Spectroscopic and Theoretical Studies of Some 3-(4′-Substituted phenylsulfanyl)-1-methyl-2-piperidones

Abstract

The analysis of the IR carbonyl bands of some 3-(4′-substituted phenylsulfanyl)-1-methyl-2-piperidones 1–6 bearing substituents: NO₂ (compound 1), Br (compound 2), Cl (compound 3), H (compound 4) Me (compound 5) and OMe (compound 6) supported by B3LYP/6-31+G(d,p) and PCM calculations along with NBO analysis (for compound 4) and X-ray diffraction (for 2) indicated the existence of two stable conformations, i.e., axial (ax) and equatorial (eq), the former corresponding to the most stable and the least polar one in the gas phase calculations. The sum of the energy contributions of the orbital interactions (NBO analysis) and the electrostatic interactions correlate well with the populations and the ν_CO frequencies of the ax and eq conformers found in the gas phase. Unusually, in solution of the non-polar solvents n-C₆H₁₄ and CCl₄, the more intense higher IR carbonyl frequency can be ascribed to the ax conformer, while the less intense lower IR doublet component to the eq one. The same ν_CO frequency trend also holds in polar solvents, that is ν_{CO (eq)} < ν_{CO (ax)}. However, a reversal of the ax/eq intensity ratio occurs going from non-polar to polar solvents, with the ax conformer component that progressively decreases with respect to the eq one in CHCl₃ and CH₂Cl₂, and is no longer detectable in the most polar solvent CH₃CN. The PCM method applied to compound 4 supports these findings. In fact, it predicts the progressive increase of the eq/ax population ratio as the relative permittivity of the solvent increases. Moreover, it indicates that the computed ν_CO frequencies of the ax and eq conformers do not change in the non–polar solvents n-C₆H₁₄ and CCl₄, while the ν_CO frequencies of the eq conformer become progressively lower than that of the ax one going from CHCl₃ to CH₂Cl₂ and to CH₃CN, in agreement with the experimental IR values. The analysis of the geometries of the ax and eq conformers shows that the carbonyl oxygen atom of the eq conformer is free for solvation, while the O_[CO]^…H_[o-Ph] hydrogen bond that takes place in the ax conformer partially hinders the approach of the solvent molecules to the carbonyl oxygen atom. Therefore, the larger solvation that occurs in the carbonyl oxygen atom of the eq conformer is responsible for the observed and calculated decrease of the corresponding frequency. The X-ray single crystal analysis of 2 indicates that this compound adopts the most polar eq geometry in the solid. In fact, in order to obtain the largest energy gain, the molecules are arranged in the crystal in a helical fashion due to dipole moment coupling along with C-H^…O and C-H^…π_Ph hydrogen bonds.

1. Introduction

The analysis of the infrared (ν_CO) band and theoretical B3LYP/6-311++G(d,p) calculations of some N,N-diethyl-2-[(4′-substituted) phenylsulfanyl acetamides Et₂NC(O)CH₂SC₆H₄-Y (Y=OMe, Me, H, Cl, Br, NO₂) [1] indicated the existence of two pairs (anti and syn) of cis (c) and gauche (g) conformers in the gas phase, with the gauche conformers being significantly more stable (ca. 90%) than the cis ones. The anti-geometry is more stable than the syn one for each pair of cis and gauche conformers. The summing up of the σ_C-S→π*_C=O, π_C=O→σ*_C-S, π*_C=O→σ*_C-S (NBO) orbital interactions is the main factor that stabilises the gauche conformers to a larger extent than the cis ones for which these interactions are absent. The similarity of the carbonyl frequencies in the gas phase for the anti and syn pairs of the cis and gauche conformers originates a carbonyl doublet in solution for which the less intense, higher frequency component is ascribed to the pair of the cis conformers and the lower frequency component corresponds to the more stable pair of the gauche conformers.

In solution of the non-polar solvents (n-C₆H₁₄ and CCl₄), the least polar gauche conformers (ca. 85%) are predominant with respect to the most polar cis ones. In the polar solvent, CHCl₃, a decrease of the gauche/cis population ratio is observed, i.e., 60% / 40%, and in the highly polar solvent, CH₃CN, the most polar cis conformer is the only one present. This trend was supported by the PCM solvation model performed by the B3LYP/6-311++G(d,p) method.

Theoretical and spectroscopic (IR, NMR and microwave) studies [2,3,4,5] of 2-piperidones showed that the piperidone ring lies in a half-chair or slightly twisted half-chair conformation. X-ray single crystal structure determination of the 3-chloro-2-piperidone [6] has shown that the piperidone ring is in a slightly twisted half-chair conformation with the 3-substituent assuming the quasi-equatorial [Cl-C-C=O (dihedral angle) = −30°] geometry. Moreover, our theoretical and X-ray single crystal studies of some 3-(4′-substituted phenylsulfonyl)-1-methyl-2-piperidones [7,8] have shown that, in these compounds, the 2-piperidone ring assumes a slightly distorted half-chair conformation. In the gas phase the 3-substituent adopts the quasi-equatorial, sin-clinal and quasi-axial geometries with respect to S-C-C=O dihedral angles of ca. 31°, 46° and 71°, respectively, while in the solid state the preferred geometry is quasi-axial with respect to the S-C-C=O dihedral angle of ca. 72°.

Aiming to throw more light on the nature of the orbital and electrostatic interactions that influence the stability of the cis-gauche conformers of 2-(4′-substituted)-phenylsulfanylamides [1], this paper reports the IR study of some six-member ring lactams, i.e., 3-(4′-substituted phenylsulfanyl)-1-methyl-2-piperidones bearing in the 4′ position the substituents, i.e., Y = NO₂ (compound 1), Br (compound 2), Cl (compound 3), H (compound 4) Me (compound 5) and OMe (compound 6) (Scheme 1) along with density functional theory (DFT), polarizable continuum model (PCM) and natural bond orbital (NBO) calculations for 4 taken as the representative compound for the whole series, along with the X-ray diffraction analysis of 2. These compounds were chosen taking into account some conformational rigidity of the piperidone ring, which allows the 3-substituent to assume almost exclusively the quasi-equatorial and axial conformations.

Scheme 1. Atom labeling of 3-(4′-substituted phenylsulfanyl)-1-methyl-2-piperidones and definition of relevant torsion angles.

Scheme 1. Atom labeling of 3-(4′-substituted phenylsulfanyl)-1-methyl-2-piperidones and definition of relevant torsion angles.

