S = 3 Ground State for a Tetranuclear Mn^IV₄O₄ Complex Mimicking the S₃ state of the Oxygen Evolving Complex

Abstract

The S₃ state is currently the last observable intermediate prior to O–O bond formation at the oxygen evolving complex (OEC) of Photosystem II, and its electronic structure has been assigned to a homovalent Mn^IV₄ core with an S = 3 ground state. While structural interpretations based on the EPR spectroscopic features of the S₃ state provide valuable mechanistic insight, corresponding synthetic and spectroscopic studies on tetranuclear complexes mirroring the Mn oxidation states of the S₃ state remain rare. Herein, we report the synthesis and characterization by XAS and multifrequency EPR spectroscopy of a Mn^IV₄O₄ cuboidal complex as a spectroscopic model of the S₃ state. Results show that this Mn^IV₄O₄ complex has an S = 3 ground state with isotropic ⁵⁵Mn hyperfine coupling constants of −75, −88, −91, and 66 MHz. These parameters are consistent with an αααβ spin topology approaching the trimer-monomer magnetic coupling model of pseudo-octahedral Mn^IV centers. Importantly, the spin ground state changes from S = 1/2 to S = 3 as the OEC is oxidized from the S₂ state to the S₃ state. This same spin state change is observed following the oxidation of the previously reported Mn^IIIMn^IV₃O₄ cuboidal complex to the Mn^IV₄O₄ complex described here. This sets a synthetic precedent for the observed low-spin to high-spin conversion in the OEC.

Geaphical Abstract

INTRODUCTION

Mechanistic studies of biological water oxidation at the oxygen evolving complex (OEC) of Photosystem II (PSII) are performed in the context of the Kok cycle of S_n (n = 0−4) states.^1–4 Starting from the dark stable S₁ state, sequential light-induced one electron oxidations lead to the progression to higher S_n states, resulting in the formation of the S₃ state, the last observable intermediate prior to dioxygen formation.^5–6 Involving a series of elementary steps that include H⁺ transfer, substrate H₂O binding, and e⁻ transfer, the S₂→S₃ transition represents a critical, sensitive step in the catalytic cycle of the OEC, as discrete changes to the OEC precede the formation of the S₃ state.^7–8 Despite being the subject of extensive biochemical, structural, spectroscopic, and computational studies, the (electronic) structure and the mechanism of formation of the S₃ state remain largely unknown.^7–23 To obtain a better understanding of the properties of the S₃ state, systematic structure-function (property) studies on relevant model complexes are necessary. Despite significant efforts to prepare tetra- and pentanuclear complexes as models of the OEC, relevant complexes in terms of structure, redox state, spectroscopy, and reactivity are rare.^24–38

On the basis of EPR, MCD, and X-ray spectroscopic studies, the electronic structure of the S₃ state has been assigned to a homovalent Mn^IV₄ core with an S = 3 spin ground state.^{5, 39–41} Two structural isomers S₃^A (dimer-of-dimers) and S₃^B (trimer-monomer), both with S = 3 spin ground states, have been invoked for the S₃ state (Figure 1a).^{5, 42} A similar structural isomerism has been proposed for the S₂ state.^{24, 43} Such proposed structural changes may lead to differences in the sign and magnitude of the magnetic exchange interactions (J_ij) between adjacent Mn centers, which in turn affect not only the spin ground state of the cluster but also the observed sign and magnitude of the projected ⁵⁵Mn hyperfine interactions (A_i).^43–44 The observed A_i for Mn ions in the S₃ state have been accommodated with the calculated J_ij for the S₃^A structure.⁵ The high-resolution (2.04 Å) structure of the S₃ state obtained at room temperature using femtosecond X-ray free electron laser (XFEL) techniques is similar to the S₃^A structure.¹⁰ Further improvements in resolution and different sample conditions need to be addressed to identify possible contributions from structural isomers such as the proposed S₃^B state.^{10, 45}

Figure 1.

a) Proposed isomers S₃^A and S₃^B of the inorganic CaMn₄O₅(OH) core of the S₃ state of the OEC. b) Representative examples of Mn^IV₄ complexes and their corresponding spin ground states.

