Synthesis,isomerization,and DFT studies of five-membered heterocyclic ketoximes

Abstract

A homologous series of five-membered heterocyclic ketoximes is synthesized by condensation reactions of 2-acetylpyrrole, 2-acetylthiophene, and 2-acetylfuran with hydroxylamine hydrochloride. These compounds were fully characterized by X-ray single-crystal diffraction, elemental analysis, ¹H NMR, IR, UV-Vis, and fluorescence spectra. Interesting isomerization phenomena are observed for these heterocyclic ketoximes by different characterization methods. By means of ¹H NMR and X-ray single-crystal diffraction, it was found that 2-acetylpyrrole oxime (1) exists as the Z-type geometric isomer, 2-acetylfuran oxime (3) exists as the E-isomer, while 2-acetylthiophene oxime (2) is identified as a mixture of Z- and E-isomers. To study the above phenomena further, quantum chemical calculations of total energy, dipole moment, E_HOMO, E_LUMO, and other parameters of each heterocyclic ketoxime were carried out by Density functional theory (DFT). The theoretical calculations were in good accordance with the experimental spectroscopic results.

Graphical abstract

Introduction

In recent years, due to the overlap of inorganic and organic chemistry, coupled with comprehensive applications of many synthetic methods, polymetallic complexes have become important in the field of coordination chemistry. Polynuclear complexes exhibit different properties from mononuclear complexes due to M–M valence electron interactions and M–L (end group or bridging ligand) interactions. Their novel structures including cages, chains, and crown ethers, and their applications in the fields of catalysis, optics, electricity, magnetism, superconductivity, and information storage, have attracted increasing attention from researchers in the field of coordination chemistry.^1–3

There are many methods to construct polynuclear complexes and these require the design and synthesis of novel organic ligands. The strategies involved in the development of novel coordination complexes can be divided into two categories: one is to design ligands containing multiple heteroatoms (such as N, O, and S atoms); the other is to introduce second bridging ligands. Ligands with multiple heteroatoms can easily react with metals to form polynuclear complexes with claw-shaped chelated structures. Such ligands have a range of electronic effects, but it is also important to limit steric effects because these can hinder the development of polycomplexation. Reducing the size of a ligand or coordinating group as much as possible helps to minimize its steric hindrance and is important in the design of novel ligands. In particular, ketoximes are of significant importance because of their small coordinating groups.

During the process of constructing monomolecular magnets, pyridine oximes have been frequently used because they can establish polynuclear systems through self-assembly with flexible and diversified structures. A large number of polynuclear transition-metal complexes with novel structures and unique properties based on these ligands have been reported.^4–6 Typical pyridine oxime ligands can be obtained by changing the substituents on the pyridine ring (e.g. the 6-position is H or Me), replacing the substituents on the side arm (e.g. H, Me, Ph, Py, NH₂, and CN), changing a symmetric structure to an unsymmetric one ((E)-picolinaldehyde oxime vs (1E,1’E)-pyridine-2,6-dicarbaldehyde dioxime), and replacing mono-oximes by di-oximes or poly-oximes ((E)-1-(pyridin-2-yl)ethan-1-one oxime vs (1E,1’E)-1,1’-(pyridine-2,6-diyl)bis(ethan-1-one) dioxime). Albert et al. ⁷ carried out self-assembly using 6-methylpyridine-2-formaldoxime with Ni(O₂CPh)₂ and Ni(acac)₂ to obtain hexanuclear, antiferromagnetic complexes [Ni₆(O₂CPh)₆(6-mepao)₆] and [Ni₆(O₂CMe)₆(6-mepao)₆]. In addition, the use of an azide group as the second bridging ligand led to the pentanuclear ferromagnetic complex [Ni₅(3-Cl-BzO)₄-(6-mepao)₄ (6-mepaoH)₂(N₃)₂] and the tetranuclear cluster [Mn^II₂Mn^III₂] ⁸ by self-assembly coordination of Ni(Cl-BzO)₂ and manganese perchlorate with 6-mepaoH, respectively. Stamatatos et al. ⁹ synthesized trinuclear Mn^III complexes by reacting manganese acetate and manganese perchlorate with mpkoH, in which the distorted triangular core formed by the mpko ligands results in ferromagnetic coupling and single molecular magnet (SMMS) properties. In addition, Theocharis and George ¹⁰ synthesized Ni₁₂ and Ni₁₄ polynuclear systems from 6-methylpyridine-2-ketone oxime, respectively. Jordi synthesized a rarely reported polynuclear Ni₁₃ ferromagnet which contained two Ni₆ rings and was bridged by one Ni^II. Jordi et al. 12 also used reactions of common nickel salts such as nickel benzoate and nickel chloride with pyridine-2-cyanooxime, and successfully obtained Ni₅ polynuclear complexes by liquid-phase diffusion by adding equimolar triethylamine. Albert et al. ¹³ utilized copper nitrate, copper benzoate, and copper acetate in reactions with the same ligand to give Cu₃, Cu₄, and their coordination polymers. Ji et al. have synthesized structural similar Ni₄ complexes by self-assembly of nickel perchlorate with pyridine-2-aminoxime and pyrazine-2-aminoxime,^14,15 respectively.

