Hydrogen-bonding patterns in two aroylthio­carbamates and two aroylimidothio­carbonates

Abstract

In O-ethyl N-benzoyl­thio­carbamate, C₁₀H₁₁NO₂S, the mol­ecules are linked into sheets by a combination of two-centre N—H⋯O and C—H⋯S hydrogen bonds and a three-centre C—H⋯(O,S) hydrogen bond. A combination of two-centre N—H⋯O and C—H⋯O hydrogen bonds links the mol­ecules of O-ethyl N-(4-methyl­benzoyl)thio­carbamate, C₁₁H₁₃NO₂S, into chains of rings, which are linked into sheets by an aromatic π–π stacking inter­action. In O,S-diethyl N-(4-methyl­benzoyl)imidothio­carbonate, C₁₃H₁₇NO₂S, pairs of mol­ecules are linked into centrosymmetric dimers by pairs of symmetry-related C—H⋯π(arene) hydrogen bonds, while the mol­ecules of O,S-diethyl N-(4-chloro­benzoyl)imidothio­carbonate, C₁₂H₁₄ClNO₂S, are linked by a single C—H⋯O hydrogen bond into simple chains, pairs of which are linked by an aromatic π–π stacking inter­action to form a ladder-type structure.

Comment

O,S-Dialkyl aroylimido­thio­carbonates are structural ana­logues of S,S-dialkyl aroylimido­thio­carbonates, which have been widely used as synthetic starting materials (Augustín et al., 1980 ▶; Sato et al., 1981 ▶; Fukada et al., 1985 ▶, 1986 ▶, 1990 ▶; Insuasty et al., 2006 ▶, 2008 ▶). We report here the structures of two O-ethyl aroylthio­carbamates, (I) and (II) (Figs. 1 ▶ a and 1b), and two O,S-diethyl aroylimidothio­carbonates, (III) and (IV) (Figs. 1 ▶ c and 1d), and we compare these structures with those of two closely related S-ethyl aroyldithio­carbamates, (V) and (VI) (see scheme) (Low et al., 2004 ▶, 2005 ▶). The structure of (I) has been briefly reported previously [Cambridge Strutural Database (Allen, 2002 ▶) refcode GIFSUK; Arslan et al., 2007 ▶)], but the authors’ primary concern was the comparison of the experimental geometry and vibrational frequencies with those calculated from first principles at various levels of theory. The hydrogen bonding was described extremely briefly in terms of only an N—H⋯O inter­action. Accordingly, we have thought it worthwhile to report here a more complete description of the hydrogen bonding in (I).

Figure 1.

The mol­ecular structures of (a) (I), (b) (II), (c) (III) and (d) (IV), showing the atom-labelling schemes. Displacement ellipsoids are drawn at the 30% probability level and H atoms are shown as small spheres of arbitrary radii.

Compounds (I) and (II) were prepared by the addition of ethanol to the corresponding inter­mediate aroylisothio­cyanate, (B) (see scheme), itself prepared by reaction of potassium thio­cyanate with the aroyl chloride, (A). Deproton­ation of the O-ethyl aroylthio­carbamates yields the ionic inter­mediate, (C), reaction of which with bromo­ethane gives (III) and (IV). The corresponding reactions of the appropriate aroylisothio­cyanate, (B), with ethanethiol rather than with ethanol produces the dithio­carbamate esters (V) and (VI) (Low et al., 2004 ▶, 2005 ▶). It is inter­esting to note that in the conversion of (A) to (B), the thio­cyanate anion reacts with the aroyl chloride exclusively via the harder N terminus, while in the formation of (III) and (IV) from (C), this anionic component of (C) reacts with bromo­ethane exclusively at the softer S centre.

