([eta]⁵-Cyclo­penta­dienyl)([eta]⁶-phenoxathiin 10,10-dioxide)iron(II) hexa­fluoridophosphate and phenoxathiin 10,10-di­oxide

In the structure of the title complex salt, [Fe(C₅H₅)(C₁₂H₈O₃S)]PF₆, the coordinated cyclo­penta­dienyl (Cp) and benzene ring planes are almost parallel, with a hinge angle between the planes of 0.8 (2)°. The hinge angle between the planes of the peripheral (coordinated and uncoordinated) benzene rings in the coordinated phenoxathiin 10,10-dioxide mol­ecule is 169.9 (2)°, and the FeCp moiety is located inside the shallow fold of the heterocycle. The hinge angle between the benzene ring planes in the free heterocycle, C₁₂H₈O₃S, is 171.49 (6)°.

S1. Comment

The title complex, (I), was obtained in an extension of the work on the synthesis of polycyclic heteroaromatic systems in the double nucleophilic aromatic substitution reaction using o-dichlorobenzene FeCp and related complexes (Sutherland et al., 1982, 1988), followed by modification of the heterocycle structure by oxidation (see Lee et al., 1986, and earlier work by the same authors). This study continues our observations of changes that result in the structure of tricyclic heterocycles containing O and/or S atoms in the central ring of the system upon FeCp complexation and the introduction of substituents into a heterocycle structure. The free heterocycle, phenoxathiin 10,10-dioxide, (II), was obtained in an oxidation of phenoxathiin with hydrogen peroxide in glacial acetic acid solution, as described by Gilman & Esmay (1952).

The unit cell of (I) contains four cationic molecules of the (phenoxathiin 10,10-dioxide) FeCp complex along with four hexafluoridophosphate counteranions. The FeCp moiety is located inside the shallow heterocycle fold in all molecules. The complexed Cp and benzene rings are nearly coplanar, with a dihedral angle of 0.8 (2)°, which is in agreement with observations for both phenoxathiin and related azaphenoxathiin FeCp complexes (Lynch et al., 1986; Sutherland et al., 1988). The centroid of the Cp ring, the Fe centre and the centroid of the benzene ring are nearly collinear with an angle 179.35 (11)° measured between the two vectors extending from the Fe atom to the centroids in the complexed rings, and this value is typical for FeCp arene complexes [see, for example, Manzur et al. (2000) and Fuentealba et al. (2007)].

The distances from the Fe^II ion to the Cp plane and to the complexed benzene ring plane are 1.645 (2) and 1.540 (2) Å, respectively. These values are close to those reported in the literature for similar FeCp complexes [see, for example, Lynch et al. (1986), Abboud et al. (1990), Fuentealba et al. (2007), Manzur et al. (2007) and Hendsbee et al. (2010)]. The distances from Fe to the C atoms of the complexed benzene ring are within the range 2.048 (4)–2.113 (4) Å, with a mean of 2.082 Å. The shortest Fe—C distance is found for a quaternary C atom bonding to an –SO₂– group, while the distance to another quaternary C atom, bonding to an O atom, appears to be the longest. The distances are within the range of reported values for FeCp complexes (Abboud et al., 1990; Piórko et al., 1994; Fuentealba et al., 2007; Manzur et al., 2007; Jenkins et al., 2008). However, they differ from the data for both phenoxathiin and azaphenoxathiin complexes, in which both the Fe—C(quaternary) distances are the longest of all six distances (Lynch et al., 1986; Sutherland et al., 1988).

The C—C distances in the complexed ring of the phenoxathiin 10,10-dioxide appear to be slightly longer than those in the uncomplexed ring, with the average distances being 1.401 and 1.387 Å, respectively. The C—S bonds extending from both the complexed and uncomplexed ring C atoms to the bridging S atom are similar in length [1.757 (4) and 1.743 (4) Å, respectively]. The C—O distances, however, are quite different, as the bond extending from the bridging O atom to the C atom of the complexed ring is significantly shorter than the C—O bond extending towards the uncomplexed ring [1.361 (5) and 1.393 (5) Å, respectively]. Both these observations agree with earlier findings for phenoxathiin and azaphenoxathiin complexes (Lynch et al., 1986; Sutherland et al., 1988) and for dibenzodioxin complexes (Piórko et al., 1994, 1995; Hendsbee et al., 2010).

The C—C bonds in the rings of the free heterocycle, phenoxathiin 10,10-dioxide, (II), have similar average lengths (1.385 and 1.387 Å) and are similar in length to the C—C bonds in the uncomplexed ring of the heterocycle FeCp complex. The S—C bond lengths from the bridging S atom to the benzene ring C atoms in the uncomplexed heterocycle are similar [1.7471 (16) and 1.7481 (18) Å] to those found in the complex. The C—O distances in the free heterocycle are 1.369 (2) and 1.371 (2) Å, more similar to the length of the C—O bond extending towards the complexed ring of the FeCp complex [1.361 (5) Å] rather than that extending towards the uncomplexed ring of the FeCp complex [1.393 (5) Å].

