Cation dynamics by ¹H and ¹³C MAS NMR in hybrid organic–inorganic (CH₃CH₂NH₃)₂CuCl₄

Abstract

To understand the dynamics of the cation in layered perovskite-type (CH₃CH₂NH₃)₂CuCl₄, the temperature-dependent chemical shifts and spin–lattice relaxation times T_1ρ in the rotating frame have been measured using ¹H magic angle spinning nuclear magnetic resonance (MAS NMR) and ¹³C cross-polarization (CP)/MAS NMR techniques. Each proton and carbon in the (CH₃CH₂NH₃)⁺ cation is distinguished in MAS NMR spectra. The Bloembergen–Purcell–Pound (BPP) curves for ¹H T_1ρ in CH₃CH₂ and NH₃, and for the ¹³C T_1ρ in CH₃ and CH₂ are revealed to have minima at low temperatures. This implies that the curves represent the CH₃ and NH₃⁺ rotational motions. The amplitude of the cationic motion is enhanced at the C-end, that is, the N-end of the organic cation is fixed to the inorganic layer through N–H⋯Cl hydrogen bonds, and T_1ρ becomes short with larger-amplitude molecular motions.

---

I. Introduction

Metal–organic hybrids, which consist of organic and inorganic components, have recently attracted much attention because these materials have many possibilities for the tailoring of their functionalities and physical properties including optical, electrical and magnetic properties by adjusting the organic and/or metal building blocks. Hybrid metal–organic compounds based on the perovskite structures are of increasing interest due to their potential use for solar cells.^1,2 However, toxicity and chemical instability issues of halide perovskites still remain as the main drawbacks for use in solar cells. The crystalline structure of compounds of the type (C_nH_2n+1NH₃)₂MCl₄, where n = 1, 2, 3 … and M represents divalent metals (M = Cu, Cd, …), may be described as a sequence of alternating organic–inorganic layers.^3–6 Many compounds in this family have been extensively investigated and have demonstrated successive phase transitions. This family of materials crystallizes in the layered perovskite structure, which consists of infinite, staggered layers of corner-sharing MCl₆ octahedra interleaved by alkylammonium cations.⁷ Because of the layered character of their structure, these crystals become appropriate substances for investigations of two-dimensional electronic systems. The cavities between the octahedra are occupied by the ammonium heads of the organic cations, which, importantly, form strong N–H⋯Cl hydrogen bonds to any of the eight chloride ions.⁸

Ethylammonium copper chloride (CH₃CH₂NH₃)₂CuCl₄ is a layered perovskite-type compound that undergoes a complicated sequence of phase transitions. Differential scanning calorimetry (DSC) data indicates several phase transitions, at 236 K (=T_C4), 330 K (=T_C3), 357 K (=T_C2), and 371 K (=T_C1), as temperature increases.^9–14 The peaks at 236 K, 330 K, and 371 K are very weak and can perhaps correspond to second-order transformations.¹³ The phase transitions in this crystal are mostly connected with changes in the arrangement of the alkylammonium chains. Fig. 1 shows the room-temperature orthorhombic crystal structure of (CH₃CH₂NH₃)₂CuCl₄.^8,15 The hybrids have the orthorhombic crystal structure with the space group Pbca, and the lattice constants are a = 7.47 Å, b = 7.35 Å, and c = 21.18 Å at room temperature.¹⁶ The CuCl₆ octahedra are strongly distorted with elongated Cu–Cl bonds orthogonal to each other on adjacent octahedra. The CuCl₆ sheets are sandwiched between two layers of alkylammonium. The structure of the organic component consists of a double layer of alkylammonium ions with their charged ends, the nitrogen atoms, oriented to the nearest CuCl₆ plane.⁴ The complete structure is constituted by corner-sharing CuCl₆ octahedra, forming the inorganic layers, and bilayers of organic cations attached to the octahedra by their NH₃ heads.^17,18

Fig. 1 Orthorhombic structure of a (CH₃CH₂NH₃)₂CuCl₄ crystal at room temperature.

