Article pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Synthesis and Characterization of Cu−Ni Mixed Metal Paddlewheels Occurring in the Metal−Organic Framework DUT-8(Ni0.98Cu0.02) for Monitoring Open-Closed-Pore Phase Transitions by X‑Band Continuous Wave Electron Paramagnetic Resonance Spectroscopy Matthias Mendt,*,† Sebastian Ehrling,‡ Irena Senkovska,‡ Stefan Kaskel,‡ and Andreas Pöppl† †
Felix Bloch Institute for Solid State Physics, Leipzig University, Linnéstrasse 5, 04103 Leipzig, Germany Department of Inorganic Chemistry, Dresden University of Technology, Bergstrasse 66, 01062 Dresden, Germany
Inorg. Chem. Downloaded from pubs.acs.org by EAST CAROLINA UNIV on 03/15/19. For personal use only.
‡
S Supporting Information *
ABSTRACT: A Cu2+-doped metal−organic framework (DUT-8(Ni0.98Cu0.02), M2(NDC)2DABCO, M = Ni, Cu, NDC = 2,6-napththalene dicarboxylate, DABCO = 1,4diazabicyclo[2.2.2]octane, DUT = Dresden University of Technology) was synthesized in the form of large (>1 μm) and small crystals ( gz′ as they are typical for Kramer doublets of half-integer spin species with zfs parameters much larger than the electron Zeeman interaction.57,58 Samples 1sol and 1act show signals J
Figure 4. Experimental EPR spectra of the solvated flexible sample 2sol (a) as well as the activated flexible sample 2act (b) and the subsequently resolvated flexible sample 2res (c). All spectra were measured at T = 14 K. Signals assigned to Ni2+-Cu2+ dimers are labeled with M, N, and O. Other minor EPR signals highlighted in the figure are assigned in the Supporting Information to minor species of defect or impurity nature.
As already mentioned, signals J, K, L, M, N, and O have the typical line shapes of EPR transitions which can be described by effective spins S′ = 1/2. Such signals typically originate from Kramer doublets of half-integer spin species.57,58 Integer spin species would give signals with different line shapes, since they cannot be described by an effective spin S′ = 1/2 formalism.66−68 The gx,y′ parameters of the respective EPR signals J to O are much too large for a monomeric transition metal ion species such as Cu2+, Ni+, or low spin Ni3+ with an electron spin S = 1/2. For such a species one would expect g-tensor E
DOI: 10.1021/acs.inorgchem.9b00123 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry principal values slightly larger than the g-value of the free electron spin ge = 2.0023 since in a corresponding metal ion complex contributions from the orbital momentum are quenched by the ligand field and contribute only to a small amount to the g-values via the admixture of higher molecular states by spin−orbit coupling.58 Except for some minor Co2+ impurities from the nickel source, divalent nickel and copper ions are the only metallic ingredients of samples 1sol, 1act, 1res, 2sol, 2act, and 2res. Monomeric Co2+ as well as Ni3+ have a 3d7 electron configuration and can therefore have either a low spin S = 1/2 configuration70 or a high spin S = 3/2 configuration.71 But both transition metal ions are expected to occur in the samples only as minor impurities species, if any. Due to the comparably large EPR intensities of the signals J, K, L, M, N, and O their explanation by species containing Co2+ or Ni3+ ions are not reasonable, as it will be further justified in a paper in progress about Co2+ doped DUT-8(Ni) and as it is already indicated by a previous publication on the monometallic DUT8(Ni).35 The only remaining reasonable interpretation of the signals J to O is their assignment to Ni2+-Cu2+ dimers. Monomeric Ni2+ has a 3d8 electron configuration and can therefore have an electron spin S = 1.57 Monomeric Cu2+ has a 3d9 electron configuration and thus an electron spin S = 1/2. As a consequence, Ni2+-Cu2+ can couple ferromagnetically having an S = 3/2 ground state or antiferromagnetically having an S = 1/2 ground state. Since we did not observe intense S = 1/2 signals for samples 1sol, 1act, 1res, 2sol, 2act, and 2res indicative