.6

61.4

2

2.18

 94.0

67.9

3

2.29

 93.2

77.9

PPh3

1.84

102.9

46.5

AsPh3

1.98

100.2

59.6

SbPh3

2.18

 97.4

67.5

BiPh3

2.28

 96.0

78.1

a Optimization of full geometries and NBO analysis were based on B3LYP/def2tzvp.

b Average bond length.

c Average bond angles.

d Average s-character of LP of the Pn atoms.

Next, the oxidation of 1 and 2 was performed; CH2Cl2 of 1 or 2 was stirred in the presence of hydrogen peroxide (H2O2) at 25 °C overnight. The oxide of 1 (1-ox) was obtained in 30% yield, whereas in the case of 2 insoluble white precipitates were generated, and oxide (2-ox) was not obtained. Antimony (III) compounds tend to form oligomeric or clustered structures in the presence of oxidants.[19] This result supports the idea that arsenic compound 1 is more stable than its heavier analogue 2. The coordination abilities of 1 and 2 were also investigated (Scheme [3]). A CH2Cl2 solution of 1 or 2 and AuCl was stirred at 25 °C for 1 h. Recrystallization from a mixture of CH2Cl2 and hexane afforded the Au complex (1-AuCl) in 74% yield. For 2, complexation did not proceed, and white precipitates, which could not be analyzed by 1H NMR spectroscopy, were obtained instead of the AuCl complex (2-AuCl). We previously reported that the Sb–C bond of 9-phenyl-9-stibafluorene, a bridged SbPh3, is cleaved by reaction with AuCl, whereas complexation of the arsenic analogue readily proceeds under the same conditions. Thus, in the present case, Sb–C bond cleavage occurred to yield decomposition products that could not be analyzed.

Zoom Image
Scheme 3 (a) Oxidation and (b) coordination of 1 and 2

Recrystallization was performed by slow mixing of dichloromethane (CH2Cl2) solutions of 1-3, 1-ox, and 1-AuCl with methanol (MeOH). Single crystals suitable for X-ray diffraction (XRD) analysis were successfully grown on 1,[20] 1-ox, and 1-AuCl (Figure [2]). By contrast, 2 and 3 formed tiny needle-like crystals and sufficient reflection was not observed. Therefore, the structures of 1, 1-ox, and 1-AuCl were compared. The C–As–C angles of 1 (95.0(1)–96.2(1)°) were larger than those of 1-ox (100.0(2)–101.9(2)°), and 1-AuCl (97.4(2)–100.7(2)°). The repulsion between the LP and electrons of the As–C bonds was relieved by the removal of LP through oxidation or coordination, thus resulting in relatively large C–As–C angles in 1-ox and 1-AuCl. Accompanied by oxidation or coordination, the As···As distances decreased to 3.4102(7) Å (1), 3.1504(6) Å (1-ox), 3.2383(8), and 3.2444(8) Å (1-AuCl).

Zoom Image
Figure 2 ORTEPs of (a) 1, (b) 1-ox, and (b) 1-AuCl. Thermal ellipsoids are drawn at the 50% probability level. For 1-AuCl, one of the three conformers is shown, and solvent molecules (CH2Cl2) are omitted for clarity.

Notably, the Au···Au distances of 1-AuCl were sufficiently short (3.0207(6) and 3.3326(6) Å) for aurophilic interaction, which is one of noncovalent metal–metal interactions.[21] The As–Au–Cl angles were slightly bent (168.91(5)–175.71(5)°) to efficiently form aurophilic interactions. Molecules of 1-AuCl were linearly arranged in the crystalline state through intermolecular aurophilic interactions (Figure [3a]). The propeller shape of 1 offered one-dimensional (1D) channel along the a-axis, which was filled with CH2Cl2 (Figure [3b]). Although we failed to remove the solvent molecules under vacuum, we demonstrated that 1, an arsenic-containing DHT, is a promising candidate for the construction of transition-metal complexes with a host space. Since aurophilic interaction often offers luminescence,[21] the photoluminescence spectra of 1-AuCl were measured at 298 K and 77 K, however, no emission was observed in the present case.

Zoom Image
Figure 3 1D supramolecular polymer structure of 1-AuCl viewed along (a) c-axis and (b) a-axis (CH2Cl2 molecules are omitted for clarity). Thermal ellipsoids are drawn at the 50% probability level.

In the present study, heavy Pn-containing DHTs 13 were synthesized without dangerous chemicals, i.e., as trichloroarsine, tert-butyllithium, or organomercury reagents, even though the isolated yields of 2 and 3 were particularly low. Oxidation by H2O2 and coordination with AuCl proceeded readily in the case of 1, whereas the decomposition of 2 occurred under the same conditions. Intermolecular aurophilic interactions produce 1D supramolecular polymer structures. The propeller shape of 1 effectively provides a 1D channel along the 1D polymer chain. We are currently investigating transition-metal complexes of 1 as host materials.


#

Conflict of Interest

The authors declare no conflict of interest.

Supporting Information


Corresponding Authors

Hiroaki Imoto
Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology
Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585
Japan   

Kensuke Naka
Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology
Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585
Japan   

Publication History

Received: 22 May 2023

Accepted after revision: 19 June 2023

Accepted Manuscript online:
19 June 2023

Article published online:
10 August 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


Zoom Image
Figure 1 Chemical structures of (a) triptycene, (b) 9-phospha-10-heterotriptycenes, and (c) heavy Pn-containing DHTs (this work).
Zoom Image
Scheme 1 Conventional synthetic routes of DHTs using (a) organomercury reagent and (b) t-BuLi
Zoom Image
Scheme 2 Synthesis of 13
Zoom Image
Scheme 3 (a) Oxidation and (b) coordination of 1 and 2
Zoom Image
Figure 2 ORTEPs of (a) 1, (b) 1-ox, and (b) 1-AuCl. Thermal ellipsoids are drawn at the 50% probability level. For 1-AuCl, one of the three conformers is shown, and solvent molecules (CH2Cl2) are omitted for clarity.
Zoom Image
Figure 3 1D supramolecular polymer structure of 1-AuCl viewed along (a) c-axis and (b) a-axis (CH2Cl2 molecules are omitted for clarity). Thermal ellipsoids are drawn at the 50% probability level.