Effects of d-orbital occupancy on the geometry of the trigonal–bipyramidal complexes [M^IICl₃(Hdabco)(dabco)]_n, where M is Mn, Co, Ni or Cu and dabco is 1,4-di­aza­bi­cyclo­[2.2.2]octane

Geometric data from (1-aza-4-azoniabicyclo­[2.2.2]octane-κN¹)(1,4-diaza­bicyclo­[2.2.2]octane-κN¹)trichloro­manganese(II), [MnCl₃(C₆H₁₃N₂)(C₆H₁₂N₂)] or [MnCl₃(Hdabco)(dabco)] (dabco is 1,4-diaza­bicyclo­[2.2.2]octane), and the cobalt(II) analogue, [CoCl₃(C₆H₁₃N₂)(C₆H₁₂N₂)], have been combined with previously reported data for the Ni and Cu analogues to show that bond-length trends across the isotypal series are consistent with a high-spin trigonal–bipyramidal system. As each transition metal is positioned on a D₃ site in the space group R32 (No. 155), two bond lengths fully define each trigonal–bipyramidal coordination geometry [Mn—Cl = 2.3903 (7) Å and Mn—N = 2.367 (2) Å, and Co—Cl = 2.3080 (8) Å and Co—N = 2.269 (3) Å].

Comment

Comparison of bond lengths in transition metal compounds shows that imperfect shielding of metal nuclei by d electrons leads to a general contraction across a series. This trend is reversed at points where d orbitals that coincide with bonds are occupied. The above effects were noted for octa­hedral metal oxides by van Santen & van Wieringen (1952) and demonstrated more recently for a series of octa­hedral complexes by Cotton et al. (1984, 1993). The latter compounds also show evidence of Jahn–Teller distortion. Chemists usually describe the above effects in terms of crystal- or ligand-field theories, which attempt to quantify the energy cost of placing an electron in a bonding d orbital (e.g. for an octa­hedral complex, the d orbitals divide into high-energy d_x²-y² and d_x², directed along the bonds, and lower-energy d_xy, d_xz and d_yz, projecting between the bonds). Importantly, the energy gap between the two sets of orbitals determines whether it is energetically favourable to pair electrons in the lowest-energy orbitals or to semi-fill all five d orbitals before starting to pair electrons (i.e. low-spin versus high-spin configurations).

The same rationale can be used for analysing the title trigonal–bipyramidal (tbp) structures, [M^IICl₃(Hdabco)(dabco)_n, where M is Mn, Co, Ni or Cu and dabco is 1,4-diazabicyclo­[2.2.2]octane. The Mn and Co structures, (I) and (II), from the current study (Fig. 1) form an isotypal series with the previously reported Ni (Petrusenko et al., 1997) and Cu analogues (Karan et al., 1999; Viossat et al., 1988). In the tbp geometry, the five d orbitals divide into three subsets, viz. low-energy d_xz and d_yz, medium-energy d_xy and d_x²-y², lying in the same plane as the three equatorial M—Cl bonds, and high-energy d_z², pointing directly along the axial M—N bonds. When the crystallographically determined M—Cl and M—N bond lengths are plotted against d-electron configuration (Fig. 2), the curves diverge at Ni^II. This is as expected for a high-spin tbp system. All bonds contract as electrons fill non-bonding d_xz and d_yz orbitals in the d⁵ to d⁷ configurations, but occupation of equatorial d_xy and d_x²-y² in d⁸ and d⁹ causes the M—Cl bonds to lengthen, whilst allowing the M—N bonds to continue contracting. The M—N bonds would not be expected to expand until all five d orbitals are fully occupied. Support for the high-spin assignment is obtained from the tbp compounds [M^IICl₃(dabcoH)₂] (M = Mn, Fe, Ni and Cu), which were previously classified as high-spin based on spectroscopic studies (Vallarino et al., 1972).

It is likely that Jahn–Teller distortion of the Ni—Cl bonds has been masked by the crystallographic threefold axis. Previous structural studies of [M^IICl₃(CH₃dabco)₂]ClO₄ (M = Ni and Cu; Rozell & Wood 1977), where all the bonds are crystallographically independent, showed a small but significant deformation in the d⁸ Ni complex. The dabco dimensions in the title Mn and Co structures show good agreement and are as expected (Tables 1 and 2). Inter­estingly, the M—N bonds are aligned with the crystal c axis in the current structures so that variations in unit-cell dimensions mirror changes in bond lengths. This suggests that spectroscopic excitation of d electrons may well have a macroscopic effect and that these materials should be investigated for opto-mechanical uses.