The structural geometry and molecular motions of the organic molecules within the layered hybrid structure is important for determining the influence of temperature on the evolution of the structural phase transitions in the perovskite structure. Physical properties in particular depend on the characteristics of metallic anion and the organic cation.

In the present study, the crystal structure and thermal stability for (CH₃CH₂NH₃)₂CuCl₄ was observed by means of conventional X-ray, thermogravimetric analysis (TGA), and optical polarizing microscopy. In order to clarify the structural geometry and dynamics of the cation in the organic–inorganic (CH₃CH₂NH₃)₂CuCl₄, we investigated the chemical shifts and the spin–lattice relaxation time T_1ρ in the rotating frame using ¹H magic angle spinning nuclear magnetic resonance (MAS NMR) and ¹³C cross-polarization (CP)/MAS NMR. The CH₃CH₂ and NH₄ groups of the CH₃CH₂NH₃ cation are distinguishable in ¹H MAS NMR spectra, and the CH₃ and CH₂ groups are distinguished by ¹³C CP/MAS NMR spectra. We investigated the ¹H and ¹³C dynamics in the (CH₃CH₂NH₃)⁺ cation near the phase-transition temperatures.

II. Experimental method

Crystals of (CH₃CH₂NH₃)₂CuCl₄ were obtained by slow evaporation at 25 °C from an aqueous solution of C₂H₅NH₂·HCl and CuCl₂·2H₂O in the stoichiometric 2 : 1 proportion. The obtained crystals were yellow square plates, typically 5 mm × 5 mm in area and 0.5 mm in thickness.

The structure of the (CH₃CH₂NH₃)₂CuCl₄ crystals was determined at room temperature with an X-ray diffraction system (PANalytical, X'pert pro MPD) with a Cu-Kα (λ = 1.5418) radiation source. Measurements were taken in a θ–2θ geometry from 10° to 60° at 45 kV and with a tube power of 40 mA. And, the TGA curve at a heating rate of 10 °C min⁻¹ was measured under N₂ atmosphere, and the mass of the powdered sample used in the TGA experiment was 11.41 mg.

The chemical shifts and the T_1ρ values for (CH₃CH₂NH₃)₂CuCl₄ were obtained by ¹H MAS NMR and ¹³C CP/MAS NMR at Larmor frequencies of ω₀/2π = 400.13 and 100.61 MHz, respectively, using Bruker 400 MHz NMR spectrometers at the Korea Basic Science Institute, Western Seoul Center. Crystalline powdered samples were placed within a 4 mm CP/MAS probe, and the MAS rate for ¹H and ¹³C measurements, to minimize spinning sideband overlap, was set to 10 kHz. The ¹H T_1ρ values were determined using a π/2−t sequence by varying the duration of spin-locking pulses. ¹³C T_1ρ values were measured by varying the duration of the spin-locking pulse applied after the CP preparation period. The width of the π/2 pulse used for measuring T_1ρ for ¹H and ¹³C was 3.7 μs, with the spin-locking field at 67.56 kHz. The chemical shifts and T_1ρ were measured over a temperature range of 180–430 K.

III. Experimental results

The measured structure at room temperature exhibited orthorhombic symmetry with cell parameters of a = 7.480 Å, b = 7.375 Å, c = 21.254 Å for (CH₃CH₂NH₃)₂CuCl₄ crystal. This result is consistent with the results reported by Steadman and Willett.¹⁶

The TGA curve of (CH₃CH₂NH₃)₂CuCl₄ is shown in Fig. 2 for measuring thermal stability. The first occurrence of mass loss begins at approximately 430 K (T_d), which is the onset of partial thermal decomposition. The second weight loss of 25.1% near 530 K is due to the removal of the CH₃CH₂NH₃Cl from the compound, leaving intermediate CH₃CH₂NH₃CuCl₃ that belongs to another known class of compounds ABX₃. Near 560 K, CuCl₂ remains as the residue and when it reaches 580 K, the total weight loss becomes 65.55%. The color of the crystal is dark yellow at room temperature although it has slightly inhomogeneous hue due to surface roughness. As the temperature increases, the color of the crystal varies from dark yellow (300 K, 350 K), brown (400 K), to dark brown (450 K, 500 K), and then they start melting at 530 K as shown in the inset in Fig. 2. The TGA and optical polarizing microscopy results show that the crystal above 430 K allows CH₃ to partially escape by the breaking the weak C–N bond.