For oximes based on heterocyclic ligands, the multiple coordination sites and strong coordination abilities make for stable five- or six-membered chelate rings after reacting with metals. In order to clarify the influence of different heterocycles on the structures and properties of ketoximes, a series of ketoximes containing the five-membered heterocycles pyrrole, furan, and thiophene (Scheme 1) were synthesized to provide analogues of the six-membered heterocyclic pyridinyl oximes. By spectral characterization, as well as DFT calculations, the most stable geometric isomers of these ketoximes have been determined and analyzed. This study of possible isomers is expected to lay an essential foundation for the construction of polynuclear metal complexes with novel structures and properties.

Scheme 1.

Synthesis of a series of five-membered heterocyclic ketoximes 1–3.

Results and discussion

In this paper, a series of five-membered heterocyclic ketoximes 1–3 are synthesized by condensation reactions of 2-acetylpyrrole, 2-acetylthiophene, and 2-acetylfuran with hydroxylamine hydrochloride, respectively. Both cis- and trans-isomers of compounds 1–3 are possible because these five-membered heterocyclic ketoximes contain a C=N bond. In order to establish the preferred isomeric structure of each compound and clarify the similarities and differences among 1–3, all compounds are fully characterized by X-ray single-crystal diffraction, elemental analysis, ¹H NMR, IR, UV-Vis, and fluorescence spectra.

In the IR spectrum of 1 (see Figure S4 in the Supporting Information), characteristic absorption bands of v_C=N 1631 cm⁻¹, v_N–O 955 cm⁻¹, v_–OH 3418 cm⁻¹, and v_pyrrole 1462, 1419, 1350, 1329 cm⁻¹ are observed for (Z)-1. The stretching vibration of OH moieties in the solid sample appears as a wide band at 3165 cm⁻¹, which can be assigned to intermolecular H-bonds. Similar absorption bands for the above characteristic functional groups are also observed in the IR spectra of (E)-2 and (E)-3 (see Figures S5 and S6 in the Supporting Information).

The UV-Vis spectra of 1, 2, and 3 (Figures 1–3) are measured in four solvents of different polarities (CHCl₃, THF, CH₂Cl₂, and CH₃CN), and the absorptions are summarized in Table 1. The UV-Vis spectra show that the absorption spectra of 1 and 3 are unimodal while 2 is bimodal, which indicates that each of 1 and 3 is a single pure isomer while 2 is a mixture of cis- and trans-isomers in the four solvents studied. This observation is consistent with that of ¹H NMR. It is generally believed that the trans-isomer has more extensive electron delocalization and stronger conjugation than the cis-isomer, where the π → π* transition is more likely to occur at a longer wavelength. Thus, each peak of 2 can be attributed to a specific geometric isomer as follows: in CHCl₃ (E)-2 absorbs at 280 nm and (Z)-2 at 265 nm; in CH₃CN (E)-2 absorbs at 279 nm and (Z)-2 at 264 nm; in THF (E)-2 absorbs at 280 nm and (Z)-2 at 265 nm; in CH₂Cl₂ (E)-2 absorbs at 280 nm and (Z)-2 at 264 nm. The polarity of the solvent also has a significant influence on the UV-Vis spectra of these compounds. It can be seen from Table 1 that λ_max of 3 is in the order: λ_max (CHCl₃) > λ_max (THF) > λ_max (CH₂Cl₂) > λ_max (CH₃CN), being opposite to the order of the polarity of the solvents CHCl₃ < THF < CH₂Cl₂ < CH₃CN. This shows that the maximum absorption wavelength of 3 is blue-shifted on increasing the solvent polarity. ¹⁶ The λ_max of 1 follows basically the same order of λ_max (CHCl₃) > λ_max (THF) > λ_max (CH₃CN) > λ_max (CH₂Cl₂) with the same rule of absorption intensity CHCl₃ > THF > CH₃CN > CH₂Cl₂ as that of 3. The λ_max of 2 changes little in the four solvents (CHCl₃ 2: 280 nm/265 nm, THF: 280 nm/265 nm; CH₂Cl₂: 280 nm/264 nm, and CH₃CN: 279 nm/264 nm), which indicates that the solvation effect of 2 is not obvious when compared with 1 and 3.