Comparative studies on structurally related Mn^IIIMn₃^IV and Mn^IV₄ complexes as models of the S₂ and S₃ states may prove beneficial, but due to synthetic difficulties, such studies have been reported for only two classes of tetranuclear Mn complexes: linear and adamantane-shaped complexes, both featuring [Mn₄O₆]ⁿ⁺ cores (Figure 1b).^46–52 For the linear [Mn^IV₄O₆(bpy)₆]⁴⁺ complex, strong pairwise antiferromagnetic interactions lead to a diamagnetic S = 0 ground state;^{46, 51} reduction to the [Mn^IIIMn^IV₃O₆(bpy)₆]³⁺ complex by radiolysis leads to an S = 1/2 ground state, featuring the characteristic multiline EPR signal at g = 2.⁵³ For the adamantane-shaped [Mn^IV₄O₆(tacn)₄]⁴⁺ complex and related ligand substitution series, magnetic susceptibility studies indicate overall ferromagnetic interactions giving rise to an S = 6 ground state.⁴⁷ EPR studies for the adamantane-shaped complexes have not been reported. An S = 5/2 ground state has been reported for a reduced adamantane-shaped complex featuring a [Mn^IIIMn^IV₃O₆]³⁺ core.⁵⁴ While both reduced linear (S = 1/2) and adamantane-shaped (S = 5/2) Mn^IIIMn^IV₃ complexes serve as spectroscopic models of the S₂ state, with two possible spin ground states S = 1/2 and S = 5/2, a Mn^IV₄ complex with an S = 3 ground state mimicking the S₃ state has not been reported.

Herein, we report the synthesis and characterization by XAS and multifrequency EPR spectroscopies of a unique Mn^IV₄O₄ cuboidal complex as a spectroscopic model of the S₃ state. Results show that the Mn^IV₄O₄ complex has an S = 3 ground state with isotropic ⁵⁵Mn hyperfine coupling constants −75, −88, −91, and 66 MHz. These parameters are consistent with an αααβ spin topology approaching the trimer-monomer magnetic coupling model of pseudo-octahedral Mn^IV centers. Importantly, the spin state change from S = 1/2 to S = 3, characteristic of the S₂→S₃ transition in the OEC, is the same spin state change observed in the oxidation of the previously reported Mn^IIIMn₃^IVO₄ cuboidal complex to the Mn^IV₄O₄ complex, providing the first synthetic precedent for the observed low-spin to high-spin transition in the OEC.

RESULTS

Synthesis.

The diamidate-bridged cuboidal complex LMn^III₂Mn^IV₂O₄(diam)(OAc) (1) was used as a precursor for the targeted Mn^IV₄O₄ complex (Scheme 1).²⁴ In propylene carbonate, the cyclic voltammogram of 1 shows a reversible oxidation to the previously characterized one electron oxidized Mn^IIIMn₃^IVO₄ complex (2) at −50 mV vs Fc/Fc⁺.²⁴ A second quasi-reversible Mn^IIIMn₃^IV/Mn^IV₄ couple is observed at +780 mV vs Fc/Fc⁺ (Figure S3). Notably, formation of the Mn^IV₄ species is not observed for analogous tris-acetate or tris-phosphinate complexes, highlighting the ability of amidate ligands in supporting high oxidation state complexes.⁵⁵ Accordingly, treatment of 1 with an optimal amount of tris(2,4-dibromophenyl)aminium hexachloroantimonate (E_1/2 = +1140 mV vs Fc/Fc⁺ in MeCN)^56–57 in thawing MeCN leads to the formation of the two electron oxidized Mn^IV₄O₄ complex (3). Addition of fewer equivalents of the aminium oxidant leads to a mixture of 2 and 3 by ¹H NMR (Figure S1), suggesting that 3 is related to 2 by a one electron oxidation. While the ESI-MS of 2 shows only one peak at m/z = 1354 consistent with the mass of [LMn₄O₄(diam)(OAc)]⁺, the ESI-MS of 3 shows an additional, major peak at m/z = 677 consistent with the mass of [LMn₄O₄(diam)(OAc)]²⁺ (Figure S2), supporting the [LMn^IV₄O₄(diam)(OAc)][SbCl₆]₂ formulation of 3. Overall, complexes 1, 2, and 3 represent a rare redox series of essentially isostructural compounds in which the Mn oxidations states of each compound mirror those in the S₁, S₂, and S₃ states, respectively.