Fig. 2 Thermogravimetric analysis of (CH₃CH₂NH₃)₂CuCl₄ (inset: color changes of a (CH₃CH₂NH₃)₂CuCl₄ crystal according to the temperature): (a) 300 K, (b) 350 K, (c) 400 K, (d) 450 K, (e) 500 K, and (f) 530 K.

The ¹H NMR spectra at a frequency of 400.13 MHz were obtained by MAS NMR. The ¹H spectrum recorded at room temperature is shown in the inset in Fig. 3; the spectrum shows two peaks at chemical shifts of δ = 0.23 and 12.12 ppm, which are assigned to the protons of the CH₃CH₂ and NH₃ groups, respectively. The spinning sidebands for CH₃CH₂ are marked with asterisks and those for NH₃ are marked with open circles. However, the different ¹H signals from CH₃ and CH₂ cannot be resolved, and therefore the combined CH₃CH₂ peak is very broad and has a larger intensity due to the overlap of the CH₃ and CH₂ peaks. The peak with the lower chemical shift is attributed to the protons in CH₃CH₂, and that of the higher chemical shift is attributed to the protons in NH₃. The ¹H chemical shifts for the alkyl and ammonium groups slowly and monotonously vary with temperature, indicating that the surrounding environments of the protons in the alkyl and ammonium groups change continuously, as shown in Fig. 3; here, the chemical shifts for protons in CH₃CH₂ and NH₃ near T_C1, T_C2, and T_C3 are nearly constant with temperature, whereas those for protons in CH₃CH₂ and NH₃ below T_C4 change more abruptly.

Fig. 3 Chemical shifts for ¹H MAS NMR of (CH₃CH₂NH₃)₂CuCl₄ as a function of temperature (inset: ¹H MAS NMR spectrum of (CH₃CH₂NH₃)₂CuCl₄ at 300 K with spinning sidebands indicated by asterisks and open circles).

The T_1ρ values for the CH₃CH₂ and NH₃ protons in (CH₃CH₂NH₃)₂CuCl₄ were obtained as a function of temperature. The magnetization traces of both the alkyl and ammonium protons may be described by a single exponential function^19–21

S(t)/S₀ = exp(−t/T_1ρ), (1)

where S(t) is the magnetization as a function of the spin-locking pulse duration t, and S₀ is the total nuclear magnetization of the proton at thermal equilibrium.¹⁹ The recovery curves for several delay times were measured, and the T_1ρ values were obtained from the slopes by the delay time vs. intensity, at several different temperatures. This analysis method was used to obtain the T_1ρ values for each proton in CH₃CH₂ and NH₃ which are plotted as a function of inverse temperature in Fig. 4. The T_1ρ values for the CH₃CH₂ and NH₃ protons in the (CH₃CH₂NH₃)⁺ cations exhibit similar trends with temperature. The proton T_1ρ data do not show evidence of a change near the phase-transition temperature; the T_1ρ values of protons in the CH₃CH₂ and NH₃ groups of (CH₃CH₂NH₃)₂CuCl₄ are almost continuous near T_C1, T_C2, and T_C3, and these values are of the order of few milliseconds. The T_1ρ values abruptly decreased with temperature in the region approaching T_C4. The relaxation time for the ¹H nucleus is minimal at 190 K and 200 K for CH₃CH₂ and NH₃, respectively. This feature of T_1ρ indicates that distinct molecular motions are present. The T_1ρ values are related to the corresponding values of the rotational correlation time, τ_C, which is a direct measure of the rate of molecular motion. For the spin–lattice relaxation time in the rotating frame, the experimental value of T_1ρ can be expressed in terms of the correlation time τ_C for the molecular motion, as suggested by the Bloembergen–Purcell–Pound (BPP) function:^19,22

