Magnetometric Characterization of Intermediates in the Solid-State Electrochemistry of Redox-Active Metal-Organic Frameworks

Qi Chen; Zhongyue Zhang; Kunio Awaga

doi:10.3791/65335

JoVE Journal > Chemistry

Quimica

Magnetometric Characterization of Intermediates in the Solid-State Electrochemistry of Redox-Active Metal-Organic Frameworks

Published: June 09, 2023

doi:

10.3791/65335

Qi Chen¹, Zhongyue Zhang², Kunio Awaga^1,3

¹Department of Chemistry, Graduate School of Science,Nagoya University, ²International Research Organization for Advanced Science and Technology,Kumamoto University, ³Integrated Research Consortium on Chemical Sciences,Nagoya University

Summary

Ex situ magnetic surveys can directly provide bulk and local information on a magnetic electrode to reveal its charge storage mechanism step by step. Herein, electron spin resonance (ESR) and magnetic susceptibility are demonstrated to monitor the evaluation of paramagnetic species and their concentration in a redox-active metal-organic framework (MOF).

Abstract

Electrochemical energy storage has been a widely discussed application of redox-active metal-organic frameworks (MOFs) in the past 5 years. Although MOFs show outstanding performance in terms of gravimetric or areal capacitance and cyclic stability, unfortunately their electrochemical mechanisms are not well understood in most cases. Traditional spectroscopic techniques, such as X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure (XAFS), have only provided vague and qualitative information about valence changes of certain elements, and the mechanisms proposed based on such information are often highly disputable. In this article, we report a series of standardized methods, including the fabrication of solid-state electrochemical cells, electrochemistry measurements, the disassembly of cells, the collection of MOF electrochemical intermediates, and physical measurements of the intermediates under the protection of inert gases. By using these methods for quantitatively clarifying the electronic and spin state evolution within a single electrochemical step of redox-active MOFs, one can provide clear insight into the nature of electrochemical energy storage mechanisms not only for MOFs, but also for all other materials with strongly correlated electronic structures.

Introduction

Since the term metal-organic framework (MOF) was introduced in the late 1990s, and especially in the 2010s, the most representative scientific concepts concerning MOFs have arisen from their structural porosity, including guest encapsulation, separation, catalytic properties, and molecule sensing¹^,²^,³^,⁴. Meanwhile, scientists were quick to realize that it is essential for MOFs to possess stimuli-responsive electronic properties in order to integrate them into modern smart devices. This idea triggered the spawning and flourishing of the conductive two-dimensional (2D) MOF family in the past 10 years, thereby opening the gate for MOFs to play key roles in electronics⁵ and, more attractively, in electrochemical energy storage devices⁶. These 2D MOFs have been incorporated as active materials in alkali metal batteries, aqueous batteries, pseudocapacitors, and supercapacitors⁷^,⁸^,⁹, and have exhibited tremendous capacity as well as excellent stability. However, to design better-performing 2D MOFs, it is crucial to understand their charge storage mechanisms in detail. Therefore, this article aims to provide a comprehensive understanding of the electrochemical mechanisms of MOFs, which can aid in the rational design of better-performing MOFs for energy storage applications.

In 2014, we first reported the solid-state electrochemical mechanisms of MOFs with redox-active sites on both metal cations and ligands¹⁰^,¹¹. These mechanisms were interpreted with the help of various in situ and ex situ spectroscopic techniques, such as X-ray photoelectron spectroscopy (XPS), X-ray absorption fine structure (XAFS), X-ray diffraction (XRD), and solid-state nuclear magnetic resonance (NMR). Since then, this research paradigm has become a trend in studies of the solid-state electrochemistry of molecular-based materials¹². These methods work fine for identifying the redox events of conventional MOFs with carboxylate bridging ligands, as the molecular orbitals and energy levels of metal cluster building blocks and organic ligands are almost independent of each other in such MOFs¹²^,¹³.

However, when encountering the strongly correlated 2D MOFs with significant π-d conjugation, the limitations of these spectroscopic methods were exposed. One of these limitations is that the band levels of most aforementioned 2D MOFs cannot be considered as a simple combination of metal clusters and ligands, but are rather a hybridization of them, while most of the spectroscopic methods only provide averaged, qualitative information about the oxidation states¹⁴. The other limitation is that the interpretation of these data is always based on the assumption of localized atomic orbitals. Therefore, the intermediate states with metal-ligand hybridization and delocalized electronic states are usually overlooked and described incorrectly with only these spectroscopic methods¹⁵. It is necessary to develop new probes for the electronic states of these electrochemical intermediates of not only 2D MOFs, but also other materials with similar conjugated or strongly correlated electronic structures, such as covalent organic frameworks¹⁶, molecular conductors, and conjugated polymers¹⁷.