Figure 1 A view of [Co^IICl3(Hdabco)(dabco)]n (50% probability displacement ellipsoids). Most H atoms have been omitted for clarity.

Figure 2 A plot of M—Cl and M—N bond lengths (Å) versus d-electron configuration in [M^IICl3(Hdabco)(dabco)]n, where M is Mn and Co (current work), Ni (Petrusenko et al., 1997), and Cu (Viossat et al., 1988; Karan et al., 1999). The M—Cl and M—N s.u. values are in the ranges 0.0007–0.002 and 0.002–0.006 Å, respectively.

Experimental

Crystals of the title Mn and Co compounds were prepared by mixing equal volumes of 0.2 M methanol solutions of dabco and MCl₂. Long needles, suitable for crystallographic investigation, formed when these solutions were subjected to vapour diffusion using diethyl ether as the anti-solvent.

Compound (I)

Crystal data

• [MnCl₃(C₆H₁₃N₂)(C₆H₁₂N₂)]

• M_r = 386.65

• Hexagonal, R 32

• a = 10.601 (2) Å

• c = 12.484 (2) Å

• V = 1215.0 (4) ų

• Z = 3

• D_x = 1.585 Mg m⁻³

• Mo Kα radiation

• μ = 1.31 mm⁻¹

• T = 223 (2) K

• Needle, colourless

• 0.3 × 0.1 × 0.1 mm

Data collection

• Enraf–Nonius CAD-4 diffractometer

• Non-profiled ω/2θ scans

• Absorption correction: ψ scan (North et al., 1968) T_min = 0.696, T_max = 0.881

• 1514 measured reflections

• 487 independent reflections

• 480 reflections with I > 2σ(I)

• R_int = 0.019

• θ_max = 25.0°

• 3 standard reflections frequency: 60 min intensity decay: 2%

Refinement

• Refinement on F²

• R[F² > 2σ(F²)] = 0.015

• wR(F²) = 0.037

• S = 1.07

• 487 reflections

• 50 parameters

• H atoms treated by a mixture of independent and constrained refinement

• w = 1/[σ²(F_o²) + (0.0248P)²] where P = (F_o² + 2F_c²)/3

• (Δ/σ)_max = 0.001

• Δρ_max = 0.16 e Å⁻³

• Δρ_min = −0.13 e Å⁻³

• Extinction correction: SHELXL97

• Extinction coefficient: 0.0011 (4)

• Absolute structure: Flack (1983), 201 Friedel pairs

• Flack parameter: 0.01 (2)

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

Mn1—N1	2.367 (2)
Mn1—Cl1	2.3903 (7)
N1—C2	1.4717 (18)
C2—C3	1.537 (2)
C3—N4	1.4806 (16)

C2—N1—C2^i	108.23 (10)
N1—C2—C3	110.77 (13)
N4—C3—C2	108.96 (12)
C3—N4—C3^i	108.97 (9)
Symmetry code: (i) -y+2, x-y+1, z.

Compound (II)

Crystal data

• [CoCl₃(C₆H₁₃N₂)(C₆H₁₂N₂)]

• M_r = 390.64

• Trigonal, R 32

• a = 10.5409 (10) Å

• c = 12.303 (2) Å

• V = 1183.9 (2) ų

• Z = 3

• D_x = 1.644 Mg m⁻³

• Mo Kα radiation

• μ = 1.59 mm⁻¹

• T = 223 (2) K

• Needle, blue

• 0.25 × 0.1 × 0.05 mm

Data collection

• Enraf–Nonius CAD-4 diffractometer

• Non-profiled ω/2θ scans

• Absorption correction: ψ scan (North et al., 1968) T_min = 0.692, T_max = 0.925

• 1463 measured reflections

• 470 independent reflections

• 457 reflections with I > 2σ(I)

• R_int = 0.029

• θ_max = 25.0°

Refinement

• Refinement on F²

• R[F² > 2σ(F²)] = 0.020

• wR(F²) = 0.041

• S = 1.1

• 470 reflections

• 50 parameters

• H atoms treated by a mixture of independent and constrained refinement

• w = 1/[σ²(F_o²) + (0.0197P)²] where P = (F_o² + 2F_c²)/3

• (Δ/σ)_max < 0.001

• Δρ_max = 0.15 e Å⁻³

• Δρ_min = −0.16 e Å⁻³

• Extinction correction: SHELXL97

• Extinction coefficient: 0.0006 (3)