Despite the close pairwise similarities between the mol­ecular constitutions within the pairs of compounds (I) and (II), (III) and (IV), and (V) and (VI), and the overall similarities within the group comprised of (I), (II), (V) and (VI), no two of these compounds are isomorphous. Thus, for example, while compounds (I) and (II) crystallize in the space groups Pna2₁ and P2₁/c, respectively, both with Z′ = 1, their close analogues (V) and (VI) crystallize in, respectively, C2/c with Z′ = 2 and P2₁/c with Z′ = 1. The unit-cell dimensions of (II) and (VI), which could well have been isomorphous and isostructural are, in fact, significantly different, particularly for the cell repeat distance a and the cell angle β. Also, (III) and (IV), which might plausibly have been expected to be isomorphous, crystallize with triclinic unit cells in which the real and reduced cell angles are very different: the cell angles in (III) are all significantly greater than 90°, while those in (IV) are all well below 90°. Hence the unit cell for (III) can be assigned as belonging to type II, designation 4R (Buerger, 1956 ▶), while that of (IV) can be assigned as type I, designation 1R.

In thio­carbamate esters (I) and (II), the mol­ecular skeletons adopt chain-extended conformations which, apart from the aryl substituents, are close to planarity, as shown by the leading torsion angles (Table 1 ▶). The most striking feature of the conformations of thio­carbonates (III) and (IV) is the orientation of the S-ethyl groups (Figs. 1 ▶ c and 1d). Whereas in (III) this ethyl group is almost coplanar with the rest of the skeleton, in (IV) this unit is almost orthogonal to the rest of the mol­ecule (Table 1 ▶). Also, whereas in thio­carbamate esters (I) and (II) the formal C=O and C=S double bonds are mutually cisoid, in dithio­carbamate esters (V) and (VI) (Low et al., 2004 ▶, 2005 ▶) these units adopt a transoid arrangement (cf. scheme).

Table 1. Selected geometric parameters (Å, °) for (I)–(IV) .

	(I)	(II)	(III)	(IV)
C1—O1	1.206 (5)	1.208 (3)	1.210 (4)	1.220 (3)
C1—N2	1.385 (5)	1.370 (3)	1.379 (4)	1.382 (3)
N2—C3	1.384 (6)	1.377 (3)	1.283 (4)	1.281 (3)
C3—S3	1.615 (4)	1.615 (3)
C3—O4	1.325 (5)	1.324 (3)
C3—S31			1.730 (4)	1.733 (3)
C3—O41			1.321 (4)	1.326 (3)
O1—C1—N2	122.4 (4)	122.3 (3)	125.7 (3)	125.1 (3)
C1—N2—C3	127.3 (4)	127.1 (2)	120.0 (3)	119.2 (2)
N2—C3—S3	127.4 (3)	128.5 (2)
N2—C3—S31			128.3 (3)	127.9 (2)
C12—C11—C1—N2	19.3 (6)	−21.4 (4)	8.3 (5)	8.9 (4)
C11—C1—N2—C3	−162.4 (4)	179.0 (2)	177.9 (3)	176.1 (2)
C1—N2—C3—O4	−179.9 (4)	−171.7 (2)
N2—C3—O4—C5	−174.4 (4)	−177.8 (2)
C3—O4—C5—C6	−171.0 (4)	−171.3 (2)
C1—N2—C3—S31			1.3 (5)	0.7 (4)
C1—N2—C3—O41			−178.8 (3)	−179.6 (2)
N2—C3—C31—C32			178.1 (3)	−172.7 (3)
C3—S31—C32—C33			−179.4 (3)	78.7 (2)
N2—C3—O41—C42			−1.3 (5)	3.2 (4)
C3—O41—C42—C43			−178.4 (3)	−172.5 (2)

In each of compounds (I)–(IV), the C—O bond distances involving atom O3, namely C3—O4 in (I) and (II), and C3—O41 in (III) and (IV), are all the same within experimental uncertainty. In addition, the N2—C3 bonds in (I) and (II) are long for their type (Allen et al., 1997a ▶). These observations, taken together, indicate that in (I) and (II) there is very little electronic delocalization from atoms N2 and O4 on to atom S3. By contrast, the C=S bond distances are 1.6586 (18) and 1.6552 (18) Å in (V), where Z′ = 2 (Low et al., 2004 ▶), and 1.631 (3) Å in (VI) (Low et al., 2005 ▶), typical of such distances in compounds containing >N—C(=S)—S– fragments (Allen et al., 1997a ▶) and consistent with the occurrence of conjugative delocalization. In (III) and (IV), the C—N distances provide a clear distinction between the formal C1—N2 single bonds and the formal N2=C3 double bonds. Despite the apparent lack of any significant polarization of the electronic structures in (I) and (II), the bond angles at atoms C1, N2 and C3 are certainly consistent with a strongly repulsive nonbonded inter­action between atoms O1 and S3. A similar pattern of bond angles is apparent in (III) and (IV).