The most common and powerful tools for assessing the electronic structures of materials are electron spin resonance (ESR) and superconducting quantum interference device (SQUID) magnetic susceptibility measurements¹⁸^,¹⁹. As both rely on unpaired electrons in the system, these tools can provide tentative information about the spin densities, spin distributions, and spin-spin interactions. ESR offers sensitive detection of unpaired electrons, while magnetic susceptibility measurement gives more quantitative signals for upper properties²⁰. Unfortunately, both techniques unavoidably face great challenges when used to analyze the electrochemical intermediates. This is because target samples are not pure, but rather a mixture of target material, conductive additive, binder, and byproduct from the electrolyte, so the obtained data²¹^,²² are the sum of contributions from both the material and the impurities. Meanwhile, most intermediates are sensitive to the environment, including air, water, certain electrolytes, or any other unpredictable perturbations; extra care is necessary while handling and measuring intermediates. Trial and error is normally necessary while dealing with a new combination of electrode material and electrolyte.

Here, we present a new paradigm, called electrochemical magnetometry, for analyzing the electronic states or spin states of 2D MOFs and similar materials using a series of techniques, utilizing electrochemistry and temperature-variable ex situ ESR spectroscopy as well as ex situ magnetic susceptibility measurements²⁰. To demonstrate the effectiveness of this approach, we use Cu₃THQ₂ (THQ = 1,2,4,5-tetrahydroxybenzoquinone; referred to as Cu-THQ), a representative 2D MOF, as an example. We explain the selection of conductive additives and electrolytes, the fabrication of electrodes and electrochemical cells, as well as details on sample handling and measurement, including possible issues during measurement. By comparing with classic characterizations such as XRD and XAFS, electrochemical magnetometry can provide a comprehensive understanding of the electrochemical mechanism of most MOFs. This approach is capable of capturing unique intermediate states and avoiding incorrect assignment of redox events. The elucidation of energy storage mechanisms using electrochemical magnetometry can also contribute to a better understanding of the structure-function relationships in MOFs, leading to more intelligent synthetic strategies for MOFs and other conjugated materials.

Protocol

1. Electrode fabrication

Synthesizing Cu-THQ MOF
NOTE: Cu-THQ MOF polycrystalline powder was synthesized via a hydrothermal method following previously published procedures¹⁴^,²⁰^,²³.
1. Put 60 mg of tetrahydroxyquinone into a 20 mL ampule, then add 10 mL of degassed water. In a separate glass vial, dissolve 110 mg of copper (II) nitrate trihydrate in another 10 mL of degassed water. Add 46 µL of the competing ligand ethylenediamine using a pipette.
  NOTE: To degas the deionized water, flow nitrogen gas for 30 min before use. If the reaction mixture is heated for too long, a Cu impurity may form with a diffraction peak appearing around 43° (Cu Kα).
2. Introduce the copper solution into the ampule containing the tetrahydroxyquinone. The color of the solution immediately changes from red to navy. Freeze, pump, and thaw²⁴ the resulting solution three times to further remove dissolved oxygen.
3. Flame seal the ampule using a torch under vacuum. Heat the solution to 60 °C for 4 h.
4. After the reaction, carefully open the ampule and remove the supernatant. Wash the precipitate with 30 mL of room-temperature deionized water 3x and 30 mL of hot deionized water (80 °C) by centrifugation 3x at 10,000 rpm for 5 min.
5. Disperse the precipitate into acetone by shaking, then filter and wash with acetone. Heat the product under vacuum at 353 K overnight to remove the residual solvent in the Cu-THQ MOF.
Preparing CuTHQ electrodes
NOTE: To distinguish between the Cu-THQ MOF and the electrode, the former is referred to as Cu-THQ, while the mixture of Cu-THQ, carbon, and binder is referred to simply as the CuTHQ.
1. To prepare the Cu-THQ/CB/PVDF electrode, dissolve 10 mg of poly(vinylidene fluoride) (PVDF) in 1.4 mL of N-methyl-2-pyrrolidone (NMP). Disperse 50 mg of Cu-THQ MOF and 40 mg of carbon black (CB) in the solution by vigorously stirring overnight. Coat the homogeneous slurry onto an Al disk with a diameter of 15 mm and a mass of ~9.7 mg.
2. To prepare the Cu-THQ/Gr/SP/SA electrode, follow the same procedure as the Cu-THQ/CB/PVDF electrode, but with a different slurry composition: Cu-THQ MOF (80 mg), sodium alginate (SA, 2 mg), and graphene/Super P (Gr/SP, 1:1.8 dilution by weight, total of 18 mg) in water/isopropanol (1:1 dilution by volume, total of 1.2 mL).
3. Dry the electrodes under vacuum at 353 K for 12 h. Vent the nitrogen gas after drying and measure the mass loading.

2. Battery assembly and post-treatment

NOTE: Due to the air-sensitive nature of electrochemical intermediates, battery assembly and post-treatment must be performed in an argon glove box with strict air-free manners.