Figure 1.

UV-Vis spectra of 1 in four solvents.

Figure 2.

UV-Vis spectra of 2 in four solvents.

Figure 3.

UV-Vis spectra of 3 in four solvents.

Table 1.

UV-Vis absorption properties of 1–3 in four solvents.

Compound	Solvent	λabs, max/nm	Intensity/a.u.
1	CHCl3	272	3.165
	CH3CN	261	3.545
	THF	262	3.043
	CH2Cl2	251	4.067
2	CHCl3	280/265	3.184/3.167
	CH3CN	279/264	1.632/1.486
	THF	280/265	2.594/2.438
	CH2Cl2	280/264	2.128/2.103
3	CHCl3	260	3.641
	CH3CN	235	3.967
	THF	259	3.097
	CH2Cl2	242	4.050

The fluorescence emission spectra of 1–3 in MeCN, produced by π* → π and π* → n electron transitions, are shown in Figure 4. Products 1–3 each show two regular peaks, 1: λ_em1 = 349 nm, λ_em2 = 603 nm; 2: λ_em1 = 365.8 nm, λ_em2 = 604 nm; 3: λ_em1 = 365 nm, λ_em2 = 603 nm. A large p–π-conjugated delocalization system is formed involving the five-membered heterocyclic rings, C=N, and lone pair electrons of the –OH oxygen atom, which is beneficial to the generation of strong π* → π fluorescence efficiency. The shorter wave λ_em with less emission intensity in each fluorescence spectrum is classified as the π* → n transition, while the longer wave λ_em with a higher emission intensity is classified as the π* → π transition. In addition, the five atoms that make up the pyrrole, thiophene, and furan rings in this series of compounds are sp² hybridized and coplanar, leading to rigid structures that facilitate strong fluorescence. This is because the rigid structure not only increases the conjugation of the π electrons, but also reduces the energy loss of non-radiative transitions such as internal conversion, intersystem crossing processes, and internal vibrations of molecules, which also reduces the quenching effect of the solvent on fluorescence. By comparing the fluorescence intensity of the three compounds, it is found that 3 is the highest, 1 is in the middle, and 2 is the weakest. The electron delocalization energies of pyrrole, furan, and thiophene rings are 88, 67, and 117 kJ/mol, respectively, which is in good agreement with the experimental results of the fluorescence intensity order: 3 > 1 > 2.

Figure 4.

Fluorescence emission spectrum of 1–3 in MeCN.

By combination of ¹H NMR spectroscopes and crystal structures, it could be found that 1–3 have distinctly different Z/E configurations (see Figures 5–7 and S1–S3). The crystal structures show that 1 has a cis-structure ((Z)-1), that is, the orientation of the oxime –OH is toward the side of the pyrrole ring and close to the N atom, while 2 and 3 both have trans-structures ((E)-2, (E)-3), that is, the oxime hydroxyl orientation is away from the thiophene and furan rings, respectively. By means of ¹H NMR in CDCl₃ solution, 1 and 3 show a full set of proton signals that correspond to pure Z- and E-isomers, respectively, while 2 displays doubling of the signals in its spectrum. These two sets of proton signals are identified as the mixture of Z- and E-isomers of 2, by integration, the contents of (E)-2 and (Z)-2 are established as 55% and 45%, respectively. ¹⁶ Regarding compound 2, the solution nuclear magnetic spectroscopy indicates there are two isomers at the same time, while X-ray single-crystal diffraction proves only E-type structure, this maybe because E-type isomer has a relative smaller solubility and a larger mass proportion (55%), both of factors help it crystallize from the solution first.

Figure 5.

ORTEP drawing of (Z)-1.

Figure 6.

ORTEP drawing of (E)-2.

Figure 7.

Crystal structure of (E)-3.

The key bond lengths and angles of 1–3 are listed in Table 2. The bond length of C(4)-C(5) in 1 is 1.443(10) Ǻ, and the corresponding bond lengths in 2 and 3 are 1.454(4) and 1.440(3) Ǻ, respectively. All these bonds are shorter than the average C–C single bond length 1.54 Ǻ. There are also some differences among the C=N bonds in 1–3: N(2)=C(5) in 1 is 1.287(9) Ǻ, and the corresponding bond lengths of N(1)=C(5) in 2 and 3 are 1.274(4) and 1.288(2) Ǻ, respectively. All of these bonds are shorter than the standard C=N double bond of 1.32 Ǻ. The above phenomena may be due to the large conjugated π system formed by the C=N with the pyrrole, thiophene, and furan rings, respectively. Among the three compounds, the torsion angles of N(1)–C(4)–C(5)–N(2) in 1, S(1)–C(4)–C(5)–N(1) in 2, and O(1)–C(4)–C(5)–N(1) in 3 are 0.6°, −5.1(4)º, and −9.8(3)º, respectively, which shows that the C=N bonds and the five-membered heterocycles (pyrrole, thiophene, and furan) are almost coplanar.