α = O(1)-C(2)-C(3)-S(4)

β = C(2)-C(3)-S(4)-C(20)

γ = C(3)-S(4)-C(20)-C(25)

δ = O(1)-C(2)-N(6)-C(10)

θ = O(1)-C(2)-N(6)-C(7)

θ’ = O(1)-C(2)-C(3)-C(9)

ω = C(2)-N(6)-C(7)-C(8)

ω’ = C(2)-C(3)-C(9)-C(8)

2. Results and Discussion

Table 1 displays the stretching frequencies and the absorbance percentage of the analytically resolved carbonyl band components for compounds 1–6 in solvents of increasing relative permittivity [9], i.e., n-C₆H₁₄ (ε = 1.9), CCl₄ (ε = 2.2), CHCl₃ (ε = 4.8), CH₂Cl₂ (ε = 9.1) and CH₃CN (ε = 38). A doublet is shown in n-hexane for derivatives 2–6 and in CCl₄ (fundamental and first overtone regions) for derivatives 1–6, with the higher frequency component being the more intense, i.e., ca. 80% and ca. 58%, respectively. However, as the relative permittivity increases, a reversal of the carbonyl doublet intensity is observed, with the lower frequency component progressively increasing up to ca. 70% in CHCl₃, ca. 88% in CH₂Cl₂ and emerges as a singlet in the highest relative permittivity solvent, CH₃CN. The solvent effect on the carbonyl band components for 4, chosen as a prototype for the series 1–6 is illustrated in Figure 1.

Figure 1. IR spectra of 3-(phenylsulfanyl)-1-methyl-2-piperidone(4) showing the analytically resolved carbonyl stretching band in: n-hexane (a), carbon tetrachloride [fundamental (b) and first overtone (c)], chloroform (d) dichloromethane (e) and acetonitrile (f).

Figure 1. IR spectra of 3-(phenylsulfanyl)-1-methyl-2-piperidone(4) showing the analytically resolved carbonyl stretching band in: n-hexane (a), carbon tetrachloride [fundamental (b) and first overtone (c)], chloroform (d) dichloromethane (e) and acetonitrile (f).

The unusual observed solvent effect (Table 1) might not be a clear evidence of the existence of conformational isomerism. However, the occurrence of two carbonyl band components in the first overtone region for 1–6 in carbon tetrachloride, with comparable intensity ratios and at frequencies twice that of the fundamental minus ca. 15 cm⁻¹ (that is two times the mechanical anharmonicity [10]), is indicative of the occurrence of two stable conformations for all compounds in this solvent, ruling out the existence of any vibrational effect in the fundamental transition of the ν_CO mode [11,12]. Therefore, the IR data strongly suggest the occurrence of two conformers for compounds 1–6 in all solvents, with the exception of CH₃CN, for which only one conformer is present.

Table 1. Frequencies (ν, cm⁻¹) and intensities of the carbonyl stretching bands in the IR spectra of 3-(4′-substituted-phenyl)sulfanyl-1-methyl-2-piperidones 1–6.

Compound		Y	n-C6H14		CCl4				CHCl3		CH2Cl2		CH3CN
			ν	Pa	ν	Pa	νb	P	ν	P	ν	P	ν	P
1		NO2	-c	-	1660	61	3300	60	1648	43	1656	9	-	-
			-	-	1652	39	3289	40	1637	57	1641	91	1645	100
2		Br	1668	84	1658	68	3299	63	1645	33	1652	12	-	-
			1656	16	1650	32	3291	37	1635	67	1641	88	1644	100
3		Cl	1668	79	1660	45	3306	49	1647	30	1656	13	-	-
			1656	21	1652	55	3292	51	1635	70	1641	87	1644	100
4		H	1669	83	1659	62	3300	59	1647	32	1656	9	-	-
			1654	17	1651	38	3290	41	1635	68	1641	91	1644	100
5		Me	1668	72	1657	57	3300	57	1645	16	1652	12	-	-
			1659	28	1651	43	3286	43	1634	84	1640	88	1643	100
6		OMe	1668	78	1660	44	3301	41	1647	25	1652	15	-	-
			1657	22	1651	56	3288	59	1635	75	1639	85	1643	100

^a Relative intensity of each component of the carbonyl doublet, expressed as a percentage of absorbance; ^b First overtone; ^c Compound slightly soluble in this solvent.

Table 1. Frequencies (ν, cm⁻¹) and intensities of the carbonyl stretching bands in the IR spectra of 3-(4′-substituted-phenyl)sulfanyl-1-methyl-2-piperidones 1–6.

Compound		Y	n-C6H14		CCl4				CHCl3		CH2Cl2		CH3CN
			ν	Pa	ν	Pa	νb	P	ν	P	ν	P	ν	P
1		NO2	-c	-	1660	61	3300	60	1648	43	1656	9	-	-
			-	-	1652	39	3289	40	1637	57	1641	91	1645	100
2		Br	1668	84	1658	68	3299	63	1645	33	1652	12	-	-
			1656	16	1650	32	3291	37	1635	67	1641	88	1644	100
3		Cl	1668	79	1660	45	3306	49	1647	30	1656	13	-	-
			1656	21	1652	55	3292	51	1635	70	1641	87	1644	100
4		H	1669	83	1659	62	3300	59	1647	32	1656	9	-	-
			1654	17	1651	38	3290	41	1635	68	1641	91	1644	100
5		Me	1668	72	1657	57	3300	57	1645	16	1652	12	-	-
			1659	28	1651	43	3286	43	1634	84	1640	88	1643	100
6		OMe	1668	78	1660	44	3301	41	1647	25	1652	15	-	-
			1657	22	1651	56	3288	59	1635	75	1639	85	1643	100

^a Relative intensity of each component of the carbonyl doublet, expressed as a percentage of absorbance; ^b First overtone; ^c Compound slightly soluble in this solvent.

Aiming to estimate the geometry of the minimum energy conformations in the gas phase, B3LYP/6-31+G(d,p) calculations were performed for compound 4, taken as representative of the whole series 1–6. Moreover, in order to estimate the solvent effect on the minimum energy conformations of 6, the PCM method was applied to fully optimise the referred conformations in the n-C₇H₁₆, CCl₄, CHCl₃, CH₂Cl₂ and CH₃CN solvents, and to compute the corresponding vibrational frequencies. The relevant data obtained for 4 are presented in Table 2.

Theoretical calculations in the gas phase show the existence of two stable conformations for compound 4. Due to the significant contribution of the [-O‒C=N+] resonance structure for the hybrid [13,14], the [C(O)NC(10)C(7)] carboxamide moiety (Scheme 1) is almost planar (δ and θ dihedral angles are ca. 3° and ca. 170°, respectively). This induces the piperidone ring to adopt a slightly distorted half-chair geometry, with the ω and ω′ dihedral angles being ca. 27° and ca. 51°, respectively (Table 2 and Figure 2) and the 3-phenylsulfanyl substituent assuming the almost axial (ax) and equatorial (eq) geometries with respect to the carbonyl group, at α angles of ca. 83° and ca. 29°, respectively. Moreover, as shown by the θ, θ′, ω and ω′ dihedral angles for the equatorial conformer (Table 2), in the gas phase of 4 and in the solid state of 2, the substituents at the 4′ position have no influence on the geometry of the lactam moiety.

Table 2. Relative energies (E, kJ mol⁻¹), dipole moments (μ, D), carbonyl frequencies (ν, cm⁻¹), and selected torsion angles (deg) calculated for the minimum energy conformations of 3-phenylthio-1-methyl-2-piperidone 4 in gas phase and in solvents with the PCM method at the B3LYP/6-31+G(d,p) level, and the X-ray geometric data of 2.