• Absolute structure: Flack (1983), 194 Friedel pairs

• Flack parameter: 0.00 (2)

Table 2
Selected geometric parameters (Å, °) for (II)

Co1—N1	2.269 (3)
Co1—Cl1	2.3080 (8)
N1—C2	1.472 (3)
C2—C3	1.528 (3)
C3—N4	1.482 (3)

C2—N1—C2^i	107.60 (16)
N1—C2—C3	111.3 (2)
N4—C3—C2	109.0 (2)
C3—N4—C3^i	108.62 (14)
Symmetry code: (i) -y, x-y, z.

Final refinements of both structures were carried out in the space group R32 (No. 155) with all non-H atoms anisotropic and H atoms isotropic, except for amine atom H4, whose U_iso(H) value was fixed at 0.028 (Mn) or 0.025 Ų (Co) due to the close proximity of a disordered symmetry-related position. In this space group, a disordered amine H atom semi-populates two sites on the threefold axis between adjacent N atoms. As previous refinements of the Cu analogue were carried out in both R32 and R3 (No. 146), which gave an improved description of the amine H atom, refinements of the current structures were also attempted in R3. However, as the amine H atoms remained disordered, the R3 refinements were discontinued. The choice of space group made negligible difference to the M—Cl and M—N bond lengths.

For both compounds, data collection: CAD-4 EXPRESS (Enraf–Nonius, 1994); cell refinement: CAD-4 EXPRESS; data reduction: XCAD4 (Harms & Wocadlo, 1995); program(s) used to solve structure: SHELXS86 (Sheldrick, 1985) for (I) and SHELXS97 (Sheldrick, 1997) for (II); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: ORTEP-3 for Windows (Farrugia, 1997); software used to prepare material for publication: WinGX (Farrugia, 1999).

Supporting information

S1. Comment

Comparison of bond lengths in transition metal compounds shows that imperfect shielding of metal nuclei by d electrons leads to a general contraction across a series. This trend is reversed at points where d orbitals that coincide with bonds are occupied. The above effects were noted for octahedral metal oxides by van Santen & van Wieringen (1952) and demonstrated more recently for a series of octahedral complexes by Cotton et al. (1984, 1993). The latter compounds also show evidence of Jahn–Teller distortion. Chemists usually describe the above effects in terms of crystal- or ligand-field theories, which attempt to quantify the energy cost of placing an electron in a bonding d orbital (e.g. for an octahedral complex the d orbitals divide into high energy d_{x2 − y2} and d_z2, directed along the bonds, and lower energy d_xy, d_xz and d_yz projecting between the bonds). Importantly, the energy gap between the two sets of orbitals determines whether it is energetically favourable to pair electrons in the lowest energy orbitals or to semi-fill all five d orbitals before starting to pair electrons (i.e. low-spin versus high-spin configurations).

The same rationale can be used for analysing the title trigonal–bipyramidal (tbp) structures, [M^II(Hdabco)(dabco)Cl₃]_n, where M = Mn, Co, Ni or Cu and dabco = 1,4-diazabicyclo[2.2.2]octane. The Mn and Co structures from the current study (Fig. 1) form an isotypal series with the previously reported Ni (Petrusenko et al., 1997) and Cu analogues (Karan et al., 1999; Viossat et al., 1988). In the tbp geometry, the five d orbitals divide into three sub-sets, viz. low-energy d_xz and d_yz, medium-energy d_xy and d_{x2 − y2} lying in the same plane as the three equatorial M—Cl bonds, and high-energy d_z2, pointing directly along the axial M—N bonds. When the crystallographically determined M—Cl and M—N bond lengths are plotted against d electron configuration (Fig. 2), the curves diverge at Ni^II. This is as expected for a high-spin tbp system. All bonds contract as electrons fill non-bonding d_xz and d_yz orbitals in the d⁵ to d⁷ configurations, but occupation of equatorial d_xy and d_{x2 − y2} in d⁸ and d⁹ cause the M—Cl bonds to lengthen, whilst allowing the M—N bonds to continue contracting. The M—N bonds would not be expected to expand until all five d orbitals are fully occupied. Support for the high-spin assignment is obtained from the tbp compounds, [M^IICl₃(dabcoH)₂] (M = Mn, Fe, Ni and Cu), which were previously classified as high-spin based on spectroscopic studies (Vallarino et al., 1972).