Table 2.

Selected bond lengths (Ǻ) and bond angles (°) for compounds 1–3.

1		2		3
N(1)–C(1)	1.331(9)	S(1)–C(1)	1.697(4)	N(1)–C(5)	1.288(2)
N(1)–C(4)	1.376(9)	S(1)–C(4)	1.710(3)	N(1)–O(2)	1.404(2)
N(2)–C(5)	1.287(9)	N(1)–C(5)	1.274(4)	O(1)–C(1)	1.365(3)
N(2)–O(1)	1.404(7)	N(1)–O(1)	1.399(3)	O(1)–C(4)	1.369(2)
O(1)–H(1A)	0.820	O(1)–H(1)	0.8200	O(2)–H(2)	0.8200
C(4)–C(5)	1.443(10)	C(4)–C(5)	1.454(4)	C(4)–C(5)	1.440(3)
C(1)–C(2)	1.326(12)	C(1)–C(2)	1.335(5)	C(5)–C(6)	1.495(2)
C(1)–N(1)–C(4)	109.8(6)	C(1)–S(1)–C(4)	92.03(18)	C(5)–N(1)–O(2)	113.06(15)
C(5)–N(2)–O(1)	113.5(6)	C(5)–N(1)–O(1)	112.2(3)	C(1)–O(1)–C(4)	106.32(16)
N(2)–O(1)–H(1A)	109.5	N(1)–O(1)–H(1)	109.5000	N(1)–O(2)–H(2)	109.5000
N(2)–C(5)–C(6)	114.7(6)	S(1)–C(1)–H(1A)	124.2000	C(2)–C(1)–O(1)	111.0(2)
N(1)–C(4)–C(5)	125.8(6)	N(1)–C(5)–C(4)	115.4(3)	O(1)–C(1)–H(1)	124.5000
N(2)–C(5)–C(4)	127.5(6)	N(1)–C(5)–C(6)	124.4(3)	O(1)–C(4)–C(5)	118.10(16)
N(2)–C(5)–C(6)	124.0(2)	C(4)–C(5)–C(6)	120.2(3)	N(1)–C(5)–C(4)	116.55(16)

The cell packing diagrams of 1–3 along the a-axis direction are shown in Figures 8–10, respectively, and the relative hydrogen bond parameters are listed in Table 3. As shown in Figure 8, there are two identical intermolecular hydrogen bonds O(1)–H(1A)···N(2) and N(1)···H(1)–O(1) with central inverse symmetry between two adjacent molecules of 1 in the cell; each hydrogen bond is formed between the oxime –OH hydrogen atom from one molecule and the imine N atom from the other molecule. These two intermolecular hydrogen bonds connect adjacent molecules into a pair of central inversely symmetric homodimeric oximes (as indicated in Figure 11), and the dimers are recombined into a layered structure by intermolecular forces. It can be seen from Figures 9 and 10 that 2 and 3 have similar intermolecular hydrogen bond and packing patterns as 1. The intermolecular hydrogen bonds in the cell of 2 are O(1)–H(1)···N(1) and N(1)···H(1)–O(1), and in the cell of 3 are O(2)-H(2)···N(1) and N(1)···H(2)-O(2).

Figure 8.

The cell packing diagram of 1 along the a-axis.

Figure 9.

The cell packing diagram of 2 along the a-axis.

Figure 10.

The cell packing diagram of 3 along the a-axis.

Figure 11.

Homodimeric oximes formed via hydrogen bonds.

Table 3.

Hydrogen bond geometry (Å, °) for 1–3.

Compound	D	H	A	D—H/Å	H. . .A/Å
1	N1	H1	O1	0.860	2.096
	O1	H1A	N2	0.820	1.953
2	O1	H1	N1	0.820	2.091
3	O2	H2	N1	0.820	2.111

Symmetry code: −x, 1 − y, 1 − z.