Assembling Li/CuTHQ coin cells
1. Cut several pieces of lithium disks with a diameter of 15.5 mm and Celgard separators with a diameter of 17 mm before assembling the battery.
2. Assemble Li/CuTHQ coin cells (CR2032) from bottom to top in the following order: negative shell, spacer (height = 0.5 mm), lithium, separator, CuTHQ electrode (prepared in step 1.2.1 or 1.2.2), spacer, spring, and positive shell (Figure 1A).
3. Before and after adding the separator, drop a total of 0.04 mL of electrolyte (1.0 M LiBF₄ in ethylene carbonate (EC)/diethyl carbonate (DEC) at 1:1 wt%). Do not use metal tweezers to hold the coin cell after it is assembled.
Preparing electrochemical intermediates
1. Compress the coin cell using the tightening screw (not sealed) with a homemade device (Figure 1B) and connect the device to the measuring cables in the glove box. Connect the instrument (outside the glove box) to the ports corresponding to the coin cell. Perform cyclic voltammetry and galvanostatic charge/discharge measurements²⁰ to achieve the intermediates at different potentials (Figure 2).
2. After electrochemical cycling, disassemble the coin cell carefully to avoid any short circuits.
3. Rinse the cycled CuTHQ electrode with 5 mL of battery-grade dimethyl carbonate (DMC). Dry the electrode naturally for 30 min. Collect the sample from the Al disk to Al foil using a clean spatula.
4. Transfer the sample powder into an ESR tube or SQUID tube through a homemade glass funnel (Figure 1C). Seal the sample tube tightly with a cap and transparent film. Alternatively, connect the sample tube to a rubber tube and seal it with a valve, followed by flame sealing the head of the sample tube under vacuum.
5. After magnetic measurements²⁰, open the sample tube and dump the sample onto Al foil. Measure the mass of the sample using an analytical balance with a resolution of 0.01 mg in air. Estimate the mass of Cu-THQ from the total mass of the sample.
  NOTE: The mass of the cycled Cu-THQ MOF is estimated to be 50% or 80% of the total mass, depending on the type of electrodes used; this estimate does not take into account the inserted Li ions and residual electrolyte.

Figure 1: The equipment used for ex situ magnetometry experiments. (A) A Photograph of a CR2032 coin cell. (B) The homemade device was used to evaluate the unsealed coin cell in the glove box. (C) Photographs of ESR and SQUID sample tubes with and without samples inside. The ESR tube consists of a 10 cm high-purity quartz tip (measurement section) and a 17 cm borosilicate glass head. There are two kinds of SQUID tubes. Tube A consists of a 2 cm x 5 cm quartz tip with a quartz diaphragm at the midpoint and a 10 cm borosilicate glass head, and tube B is a plastic tube (20 cm long) with a plastic diaphragm at the midpoint. All sample tubes have an outer diameter of 5 mm. Please click here to view a larger version of this figure.

3. Registration of ESR spectra at variable temperatures

Recording the ESR spectra at room temperature
1. Once the ESR spectrometer is ready, insert the prepared sample tube into the microwave cavity and center the sample. Autotune the microwave phase, coupling, and frequency to reach the cavity's resonance condition. Check the Q-dip in the center of the screen for a symmetric shape and maximal depth.
  NOTE: If the sample contains too much conductive carbon, such as carbon black, the autotune process may fail or result in a small quality factor (Q-value) of the cavity. The typical mass of the sample is 3 mg.
2. Choose the optimal parameters, such as: microwave: power; magnetic field: sweep time; center field: sweep width; modulation: frequency, width; channel: amplitude, time constant. Then, sweep the magnetic field and record the ESR spectrum. The typical values of the measurement parameters are shown in Figure 3 and Figure 4.
3. Adjust the insertion amount of the Mn marker to 800. Repeat steps 3.1.1 and 3.1.2 to record an ESR spectrum with the Mn marker. Calibrate the magnetic field by employing six hyperfine lines for the Mn(II) ions.
Line shape analysis of Cu-THQ
1. Import the ESR dataset into Python (version 3.9.7). Normalize the ESR spectrum by dividing the intensity by the sample mass, the square root of the microwave power, the modulation width, and the amplitude.
2. Fit the calibrated and normalized ESR spectrum of as-prepared Cu-THQ MOF to the axially symmetric Lorentzian function²⁵:
  
  Where N is a scale factor that includes instrument parameters g_ll and H_II constants, ( and ) are the parallel (perpendicular) component of Lander g-factor and the corresponding resonance magnetic field, ΔH_pp is the peak-to-peak line width, and H_r is an integral variable.
  NOTE: The Python codes for the Lorentzian function are available in Supplementary Coding File 1 (which is named as AxialLorentz).
3. Obtain the anisotropic g-value and peak-to-peak line width for the axially symmetric Cu(II) ions.
4. Fit the calibrated and normalized ESR spectrum with the Lorentzian function for the radical samples. Obtain the isotropic g-value and peak-to-peak line width for the radicals.
  