	Conf. a	E b	P c	μ	νCO	Torsion angles/º d
α						β	γ	δ	θ	θ’	ω	ω'
gas	ax	0	96.8	3.62	1713.4	82.0	−101.3	48.2	2.13	167.9	−154.3	23.6	−50.8
	eq	8.45	3.2	5.39	1720.1	30.4	−176.5	73.2	−3.90	−170.8	157.1	−23.7	50.8
C7H16	ax	0	84.4	4.08	1696.7	82.0	−101.9	48.4	2.06	168.2	−154.5	23.3	−50.7
	eq	4.20	15.6	6.21	1699.0	31.6	−177.4	72.9	−3.74	−172.2	158.2	−22.8	50.3
CCl­4	ax	0	80.0	4.16	1694.2	81.9	−102.1	49.1	2.06	168.1	−154.5	23.4	−50.7
	eq	3.43	20.0	6.37	1694.3	31.1	−176.2	73.3	−3.75	−171.7	157.7	−23.1	50.6
CHCl3	ax	0.17	48.3	4.61	1676.8	81.9	−102.4	47.7	2.11	168.3	−154.5	23.2	−50.7
	eq	0	51.7	7.20	1673.0	33.4	−178.1	72.1	−3.28	−174.9	159.9	−20.6	49.7
CH2Cl2	ax	2.01	30.8	4.89	1668.9	80.6	−104.3	48.5	2.57	169.3	−155.7	23.1	−50.1
	eq	0	69.2	7.62	1663.8	32.7	−178.4	71.4	−3.20	−174.3	159.1	−20.8	50.9
CH3CN	ax	3.93	16.9	5.18	1659.5	80.0	−105.6	48.9	2.60	169.8	−156.2	22.6	−49.8
	eq	0	83.1	8.10	1652.3	32.1	−178.4	70.7	−3.11	−173.2	158.5	−21.7	50.2
X-ray e	-	-	-	-	-	31.0 (3)	−176.2 (2)	−155.5(2)	−4.1 (4)	−177.6(3)	157.2(3)	−14.3(4)	53.1 (3)

^a Conformer designations ax and eq refers to the axial and equatorial conformers, respectively; ^b Relative energy; ^c Molar fraction in percentage; ^d See Scheme 1; ^e Refers to compound 2.

Table 2. Relative energies (E, kJ mol⁻¹), dipole moments (μ, D), carbonyl frequencies (ν, cm⁻¹), and selected torsion angles (deg) calculated for the minimum energy conformations of 3-phenylthio-1-methyl-2-piperidone 4 in gas phase and in solvents with the PCM method at the B3LYP/6-31+G(d,p) level, and the X-ray geometric data of 2.

	Conf. a	E b	P c	μ	νCO	Torsion angles/º d
α						β	γ	δ	θ	θ’	ω	ω'
gas	ax	0	96.8	3.62	1713.4	82.0	−101.3	48.2	2.13	167.9	−154.3	23.6	−50.8
	eq	8.45	3.2	5.39	1720.1	30.4	−176.5	73.2	−3.90	−170.8	157.1	−23.7	50.8
C7H16	ax	0	84.4	4.08	1696.7	82.0	−101.9	48.4	2.06	168.2	−154.5	23.3	−50.7
	eq	4.20	15.6	6.21	1699.0	31.6	−177.4	72.9	−3.74	−172.2	158.2	−22.8	50.3
CCl­4	ax	0	80.0	4.16	1694.2	81.9	−102.1	49.1	2.06	168.1	−154.5	23.4	−50.7
	eq	3.43	20.0	6.37	1694.3	31.1	−176.2	73.3	−3.75	−171.7	157.7	−23.1	50.6
CHCl3	ax	0.17	48.3	4.61	1676.8	81.9	−102.4	47.7	2.11	168.3	−154.5	23.2	−50.7
	eq	0	51.7	7.20	1673.0	33.4	−178.1	72.1	−3.28	−174.9	159.9	−20.6	49.7
CH2Cl2	ax	2.01	30.8	4.89	1668.9	80.6	−104.3	48.5	2.57	169.3	−155.7	23.1	−50.1
	eq	0	69.2	7.62	1663.8	32.7	−178.4	71.4	−3.20	−174.3	159.1	−20.8	50.9
CH3CN	ax	3.93	16.9	5.18	1659.5	80.0	−105.6	48.9	2.60	169.8	−156.2	22.6	−49.8
	eq	0	83.1	8.10	1652.3	32.1	−178.4	70.7	−3.11	−173.2	158.5	−21.7	50.2
X-ray e	-	-	-	-	-	31.0 (3)	−176.2 (2)	−155.5(2)	−4.1 (4)	−177.6(3)	157.2(3)	−14.3(4)	53.1 (3)

^a Conformer designations ax and eq refers to the axial and equatorial conformers, respectively; ^b Relative energy; ^c Molar fraction in percentage; ^d See Scheme 1; ^e Refers to compound 2.

The ax conformer is the most stable (ca. 97%) and the least polar (ca. 3.6 D) and the eq conformer is the least stable (ca. 3%) and the most polar (ca. 5.4 D). Moreover, in the gas phase of 4, the carbonyl frequencies for the ax and eq conformers are ca. 1713 cm⁻¹ and ca. 1720 cm⁻¹, respectively. As expected, this trend is in agreement with the computations of the N,N-diethyl-2-[(4′-substituted) phenylsulfanyl acetamides [1] for which, in the gas phase, the less polar gauche conformers are the most stable (lower carbonyl frequency) than the more polar cis conformers (higher carbonyl frequency) (see above). However, the theoretical (gas phase) carbonyl frequencies and the relative population values for the axial and equatorial conformers for 4 (Table 2) do not match with the IR frequencies and the relative intensities of each carbonyl component in solution (Table 1). Unusually, it seems reasonable to assign the more intense (ca. 80%) higher carbonyl frequency component in solution of non-polar solvent (n-C₆H₁₄) to the ax conformer and the less intense lower carbonyl frequency component to the eq one.

Table 2 shows that the torsional angles α-ω’ for the ax and eq conformers for 4 remain almost the same going from the gas phase to solvents of increasing relative permittivity, i.e., from n-C₇H₁₆ to CH₃CN. Furthermore, in this direction, there is a progressive decrease of the calculated ax/eq conformer population ratio, i.e., 97%/3% (gas), 84%/16% (n-C₇H₁₆), 80%/20% (CCl₄), 48%/52% (CHCl₃), 31%/69% (CH₂Cl₂) and 17%/83% (CH₃CN) followed by a simultaneous decrease of carbonyl frequency shifts (Δν = ν_eq – ν_ax) between the eq and ax conformer frequencies that is: ca. 7 cm⁻¹ (gas), ca. 2 cm⁻¹ (n-C₇H₁₆), ca. 0 cm⁻¹ (CCl₄), ca. −4 cm⁻¹ (CHCl₃), ca. −5 cm⁻¹ (CH₂Cl₂) and ca. −7 cm⁻¹ (CH₃CN). These trends cannot be justified on the grounds of the dipole moment (μ) analysis as the dipole moment for the ax and eq conformers increases almost to the same extent going from gas to acetonitrile solution, i.e., ca. 2.7 D.