The molecular optimization calculation was performed with the path selection of DFT/Opt + Freq /UB3LYP and the 6-31G base set. The cis–trans-isomers of 1–3 are theoretically optimized to calculate the lowest total energy, dipole moment, and central atomic charge of the molecule. The results are shown in Table 4. The data for ∆E indicate that the total energy of (Z)-1 is significantly lower than that of (E)-1 (∆E₁ = 4.0132 kJ/mol), while that of (E)-3 is lower than that of (Z)-3 (∆E₃ = 3.7128 kJ/mol), so (Z)-1 and (E)-3 are more dominant in stability. These theoretical results are consistent with the experimental conclusions from ¹H NMR and XRD, and so on. In addition, the energy of (Z)-2 is slightly lower than that of (E)-2 (∆E₂ = 1.9182 kJ/mol), which indicates that the two isomers have almost the same stability, so they may co-exist. This theoretical prediction is confirmed by the ¹H NMR spectrum, in which signals for both (Z)-2 and (E)-2 appear. The most stable molecular structures of 1–3 through the Gaussian optimization are shown as Figure 12.

Table 4.

A summary of the structure optimization calculations for 1–3.

Calcd	Total energy (kJ/mol)	Dipole moment (debye)	Central atom charge (e)	∆E (kJ/mol)
(Z)-1	−1,089,542.6673	1.4709	−0.9695	4.0132
(E)-1	−1,089,538.6541	1.8254	−0.9544
(Z)-2	−1,149,533.6485	0.8485	0.4264	1.9182
(E)-2	−1,149,531.7303	0.8271	0.4368
(Z)-3	−1,997,503.7299	1.1321	−0.5221	3.7128
(E)-3	−1,997,507.4427	1.1794	−0.5226

Figure 12.

The most stable molecular structures of 1–3 through the Gaussian optimization.

Table 5 lists some important bond lengths and angles from the theoretical calculations and crystal structure determinations. It is found that most of the theoretical bonds are quite close, while a few are larger than those measured by XRD. This difference can be explained by the different treatment states employed. The Gaussian calculation studies use isolated gaseous molecules, while single-crystal diffraction performs analysis of a large number of molecules stacked in the solid state. The multiple particles in a crystal cell will inevitably interact with each other, thus affecting the molecular parameters. However, these factors are not taken into account in general gas-phase single molecule optimization, so there will be a difference between the two approaches. The theoretical bond angles are basically consistent with experimental angles, showing that the geometric parameter calculations were a good approximation with the experimental structure.

Table 5.

Selected theoretical and experimental bond lengths (Å) and angles (°) of 1–3.

Bond length/angle	1 (calcd)	1 (exp)	2 (calcd)	2 (exp)	3 (calcd)	3 (exp)
(N1/S1/O1)–C1	1.376	1.313	1.795	1.697	1.366	1.365
(N1/S1/O1)–C4	1.402	1.376	1.822	1.710	1.412	1.369
C4–C5	1.448	1.443	1.461	1.454	1.447	1.440
C5–N2	1.305	1.287	1.303	1.274	1.303	1.288
N2–O1	1.462	1.404	1.459	1.399	1.463	1.404
O(1)–H(1A)	0.977	0.820	0.977	0.820	0.977	0.820
N(2)–O(1)–H(1A)	102.283	109.500	102.707	109.522	102.376	109.500
O(1)–N(2)–C(5)	112.366	113.500	110.260	112.236	112.260	113.062
N(2)–C(5)–C(4)	128.006	127.500	115.345	115.402	126.150	116.551
C(4)–C(5)–C(6)	119.017	117.800	122.065	120.217	119.343	119.156

According to Frontier Molecular Orbital Theory, ¹⁷ in a molecule, the energies of the highest occupied molecular orbital (HOMO) and the lowest occupied molecular orbital (LUMO) determine the electron gain and loss and transfer ability, as well as the reactivity and spatial orientation of intermolecular reactions. In order to predict the coordination ability of these ketoximes as chelating ligands, the Frontier Molecular Orbital Energies: E_HOMO, E_LUMO, and energy gap E (∆E = E_LUMO − E_HOMO) ¹⁸ of four optimized isomers of 1–3 are further calculated with the Gaussian09W software at the B3LYP/6-31G level, as shown in Table 6. The value of E_HOMO is in the order of (E)-2 > (Z)-2 > (E)-3 > (Z)-1, while E_LUMO is in the order of (E)-2 > (Z)-2 > (Z)-1 > (E)-3. In general, the organic ligand donates 1–3 its lone pair of electrons to the empty valence electron orbital of the central atom or ion to form a coordination compound. The larger the E_HOMO, the more easily electrons are donated, that is, (E)-2 and (Z)-2 are more active than (Z)-1 and (E)-3 as far as the coordination ability is concerned. The greater the energy of E_LUMO, the more likely electrons are accepted from the central atom to form back donating bonding and strengthen the coordination bonds further. In addition, theoretical calculations showed that there was little difference in the frontier orbital energies between (E)-2 and (Z)-2, which shows that these two isomers have almost the same activity. The above predictions lay foundations for the coordination reactions of five-membered heterocyclic ketoximes and transition metals, so as to obtain polynuclear self-assembled metal complexes with novel structures.