  This is named as SymLorentz in Supplementary Coding File 1.
Quantitating the radical concentration
1. Grind 3.45 mg of 4-hydroxy-2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPOL) and 96.55 mg of KBr together in an agate mortar until a homogeneous mixture is achieved. Place 1 mg (0.2 µmol), 2 mg (0.4 µmol), and 4 mg (0.8 µmol) of TEMPOL/KBr mixtures into three ESR sample tubes, respectively.
2. Follow steps 3.1.1 and 3.1.2 to record the ESR spectra for the TEMPOL/KBr standards.
3. Conduct a linear baseline fitting between the double integration of the ESR spectra and the number of spins in the TEMPOL/KBr standards. Determine the number of spins in the cycled Cu-THQ using the linear baseline of the TEMPOL/KBr standards²⁶.
Recording the ESR spectra at low temperatures
NOTE: Use liquid helium to achieve a cryogenic temperature. It is necessary to wear cryogenic gloves when working with liquid helium.
1. Check the ESR spectrum at room temperature first by following step 3.1.
2. Evacuate the thermal shield to a high vacuum level. Purge the microwave cavity using nitrogen gas to avoid condensation.
3. Introduce liquid helium from the vessel into the cryostat. Gradually cool the sample to the lowest temperature (around 10 K). Wait 30 min to achieve thermal equilibrium.
4. Record the temperature-dependent ESR spectra during warming. Confirm that the ESR spectrum does not suffer from the power saturation effect at low temperature, and the ratio between signal intensity (peak-to-peak height) and the square root of microwave power remains constant in the absence of power saturation.
  NOTE: When power is saturated, the signal intensity increases more slowly than the square root of the microwave power. The sampling density could gradually decrease as the temperature rises.

4. Magnetic susceptibility measurements

Attach the sample tube to the bottom of the sample rod. Make sure that the surface of the sample tube is clean.
Purge the sample chamber and insert the sample tube into the SQUID. Apply a magnetic field and center the sample within the detection coil. Remove the external magnetic field after centering.
NOTE: If the spin concentration is too low to detect, consider increasing the magnetic field or centering after cooling to 2 K. The typical sample mass for SQUID measurements is approximately 6 mg.
Cool the system to 20 K at a rate of 10 K/min. Pause cooling for 30 min, then further cool to 2 K for 1 h.
Measure the magnetic susceptibility of the cycled CuTHQ electrode under a magnetic field of 1,000 Oe while warming to 300 K; this is referred to as the zero-field-cooled (ZFC) process. Next, cool to 2 K again and record the magnetic susceptibility in the field-cooled (FC) process.
Repeat steps 4.1 to 4.4 with CuTHQ electrodes cycled at different reduction degrees.
Measure the magnetic susceptibility of the carbon materials (Gr/SP) under the same conditions. Use this result to compensate for the magnetic susceptibility of the CuTHQ electrodes.
Fit the temperature dependence of magnetic susceptibility to the modified Curie-Weiss law:

Where χ_m is the molar magnetic susceptibility, C_m is the molar Curie constant, θ is the Weiss temperature, and χ₀ is a temperature-independent term.

Representative Results

Our previous work included a detailed discussion of ex situ ESR spectroscopy and ex situ magnetic susceptibility measurements for electrochemically cycled CuTHQ²⁰. Here, we present the most representative and detailed results that can be obtained following the protocol described in this paper.

Figure 2: Electrochemical performances of the Li/CuTHQ batteries. (A) The first discharge/charge curves of the Cu-THQ/CB/PVDF electrode (red lines) and the Cu-THQ/Gr/SP/SA (teal dashed lines) under a current density of 50 mA/g. The lithium-ion content (x) was calculated based on the ratio of the partial capacity to a theoretical capacity of 130 mAh/g per electron. (B) The differential capacity (dQ/dV) curves of the Cu-THQ/CB/PVDF electrode. (C) The cyclic voltammogram for the Cu-THQ/CB/PVDF electrode, cycling at a scan rate of 0.1 mV/s during the first three scans, is shown. Please click here to view a larger version of this figure.

We found that the carbon and binder do not affect the number of electrons transferred, and that the Li/CuTHQ battery delivers a specific capacity of 390 mAh/g in the first discharge process (Figure 2A). This value is exceedingly high and corresponds to a three-electron reduction (130 mAh/g for a one-electron reduction). The high capacity benefits from the variable valences of the Cu ions, the unsaturated π-electron bond in the THQ ligand, and the π-d conjugation of the network¹⁵. Differential capacity (dQ/dV) analysis and cyclic voltammetry (CV) of the CuTHQ/CB/PVDF electrode are shown in Figure 2B and Figure 2C, respectively. Three electronic states, namely the Cu(II)-related state, the π-d conjugated state, and the delocalized π-electron state, were proposed to account for the three redox peaks in the dQ/dV and CV curves when the potential was varied from 4.0 V to 1.5 V.