A double nucleophilic aromatic substitution reaction yielding tricyclic heterocycle complexes may result in the formation of FeCp-in-fold, FeCp-out-of-fold, or both isomeric molecules of the nonplanar tricyclic heterocycle in the solid state. Examples of all three cases may be found in the literature, and all reports provide crystallographic data supporting this a statement. The earlier studies of the synthesis and structure of dibenzodioxin and thianthrene FeCp complexes suggested that only FeCp-in-fold molecules are formed in such a reaction. This conclusion was based on the results of several crystallographic studies (Simonsen et al., 1985; Abboud et al., 1990; Christie et al., 1994; Piórko et al., 1994, 1995). Recently, we reported that a double nucleophilic substitution reaction leading to the formation of (1,2,3,4,4a,10a-η)-1-methylthianthrene FeCp hexafluoridophosphate gave rise to a mixture of both in-fold and out-of-fold isomers, as found in a crystallographic study of the reaction products (Hendsbee et al., 2009). The only previously reported out-of-fold thianthrene complex was obtained, along with its in-fold isomer, in a different reaction, a photolytic demetallation of a mixture containing cis and trans di-(η⁵-Cp)(η⁶,η⁶-thianthrene)(iron)₂ bis-hexafluoridophosphates, which were prepared in a ligand-exchange reaction (see Abboud et al., 1990).

Two phenoxathiin complexes which were obtained using a double nucleophilic substitution reaction have been reported in the literature to date. Both (phenoxathiin)FeCpPF₆ (Lynch et al., 1986) and (5a,6,7,8,9,9a-η-1,4-benzoxathiino[3,2-b]pyridine)FeCpPF₆ (Sutherland et al., 1988) contain, in the solid state, only out-of-fold FeCp moieties. In both complexes, the FeCp moiety is located outside the shallow heterocycle fold, with dihedral angles of 178.7 (1) and 176.8 (1)° for the phenoxathiin and azaphenoxathiin complexes, respectively. The dihedral angle for the free phenoxathiin molecule was reported as 138° (Hosoya, 1966) and as 142.3° (Fitzgerald et al., 1991; 223 K). This angle has yet to be reported for the uncomplexed azaphenoxathiin molecule. It appears then that FeCp complexation flattens the phenoxathiin skeleton. In this study it was found that the FeCp moiety is located inside the phenoxathiin 10,10-dioxide fold, with a dihedral angle of 169.9 (2)° between the two external benzene rings. Thus, oxidation of the S atom in the central ring appears to counteract the effect of FeCp complexation, causing more pronounced folding of the heterocycle molecule, and apparently converting the FeCp-out-of-fold isomer into the FeCp-in-fold one. For free phenoxathiin 10,10-dioxide we found the dihedral angle to be 171.49 (6)°, which means that when this molecule is complexed to FeCp it is slightly more folded. This is the first firmly confirmed example, and only the second case, in which FeCp complexation appears to increase folding of the tricyclic heterocycle molecule. In the earlier case this effect was observed in the structure of a methylthianthrene molecule carrying a methyl group in an uncomplexed benzene ring. The structure of the free heterocycle, 2-methylthianthrene, has not yet been reported, so the folding angle of a parent thianthrene molecule was used for comparison (Simonsen et al., 1985). A similar effect was reported for a structurally related thioxanthene molecule, with a methylene group replacing the O atom in the central ring. Literature reports indicate that oxidation of the S atom to a dioxide results in slightly more pronounced folding of the molecule. For thioxanthene, the dihedral angle was reported as 135.3 (1)° (Gillean et al., 1973), while for thioxanthene 10,10-dioxide this angle was 133.9° (Chu & Chung, 1974).

We suggest that both the increased folding and the location of the FeCp moiety inside the fold may be requirements for minimizing the interaction of S-bonded O atoms with Fe in a complex. In this apparently favoured in-fold isomer, the distance from Fe to the proximal O atom will be longer than the analogous distance in the out-of-fold molecule. With a relatively small dihedral angle in the starting phenoxathiin molecule, a thermal flipping of this molecule during oxidation, which results in inversion of the FeCp-out-of-fold isomer to the FeCp-in-fold isomer, appears to be possible and favoured at an elevated reaction temperature. However, this inversion process may be difficult to observe experimentally.

Related literature

For related literature, see: Abboud et al. (1990); Abd-El-Aziz & Epp (1995); Abd-El-Aziz, Boraie, Al-Salem, Sadek & Epp (1997); Christie et al. (1994); Chu & Chung (1974); Fitzgerald et al. (1991); Fuentealba et al. (2007); Gillean et al. (1973); Gilman & Esmay (1952); Hendsbee et al. (2009, 2010); Hosoya (1966); Jenkins et al. (2008); Lee et al. (1982); Lee, Abd El Aziz, Chowdhury, Piórko & Sutherland (1986); Lee, Chowdhury, Piórko & Sutherland (1986); Lynch et al. (1986); Manzur et al. (2000, 2007); Piórko et al. (1994, 1995); Simonsen et al. (1985); Sutherland et al. (1982, 1988).

S2. Experimental

The precursor (phenoxathiin)FeCp hexafluoridophosphate complex was obtained in the double nucleophilic aromatic substitution reaction of the o-dichlorobenzene FeCp complex with 2-mercaptophenol, as described in the literature (Sutherland et al., 1982). The title complex salt was obtained in an oxidation of the precursor phenoxathiin complex using hydrogen peroxide in trifluoroacetic acid. This method, while used for oxidation of the amino group to a nitro group in similar complexes (see Lee et al., 1982, and later work by the same authors) and, more recently, for simultaneous oxidation of both amino and alkyl groups to nitro and carboxy groups, respectively (Abd-El-Aziz et al., 1997; Abd-El-Aziz & Epp, 1995), has not been previously reported in the oxidation of sulfur-containing FeCp complexes, although it has been generally used in the oxidation of sulfur-containing heterocycles using glacial acetic acid as a solvent [see, for example, oxidation of dibenzothiophene and phenoxathiin to the corresponding dioxides by Gilman & Esmay (1952)]. The reaction gave 68% yield and the crystal suitable for X-ray examination was obtained from the acetone–diethyl ether–dichloromethane solution at 280 (2) K. The same complex was also obtained in 63% yield in an alternative oxidation of the phenoxathiin complex with m-chloroperbenzoic acid, according to the procedure of Lee, Abd El Aziz et al. (1986). Phenoxathiin 10,10-dioxide, (II), was obtained by oxidation of the precursor phenoxathiin using 30% hydrogen peroxide in glacial acetic acid (Gilman & Esmay, 1952) with a total yield of 98%. The crystal suitable for X-ray examination was obtained upon cooling the reaction mixture to room temperature. Experimental details and analytical data for both the complex (I) and free heterocycle (II) are provided in the Supplementary materials.