Table 6.

Frontier molecular orbital energies of 1–3.

	ELUMO (a.u.)	EHOMO (a.u.)	∆E (eV)
(Z)-1	−0.16403	−0.29990	3.7111
(Z)-2	−0.16351	−0.28413	3.2861
(E)-2	−0.16201	−0.28322	3.2981
(E)-3	−0.16863	−0.29721	3.4986

Conclusion

In this paper, a series of five-membered heterocyclic ketoximes 1–3 has been synthesized by the Schiff base condensation. The ¹H NMR spectra of 1 and 3 show a full set of proton signals that are identified as pure (Z)- 1 or (E)- 3 isomers, while 2 displays double-image signals, which are identified as being due to a mixture of (Z)- 2 (55%) and (E)- 2 (45%) isomers. The UV-Vis spectra of 1 and 3 measured in different solvents are unimodal, while 2 is bimodal, indicating that 1 and 3 are single pure isomers in the four solvents examined, while 2 is a mixture of cis–trans-isomers. These results are consistent with the results analyzed by ¹H NMR. Comparing UV-Vis spectra in different solvents, it was found that a blue shift of λ_max occurs on increasing the solvent polarity, which may be due to solvation between the compound and polar solvent. The fluorescence emission spectra show that each of 1–3 has two regular peaks, which are due to π* → π and π* → n electron transitions. A large π-conjugated system and rigid structure are beneficial to the generation of fluorescence. The crystal structure of 1 indicates a cis-structure, while 2 and 3 have trans-structures. By molecular optimization calculations, the data of ∆E indicate that (Z)-1 and (E)-3 are more dominant in stability, which is consistent with the experimental conclusions from ¹H NMR and XRD and so on. The energies of the two isomers of (Z)-2 and (E)-2 indicate that the isomers have almost the same stability, so they may co-exist. This theoretical predict is consistent with the ¹H NMR data, in which signals due to both of (Z)-2 and (E)-2 appear. From Frontier Molecular Orbital Energy calculations of the four optimized isomers of 1–3, it is predicted that (E)-2 and (Z)-2 are more active than (Z)-1 or (E)-3 as far as the coordination ability is concerned.

Experimental

General procedures and materials

IR spectra were obtained with a Perkin–Elmer FTIR 2000 spectrometer. C, H, and N analyses were performed with an HP-MOD 1106 microanalyzer. ¹H NMR spectra were recorded with a Bruker AVANCE III 500 spectrometer. The UV spectra were recorded on a UV-Vis spectrophotometer UV-2600. Fluorescence spectra were recorded on the Seamer Fisher fluorescence spectrophotometer. Melting points were measured with an X-5 micro-melting point measuring instrument. Single-crystal data were measured using an X-ray single-crystal surface detector in a BRUKER SMART APEX II CCD single-crystal diffractometer. 2-Acetylpyrrole and 2-acetylthiophene were purchased from the Acros Co., and 2-acetylfuran was purchased from the Alfa Aesar Chemical Co., Ltd. Hydroxylamine hydrochloride was purchased from the Tianjin Tianli Chemical Reagent Co., Ltd and used as received. All reagents were used without further purification.

Synthesis of five-membered heterocyclic ketoximes 1–3 ¹⁹

(Z)-1-(1H-pyrrol-2-yl)ethanone oxime (1)

2-Acetylpyrrole (0.50 g, 4.59 mmol) and hydroxylamine hydrochloride (0.63 g, 9.18 mmol) in a molar ratio of 1:2 were dissolved in 15 mL of distilled water, and then mixed in a 50-mL round-bottom flask. After heating and stirring in a water bath at 55 °C for 8 h, aqueous sodium hydroxide solution (5 mL, 0.36 g, 9.18 mmol) was added dropwise to the mixture using a dropping funnel. TLC was used to monitor the progress of the reaction to determine the endpoint. The obtained solution was left overnight, during which time a large number of fine, white, needle-like crystals appeared. After being filtered and dried, a white solid (0.32 g, 64%) was obtained. M.p. 416–418 K. (Z)-1: IR (KBr, cm⁻¹): v_C═N 1630.7, ν_–OH 3417.6, 3130.5, ν_pyrrole 1461.8, 1419.1, 1349.8, 1329.2. (Z)-1: ¹H NMR (400 MHz, CDCl₃): 10.64 (s, 1H, HO–N=C), 7.26 (s, 1H, N–H), 7.26 (s, 1H, N–H), 6.98 (d, J = 3.42 Hz, 1H, X–(CH)–CH=CH–), 6.57 (d, J = 3.34 Hz, X–CH–CH=(CH)–), 7.04 (dd, J = 2.04, 2.68 Hz, 1H, X–CH–(CH)=CH–), 2.23 (d, J = 4 Hz, 3H, –N=C(CH₃)). Anal. calcd for C₆H₈N₂O: C, 58.05; H, 6.50; N, 22.57; found: C, 57.77; H, 6.35; N, 22.37. On dissolving an appropriate amount of 1 in 4 mL of a mixed solvent (V(ethyl acetate)/V(petroleum ether) = 3:1), a colorless rod-like crystal suitable for X-ray single-crystal diffraction analysis was grown after a week.