The supra­molecular aggregation in (I) and (II) is dominated by a combination of N—H⋯O and C—H⋯O hydrogen bonds, utilizing the same acceptor O1 atom in each compound (Table 2 ▶). These inter­actions are augmented, in the case of (I) only, by two C—H⋯S inter­actions. In (I), mol­ecules related by the a-glide plane at y = are linked by a combination of a two-centre N—H⋯O hydrogen bond and a weaker three-centre C—H⋯(O,S) hydrogen bond to form a ribbon containing alternating (6) (Bernstein et al., 1995 ▶) and (7) rings and running parallel to the [100] direction (Fig. 2 ▶). A second C—H⋯S inter­action links mol­ecules related by the n-glide plane at x = into a simple C(7) chain running parallel to the [01] direction (Fig. 3 ▶). The combination of the ribbon along [100] and the chain along [01] generates a hydrogen-bonded sheet parallel to (011), but there are no direction-specific inter­actions between adjacent sheets. In particular, both C—H⋯π(arene) hydrogen bonds and aromatic π–π stacking inter­actions are absent from the crystal structure of (I).

Table 2. Hydrogen bonds and short inter­molecular contacts (Å, °) for (I)–(IV) .

Cg1 is the centroid of the C11–C16 ring.

Compound	D—H⋯A	D—H	H⋯A	D⋯A	D—H⋯A
(I)	N2—H2⋯O1i	0.88	2.05	2.921 (4)	170
	C12—H12⋯S3i	0.95	2.83	3.606 (4)	139
	C12—H12⋯O1i	0.95	2.51	3.192 (5)	129
	C16—H16⋯S3ii	0.95	2.86	3.576 (5)	134
(II)	N2—H2⋯O1iii	0.88	2.03	2.896 (3)	167
	C12—H12⋯O1iii	0.95	2.32	3.075 (3)	136
(III)	C32—H32B⋯Cg1iv	0.99	2.62	3.554 (5)	158
(IV)	C13—H13⋯O1v	0.95	2.45	3.345 (4)	156

Symmetry codes: (i) − + x,  − y, z; (ii)  − x,  + y, − + z; (iii) x,  − y, − + z; (iv) 1 − x, 1 − y, 1 − z; (v) −1 + x, y, z.

Figure 2.

Part of the crystal structure of (I), showing the formation of a ribbon along [100] containing (6) and (7) rings. For the sake of clarity, H atoms not involved in the motifs shown have been omitted. Atoms marked with an asterisk (*), a hash symbol (#), a dollar sign ($) or an ampersand (&) are at the symmetry positions (− + x,  − y, z), (−1 + x, y, z), ( + x,  − y, z) and (1 + x, y, z), respectively.

Figure 3.

Part of the crystal structure of (I), showing the formation of a C(7) chain along [01]. For the sake of clarity, H atoms not involved in the motifs shown have been omitted. Atoms marked with an asterisk (*), a hash symbol (#) or a dollar sign ($) are at the symmetry positions ( − x,  + y, − + z), ( − x, −  + y,  + z) and (x, −1 + y, 1 + z), respectively.

In (II), a combination of N—H⋯O and C—H⋯O hydrogen bonds, similar to the corresponding combination in (I), links mol­ecules related by the c-glide plane at y = into a chain of (7) rings running parallel to the [001] direction (Fig. 4 ▶). The closest inter­molecular contacts between atom S3 and the H atoms in (II) corresponding to the C—H⋯S inter­actions in (I) all have H⋯S distances well in excess of 3 Å, and hence they cannot be regarded as structurally significant. Although significant C—H⋯S inter­actions are absent from the crystal structure of (II), the hydrogen-bonded chains are linked by an aromatic π–π stacking inter­action. The aryl rings in the mol­ecules at (x, y, z) and (1 − x, 1 − y, 1 − z) are parallel by symmetry, with an inter­planar spacing of 3.479 (2) Å. The corresponding ring-centroid separation is 3.746 (2) Å and the ring-centroid offset (slippage) is 1.389 (2) Å. The effect of this stacking inter­action is to link the hydrogen-bonded chains into a sheet parallel to (100) (Fig. 5 ▶).