S3. Refinement

H atoms were placed in geometrically idealized positions, with C—H = 0.95 [all uncoordinated and coordinated aromatic C atoms in (I), and all C atoms in free heterocycle (II)] or 1.00 Å [cyclopentadienyl C atoms in (I)], and constrained to ride on their parent C atoms, with U_iso(H) = 1.2U_eq(C). For the counter-anion PF₆^- group in (I), the F atoms were restrained to have the same U^ij components within a standard uncertainty of 0.05 Ų. The complexed heterocycle structure contains a cyclopentadienyl (Cp) ring which is π-bonded to an Fe atom, and the thermal motion of this Cp ring resulted in unsatisfactory anisotropic displacement parameters for atoms C11, C12, C13 and C15. This was resolved through the use of the restraints applied to the refinement of these atoms: the U^ij components of these atoms were restrained to be equal within 0.004 Ų, and their anisotropic displacement parameters were restrained to be equal within 0.002 Ų.

Computing details

For both compounds, data collection: APEX2 (Bruker, 2008); cell refinement: SAINT (Bruker, 2008); data reduction: SAINT (Bruker, 2008); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: ORTEP-3 for Windows (Farrugia, 1997); software used to prepare material for publication: SHELXL97 (Sheldrick, 2008).

Figures

	Fig. 1. The cation and anion of complex (I), showing the atom-labelling scheme. Displacement ellipsoids are drawn at the 50% probability level. H atoms have been omitted for clarity.
	Fig. 2. The free heterocycle, (II), showing the atom-labelling scheme. Displacement ellipsoids are drawn at the 50% probability level. H atoms have been omitted for clarity.

(I) (η⁵-Cyclopentadienyl)(η⁶-phenoxathiin 10,10-dioxide)iron(II) hexafluoridophosphate

Crystal data

[Fe(C5H5)(C12H8O3S)]PF6	F(000) = 1000
Mr = 498.16	Dx = 1.842 Mg m^−3
Monoclinic, Cc	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: C -2yc	Cell parameters from 5464 reflections
a = 10.059 (2) Å	θ = 2.6–27.8°
b = 13.618 (3) Å	µ = 1.12 mm^−1
c = 13.544 (4) Å	T = 100 K
β = 104.446 (2)°	Block, yellow
V = 1796.6 (8) Å^3	0.30 × 0.18 × 0.18 mm
Z = 4

Data collection

Bruker APEXII CCD area-detector diffractometer	4277 independent reflections
Radiation source: fine-focus sealed tube	3615 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.046
ϕ and ω scans	θmax = 28.4°, θmin = 2.6°
Absorption correction: multi-scan (SADABS; Bruker, 2008)	h = −13→13
Tmin = 0.521, Tmax = 0.746	k = −17→17
10650 measured reflections	l = −17→18

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: inferred from neighbouring sites
R[F^2 > 2σ(F^2)] = 0.046	H-atom parameters constrained
wR(F^2) = 0.089	w = 1/[σ^2(Fo^2) + (0.P)^2] where P = (Fo^2 + 2Fc^2)/3
S = 1.28	(Δ/σ)max = 0.010
4277 reflections	Δρmax = 0.87 e Å^−3
262 parameters	Δρmin = −0.28 e Å^−3
52 restraints	Absolute structure: Flack (1983), with how many Friedel pairs?
Primary atom site location: structure-invariant direct methods	Absolute structure parameter: −0.037 (19)

Special details

Experimental. Solid (phenoxathiin)FeCp hexafluoridophosphate (0.932 g, 2.0 mmol) was added to a mixture of trifluoroacetic acid (20 ml) and 30% hydrogen peroxide (20 ml). The resulting mixture was stirred for 3 h at 333 K, cooled to room temperature and poured into water (120 ml). The solution was extracted with dichloromethane (3 × 20 ml) and then with a 4:1 (v/v) mixture of dichloromethane and nitromethane. The combined extracts were evaporated at reduced pressure, and then acetone (30 ml) was added to the flask and the solvent removed again by evaporation. The addition of acetone and evaporation were repeated twice more to remove water. Finally, the crude product was dissolved in a minimum volume of acetone, diethyl ether was added to start precipitation, and dichloromethane was added to clear the solution. The final solution was kept in a refrigerator at 280 (2) K with the occasional addition of a small volume of diethyl ether until crystals of (I) suitable for data collection appeared in the flask. After collecting these crystals, the complex remaining in solution was precipitated upon addition of an excess of diethyl ether. The product was collected on a sintered glass filter, washed with diethyl ether and dried to give a yellow–orange powder. The total mass of recovered product was 0.678 g (68% yield).

