Patent Application
20240230171
Currently, more than 20% of the world's electricity is used for cooling, including for the storage of foods and medicine, air conditioning of populated spaces and buildings, and more recently, removing heat from electronics and data centers. With the world's increasing population demanding more comfortable living conditions and smaller, faster, and more powerful electronics, the demand for cooling is expected to continue to increase. Conventional cooling technologies, which have been around since the early 19th century, rely primarily on vapor compression cycles of hydrofluorocarbons that have a high global warming potential. In addition, next-generation electronics require aggressive cooling for sufficient thermal management, but current cooling methods cannot be scaled down to the dimensions of microchips. Thus, there is a pressing need to find new cooling technologies that are environmentally friendly and amenable to miniaturization.
Solid-state cooling based on caloric materials offers the potential to overcome many challenges associated with traditional cooling technologies. Caloric materials are a class of solids that undergo solid-solid phase transitions driven by magnetic, electrical, or mechanical stimuli. In a typical cooling cycle, the stimulus is applied adiabatically to induce a phase change—typically to a more ordered state—which leads to a large increase in temperature. Once the temperature has re-equilibrated by dissipating excess heat to a heat sink, the stimulus is removed, reverting the caloric material to its original phase—now at a lower temperature—that can be used to remove heat from a heat source. Because these phase transitions occur entirely in the solid state, refrigeration cycles can be achieved without using greenhouse gases. Until recently, materials exhibiting magnetocaloric, elastocaloric, and electrocaloric effects have been viewed as most promising. However, in magnetocaloric cooling, the requirement of a large magnetic field (H>2 T) and the cost of expensive rare-earth magnetic materials have prevented widespread industrial and commercial applications. In elastocaloric materials, a short fatigue life has limited their utilization. Electrocaloric materials have also lagged behind as they require the energetically expensive production of electric fields, whose value is limited by the breakdown field.
To date, many materials such as natural rubbers, shape-memory alloys and intermetallic compounds, antiferromagnetic compound (Mn3GaN), ionic conductors (AgI), ferroelectric ceramic (BaTiO3), ferrielectric organic salts, organic molecule-based switchable dielectrics, 3D hybrid perovskites (with a general chemical formula of ABX3), and organic plastic crystals have been explored as barocaloric materials. However, these materials are often not ideal for practical refrigeration due to low thermodynamic efficiencies, small entropy changes associated with pressure-induced transitions (dozens of joules per kilogram per kelvin), small latent heats, the need to operate at non-ambient temperatures, large thermal hysteresis, high operating pressures, and a lack of mechanical durability and synthetic tunability. Most importantly, rational synthetic manipulation of these materials to establish fundamental structure-property relationships and to improve their performance is difficult.
Thus, there is a need for new methods, compositions, and systems for barocaloric applications.
The invention provides methods, compositions, and systems for barocaloric applications, e.g., cooling, heating, and energy storage.
In one aspect, the invention provides a method of heating or cooling employing a barocaloric cycle including providing heat energy to a composition including an organic layer including optionally substituted C>3 alkyl chains (e.g., C>4 alkyl chains), wherein the organic layer is in a disordered state and wherein the organic layer is between first and second inorganic layers or includes a head group capable of hydrogen bonding, halogen bonding, and/or electrostatic interaction with a counterion (e.g., an anion such as a halide); applying compression to the composition to induce the organic layer to undergo an exothermic phase transition to an ordered state, releasing latent heat; removing the latent heat while the composition is compressed; and removing the compression to allow the composition to revert to the disordered state.
In some embodiments, the composition includes first and second inorganic layers separated by the organic layer. In some embodiments, the compression is hydrostatic or mechanical and/or the latent heat is removed by a heat sink. In some embodiments, the organic layer includes a C>3 alkyl ammonium species (e.g., C>4 alkyl, e.g., C4-36, e.g., C4-18), such as a species selected from:
In some embodiments, the organic layer is an organic bilayer. In some embodiments, the organic layer includes a compound of formula (CnH2n+1)(CmH2m+1)NH2X, where n is 1-3 or 4-36 and m is 4-36; and X is a monoanionic species (e.g., a halide (e.g., F, Cl, Br, or I) or a non-halide anion, such as NO3—, ClO3—, ClO4—, H2PO4—, HSO4—, CN—, HCOO—, N3—, N(CN)2—, BF4—, BH4—, PF6—, SCN—, or OCN—). In certain embodiments, n=m and n=4-36. In certain embodiments, n=1-3 and m=4-36 (e.g., where n=1 and m=6, 8, 10, 12, or 18).
In some embodiments, the composition is a 2D perovskite. In some embodiments, the 2D perovskite includes a transition metal halide. In some embodiments, the 2D perovskite includes Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg. In some embodiments, the 2D perovskite includes a tetrahedral or octahedral transition metal complex. In some embodiments, the halide of the transition metal halide is F, Cl, Br, or I. In some embodiments, the transition metal halide includes a monovalent metal cation and a trivalent metal cation. In some embodiments, the 2D perovskite is of formula [(R1)x(R2)1−x]2MXyX′4−y, where R1 and R2 are independently optionally substituted alkylammonium species, X and X′ are different halides, x is a real number between 0-1, y is 0-4, M is a transition metal (e.g., Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg); and where if y=0 or 4, R1≠R2 and x≠0 or 1.
In some embodiments, the organic layer includes two different molecular structures. In some embodiments, the first and second inorganic layers include a silicate. In some embodiments, the composition includes a metal alkyl phosphonate salt. In some embodiments, the composition includes a compound of the following table:
DA=decylammonium, NA=nonylammonium, and UA=undecylammonium.
In some embodiments, the compression results from a pressure change of less than 500 bar, e.g., less than 300 bar. In some embodiments, the compression results in a reversible entropy change of more than 200 J kg−1 K−1.
In some embodiments, the compression is provided using a pressure transmitting medium (PTM). In some embodiments, the method includes providing a gas to the PTM that induces a change in a thermal property of the composition. In some embodiments, the change in thermal property is a lowering of a phase transition temperature and/or a barocaloric effect inversion. In some embodiments, the method further includes removing the gas from the PTM. In some embodiments, the gas is an inert gas that permeates into a free volume of the organic layer. In some embodiments, permeated gas interacts with the composition. In some embodiments, permeation and interaction of the gas with the composition together induce a lowering of a phase transition and/or a barocaloric effect inversion. In some embodiments, the gas is nitrogen, argon, krypton, xenon, methane, ethane, propane, butane, sulfur hexafluoride, or carbon dioxide.
In an aspect, the invention provides a method of storing thermal energy employing by providing a composition including an organic layer including optionally substituted C>3 alkyl chains (e.g., C>4 alkyl chains, e.g., C4-36 alkyl chains) at a first temperature and a first pressure, wherein the composition is in an ordered state and wherein the organic layer is between first and second inorganic layers or includes a head group capable of hydrogen bonding, halogen bonding, and/or electrostatic interaction with a counterion; and reducing compression on the composition to a second pressure to induce a phase transition in the composition to a disordered state, thereby storing energy.
In some embodiments, the method further includes increasing compression on the composition to apply a third pressure to revert the composition to an ordered state and release heat energy. In some embodiments, the composition includes first and second inorganic layers separated by the organic layer. In some embodiments, the compression is hydrostatic or mechanical. In some embodiments, the organic layer includes a C>3 alkyl (e.g., C>4 alkyl, e.g., C4-36 alkyl) ammonium species, such as:
In some embodiments, the organic layer is an organic bilayer. In some embodiments, the organic layer includes a compound of formula (CnH2n+1)(CmH2m+1)NH2X, where n is 1-3 or 4-36 and m is 4-36; and X is a monoanionic species (e.g., a halide (e.g., F, Cl, Br, or I) or a non-halide anion, such as NO3—, ClO3—, ClO4—, H2PO4—, HSO4—, CN—, HCOO—, N3—, N(CN)2—, BF4—, BH4—, PF6—, SCN—, or OCN—). In certain embodiments, n=m and n=4-36. In certain embodiments, n=1-3 and m=4-36 (e.g., where n=1 and m=6, 8, 10, or 12).
In some embodiments, the composition is a 2D perovskite. In some embodiments, the 2D perovskite includes a transition metal halide. In some embodiments, the 2D perovskite includes Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg. In some embodiments, the 2D perovskite includes a tetrahedral or octahedral transition metal complex. In some embodiments, the halide of the transition metal halide is F, Cl, Br, or I. In some embodiments, the transition metal halide includes a monovalent metal cation and a trivalent metal cation. In some embodiments, the 2D perovskite is of formula [(R1)x(R2)1−x]2MXyX′4−y, where R1 and R2 are independently optionally substituted alkylammonium species, X and X′ are different halides, x is a real number between 0-1, y is 0-4, M is a transition metal (e.g., Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg); and where if y=0 or 4, R1≠R2 and x≠0 or 1.