A close match was not found between the computed ax/eq population ratios (PCM) and the experimental IR ones. However, the progressive decrease of the prevalence of the population of the ax conformer over the eq one going from n-C₇H₁₆ to CCl₄, along with the progressive increase of the prevalence of the population of the eq conformer over the ax one going from CHCl₃ to CH₂Cl₂ and to CH₃CN, seems to be in line with the experimental IR trend of the ax/eq population ratio. The fact that the calculated (PCM) carbonyl frequencies for the ax and eq conformers becomes practically coincident in the non-polar solvents n-C₇H₁₆ and CCl₄, and that the ν_CO frequency of the eq conformer becomes progressively smaller than the ν_CO frequency of the ax conformer going from CHCl₃ to CH₂Cl₂ and to CH₃CN, gives support to the experimental (IR) ax-eq carbonyl frequencies assignments (see above).

It should be pointed out that MP2/6-31+G(d,p) calculations for 4 [15] have shown that the population ratio for the ax/eq conformers (ca. 78%/22%) as well as the torsional angles α-ω’ values for both conformers, are close to those obtained at the B3LYP/6-31+G(d,p) level (Table 2). The computed MP2 carbonyl frequencies indicate that the ax and eq conformers have almost the same carbonyl frequency of ca. 1716 cm⁻¹. Moreover the single-point PCM calculations at the MP2 level [16] for the ax and eq conformers of 4 have shown that the solvent effect on the ax/eq population ratio is quite similar to that obtained with the B3LYP method.

Table 3 presents the ChELPG atomic charges [17] for selected atoms computed at the B3LYP/6-31+G(d,p) level for compound 4, while Table 4 displays the interatomic distances between some selected atoms and the difference between these contacts and the sum of the van der Waals (ΣvdW) radii.

Table 3. ChElPG charge (e) at selected atoms obtained at the B3LYP/6-31+G(d,p) level for 3-phenylsulfanyl-1-methyl-piperidone 4.

Conf.	O(1)	C(2)	S(4)	N(6)	H(26)	H(30)	H(5)	H(11)	H(16)	H(17)
ax	−0.557	0.525	−0.321	−0.187	0.131	0.102	0.048	0.088	0.028	0.065
eq	−0.537	0.605	−0.320	−0.311	0.103	0.061	0.027	0.058	0.030	0.045

Table 3. ChElPG charge (e) at selected atoms obtained at the B3LYP/6-31+G(d,p) level for 3-phenylsulfanyl-1-methyl-piperidone 4.

Conf.	O(1)	C(2)	S(4)	N(6)	H(26)	H(30)	H(5)	H(11)	H(16)	H(17)
ax	−0.557	0.525	−0.321	−0.187	0.131	0.102	0.048	0.088	0.028	0.065
eq	−0.537	0.605	−0.320	−0.311	0.103	0.061	0.027	0.058	0.030	0.045

Table 4. Selected interatomic distances (Å) for the ax and eq conformers of 3-phenylsulfanyl-1-methyl-2-piperidone 4 at the B3LYP/6-31+G(d,p) level.

Conf. a	O[1]...S[4] b	Δl c	O[1]...H[11] d	Δl	O[1]...H[26] d	Δl	O[1]...H[5] d	Δl
ax	3.37	+0.05	2.28	−0.44	2.35	−0.37	2.52	−0.20
eq	2.89	−0.43	2.27	−0.45	4.75	+2.03	2.83	+0.11

^a Conformer designation; ^b Sum of the van der Waals radii = 3.32 Å; ^c Difference between non bonded atoms distance and the sum of their van der Waals radii; ^d Sum of the van der Waals radii = 2.72 Å.

Table 4. Selected interatomic distances (Å) for the ax and eq conformers of 3-phenylsulfanyl-1-methyl-2-piperidone 4 at the B3LYP/6-31+G(d,p) level.

Conf. a	O[1]...S[4] b	Δl c	O[1]...H[11] d	Δl	O[1]...H[26] d	Δl	O[1]...H[5] d	Δl
ax	3.37	+0.05	2.28	−0.44	2.35	−0.37	2.52	−0.20
eq	2.89	−0.43	2.27	−0.45	4.75	+2.03	2.83	+0.11

^a Conformer designation; ^b Sum of the van der Waals radii = 3.32 Å; ^c Difference between non bonded atoms distance and the sum of their van der Waals radii; ^d Sum of the van der Waals radii = 2.72 Å.

Due to the decrease of the α dihedral angle going from the ax (ca. 80°) to the eq (ca. 31°) conformer there is in the eq conformer a short contact between the negatively charged O[1] (≅ −0.55e)…S[4] (≅ −0.32e) atoms whose interatomic distance is significantly shorter than the sum of the van der Waals radii (ΣvdW) (Δl ≅ −0.43Å) while for the ax conformer the same contact is close to the ΣvdW radii (Δl ≅ +0.05Å). The referred shorter O^…S contact of the eq conformer should be responsible for the stronger Repulsive Field Effect [11,18] between the C^δ+=O^δ- and C^δ+─S^δ− dipoles in the eq conformer, being responsible for both the destabilisation of this conformer relative to the ax one and for the higher carbonyl frequency observed in the gas phase of the eq conformer with respect to that of the ax one whose carbonyl frequency shift (ν_eq – ν_ax) is ca. +7 cm⁻¹ (Table 2).

As expected from the geometry of the carboxamide moiety of the 1-methyl-2-piperidones, there is a short contact between the oppositely charged O[1] (≅ −0.55e)^…H[11] (≅ 0.09e) atoms whose interatomic distance is significantly shorter than the ΣvdW radii (Δl ≅ −0.44Å; C-H^…O angle ca. 103°) for bothax and eq conformers (Figure 2). The O[1] (≅ −0.55e) and H[5] (≅ 0.05e) contact is shorter than ΣvdW radii by (Δl ≅ −0.20Å; C-H^…O angle ca. 84°) for theax conformer only. Moreover, as a consequence of the smaller γ dihedral angle in the ax conformer (ca. 48°) as compared with that of the eq one (ca. 73°) for 4 (Table 2) the positively charged o-phenylsulfanyl hydrogen atom H[26] (≅ 0.13e) gets closer to the negatively charged carbonyl oxygen atom O[1] (≅ −0.55e) Therefore, the interatomic distance becomes shorter than the ΣvdW radii by (Δl ≅ −0.37 Å; C-H^…O angle ca. 148°) for the ax conformer and larger than the ΣvdW radii (Δl ≅ +2.0 Å) for the eq one.

Figure 2. Axial (ax) (a) and equatorial (eq) (b) conformers for 4 obtained at the B3LYP/6-31+G(d,p) level.

Figure 2. Axial (ax) (a) and equatorial (eq) (b) conformers for 4 obtained at the B3LYP/6-31+G(d,p) level.