Analytical data for the complex. Elemental analysis, calculated: C 40.97%, H 2.63%; found: C 41.06%, H 2.33%. ¹H NMR (300.133 MHz, acetone-d₆, δ, p.p.m.): Cp 5.25 (5H, s); complexed benzene ring 6.82 (1 H, t, J = 6.0 Hz), 6.90 (1 H, t, J = 6.4 Hz, d, J = 1.1 Hz), 7.19 (1H, d, J = 6.4 Hz), 7.28 (1 H, d, J = 6.0 Hz, d, J = 1.1 Hz); uncomplexed benzene ring 7.74 (1 H, d, J = 8.6 Hz), 7.6 (1 H, degenerate t, J = 7.8 Hz, d, J = 0.9 Hz), 8.02 (1 H, d, J = 8.6 Hz, d, J = 7.6 Hz, d, J = 1.8 Hz), 8.23 (1H, d, J = 7.8 Hz, d, J = 1.8 Hz). ¹³C NMR (75.469 MHz, acetone-d₆, δ, p.p.m.): 79.4 (Cp), 80.5, 81.1, 87.3, 90.5, 94.1 (C of complexed benzene ring), 94.1, 125.0 (quaternary C of complexed benzene ring), 121.0, 124.4, 128.6, 137.6 (C of uncomplexed benzene ring), 127.5, 152.1 (quaternary C of uncomplexed benzene ring).

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

The H atoms were placed in geometrically idealized positions, with C—H = 0.95 (all uncoordinated and coordinated aromatic C atoms) or 1.00 Å (cyclopentadienyl C atoms), and constrained to ride on their parent C atoms, with U_iso(H) = 1.2U_eq(C). For the counter-anion PF₆ group in (I), the F atoms were restrained to have the same U_ij components within a standard uncertainty of 0.05 Ų. The complexed heterocycle structure contains a cyclopentadienyl (Cp) ring which is π-bonded to an Fe atom, and the thermal motion of this Cp ring resulted in unsatisfactory anisotropic displacement parameters for atoms C11, C12, C13 and C15. This was resolved through use of the restrains applied to the refinement of these atoms: the U_ij components of these atoms were restrained to be equal to within 0.004 Ų, and their anisotropic

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
Fe1	0.06320 (5)	0.09010 (4)	0.70531 (4)	0.02552 (14)
C1	0.2221 (4)	0.0277 (3)	0.8146 (3)	0.0263 (9)
H1	0.2245	−0.0240	0.8621	0.032*
C2	0.2422 (4)	0.0078 (3)	0.7182 (3)	0.0267 (9)
H2	0.2583	−0.0574	0.6990	0.032*
C3	0.2381 (4)	0.0857 (3)	0.6507 (3)	0.0290 (9)
H3	0.2517	0.0723	0.5851	0.035*
C4	0.2149 (4)	0.1826 (3)	0.6758 (3)	0.0300 (9)
H4	0.2102	0.2336	0.6272	0.036*
C4A	0.1987 (4)	0.2036 (3)	0.7731 (3)	0.0264 (9)
C10A	0.1981 (4)	0.1254 (3)	0.8407 (3)	0.0225 (8)
O5	0.1739 (3)	0.2994 (2)	0.7910 (2)	0.0331 (7)
C5A	0.1270 (4)	0.3280 (3)	0.8751 (3)	0.0257 (9)
C6	0.0945 (4)	0.4269 (3)	0.8774 (3)	0.0304 (9)
H6	0.1033	0.4689	0.8234	0.037*
C7	0.0500 (4)	0.4635 (3)	0.9572 (4)	0.0373 (11)
H7	0.0297	0.5315	0.9592	0.045*
C8	0.0337 (4)	0.4018 (3)	1.0364 (3)	0.0351 (10)
H8	0.0010	0.4277	1.0911	0.042*
C9	0.0655 (4)	0.3034 (3)	1.0342 (3)	0.0298 (9)
H9	0.0562	0.2613	1.0880	0.036*
C9A	0.1116 (4)	0.2659 (3)	0.9522 (3)	0.0223 (8)
S10	0.16192 (11)	0.14315 (7)	0.95985 (8)	0.0269 (2)
O2	0.0495 (3)	0.0825 (2)	0.9681 (2)	0.0399 (8)
O3	0.2878 (3)	0.1301 (2)	1.0364 (2)	0.0337 (7)
C11	−0.1138 (5)	0.1583 (4)	0.6355 (4)	0.0490 (12)