In some embodiments, the organic layer includes two different molecular structures. In some embodiments, the inorganic layer includes a silicate. In some embodiments, the composition includes a metal alkyl phosphonate salt.
In some embodiments, the composition includes a compound of the following table:
In some embodiments, the compression is provided using a pressure transmitting medium (PTM) and the method includes providing a gas to the PTM that induces a change in a thermal property of the composition. In some embodiments, the gas is an inert gas that is able to permeate a free volume of the organic layer. In some embodiments, the permeated gas interacts with the composition. In some embodiments, permeation and interaction of the gas with the composition together induce a lowering of a phase transition and/or a barocaloric effect inversion. In some embodiments, the gas is nitrogen, argon, krypton, xenon, methane, ethane, propane, butane, sulfur hexafluoride, or carbon dioxide.
In some embodiments, the change in thermal property is a lowering of a phase transition temperature and/or a barocaloric effect inversion.
In an aspect, the invention provides a 2D perovskite composition including first and second layers of a transition metal halide and an organic layer including a C>3 alkyl (e.g., C>4 alkyl, e.g., C4-36 alkyl chains) ammonium species selected from:
In an aspect, the invention provides a 2D perovskite composition having formula [(R1)x(R2)1−x]2MXyX′4−y, where R1 and R2 are independently optionally substituted alkylammonium species, X and X′ are different halides, x is between 0-1, y is 0-4, M is a transition metal (e.g., Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg); and where if y=0 or 4, R1≠R2 and x≠0 or 1.
In some embodiments, R1 and R2 are independently alkylammonium species of formula C2H2n+1NH3+, where n>3 (e.g., n>4, e.g., C4-36 alkyl). In some embodiments, the 2D perovskite composition has a formula selected from: (NA)2CuCl3Br; (NA)2CuCl2Br2; (NA)2CuClBr3; (DA)2CuCl3Br; (DA)2CuCl2Br2; (DA)2CuClBr3; [(NA)0.75(DA)0.25]2CuCl4; [(NA)0.5(DA)0.5]2CuCl4; [(NA)0.25(DA)0.75]2CuCl4; [(NA)0.25(UA)0.75]2CuCl4; [(NA)0.5(UA)0.5]2CuCl4; or [(NA)0.5(DA)0.5]2CuCl2Br2. In some embodiments, R1 and R2 are independently alkylammonium species selected from:
In some embodiments, X is Cl, and X′ is Br.
In an aspect, the invention provides a barocaloric system including a composition including an organic layer including optionally substituted C>3 alkyl chains (e.g., C>4 alkyl chains, e.g., C4-36 alkyl chains), wherein the organic layer is between first and second inorganic layers or includes a head group capable of hydrogen bonding, halogen bonding, and/or electrostatic interaction with a counterion; and a source of compression.
In some embodiments, the system includes first and second inorganic layers separated by the organic bilayer. In some embodiments, the organic layer includes a compound of formula (CnH2n+1)(CmH2m+1)NH2X, where n is 1-3 or 4-36 and m is 4-36; and where X is a monoanionic species (e.g., a halide (e.g., F, Cl, Br, or I) or a non-halide anion, such as NO3—, ClO3—, ClO4—, H2PO4—, HSO4—, CN—, HCOO—, N3—, N(CN)2—, BF4—, BH4—, PF6—, SCN—, or OCN—). In some embodiments, n=m and n=4-36. In some embodiments, n=1-3 and m=4-36 (e.g., where n=1 and m=6, 8, 10, or 12).
In some embodiments, the source of compression is hydrostatic or mechanical. In some embodiments, the system further includes a heat sink.
In an aspect, the invention provides a barocaloric system. The system includes a composition including an organic layer including optionally substituted C>3 alkyl chains. The system further includes a pressure transmitting medium including one or more gases, at least one of which induces a change in a thermal property of the organic layer, and a source of compression.
In some embodiments, the at least one gas is an inert gas that is able to permeate a free volume of the organic layer. In some embodiments, the permeated gas interacts with the composition. In some embodiments, an extent of permeation and interaction of the at least one gas with the composition together induce a lowering of a phase transition and/or a barocaloric effect inversion. In some embodiments, the change in thermal property is a lowering of a phase transition temperature and/or a barocaloric effect inversion. In some embodiments, the system includes a pump for controlling the amount of the at least one gas. In some embodiments, the system includes a heat sink. In some embodiments, the system further includes a second organic layer that does not undergo the barocaloric effect inversion. In some embodiments, the at least one gas is nitrogen, argon, krypton, xenon, methane, ethane, propane, butane, sulfur hexafluoride, or carbon dioxide.
The term “about,” as used herein, refers to ±10% of a recited value.
The term “alkyl,” as used herein, is meant straight chain or branched saturated groups of carbons. Alkyl groups are exemplified by n-, sec-, iso- and tert-butyl, neopentyl, nonyl, decyl, and the like, and may be optionally substituted with one or more, substituents. Alkyl groups of the invention may include 1 or more carbon atoms, e.g., greater than 2, e.g., 6-15, such as 8-12, in the main chain. Carbon atoms in the main chain may be interrupted with one or more heteroatoms, e.g., O, S, or N.
By “aryl” is meant an aromatic cyclic group in which the ring atoms are all carbon. Exemplary aryl groups include phenyl, naphthyl, and anthracenyl. Aryl groups may be optionally substituted with one or more substituents.
By “carbocyclyl” is meant a non-aromatic cyclic group in which the ring atoms are all carbon. Exemplary carbocyclyl groups include cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl, and cyclooctyl. Carbocyclyl groups may be optionally substituted with one or more substituents.
By “halo” is meant, fluoro, chloro, bromo, or iodo.
By “heteroaryl” is meant an aromatic cyclic group in which the ring atoms include at least one carbon and at least one O, N, or S atom, provided that at least three ring atoms are present. Exemplary heteroaryl groups include oxazolyl, isoxazolyl, tetrazolyl, pyridyl, thienyl, furyl, pyrrolyl, imidazolyl, pyrimidinyl, thiazolyl, indolyl, quinolinyl, isoquinolinyl, benzofuryl, benzothienyl, pyrazolyl, pyrazinyl, pyridazinyl, isothiazolyl, benzimidazolyl, benzothiazolyl, benzoxazolyl, oxadiazolyl, thiadiazolyl, and triazolyl. Heteroaryl groups may be optionally substituted with one or more substituents.
By “heterocyclyl” is meant a non-aromatic cyclic group in which the ring atoms include at least one carbon and at least one O, N, or S atom, provided that at least three ring atoms are present. Exemplary heterocyclyl groups include epoxide, thiiranyl, aziridinyl, azetidinyl, thietanyl, dioxetanyl, morpholinyl, thiomorpholinyl, piperazinyl, piperidinyl, pyrrolidinyl, tetrahydropyranyl, tetrahydrofuranyl, dihydrofuranyl, tetrahydrothienyl, dihydrothienyl, dihydroindolyl, tetrahydroquinolyl, tetrahydroisoquinolyl, pyranyl, pyrazolinyl, pyrazolidinyl, dihydropyranyl, tetrahydroquinolyl, imidazolinyl, imidazolidinyl, pyrrolinyl, oxazolidinyl, isoxazolidinyl, thiazolidinyl, isothiazolidinyl, dithiazolyl, and 1,3-dioxanyl. Heterocyclyl groups may be optionally substituted with one or more substituents.
Optional substituents include halo, optionally substituted C3-10 carbocyclyl; optionally substituted C1-9 heterocyclyl having one to four heteroatoms independently selected from O, N, and S; optionally substituted C6-20 aryl; optionally substituted C1-9 heteroaryl having one to four heteroatoms independently selected from O, N, and S; —CN: —NO2; —ORa; —N(Ra)2; —C(═O)Ra; —C(═O)ORa; —S(═O)2Ra; —S(═O)2ORa; —P(═O)Ra2; —O—P(═O)(ORa)2, or —P(═O)(ORa)2, or an ion thereof; wherein each Ra is independently H, optionally substituted C1-6 alkyl; optionally substituted C3-10 carbocyclyl; optionally substituted C1-9 heterocyclyl having one to four heteroatoms independently selected from O, N, and S; optionally substituted C6-20 aryl; or optionally substituted C1-9 heteroaryl having one to four heteroatoms independently selected from O, N, and S.