It is likely that Jahn–Teller distortion of the Ni—Cl bonds has been masked by the crystallographic threefold axis. Previous structural studies of [M^II(CH₃dabco)₂Cl₃]ClO₄ (M = Ni and Cu; Rozell & Wood 1977), where all the bonds are crystallographically independent, showed a small but significant deformation in the d⁸ Ni complex. The dabco dimensions in the title Mn and Co structures show good agreement and are as expected (Table 2). Interestingly, the M—N bonds are aligned with the crystal c axis in the current structures so that variations in unit-cell dimensions mirror changes in bond lengths. This suggests that spectroscopic excitation of d electrons may well have a macroscopic effect and that these materials should be investigated for opto-mechanical uses.

S2. Experimental

Crystals of the title Mn and Co compounds were prepared by mixing equal volumes of 0.02 M methanol solutions of dabco and MCl₂. Long needles, suitable for crystallographic investigation, formed when these solutions were subjected to vapour diffusion using diethyl ether as the anti-solvent.

S3. Refinement

Final refinements of both structures were carried out in space group R32 (No. 155) with all non-H atoms anisotropic and H atoms isotropic, except for atom H4, whose U_iso(H) value was fixed at 0.028 Ų (Mn) or 0.025 Ų (Co) due to the close proximity of a disordered, symmetry-related ghost. In this space group, a disordered amine H atom semi-populates two sites on the threefold axis between adjacent N atoms. As previous refinements of the Cu analogue were carried out in both R32 and R3 (No. 146), which gave an improved description of the amine H atom, refinements of the current structures were also attempted in R3. However, as the amine H atoms remained disordered, the R3 refinements were discontinued. The choice of space group made negligible difference to the M—Cl and M—N bond lengths.

Computing details

For both compounds, data collection: CAD-4 EXPRESS (Enraf–Nonius, 1994); cell refinement: CAD-4 EXPRESS; data reduction: XCAD4 (Harms & Wocadlo, 1995). Program(s) used to solve structure: SHELXS86 (Sheldrick, 1986) for (I); SHELXS97 (Sheldrick, 1997) for (II). For both compounds, program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: ORTEP-3 for Windows (Farrugia, 1997); software used to prepare material for publication: WinGX (Farrugia, 1999).

Figures

	Fig. 1. A view of [Co^II(Hdabco)(dabco)Cl3]n, (50% probability displacement ellipsoids). Most H atoms have been omitted for clarity.
	Fig. 2. A plot of M—Cl and M—N bond lengths (Å) versus d-electron configuration in [M^II(Hdabco)(dabco)Cl3]n, where M is Mn and Co (current work), Ni (Petrusenko et al., 1997), and Cu (Viossat et al., 1988; Karan et al., 1999). The M—Cl and M—N s.u. values are in the ranges 0.0007–0.002, and 0.002–0.006 Å, respectively.

(I) (1-aza-4-azoniabicyclo[2.2.2]octane-κN¹) (1,4-diazabicyclo[2.2.2]octane-κN¹)trichloromanganese(II)

Crystal data

[MnCl3(C6H13N2)(C6H12N2)]	Dx = 1.585 Mg m^−3
Mr = 386.65	Mo Kα radiation, λ = 0.71073 Å
Hexagonal, R32	Cell parameters from 25 reflections
Hall symbol: R 3 2"	θ = 12.6–15.4°
a = 10.601 (2) Å	µ = 1.31 mm^−1
c = 12.484 (2) Å	T = 223 K
V = 1215.0 (4) Å^3	Needle, colourless
Z = 3	0.3 × 0.1 × 0.1 mm
F(000) = 603

Data collection

Enraf–Nonius CAD-4 diffractometer	Rint = 0.019
non–profiled ω/2θ scans	θmax = 25.0°, θmin = 3.8°
Absorption correction: ψ scan North et al., 1968	h = −12→10
Tmin = 0.696, Tmax = 0.881	k = 0→12
1514 measured reflections	l = −14→14
487 independent reflections	3 standard reflections every 60 min
480 reflections with I > 2σ(I)	intensity decay: 2%

