Novel Structures and Magnetic Properties of Two [Mn₂] Complexes with 2,4-di-2-pyridyl-2,4-pentanediol as the Ligand

Abstract

Two ligands, 2,4-di-2-pyridyl-2,4-pentanediol (rD and mD), were employed to synthesize two Mn₂ complexes, [Mn₂(rD)₂Br₂] (1) and [Mn₂(mD)₂(H₂O)₂]Br₂ (2). Compound 1 crystallized in a tetragonal space group, P4₁2₁2, with a novel hamburger shaped structure. A detailed study indicated that compound 1 did not contain a metal–metal bond, but antiferromagnetic coupling was observed between the Mn(III) ions. Compound 2 crystallized in a monoclinic space group, C2/c, with one Mn(II) and the other with Mn(IV). The two manganese ions were bridged by two alkoxide ligands, resulting in ferromagnetic coupling. Magnetic property studies confirm the above assignments.

1. Introduction

Manganese containing coordination complexes are of great interest due to their unique magnetic properties [1,2,3] as well as their variety of functions in chemistry [4,5,6]. A very large number of oxygen bridging compounds have been synthesized [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24]. However, in a survey of the literature, all of the manganese ions in these complexes are connected with neighboring manganese ions through bridging atoms. The only example of a complex where manganese atoms are in direct contact is the well known (OC)₅Mn-Mn(CO)₅ complex, which contains a single metal–metal bond between the two Mn units [25]. Because the unpaired electrons on the metal centers are paired, this moiety is therefore diamagnetic. Except for this example, there are no reports in the literature of a capped form of a structure with a direct Mn–Mn bond. Nevertheless, in our attempts to use the 2-(2-pyridyl)-isopropanol (dmhmp-H) ligand to synthesize a manganese cluster, we accidently but repeatedly isolated a small amount of a capped form of an Mn₂ complex. Although only small amounts of a dark-green complex were isolated, elemental analysis as well as an XRD powder pattern indicated that it was a highly pure compound. Due to its special structure, we attempted to develop a method to systematically prepare this compound. Based on the X-ray structure, we concluded that the ligand was a racemate of 2,4-di-2-pyridyl-2,4-pentanediol (rD). Numerous efforts were made to optimize the yield, but they were not successful [26]. In a search of the literature, we found that Shopov and co-workers [27] encountered a situation that was very similar to ours, i.e., the most efficient route for preparing such complexes is to start by preparing dmhmp-H from a Grignard reagent [27,28]. We then used MeMgI as the starting material and purified the product by distillation followed by flash column chromatography. Approximately 1% of the racemate and the meso 2,4-di-2-pyridyl-2,4-pentanediol (rD and mD) were obtained, respectively. Armed with the pure ligands, we were then able to generate isolable amounts of the capped form of the [Mn₂(rD)₂Br₂] (1) complex. For a comparison in magnetic property studies, we also used the same procedure for preparing compound 1 to synthesize [Mn₂(mD)₂(H₂O)₂]Br₂ (2). The magnetic properties of compound 2 remained unreported, although a similar compound with different counter ions, [Mn₂(mD)₂(H₂O)₂](NO₃)_2, was reported by Shopov and co-workers [27]. In this report, we describe the novel structure of compound 1 and the magnetic properties of both compounds 1 and 2.

2. Results and Discussion

2.1. Description of Structures

The 2,4-di-2-pyridyl-2,4-pentanediol (rD and mD) ligands (see Scheme 1) were crucial for preparing these complexes. With the same synthetic approach, different ligands (rD and mD) led to different structures. Numerous efforts were made to synthesize these ligands, but none were successful. Finally, we realized that the best strategy for producing the products was the original reaction condition used for the synthesis of dmhmpH. The Grignard reagent automatically led to minor but promising amounts of rD and mD. After careful work on distillation and chromatography, a reasonable amount of rD and mD ligands could be obtained.

Compound 1 crystallized in a tetragonal space group, P4₁2₁2. X-ray refinement data for the complex are listed in

Table 1. Refined crystal data and unit cell information.