Figure 4.

Part of the crystal structure of (II), showing the formation of a chain of (7) rings along [001]. For the sake of clarity, H atoms not involved in the motifs shown have been omitted. Atoms marked with an asterisk (*), a hash symbol (#), a dollar sign ($) or an ampersand (&) are at the symmetry positions (x,  − y, − + z), (x, y, −1 + z), (x,  − y,  + z) and (x, y, 1 + z), respectively.

Figure 5.

A stereoview of part of the crystal structure of (II), showing the π-stacking of hydrogen-bonded chains along [001] to form a sheet parallel to (100). For the sake of clarity, H atoms not involved in the motifs shown have been omitted.

Since no N—H bonds are present in (III) and (IV), the modes of supra­molecular aggregation in these compounds necessarily differ substantially from those observed in (I) and (II). In (III), a single C—H⋯π(arene) hydrogen bond links pairs of mol­ecules into centrosymmetric dimers (Fig. 6 ▶). However, π–π stacking inter­actions are absent and there are no direction-specific inter­actions between the dimeric units. By contrast, the crystal structure of (IV) contains no C—H⋯π(arene) hydrogen bonds, but instead chains built from C—H⋯O hydrogen bonds are linked in pairs to form a ladder-type structure, in which mol­ecules related by translation form C(6) chains running parallel to the [100] direction. The aryl rings of the mol­ecules at (x, y, z) and (1 − x, 1 − y, 1 − z) are strictly parallel, with an inter­planar spacing of 3.428 (2) Å. The ring-centroid separation and offset are 3.787 (2) and 1.609 (2) Å, respectively, so that pairs of anti­parallel C(6) chains are weakly linked (Fig. 7 ▶).

Figure 6.

Part of the crystal structure of (III), showing the formation of a centrosymmetric dimer by means of symmetry-related C—H⋯π(arene) hydrogen bonds. For the sake of clarity, H atoms bonded to C atoms which are not involved in the motif shown have been omitted. The atom marked with an asterisk (*) is at the symmetry position (1 − x, 1 − y, 1 − z).

Figure 7.

A stereoview of part of the crystal structure of (IV), showing a π-stacked pair of anti­parallel hydrogen-bonded chains along [100]. For the sake of clarity, H atoms not involved in the motif shown have been omitted.

It is of inter­est briefly to compare the supra­molecular aggregation in (V) and (VI) (Low et al., 2004 ▶, 2005 ▶) with that in (I)–(IV). Compound (V) crystallizes with Z′ = 2 in space group C2/c (Low et al., 2004 ▶) and each type of mol­ecule independently forms a cyclic (8) dimer. These dimers are built using two symmetry-related N—H⋯S hydrogen bonds, with the thione-type S atoms as the acceptors in both types of dimer. One type of dimer contains mol­ecules related by inversion and the other contains mol­ecules related by a twofold rotation axis, and the two independent types of dimer are linked into chains by a single C—H⋯π(arene) hydrogen bond. It is striking that there is no participation by the amidic O atom in the hydrogen bonding in (V). In compound (VI), on the other hand, where Z′ = 1 (Low et al., 2005 ▶), a combination of N—H⋯O and C—H⋯O hydrogen bonds generates a chain of (7) rings along [001], and chains of this type are linked into a sheet by a single aromatic π–π stacking inter­action. Thus, while (II) and (VI) are not isomorphous, and while their mol­ecules adopt different conformations, nonetheless their crystal structures exhibit very similar patterns of inter­molecular inter­action.

The only hydrogen bonds in this series of compounds, viz. (I)–(VI), which involve S atoms as the acceptors utilize the thione-type S atoms in (I) and (V); the two-coordinate S atoms in compounds (III)–(VI) do not participate in any hydrogen-bond formation. This observation is certainly consistent with the deductions drawn from database analyses (Allen et al., 1997a ▶,b ▶) that, while thione-type S atoms can in some circumstances act as effective hydrogen-bond acceptors, two-coordinate S atoms are, in general, very poor acceptors.