for low spin Ni2+-Cu2+ dimers, we assign all signals J to O to characteristic S = 3/2 species of ferromagnetically coupled Ni2+-Cu2+ dimers. In particular, the effective g-values of those signals proof that these signals originate from the corresponding |mS = ± 1/2⟩ Kramer doublets.60,72 We analyze now the observed signals J to O in more detail: Signal J of the rigid solvated sample 1sol at T = 14 K can be simulated by a single Ni2+-Cu2+ (S = 3/2) species A1sol as shown in Figure 5a. Its zfs tensor is almost axially symmetric within the spectral resolution, and its g-tensor has principal values gx,y ≈ 1.97 (Table 1). Signal K of the rigid activated samples 1act at T = 14 K can be simulated by a single Ni2+-Cu2+ (S = 3/2) species A1act (Figure 5b). Again, its zfs tensor is almost axially symmetric within the spectral resolution, and its g-tensor principal values gx,y ≈ 2.00 are slightly larger than those of species A1sol (Table 1). The spectrum in Figure 5b shows also signals U1act and V1act from minor impurity species (for an assignment see Supporting Information). Signal L of the rigid DMF solvated sample 1res at T = 7 K shows at its high field end a small shoulder (Figure 5c). The temperature dependence of signal L in the Supporting Information Figure S7 shows that this shoulder remains at significant intensity until T = 35 K. However, at this temperature the intense maximum of L at B = 165 mT has been distinctly decreased relatively to this shoulder. We therefore attribute signal L to a superposition of the EPR signals of two Ni2+-Cu2+ (S = 3/2) species A1res and B1res with principal values gx,y ≈ 1.94 and gx,y ≈ 1.75 and almost axially symmetric zfs tensors (see Figure 5c and Table 1). The gx,y′ powder edge singularity of signal M, occurring in EPR spectra of the solvated flexible sample 2sol at T = 14 K, shows also a small shoulder at its high field end (Figure 6a,b). This small splitting can be interpreted in two ways. In scenario I, signal M is a superposition of the EPR signals of two Ni2+-
Figure 5. Experimental and simulated EPR spectra of (a) the DMF solvated rigid sample 1sol at T = 14 K, (b) the activated rigid sample 1act at T = 14 K, and (c) the DMF resolvated rigid sample 1res measured at T = 7 K. In (a) the simulated signal of species A1sol is shown. In (b) the simulated signal (sum) is a superposition of the simulated signals of the electron spin S = 3/2 species A1act and the electron spin S = 1/2 species U1act and V1act (see Supporting Information for assignment) as indicated in the figure. In (c) the simulated signal (sum) is a superposition of the S = 3/2 species A1res and B1res. Minor EPR active species, highlighted in (a) and (c), are assigned in the Supporting Information to species of defect or impurity nature.
Cu2+ (S = 3/2) species A2sol,I and B2sol,I with almost axially symmetric zfs tensor (see Figure 6a and Table 1). Their gtensor principal values are gx,y ≈ 2.02 and gx,y ≈ 1.9, respectively (Table 1). In scenarios II, signal M is explained as the EPR signal of a single Ni2+-Cu2+ (S = 3/2) species A2sol,II F
DOI: 10.1021/acs.inorgchem.9b00123 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Table 1. Experimentally Derived Spin Hamiltonian Parameters of Different Species Explaining the EPR Signals J, K, L, M, N, and O in the Samples 1sol, 1act, 1res, 2sol, 2act, and 2res, Respectivelya sample/signal
scenario
DUT-8(Cu) 1sol/J 1act/K 1res/L 2sol/M
I
2act/N
II I II III
2res/O
I II
species
gx,yb
gz
D (MHz)
E/D
X (S = 1) A1sol (S = 3/2) A1act (S = 3/2) A1res (S = 3/2) B1res (S = 3/2) A2sol,I (S = 3/2) B2sol,I (S = 3/2) A2sol,II (S = 3/2) B2act,I (S = 3/2) C2act,I (S = 3/2) C2act,II (S = 3/2) D2act,III (S = 3/2) E2act,III (S = 3/2) A2res,I (S = 3/2) B2res,I (S = 3/2) A2res,II (S = 3/2)
2.065 ± 0.015 1.97 ± 0.02 2.00 ± 0.02 1.94 ± 0.04 1.70 ± 0.05 2.02 ± 0.02 1.88 ± 0.12 1.97 ± 0.02 1.86 ± 0.1 2.23 ± 0.03 2.24 ± 0.04 2.52 ± 0.04 1.85 ± 0.04 2.01 ± 0.05 1.85 ± 0.15 1.91 ± 0.04
2.38 ± 0.01 24000c >23000c >20000c