Barocaloric effects—thermal changes driven by hydrostatic pressure—offer particularly simple and energy-efficient ways to achieve solid-state cooling. As first-order phase transitions involve a latent heat, large barocaloric effects are expected to occur near the phase transition temperature when the first-order transition is induced by applied hydrostatic pressure. To exhibit a large barocaloric effect, a material must meet the following three requirements: 1) first-order phase transition with large latent heat (Gtr), large entropy change (ΔStr), and transition temperature Ttr close to a desired operational temperature; 2) high sensitivity of the phase transition to applied pressure (i.e., high barocaloric coefficient δTtr/δP); and, 3) low thermal hysteresis. As cooling devices operate cyclically, the barocaloric effects must be driven reversibly upon a sequence of application and removal of external pressure; therefore, thermal hysteresis dramatically affects the cooling performance, determining the minimum pressure needed to achieve reversible barocaloric effects (prev). Usually, prev is proportional to the thermal hysteresis width. Since it is practically beneficial to have low operating pressure (prev), identifying barocaloric materials with low thermal hysteresis is highly desirable. Note, additionally, that for a first-order phase transition, the barocaloric coefficient can be calculated using the Clausius-Clapeyron relation δTtr/δP=ΔVtr/ΔStr where ΔVtr=volume change of the phase transition. Thus, a material must have a high ΔVtr/ΔStr to have a high barocaloric coefficient.
The invention provides highly generalizable approaches to realizing a new class of materials that display colossal—yet tunable—barocaloric effects at relatively low operating pressures. Specifically, this invention describes the use of reversible, solid-solid, order-disorder phase transitions, e.g., in layered, two-dimensional (2D) organic-inorganic hybrid materials for barocaloric cooling (
The propensity for confined chain melting to drive large barocaloric effect is based on: (i) the wide range of organic phase-change materials that undergo phase transitions near ambient temperature with a large volume change (ΔVtr), latent heat (Qtr), and entropy change (ΔStr); (ii) the organic phase-change materials, when templated with inorganic substances, remain solid-state during the phase change, exhibiting reversible, solid-solid phase transitions with all beneficial phase-change properties (i.e., high ΔVtr, Qtr, ΔStr) retained. Unlike magnetocaloric, electrocaloric, and elastocaloric effects, the barocaloric effect is not system-selective and in principle can be observed in any materials, as the free energy of a system always depends on pressure. In summary, this invention reports that order-disorder transition of normal, branched, or functionalized organic chains—on or within inorganic structural templates—can be used as a new mechanism to achieve large barocaloric effects relevant to solid-state cooling.
Layered materials such as two-dimensional (2-D) metal-halide perovskites of the form (R-NH3)2MX4 (R=CnH2n+1; n>3 (e.g., >4); M=Mn, Fe, Cu, Cd, Pb; X=F, Cl, Br, or I) can undergo chain-melting transitions in the solid state. In these compounds, sheets of corner-sharing MX6 octahedra create anionic pockets-defined by the axial halides of four adjacent metal centers—that template the arrangement of bilayers of alkylammonium cations through charge-assisted hydrogen bonds. When long-chained hydrocarbon molecules (n>3, e.g., >4) are incorporated, many layered perovskites undergo thermally induced, first-order phase transitions between ordered and disordered states driven by what is effectively a partial melting transition of the hydrocarbon bilayers. As such 2-D perovskites can serve as a highly tunable solid-state platform to leverage the large changes in entropy and enthalpy that accompany hydrocarbon chain-melting transitions for barocaloric cooling (
In one embodiment, this invention describes barocaloric effects in 2D hybrid perovskites, with a general chemical formula of (R-NH3)2MX4, which contain layers of first-row transition metal halides [MX4]2− (M=Mn, Fe, Co, Cu, Zn; X=Cl, Br, or I) connected by bilayers of ammonium cations (R-NH3+).
In these compounds, organic bilayers are confined by metal-halide inorganic layers. Inorganic layers can assemble (and confine organic layers) via, e.g., hydrogen bonds (e.g., between R-NH3+ . . . X− groups), electrostatic attraction, or a combination thereof. Van der Waals interactions between R groups (see, e.g.,
Many 2D perovskites with long-chain organic molecules (R=CnH2n+1, n>4) are known to undergo thermally-induced, reversible, solid-solid phase transitions near room temperature (20-90° C.) due to the ordering and disordering the organic molecules. As these solid-solid transitions are often accompanied by a large latent heat (>60 kJ kg−1) and entropy change (>200 J kg−1 K−1), layered hybrid perovskites may be employed as solid-solid phase-change materials for passive thermal management and thermal energy storage.
The main phase transition of these compounds involves disordering of organic chains, often referred to as a “chain-melting” transition. The mechanism for “chain melting”—defined as the rapid diffusion of a kink (one or more gauche bonds) up and down along the C—C bonds within an organic chain—has been extensively studied at ambient pressure by various structural, thermal, and spectroscopic techniques, including powder and single-crystal X-ray diffraction, differential scanning calorimetry, dielectric measurements, and infrared and Raman spectroscopies. The dynamics of the ammonium chains have been further investigated at ambient pressure by solid-state NMR techniques and inelastic neutron scattering. In the low-temperature, ordered phase, the confined chains are tilted because their cross-sectional area is less than the area of the halide square (˜5×5 Å2) afforded by the 2D inorganic lattice.
In the high-temperature, disordered phase, the chains are disordered and effectively occupy the whole cross-sectional area available to them. This results in a large expansion of the interlayer spacing. As the intralayer distances are essentially unchanged due to the robust inorganic templates, this directly leads to large increase in volume (ΔVtr/V0=7-10%). These studies show that the large latent heat, entropy change, and volume change of main phase transition observed in the organic-inorganic hybrid materials originate from confined chain melting process.
A representative layered perovskite (C10H21NH3)2MnCl4 undergoes solid-solid, reversible phase transition near room temperature (35° C.) with large entropy change (e.g., ˜221 J kg−1 K−1) and volume expansion (e.g., ˜7.3%). Based on thermodynamics calculations (Clausius-Clapeyron relation), we identified that the phase transition is expected to be highly sensitive to applied pressure (e.g., δTtr/δP˜28 K kbar−1). As such, inducing the disorder-to-order transition by adiabatically applying pressure is expected to result in adiabatic temperature increase, e.g., (ΔTad) of ˜30 K. Taken together, this analysis suggested that (C10H21NH3)2MnCl4 should exhibit colossal barocaloric effect.
To experimentally confirm the existence of large barocaloric effects in layered perovskites, we measured powder X-ray diffraction (PXRD) patterns as a function of temperature and applied pressure for the 2D layered perovskites (C10H21NH3)2MnCl4 to evaluate the pressure dependence of the transition temperature (Ttr) between ordered and disordered phases of this material. As the thermally induced order-disorder phase transition leads to a clear shift in powder diffractions peak positions (
The class of existing and potential 2D layered perovskites provide access to a tremendous structural and chemical diversity through the judicious selection of the inorganic and organic moieties that constitute each material. As such, we may control the confined chain melting in these materials leading to not only colossal but also highly tunable barocaloric effects. Indeed, the thermodynamics and kinetics of these pressure-induced phase transitions are controllable, e.g., by modifying: 1) the molecular interactions between the inorganic layers, 2) the flexibility of the organic chains, and 3) the free volume within the organic bilayers.
For instance, we have synthesized a series of new layered perovskites—containing Cu, Mn, and Fe metal centers ligated to Cl− anions—with organic molecules incorporated as bilayers between the metal-chloride sheets that include oxygen-substituted alkyl chains (C3OC4), functionalized phenylalkylamines (C4Ph and C6Ph), alkyl chains incorporating an ester group (Cn—COO—C2; n=9, 10, 11), and alkyl chains functionalized with alcohols (CnOH; n=5, 6, 8) that create hydrogen bonding networks within the bilayers (see, e.g.,
OC4
)
Mn
Mn
(COO)C2
F
)
F
)
OC4)
Cu
OH
Ar)2Cu
OH
OAr)
Cu
OH
OH)2Cu
OC4OH
OH)
Cu
OC
OH
OH)
Mn
OPh
indicates data missing or illegible when filed
Table 1 shows a library of long-chain ammonium cations incorporated into layered perovskites. Structures of long-chain ammonium cations studied in our laboratory (left) and thermal properties of layered perovskites incorporating the ammonium cation chains (right). Ttr, transition temperature; ΔStransition, entropy of phase transition; Qtr, latent heat of phase transition. (R)2M denotes layered perovskite (R-NH3)2MCl4.