Empirical Formula	C30H32N4O4Br2Mn2	C30H44N4O10Br2Mn2
Fw	782.3	890.39
Space Group	P41212	C2/c
a, Å	10.1069(5)	27.9847(12)
b, Å	10.1069(5)	8.3508(3)
c, Å	29.1085(16)	19.8000(8)
α, deg	90	90
β, deg	90	129.3794(10)
γ, deg	90	90
V, (Å3)	2973.4(3)	3576.6(2)
Z	4	4
ρ(calc), Mg/m3	1.748	1.654
T, K	150(2)	150(2)
λ, Å	0.71073	0.71073
θ range (deg)	2.133–27.499	2.614–30.000
hkl ranges	−10<=h<=12, −13<=k<=13, −37<=l<=37	−39<=h<=39, −11<=k<=11, −27<=l<=27
Reflections collected	18008	17265
Independent reflections	3411 [R(int) = 0.0354]	5220 [R(int) = 0.0262]
μ, mm−1	3.580	2.999
GOF	1.139	1.052
R1 [I > 2σ(I)]	0.0224	0.0228
wR2	0.0547	0.0571
Flack parameter	0.009(5)

R₁ = Σ||F₀| − |F_c||/Σ|F₀|; wR₂ = [Σw(F₀² − F_c²)²/Σw(F₀²)²]^1/2; GOF = $\sqrt{\frac{{{\displaystyle \sum }}_{hkl}w{\Delta }^2}{m-n}}$

. The structural plot of complex 1 is presented in Figure 1. The structure clearly shows that two manganese ions were penta-coordinated and approached each other through the flat vacancy site of the square-pyramidal structure. Two rD ligands held these two ions in a hamburger form through the nitrogen atoms of the pyridine and the oxygen atom from the alkoxide arm. Two bromide ions functioned as caps from both the top and the bottom of the complex. The metal–ligand bond lengths are shown in

Table 2. Selected bond distances (Å) and angles (deg) of compound 1.

Mn(1)-O(1)	1.846(2)	Mn(2)-O(2)	1.854(2)
Mn(1)-O(1)#1	1.846(2)	Mn(2)-O(2)#1	1.854(2)
Mn(1)-N(1)	2.003(2)	Mn(2)-N(2)	2.004(2)
Mn(1)-N(1)#1	2.003(2)	Mn(2)-N(2)#1	2.004(2)
Mn(1)-Br(1)	2.5853(7)	Mn(2)-Br(2)	2.5710(7)
O(1)-Mn(1)-O(1)#1	152.66(13)	O(2)-Mn(2)-O(2)#1	155.75(15)
O(1)-Mn(1)-N(1)	82.98(10)	O(2)-Mn(2)-N(2)	82.57(9)
O(1)#1-Mn(1)-N(1)	92.34(10)	O(2)#1-Mn(2)-N(2)	93.92(9)
O(1)-Mn(1)-N(1)#1	92.34(10)	O(2)-Mn(2)-N(2)#1	93.92(9)
O(1)#1-Mn(1)-N(1)#1	82.98(10)	O(2)#1-Mn(2)-N(2)#1	82.58(9)
N(1)-Mn(1)-N(1)#1	160.16(13)	N(2)-Mn(2)-N(2)#1	163.33(16)
O(1)-Mn(1)-Br(1)	103.67(7)	O(2)-Mn(2)-Br(2)	102.12(8)
O(1)#1-Mn(1)-Br(1)	103.67(7)	O(2)#1-Mn(2)-Br(2)	102.12(8)
N(1)-Mn(1)-Br(1)	99.92(6)	N(2)-Mn(2)-Br(2)	98.33(8)
N(1)#1-Mn(1)-Br(1)	99.92(6)	N(2)#1-Mn(2)-Br(2)	98.33(8)

, which shows that the equatorial metal–ligand bond lengths were from 1.846 to 2.003 Å, while the axial Mn–Br bond was 2.5853(7) Å [29,30]. This character clearly indicated a z-out form of the Jahn–Teller distortion of Mn(III) ion. Both the bond valence sum (BVS) calculation (see supporting information) and the charge balance supported our charge assignments. We know that the Mn(III) ions have the same electron configuration as Cr(II), in which the Cr–Cr quadrupole bonds are seen in face-to-face forms of many chromium dimers [31,32]. Nevertheless, the metal–metal distances were maintained at a distance of 3.69 Å from each other, which prevented significant metal–metal bond formation. We therefore became curious about the metal–metal interactions at such a distance. In the discussion below, we conclude that, instead of metal–metal bonds, there was strong antiferromagnetic coupling between these two Mn(III) ions.