Scheme 1.

Synthesis of complexes 1~3 studied in this work.

X-ray spectroscopy.

Frozen solution Mn K-edge X-ray absorption near-edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS) were used to further characterize metal oxidation states and to provide evidence of structural similarity between 2 and 3 in solution (Figure 2). Absorption edge energies were determined from the second-derivative zero-crossings, giving the following values (eV): 6552.4 (2) and 6553.2 (3). The positive shift in the edge energy by 0.8 eV is consistent with a Mn centered oxidation, supporting the Mn^IV₄ assignment for 3. In comparison, the edge energy of the S₂ and S₃ states of cyanobacteria PSII are 6554.1 and 6554.4 eV, respectively.⁵⁸ The smaller edge energy shift in the S₂→S₃ transition in PSII could be due to changes in Mn coordination sphere, such as binding of substrate water.¹⁰ The k³-weighted EXAFS spectra of 2 and 3 are nearly superimposable (Figure 2b, see SI for fit). A similar set of parameters (Table S1) were used to fit the EXAFS spectra of 2 and 3, and results show that within error, average Mn-ligand and Mn-Mn distances are similar between 2 and 3. While small geometrical changes in 3 are not resolved by EXAFS, such as the contraction of a specific Mn-oxo distance (by c.a. 0.2 Å) upon oxidation from Mn^III₂Mn^IV₂ (1) to Mn^IIIMn^IV₃ (2), indicative of the Jahn-Teller-like effect in pseudo-octahedral Mn^III centers, the X-ray spectroscopic data support the structural integrity of 3 as a cuboidal Mn^IV₄O₄ cluster.

Figure 2.

Mn K-edge XAS spectra for complexes 2 (black lines) and 3 (red dotted lines). a) Normalized XANES data, b) k³-weighted EXAFS data, and c) Fourier transforms of k³-weighted EXAFS data.

EPR spectroscopy.

To obtain a better understanding of the spin ground state and the nature of the magnetic coupling interactions between Mn centers, D-band (130 GHz) EPR studies were conducted on frozen solution samples of 3 (Figure 3). Spanning approximately 4 T, the spectrum features five resolved lines (inflection points; see Figure S6 for pseudomodulated spectrum) at 3.4, 4, 5.2, 6, and 6.9 T, with the central feature at 4.63 T (g = 2.01) overlapping with a Mn^II impurity vide infra. On the basis of the rising edge energy of the XANES spectrum for complex 3 (Figure 2a), the average Mn oxidation state in the sample is increasing, with a small Mn^II impurity contributing to the EPR signal at the temperatures studied. EDNMR at the central 4.63 T feature (Figure S8) resolves NMR transitions at approximately 302, 417, and 544 MHz, characteristic of Mn^II (S = 5/2) with A_iso = 240 MHz. The Mn^II feature is simulated with ZFS parameters D and E/D of 0.06 cm⁻¹ and 0.167 respectively, in line with reported spin Hamiltonian parameters of Mn^II.^59–60 Excluding this central feature, the large spectral breadth of the EPR envelope is indicative of a complex with a large zero-field splitting (ZFS) parameter, while the distribution of the EPR spectrum is suggestive of a positive ZFS parameter and small rhombicity.

Figure 3.

D-band EPR spectrum of complex 3 collected at 1.6 K. Spectral data are shown in black and the simulated spectrum is represented by the dashed red line and other colored traces. The simulated Mn(II) contribution is shown in light blue and the m_S transitions that contribute to the complex 3 simulation are shown in blue, purple, and orange. Simulation parameters: S = 3, g = 1.97, D = 0.4 cm⁻¹, E/D = 0.1. (Inset) Electron-spin nutation curves of complex 3 (blue) collected at 7 T and BDPA (black).