These newly synthesized 2D hybrid perovskites exhibit reversible, thermally-induced phase transitions driven by chain melting. Most notably, the synthetic modifications of the organic chain enabled the tuning of transition temperature between −30° C. to 120° C. without compromising beneficial thermodynamic properties (e.g., large latent heat and entropy change). These results demonstrate that the barocaloric properties of layered perovskites can be readily tuned through synthetic modifications. Barocaloric effects of the functionalized perovskites synthesized in our laboratory are summarized in Table 5.
In addition, we have synthesized compositionally engineered mixed halide 2D metal-halide perovskites, e.g., replacing all—or a portion—of Cl anions with Br anions for mixed-halide systems, e.g., having formula [(R1)x(R2)1−x]2MXyX′4−y, where R1 and R2 are long chain alkylammonium species(e.g., CnH2n+1NH3+, where n>3, e.g., >4, e.g., NA or DA) and where X and X′ are different halides, e.g., selected from Cl, Br, or I, e.g., (R-NH3)2MCl4−yBry (0<y≤4), e.g., where M is a transition metal (e.g., Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Nb, Mo, Rh, Pd, Cd, Re, Pt, or Hg) and R is NA or DA. Mixed halide compounds may also be “double mixed”, e.g., containing two halides (e.g., Cl and Br) and two different alkylammonium species. The different alkylammonium species may be in non-integer ratios, relative to the metal center (e.g., [(NA)0.75(DA)0.25]2CuCl4, [(NA)0.5(DA)0.5]2CuCl4, or [(NA)0.25(DA)0.75]2CuCl4, [(NA)0.25(UA)0.75]2CuCl4, [(NA)0.5(UA)0.5]2CuCl4, or [(NA)0.5(DA)0.5]2CuCl2Br2). In compounds of the formula [(R1)x(R2)1−x]2MXyX′4−y, ‘y’ may be 0-4 (e.g., 0, 1, 2, 3, or 4) and ‘x’ may be between 0-1, e.g., about 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, or 0.95. Thermal and structural data on the mixed 2D metal-halide perovskite materials in shown in Tables 21-25. Some mixed 2D metal-halide perovskite materials are compared to non-mixed 2D metal-halide perovskite materials in
In addition, additional functionalized layered perovskites may be formed by (i) combining two different modifications in a single chain (e.g., C3OC4OH) (ii) fluorinating the chain, (iii) synthesizing layered halide double perovskites of the form (R-NH3)4MM′X8 where M is a monovalent cation such as Na+ and M′ is a trivalent cation such as Fe3+, X=Cl—, Br—. In some embodiments, one or more Cl, Br, or I halides may be replaced by F. In addition, additional functionalized layered perovskites may be formed from non-halide anions, e.g., CN—, HCOO—, N3—, N(CN)2—, BF4—, BH4—, PF6—, SCN—, OCN—. Incorporation of the larger anions can increase the pocket size for cations (e.g., the alkylammonium species), thus relatively larger cations e.g., the largest cations of Table 1, or even dialkylammonium cations described herein.
In addition to synthetic tunability and chemical diversity, layered halide perovskites display a number of other properties that are advantageous for practical solid-state cooling. First, the “soft” nature and high solution processability of layered halide perovskites enables thin films to be easily fabricated. The high processability not only presents rich opportunities for the design of miniaturized cooling devices but may also allow the invention to take advantage of microscopic mechanisms for barocaloric effects across various length scales (from bulk powders to thin films to atomically thin layers) and materials forms (from single crystals to microcrystalline powders to thin films). Second, phase transitions in appropriately designed layered perovskites display small thermal hysteresis (<4 K). Thus, reversible barocaloric effects can be achieved at small operating pressure prev. Given that these layered perovskites display high barocaloric coefficients (>20 K kbar−1), this suggests that colossal reversible barocaloric effects can be realized at easily accessible pressures (<100 bar) (
For comparison, organic plastic crystals—another class of colossal barocaloric materials—require relatively high operating pressure of 500-1,000 bar for reversible barocaloric effects. These two features—high processability and low operating pressure—allow various device architectures with minimal design constraints.
The confined chain melting described herein is a general mechanism to achieve colossal barocaloric effects because large latent heat, entropy change, and volume expansion—prerequisites to colossal barocaloric effects—can simultaneously emerge when long-chain molecules are forced to undergo large conformational changes in confined space. Note that the analysis and experiments described in this invention are readily applicable to other types of layered materials beyond hybrid perovskites. Any organic and inorganic materials that can be assembled into layers of organic material between layers of inorganic material are suitable to be used in the invention. Alternatively or in addition, organometallic materials including a layer-forming inorganic component and an organic component including long, optionally substituted alkyl chains may be used, for example, the following barocaloric materials: di-n-alkylammonium salts (e.g., compounds of formula (CnH2n+1)2NH2X (n>3 (e.g., >4, e.g., C4-36 alkyl chains) or compounds of formula (CnH2n+1)(CmH2m+1)NH2X, where n is 1-3 or 4-36 and m is 4-36, e.g., where n=m, or where n=1-3 (e.g., 1) and m=4-36 (e.g., 6, 8, 10, or 12); where X is a monoanionic species, e.g., a halide (e.g., F, Cl, Br, or I) or a non-halide anion, such as NO3—, ClO3—, ClO4—, H2PO4—, HSO4—, CN—, HCOO—, N3—, N(CN)2—, BF4—, BH4—, PF6—, SCN—, or OCN—), see
Chain-melting phase transitions and their pressure sensitivity (dTtr/dP), discussed here in the context of barocaloric cooling (using the 2-D perovskites as proof-of-concept examples), offer a new mechanism to realize highly efficient and tunable thermal energy storage, whereby application of hydrostatic pressure can be used to tune the storage temperature, Tstor, and release temperature, Trel (
The material properties required for efficient PT-TES are similar to those required for efficient barocaloric cooling. Materials for PT-TES should undergo solid-state phase transitions with high thermal changes (ΔStr), high sensitivity to pressure (dTtr/dP), and small hysteresis (ΔThys), because the temperature span of the PT-TES operation (ΔTspan), defined here as the difference between Trel and Tstor, is calculated by, ΔTspan=(dTtr/dP)×ΔP−ΔThys. Note that PT-TES materials will benefit from having large ΔTspan and small ΔP, and materials with high dTtr/dP and low ΔThys, as well as large ΔStr and ΔHtr, will be suitable for this. Another difference is, unlike barocaloric cooling (which requires phase transition temperatures near or below ambient temperature), thermal energy storage materials are needed across a broad temperature range. In this context, compositions of the invention, e.g., the two-dimensional metal-halide perovskites—first highlighted here in the context of barocaloric cooling due to their large thermal changes (ΔStr and ΔHtr), high pressure sensitivity (dTtr/dP), and small hysteresis (ΔThys)—provide a versatile platform to achieve efficient the PT-TES in the broad temperature range, as their transition temperatures and sensitivity to pressure can be readily manipulated by changing the length of hydrocarbon chains to cover a wide temperature range (from 250 K to 400 K) (Tables 2 to 5). Layered compounds of the invention, e.g., with long-chain hydrocarbons, will be highly competitive PT-TES materials, due to their beneficial phase-change properties, synthetic tunability, and pressure dependence, similar to those of the 2-D perovskites (Table 20) exemplified herein.
As shown in
The invention also includes methods of enhancing barocaloric cycles based on the properties of the pressure-transmitting medium. The pressure transmitting medium (PTM) can affect the properties of the phase transitions of the barocaloric cycles in materials including long-chain hydrocarbons. The effects include inverse barocaloric effects for compounds (e.g., with long-chain hydrocarbons) that undergo reversible chain-melting transitions. Gaseous PTMs may induce changes in thermal properties of materials with long alkyl chains (e.g., those of the invention) by permeating into and interacting with the materials of the composition, e.g., by permeating into free volume in the organic layer. Gases that can permeate into the composition are preferably inert gases that can also interact with the composition at the microscopic level (e.g., non-covalently interact, e.g., via Van der Waal's-type interactions, e.g., via dispersion forces). Both the extent of permeation (e.g., amount of gas molecules in the free volume/interacting with the composition) and degree and nature of interaction (e.g., strength of interaction, e.g., determined by a molecule or atom's size, shape, polarizability, etc.) can determine the effect of the PTM on thermal transitions of the composition. Exemplary PTM gases include nitrogen, argon, krypton, xenon, methane, ethane, propane, butane, sulfur hexafluoride, or carbon dioxide. Gases that permeate and interact sufficiently with the composition (e.g., argon into (NA)2CuBr4)) can cause the barocaloric effect of the composition to be inverted.