Compound 2 crystallized in a monoclinic C2/c space group. X-ray refinement data for the complex are also listed in Table 1. The structural plot of complex 2 is displayed in Figure 2. Two manganese ions were held by two double deprotonated ligands with one bound by four alkoxy-groups and two pyridines, the other bound by two bridging alkoxides, two pyridines, and coordinates to two water molecules. Both of the manganese ions were six coordinated, but their charges were quite different. The Mn(1) ion had two metal–ligand bonds with distances of 1.85 Å, two with distances of 1.91 Å, and bound with pyridine at a distance of 1.99 Å. Since all six metal–ligand bonds were short, we assigned this atom as an Mn(IV) ion. In contrast, Mn(2) bound with a bridging alkoxide at a distance of 2.13 Å, with two water molecules with a distance of 2.18 Å and two pyridines at 2.31 Å. These metal–ligand bonds were obviously long and were therefore assigned as an Mn(II) ion. (See

Table 3. Selected bond distances (Å) and angles (deg) of compound 2.

Mn(1)-O(3)#1	2.1223(10)	Mn(2)-O(2)#1	1.8435(9)
Mn(1)-O(3)	2.1223(10)	Mn(2)-O(2)	1.8436(9)
Mn(1)-O(1)	2.1832(9)	Mn(2)-O(1)	1.9049(8)
Mn(1)-O(1)#1	2.1832(9)	Mn(2)-O(1)#1	1.9049(8)
Mn(1)-N(1)	2.3073(11)	Mn(2)-N(2)#1	1.9862(11)
Mn(1)-N(1)#1	2.3073(11)	Mn(2)-N(2)	1.9862(11)
O(3)#1-Mn(1)-O(3)	90.13(6)	O(1)-Mn(2)-O(1)#1	85.63(5)
O(3)#1-Mn(1)-O(1)	102.84(4)	O(2)#1-Mn(2)-N(2)#1	80.90(4)
O(3)-Mn(1)-O(1)	156.24(4)	O(2)-Mn(2)-N(2)#1	93.45(4)
O(3)#1-Mn(1)-O(1)#1	156.24(4)	O(1)-Mn(2)-N(2)#1	92.28(4)
O(3)-Mn(1)-O(1)#1	102.84(4)	O(1)#1-Mn(2)-N(2)#1	93.67(4)
O(1)-Mn(1)-O(1)#1	72.74(4)	O(2)#1-Mn(2)-N(2)	93.45(4)
O(3)#1-Mn(1)-N(1)	99.60(4)	O(2)-Mn(2)-N(2)	80.90(4)
O(3)-Mn(1)-N(1)	86.97(4)	O(1)-Mn(2)-N(2)	93.67(4)
O(1)-Mn(1)-N(1)	71.43(4)	O(1)#1-Mn(2)-N(2)	92.28(4)
O(1)#1-Mn(1)-N(1)	100.88(4)	N(2)#1-Mn(2)-N(2)	171.90(6)
O(3)#1-Mn(1)-N(1)#1	86.96(4)	O(2)#1-Mn(2)-Mn(1)	133.89(3)
O(3)-Mn(1)-N(1)#1	99.60(4)	O(2)-Mn(2)-Mn(1)	133.89(3)
O(1)-Mn(1)-N(1)#1	100.87(4)	O(1)-Mn(2)-Mn(1)	42.82(3)
O(1)#1-Mn(1)-N(1)#1	71.43(4)	O(1)#1-Mn(2)-Mn(1)	42.82(3)
N(1)-Mn(1)-N(1)#1	170.75(6)	N(2)#1-Mn(2)-Mn(1)	94.05(3)
O(3)#1-Mn(1)-Mn(2)	134.94(3)	N(2)-Mn(2)-Mn(1)	94.05(3)
O(3)-Mn(1)-Mn(2)	134.94(3)	C(6)-O(1)-Mn(2)	127.13(7)
O(1)-Mn(1)-Mn(2)	36.37(2)	C(6)-O(1)-Mn(1)	115.89(7)
O(1)#1-Mn(1)-Mn(2)	36.37(2)	Mn(2)-O(1)-Mn(1)	100.81(4)
N(1)-Mn(1)-Mn(2)	85.38(3)	C(8)-O(2)-Mn(2)	111.86(7)
N(1)#1-Mn(1)-Mn(2)	85.38(3)	C(5)-N(1)-C(1)	118.48(11)
O(2)#1-Mn(2)-O(2)	92.22(5)	C(5)-N(1)-Mn(1)	114.76(8)
O(2)#1-Mn(2)-O(1)	172.42(4)	C(1)-N(1)-Mn(1)	125.45(9)
O(2)-Mn(2)-O(1)	91.48(4)	C(13)-N(2)-C(9)	121.39(11)
O(2)#1-Mn(2)-O(1)#1	91.48(4)	C(13)-N(2)-Mn(2)	126.02(9)
O(2)-Mn(2)-O(1)#1	172.42(4)	C(9)-N(2)-Mn(2)	112.58(8)

) This assignment was also supported by bond-valence-sum (BVS) calculations. As described below, we concluded that, as the half filled d orbital in the Mn(II) ion overlapped the empty e_g orbital of the neighbor Mn(IV) ion, this actually triggered the ferromagnetic coupling of compound 2.