From these H^δ+…O^δ- contacts analysis it may be concluded that the O[1]^…H[11] short contact stabilises electrostatically both the eq and ax conformers to the same extent. However, the O[1]^…H[5] contact and mainly the O[1]^...H[26] one stabilise electrostatically the ax conformer only.

In order to rationalise the orbital interactions that stabilise the ax and eq conformers, NBO analysis [19] was performed and the results obtained for selected NBO energy interactions between donor and acceptor orbitals [20] are reported in Table 5.

Table 5. Comparison of significant NBO energies (kcal mol⁻¹) of the corresponding interacting orbitals for the conformers ax and eq of 3-phenylsulfanyl-1-methyl-2-piperidone 4 at the B3LYP/6-31+G(d,p) level.

Orbitals	ax	eq
LPN6→π*C2=O1	65.3	50.6
LPO1→σ*C2-N6	25.4	25.5
LPO1→σ*C2-C3	18.9	20.3
LPO1→σ*C21-H26	3.0	-
LPO1→σ*C10-H11	1.0	1.2
LPO1→ σ*S4-C20	- a	0.9
LPS4→ π*C2=O1	2.0	-
LPS4→ π *C20-C21	8.3	2.7
σC3-S4→π*C2=O1	4.2	0.7
σC3-S4→σ*C2-N6	-	3.4
πC2=O1→σ*C3-S4	1.7	0.6
π*C2=O1→σ*C3-S4	5.5	0.8
∑ Eb	135.3	106.7

^a Interaction energy smaller than 0.5 kcal mol⁻¹; ^b Summing up of the energies of the interacting orbitals.

Table 5. Comparison of significant NBO energies (kcal mol⁻¹) of the corresponding interacting orbitals for the conformers ax and eq of 3-phenylsulfanyl-1-methyl-2-piperidone 4 at the B3LYP/6-31+G(d,p) level.

Orbitals	ax	eq
LPN6→π*C2=O1	65.3	50.6
LPO1→σ*C2-N6	25.4	25.5
LPO1→σ*C2-C3	18.9	20.3
LPO1→σ*C21-H26	3.0	-
LPO1→σ*C10-H11	1.0	1.2
LPO1→ σ*S4-C20	- a	0.9
LPS4→ π*C2=O1	2.0	-
LPS4→ π *C20-C21	8.3	2.7
σC3-S4→π*C2=O1	4.2	0.7
σC3-S4→σ*C2-N6	-	3.4
πC2=O1→σ*C3-S4	1.7	0.6
π*C2=O1→σ*C3-S4	5.5	0.8
∑ Eb	135.3	106.7

^a Interaction energy smaller than 0.5 kcal mol⁻¹; ^b Summing up of the energies of the interacting orbitals.

The most important orbital interaction, the LP_N6→π*_C2=O1 corresponding to the [O=C–N ↔ ⁻O–C=N⁺] conjugation, has a value of ca. 65 kcal moL⁻¹ for the ax conformer, significantly higher than that of ca. 51 kcal mol⁻¹ for theeq one. The smaller orbital interaction value for the latter conformer is in agreement with the smaller contribution of the [⁻O–C=N⁺] canonical form to the resonance hybrid due to the strong Repulsive Field Effect between the C^δ+=O^δ- and C^δ+-S^δ− dipoles, as outlined above. On the other hand, the very short O^δ−[1]^…H ^δ+[13] contact with a suitable C─H^…O angle of ca. 148° in the ax conformer determines the LP_O1→σ*_C21-H26 orbital interaction (ca. 3 kcal mol⁻¹) (hydrogen bond) in the ax conformer. This interaction increases the weight of the polar [^-O‒C=N⁺] canonical form for the resonance hybrid and is partially responsible for the higher value of the LP_N6→π*_C2=O1 orbital interaction cited above. In addition, the higher occupancy value [20] of the σ*_C21-H26 orbital in the ax conformer (0.134 vs. 0.014) gives further support for the occurrence of the significant LP_O1→σ*_C21-H26 hydrogen bond interaction.

In the carboxamide moiety, there are two additional higher energy interactions (through bond coupling [21]), LP_O1→σ*_C2-N6 and LP_O1→ σ*_C2-C3, at mean values of ca.26 kcal mol⁻¹ and ca.20 kcal mol⁻¹, respectively, for the eq and ax conformers.

The more favourable γ dihedral angle of the phenylsulfanyl group (ca. 48°) for the ax conformer allows a reasonable 3p_S-π_Ph* orbital overlap that contributes to a larger stabilisation of the ax conformer with respect to the eq one [LP_S4→π*_C20-C21 orbital interaction of ca. 8.3 kcal mol⁻¹ (ax) and ca.2.7 kcal mol⁻¹ (eq)].

The O[1]^…H[11] short contacts are responsible for the weak LP_O1→σ*_C10-H11 orbital interactions of ca. 1.1 kcal mol⁻¹ that stabilise to the same extent both conformers.

It should be pointed out that the eq conformer is stabilised through two weak LP_O1→σ*_S4-C20 and σ_C3-S4→σ*_C2-N6 orbital interactions whose delocalisation energy values are ca. 0.9 kcal mol⁻¹ and 3.4 kcal mol⁻¹, respectively.

The suitable geometry of the ax conformer allows the occurrence of superjacent LP_S4→ π*_C2-O1 orbital interaction [22] whose value is 2.0 kcal mol⁻¹, which is absent for the eq conformer. Moreover, in the [O=C-C-S] moiety there are three π(π*)/σ(σ*) orbital interactions: π_C2=O1→σ*_C3S4, σ_C3-S4→π*_C2=O1 and π*_C2=O1→σ*_C3-S4 whose energy values progressively decrease as the π/σ overlap decreases, i.e., they present the maximum value for the ax conformer and the minimum one for the eq one. In fact, the sum of the energies of these interactions displays the value ca.11.4 kcal mol⁻¹ and ca. 2.1 kcal mol⁻¹ for the ax and eq conformers, respectively.

In conclusion, since the total sum of the delocalisation energies of the selected interactions has the maximum value of ca. 135 kcal mol⁻¹ for the ax conformer and significantly decreases to ca. 107 kcal mol⁻¹ for the eq one, the NBO analysis and the trend of the electrostatic interactions are in agreement with the higher relative abundance of the ax (ca. 97%) conformer with respect to the eq one, found in the gas phase, whose computed carbonyl frequencies are ca. 1713 cm⁻¹ and 1720 cm⁻¹, respectively.

However, there is no matching between the IR experimental and the theoretical gas phase data. As pointed out above, since a reversal of the IR carbonyl frequencies has been observed experimentally in solution, the higher frequency component has been ascribed to the ax conformer, while the lower frequency component to the eq one. This behaviour may be rationalised as follows.

A close inspection of the ax (a) and eq (b) conformers in the gas phase (Figure 2 and Scheme 2) shows that the eq conformer carbonyl oxygen atom is free for solvation, while the hydrogen bond that takes place between the ortho-phenylsulfanyl hydrogen atom and the carbonyl oxygen atom lone pair in the ax conformer partially prevents the solvent molecules' approach to the carbonyl oxygen atom.