H11	−0.1233	0.2285	0.6135	0.059*
C12	−0.1239 (5)	0.1223 (4)	0.7294 (4)	0.0488 (12)
H12	−0.1399	0.1631	0.7868	0.059*
C13	−0.1011 (5)	0.0201 (3)	0.7330 (4)	0.0414 (11)
H13	−0.1022	−0.0254	0.7907	0.050*
C14	−0.0871 (4)	−0.0056 (3)	0.6336 (3)	0.0338 (10)
H14	−0.0713	−0.0734	0.6106	0.041*
C15	−0.0910 (5)	0.0790 (4)	0.5764 (4)	0.0406 (11)
H15	−0.0809	0.0826	0.5049	0.049*
P1	0.09922 (12)	0.75056 (8)	0.85832 (9)	0.0312 (3)
F1	0.0852 (4)	0.65239 (19)	0.7938 (2)	0.0636 (9)
F2	−0.0128 (3)	0.7119 (2)	0.9118 (2)	0.0519 (7)
F3	0.1117 (3)	0.84921 (19)	0.9229 (2)	0.0563 (8)
F4	0.2081 (3)	0.79130 (19)	0.8012 (2)	0.0554 (8)
F5	−0.0164 (3)	0.7989 (2)	0.7700 (2)	0.0651 (9)
F6	0.2163 (3)	0.7011 (2)	0.9436 (3)	0.0675 (9)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Fe1	0.0271 (3)	0.0232 (3)	0.0247 (3)	0.0025 (3)	0.0034 (2)	−0.0011 (3)
P1	0.0380 (6)	0.0224 (5)	0.0351 (6)	−0.0033 (5)	0.0128 (5)	−0.0055 (5)
F1	0.120 (3)	0.0206 (14)	0.0559 (19)	−0.0082 (16)	0.032 (2)	−0.0068 (12)
C9	0.028 (2)	0.030 (2)	0.031 (2)	−0.0048 (18)	0.0051 (18)	−0.0025 (19)
F2	0.0525 (17)	0.0442 (17)	0.067 (2)	−0.0053 (14)	0.0303 (15)	0.0102 (15)
O2	0.056 (2)	0.0287 (17)	0.0425 (18)	−0.0058 (15)	0.0268 (16)	0.0005 (14)
C8	0.029 (2)	0.037 (3)	0.040 (3)	0.000 (2)	0.010 (2)	−0.018 (2)
F3	0.093 (2)	0.0319 (15)	0.0565 (19)	−0.0163 (16)	0.0426 (17)	−0.0179 (13)
O3	0.0502 (19)	0.0287 (16)	0.0211 (15)	0.0073 (14)	0.0064 (14)	0.0055 (12)
C7	0.035 (3)	0.019 (2)	0.051 (3)	0.0090 (19)	−0.001 (2)	−0.006 (2)
F4	0.078 (2)	0.0311 (15)	0.076 (2)	−0.0032 (14)	0.0537 (18)	−0.0041 (14)
C6	0.034 (2)	0.021 (2)	0.032 (2)	0.0004 (19)	−0.0001 (18)	0.0043 (18)
C5A	0.031 (2)	0.0204 (19)	0.022 (2)	−0.0034 (17)	−0.0002 (17)	−0.0013 (16)
F5	0.065 (2)	0.053 (2)	0.069 (2)	0.0000 (16)	0.0021 (17)	0.0246 (17)
O5	0.0511 (19)	0.0234 (15)	0.0254 (15)	−0.0016 (14)	0.0109 (13)	0.0018 (12)
C4A	0.031 (2)	0.025 (2)	0.024 (2)	−0.0032 (17)	0.0083 (17)	−0.0038 (17)
C4	0.030 (2)	0.033 (2)	0.027 (2)	−0.0030 (19)	0.0064 (18)	−0.0027 (18)
F6	0.055 (2)	0.058 (2)	0.078 (2)	0.0035 (16)	−0.0051 (17)	0.0206 (18)
C3	0.025 (2)	0.041 (3)	0.0207 (19)	−0.0009 (19)	0.0066 (16)	−0.0068 (18)
C2	0.021 (2)	0.023 (2)	0.032 (2)	0.0046 (16)	−0.0006 (17)	−0.0038 (17)
C10A	0.022 (2)	0.026 (2)	0.0188 (19)	−0.0007 (16)	0.0027 (16)	−0.0059 (16)
C1	0.024 (2)	0.027 (2)	0.024 (2)	−0.0013 (18)	−0.0015 (17)	0.0041 (17)
S10	0.0403 (6)	0.0174 (5)	0.0247 (5)	−0.0005 (4)	0.0112 (4)	0.0002 (4)
C9A	0.027 (2)	0.0206 (19)	0.0201 (19)	−0.0008 (16)	0.0067 (17)	−0.0028 (16)
C11	0.036 (2)	0.035 (2)	0.063 (3)	0.007 (2)	−0.012 (2)	0.0089 (19)
C12	0.032 (2)	0.050 (2)	0.065 (3)	0.003 (2)	0.015 (2)	−0.023 (2)
C13	0.039 (2)	0.042 (2)	0.054 (3)	−0.011 (2)	0.032 (2)	−0.012 (2)
C14	0.031 (2)	0.031 (2)	0.038 (3)	0.0011 (19)	0.006 (2)	−0.0076 (19)
C15	0.028 (2)	0.044 (2)	0.043 (3)	0.0003 (19)	−0.0042 (18)	0.0102 (18)