The magnetic properties of reduced Cu-THQ MOF are conducted on the electrochemically cycled electrode, which is a mixture of Cu-THQ MOF, conductive carbon, and binder. For the ESR studies, we use the ESR silent CB to prepare the CuTHQ electrode. As for the magnetic susceptibility measurements, we replace the CB with a Gr/SP mixture, the magnetic susceptibility of which is negligible at room temperature.

Figure 3: Ex situ ESR spectra obtained during the first discharge of Cu-THQ/CB/PVDF electrodes. (A) The image displays the normalized ESR spectra for the corresponding points in B, with solid lines representing the best-fitted lines. The top insets show the piecewise fits of Cu(II) (left) and radical (right) signals for the electrode discharged to 1.9 V, while the bottom inset presents the peak-differentiating and imitating spectra of the electrode cycled to 1.5 V. The measurement conditions were a microwave power of 0.5 mW and modulation width of 0.5 mT. (B) The image depicts the proposed electrochemical mechanism for the Cu-THQ MOF. Please click here to view a larger version of this figure.

Based on the ex situ ESR spectra, we have identified the evolution of paramagnetic species by analyzing the g-value, peak-to-peak line width, and line shape during the discharge process. Figure 3 presents representative ESR spectra for the Cu(II) and organic radical signals. The prepared Cu-THQ MOF exhibited a broad ESR line with a peak-to-peak line width of 37 mT, centered at a Landé g-factor of 2.11. Upon discharging to 2.35 V, the Cu(II) signal split into two components, perpendicular 2.06 and parallel 2.25, due to the g-tensor anisotropy in an axially symmetric crystal field²⁷. Further discharge to 1.9 V resulted in an additional Lorentzian line at g = 2.0047 with a line width of 0.66 mT, and could be attributed to the organic radical generated by the partial reduction of quinone to semiquinone. When Cu-THQ was deeply reduced to 1.5 V, the Cu(II) signal disappeared, and only the radical signal remained, indicating that the Cu(II) ions were fully reduced to Cu(I). Fitting the Lorentzian function suggested that the radical signal contained two contributions: a narrow line and a slightly broad line with line widths of 0.73 and 2.98 mT, respectively.

Figure 4: Quantitative determination of the spin concentration for the prepared Cu-THQ and the electrode discharged to 1.5 V. (A) The image shows the ESR spectra of the samples and TEMPOL standard, with the ESR spectra of both samples scaled up to match the intensity of the Mn marker's signal with that of the standards. The measurement conditions were a microwave power of 0.5 mW, modulation width of 0.5 mT, and Mn marker of 800. (B,C) The images are zoomed-in views of A, displaying the Cu(II) (B) and radical signal (C). (D) This image presents the linear regression analysis of the double integral and the number of spins for the standards. According to the fitting result, the number of spins for the prepared Cu-THQ was determined to be 1.08, while that for the 1.5 V sample was 0.017. Please click here to view a larger version of this figure.

Quantitative ESR measurements were performed to determine the spin concentrations of the CuTHQ electrodes, as shown in Figure 4. The spin intensity of the fully discharged state (discharged to 1.5 V) was estimated to be 1.7% per CuO₄ unit. The same methodology was used to analyze the prepared Cu-THQ MOF and confirm the valence state of the Cu ions. The results showed that 96% of Cu ions in the prepared Cu-THQ were paramagnetic Cu(II), consistent with our previous magnetic susceptibility findings that 99% of Cu sites in the prepared Cu-THQ were occupied by the Cu(II) ions²⁰.

Figure 5: Temperature dependence of ESR spectra for the corresponding states in Figure 3. (A,B) The images show the temperature dependence of Cu(II) lines for the prepared Cu-THQ and the electrode discharged to 2.35 V, respectively, in the temperature range from ~10 K to 300 K, with modulation fields of 0.5 mT and 2.0 mT, respectively. (C) This image illustrates the Curie-like behavior of the localized radical signal for the electrode discharged to 1.9 V. The measurement conductions were a microwave power of 0.1 mW, modulation width of 0.2 mT, and amplitude of 100. (D) This image shows the temperature-independent feature of the additional π-electron signal, with measurement conductions of a microwave power of 0.08 mW, modulation width of 0.2 mT, and amplitude of 50. Please click here to view a larger version of this figure.

Figure 5 displays the temperature-dependent ESR spectra of the prepared Cu-THQ MOF and CuTHQ electrodes that were discharged at 2.35 V, 1.9 V, and 1.5 V. This information is essential for understanding the physical properties of paramagnetic spins, such as their ground spin states (spin crossover), magnetic states (Curie spins, Pauli spins, or superparamagnetic spins), exchange and dipolar interactions, and crystal field symmetry (structural phase transition)²⁸.