Refinement

Refinement on F^2	w = 1/[σ^2(Fo^2) + (0.0248P)^2] where P = (Fo^2 + 2Fc^2)/3
Least-squares matrix: full	(Δ/σ)max = 0.001
R[F^2 > 2σ(F^2)] = 0.015	Δρmax = 0.16 e Å^−3
wR(F^2) = 0.037	Δρmin = −0.13 e Å^−3
S = 1.07	Extinction correction: SHELXL97, Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
487 reflections	Extinction coefficient: 0.0011 (4)
50 parameters	Absolute structure: Flack (1983), 201 Friedel pairs
0 restraints	Absolute structure parameter: 0.01 (2)
H atoms treated by a mixture of independent and constrained refinement

Special details

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.

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

	x	y	z	Uiso*/Ueq	Occ. (<1)
Mn1	1	1	0	0.01470 (17)
Cl1	0.77452 (5)	1	0	0.02454 (18)
N1	1	1	0.18956 (16)	0.0168 (4)
C2	0.86318 (18)	0.8784 (2)	0.23120 (13)	0.0229 (4)
C3	0.85547 (15)	0.88817 (17)	0.35364 (13)	0.0198 (4)
N4	1	1	0.39414 (15)	0.0173 (4)
H2A	0.784 (2)	0.874 (2)	0.198 (2)	0.042 (5)*
H2B	0.861 (2)	0.793 (2)	0.2143 (16)	0.033 (5)*
H3A	0.7894 (18)	0.9161 (15)	0.3741 (12)	0.015 (4)*
H3B	0.8348 (17)	0.800 (2)	0.3875 (13)	0.021 (4)*
H4	1	1	0.482 (6)	0.028*	0.5

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Mn1	0.0160 (2)	0.0160 (2)	0.0122 (3)	0.00798 (11)	0	0
Cl1	0.0219 (2)	0.0362 (3)	0.0203 (3)	0.01812 (17)	0.00005 (10)	0.0001 (2)
N1	0.0184 (6)	0.0184 (6)	0.0136 (9)	0.0092 (3)	0	0
C2	0.0205 (8)	0.0234 (8)	0.0152 (8)	0.0037 (7)	−0.0008 (6)	0.0007 (6)
C3	0.0199 (9)	0.0216 (8)	0.0154 (7)	0.0086 (7)	0.0018 (6)	0.0017 (5)
N4	0.0192 (7)	0.0192 (7)	0.0137 (11)	0.0096 (3)	0	0

Geometric parameters (Å, º)

Mn1—N1^i	2.367 (2)	C2—H2A	0.91 (2)
Mn1—N1	2.367 (2)	C2—H2B	0.92 (2)
Mn1—Cl1^ii	2.3903 (7)	C3—N4	1.4806 (16)
Mn1—Cl1	2.3903 (7)	C3—H3A	0.923 (18)
Mn1—Cl1^iii	2.3903 (7)	C3—H3B	0.95 (2)
N1—C2^ii	1.4717 (18)	N4—C3^ii	1.4806 (16)
N1—C2	1.4717 (18)	N4—C3^iii	1.4806 (16)
N1—C2^iii	1.4717 (18)	N4—H4	1.09 (7)
C2—C3	1.537 (2)
N1^i—Mn1—N1	180	N1—C2—H2A	111.2 (14)
N1^i—Mn1—Cl1^ii	90	C3—C2—H2A	111.9 (14)
N1—Mn1—Cl1^ii	90	N1—C2—H2B	108.0 (13)
N1^i—Mn1—Cl1	90	C3—C2—H2B	108.5 (13)
N1—Mn1—Cl1	90	H2A—C2—H2B	106.3 (17)
Cl1^ii—Mn1—Cl1	120	N4—C3—C2	108.96 (12)
N1^i—Mn1—Cl1^iii	90	N4—C3—H3A	107.2 (9)
N1—Mn1—Cl1^iii	90	C2—C3—H3A	111.8 (9)
Cl1^ii—Mn1—Cl1^iii	120	N4—C3—H3B	106.0 (10)
Cl1—Mn1—Cl1^iii	120	C2—C3—H3B	111.7 (11)
C2^ii—N1—C2	108.23 (10)	H3A—C3—H3B	110.9 (13)
C2^ii—N1—C2^iii	108.23 (10)	C3—N4—C3^ii	108.97 (9)
C2—N1—C2^iii	108.23 (10)	C3—N4—C3^iii	108.97 (9)
C2^ii—N1—Mn1	110.68 (10)	C3^ii—N4—C3^iii	108.97 (9)
C2—N1—Mn1	110.68 (10)	C3—N4—H4	109.97 (10)
C2^iii—N1—Mn1	110.68 (10)	C3^ii—N4—H4	109.97 (9)
N1—C2—C3	110.77 (13)	C3^iii—N4—H4	109.97 (9)
Cl1^ii—Mn1—N1—C2^ii	54.17 (8)	Cl1^iii—Mn1—N1—C2^iii	54.17 (8)
Cl1—Mn1—N1—C2^ii	−65.83 (8)	C2^ii—N1—C2—C3	−52.13 (16)
Cl1^iii—Mn1—N1—C2^ii	174.17 (8)	C2^iii—N1—C2—C3	64.96 (14)
Cl1^ii—Mn1—N1—C2	174.17 (8)	Mn1—N1—C2—C3	−173.58 (10)
Cl1—Mn1—N1—C2	54.17 (8)	N1—C2—C3—N4	−10.99 (17)
Cl1^iii—Mn1—N1—C2	−65.83 (8)	C2—C3—N4—C3^ii	65.67 (15)
Cl1^ii—Mn1—N1—C2^iii	−65.83 (8)	C2—C3—N4—C3^iii	−53.12 (15)
Cl1—Mn1—N1—C2^iii	174.17 (8)

