HC₃As, the simplest arsadiyne

Abstract

1-Arsabutadiyne (2-propynylidynearsine, HC₃As), is efficiently produced by photolysis of propynylarsine isolated in solid argon. The observed infrared absorption spectra and predicted molecular parameters of HC₃As and HC₃P show significant similarities, but large differences compared to HC₃N.

---

Cyanoacetylene, HC₃N, first synthesized over a century ago,¹ has been extensively characterized, partly because of its importance for molecular astrophysics. The molecule has been detected^2–6 in galactic gas clouds, circumstellar envelopes, comets, the atmosphere of Titan, and also in extragalactic locations. It constitutes the first element in the series of rod-like nitriles H-(CC)_n-CN represented in space up to at least n = 4.^7,8 Spectroscopic data available for HC₃P are sparse, limited to the gas-phase rotational (microwave) domain⁹ and to the vibrational (infrared) spectrum of the cryogenically isolated molecule.¹⁰ The arsenic analogue of cyanoacetylene is expected to be less stable than HC₃N and HC₃P, as it features a monocoordinated arsenic atom with inherently unfavorable π-bonds between the 4p orbital of arsenic and the 2p orbital of carbon. To the best of our knowledge, no theoretical or experimental studies on HC₃As have been reported.

Typically, molecules containing the –C≡As moiety are kinetically unstable,¹¹ therefore laboratory studies of such compounds are rare. Their instability can be overcome by protecting the arsaalkyne centre with bulky substituents and by electron delocalization, as demonstrated for 2-(2,3,6-tri-tert-butylphenyl)-1-arsaethyne.¹² The simplest arsaalkyne, HC≡As, has been detected in the gas phase with photoelectron spectroscopy and mass spectrometry.¹³ Arsaalkynes CH₃CAs¹⁴ and CH₃CH₂CAs,¹⁵ which exhibit half-lives of less than 30 minutes in room-temperature solutions, have been characterized by NMR spectroscopy with additional infrared and microwave spectroscopic measurements made only for CH₃CAs. To the best of our knowledge, no other reports of experimental observation of free arsaalkynes can be found in the literature.

Our previous studies demonstrated that small pnictogen-containing molecules such as HCCPH₂, HCCAsH₂, and HCCSbH₂ undergo efficient photodehydrogenation to yield the corresponding radicals: HCCP, HCCAs, and HCCSb, respectively.^16,17 More recently, we reported on photolysis of CH₃CH₂CP in a solid argon matrix, where HC₃P was produced via an intermediate species, 1-propynylphosphine.¹⁰ In the present work, we adopt a similar strategy using 1-propynylarsine (CH₃CCAsH₂) as a precursor for HC₃As (Scheme 1).

Scheme 1 Photochemical generation of 1-arsabutadiyne.

As propynylarsine is prone to degradation even when stored at −78 °C, it was freshly synthesized prior to each experiment from propynyldichloroarsine following the procedure described by Lassalle et al.¹⁵ and collected in a glass trap at −196 °C.

For matrix isolation studies, a mixture of argon and propynylarsine sublimed from the trap (molar ratio 500 : 1) was deposited onto a caesium iodide window maintained at 10 K within a closed-cycle helium cryostat. The main impurities detected in the cryogenic sample were AsH₃ and traces of 1-arsabutyne (CH₃CH₂C≡As; an isomer of propynylarsine).

A detailed description of this experimental setup and the applied quantum chemical methods is provided in the SI.

IR absorption features produced by photolyzing (Hg lamp, 254 nm) the precursor molecules isolated in cryogenic argon matrices are compared in Fig. 1 with anharmonic frequencies and IR absorption band intensities computed at the coupled cluster level of theory. A set of easily discernible, mutually correlated photoproduct bands developed upon UV irradiation. Based on theory, they are unambiguously attributed to 1-arsabutadiyne (Fig. 1, 2; Table 1). All bands of this simplest arsadiyne having the predicted intensities greater than 1 km mol⁻¹ and falling in our detection range were observed, including the 1st overtone of δ(CCH). However, accurate determination of the ν_CAs frequency in the vicinity of 1436 cm⁻¹ was hampered by the overlapping methyl group vibration bands of the precursor molecule (SI, S8c). After ca. 20 hours of irradiation, the concentration of photoproduced HC₃As reached a plateau and the conversion ratio of propynylarsine to arsabutadiyne could be estimated as 15 to 25 percent, based on theoretically predicted IR band intensities of both compounds (see S4 of SI and Table 1).

Fig. 1 Identification of HC₃As via IR absorption. Bottom: difference spectrum showing the net effect of 20 hours of 254 nm photolysis of 1-propynylarsine isolated in solid argon (precursor bands point downwards, product bands point upwards). Top: spectrum predicted at the VPT2/CCSD(T)/aug-cc-pVTZ level of theory. Diamonds mark the bands of tentatively assigned products: CH₂CHCAs and CH₂CCHAsH₂ (SI S9 and S5). Asterisks indicate artifact features originating in the irradiated CsI substrate window.