In the prepared Cu-THQ MOF, the ESR line width of the Cu(II) signal narrowed gradually as the temperature drops, while the g-value remained constant. This can be attributed to the large antiferromagnetic interaction (the Weiss temperature is -18.1 K) rather than magnetic ordering. In the case of the 0.65-electron reduced sample (discharge to 2.35 V), both the line width and the g-value remained unchanged due to the low spin density and the weak magnetic interaction. The temperature dependence of the ESR spectra of the electrode discharged to 1.9 V follows the Curie law, indicating that the radical spins are localized. This characteristic was also observed on the narrow line in the electrode discharged to 1.5 V, suggesting the extrinsic defect nature of the organic radicals. In contrast to the narrow line, we discovered that the spin intensity of the broad line was independent of temperature between 100 K and 300 K. This suggests that the electron spin is delocalized/hopping in the conjugated network¹⁷^,²⁹.

Figure 6: ESR spin susceptibility and SQUID magnetic susceptibility. The ESR spin susceptibility and SQUID magnetic susceptibility of the prepared CuTHQ (A) and the 1.5 V reduced CuTHQ (B). The external magnetic field of SQUID measurements was 1,000 Oe, and a diamagnetic correction of -8 x 10^-5 emu/mol was applied to the raw data. The dashed lines in both panels represent the best fits by a modified Curie-Wiess law. Please click here to view a larger version of this figure.

Figure 6A,B compares the temperature dependence of the normalized ESR spin susceptibility (χ_sT) and SQUID magnetic susceptibility (χ_mT) for both as-prepared Cu-THQ and the fully discharged state. A modified Curie-Weiss law (dashed line) was used to fit the χ_mT versus T plot, yielding a Curie constant of 0.039 emu K/mol and a temperature-independent paramagnetic (TIP) term of 1.02 x 10^-3 emu/mol. The significant TIP term observed in the χ_sT versus T plot can be mainly attributed to the electrochemical doping of the π-d conjugated framework with delocalized π electrons. Notably, graphite, the most commonly used anode in commercial Li-ion batteries, also demonstrates a charge storage mechanism associated with delocalized π electrons³⁰, which suggests that such a mechanism could occur in 2D fully π-d conjugated MOFs.

Supplementary Coding File 1: Python code for Lorentzian function. Please click here to download this File.

Discussion

To produce cathodes, it is necessary to mix the active material with conductive carbon to achieve a low polarization during the electrochemical process. The carbon additive is the first critical point for ex situ magnetometry; if the carbon has radical defects, the emergence of the electrochemically induced organic radical cannot be observed in the ESR spectrum. This makes it difficult to precisely determine the spin concentration or organic radical concentration, since these two types of radicals have similar g-values, and their ESR lines may overlap. Additionally, if the carbon contains even a small amount of ferromagnetic impurity, its magnetic susceptibility can dominate in the high-temperature region. Furthermore, carbon additives can absorb microwave energy in the X-band³¹, which limits the use of ex situ ESR spectroscopy and introduces an error in the quantitative determination of the radical concentration. In other words, the actual microwave exposure of the ESR sample is lower than expected.

The second critical point is related to SQUID magnetometry. Almost all conductive carbons exhibit a Curie tail in the temperature dependence of magnetic susceptibility, due to the paramagnetic impurities²⁰. Consequently, it is necessary to measure and subtract the diamagnetic contribution of carbons and binders from the raw data. In the previous study²⁰, we found that a mixture of graphene and Super P in a mass ratio of 1:1.8 has negligible susceptibility at room temperature. This mixture can be used as a carbon additive to improve the accuracy of magnetic susceptibility measurements.

The third critical point concerns the reproducibility of electrochemical behaviors. On the one hand, in ex situ measurements, we assembled numerous Li/CuTHQ coin cells to produce cycled samples with various redox states. If coin cells exhibit different discharge-charge profiles, the electron doping level can be ambiguous. On the other hand, since the cycled Cu-THQ are sensitive to air, their ESR spectra may change significantly when contaminated. Therefore, battery assembling, testing, and post-treatment were performed in a restricted air-free manner and using air-free solvents.

Moreover, the thermal stability of intermediates is another critical point to consider³². In our cases, the annealing effect decreased the concentration of defect radicals during low-temperature ESR measurements. We observed that an annealing effect occurred when the sample was kept under a vacuum or subjected to a cooling/heating cycle. Hence, we presented only the low-temperature results for the electrode discharged to 1.9 V.

Compared to other spectroscopic techniques, such as XPS, X-ray absorption spectroscopy (XAS), and XRD, ESR has an advantage in identifying small local defect radicals and distinguishing paramagnetic species. Furthermore, combining ESR spectroscopy and SQUID magnetometry allows for quantitative monitoring of the evolution of the spin concentration of magnetic centers during electrochemical cycling. However, the active material or its reduced/oxidized products must be magnetic for effective magnetic measurements. It's important to note that non-Kramer ions, such as S = 1: V(III) and Ni(II); S = 2: Cr(II), Mn(III), Fe(II), and Co(III) (unless in a highly symmetric crystal field), cannot be detected by X-band ESR, due to the large energy gap between the singlet ground state and doublet excited state³³.