In our recent studies, we discovered that the pressure sensitivity (dT/dP) of chain-melting transitions in representative 2-D perovskites—(DA)2MnCl4 and (NA)2CuBr4—depends on the identity of pressure-transmitting medium (PTM). Specifically, as shown in
These results fully confirm the validity of this phenomenon and establish that (NA)2CuBr4—and other related barocaloric compounds (e.g., with long-chain hydrocarbons), e.g., those described herein, more broadly—displays highly reversible, giant inverse barocaloric effects under a pressure-transmitting medium that can interact with hydrocarbon chains (and the lattice).
Most inverse barocaloric effects arise from a decrease in volume upon increase in a degree of freedom (during thermally-induced phase transitions). The mechanism described herein is unique for the following reasons. First, the material (i.e., the host lattice, e.g., a barocaloric material of the invention) still undergoes a volume expansion upon the thermally-induced order-to-disorder transition, and the inverse barocaloric effect is entirely driven by the permeation and absorption of the PTM into the lattice. Second, the magnitude of inverse barocaloric effects can be tuned via judicious selection of pressure medium. As shown in
Methods of the invention may include providing heat energy (e.g., from a room, an AC system, heat transfer medium, heat pump, heat sink, etc.) to a composition of the invention (e.g., a 2D perovskite). The heat energy may cause alkyl chains in the composition to undergo a phase transition (e.g., from an ordered to a disordered state, e.g., in a thermal energy storage system) or there may be no phase transition until pressure is applied (e.g., in a barocaloric cooling system). Methods may be for refrigeration or heating.
In a barocaloric cooling system or method, providing compression to the composition releases latent heat in the composition, which may be removed, e.g., via a heat sink, e.g., a high surface area, high conductivity medium in thermal contact with the composition which may be itself cooled by, e.g., a fan. Removal of the heat is performed while the composition is still compressed, and removal of the compression allows the composition to return to a disordered state, cooling the composition as the endothermic transition occurs. At this point the cycle may be repeated with input of new heat energy.
In a barocaloric thermal energy storage system heat energy is provided to a composition of the invention causing it to undergo a phase transition to a disordered state. The disordered state is then modified by the application of compression in order to change the temperature at which heat is released.
Methods of the invention may also include selecting or otherwise controlling the pressure transmitting medium to modulate the barocaloric cycle. For example, selecting a gas (e.g., a high polarizability gas) that sufficiently permeates and interacts with the composition at the microscopic level as the PTM to change the temperature of phase transitions in the barocaloric material, or to induce inverse barocaloric effects such as described herein. Methods may include modulating the barocaloric cycle by altering a ratio of polarizable and on-polarizable gases used as a mixed in a PTM. Methods may include selecting a gas as the PTM that does not interact, or minimally interacts, with the composition (e.g., He), e.g., to not induce changes in thermal properties, or to revert changes caused by an interacting gas.
Systems of the invention may include components to provide compressive force to the composition, e.g., pumps, pistons, actuators (e.g., mechanical, hydraulic, or pneumatic, etc., actuators), presses (e.g., mechanical, hydraulic, or pneumatic, etc., presses), piezoelectric actuators, levers, etc. Systems may also include components to transfer or remove heat energy, e.g., pumps, heat sinks, thermoelectrics, fans, chiller pumps, etc. A system of the invention may also include a power source, e.g., to power the source of compressive force, the cooling or heat transfer components, etc. Systems of the invention may include a pressure transmitting medium (PTM), e.g., a gas. The PTM may be a non- or minimally-interacting gas (e.g., a low polarizability gas, e.g., He) or a polarizable gas (e.g., N2, Ar, Kr, Xe, methane, ethane, propane, butane, sulfur hexafluoride, or carbon dioxide). Systems of the invention may include a pump for controlling a pressure transmitting medium (e.g., a mixture of gases), such as pumps, gas reservoirs (e.g., tanks, cylinders, etc.), pressure sensors, actuators, valves, etc. The PTM may not be a gas, for example, the PTM may be an oil, e.g., a fluorocarbon oil, silicone oil, etc.).
The 2-D perovskite (DA)2MnCl4 (DA=decylammonium) was selected as a barocaloric material because of its large phase-transition entropy (ΔStr=230 J kg−1 K−1) and enthalpy (ΔHtr=71 kJ kg−1), near-ambient phase transition temperature (Ttr=310 K), and lightweight, nontoxic elemental composition. At ambient temperature and pressure, (DA)2MnCl4 adopts an ordered monoclinic structure (low-temperature, LT, phase) with bilayers of hydrocarbon chains—each of which contain a single gauche C—C bond (C2-C3) and seven trans C—C bonds—aligned parallel to one another and tilted 48.3(1)° with respect to the Mn—Cl plane (
Differential scanning calorimetry (DSC) measurements (
To directly evaluate the pressure dependence of the phase transition temperature, isobaric DSC experiments were performed under applied hydrostatic pressures of up to 150 bar using He as the pressure-transmitting medium. The phase transition shifts to higher temperatures as the pressure is increased, with a measured dTtr/dP of 22.1±0.7 K kbar−1 during heating and 20.6±0.8 K kbar−1 during cooling (
To investigate the origin of the different experimental and predicted barocaloric coefficients, we used helium pycnometry to directly measure the volume change of (DA)2MnCl4 during the phase transition (
Under the cyclic operating conditions of a barocaloric cooling system, the lowest possible operating pressure is set by the pressure that must be applied to induce a reversible entropy change, Prev, when cycling to and from ambient pressure. For a conventional barocaloric effect, Prev corresponds to the pressure at which the onset temperature of the exothermic phase transition is equal to the onset temperature of the endothermic phase transition at 1 bar. As such, Prev is proportional to the thermal hysteresis at 1 bar and inversely proportional to the barocaloric coefficient for the exothermic transition, with Prev=ΔThys/(dTtr/dP)cooling (16). Owing to its low ΔThys and high dTtr/dP, (DA)2MnCl4 has a predicted Prev of just 66 bar.
This low Prev was further confirmed by calculating the isothermal entropy changes (ΔSit). To do so, we first obtained isobaric entropy changes (ΔSib) associated with the chain-melting transition as a function of temperature and pressure by integrating the HP-DSC heat flow signal, Q, obtained at a scan rate of {dot over (T)} over the temperature range from Ti to Tf:
ΔSit curves were then calculated as the difference between ΔSib at ambient pressure and ΔSib at elevated pressure, with ΔSib values obtained from heating data corresponding to the disordering transition induced by a decrease in pressure (ΔSit>0) and ΔSib values from cooling data corresponding to the ordering transition induced by an increase in pressure (ΔSit<0). Excitingly, these ΔSit curves show that a non-zero reversible entropy change can be induced below 80 bar, and the full entropy of the phase transition can be induced irreversibly by applying a pressure of just 100 bar (
Although quasi-direct methods of calculating isothermal changes-as well as adiabatic ones-from isobaric experiments are commonly used to evaluate barocaloric materials due to the challenge of maintaining isothermality—or adiabaticity—during direct variable pressure measurements, we performed quasi-isothermal HP-DSC experiments to more directly evaluate Prev by measuring pressure, rather than thermal, hysteresis. Specifically, we measured heat flow signals while cycling the pressure between 1 bar and 150 bar at 311 K. Note that isothermality is maintained until the phase transition onset pressure, which allows us to accurately determine pressure hysteresis. By comparing the onset pressures for compression-induced exotherms and decompression-induced endotherms, we were able to directly measure a pressure hysteresis for (DA)2MnCl4 of 70 bar, which is in excellent agreement with the predicted value of 73 bar at 311 K (
In an effort to target barocaloric materials with large reversible entropy changes at even lower pressures, we searched for a 2-D perovskite that undergoes a chain-melting transition with a thermal hysteresis of less than 1 K. Unlike (DA)2MnCl4, however, the total entropy of the chain-melting transition in most 2-D perovskites is partitioned across one or more minor—lower entropy—phase transitions in addition to the principal transition (
Since Br anions can be more readily accommodated within Cu, rather than Mn, 2-D perovskites, we targeted a (CnH2n+1NH3)2CuBr4 compound with alkylammonium cations of appropriate length to place the chain-melting temperature near ambient temperature. We found that the new 2-D perovskite (NA)2CuBr4 (NA=nonylammonium) undergoes a chain melting transition at 303 K with a high ΔStr (78 J kg−1 K−1), hysteresis of only 0.4 K, and no minor phase transitions within at least 60 K of the principal transition (
Isobaric HP-DSC experiments confirmed that (NA)2CuBr4 features a high barocaloric coefficient with dTtr/dP of 26.9±0.4 K kbar−1 and 26.5±0.5 K kbar−1 during heating and cooling, respectively (
To provide additional insight into the structural and chemical factors that influence barocaloric effects in 2-D perovskites, we used X-ray crystallography and IR spectroscopy to compare the nature of the chain-melting transition in (NA)2CuBr4 and (DA)2MnCl4. In particular, we hypothesized that the increased size of the halide pocket in (NA)2CuBr4 (30.5 Å2) relative to (DA)2MnCl4 (26.3 Å2)—coupled with weaker N—H . . . Br hydrogen bonds at the organic-inorganic interface—would lead to more conformational disorder in the LT phase of the Cu compound. This would reduce the entropy difference between the LT and HT phases and explain the 56% lower molar entropy change of (NA)2CuBr4, as well as the lower entropy changes generally observed across longer-chain (CnH2n+1NH3)2CuBr4 (n=11-16) compounds compared to (CnH2n+1NH3)2MCl4 (M=Mn, Cu, Cd) compounds of the same lengths (Tables 2 to 4).