Electron spin nutation experiments can be used to assign the spin ground state of EPR active compounds.^{5, 61} A variable microwave preparation pulse is applied and followed by a Hahn echo sequence. This preparation pulse nutates the electron spin, causing the magnitude of the observed spin echo to oscillate. The period of this oscillation is inversely proportional to the magnitude of the effective spin. Electron spin nutation experiments on the high-field edge of the EPR envelope (7 T) were performed to assign the spin state of 3 (inset of Figure 3). An S = 1/2 spin standard is needed to assign the spin state of the measured EPR transition: α,γ-bisdiphenylene-β-phenylallyl (BDPA), an S = 1/2 radical, was used to this effect. The nutation period of 3 is shorter than that of BDPA by approximately a factor of 2.8. This value is in good agreement with the expected factor of √6 ≈ 2.45 for the |−3⟩→|−2⟩ transition in an S = 3 spin system.⁵ The slight discrepancy likely arises because the external spin standard BDPA is measured separately from 3. The conversion factor, the difference between the incident power and the power at the sample, cannot be assumed to be identical from sample to sample as is the case of the internal Y_D radical species present in PSII samples poised in the S₃ state.⁵ Despite the slight discrepancy in the expected nutation period for 3, an intense feature at g = 12 in the parallel mode X-band EPR spectrum of 3 (Figure 4) supports the S = 3 spin state assignment. The g = 12 signal is reminiscent of the spectrum observed for the S₃ state of T. Elongatus PSII and other S = 3 systems.^{5–6, 62}

Figure 4.

Parallel mode X-band CW-EPR of complex 3. Spectral data are shown in black and the simulated spectrum is represented by the dashed red line. Simulation parameters: S = 3, g = 1.97, D = 0.4 cm⁻¹, E/D = 0.1.

Spin Hamiltonian simulations of 3 were performed using an S = 3 spin state and an isotropic g-value of g = 1.97. In higher spin systems (S > 1/2), the breadth of EPR spectra depends largely on the axial ZFS parameter D, which removes the degeneracy of the |±m_s⟩ states; following the electron Zeeman effect, the magnitude of D determines the magnetic field position of a given EPR transition (Figure S5). Based on the breadth of the EPR envelope, the ZFS parameter in 3 was estimated to be D = +0.4 cm⁻¹. The spacing of the turning points (individual m_S levels) is dictated by the ratio of E/D, where E represents the rhombicity of the ZFS. Based on the observed distribution, E/D was estimated to be 0.1. The observed ZFS in 3 is larger than that of the S₃ state of T. Elongatus and chemically modified T. Elongatus PSII (|D| = 0.175 and 0.281 cm⁻¹, respectively)^{5, 63} and that of the biomimetic CaMn₃^IVO₄ model (|D| = 0.068 cm⁻¹)³⁶. Monomeric octahedral Mn^IV compounds are useful benchmarks for interpreting spin Hamiltonian parameters of more complex systems such as the S₃ state.⁵ Typical ZFS values for monomeric octahedral Mn^IV compounds are relatively small, on the order of 0.2 cm⁻¹, due to the near spherical symmetry of the ($t_{2g}^3{e}_g^0$) electronic configuration,^64–65 but values in the range of |D| = 0.17~2.3 cm⁻¹ have been reported for six-coordinate mononuclear Mn^IV complexes.^66–68

EDNMR spectroscopy.