Fig. 2 Vibrational frequencies observed in solid Ar for HC₃X compounds (X = As, P, N) and theoretically derived (VPT2/CCSD(T)/aug-cc-pVTZ). Points indicated with arrows refer to the C≡X stretching mode. See Fig. S1 of SI for graphical representations of the fundamental modes. Experimental data come from ref. 18 (HC₃N), ref. 10 (HC₃P), and this work (HC₃As).

Table 1 Measured and theoretically derived vibrational wavenumbers (cm⁻¹) of HC₃As, compared to those reported for HC₃P and HC₃N. Values in parentheses are the theoretical (in km mol⁻¹) and experimental (relative) IR band intensities. X stands for pnictogen. See Fig. S1f or the graphical representation of fundamental vibrations

Mode^a	H–C≡C–C≡As			H–C≡C–C≡P			H–C≡C–C≡N
	Theory (CCSD(T)/aug-cc-pVTZ)		Ar matrix	Theory (CCSD(T)/aug-cc-pVTZ)		Ar matrix^10	Theory (CCSD(T)/aug-cc-pVTZ)		Ar matrix^18
	Harmonic	VPT2		Harmonic	VPT2		Harmonic	VPT2
a S2 (SI) provides the graphical representation of all fundamental modes.b Estimation of intensity and the determination of exact maxima of the band was hampered by an overlapping propynylarsine band.c Gas phase; ref. 21.
ν(CH)	3436	3328	3323.4, 3321.1	3445	3333	3322.7	3443	3332	3314.9, 3315.9
	(83)	(71)	(100)	(79)	(68)		(71)	(64)
ν(CC)	2106	2072	2062.7	2127	2082	2062.8	2121	2089	2076.5
	(15)	(11)	(6)	(18)	(12)		(2)	(1.7)
ν(CX)	1465	1445	1436^b	1553	1531	1524.3	2337	2297	2268.7
	(21)	(23)		(30)	(30)		(14)	(11)
ν(CCX)	567	563	Not detected	693	683	Not detected	888	873	Not detected
	(1)	(1)		(0.5)	(0.1)		(0.2)	(0.0)
δ(CCH)	603	624	630.0, 629.0	616	621	611.2, 611.8, 613.1, 614.0	675	671	665.7, 667.3
	(70)	(73)	(95)	(69)	(71)		(73)	(71)
2δ(CCH)	—	1259	1247.0	—	1241	1212.2	—	1334	1318.3
		(31)	(14)		(27)			(18)
δ(CCCX) zig-zag	500	513	Not detected	506	511	Not detected	548	535	502.1, 503.5
	(2)	(0.2)		(2)	(1)		(9)	(12)
δ(CCCX) banana	197	196	Out of range	209	203	Out of range	237	225	222.4^c
	(20)	(18)		(16)	(15)		(0.4)	(0.4)

---

Data collected in Fig. 2 and Table 1 allow comparison of the measured IR bands of HC₃As with those of its phosphorus-¹⁰ and nitrogen-bearing¹⁸ analogues observed in solid Ar. The measured frequencies of ν_CH and ν_CC stretching vibrations are practically insensitive to the replacement of arsenic with either phosphorus or nitrogen (see Fig. S1 of SI for a graphical representation of the fundamental modes). More pronounced (i.e., on the order of several percent) are the corresponding changes revealed for the δ_CCH mode and its first overtone. There, the frequency does not decrease monotonically from the lightest to the heaviest compound but reaches a minimum for HC₃P. As shown in Table 1 (see also Fig. S2 of SI), this non-trivial effect is reproduced with calculations, provided that the anharmonic approach is applied. The most pronounced influence of pnictogen atom exchange on the IR spectrum is observed in HC₃X molecules (X denoting pnictogen) for the ν_CX stretching band (Fig. 2; see also Fig. S3 of SI). Compared to cyanoacetylene, the respective frequency is reduced by 33% in HC₃P and 37% in HC₃As, in agreement with the theoretical result (33% and 36%, respectively).

Differences in vibrational frequencies observed for HC₃X-family molecules reflect the varying masses of X atoms and electronic charge distributions. Phosphorus and arsenic have similar, low electronegativities, while nitrogen is more electronegative than carbon. Some of the ensuing molecular properties can be traced with natural bond orbital (NBO) analysis. The computed natural atomic charges (Fig. 3) indicate strongly positive values on the phosphorus and arsenic centers, in contrast to the negative charge on nitrogen. This is reflected in CCSD(T)/aug-cc-pVTZ predictions of the equilibrium electric dipole moment: 3.72, 0.75, and 0.43 debyes for HC₃N, HC₃P, and HC₃As, respectively.

Fig. 3 CCSD(T)/cc-pVTZ interatomic distances predicted for three HC₃X molecules, together with the select natural bond orbital analysis results: Wiberg bond indices (italics) and natural atomic charges (red).