As anticipated, the atoms in the NA chains of (NA)2CuBr4 have much larger atomic displacement parameters in the LT phase than those in the DA chains of the LT phase of (DA)2MnCl4 (
Although conformational disorder in the LT phase leads to a decreased entropy change, it also likely contributes to the enhanced reversibility of the (NA)2CuBr4 chain-melting transition through two primary effects. First, the higher degree of disorder in the alkylammonium chains in the LT phase-along with the softer nature of Br anions—should make the (NA)2CuBr4 lattice more compressible than the (DA)2MnCl4 lattice. Since the barocaloric coefficient dTtr/dP of a solid tends to increase with increasing compressibility (43, 44), this would be expected to make the phase transition in (NA)2CuBr4 more sensitive to pressure. Indeed, dTtr/dP for the transition to the LT phase of (NA)2CuBr4 is 29% higher than for (DA)2MnCl4. Second, the presence of certain configurational degrees of freedom, such as gt2n+1g′ kinks, in both the LT and HT phases should render the two phases more compatible, lowering both isobaric and isothermal hysteresis. In any case, both compounds display, near room temperature, large and reversible barocaloric cooling, represented by their materials properties Ttr, ΔStr, and Prev, and are highly competitive with other leading barocaloric materials (
In addition to its influence on operating pressure, hysteresis, which leads to dissipative heat losses, adversely impacts the second-law efficiency and coefficient of performance (COP) of any caloric cooling cycle. The impact of hysteresis on efficiency can be quantified by calculating the idealized thermodynamic efficiency, η, of a caloric material—relative to the Carnot efficiency—using a simple material model that integrates the dissipative losses in a Carnot-like cycle:
Based on this model, caloric materials with ΔThys/ΔTad,max of less than 10% will have second-law efficiencies competitive with those of conventional vapor compression-based systems (˜85%). Excitingly, (DA)2MnCl4 and (NA)2CuBr4 display second-law efficiencies of 88 and 93%, while the values found in most barocaloric compounds range between 40 and 65% (Table 19), either because of large ΔThys or a low ΔStr that leads to a small ΔTad,max. Additionally, both compounds display the largest values of barocaloric strength—the reversible isothermal entropy change ΔSit,rev normalized by the driving pressure—than have been realized for barocaloric materials (Table 19).
These results highlight exciting opportunities to exploit the tunability of 2-D perovskites to independently manipulate phase-change hysteresis, entropy, and sensitivity to pressure for improved barocaloric performance. For instance, it should be possible to realize chain-melting transitions with even higher entropy changes through functionalization of the organic bilayers—such as by introducing aromatic groups or hydrogen bond donor-acceptor pairs—and even lower hysteresis through modification of the organic-inorganic interface—such as by incorporating mixtures of different halide anions or introducing defects—or through leveraging multicaloric effects. In addition, the anisotropic nature of the chain-melting transition in (DA)2MnCl4 and (NA)2CuBr4—wherein an increase in interlayer spacing along a single direction accounts for ˜80% of the volume change-suggests that uniaxial stress, which can be readily applied through mechanical actuation, may be able to drive large elastocaloric effects in 2-D perovskites.
The results and materials discussed herein were obtained according to the following methods.
All manipulations were conducted in air unless otherwise noted. Anhydrous diethyl ether was obtained from a Pure Process Technology anhydrous solvent system. Anhydrous methanol and ethanol were purchased from a commercial vendor and used as received. All other reagents were purchased from commercial vendors and used as received. Single crystal diffraction data was collected using a Bruker D8, SMART APEX II, or APEX DUO instrument. IR spectra were obtained on a Bruker ALPHA II Platinum ATR with a variable temperature stage. Thermogravimetric analysis (TGA) experiments were performed using a TA Instruments TGA550. Abbreviation used: DA=decylammonium, (C10H21NH3)2MnCl4=(DA)2MnCl4, (NA)=nonylammonium, (C9H19NH3)2CuBr4=(NA)2CuBr4.
(DA)2MnCl4·: Decylamine (≥99.0%) and hydrochloric acid (HCl) solution (37 wt %) were purchased from Sigma Aldrich and used without further purification. C10H21NH3Cl was first synthesized by adding HCl solution (550 μL, 6.6 mmol) into decylamine (1.1 mL, 5.5 mmol) in ca. 5 mL ethanol in a cold-water bath. After evaporating the solvent at reduced pressure, the resulting white powder of (DA)Cl was washed with diethyl ether and vacuum dried at room temperature for a day. Crystalline powders of (DA)2MnCl4 were prepared by the cooling of an ethanol solution containing a stoichiometric quantity of the manganese(II) chloride and (DA)Cl, as previously reported (e.g., in M. R. Ciajolo, et al., Comparative Studies of Layer Structures: The Crystal Structure of Bis(Monodecylammonium)tetrachloromanganate(II). Gazzetta Chimica Italiana. 106, 807 (1976), and H. Arend, et al., Layer perovskites of the (CnH2n+1NH3)2MX4 and NH3(CH2)mNH3MX4 families with M=Cd, Cu, Fe, Mn or Pd and X=Cl or Br: Importance, solubilities and simple growth techniques. J. Cryst. Growth. 43, 213-223 (1978)). (DA)Cl (96.9 mg, 0.5 mmol) was dissolved in 4.0 mL of ethanol. After several minutes of stirring, MnCl2·4H2O (49.5 mg, 0.25 mmol) was added to the solution, and the solution was heated to 65° C. After the cooling this solution to room temperature at a rate of 4 K h−1, pale pink crystals were obtained. The crystals were filtered and washed with diethyl ether (5×10 mL) and held at reduced pressure for 6 h to afford 45.2 g (35.2% yield) of product. Crystals suitable for structure determination were obtained by slow cooling at a rate of 2 K h−1.
(NA)2CuBr4: Nonylamine (≥99.5%) and hydrobromic acid (HBr) solution (48 wt %, 8.8 M) were purchased from Sigma Aldrich and used without further purification. C9H19NH3Br was first synthesized by adding HBr solution (545 μL, 4.8 mmol) into nonylamine (733 μL, 4.0 mmol) in ca. 5 mL ethanol in a cold-water bath. The solvent was removed at reduced pressure to yield colorless powder of (NA)Br. The powder was washed with diethyl ether and vacuum dried at room temperature for a day. CuBr2 (402 mg, 1.8 mmol) and C9H19NH3Br (807 mg, 3.6 mol) were dissolved in 2 mL of ethanol. The solution was slowly cooled from 65° C. to room temperature at a rate of 4 K h−1. The saturated solution was then stored at below ambient temperature (5° C.) for 1 hour. The resulting dark purple precipitate was filtered and washed with diethyl ether (5×10 mL). The dark purple crystalline powder was held at reduced pressure for 12 h to remove moisture. Crystals suitable for structure determination was obtained by slow evaporation of a 1-mL solution of (NA)2CuBr4 (202 mg, 0.3 mmol) in methanol. Anal. Calcd. for (C9H19NH3)2CuBr4: C: 32.19%, H: 6.60%, N: 4.17%, Br: 47.58%, Found: C: 31.84%, H: 6.69%, N: 4.43%, Br: 47.76%.
DSC at ambient pressure: A Discovery 2500 DSC with an RCS 90 cooling system (TA Instruments) was used to measure the transition temperatures and gravimetric enthalpies for all compounds. The DSC baseline and cell thermal parameters were calibrated using sapphire discs. The temperature and cell constant were calibrated using an indium standard. All DSC samples were prepared in air using 3-15 mg of sample and were hermetically sealed in aluminum pans (purchased from TA instruments). The sample was scanned under a dynamic flow of N2 (50 mL min−1). An empty, hermetically sealed aluminum pan we used as a reference.