T_1ρ⁻¹ = (N/20)(γ_Hγ_Cħ/r_H–C³)²{4τ_C/(1 + ω₁²τ_C²) + τ_C/[1 + (ω_H − ω_C)²τ_C²] + 3τ_C/[1 + (ω_C²τ_C²)] + 6τ_C/[1 + (ω_H + ω_C)²τ_C²] + 6τ_C/[1 + ω_H²τ_C²]}. (2)

Fig. 4 ¹H spin–lattice relaxation times T_1ρ in the rotating frame for the CH₃CH₂ and NH₃ groups of (CH₃CH₂NH₃)₂CuCl₄ as a function of inverse temperature.

Here, γ_H and γ_C are the gyromagnetic ratios for the ¹H and ¹³C nuclei, respectively; N is the number of directly bound protons; r_H–C is the H–C internuclear distance; ħ is the reduced Planck constant; ω_H and ω_C are the Larmor frequencies of ¹H and ¹³C, respectively; and ω₁ is the frequency of the spin-locking field. We analyzed our data assuming that T_1ρ would show a minimum when ω₁τ_C = 1, and that the BPP relation between T_1ρ and the characteristic frequency ω₁ could be applied. We sensitively controlled the minima in the T_1ρ temperature variations and the slopes around the minima. From these results, the value of (γ_Hγ_Cħ/r_H–C³)² for the proton in eqn (2) was obtained. We then calculated the temperature dependences of the τ_C values for protons by using the obtained values of (γ_Hγ_Cħ/r_H–C³)². The temperature dependence of τ_C follows a simple Arrhenius equation:

τ_C = τ₀ exp(−E_a/RT), (3)

where τ₀ is a pre-exponential factor, T is the temperature, R is the gas constant, and E_a is the activation energy. Thus, the slope of the linear portion of a semi-log plot should yield E_a. The E_a value for the rotational motion can be obtained from the log τ_C vs. 1000/T curve shown in Fig. 5; we obtained E_a = 12.19 ± 1.30 kJ mol⁻¹ and E_a = 8.33 ± 0.50 kJ mol⁻¹ for CH₃CH₂ and NH₃, respectively. The rotational motion for alkyl groups is activated, whereas the rotational motion for ammonium groups at the end of the organic cation is less strongly activated.

Fig. 5 Arrhenius plots of the natural logarithm of the correlation time for each ¹H and ¹³C of (CH₃CH₂NH₃)₂CuCl₄ as a function of inverse temperature.

The structural analysis of the carbons in (CH₃CH₂NH₃)₂CuCl₄ was performed by ¹³C CP/MAS NMR, and the corresponding spectrum is shown in Fig. 6, as a function of temperature; the ¹³C CP/MAS NMR spectrum at room temperature shows two signals at chemical shifts of δ = 50.77 ppm and δ = 113.50 ppm with respect to tetramethysilane (TMS), which can be assigned to CH₃ and CH₂, respectively. The ¹³C chemical shift of CH₂ abruptly shifts with temperature, whereas that of CH₃ changes only much less with temperature. The full width at half maximum (FWHM) linewidths for the ¹³C of CH₃ and CH₂ in Fig. 7 showed a monotonic decrease with increasing temperature, with no particular anomalies attributable to the phase transitions. The linewidth of the ¹³C signal assigned to CH₃ is broad compared to that of CH₂, and the linewidth narrows significantly with increasing temperature. This narrowing of the ¹³C linewidths is attributed to internal motions that the line widths follow the same temperature dependence as some internal motions, hence the motions are responsible for the line widths.

Fig. 6 ¹³C CP/MAS NMR spectra of (CH₃CH₂NH₃)₂CuCl₄ measured at different temperatures.

Fig. 7 Temperature dependences of line widths of ¹³C NMR spectra of CH₃ and CH₂ in (CH₃CH₂NH₃)₂CuCl₄.