Geometric parameters (Å, º)

Fe1—C11	2.021 (5)	C6—H6	0.9500
Fe1—C13	2.022 (4)	C5A—C9A	1.382 (5)
Fe1—C15	2.032 (5)	C5A—O5	1.393 (5)
Fe1—C12	2.037 (5)	O5—C4A	1.361 (5)
Fe1—C10A	2.048 (4)	C4A—C10A	1.406 (6)
Fe1—C14	2.048 (4)	C4A—C4	1.398 (5)
Fe1—C1	2.071 (4)	C4—C3	1.397 (6)
Fe1—C3	2.073 (4)	C4—H4	0.9500
Fe1—C2	2.091 (4)	C3—C2	1.394 (6)
Fe1—C4	2.093 (4)	C3—H3	0.9500
Fe1—C4A	2.113 (4)	C2—C1	1.397 (5)
P1—F2	1.575 (3)	C2—H2	0.9500
P1—F6	1.580 (3)	C10A—C1	1.413 (5)
P1—F1	1.584 (3)	C10A—S10	1.757 (4)
P1—F5	1.588 (3)	C1—H1	0.9500
P1—F4	1.590 (3)	S10—C9A	1.743 (4)
P1—F3	1.591 (3)	C11—C12	1.390 (8)
C9—C8	1.380 (6)	C11—C15	1.396 (7)
C9—C9A	1.402 (5)	C11—H11	1.0000
C9—H9	0.9500	C12—C13	1.409 (7)
O2—S10	1.427 (3)	C12—H12	1.0000
C8—C7	1.403 (7)	C13—C14	1.432 (6)
C8—H8	0.9500	C13—H13	1.0000
O3—S10	1.434 (3)	C14—C15	1.382 (6)
C7—C6	1.364 (6)	C14—H14	1.0000
C7—H7	0.9500	C15—H15	1.0000
C6—C5A	1.388 (5)
C11—Fe1—C13	68.8 (2)	C5A—C6—H6	120.1
C11—Fe1—C15	40.3 (2)	C9A—C5A—C6	120.4 (4)
C13—Fe1—C15	68.9 (2)	C9A—C5A—O5	124.9 (3)
C11—Fe1—C12	40.1 (2)	C6—C5A—O5	114.8 (3)
C13—Fe1—C12	40.63 (19)	C4A—O5—C5A	121.9 (3)
C15—Fe1—C12	67.4 (2)	O5—C4A—C10A	125.4 (3)
C11—Fe1—C10A	128.37 (19)	O5—C4A—C4	115.8 (3)
C13—Fe1—C10A	109.52 (18)	C10A—C4A—C4	118.7 (4)
C15—Fe1—C10A	168.67 (19)	O5—C4A—Fe1	130.7 (3)
C12—Fe1—C10A	103.79 (19)	C10A—C4A—Fe1	67.8 (2)
C11—Fe1—C14	67.06 (19)	C4—C4A—Fe1	69.8 (2)
C13—Fe1—C14	41.19 (18)	C3—C4—C4A	119.3 (4)
C15—Fe1—C14	39.61 (18)	C3—C4—Fe1	69.6 (2)
C12—Fe1—C14	67.09 (19)	C4A—C4—Fe1	71.4 (2)
C10A—Fe1—C14	145.08 (18)	C3—C4—H4	120.4
C11—Fe1—C1	162.9 (2)	C4A—C4—H4	120.4
C13—Fe1—C1	101.23 (18)	Fe1—C4—H4	131.3
C15—Fe1—C1	150.64 (19)	C4—C3—C2	122.6 (4)
C12—Fe1—C1	123.6 (2)	C4—C3—Fe1	71.2 (2)
C10A—Fe1—C1	40.13 (15)	C2—C3—Fe1	71.1 (2)
C14—Fe1—C1	115.14 (17)	C4—C3—H3	118.7
C11—Fe1—C3	125.0 (2)	C2—C3—H3	118.7
C13—Fe1—C3	149.12 (18)	Fe1—C3—H3	132.3
C15—Fe1—C3	103.03 (18)	C1—C2—C3	118.6 (4)
C12—Fe1—C3	164.6 (2)	C1—C2—Fe1	69.6 (2)
C10A—Fe1—C3	83.95 (16)	C3—C2—Fe1	69.7 (2)
C14—Fe1—C3	113.83 (17)	C1—C2—H2	120.7
C1—Fe1—C3	70.77 (16)	C3—C2—H2	120.7
C11—Fe1—C2	157.6 (2)	Fe1—C2—H2	132.8
C13—Fe1—C2	117.29 (17)	C4A—C10A—C1	121.4 (4)
C15—Fe1—C2	119.26 (18)	C4A—C10A—S10	122.0 (3)
C12—Fe1—C2	155.9 (2)	C1—C10A—S10	116.6 (3)
C10A—Fe1—C2	71.75 (16)	C4A—C10A—Fe1	72.8 (2)
C14—Fe1—C2	102.53 (16)	C1—C10A—Fe1	70.8 (2)
C1—Fe1—C2	39.22 (15)	S10—C10A—Fe1	127.7 (2)
C3—Fe1—C2	39.12 (16)	C2—C1—C10A	119.4 (4)
C11—Fe1—C4	103.50 (19)	C2—C1—Fe1	71.2 (2)
C13—Fe1—C4	171.01 (19)	C10A—C1—Fe1	69.1 (2)
C15—Fe1—C4	108.51 (18)	C2—C1—H1	120.3
C12—Fe1—C4	130.4 (2)	C10A—C1—H1	120.3
C10A—Fe1—C4	71.23 (16)	Fe1—C1—H1	132.3
C14—Fe1—C4	141.19 (18)	O2—S10—O3	116.44 (19)
C1—Fe1—C4	85.08 (16)	O2—S10—C9A	109.62 (19)
C3—Fe1—C4	39.17 (16)	O3—S10—C9A	110.67 (19)
C2—Fe1—C4	71.62 (16)	O2—S10—C10A	110.01 (19)
C11—Fe1—C4A	105.3 (2)	O3—S10—C10A	107.43 (18)
C13—Fe1—C4A	137.11 (17)	C9A—S10—C10A	101.62 (19)
C15—Fe1—C4A	134.38 (18)	C5A—C9A—C9	119.9 (4)
C12—Fe1—C4A	107.13 (18)	C5A—C9A—S10	123.1 (3)
C10A—Fe1—C4A	39.46 (16)	C9—C9A—S10	116.8 (3)
C14—Fe1—C4A	172.40 (17)	C12—C11—C15	108.2 (4)
C1—Fe1—C4A	71.99 (16)	C12—C11—Fe1	70.6 (3)
C3—Fe1—C4A	70.33 (16)	C15—C11—Fe1	70.2 (3)
C2—Fe1—C4A	84.69 (16)	C12—C11—H11	125.9
C4—Fe1—C4A	38.82 (14)	C15—C11—H11	125.9
F2—P1—F6	90.56 (17)	Fe1—C11—H11	125.9
F2—P1—F1	89.76 (17)	C11—C12—C13	109.3 (4)
F6—P1—F1	89.48 (18)	C11—C12—Fe1	69.4 (3)
F2—P1—F5	90.66 (17)	C13—C12—Fe1	69.1 (3)
F6—P1—F5	178.2 (2)	C11—C12—H12	125.3
F1—P1—F5	89.23 (18)	C13—C12—H12	125.3
F2—P1—F4	178.0 (2)	Fe1—C12—H12	125.3
F6—P1—F4	91.48 (19)	C12—C13—C14	105.2 (4)
F1—P1—F4	90.22 (17)	C12—C13—Fe1	70.3 (3)
F5—P1—F4	87.30 (18)	C14—C13—Fe1	70.4 (3)
F2—P1—F3	89.83 (16)	C12—C13—H13	127.3
F6—P1—F3	90.90 (18)	C14—C13—H13	127.3
F1—P1—F3	179.4 (2)	Fe1—C13—H13	127.3
F5—P1—F3	90.40 (18)	C15—C14—C13	109.1 (4)
F4—P1—F3	90.17 (15)	C15—C14—Fe1	69.6 (3)
C8—C9—C9A	119.6 (4)	C13—C14—Fe1	68.4 (2)
C8—C9—H9	120.2	C15—C14—H14	125.4
C9A—C9—H9	120.2	C13—C14—H14	125.4
C9—C8—C7	119.6 (4)	Fe1—C14—H14	125.4
C9—C8—H8	120.2	C14—C15—C11	108.0 (5)
C7—C8—H8	120.2	C14—C15—Fe1	70.8 (3)
C6—C7—C8	120.7 (4)	C11—C15—Fe1	69.5 (3)
C6—C7—H7	119.6	C14—C15—H15	126.0
C8—C7—H7	119.6	C11—C15—H15	126.0
C7—C6—C5A	119.9 (4)	Fe1—C15—H15	126.0
C7—C6—H6	120.1