Determination of Ttr and gravimetric ΔHtr and ΔStr: Transition temperatures, Ttr, and enthalpies of transition, ΔHtr, were determined using the TA Instrument TRIOS or Netzsch Proteus software. Peaks were selected for analysis by defining a temperature range containing the peak of interest. The lower bound and upper bounds of the temperature range were chosen to encompass the phase transition, which starts with a deviation from the baseline and ends with a return to baseline.
Prior to determination of Ttr or ΔHtr, a baseline, which models the heat flow in the absence of transition, must be generated to approximate the baseline in the transition region in the absence of a transition. A baseline is generated within the defined temperature range using various option that determine the slope of the lower and higher temperature limits and shape of the baseline. When possible, baselines were generated using mutual tangent slopes at both the upper and lower temperature limits with a sigmoidal baseline, which we found to produce most physically reasonable baselines.
The extrapolated onset temperature was reported as the transition temperature, as is standard in DSC data analysis, because the onset temperature—unlike the peak temperature—is relatively independent of experimental parameters like the heating rate or sample mass. The onset temperature is determined by identifying the region of the onset melting peak that has the highest slope, defining a tangent to that region, and then extending the tangent to the generated baseline. The intersection between the baseline and the tangent is the onset temperature. Endotherms were integrated between the upper and lower temperature limits with the baseline subtracted to provide ΔHtr, and ΔStr was calculated through ΔStr=ΔHtr/ttr. If physically reasonable limits were chosen, the onset transition temperatures and ΔHtr did not depend strongly on the choice of the temperature limits, and such variations were within the error of the measurements, which is estimated to be <0.5% for Ttr and <2% for ΔHtr. Note that volumetric ΔHtr were calculated from gravimetric quantities using crystallographic densities at ambient temperature.
DSC at applied hydrostatic pressure: High-pressure DSC measurements at the pressure range between 1 to 150 bar were carried out in a DSC 204 HP Phoenix® (Netzsch). The temperature and heat flow were calibrated at each pressure using an indium standard. The temperature and cell constant were calibrated at each pressure using an indium standard. Helium gas was used as a pressure-transmitting medium. All DSC samples were prepared in air using 3-10 mg of sample and were sealed in aluminum pans (purchased from Netzsch) with a pierced lid. An empty, aluminum pan with a pinhole was used as a reference. All measurements were carried out in a dynamic gas environment with a 50 ml min−1 He. Otherwise noted, heating and cooling rates of 2 K min−1 were used during isobaric measurements.
During the pressure cycling experiments, heat flow signals were measured over time under the repeated application and removal of a hydrostatic pressure of 150 bar at 311 K for (DA)2MnCl4 and 307 K for (NA)2CuBr4. The pressure linearly increased at a rate of 6 bar min−1 and asymptotically decreased at an average rate of 13 bar min−1. During the pressure change, quasi-isothermal conditions were maintained, where a small change in temperature (<1 K) induced by gas compression and decompression is compensated by external thermal control measures. The pressurization and depressurization processes are associated with average temperature fluctuations of 0.3 K and 0.7 K, respectively. To distinguish the heat flow signals associated with pressure-induced phase transitions of samples from those associated with compression and decompression of pressure transmitting medium (He gas), (C12H25NH3)2MnCl4, prepared from the previously reported procedure (G. F. Needham, et al., J. Phys. Chem. 88, 674-680 (1984), was used as a blank sample because it undergoes transitions at temperatures (Ttr, major=331 K and Ttr,minor=335 K) well above the set temperatures for (DA)2MnCl4 and (NA)2CuBr4. The heat flow signals measured from the blank during the pressure change at the set temperature were modeled as a baseline for the sample data. By subtracting the features in the heat flow associated with the gas compression and decompression from the sample data, we were able to determine the onset pressure associated with the pressure-induced transitions during compression and decompression processes. Because maintaining isothermal conditions becomes challenging once the phase transition is induced and both pressure and temperature change drive the endothermic and exothermic transitions to completion, we did not integrate the heat flow signals and focused only on using the information to determine the onset pressures for transitions.
On the barocaloric effects outside the transition: We note that the ΔSit values obtained through our HP-DSC experiments do not include the contributions from the additional barocaloric effects (ΔS+) that arise from each phase. This additional contribution can be estimated through ΔS+=−[(∂V/∂T)p=0]ΔP, where ΔP and (∂V/∂T)p=0 denote a driving pressure and a thermal expansion at the ambient pressure, respectively. Note that the isothermal entropy contribution is derived from the Maxwell relation (∂V/∂T)p=−(∂V/∂P)T with an assumption that the volume expansion is independent of pressure. The ΔS+ values are estimated to be ˜3 and ˜4 J kg−1 K−1 at the ordered and disordered phases, respectively, under 150 bar driving pressures. Although these values are small in comparison with ΔStr, these contributions can be large at higher driving pressure, such as ˜# at 400 bar, because of large thermal expansion coefficients (˜10−4 K−1)
Sample density was determined using an InstruQuest Inc. μ-ThermoPyc variable temperature He pycnometer. In a typical measurement, ca. 150 mg of crystalline sample were transferred to the sample holder, and the sample mass obtained. The holder was then placed in the instrument test chamber and the headspace evacuated and refilled five times to obtain a pure He atmosphere. The sample was then cycled multiple times through the order-disorder transition with the chamber volume determined every 2-5° C. away from the transition and every 0.5-1.0° C. close to the transition. For each point, the temperature was fully equilibrated with a standard deviation of no more than 0.2° C. prior to volume measurement. At each temperature, the chamber volume was measured five times to obtain good statistics. Sample volume was then determined by subtracting the average observed chamber volume from the volume of the empty sample holder which had been measured previously. The sample mass was redetermined after the measurement and found to have decreased by no more than 0.5 mg, likely due to loss of adsorbed water. Uncertainties of the reported densities were determined by propagation of the standard deviations of the empty and filled chamber volumes and the sample mass.
X-ray diffraction analyses were performed on a single crystal coated with Paratone-N oil and mounted on a MiTeGen microloops, at different temperatures (100 to 335 K) controlled by an Oxford Cryostreams nitrogen flow apparatus. Crystals were mounted at 270 K, and 270 K data sets were collected. Crystals were then cooled to 100 K for 100 K data collection. After 100 K data sets, high-temperature data sets were collected, 330 K for (DA)2MnCl4 and 335 K for (NA)2CuBr4. The temperature was manipulated at a rate of 60 K h−1. The intensities of the reflections were primarily collected by a Bruker D8 diffractometer with CMOS area detector (Mokα radiation, λ=0.71073 Å). The collection method involved 0.5° scans in ω at 23° in 2θ° with a detector distance at 9 cm for (DA)2MnCl4 and 8 cm for (NA)2CuBr4. Data integration down to 0.84 Å resolution was carried out using SAINT V8.37A with reflection spot size optimization (9). Most crystals were either single or merohedrally twinned and absorption corrections were made with the program SADABS. All single crystal structures were solved by the Intrinsic Phasing methods and refined by least-squares methods again F2 using SHELXT-2014 and SHELXL-2018 with OLEX 2 interface. Thermal parameters were refined anisotropically for all non-hydrogen atoms. Hydrogen atoms were placed geometrically and refined using a riding model for all structures. Crystal data as well as details of data collection and refinement are summarized in Tables 15 and 16, and geometric parameters are shown in Tables 10 to 14.
Notes on data quality, twining, and disorders: The phase transition from low-temperature phase to high-temperature (HT) phase during heating often resulted in fracturing and twining of the crystals, giving rise to the decay of crystal quality. Due to the large thermal motions of the long alkyl ammonium cations, the geometric parameters calculated for the HT phase structures should be treated as estimates, as pointed out in the literature.
Powder X-ray diffraction data for (DA)2MnCl4 and (NA)2CuBr4 were collected on beamline 17-BM-B at the Advanced Photon Source (APS) at Argonne National Laboratory. All X-ray wavelengths were between 0.24 Å and 0.45 Å, and are specified for each experiment in the relevant figures and tables. For variable temperature and pressure experiments, approximately 10 mg of samples was loaded into a sapphire capillary (1.524 mm×1.07 mm×50.8 mm, Saint-Gobain Crystals). Each capillary was attached to a custom-designed flow cell equipped with a gas valve, which was mounted onto the goniometer head and connected to a syringe pump that the applied the hydrostatic pressure of Helium (80-360 bar). Sample temperature was controlled by an Oxford Cryostream (Oxford Cryostream 800+). Prior to measurement, samples were evacuated by flowing ambient pressure of He gas for 10 minutes and annealed by heating 20 K above Ttr and cooling back to ambient temperature, at a rate of 6 K min−1 in the cryostream. The internal sample temperature was monitored via a K-type thermocouple (TC) that maintained the thermal contact with the powder sample within the capillary. Otherwise noted, the samples were heated and cooled by the cryostream at a rate of 6 K min−1, which resulted in the rate of ca. 3 K min−1 in the TC temperature due to the temperature gradient. Diffraction patterns were analyzed using the software TOPAS-R (Bruker AXS, version 3.0, 2005). Diffraction patterns at select temperature were indexed, and Le Bail refinements were performed to extract unit cell parameters.