(Z), (E)-1-(1H-thiophen-2-yl)ethanone oxime (2)

2-Acetylthiophene (2 mL, 1.00 g, 7.92 mmol) in anhydrous ethanol 10 mL was mixed with a solution of hydroxylamine hydrochloride (1.10 g, 15.84 mmol) in distilled water (10 mL) in a 50-mL round-bottom flask. After heating and stirring in an oil bath at 78 °C for 5 h, aqueous sodium hydroxide solution (5 mL, 0.63 g, 15.84 mmol) was added dropwise to the mixture using a dropping funnel. TLC was used to monitor the progress of the reaction to determine the endpoint. The obtained solution was transferred to a 500-mL beaker, diluted with 400 mL of distilled water, and after standing overnight, a large amount of a white powder separated. After being filtered and dried, the obtained crude product was recrystallized from ethyl acetate 10 mL to give a white solid (0.64 g, 57%). M.p. 386–388 K. (E)-2: IR (KBr, cm⁻¹): ν_C=N 1631.5, ν_–OH 3167.6, ν_thiophene 1419.0, 1349.0, 1298.5. (E)-2: ¹H NMR (400 MHz, CDCl₃): 9.32 (s, 1H, HO–N=C), 7.52 (d, J = 2.04 Hz, 1H, X–(CH)–CH=CH–), 7.27 (d, J = 2.68 Hz, 1H, X–CH–CH=(CH)–), 6.54 (dd, J = 1.67, 1.81 Hz, 1H, X–CH–(CH)=CH–), 2.32 (s, 3H, –N=C(CH₃)). (Z)-2: ¹H NMR (400 MHz, CDCl₃): δ 9.36 (s, 1H, HO–N=C), 7.58 (d, J = 1.28 Hz, 1H, X–(CH)–CH=CH–), 7.29 (d, J = 1.09 Hz, 1H, X–CH–CH=(CH)–), 6.92 (dd, J = 3.34, 3.42 Hz, 1H, X–CH–(CH)=CH–), 2.40 (s, 3H, –N=C(CH₃)). Anal. calcd for C₆H₇NOS: C, 51.04; H, 5.00; N, 9.92; found: C, 51.11; H, 5.02; N, 9.89. On dissolving an appropriate amount of 2 in 4 mL of a mixed solvent (V(ethanol)/V(petroleum ether) = 3:1), a colorless, acicular crystal suitable for X-ray single-crystal diffraction analysis was grown after a week.

Synthesis of (E)-1-(1H-furan-2-yl)ethanone oxime (3)

2-Acetylfuran (2.3 mL, 1.00 g, 9.08 mmol) was added in 10 mL of anhydrous ethanol and hydroxylamine hydrochloride (1.26 g, 18.16 mmol) in 10 mL of a distilled water solution was mixed in a 50-mL round-bottom flask. After heating and stirring in an oil bath at 120 °C for 6 h, aqueous sodium hydroxide (5 mL, 0.72 g, 18.16 mmol) was added dropwise to the mixture using a drop funnel, the initial reaction solution was reddish brown and then turned dark yellow, and TLC was used to monitor the progress of reaction to determine the endpoint. The obtained solution was transferred to a 500-mL beaker and diluted with 400 mL distilled water. A large amount of light yellow powder was precipitated overnight, the filter residue was recrystallized with 10 mL ethyl acetate to get light yellow solid (0.82 g, 73%). M.p. 384–385 K. (E)-3: IR (KBr, cm⁻¹): ν_C=N 1627.1, ν_–OH 3387.5; ν_furan 1493.1, 1349.4, 1299.7. (E)-3: ¹H NMR (400 MHz, CDCl₃): 7.30 (s, 1H, HO–N=C), 7.46 (d, J = 1.81 Hz, 1H, X–(CH)–CH=CH–), 6.64 (d, J = 1.67 Hz, 1H, X–CH–CH=(CH)–), 7.12 (dd, J = 1.09, 1.28 Hz, 1H, X–CH–(CH)=CH–), 2.22 (s, 3 H, –N=C(CH₃)). Anal. calcd for C₆H₇NOS: C, 57.59; H, 5.64; N, 11.19; found: C, 57.69; H, 5.60; N, 11.02. On dissolving a proper amount of 3 in 4 mL of a mixed solvent (V(carbon tetrachloride)/V(petroleum ether) = 3:1), a yellowish acicular crystal suitable for X-ray single-crystal diffraction analysis was grown after a week.