(II) Phenoxathiin 10,10-dioxide

Crystal data

C12H8O3S	Z = 2
Mr = 232.24	F(000) = 240
Triclinic, P1	Dx = 1.554 Mg m^−3
Hall symbol: -P 1	Melting point: 419 K
a = 7.2067 (10) Å	Mo Kα radiation, λ = 0.71073 Å
b = 7.9568 (11) Å	Cell parameters from 4033 reflections
c = 8.9360 (13) Å	θ = 2.4–28.2°
α = 102.475 (1)°	µ = 0.31 mm^−1
β = 95.493 (2)°	T = 150 K
γ = 93.399 (2)°	Needle, colourless
V = 496.33 (12) Å^3	0.22 × 0.20 × 0.19 mm

Data collection

Bruker APEXII CCD area-detector diffractometer	1746 independent reflections
Radiation source: fine-focus sealed tube	1592 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.019
ϕ and ω scans	θmax = 25.0°, θmin = 2.4°
Absorption correction: multi-scan (SADABS; Bruker; 2008)	h = −8→8
Tmin = 0.665, Tmax = 0.746	k = −9→9
4816 measured reflections	l = −10→10

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.030	Hydrogen site location: inferred from neighbouring sites
wR(F^2) = 0.072	H-atom parameters constrained
S = 0.97	w = 1/[σ^2(Fo^2) + (0.0265P)^2 + 0.3858P] where P = (Fo^2 + 2Fc^2)/3
1746 reflections	(Δ/σ)max = 0.001
145 parameters	Δρmax = 0.33 e Å^−3
0 restraints	Δρmin = −0.31 e Å^−3

Special details

Experimental. Phenoxathiin 10,10-dioxide, (II), was obtained by oxidation of the precursor phenoxathiin using 30% hydrogen peroxide in glacial acetic acid (Gilman & Esmay, 1952). The mixture was refluxed for 1 h and crystals grew from the solution upon cooling overnight. An additional quantity of the product was recovered from the mother liquor diluted with water upon extraction with chloroform. The total yield of the product was 98%.

Analytical data for the free heterocycle: m.p. 419 (1) K (419–420 K; Gilman & Esmay, 1952). ¹H NMR (300.133 MHz, chloroform-d, δ, p.p.m.): 7.40 (2 H, overlapping t and d, J ca 7.9 Hz), 7.64 (1 H, quasi-t, J ca 7.9 Hz, d, J = 1.4 Hz, d, 1.1 Hz), 8.06 (1 H, d, J = 7.9 Hz, d, J = 1.1 Hz).

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger.

The H atoms were placed in geometrically idealized positions with C–H distances of 0.95 Å (all C atoms) and protons were constrained to ride on the parent C atom with U_iso(H) = 1.2U_eq(C).

