ENDOR Characterization of (N2)FeII(μ-H)2FeI(N2)-: A Spectroscopic Model for N2 Binding by the Di-μ-hydrido Nitrogenase Janus Intermediate

Abstract

The biomimetic diiron complex 4-(N₂)₂, featuring two terminally bound Fe-N₂ centers bridged by two hydrides, serves as a model for two possible states along the pathway by which the enzyme nitrogenase reduces N₂. One is the Janus intermediate E₄(4H), which has accumulated 4[e-/H+], stored as two [Fe-H-Fe] bridging hydrides, and is activated to bind and reduce N₂ through reductive elimination (RE) of the hydride ligands as H₂. The second is a possible RE intermediate. ¹H and ¹⁴N 35 GHz ENDOR measurements confirm that the formally Fe(II)/Fe(I) 4-(N₂)₂ complex exhibits a fully delocalized, Robin-Day type-III mixed valency. The two bridging hydrides exhibit a fully rhombic dipolar tensor form, T ≈ [- t, + t, 0]. The rhombic form is reproduced by a simple point-dipole model for dipolar interactions between a bridging hydride and its "anchor" Fe ions, confirming validity of this model and demonstrating that observation of a rhombic form is a convenient diagnostic signature for the identification of such core structures in biological centers such as nitrogenase. Furthermore, interpretation of the ¹H measurements with the anchor model maps the g tensor onto the molecular frame, an important function of these equations for application to nitrogenase. Analysis of the hyperfine and quadrupole coupling to the bound ¹⁴N of N₂ provides a reference for nitrogen-bound nitrogenase intermediates and is of chemical significance, as it gives a quantitative estimate of the amount of charge transferred between Fe and coordinated N, a key element in N₂ activation for reduction.

Figure 1

FeMo-co, [7Fe-9S-Mo-C-homocitrate] with cartoon of active Fe 2,3,6,7 FeMo-co face. PDB, 2AFI.

Figure 2. Schematic of re/oa:

Fe 2,3,6,7 FeMo-co face. The positioning of hydrides is a matter of current investigation and shown is our current best guess; likewise for position of protons in E₄(4H) and the hemilability of bridging sulfur (see text); ‘2N2H’ denotes a diazene-level intermediate.

Figure 3

X-band (top) and the numerical derivative of 35 GHz (bottom) CW EPR spectra for 4-(N₂)₂: (black) (μ−¹H)₂; (red) (μ−²H)₂). Experimental conditions: X-band, microwave frequency, 9.215 GHz, modulation amplitude, 10 G; temperature, 77 K. ‘35 GHz’ microwave frequency, 35.084 GHz; microwave power, 10 μW; modulation amplitude 2 G; time constant, 32 ms; temperature, 2 K.

Figure 4

35 GHz stochastic CW ¹H (black) and ²H (red) ENDOR spectra at g₁ for 4-(N₂)₂. Experimental conditions: sample time, 0.75 ms; delay time, 0.5 ms; RF-on time, 0.5 ms; modulation amplitude, 4 G; microwave frequency, 35.084 GHz; microwave power, 10 μW; temperature 2 K. (*) in ²H ENDOR spectrum indicates ν₋ of ³¹P_A.

Figure 5

2D field-frequency pattern of 35 GHz stochastic-CW ¹H ENDOR spectra of 4-(N₂)₂. Experimental conditions: sample time, 0.75 ms; delay time, 0.5 ms; RF-on time, 0.5 ms; modulation amplitude, 4 G; microwave frequency, 35.084 GHz; microwave power, 10 μW; temperature 2K. Simulation (Blue): g = [2.155, 2.067, 2.038], A = [87, 42, 70] MHz, and (α, β,γ) = (0, 10, 0).

Figure 6

Core structure of 4-(N₂)₂ superimposed with g-frame and dipolar tensor T-frame of bridging hydrides.

Figure 7

2D pattern of stochastic ³¹P ENDOR (Black) and the simulation of P_A (red) and P_B (blue). Experimental conditions: sample time, 0.75 ms; delay time, 0.5 ms; RF-on time, 0.5 ms; modulation amplitude, 4 G; microwave frequency, 35.084 GHz; microwave power, 10 μW; temperature 2K. Simulation: P_A, A = +[16.5, 25, 13] MHz, (α, β, γ) = (30, 10, 0); P_B, A = +[27, 38, 26] MHz, (α, β, γ) = (45, 20, 0).

Figure 8

35 GHz Mims ¹⁴N ENDOR spectra of 4-(¹⁴N₂)₂ collected at the single-crystal field for g₃, showing suppression of the entire ¹⁴N signal when τ = 1/A (eq 2): Upper/Lower spectra, τ = 350/555 ns. The signal intensities of the two spectra are normalized to their 2-pulse ESE intensity. Experimental conditions: microwave frequency, 35.075 GHz; π/2 = 30 ns; t_RF = 60 μs, and RF randomly hopped; repetition time, 25 ms; temperature, 2 K. The arrows represent Mims-hole suppression pattern.

Figure 9

Partial 2D field-frequency 35 GHz Mims ¹⁴N ENDOR spectra of 4-(N₂)₂ (black). Experimental conditions: microwave frequency, 35.075 GHz; π/2 = 30 ns; τ = 350 ns; t_RF = 60 μs, and RF randomly hopped; repetition time, 25 ms; temperature, 2K. Simulations (blue): A = [0.1, 1.8, 1.8] MHz, P = [−1.71, 0.75, 0.96] MHz, (α, β,γ) = (0, 0, 0).

Chart 1

Scheme 1

Schematic representation of metric parameters used to calculate the proton dipolar tensor within the defined molecular frame.