Quasi 2D fully π-d conjugated MOFs have an ultra-high specific capacity in various electrochemical devices and have been extensively studied using various spectroscopic techniques⁷^,⁸^,⁹. However, understanding the charge storage mechanism in such strongly correlated systems remains incomplete. Electrochemical magnetometry can play an irreplaceable role in elucidating the electrochemical mechanisms of MOFs based on paramagnetic metal centers and free radical ligands. Specifically, through ESR spectroscopy and SQUID magnetometry, we have uncovered an underlying charge storage mechanism involving delocalized π electrons and elucidated the extra capacity in the strongly correlated 2D Cu-THQ MOF. Further efforts should be made to develop a carbon-free and binder-free electrode, such as the conductive substrate covered with a monolayer MOF, for solid-state electrochemistry. This would be a major step toward achieving non-trivial magnetic ground states and physical properties through electrochemical modification.

Divulgaciones

The authors have nothing to disclose.

Acknowledgements

This study was supported by a Japan Society for the Promotion of Science (JSPS) KAKENHI Grant (JP20H05621). Z. Zhang also thanks the Tatematsu Foundation and Toyota Riken scholarship for financial support.

Materials

1-Methyl-2-pyrrolidone	FUJIFILM Wako Chemicals	139-17611	Super Dehydrated
1mol/L LiBF4 EC:DEC (1:1 v/v%)	Kishida	LBG-96533	electrolyte
4-Hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl	FUJIFILM Wako Chemicals	089-04191	TEMPOL, for Spin Labeling
Ampule tube	Maruemu Corporation	5-124-05	20mL
Carbon black, Super P Conductive	Alfa Aesar	H30253
Conductive Carbon Black	Mitsubishi Chemical
Copper (II) Nitrate Trihydrate	FUJIFILM Wako Chemicals	033-12502	deleterious substances
Dimethyl Carbonate	FUJIFILM Wako Chemicals	046-31935	battery grade
Ethylenediamine	FUJIFILM Wako Chemicals	053-00936	deleterious substances
Graphene Nanoplatelets	Tokyo Chemical Industry	G0442	6-8nm(thick), 15µm(wide)
Poly(vinylidene fluoride)	Sigma Aldrich	182702
Potassium Bromide	FUJIFILM Wako Chemicals	165-17111	for Infrared Spectrophotometry
Sodium Alginate	FUJIFILM Wako Chemicals	199-09961	500-600 cP
SQUID Magnetometer	Quantum Design	MPMS-XL 5
Tetrahydroxy-1,4-benzoquinone Hydrate	Tokyo Chemical Industry	T1090
X-Band ESR	JEOL	JES-F A200