The Wiberg bond indices¹⁹ show a progressive decrease in the carbon–pnictogen bond order from HC₃N to HC₃As, consistent with the expected weakening of π-interactions.¹¹

While some traces of other products, apart from HC₃As, were found in the irradiated sample, it is worth noting that the overall photolysis yield, and hence the number of detectable products, is significantly lower here than in a recently reported¹⁰ study on propynylphosphine in solid Ar. We identified allenylarsine, CH₂CCHAsH₂ (based on weak bands at 838.0 cm⁻¹ and 1956.2 cm⁻¹) (SI S5), just as allenylphosphine was the main product (along with HC₃P) of propynylphosphine photolysis. Similarly, vinylarsaethyne, CH₂CHCAs, was observed (tentatively), analogous to CH₂CHCP obtained¹⁰ by single photodehydrogenation of propynylphosphine. Vinylarsaethyne (SI S9) is a plausible intermediate en route to HC₃As.

Traces of HCCAs and methane were also detected upon photolysis. That pair of compounds may arise (by analogy with the photochemistry of propynylphosphine) through decomposition of a transiently formed isomer, most likely HCCAs(H)CH₃. The characterization of HCCAs and methane was based on their CH stretching¹⁷ and bending bands²⁰ at 3284.5 cm⁻¹ and 1304.5 cm⁻¹ (SI S10).

Arsabutyne (CH₃CH₂C≡As), observed as an impurity in the unphotolyzed sample (SI Fig. S6), most likely isomerizes to propynylarsine upon irradiation and further decomposes to HC₃As.

The identification of HC₃As, the major product of propynylarsine photolysis, opens the way to its further chemical and spectroscopic characterization.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data supporting the findings of this study are available within the article and the supplementary information (SI). Additional raw experimental and computational data are available from the corresponding author upon reasonable request.

Supplementary information is available. See DOI: https://doi.org/10.1039/d5dt02772a.

References

1. C. Moureu and J.-C. Bongrand, C. R. Acad. Sci., 1910, 151, 946.
2. B. E. Turner, Astrophys. J., 1971, 163, L35.
3. V. Bujarrabal, M. Guelin, M. Morris and P. Thaddeus, Astron. Astrophys., 1981, 99, 239.
4. D. Bockelée-Morvan, D. C. Lis, J. E. Wink, D. Despois, J. Crovisier, R. Bachiller, D. J. Benford, N. Biver, P. Colom, J. K. Davies, E. Gérard, B. Germain, M. Houde, D. Mehringer, R. Moreno, G. Paubert, T. G. Phillips and H. Rauer, Astron. Astrophys., 2000, 353, 1101.
5. A. Coustenis, B. Bézard, D. Gautier, A. Marten and R. Samuelson, Icarus, 1991, 89, 152.
6. J. E. Lindberg, S. Aalto, F. Costagliola, J.-P. Pérez-Beaupuits, R. Monje and S. Muller, Astron. Astrophys., 2011, 527, A150.
7. R. L. Snell, F. P. Schloerb, J. S. Young, A. Hjalmarson and P. Friberg, Astrophys. J., 1981, 244, 45.
8. N. W. Broten, T. Oka, L. W. Avery, J. M. MacLeod and H. Kroto, Astrophys. J., 1978, 223, L105.
9. L. Bizzocchi, C. Degli Esposti and P. Botschwina, Chem. Phys. Lett., 2000, 319, 411.
10. A.-L. Lawzer, E. Ganesan, T. Custer, J.-C. Guillemin and R. Kołos, Phys. Chem. Chem. Phys., 2025, 27, 7556.
11. L. Weber, Chem. Ber., 1996, 129, 367–379; L. L. Lohr and A. C. Scheiner, J. Mol. Struct.: THEOCHEM, 1984, 18, 195; T. Marino, M. C. Michelini, N. Russo, E. Sicilia and M. Toscano, Theor. Chem. Acc., 2012, 131, 1141.
12. G. Märkl and H. Sejpka, Angew. Chem., Int. Ed. Engl., 1986, 25, 264 (Angew. Chem., 1986, 98, 286).
13. J.-C. Guillemin, A. Chrostowska, A. Dargelos, T. X. M. Nguyen, A. Graciaa and P. Guenot, Chem. Commun., 2008, 4204.
14. J.-C. Guillemin, L. Lassalle, P. Dréan, G. Wlodarczak and J. Demaison, J. Am. Chem. Soc., 1994, 116, 8930.
15. L. Lassalle, S. Legoupy and J.-C. Guillemin, Inorg. Chem., 1995, 34, 5694.
16. A.-L. Lawzer, T. Custer, J.-C. Guillemin and R. Kołos, Angew. Chem., Int. Ed., 2021, 60, 6400.
17. A.-L. Lawzer, E. Ganesan, M. Gronowski, T. Custer, J.-C. Guillemin and R. Kołos, Chem. – Eur. J., 2023, 29, e202300887.
18. F. Borget, T. Chiavassa, A. Allouche, F. Marinelli and J.-P. Aycard, J. Am. Chem. Soc., 2001, 123, 10668.
19. K. B. Wiberg, Tetrahedron, 1968, 24, 1083.
20. M. G. Govender and T. A. Ford, J. Mol. Struct., 2000, 550, 445–454.
21. A. Jolly, Y. Benilan and A. Fayt, J. Mol. Spectrosc., 2007, 242, 46–54.

---