To further confirm the assignments of the charges for the metal ions, BVS [33] approximations were carried on both compounds 1 and 2. The calculations were based on Equation (1), where the parameter B was an experience constant with the value 0.37 and the r₀ values were obtained from literature reports and are listed in

Table 4. r₀’s value for Mn with different charges bonds to O, N, and Br atoms.

MnII-O	MnII-N	MnII-Br
1.765	1.849	2.34
MnIII-O	MnIII-N	MnIII-Br
1.732	1.837	2.315
MnIV-O	MnIV-N	MnIV-Br
1.75	1.822	2.41

. The metal–ligand bond lengths listed in Table 2 and Table 3 were then substituted into Equation (1), which indicated that both manganese ions in compound 1 were Mn(III) ions, whereas the metal centers in compound 2 were Mn(II) and Mn(IV) ions [34,35,36,37].

$$\mathrm{V}={\displaystyle \sum }_i{S}_i={\displaystyle \sum }_i\exp (\frac{r_0-r}{B})$$

(1)

Powder X-ray Diffraction Patterns: To demonstrate the integrity of the bulk samples of compounds 1 and 2, series powder X-ray diffraction experiments were performed. Powder X-ray diffraction patterns were collected at the Taiwan Photon Source (TPS) of the National Synchrotron Radiation Research Center (NSRRC). The powder X-ray diffraction pattern of compounds 1 and 2 are shown in Figure 3. The simulated powder patterns of the single crystal data were also plotted together for comparison. The experimental synchrotron radiation X-ray powder diffraction patterns of both compounds 1 and 2 were very close to the single crystal simulation results. This result clearly showed that the bulk samples had good integrity.

2.2. Magnetic Properties

The magnetic properties of both compounds 1 and 2 could be described as follows: magnetic susceptibility measurements were carried on both compound 1 and compound 2. Polycrystalline samples restrained in eicosane were measured under a magnetic field of 1000 G in the temperature range 2–300 K. Figure 4 illustrates the χ_MT vs. T plot of compound 1. The χ_MT value was 5.69 cm³K mol⁻¹, which was somewhat smaller than the spin only value of 6.0 cm³K mol⁻¹ for two uncoupled Mn(III) ions. This could be attributed to antiferromagnetic coupling, even at room temperature. The χ_MT value slowly decreased to 3.76 cm³K mol⁻¹ at 50 K and then decreased very rapidly to 0.17 cm³K mol⁻¹ upon further cooling to 2 K. The magnetic behaviors were then interpreted by the Kambe model [38] with:

H = gβB⋅(S₁ + S₂) − 2J (S₁·S₂)

(2)

where β is Bohr magneton, B is the magnetic field, and both S₁ and S₂ have the values of 2. The best fitting is presented by the solid red line in Figure 4. The parameters obtained by the fitting gave g = 2.0 and J = −5.0 K, which were in good agreement with the properties of the two Mn(III) ions. Interestingly, the electron configuration of Mn(III) was high spin d⁴, which is isoelectronic to Cr(II). By experience, two Cr(II)s in a capped form tend to build up a metal–metal quadrupole bond [39]. However, in our case, it was much more difficult for the Mn(III) to form a metal–metal bond than Cr(II) due to Mn(III) being a harder Lewis acid, whereas the Cr(II) is relatively soft. In addition, in compound 1, the two Mn(III) ions were separated by a large distance (3.69 Å) by the binding site of ligands. All of these factors reduced the chance to form an actual metal–metal bond, but strong antiferromagnetic couplings were clearly observed.