Symmetry codes: (i) x−y+1, −y+2, −z; (ii) −x+y+1, −x+2, z; (iii) −y+2, x−y+1, z.

(II) (1-aza-4-azoniabicyclo[2.2.2]octane-κN¹) (1,4-diazabicyclo[2.2.2]octane-κN¹)trichlorocobalt(II)

Crystal data

[CoCl3(C6H13N2)(C6H12N2)]	Dx = 1.644 Mg m^−3
Mr = 390.64	Mo Kα radiation, λ = 0.71073 Å
Trigonal, R32	Cell parameters from 25 reflections
Hall symbol: R 3 2"	θ = 6.1–9.2°
a = 10.5409 (10) Å	µ = 1.59 mm^−1
c = 12.303 (2) Å	T = 223 K
V = 1183.9 (2) Å^3	Prism, blue
Z = 3	0.25 × 0.1 × 0.05 mm
F(000) = 609

Data collection

Enraf–Nonius CAD-4 diffractometer	457 reflections with I > 2σ(I)
non–profiled ω/2θ scans	Rint = 0.029
Absorption correction: ψ scan North et al., 1968	θmax = 25.0°, θmin = 3.9°
Tmin = 0.692, Tmax = 0.925	h = 0→12
1463 measured reflections	k = −12→10
470 independent reflections	l = −14→14

Refinement

Refinement on F^2	w = 1/[σ^2(Fo^2) + (0.0197P)^2] where P = (Fo^2 + 2Fc^2)/3
Least-squares matrix: full	(Δ/σ)max < 0.001
R[F^2 > 2σ(F^2)] = 0.020	Δρmax = 0.15 e Å^−3
wR(F^2) = 0.041	Δρmin = −0.16 e Å^−3
S = 1.1	Extinction correction: SHELXL97, Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
470 reflections	Extinction coefficient: 0.0006 (3)
50 parameters	Absolute structure: Flack (1983), 194 Friedel pairs
0 restraints	Absolute structure parameter: 0.00 (2)
H atoms treated by a mixture of independent and constrained refinement

Special details

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.

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

	x	y	z	Uiso*/Ueq	Occ. (<1)
Co1	0	0	1	0.0158 (2)
Cl1	0.21896 (8)	0	1	0.0238 (2)
N1	0	0	0.8155 (2)	0.0163 (6)
C2	0.1356 (3)	0.1238 (3)	0.7721 (2)	0.0235 (6)
C3	0.1450 (3)	0.1127 (3)	0.6488 (2)	0.0211 (6)
N4	0	0	0.6070 (2)	0.0168 (6)
H2A	0.213 (3)	0.126 (3)	0.806 (2)	0.037 (8)*
H2B	0.141 (3)	0.208 (3)	0.789 (2)	0.020 (7)*
H3A	0.206 (3)	0.082 (2)	0.6304 (18)	0.019 (7)*
H3B	0.172 (2)	0.198 (3)	0.6132 (19)	0.025 (6)*
H4	0	0	0.527 (6)	0.025*	0.5