To obtain the ¹³C T_1ρ values, the nuclear magnetization was also measured at several temperatures as a function of delay time. The signal intensity of the nuclear magnetization recovery curves for ¹³C is described by a single exponential function as in eqn (1) at all temperatures. The ¹³C T_1ρ values for CH₃ and CH₂ in (CH₃CH₂NH₃)₂CuCl₄ are plotted as a function of inverse temperature in Fig. 8. The temperature dependences of the ¹³C MAS NMR T_1ρ values seem to be similar. The T_1ρ values for CH₃ and CH₂ both increase with temperature in the same manner; whereas, the ¹³C T_1ρ values near the phase-transition temperatures are approximately continuous. The T_1ρ values for CH₃ and CH₂ at room temperature are 33.85 ms and 109.40 ms, respectively. The amplitude of the cationic motion is enhanced at its CH₃ end, and the central CH₂ moiety is fixed to the NH₃ group in the organic cation. The T_1ρ curve below T_C4 can be reproduced by BPP theory. The BPP curves for CH₃ and CH₂, showing minima at low temperatures, is almost the same as those of the CH₃CH₂ and NH₃ shifts of the ¹H MAS NMR measurements. E_a for the rotational motion of CH₃ and CH₂ can be obtained from the log τ_C vs. 1000/T curve shown in Fig. 5; we obtained E_a = 21.35 ± 0.45 kJ mol⁻¹ for CH₃ and E_a = 19.72 ± 1.76 kJ mol⁻¹ for CH₂, respectively, which, considering their error ranges, are the same values.

Fig. 8 ¹³C spin–lattice relaxation times T_1ρ in the rotating frame for CH₃ and CH₂ in (CH₃CH₂NH₃)₂CuCl₄ as a function of inverse temperature.

IV. Conclusion

We discuss the molecular motions for cation of Cu-based hybrid materials, where we replace Pb with nontoxic Cu metal for lead-free perovskite solar cells, and investigate their potential toward solar cell applications based on ionic dynamics of the cation in hybrid organic–inorganic (CH₃CH₂NH₃)₂CuCl₄ by NMR studies. The cation dynamics and interionic interactions through hydrogen bonds are expected to be closely related with the physical properties due to the potential applications. The cation dynamics in a layered perovskite-type (CH₃CH₂NH₃)₂CuCl₄ were investigated as a function of temperature by ¹H MAS NMR and ¹³C CP/MAS NMR experiments. The CH₃CH₂ and NH₄ units in the CH₃CH₂NH₃ cation were distinguished by the ¹H MAS NMR spectra, and the CH₃ and CH₂ units in the CH₃CH₂NH₃ cation were also clearly distinguished in the ¹³C CP/MAS NMR spectra. To obtain detailed information about the cation dynamics of this crystal, the spin–lattice relaxation time T_1ρ in the rotating frame for both ¹H and ¹³C were measured, revealing that these atoms undergo rotational motions at low temperatures. The BPP curves for the ¹H T_1ρ in CH₃CH₂ and NH₃, and for the ¹³C T_1ρ in CH₃ and CH₂, were shown to have a minimum at low temperatures; the T_1ρ of ¹H and ¹³C showed a minimum and is governed by the tumbling motion of the CH₃CH₂ and NH₃ groups, indicating that the ¹H and ¹³C atoms in the CH₃CH₂NH₃⁺ groups exhibit high mobility at low temperatures. The molecular motions for ¹H and ¹³C in the CH₃CH₂NH₃⁺ cation were very free at low temperatures. T_1ρ provides insight into the changes in the cation reorientation rates at low temperature.