Scheme 2. Differential solvent effect for 4 showing that the carbonyl oxygen atom in the axial conformer (a) is partially hindered for the approach of the solvent molecules due to [H_(o-Ph)…O_(CO)] hydrogen bond in comparison to the equatorial conformer (b) for which the carbonyl oxygen atom is free for salvation.

Scheme 2. Differential solvent effect for 4 showing that the carbonyl oxygen atom in the axial conformer (a) is partially hindered for the approach of the solvent molecules due to [H_(o-Ph)…O_(CO)] hydrogen bond in comparison to the equatorial conformer (b) for which the carbonyl oxygen atom is free for salvation.

Therefore, the larger solvation at the carbonyl oxygen atom in the more polar eq conformer contributes to the decrease of the carbonyl bond order and, as a consequence, lowers its frequency to a larger extent than that of the ax conformer.

X-ray single crystal analysis of 2 showed that in the solid state this compound assumes the equatorial geometry with respect to the torsion angle α (Scheme 1) and the 2-piperidone ring, similar to the gas phase of 4, displays a slightly distorted half-chair conformation (Table 2 and Figure 3).

Figure 3. The molecular structure of 2 showing atom labelling scheme and displacement ellipsoids at the 50% probability level (arbitrary spheres for the H-atoms).

Figure 3. The molecular structure of 2 showing atom labelling scheme and displacement ellipsoids at the 50% probability level (arbitrary spheres for the H-atoms).

The absolute values of the relevant torsional angles (α-ω’) in the solid state of 2 (Table 2) are similar to the corresponding torsional angles in the gas phase of the equatorial conformer of 4, except for the absolute value of the γ angle of ca. 155° for 2, which differs from the γ angle of ca. 73° for 4. The larger γ torsional angle for 2 allows the hydrogen bonding between the π phenyl orbital and the methylene [H14] atom (Figure 4).

Figure 4. A view of the supramolecular array of 2 where the C-H…O and C-H….π_Ph interactions are shown.

Figure 4. A view of the supramolecular array of 2 where the C-H…O and C-H….π_Ph interactions are shown.

In fact, in order to obtain the largest energy gain from the crystal packing, the molecules of the equatorial conformer of 2 are stabilised in the crystal through dipole moment coupling along with two hydrogen bond interactions in a helical fashion, i.e., C-H^…O (C3ⁱ-H5^i…O1: C3^i…O1 = 3.346(3), H5^i…O1 = 2.43 Å; C3ⁱ-H5^i…O1 = 155°; symmetry operation i = 1/2 + x, 1/2-y, 1-z) and C-H^…π (C7^…Cgⁱⁱ = 3.618(4), H14^…Cgⁱⁱ = 2.72 Å; C7-H14^…Cgⁱⁱ = 155°; symmetry operation ii = x − 1/2, −1/2−y, 1 − z). Additionally, similar to the gas phase of 4, in the crystal of 2 an intramolecular short contact [C10-H11^...O1: C^...O = 2.676(4); H^…O = 2.23 Å; C-H^...O = 107°] was found.

3. Experimental

3.1. Materials

All solvents for IR measurements were spectrograde and used without further purification. The 3-[4′-substituted phenylsulfanyl)]-1-methyl-2-piperidones 1–3, 5 and 6 are new compounds. Derivatives 2–6 were obtained from the reaction of 1-methyl-2-piperidone (a) with lithium diisopropylamide (b), in tetrahydrofuran (THF), followed by the addition of the THF solution of 4-substituted diphenyl disulphide (c), in the presence of hexamethylphosphoramide (d), in the molar ratio of 1(a):2(b):1(c):1(d) as described in the literature for 4 [23]. The appropriate diphenyl disulphide was prepared by the reaction of 4-substituted-thiophenol with bromine in dichloromethane on hydrated silica gel support [24]. The pure phenylsulfanyl piperidones 2–6 were obtained in 20–45% yield. The 4′-[nitrophenylsulfanyl]-1-methyl-2-piperidone 1 was obtained as follows: initially, the THF solution of the lithium enolate of 1-methyl-2-piperidone obtained as described above [23] reacted with the tosylate chloride [25] in a 1:1 ratio to obtain the 3-chloro-1-methyl-2-piperidone. This crude product was allowed to react with an ethanolic solution of 4-nitrothiophenol in the presence of equimolar proportion of potassium hydroxide. After usual workup and recrystallization from ethanol-water, the pure compound 1 was obtained in 24% yield. Suitable crystals for X-ray analysis for 2 were obtained by vapour diffusion from chloroform/n-hexane at 283K. The ¹H- and ¹³C-NMR data for compounds 1–6 are collated in Table 6 and elemental analysis results are shown in Table 7.

3.2. IR Measurements

The IR spectra were obtained on a FTIR Michelson-Bomem-MB100 spectrophotometer, with 1.0 cm^-1 resolution. The carbonyl region (1800–1600 cm⁻¹) was recorded for n-hexane, carbon tetrachloride, chloroform, dichloromethane and acetonitrile solutions, at a concentration of 2.0 × 10⁻² mol dm⁻³, using a 0.519 mm sodium chloride cell. The carbonyl first overtone region (3500–3100 cm⁻¹) was recorded for carbon tetrachloride and chloroform solutions (2.0 × 10⁻² mol dm⁻³) using a 1.00 cm quartz cell. The overlapped carbonyl bands (fundamental and first overtone) were deconvoluted by means of the Grams/32 curve-fitting program, version 4.04 Level II [26]. The populations of the axial and equatorial conformers were estimated from the maximum of each component of the resolved carbonyl doublet expressed in percentage of absorbance, assuming equal molar absorptivity coefficients for all the studied compounds 1–6, in each solvent.

Table 6. Physical, ¹H- and ¹³C-NMR data for 3-[(4′-substituted-phenyl)sulfanyl]-1-methyl-2-piperidones 1–6.