Crystal structure determination and refinement of 1–3

The single crystals of 1, 2, and 3 with appropriate volumes (1: 0.37 mm × 0.30 mm × 0.24 mm; 2: 0.37 mm × 0.29 mm × 0.14 mm; 3: 0.38 mm × 0.30 mm × 0.24 mm) and suitable for X-ray single-crystal diffraction analysis were analyzed, using MoKα radiation (λ = 0.071073 nm) at 296 (2) K, in ω/2θ scanning mode. Total diffraction points (1: 3043, 2: 3213, 3: 3128), including independent diffraction points (1: 1130 points (R_int = 0.1269), 2: 1213 points (R_int = 0.0438), 3: 1124 points (R_int = 0.0270)), were collected in three ranges (1: 2.80°⩽ θ ⩽ 25.10°, 2: 2.53° ⩽ θ ⩽ 25.09°, 3: 2.53° ⩽ θ ⩽ 25.10°). The intensity data were corrected by SADABS, and the crystal structures were obtained by direct methods. All calculations were accomplished with the by SHELX-97 program. ²⁰ All hydrogen atoms were obtained by difference FOURIER synthesis and theoretical hydrogenation. The coordinates of the main non-hydrogen atoms were obtained according to the E diagram, and the other non-hydrogen atom coordinates were solved step-by-step by several rounds of difference FOURIER synthesis. The coordinates of all non-hydrogen atoms and their anisotropic temperature factors were modified by the full matrix least square method.

Crystals 1, 2, and 3 belong to the monoclinic system, and their data are as follows: 1: P2 (1) space group, unit cell parameters: a = 4.634(7), b = 7.836(10) Å, c = 17.44(2) Å; α =90°, β = 90°, γ = 90°; V = 633.3(14) ų, Z = 4, Dc = 1.302 mg/m³, μ = 0.092 mm⁻¹, and F(000) = 264. The range of diffraction indexes is −4 ⩽ h ⩽ 5, −9 ⩽ k ⩽7, −16 ⩽ l ⩽20, and the final structure deviation factors are R₁ = 0.1092, wR₂ = 0.2629, and GOF = 1.064. 2: P2 (1) space group, unit cell parameters: a = 8.420(5), b = 14.969(10) Å, c = 5.667(4) Å; α = 90°, β = 107.148(10)°, γ = 90°; V = 682.5(8) ų, Z = 4, Dc = 1.374 mg/m³, μ = 0.385 mm⁻¹, and F(000) = 296. The range of diffraction indexes is −10 ⩽ h ⩽ 10, −15 ⩽ k ⩽ 17, −6 ⩽ l ⩽ 6. The final structural deviation factors are R₁ = 0.0554, wR₂ = 0.1260, and GOF = 1.095. 3: P2 (1) space group, unit cell parameters: a = 8.329(3), b = 14.340(5) Å, c = 5.5349(18) Å; α = 90°, β = 104.753(5)°, γ = 90°; V = 633.3(14) ų, Z = 4, Dc = 1.300 mg/m³, μ = 0.092 mm⁻¹, and F(000) = 264. The range of diffraction indexes is −9 ⩽ h ⩽ 8, −11 ⩽ k ⩽ 17, −6 ⩽ l ⩽ 6. The final structural deviation factors are R₁ = 0.0489, wR₂ = 0.1240, and GOF = 1.051.

Computational details

All calculations were performed with the Gaussian 09 software package, using the B3LYP hybrid. The functional included a mixture of Hartree–Fock exchange with DFT exchange correlation. The geometry optimizations were accomplished without symmetry constraints using the 6-31G basis set for all atoms. Single-point energy calculations were performed on the B3LYP geometries using the same functional and a standard 6-31G basis set for all elements.^21–24

Supplemental Material

sj-docx-1-chl-10.1177_17475198211032538 – Supplemental material for Synthesis, isomerization, and DFT studies of five-membered heterocyclic ketoximes

Supplemental material, sj-docx-1-chl-10.1177_17475198211032538 for Synthesis, isomerization, and DFT studies of five-membered heterocyclic ketoximes by Biyun Su, Qiaoqiao Han, Xiaoteng Li, Yifan Hou, Jindi Wu, Li Wang and Liqin Ding in Journal of Chemical Research