Figure 5 shows a χ_MT vs. T plot for compound 2. The χ_MT value was 6.39 cm³K mol⁻¹, which was slightly higher than the spin only value of 6.25 cm³K mol⁻¹ for one uncoupled Mn(IV) and one uncoupled Mn(II). This behavior suggested that ferromagnetic coupling occurred, even at room temperature. The χ_MT value then increased to 7.98 cm³K mol⁻¹ at 15 K and then decreased to 3.58 cm³K mol⁻¹ at 2 K. Such behavior is usually attributed to zero-field splitting or intermolecular antiferromagnetic coupling. We then collected magnetic susceptibility data and fitting by utilizing the Kambe model to account for Mn(IV)-Mn(II) above 15 K. The fitting for this appears as a solid red line in Figure 5. The best fitting clearly indicated a ferromagnetic coupling J = +1.8 K, g = 2.0. Judging from the structure, we know there were two unpaired electrons located in the e_g orbitals (d_x2-y2 and d_z2) within the Mn(II) center, whereas these orbitals were empty in the Mn(IV) center. When these two metal ions produced bridging through the x and the y axis, as shown in Scheme 2, the occupied e_g orbitals in Mn(II) right overlapped with the empty e_g orbitals in Mn(IV), which resulted in ferromagnetic coupling between these two ions.

To further examine the spin number of the molecule, reduced magnetization measurements were carried on the polycrystalline sample of compound 2 in the temperature range 2~4 K under a magnetic field 2~6 Tesla. The results are shown in Figure 6, where the solid red lines represent the best fitting based on the parameters g = 1.95, S = 4, and D = −0.78 K. The spin value of S = 4 saturated under high-field and low temperatures further confirmed the existence of ferromagnetic coupling between the Mn(II) and Mn(IV) ions.

The alternating current (AC) magnetic susceptibility measurements were also carried out on both compounds 1 and 2 under 1000 Hz of a 3.5 G AC magnetic field in the temperature range 1.9~10 K. Whether a 0 G or a 500 G direct current (DC) magnetic field was applied, no significant out-of-phase signals were detected. This measurement further confirmed that neither 1 nor 2 showed single-molecule magnet (SMM) behavior.

3. Experimental Methods

All chemicals used in this study were of commercial grade and were used without further purification.

3.1. Ligand Synthesis

To a two-neck 250 mL round bottom flask charged with magnesium powder (7.29 g, 300 mmol), 120 mL of anhydrous ethyl-ether was added under a nitrogen atmosphere. Methyl iodide (64 g in 450 mmol) was slowly added over a period of 2~3 h to form the Grignard reagent. Then, 2-Acetylpyridine (18.18 g in 150 mmol) was added to the reactant solution, and the temperature was controlled at –8 °C The reaction was carried out for 5 h and then quenched by adding ice water in an ice bath. Without neutralization, the mixture was extracted by 200 mL ethyl acetate five times. The organic layer was dried over anhydrous MgSO₄. After evaporating the organic solvent, the dmhmp-H ligand was removed by distillation under reduced pressure (yield about 70%). The residues were then separated by flash column chromatography eluted with ethyl acetate/hexane (1:1) solution to obtain 1.7% rD, and 1.8% mD was then obtained by eluting pure ethyl acetate.

rD:¹H NMR (300 MHz, CDCl₃) δ 1.17 (s, 6H, CH₃), δ 2.55 (s, 2H HC-H), δ 6.24 (br, 2H, OH), δ 7.18 (qd, 2H, Py-H), δ 7.61 (td, 2H, Py-H), δ 7.69 (td, 2H, Py-H), δ 8.50 (m, 2H, Py-H). ¹³C NMR CDCl₃: δ 166.31, δ 147.34, δ 137.05, δ 121.68, δ 119.44, δ 75.58, δ 50.97, δ 30.56. HRMS (FT-ICR): [C₁₅H₁₈N₂O₂]H⁺: 259.1441. Found 259.1438.

mD:¹H NMR (300 MHz, CDCl₃) δ 1.46 (s, 6H, CH₃), δ 2.54; 3.02 (d, HC-H), δ 6.26 (br, 2H, OH), δ 6.78 (qd, 2H, Py-H), δ 7.16 (td, 2H, Py-H), δ 7.31 (td, 2H, Py-H), δ 8.03 (qd, 2H, Py-H). ¹³C NMR CDCl₃: δ 164.16, δ 146.66, δ 135.91, δ 120.73, δ 119.89, δ 75.55, δ 50.32, δ 32.68. HRMS (FT-ICR): [C₁₅H₁₈N₂O₂]H⁺: 259.1441. Found 259.1436.