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Co1	0.0158 (3)	0.0158 (3)	0.0158 (4)	0.00792 (15)	0	0
Cl1	0.0204 (4)	0.0356 (5)	0.0205 (5)	0.0178 (3)	−0.00067 (17)	−0.0013 (3)
N1	0.0167 (9)	0.0167 (9)	0.0155 (15)	0.0083 (4)	0	0
C2	0.0217 (14)	0.0222 (14)	0.0185 (13)	0.0049 (11)	−0.0019 (10)	0.0003 (11)
C3	0.0195 (16)	0.0240 (14)	0.0182 (11)	0.0097 (11)	0.0028 (10)	0.0028 (9)
N4	0.0186 (10)	0.0186 (10)	0.0134 (16)	0.0093 (5)	0	0

Geometric parameters (Å, º)

Co1—N1^i	2.269 (3)	C2—H2A	0.91 (3)
Co1—N1	2.269 (3)	C2—H2B	0.89 (3)
Co1—Cl1^ii	2.3080 (9)	C3—N4	1.482 (3)
Co1—Cl1	2.3080 (8)	C3—H3A	0.88 (3)
Co1—Cl1^iii	2.3080 (8)	C3—H3B	0.91 (3)
N1—C2	1.472 (3)	N4—C3^iii	1.482 (3)
N1—C2^ii	1.472 (3)	N4—C3^ii	1.482 (3)
N1—C2^iii	1.472 (3)	N4—H4	0.98 (7)
C2—C3	1.528 (3)
N1^i—Co1—N1	180	N1—C2—H2A	108.3 (18)
N1^i—Co1—Cl1^ii	90.0000 (10)	C3—C2—H2A	111.7 (17)
N1—Co1—Cl1^ii	90.0000 (10)	N1—C2—H2B	110.7 (18)
N1^i—Co1—Cl1	90	C3—C2—H2B	109.6 (17)
N1—Co1—Cl1	90	H2A—C2—H2B	105 (2)
Cl1^ii—Co1—Cl1	120	N4—C3—C2	109.0 (2)
N1^i—Co1—Cl1^iii	90	N4—C3—H3A	105.3 (14)
N1—Co1—Cl1^iii	90	C2—C3—H3A	111.8 (15)
Cl1^ii—Co1—Cl1^iii	120	N4—C3—H3B	108.0 (14)
Cl1—Co1—Cl1^iii	120	C2—C3—H3B	113.8 (17)
C2—N1—C2^ii	107.60 (16)	H3A—C3—H3B	109 (2)
C2—N1—C2^iii	107.60 (16)	C3—N4—C3^iii	108.62 (14)
C2^ii—N1—C2^iii	107.60 (16)	C3—N4—C3^ii	108.62 (14)
C2—N1—Co1	111.29 (15)	C3^iii—N4—C3^ii	108.62 (14)
C2^ii—N1—Co1	111.29 (15)	C3—N4—H4	110.31 (14)
C2^iii—N1—Co1	111.29 (15)	C3^iii—N4—H4	110.31 (14)
N1—C2—C3	111.3 (2)	C3^ii—N4—H4	110.31 (14)
Cl1^ii—Co1—N1—C2	−175.50 (14)	Cl1^iii—Co1—N1—C2^iii	−55.50 (14)
Cl1—Co1—N1—C2	−55.50 (14)	C2^ii—N1—C2—C3	50.3 (3)
Cl1^iii—Co1—N1—C2	64.50 (14)	C2^iii—N1—C2—C3	−65.4 (2)
Cl1^ii—Co1—N1—C2^ii	−55.50 (14)	Co1—N1—C2—C3	172.43 (17)
Cl1—Co1—N1—C2^ii	64.50 (14)	N1—C2—C3—N4	13.0 (3)
Cl1^iii—Co1—N1—C2^ii	−175.50 (14)	C2—C3—N4—C3^iii	51.6 (2)
Cl1^ii—Co1—N1—C2^iii	64.50 (14)	C2—C3—N4—C3^ii	−66.3 (2)
Cl1—Co1—N1—C2^iii	−175.50 (14)

Symmetry codes: (i) x−y, −y, −z+2; (ii) −x+y, −x, z; (iii) −y, x−y, z.