Experimental

For the synthesis of (I) and (II), the appropriate aroyl chloride (0.043 mol) was added to a solution of potassium thio­cyanate (0.043 mol) in acetonitrile (75 ml). This mixture was heated under reflux for 15 min to afford the corresponding aroyl isothio­cyanate, which was not isolated. After cooling the inter­mediate solution to 273 K under an inert atmosphere, dry ethanol (0.47 mol) was added, and this mixture was then stirred at ambient temperature for 24 h. Ice–water was added to the reaction mixture and the resulting light-green solid was collected by filtration, washed with water, dried under reduced pressure and finally crystallized by slow evaporation, at ambient temperature and in air, of a solution in n-hexane, to give crystals suitable for single-crystal X-ray diffraction. For (I): yield 95%, m.p. 345 K; for (II): yield 92%, m.p. 331 K.

For the synthesis of (III) and (IV), a slight excess of sodium hydride (60% suspension in oil, 0.020 mol) was added under an inert atmosphere to an ice-cold solution of the corresponding O-ethyl aroylimido­thio­carbonate (0.010 mol) in N,N-dimethyl­formamide (6 ml). This mixture was stirred for 45 min at ambient temperature, and then bromo­ethane (0.012 mol) was added slowly and the stirring was continued for a further 30 min. Ice–water was added to the reaction mixture and the resulting colourless solid was collected by filtration, washed with water, dried under reduced pressure and finally crystallized by slow evaporation, at ambient temperature and in air, of a solution in dry ethanol, to give crystals suitable for single-crystal X-ray diffraction. For (III): yield 94%, m.p. 333 K; for (IV): yield 95%, m.p. 373 K.

Compound (I)

Crystal data

• C₁₀H₁₁NO₂S

• M _r = 209.27

• Orthorhombic,

• a = 9.9418 (8) Å

• b = 9.3619 (5) Å

• c = 10.9337 (13) Å

• V = 1017.65 (16) ų

• Z = 4

• Mo Kα radiation

• μ = 0.29 mm⁻¹

• T = 120 K

• 0.50 × 0.42 × 0.41 mm

Data collection

• Bruker–Nonius KappaCCD area-detector diffractometer

• Absorption correction: multi-scan (SADABS; Sheldrick, 2003 ▶) T _min = 0.868, T _max = 0.890

• 13852 measured reflections

• 1889 independent reflections

• 1284 reflections with I > 2σ(I)

• R _int = 0.074

Refinement

• R[F ² > 2σ(F ²)] = 0.048

• wR(F ²) = 0.130

• S = 1.14

• 1889 reflections

• 128 parameters

• 1 restraint

• H-atom parameters constrained

• Δρ_max = 0.45 e Å⁻³

• Δρ_min = −0.29 e Å⁻³

• Absolute structure: Flack (1983 ▶), with 890 Bijvoet pairs

• Flack parameter: 0.08 (17)

Compound (II)

Crystal data

• C₁₁H₁₃NO₂S

• M _r = 223.29

• Monoclinic,

• a = 12.2845 (14) Å

• b = 9.1431 (17) Å

• c = 9.7897 (4) Å

• β = 90.182 (7)°

• V = 1099.6 (2) ų

• Z = 4

• Mo Kα radiation

• μ = 0.27 mm⁻¹

• T = 120 K

• 0.34 × 0.30 × 0.24 mm

Data collection

• Bruker–Nonius KappaCCD area-detector diffractometer

• Absorption correction: multi-scan (SADABS; Sheldrick, 2003 ▶) T _min = 0.918, T _max = 0.937

• 11469 measured reflections

• 2026 independent reflections

• 1334 reflections with I > 2σ(I)

• R _int = 0.051

Refinement

• R[F ² > 2σ(F ²)] = 0.049

• wR(F ²) = 0.134

• S = 1.07

• 2026 reflections

• 138 parameters

• H-atom parameters constrained

• Δρ_max = 0.31 e Å⁻³

• Δρ_min = −0.29 e Å⁻³

Compound (III)