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
C1	0.1730 (2)	0.8723 (2)	0.5920 (2)	0.0319 (4)
H1	0.1667	0.9708	0.5481	0.038*
C2	0.1331 (2)	0.8833 (2)	0.7410 (2)	0.0352 (4)
H2	0.0994	0.9890	0.8000	0.042*
C3	0.1422 (2)	0.7390 (2)	0.8051 (2)	0.0341 (4)
H3	0.1144	0.7466	0.9080	0.041*
C4	0.1912 (2)	0.5853 (2)	0.7204 (2)	0.0314 (4)
H4	0.1967	0.4872	0.7648	0.038*
C4A	0.2327 (2)	0.5738 (2)	0.56992 (19)	0.0263 (4)
C10A	0.2227 (2)	0.7172 (2)	0.5048 (2)	0.0263 (4)
O5	0.28352 (18)	0.41578 (14)	0.49805 (13)	0.0318 (3)
C5A	0.3064 (2)	0.3761 (2)	0.34446 (19)	0.0250 (4)
C9A	0.2983 (2)	0.4920 (2)	0.24852 (19)	0.0254 (4)
C6	0.3378 (2)	0.2051 (2)	0.2853 (2)	0.0291 (4)
H6	0.3466	0.1262	0.3512	0.035*
C7	0.3561 (2)	0.1507 (2)	0.1311 (2)	0.0334 (4)
H7	0.3767	0.0335	0.0909	0.040*
C8	0.3448 (3)	0.2646 (2)	0.0331 (2)	0.0362 (4)
H8	0.3563	0.2254	−0.0735	0.043*
C9	0.3168 (3)	0.4345 (2)	0.0921 (2)	0.0326 (4)
H9	0.3101	0.5132	0.0259	0.039*
S10	0.27213 (6)	0.71137 (5)	0.31608 (5)	0.02954 (14)
O2	0.1129 (2)	0.75920 (17)	0.22942 (15)	0.0440 (4)
O3	0.44674 (19)	0.80965 (15)	0.31868 (16)	0.0407 (3)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0291 (9)	0.0244 (9)	0.0412 (10)	0.0039 (7)	−0.0009 (8)	0.0065 (7)
C2	0.0277 (9)	0.0314 (10)	0.0417 (11)	0.0053 (7)	0.0013 (8)	−0.0016 (8)
C3	0.0264 (9)	0.0409 (10)	0.0318 (9)	0.0000 (8)	0.0027 (7)	0.0023 (8)
C4	0.0295 (9)	0.0318 (9)	0.0332 (9)	−0.0005 (7)	0.0009 (7)	0.0102 (8)
C4A	0.0248 (9)	0.0224 (8)	0.0313 (9)	0.0019 (6)	−0.0002 (7)	0.0063 (7)
C10A	0.0240 (8)	0.0232 (8)	0.0318 (9)	0.0022 (6)	0.0001 (7)	0.0076 (7)
O5	0.0475 (8)	0.0207 (6)	0.0302 (6)	0.0087 (5)	0.0065 (5)	0.0096 (5)
C5A	0.0238 (8)	0.0216 (8)	0.0296 (9)	0.0009 (6)	0.0014 (7)	0.0068 (7)
C9A	0.0251 (8)	0.0197 (8)	0.0320 (9)	0.0015 (6)	0.0012 (7)	0.0078 (7)
C6	0.0264 (9)	0.0213 (8)	0.0413 (10)	0.0020 (7)	0.0042 (7)	0.0106 (7)
C7	0.0315 (9)	0.0216 (9)	0.0447 (11)	0.0008 (7)	0.0086 (8)	0.0009 (8)
C8	0.0401 (10)	0.0333 (10)	0.0328 (10)	−0.0019 (8)	0.0074 (8)	0.0021 (8)
C9	0.0370 (10)	0.0300 (9)	0.0322 (9)	−0.0004 (8)	0.0032 (8)	0.0108 (7)
S10	0.0394 (3)	0.0192 (2)	0.0325 (2)	0.00554 (17)	0.00337 (18)	0.01074 (17)
O2	0.0588 (9)	0.0371 (7)	0.0399 (8)	0.0206 (7)	−0.0021 (6)	0.0157 (6)
O3	0.0529 (8)	0.0235 (6)	0.0481 (8)	−0.0036 (6)	0.0155 (6)	0.0111 (6)

Geometric parameters (Å, º)

C1—C2	1.374 (3)	C5A—C6	1.389 (2)
C1—H1	0.9500	C6—C7	1.374 (3)
C2—H2	0.9500	C6—H6	0.9500
C3—C2	1.392 (3)	C7—C8	1.390 (3)
C3—H3	0.9500	C7—H7	0.9500
C4—C3	1.376 (3)	C8—H8	0.9500
C4—H4	0.9500	C9—C8	1.375 (3)
C4A—C4	1.391 (2)	C9—H9	0.9500
C10A—C4A	1.390 (2)	C9A—C9	1.396 (2)
C10A—C1	1.397 (2)	S10—C9A	1.7471 (16)
O5—C4A	1.371 (2)	S10—C10A	1.7481 (18)
O5—C5A	1.369 (2)	S10—O2	1.4388 (13)
C5A—C9A	1.388 (2)	S10—O3	1.4371 (14)
C1—C2—C3	119.80 (17)	C5A—C6—H6	120.1
C1—C2—H2	120.1	C5A—C9A—S10	122.70 (13)
C1—C10A—S10	118.18 (13)	C6—C5A—C9A	119.95 (16)
C2—C1—C10A	120.30 (16)	C6—C7—C8	120.94 (16)
C2—C1—H1	119.8	C6—C7—H7	119.5
C2—C3—H3	119.8	C7—C6—C5A	119.71 (16)
C3—C4—C4A	119.93 (17)	C7—C6—H6	120.1
C3—C4—H4	120.0	C7—C8—H8	120.3
C3—C2—H2	120.1	C8—C9—C9A	120.41 (16)
C4—C4A—C10A	119.91 (16)	C8—C9—H9	119.8
C4—C3—C2	120.49 (17)	C8—C7—H7	119.5
C4—C3—H3	119.8	C9—C9A—S10	117.69 (13)
C4A—C10A—C1	119.57 (16)	C9—C8—C7	119.37 (17)
C4A—C10A—S10	122.25 (13)	C9—C8—H8	120.3
C4A—C4—H4	120.0	C9A—S10—C10A	101.57 (8)
C10A—C1—H1	119.8	C9A—C9—H9	119.8
O5—C4A—C4	114.81 (14)	O2—S10—C10A	109.52 (8)
O5—C4A—C10A	125.28 (15)	O2—S10—C9A	109.30 (8)
O5—C5A—C6	115.17 (14)	O3—S10—O2	116.26 (8)
O5—C5A—C9A	124.88 (15)	O3—S10—C9A	109.47 (8)
C5A—O5—C4A	122.20 (13)	O3—S10—C10A	109.71 (8)
C5A—C9A—C9	119.58 (15)