Compd.	Y	Mp (°C)	1H- and 13C-NMR a
1	NO2	77–78	1H-NMR: 8.14–8.12 (m, 2H), 7.64–7.63 (m, 2H), 4.06 (t, 1H, J = 5.5 Hz), 3.41–3.32 (m, 2H), 2.99 (s, 3H), 2.30–2.27 (m, 1H), 2.09–2.03 (m, 2H), 1.92–1.88 (m, 1H). 13C-NMR: 167.83, 146.07, 145.77, 128.67, 123.86, 49.82, 46.90, 35.36, 28.68, 20.76.
2	Br	64–64		1H-NMR: 7.44–7.40 (m, 4H), 3.82 (t, 1H, J = 5.7Hz), 3.33–3.26 (m. 2H), 2.96 (s, 3H), 2.11–2.04 (m, 2H), 1.99–1.94 (m, 1H), 1.82–1.77 (m, 1H). 13C-NMR: 168.36, 134.27, 134.07, 132.19, 121.82, 50.06, 49.02, 35.47, 28.83, 20.59.
3	Cl	b	1H-NMR: 7.53–7.51 (m, 2H), 7.30–7.29 (m, 2H), 3.84 (t, 1H, J = 5.7Hz), 3.36–3.28 (m, 2H), 2.99 (s, 3H), 2.17–2.08 (m, 2H), 2.02–1.96 (m, 1H), 1.83–1.80 (m, 1H) 13C-NMR: 168.19, 133.77, 133.64, 133.33, 129.06, 49.86, 48.96, 35.27, 28.62, 20.38.
4	H	38–39		1H-NMR: 7.56–7.54 (m, 2H), 7.32–7.28 (m, 2H), 7.26–7.24 (m, 1H), 3.88 (t, 1H, J = 5.0 Hz), 3.30–3.24 (m, 2H), 2.97 (s, 3H), 2.12–2.07 (m, 2H), 1.98–1.94 (m, 1H), 1.77–1.74 (m, 1H). 13C-NMR: 168.47, 134.96, 132.61, 129.12, 127.61, 50.08, 49.03, 35.44, 28.75, 20.46.
5	CH3	44–45	1H-NMR: 7.47–7.46 (m, 2H), 7.15–7.13 (m, 2H), 3.83 (t, 1H, J = 5,5Hz), 3.34–3.25 (m, 2H), 2.98 (s, 3H), 2.35 (s, 3H), 2.13–2.05 (m, 2H), 1.96-1.92 (m, 1H), 1.80–1.75 (m, 1H) 13C-NMR: 168.37, 137.77, 133.25, 130.79, 129.71, 49.89, 49.21, 35.23, 28.35, 21.12, 20.12.
6	OCH3	b		1H-NMR: δ (ppm): 7.51–7.49 (m, 2H), 6.85–6.83 (m, 2H), 3.79 (s, 3H), 3,71 (t, 1H, J = 6,0Hz), 3.25–3,23 (m, 2H), 2.94 (s, 3H), 2.10–2,01 (m, 2H), 1.94–1,89 (m, 1H), 1.75–1,71 (m, 1H). 13C-NMR: 168.36, 159.85, 135.93, 124.61, 114.50, 55.31, 49.89, 49.80, 35.20, 28.28, 20.12

^a1H- and ¹³C- chemical shifts in ppm relative to TMS and coupling constants in Hz at 500 MHz. in CDCl₃; ^b liquid at room temperature; ^c From [4]: b.p 130–133° / 0.05 Torr.

Table 6. Physical, ¹H- and ¹³C-NMR data for 3-[(4′-substituted-phenyl)sulfanyl]-1-methyl-2-piperidones 1–6.

Compd.	Y	Mp (°C)	1H- and 13C-NMR a
1	NO2	77–78	1H-NMR: 8.14–8.12 (m, 2H), 7.64–7.63 (m, 2H), 4.06 (t, 1H, J = 5.5 Hz), 3.41–3.32 (m, 2H), 2.99 (s, 3H), 2.30–2.27 (m, 1H), 2.09–2.03 (m, 2H), 1.92–1.88 (m, 1H). 13C-NMR: 167.83, 146.07, 145.77, 128.67, 123.86, 49.82, 46.90, 35.36, 28.68, 20.76.
2	Br	64–64		1H-NMR: 7.44–7.40 (m, 4H), 3.82 (t, 1H, J = 5.7Hz), 3.33–3.26 (m. 2H), 2.96 (s, 3H), 2.11–2.04 (m, 2H), 1.99–1.94 (m, 1H), 1.82–1.77 (m, 1H). 13C-NMR: 168.36, 134.27, 134.07, 132.19, 121.82, 50.06, 49.02, 35.47, 28.83, 20.59.
3	Cl	b	1H-NMR: 7.53–7.51 (m, 2H), 7.30–7.29 (m, 2H), 3.84 (t, 1H, J = 5.7Hz), 3.36–3.28 (m, 2H), 2.99 (s, 3H), 2.17–2.08 (m, 2H), 2.02–1.96 (m, 1H), 1.83–1.80 (m, 1H) 13C-NMR: 168.19, 133.77, 133.64, 133.33, 129.06, 49.86, 48.96, 35.27, 28.62, 20.38.
4	H	38–39		1H-NMR: 7.56–7.54 (m, 2H), 7.32–7.28 (m, 2H), 7.26–7.24 (m, 1H), 3.88 (t, 1H, J = 5.0 Hz), 3.30–3.24 (m, 2H), 2.97 (s, 3H), 2.12–2.07 (m, 2H), 1.98–1.94 (m, 1H), 1.77–1.74 (m, 1H). 13C-NMR: 168.47, 134.96, 132.61, 129.12, 127.61, 50.08, 49.03, 35.44, 28.75, 20.46.
5	CH3	44–45	1H-NMR: 7.47–7.46 (m, 2H), 7.15–7.13 (m, 2H), 3.83 (t, 1H, J = 5,5Hz), 3.34–3.25 (m, 2H), 2.98 (s, 3H), 2.35 (s, 3H), 2.13–2.05 (m, 2H), 1.96-1.92 (m, 1H), 1.80–1.75 (m, 1H) 13C-NMR: 168.37, 137.77, 133.25, 130.79, 129.71, 49.89, 49.21, 35.23, 28.35, 21.12, 20.12.
6	OCH3	b		1H-NMR: δ (ppm): 7.51–7.49 (m, 2H), 6.85–6.83 (m, 2H), 3.79 (s, 3H), 3,71 (t, 1H, J = 6,0Hz), 3.25–3,23 (m, 2H), 2.94 (s, 3H), 2.10–2,01 (m, 2H), 1.94–1,89 (m, 1H), 1.75–1,71 (m, 1H). 13C-NMR: 168.36, 159.85, 135.93, 124.61, 114.50, 55.31, 49.89, 49.80, 35.20, 28.28, 20.12

^a1H- and ¹³C- chemical shifts in ppm relative to TMS and coupling constants in Hz at 500 MHz. in CDCl₃; ^b liquid at room temperature; ^c From [4]: b.p 130–133° / 0.05 Torr.

Table 7. Elemental analysis data for 3-[(4′-substituted-phenyl)sulfanyl]-1-methyl-2-piperidones 1–6.

Compd	Y	Molecular formula		Analysis (%)
				C	H	N
1	NO2	C12H14N2O3S	Calc.Found	54.1254.34	5.305.23	10.5210.46
2	Br						C12H14BrNOS	Calc. Found	48.0147.91	4.704.77	4.674.79
3	Cl	C12H14ClNOS	Calc. Found	56.3556.26	5.525.22	5.485.27
4	H						C12H15NOS	Calc. Found	65.1265.12	6.836.56	6.336.46
5	CH3	C13H17NOS	Calc. Found	66.3466.23	7.287.20	5.956.01
6	OCH3						C13H17NO2S	Calc. Found	62.1262.13	6.826.81	5.575.55

Table 7. Elemental analysis data for 3-[(4′-substituted-phenyl)sulfanyl]-1-methyl-2-piperidones 1–6.