Crystal data

• C₁₃H₁₇NO₂S

• M _r = 251.35

• Triclinic,

• a = 7.2470 (16) Å

• b = 8.8870 (7) Å

• c = 10.737 (1) Å

• α = 100.691 (7)°

• β = 99.984 (13)°

• γ = 107.307 (11)°

• V = 629.19 (17) ų

• Z = 2

• Mo Kα radiation

• μ = 0.25 mm⁻¹

• T = 120 K

• 0.42 × 0.33 × 0.26 mm

Data collection

• Bruker–Nonius KappaCCD area-detector diffractometer

• Absorption correction: multi-scan (SADABS; Sheldrick, 2003 ▶) T _min = 0.889, T _max = 0.938

• 2460 independent reflections

• 2064 reflections with I > 2σ(I)

Refinement

• R[F ² > 2σ(F ²)] = 0.060

• wR(F ²) = 0.167

• S = 1.18

• 2460 reflections

• 158 parameters

• H-atom parameters constrained

• Δρ_max = 0.71 e Å⁻³

• Δρ_min = −0.46 e Å⁻³

Compound (IV)

Crystal data

• C₁₂H₁₄ClNO₂S

• M _r = 271.75

• Triclinic,

• a = 7.8028 (14) Å

• b = 9.480 (2) Å

• c = 9.860 (2) Å

• α = 69.545 (13)°

• β = 75.683 (14)°

• γ = 69.548 (15)°

• V = 633.7 (2) ų

• Z = 2

• Mo Kα radiation

• μ = 0.46 mm⁻¹

• T = 120 K

• 0.52 × 0.45 × 0.29 mm

Data collection

• Bruker–Nonius KappaCCD area-detector diffractometer

• Absorption correction: multi-scan (SADABS; Sheldrick, 2003 ▶) T _min = 0.778, T _max = 0.879

• 14986 measured reflections

• 2363 independent reflections

• 1707 reflections with I > 2σ(I)

• R _int = 0.051

Refinement

• R[F ² > 2σ(F ²)] = 0.045

• wR(F ²) = 0.119

• S = 1.12

• 2363 reflections

• 156 parameters

• H-atom parameters constrained

• Δρ_max = 0.49 e Å⁻³

• Δρ_min = −0.34 e Å⁻³

All H atoms were located in difference maps and then treated as riding atoms in geometrically idealized positions, with C—H = 0.95 (aromatic), 0.98 (CH₃) or 0.99 Å (CH₂) and N—H = 0.88 Å, and with U _iso(H) = kU _eq(carrier), where k = 1.5 for the methyl groups, which were permitted to rotate but not to tilt, and 1.2 for all other H atoms. For (I), the correct orientation of the structure with respect to the polar-axis direction was established by means of the Flack x parameter (Flack, 1983 ▶), x = 0.08 (17), and the Hooft y parameter (Hooft et al., 2008 ▶), y = 0.03 (5), for 99.6% coverage of the Bijvoet pairs. Compound (III) was handled as a nonmerohedral twin, in which the two twin components are related by the matrix (1.000, 0.557, 0.320/0.000, −1.000, 0.000/0.000, 0.000, −1.000). Using the original reflection file (15150 measured reflections, R _int = 0.0581), a modified file (2460 reflections, R _int = 0.0000) was prepared using the TwinRotMat option in PLATON (Spek, 2009 ▶) and then used in conjunction with the HKLF 5 option in SHELXL97 (Sheldrick, 2008 ▶), giving twin fractions of 0.268 (5) and 0.732 (5).

For all compounds, data collection: COLLECT (Nonius, 1999 ▶); cell refinement: DIRAX/LSQ (Duisenberg et al., 2000 ▶); data reduction: EVALCCD (Duisenberg et al., 2003 ▶). Program(s) used to solve structure: SHELXS97 (Sheldrick, 2008 ▶) for (I); SIR2004 (Burla et al., 2005 ▶) for (II), (III) and (IV). For all compounds, program(s) used to refine structure: SHELXL97 (Sheldrick, 2008 ▶); molecular graphics: PLATON (Spek, 2009 ▶); software used to prepare material for publication: SHELXL97 and PLATON.