The ¹³C T_1ρ values in CH₃ increased with temperature, a trend that has been observed in alkyl chains attached to the (CH₃CH₂NH₃) cation due to its greater mobility toward its free end. The CH₃CH₂NH₃ cationic motion is enhanced at the opposing end of the cation to the NH₄⁺ group probably because this group is bound to the inorganic layer through the N–H⋯Cl hydrogen bonds. The ¹³C T_1ρ is usually dominated by the fluctuation of the anisotropic chemical shift, and it becomes shorter with larger-amplitude molecular motions. This implies that the amplitude of the cationic motion is enhanced at the C-end, that is, the N-end of the organic cation is fixed at the inorganic layer through N–H⋯Cl hydrogen bonds. The cationic motion, being associated with the fluctuation of the molecular axis, is expected to be gradually excited with increasing temperature.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the Basic Science Research program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science, and Technology (2018R1D1A1B07041593 and 2016R1A6A1A03012069).

References

1. A. M. Elseman, A. E. Shalan, S. Sajid, M. M. Rashad, A. M. Hassan and M. Li, ACS Appl. Mater. Interfaces, 2018, 10, 11699.
2. J. A. Aramburu, P. Garcia-Fernandez, N. R. Mathiesen, J. M. Garcia-Lastra and M. Moreno, J. Phys. Chem. C, 2018, 122, 5071.
3. A. Caretta, M. C. Donker, A. O. Polyakov, T. T. M. Palstra and P. H. M. van Loosdrecht, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 91, 20405.
4. E. Suprayoga, A. A. Nugroho, A. O. Polyakov, T. T. M. Palstra and I. Watanabe, J. Phys.: Conf. Ser., 2014, 551, 12054.
5. J. Ding, H. Li, L. Wen, X. Kang, H. Li and J. Zhang, J. Magn. Magn. Mater., 2013, 346, 91.
6. R. Lefi, F. B. Naser and H. Guermazi, J. Alloys Compd., 2017, 696, 1244.
7. K. Ohwada, K. Ishii, T. Inami, Y. Murakami, T. Shobu, H. Ohsumi, N. Ikeda and Y. Ohishi, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 14123.
8. B. Kundys, A. Lappas, M. Viret, V. Kapustianyk, V. Rudyk, S. Semak, Ch. Simon and I. Bakaimi, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 224434.
9. V. Kapustianyk, V. Rudyk and M. Partyka, Phys. Status Solidi B, 2007, 244, 2151.
10. V. B. Kapustianik, V. V. Bazhan and Yu. M. Korchak, Phys. Status Solidi B, 2002, 234, 674.
11. W. Kleemann, F. J. Schafer, E. Karajamaki, R. Laiho and T. Levola, Physica B, 1983, 119, 269.
12. V. Kapustianik, Yu. Korchak, I. Polovinko, R. Tchukvinskyi, Z. Czapla and S. Dacko, Phys. Status Solidi B, 1998, 207, 95.
13. C. B. Mohamed, K. Karoui, F. Jomni, K. Guidara and A. B. Rhaiem, J. Mol. Struct., 2015, 1082, 38.
14. I. R. Jahn, K. Knorr and J. Ihringer, J. Phys.: Condens. Matter, 1989, 1, 6005.
15. P. Zolfaghari, G. A. de Wijs and R. A. de Groot, J. Phys.: Condens. Matter, 2013, 25, 295502.
16. J. P. Steadman and R. D. Willett, Inorg. Chim. Acta, 1970, 4, 367.
17. A. O. Polyakov, A. H. Arkenbout, J. Baas, G. R. Blake, A. Meetsma, A. Caretta, P. H. M. van Loosdrecht and T. T. M. Palstra, Chem. Mater., 2012, 24, 133.
18. A. Caretta, R. Miranti, R. W. A. Havenith, E. Rampi, M. C. Donker, G. R. Blake, M. Montagnese, A. O. Polyakov, R. Broer, T. T. M. Palstra and P. H. M. van Loosdrecht, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 89, 24301.
19. J. L. Koenig, Spectroscopy of Polymers, Elsevier, New York, 1999.
20. V. J. Mcbrierty and K. J. Packer, Nuclear Magnetic Resonance in Solid Polymers, Cambridge University Press, Cambridge, 1993.
21. A. R. Lim, RSC Adv., 2018, 8, 18656.
22. A. R. Lim, RSC Adv., 2017, 7, 55276.

---