Referencias

Lee, J., et al. Metal-organic framework materials as catalysts. Chemical Society Reviews. 38 (5), 1450-1459 (2009).
Dolgopolova, E. A., Rice, A. M., Martin, C. R., Shustova, N. B. Photochemistry and photophysics of MOFs: steps towards MOF-based sensing enhancements. Chemical Society Reviews. 47 (13), 4710-4728 (2018).
Qian, Q., et al. MOF-based membranes for gas separations. Chemical Reviews. 120 (16), 8161-8266 (2020).
Wang, Q., Astruc, D. State of the art and prospects in metal-organic framework (MOF)-based and MOF-derived nanocatalysis. Chemical Reviews. 120 (2), 1438-1511 (2020).
Wang, M., Dong, R., Feng, X. Two-dimensional conjugated metal-organic frameworks (2D c-MOFs): chemistry and function for MOFtronics. Chemical Society Reviews. 50 (4), 2764-2793 (2021).
Baumann, A. E., Burns, D. A., Liu, B., Thoi, V. S. Metal-organic framework functionalization and design strategies for advanced electrochemical energy storage devices. Communications Chemistry. 2 (1), 86 (2019).
Nam, K. W., et al. Conductive 2D metal-organic framework for high-performance cathodes in aqueous rechargeable zinc batteries. Nature Communications. 10 (1), 4948 (2019).
Sheberla, D., et al. Conductive MOF electrodes for stable supercapacitors with high areal capacitance. Nature Materials. 16 (2), 220-224 (2017).
Wang, Z., et al. Ultrathin two-dimensional conjugated metal-organic framework single-crystalline nanosheets enabled by surfactant-assisted synthesis. Chemical Science. 11 (29), 7665-7671 (2020).
Zhang, Z., Yoshikawa, H., Awaga, K. Monitoring the solid-state electrochemistry of Cu(2,7-AQDC) (AQDC = anthraquinone dicarboxylate) in a lithium battery: Coexistence of metal and ligand redox activities in a metal-organic framework. Journal of the American Chemical Society. 136 (46), 16112-16115 (2014).
Zhang, Z., Yoshikawa, H., Awaga, K. Discovery of a "bipolar charging" mechanism in the solid-state electrochemical process of a flexible metal-organic framework. Chemistry of Materials. 28 (5), 1298-1303 (2016).
Li, C., Hu, X., Hu, B. Cobalt(II) dicarboxylate-based metal-organic framework for long-cycling and high-rate potassium-ion battery anode. Electrochimica Acta. 253, 439-444 (2017).
Liu, J., et al. Reversible formation of coordination bonds in Sn-based metal-organic frameworks for high-performance lithium storage. Nature Communications. 12 (1), 3131 (2021).
Jiang, Q., et al. A redox-active 2D metal-organic framework for efficient lithium storage with extraordinary high capacity. Angewandte Chemie. 59 (13), 5273-5277 (2020).
Sakaushi, K., Nishihara, H. Two-dimensional π-conjugated frameworks as a model system to unveil a multielectron-transfer-based energy storage mechanism. Accounts of Chemical Research. 54 (15), 3003-3015 (2021).
Li, H., et al. 2D organic radical conjugated skeletons with paramagnetic behaviors. Advanced Materials Interfaces. 8 (18), 2100943 (2021).
Peeks, M. D., et al. Electronic delocalization in the radical cations of porphyrin oligomer molecular wires. Journal of the American Chemical Society. 139 (30), 10461-10471 (2017).
Krug von Nidda, H. A., et al. Anisotropic exchange in LiCuVO4 probed by ESR. Physical Review B. 65 (13), 134445 (2002).
Zeng, Z., et al. Pro-aromatic and anti-aromatic π-conjugated molecules: An irresistible wish to be diradicals. Chemical Society Reviews. 44 (18), 6578-6596 (2015).
Chen, Q., Adeniran, O., Liu, Z. F., Zhang, Z., Awaga, K. Graphite-like charge storage mechanism in a 2D π-d conjugated metal-organic framework revealed by stepwise magnetic monitoring. Journal of the American Chemical Society. 145 (2), 1062-1071 (2023).
Julien, C. M., Mauger, A., Groult, H., Zhang, X., Gendron, F. LiCo1-yByO2 as cathode materials for rechargeable lithium batteries. Chemistry of Materials. 23 (2), 208-218 (2011).
Niemöller, A., Jakes, P., Eichel, R. A., Granwehr, J. In operando EPR investigation of redox mechanisms in LiCoO2. Chemical Physics Letters. 716, 231-236 (2019).
Park, J., et al. Synthetic routes for a 2D semiconductive copper hexahydroxybenzene metal-organic framework. Journal of the American Chemical Society. 140 (44), 14533-14537 (2018).
Rondeau, R. E. A technique for degassing liquid samples. Journal of Chemical Education. 44 (9), 530 (1967).
Flores-Llamas, H. Inhomogeneously broadened EPR lineshape of axial powder. Applied Magnetic Resonance. 9 (2), 289-298 (1995).
Sun, L., et al. Room-temperature quantitative quantum sensing of lithium ions with a radical-embedded metal-organic framework. Journal of the American Chemical Society. 144 (41), 19008-19016 (2022).
Chen, Y., et al. Successive storage of cations and anions by ligands of π-d-conjugated coordination polymers enabling robust sodium-ion batteries. Angewandte Chemie. 60 (34), 18769-18776 (2021).
Roessler, M. M., Salvadori, E. Principles and applications of EPR spectroscopy in the chemical sciences. Chemical Society Reviews. 47 (8), 2534-2553 (2018).
Ji, X., et al. Pauli paramagnetism of stable analogues of pernigraniline salt featuring ladder-type constitution. Journal of the American Chemical Society. 142 (1), 641-648 (2020).
Noel, M., Santhanam, R. Electrochemistry of graphite intercalation compounds. Journal of Power Sources. 72 (1), 53-65 (1998).
Wu, K. H., Ting, T. H., Wang, G. P., Ho, W. D., Shih, C. C. Effect of carbon black content on electrical and microwave absorbing properties of polyaniline/carbon black nanocomposites. Polymer Degradation and Stability. 93 (2), 483-488 (2008).
Yao, M., Taguchi, N., Ando, H., Takeichi, N., Kiyobayashi, T. Improved gravimetric energy density and cycle life in organic lithium-ion batteries with naphthazarin-based electrode materials. Communications Materials. 1 (1), 70 (2020).
Krzystek, J., et al. EPR spectra from "EPR-silent" species: High-frequency and high-field EPR spectroscopy of pseudotetrahedral complexes of nickel(II). Inorganic Chemistry. 41 (17), 4478-4487 (2002).

Play Video

PDF

DOI

DOWNLOAD MATERIALS LIST

Citar este artículo

Chen, Q., Zhang, Z., Awaga, K. Magnetometric Characterization of Intermediates in the Solid-State Electrochemistry of Redox-Active Metal-Organic Frameworks. J. Vis. Exp. (196), e65335, doi:10.3791/65335 (2023).