To gain a better understanding of the ⁵⁵Mn hyperfine interactions in 3, electron-electron double resonance detected NMR (EDNMR) spectra were collected at selected field positions within the D-band EPR spectrum (Figure 5). EDNMR utilizes a high-turning angle (HTA) pulse to excite formally forbidden transitions (Δm_S = ±1, Δm_I = ±1, ±2, etc.), which can be then detected via a Hahn echo sequence.^69–70 The HTA pulse is of a different excitation frequency than pulses in the Hahn echo sequence. When the difference in these frequencies corresponds to a nuclear transition frequency, a decrease in the Hahn echo is observed, which is interpreted as a peak in Figure 5. As with electron nuclear double resonance (ENDOR), EDNMR allows for the detection of nuclear transitions and has been shown to be an important spectroscopic tool at high microwave frequencies.⁷¹ The field dependent EDNMR spectra of 3 display a number of transitions ranging from 10 to 320 MHz (Figure 5). These features arise from the hyperfine coupling interaction A of the ⁵⁵Mn nuclei (I = 5/2) and the S = 3 electron spin of complex 3. The positions of these transitions are dictated by A and the Larmor frequency (ν_n) of the ⁵⁵Mn nuclei. Depending on the specific |m_s⟩ level being pumped, features in the field-dependent EDNMR are observed at multiples of |A| ± ν_n (see SI). Spin Hamiltonian simulations of the ⁵⁵Mn EDNMR are shown in the colored traces in Figure 5, and the corresponding A values are listed in Table 1. One positive and three negative A values are needed to accurately simulate the EDNMR spectra. This is evidenced by the field-dependent shift of the peaks at 200 (Mn2, blue peaks, Figure 5) and 150 MHz at 7 T (Mn4, cyan peaks, Figure 5) to higher frequencies with decreasing magnetic field, indicating that both Mn2 and Mn4 have negative A values. The shift of the peak at 250 MHz (Mn1, yellow peaks, Figure 5) at 7 T to lower frequencies with decreasing magnetic field indicates that Mn1 has a positive A value. To obtain |A|, for example, the yellow peak at 270 MHz at 7 T corresponds to |A| = 65 MHz according to the formula n|A| + ν_n, where n is 3 for the m_S = |−3⟩ to |−2⟩ manifold and ν_n is 74 MHz (ν_n(⁵⁵Mn) varies linearly with magnetic field as 10.554 MHz/T).

Figure 5.

(Left) EPR spectrum of 3 showing the field positions (*) where EDNMR spectra were collected. (Right) Field dependent EDNMR of 3. Experimental data are shown in black traces. Spectral simulations are shown in the colored traces. The simulated hyperfine values are listed in Table 1.

Table 1.

Isotropic ⁵⁵Mn hyperfine interactions (MHz) in complex 3, different models of the S₃ state^{5, 63}, and the CaMn₃^IVO₄ model complex³⁶.

Projected hyperfine interactions (MHz)
	Mn1	Mn2	Mn3	Mn4
3	66	−91	−88	−75
S3MeOH a	62	−102	−99	−95
CaMn3IVO4 model,b		−87	−87	−76
S3B c	86	−75	−79	−65
S3A d	−99	−96	−26 or 7	≤ 5
On-site hyperfine interactions (MHz)
3e	−178	−194	−187	−174
S3MeOH e	−166	−222	−216	−207
CaMn3IVO4 modelf		−185	−185	−179
S3B g	−232	−161	−169	−148
Spin projection factors ρ
Trimer-monomer modelh	−0.37	0.47	0.47	0.43

^aReported ⁵⁵Mn hyperfine interactions in the MeOH-modified form of the S₃ state.⁶³

^bProjected ⁵⁵Mn hyperfine interactions for the CaMn₃^IVO₄ model complex using the trimer-monomer model spin projection factors.³⁶

^cProjected ⁵⁵Mn hyperfine interactions in the calculated trimer-monomer S₃^B model of the S₃ state.⁵

^dReported ⁵⁵Mn hyperfine interactions for the S₃^A form of the S₃ state.⁵

^eEstimated on-site ⁵⁵Mn hyperfine interactions assuming a trimer-monomer model.⁶³

^fReported ⁵⁵Mn hyperfine interactions in an S = 9/2 CaMn₃^IVO₄ model complex.³⁶

^gCalculated on-site ⁵⁵Mn hyperfine interactions in the trimer-monomer S₃^B model of the S₃ state.⁵

^hRef 5.