3.2. Materials Synthesis

[Mn₂(rD)₂Br₂] (1) MnBr₂·4H₂O (0.52 g, 1.85 mmol) and ligand rD (0.45 g, 0.88 mmol) were added to 20 mL of acetonitrile. To the mixture, 25% NMe₄OH (0.53 g, 1.46 mmol) was added. The resulting solution was stirred 1~2 h and then passed through a filter. The filtrate was allowed to stand 2~4 days, and dark green crystals suitable for X-ray crystallography analysis were obtained (yield: 80%). Elemental Analysis: (C₃₀H₃₂N₄O₄Mn₂Br₂), Found (Calc.):C, 45.87% (46.06%), H, 4.08% (4.12%), N, 7.37% (7.16%). IR (cm⁻¹): 3448 (b), 3079 (s), 2971 (s), 2942 (w), 2921 (w), 1607 (s), 1475 (s), 1181 (s), 1094 (s), 1052 (s), 943 (s), 788 (s), 685 (s), 616 (s), 552 (w), 516 (w).

[Mn₂(mD)₂(H₂O)₂]Br₂(H₂O)₄ (2) The same reaction conditions that were used for the preparation of compound (1) were used here. MnBr₂ 4H₂O (0.52 g, 1.85 mmol) and the ligand mD (0.45 g, 0.88 mmol) were added to 20 mL acetonitrile. Then, 25% NMe₄OH (0.53 g, 1.46 mmol) was added to the mixture. The resulting solution was stirred for 1~2 h and then passed through a filter. The filtrate was allowed to stand for 2~4 days, whereupon pink crystals suitable for X-ray crystallographic analysis were formed (yield: 80%). Elemental Analysis: (C₃₀H₄₄N₄O₁₀Mn₂Br₂), Found (Calc.):C, 40.12% (40.46%), H, 4.73% (4.98%), N, 6.19% (6.29%). IR (cm⁻¹): 3373 (b), 3068 (b), 1596 (s), 1476 (s), 1435 (s), 1182 (s), 1114 (s), 1097 (s), 1045 (s), 929 (s), 863 (s), 780 (s), 759 (s), 650 (s), 606 (s), 579 (w), 549 (w), 513 (w).

3.3. Physical Property Measurements

NMR spectra were measured on a Bruker AV-300 spectrometer. Infrared spectra were collected using a Perkin Elmer 1600 spectrometer using KBr pellets in the range of 500~4000 cm−¹.

DC magnetic susceptibility data were conducted by using a Quantum Design MPMS7 system. Eicosane was employed to restrain the sample to prevent torqueing. The background was corrected by a gel cap charged with eicosane. The diamagnetic correction was estimated by Pascal’s constants [40,41].

Elemental analyses were performed by an elemental vario EL cube. All of the measurements regarding magnetic properties and elemental analysis were carried out at the National Taiwan University Instrument Centre, College of Science.

For X-ray crystallography, single-crystal X-ray diffraction data collection was carried out on a Bruker D8 VENTURE CCD diffractometer equipped by Mo, λ = 0.71073 Å as light source. The temperatures were controlled at 150(2) K using an Oxford Cryosystems Cooler. The absorption correction was done by using the SADABS [42] (Bruker 2016) program, which is based on symmetry-equivalent reflections. The structures were solved by direct methods and refined with a full-matrix least-squares technique within the Shelxs-2018 [43] program and refined by the Shelxl-2018 program. [44] All non-hydrogen atoms were refined anisotropically. The hydrogen atoms were set in calculated positions and refined using the riding model. The refinement parameters are summarized in Table 1. CCDC numbers: 1922174 for 1; 1922175 for 2.

4. Conclusions

In this paper, we report on the preparation of two 2,4-di-2-pyridyl-2,4-pentanediol (rD and mD) ligands, which were then used to prepare two Mn₂ complexes: [Mn₂(rD)₂Br₂] (1) and [Mn₂(mD)₂(H₂O)₂]Br₂ (2). In compound 1, the rD ligand bound two Mn(III) ions in a hamburger form. Both of the Mn(III) units were five coordinated and approached each other through the flat vacancy site of the square-pyramidal. However, this did not lead to the formation of any metal–metal bond between these two ions. Upon a magnetic property examination, antiferromagnetic coupling was observed in compound 1. In compound 2, the half filled e_g orbitals in Mn(II) were significantly overlapped with the empty e_g orbitals in Mn(IV). This interaction significantly triggered the ferromagnetic coupling of compound 2, which led to a molecular spin S = 4 and D = −0.78 K. However, this property did not lead to SMM behavior for compound 2. This manuscript reports not only on a systematic approach for producing both rD and mD ligands but also on their use in forming complexes with novel structures and magnetic properties.