Compd	Y	Molecular formula		Analysis (%)
				C	H	N
1	NO2	C12H14N2O3S	Calc.Found	54.1254.34	5.305.23	10.5210.46
2	Br						C12H14BrNOS	Calc. Found	48.0147.91	4.704.77	4.674.79
3	Cl	C12H14ClNOS	Calc. Found	56.3556.26	5.525.22	5.485.27
4	H						C12H15NOS	Calc. Found	65.1265.12	6.836.56	6.336.46
5	CH3	C13H17NOS	Calc. Found	66.3466.23	7.287.20	5.956.01
6	OCH3						C13H17NO2S	Calc. Found	62.1262.13	6.826.81	5.575.55

3.3. X-ray Measurements

Crystal data for C₁₂H₁₄BrNOS (2): M = 300.21, T = 290(2) K, orthorhombic, P2₁2₁2₁, a = 7.3328(5), b = 7.7903(6), c = 21.502(2)Å, V = 1228.30(17) ų, Z = 4, D_x = 1.623 g cm⁻³, F(000) = 608, λ (Mo Kα) = 0.71073, μ = 3.495 mm⁻¹, R = 0.0275.

3.3.1. Data Collection and Processing

Data were collected on a Bruker APEX-II CCD diffractometer using Mo Kα radiation so that θ_max = 25.4°, no. of unique data = 2229, no. of parameters = 146, R (1964 data with I ≥ 2σ(I)) = 0.0275, wR (all data) = 0.0663. The structure was solved by direct methods [27] and refined with anisotropic displacement parameters for non- hydrogen atoms. The H atoms were geometrically placed (C—H = 0.93–0.98 Å) and refined as riding with Uiso(H) = 1.2–1.5Ueq(C). The weighting scheme used was w = 1/[σ²(F_o²) + 0.0193P² + 0.000P] where P = (F_o² + 2F_c²)/3) with SHELXL-97 [28] on F². The programs WinGX [29], ORTEP3 for Windows [30], PLATON [31], MarvinSketch 5.1.10 [32] and DIAMOND [33] were used for geometric calculations and to prepare crystallographic material for publication and depositing. Crystallographic data for the structural analysis have been deposited with the Cambridge Crystallographic Data Centre as CCDC 934425. Copies of this information may be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: 44 1223 336 033; e-mail: [email protected] or www: http://www.ccdc.cam.ac.uk).

3.4. Theoretical Calculations

All calculations were carried out (at 298 K) using methods and basis sets implemented in the GAUSSIAN package of programs (G03.E01) [34]. Full geometry optimisations (GJF files for the eq and ax conformers of 4 are available in the Supplementary material as eq.gjf and ax.gjf) and analytical vibrational frequency calculations were performed on all the orientations with respect to the carbonyl group resultant from a systematic conformational search using the ab initio Hartree–Fock method with the 3–21G basis set [35,36,37,38,39], allowing complete relaxation of all internal parameters. The obtained conformations were refined using B3LYP and MP2 methods with the 6-31+G(d,p) basis set [35,36,37,38,39]. Frequency analyses were carried out to verify the nature of the minimum state of all the stationary points obtained. To estimate the solvation effects on the relative stability of the most relevant conformers, a full optimisation and frequency calculations were done starting from the gas-phase optimised structures using the PCM model [40] at the B3LYP/6-31+G(d,p) level as implemented in the GAUSSIAN 03. Additionally, the solvent effect was estimated by PCM single-point calculations MP2/6-31+G(d,p) [41] using the corresponding gas-phase optimised structures. The NBO 3.1 program [19] was used as implemented in the GAUSSIAN 03 package, and the reported NBO delocalisation energies (E2) are those given by second-order perturbation theory. The partial atomic charges were calculated using a grid based method (ChElPG) [42].

4. Conclusions

The preferred conformations of some 3-(4′-substituted phenylsulfanyl)-1-methyl-2-piperidones 1–6 bearing as substituents NO₂ (compound 1), Br (compound 2), Cl (compound 3), H (compound 4) Me (compound 5) and OMe (compound 6) were determined by ν_CO IR analysis, B3LYP/6-31+G(d,p) and PCM calculations along with NBO analysis (for 4) and X-ray diffraction (for 2). Theoretical data indicated the existence of two stable conformations. The 2-piperidone ring of the title compounds assumes a slightly distorted half-chair geometry, while the phenylsulfanyl substituent adopts the almost axial (ax) and equatorial (eq) geometries with respect to the carbonyl group. The ax conformer is the most stable (ca. 97%), the least polar, and exhibits a lower ν_CO frequency.

Unusually, the higher ν_CO frequency component can be ascribed to the ax conformer in solution of non-polar (C₆H₁₄ and CCl₄) and polar solvents (CHCl₃ and CH₂Cl₂). As far as the relative abundance is concerned, the ax conformer component progressively decreases with respect to the eq one as the relative permittivity increases, and is no longer detectable in the most polar solvent CH₃CN.

The PCM method was applied to derivative 4 to fully optimise the ax and eq conformations in all solvents. The torsional angles α-ω’ of the ax and eq conformers remain practically the same going from the gas phase to solvents of increasing relative permittivity.

The progressive decrease of the ax/eq population ratio as the solvent polarity increases is in line with the experimental IR trend. Moreover, the computed PCM ν_CO frequencies of the two conformers do not significantly differ in the non-polar solvents. In contrast, the ν_CO frequency of the eq conformer becomes progressively smaller than that of the ax one as the solvent polarity increases. These results further support the experimental IR frequency assignments. The lower ν_CO IR frequency of the eq conformer in solution may be rationalised by the analysis of the computed geometries, which indicate that in the eq conformer the carbonyl oxygen atom is free for solvation, while in the ax conformer the presence of the [O_CO^…H_o-Ph] hydrogen bond partially prevents the solvent molecules' approach to the carbonyl oxygen atom. Thus, the larger solvation at the carbonyl oxygen atom of the most polar eq conformer decreases the carbonyl bond order and consequently its frequency to a larger extent than in the ax conformer.

The sum of the energy contributions of the orbital interactions (NBO analysis) and the electrostatic interactions correlates well with the populations and the ν_CO frequencies of both conformers in the gas phase. The smaller contribution of the LP_N→π*_CO orbital interaction for the eq conformer with respect to the ax conformer is in line with the smaller contribution of the [⁻O‒C=N⁺] canonical form to the resonance hybrid due to the strong Repulsive Field Effect between the C^δ+=O^δ− and C^δ+-S^δ− dipoles, which destabilises the eq conformer and increases its ν_CO frequency relative to that of the ax conformer.

X-ray single crystal analysis of 2 indicates that this compound assumes, in the solid, the most polar eq geometry with respect to the [O=C-C-SPh] moiety and that the 2-piperidone ring displays a slightly distorted half-chair conformation. In order to obtain the largest energy gain from the crystal packing, the molecules of the eq conformer of 2 are arranged in the solid through a helical fashion due to hydrogen bond interactions, i.e., C(3)-H(5)^…O(1) and C(7)-H(14)^…π_(Ph).