Typical Mn^IV ions in octahedral coordination environments exhibit little ⁵⁵Mn hyperfine anisotropy by virtue of the spherical symmetry of the half-filled t_2g set of d-orbitals, and thus produce sharp ⁵⁵Mn NMR lines.⁵ The EDNMR of 3 is complicated by multiple overlapping transitions, but at fields greater than 5 T, distinct, sharp transitions are observable, characteristic of an all-octahedral Mn^IV₄ complex. Due to the relative breadth of the EDNMR peaks (e.g. FWHM = 34 MHz for the feature centered at 197 MHz in the EDNMR trace recorded at 7 T) it is difficult to precisely determine the hyperfine anisotropy of all the individual Mn^IV ions. However, inspection of transitions at 150 MHz and 270 MHz in the EDNMR trace recorded at 7 T suggests that the hyperfine anisotropy for Mn1 and Mn4 is on the order of 10 MHz. This hyperfine anisotropy is in line with reported values of monomeric octahedral Mn^IV as well as the S₃ state.⁵ The site-specific anisotropy can be calculated from the projection factors listed in Table 1 and yield values of approximately 25 MHz, which are in good agreement with rhombically distorted octahedral Mn^IV in SnO₂/TiO₂.^64,72–73 Mn^IV in SnO₂/TiO₂ similarly displays ZFS values on the order of a wavenumber^{64, 72–73} suggesting that a distortion of the octahedral coordination environment of the individual Mn^IV ions in 3 leads to the observed ZFS.

In exchange-coupled systems, the observed sign and magnitude of the hyperfine coupling interactions reflect the nature of the magnetic exchange interactions between the metal centers. For 3, a trimer-monomer magnetic coupling model of pseudo-octahedral Mn^IV centers accurately describes the observed ⁵⁵Mn hyperfine interactions. In the trimer-monomer model, three Mn^IV (S = 3/2) centers couple ferromagnetically, giving an S = 9/2 fragment. This fragment then couples antiferromagnetically with the fourth Mn^IV center to give an overall S = 3 spin ground state. Importantly, computational studies on the S₃^B model of the S₃ state with a trimer-monomer magnetic coupling scheme produces spin projection factors that estimate the observed A to be on the order of |70−90| MHz, with a unique A holding an opposite sign to the rest. This model is in good agreement with our EDNMR results (Table 1). Additionally, scaling the observed hyperfine coupling interactions in the ferromagnetically coupled S = 9/2 CaMn₃^IVO₄ complex using the spin projection factors for the trimer-monomer model yield ⁵⁵Mn hyperfine interactions that are in excellent agreement with the values obtained for 3. Altogether, the EDNMR data of 3 is consistent with the trimer-monomer magnetic coupling model of pseudo-octahedral Mn^IV centers in which the ferromagnetically coupled subunit comprised of Mn2, Mn3, and Mn4 couples antiferromagnetically to Mn1 to give an overall S = 3 spin ground state.

DISCUSSION

In the Kok cycle, each S_n state adopts a characteristic spin ground state that is intimately connected to the Mn oxidation states and the overall structure of the Mn-oxo core.⁷⁴ Within the ensemble of the various S_n states present in PSII samples, such differences in spin state provide unique, differentiating spectroscopic features that can be exploited in EPR studies of the OEC. In the case of the S₃ state, an effective S = 3 spin ground state gives rise to a broad EPR signal that distinguishes the S₃ state from the S₂ state, which features a narrower EPR signal rising from an S = 1/2 ground state.⁵ The higher S = 3 spin state of the S₃ state is notable, as the OEC adopts low spin state electronic structures in the earlier S₀ (S = 1/2) and S₁ (S = 0) states.^75–76 In the S₃ state, the small anisotropy of the ⁵⁵Mn hyperfine interactions has been interpreted as rising from the presence of all-octahedral Mn^IV centers. The relative sign and magnitude of the ⁵⁵Mn hyperfine interactions inform about the magnetic exchange coupling interactions between the Mn^IV centers. In turn, magnetic coupling interactions are highly sensitive to the structure of the complex, and two limiting models S₃^A and S₃^B have been considered computationally, with S₃^A better accommodating the experimental data.⁵ Recently, an altered form of the S₃ state was generated via treatment with methanol.⁶³ This chemically modified S₃^MeOH state exhibits a larger ZFS than that of untreated form (|D| = 0.281 vs 0.175 cm⁻¹) by virtue of a five-coordinate Mn^IV at the dangler Mn site. This form of the S₃ state with a larger ZFS is also observed in the untreated S₃ state, but in a lower concentration. Importantly, the isotropic ⁵⁵Mn hyperfine interactions in S₃^MeOH are consistent with the trimer-monomer magnetic coupling model similar to the S₃^B form (Figure 1a). Reported ⁵⁵Mn hyperfine interactions of S₃^MeOH are on the order of |60−100| MHz, in close agreement with our experimental values (Table 1).⁶³

Oxidation of the previously reported Mn^IIIMn₃^IVO₄ (2) complex featuring an S = 1/2 ground state²⁴ leads to a Mn^IV₄O₄ (3) complex with an S = 3 ground state, modeling the spin state change behavior of the S₂→S₃ transition in the OEC. The structure of 3 and its Mn^IV₄ oxidation state assignment is supported by the rising edge energy of the XANES spectrum and the similar EXAFS spectrum between 2 and 3. EPR spectra at both X- and D-band frequencies support the S = 3 spin state assignment. Most importantly, observed ⁵⁵Mn hyperfine interactions (A) support a magnetic coupling model proposed for the trimer-monomer model for S₃^B and and S₃^MeOH, in which a ferromagnetically coupled trimetallic S = 9/2 subunit is coupled antiferromagnetically to the fourth Mn center to yield an S = 3 effective spin ground state. The intrinsic, on-site ⁵⁵Mn hyperfine interaction (a), which is a function of the oxidation state of the Mn center and its coordination geometry, is scaled according to a spin projection factor (ρ) that reflects the nature of the magnetic exchange coupling within the cluster. In addition to the reported a ≈ 180 MHz for a ferromagnetically coupled S = 9/2 CaMn₃^IVO₄ complex, on-site Mn^IV hyperfine values of a ≈ 200 MHz have been reported for mononuclear and dinuclear Mn^IV complexes.^{71, 77–79} As shown in Table 1, the observed A for 3 is in good agreement with the value obtained by scaling the on-site Mn^IV hyperfine interaction by the spin projection factor of the trimer-monomer model. Similar to the spin projection factor ρ, a second coefficient κ_i is needed to scale the site fine structure d_i to the measured zero field splitting D, according to the expression D = Σκ_id_i.^{5, 63} Assuming a site value of d_i ≈ 1 cm⁻¹ for two sites, with the remainder being d_i ≈ 0.3 cm⁻¹, a total D of 0.4 cm⁻¹ can be achieved using the κ_i for the trimer-monomer model⁶³: D = (0.15)(1 cm⁻¹) + (0.15)(1 cm⁻¹) + (0.15)(0.3 cm⁻¹) + (0.08)(0.3 cm⁻¹) = 0.37 cm⁻¹. This rough analysis illustrates the point that the measured D of 0.4 cm⁻¹ likely arises from multiple distorted octahedral sites in 3, as further evident from the relatively larger estimated hyperfine anisotropy of 3.

In conclusion, a unique Mn^IV₄O₄ cuboidal complex has been synthesized and characterized by XAS, CW-EPR, and pulsed-EPR spectroscopies. To our knowledge this is the first set of experimental studies that directly probes the electronic structure of a tetranuclear Mn^IV₄ complex with an S = 3 spin ground state mimicking the S₃ state of the OEC. Our studies provide a synthetic precedent for the S = 1/2→3 spin state change that is characteristic of the S₂→S₃ transition in the OEC. The magnetic coupling scheme in the Mn^IV₄O₄ cuboidal complex resemble that of a recently characterized form of the S₃ state of the OEC.⁶³