Conformational entropy: For an alkylammonium chain CnH2n+1NH3+, there are n-2 rotatable C—C bonds that can contribute to formation of different conformers. Note that the conformation of alkylammonium chains can be described through a sequence of dihedral angles often referred to as Hoffmann's notation, with the terms g+, g−, and t denoting dihedral angles of approximately +60 (gauche), −60 (gauche), and 180° (trans), respectively. In principle, each C—C bond should be able to equally access these three rotameric states that correspond to local energy minima, and the change in solid-state configurational entropy can be described as ΔSconfiguration=R In W, where W is the ratio between the number of configurations in the disordered and ordered chains (That is, W=Ndisorder/Norder). Thus, the entropy change associated with the conformational disordering of an alkylammonium chain can be estimated as R In 3n−2.
However, depending on how chains are packed and how neighboring chains influence one another in each phase, the average number of “accessible” rotameric states for each C—C bond—defined as the chain flexibility number ϕ—can deviate from 3. Note that, in linear organic molecules, the flexibility number depends on the energetic difference between each conformer and a flexibility number of 2.85 has been used to predict the melting thermodynamics. In two-dimensional perovskites, a few structural features—steric restrictions (imposed by halide pockets and neighboring chains), correlations between torsions, and residual degrees of freedom present in C—C bonds in the ordered phase—can partially limit conformational degrees of freedom, reducing the number of C—C bonds associated with the chain melting processes by a restriction parameter β. The number of accessible conformations at the disordered phase can then be approximated as (ϕ)n−2−β, which includes the conformers arising from kink {gtg′} formation and cooperative torsion along the chain axis. Here, we assume that the two parameters ϕ and β are independent. Note that the entropy change associated with a reorientational motion (flipping) of the entire alkylammonium chains between two energetically equivalent orientations within the metal-halide pocket can be accounted for by adding the term R In 2. This chain-flipping can occur as an isolated minor transition or be coupled to the major order-disorder transition. In addition to the flipping, the major conformation disordering can be often distributed across multiple, successive transitions, and the total number of structural transitions of a compound is here referred to as m. The contributions from both flipping (ΔSflipping) and chain melting (ΔSmelting) to the total entropy change, ΔStotal, are multiplied by 2, because 2 mole of CnH2n+1NH3+ chains are present in 1 mole of perovskite. The contribution from any change in entropy associated with the MX6 octahedra is assumed to be negligible. Thus, the total difference in entropy associated m structural transitions in 2-D perovskites between low-temperature (LT) “ordered” and high-temperature (HT) “disordered” phases can be expressed through the following relationship:
ΔStotal=Σi=1mΔStr,i=ΔSflipping+ΔSmelting=2RIn2+2RIn(ϕ)n−2−β (1)
By fitting ΔStotal as a function of chain length n, we were able to determine the flexibility number ϕ and the restriction parameter β for two-dimensional perovskite (CnH2n+1NH3)2MX4 (M=Mn, Cu, Cd; X=Cl, Br). ΔStotal values are tabulated in Tables 2 to 4, and fitting parameters are summarized in Table 6. With the expectations that odd and even chain lengths should exhibit different thermodynamic trends, odd-numbered and even-numbered analogs were fit separately. We note that, somewhat surprisingly, the structural origins of odd-even effects of the phase-change thermodynamics in 2-D perovskites are not fully understood and require further investigations.
(DA)2MnCl4: From the fitting parameters (ϕ=3.0 and β=2.1, R2>0.99) for (CnH2n+1NH3)2MnCl4, (n=8-15), we can estimate that each decylammonium chain can access 35.9 (˜650) distinct configurations as a result of conformational disorder at the HT phase. In addition, the total entropy change associated with the single-step transition at 310 K can be calculated as ΔStotal,calc=ΔSflipping+ΔSmelting=2R In 2+2R In (3)5.9=119 J K−1 mol−1, which agrees with the experimentally measured ΔStotal,exp of 118 J K−1 mol−1. This estimation is further supported by the single-crystal structure of (DA)2MnCl4 at the HT phase (330 K), where each alkylammonium chain is disordered over two energetically equivalent positions and has a conformation with six C—C dihedral angles of 150-166° that deviate from the ideal trans dihedral angle of 180° and two C—C dihedral angles (C1-C2, 174°; C3-C4, 180°) close to the trans angle. (Table 13). This deviation may indicate fast “trans-gauche” rotations around C—C bonds and/or cooperative torsion along the chain axis. It is worth emphasizing that the chain length dependence of ΔStotal does not display a pronounced odd-even effect in 2-D Mn—Cl perovskites.
We also note that the conformational disordering is likely to be associated with the formation of one {gtg′} kink per chain (on average) near the chain ends and the most probable conformer is {t4gtg′t}, as suggested by incoherent neutron scattering experiments and intramolecular energy calculations. At higher temperature, some chains could adopt energetically less stable forms, in which the kink defects are located near the polar head groups, such as {t3gtg′t)} and {t2gt3g′t}. These proposed conformers are shown in the conceptual illustration in
The single-crystal structure at the HT phase also supports the proposed model that formation of kink is favored near the chain ends. In
(NA)2CuBr4: We note that the transition entropy of (NA)2CuBr4 is about 48% and 56% of ΔStotal of (C9)2MnCl4 and (C9)2CuCl4, respectively. This trend is also observed in the previously reported thermal data of (Cn)2CuBr4 (n=11-16), which shows that the transition entropies of 2-D Cu—Br perovskites are only 40-60% of those reported in Cu—Cl and Mn—Cl analogs. Fitting ΔStotal with chain length n results in ϕ=2.1 and 2.2 and β=5.1 and 5.5, for odd-numbered and even-numbered chains, respectively (Table 6). The lower value of ϕ and higher value of β, compared to those obtained from Mn—Cl and Cu—Cl series, indicate that the difference in solid-state conformational entropy between LT and HT phases is smaller in (Cn)2CuBr4, and the difference likely originates from the smaller difference in the flexibility of the chain and the number of newly rotating C—C bonds.
The LT phase crystal structure indicates that the smaller difference in solid-state entropy between LT and HT phase may result from the higher degree of disorder present in the nonylammonium chains at the LT phase. In the LT phase, each of the two chain conformations (chain A with C1-C2 gauche bond and chain B with C2-C3 gauche bond) is modeled with two-part disorder (Table 14). Note that the atomic positions in chain A and chain B were refined to 35/65% and 53/47% occupancies, respectively. The analysis on the chain conformations shows that (i) the chains are distorted near the methyl ends with C7-C8 dihedral angles of and +159°/−170° in chain A (Part 1/Part 2) and +164°/−164° in chain B and (ii) the chain A displays additional distortion in the C3-C4 bond (−159°/+169°). These results illustrate that configurational disorder is present in the LT phase of (NA)2CuBr4. In addition, the chains also display Uequiv values higher than those in decylammonium chain of (DA)2MnCl4 at the LT phase, whereas the Uequiv values at the HT phase were similar in both compounds (
Comparison with 2-D Cd—Cl and Cu—Cl perovskites. Although the chain length dependences of ΔStotal in 2-D Cd—Cl and Cu—Cl perovskites exhibit similar trends to that observed in the Mn—Cl series, with ϕ˜3 and β˜2, the trends slightly deviate from linearity and/or display pronounced odd-even effects. Even with the same chain length, these analogs display different phase transition behaviors, depending on the identity of metals. For example, unlike (C10)2MnCl4, both (C10)2CuCl4 and (C10)2CdCl4 display two-step transitions, and the major transition in (C10)2CuCl4 is followed by a minor transition, whereas the major transition in (C10)2CdCl4 is preceded by a minor transition. These compounds also display differences in chain conformations at room temperature: (C10)2MnCl4 with B conformer (C2-C3 gauche), (C10)2CdCl4 with both A conformer (C1-C2 gauche) and B conformer), and (C10)2CuCl4 with A, B, and all trans {t8} conformers. As expected, these changes translate to the most probable conformational disorder at the HT phase. Overall, these observations indicate that (i) the trends in thermodynamics of chain melting transitions are sensitive to chain packing and metals and (ii) the stepwise conformational disordering transitions observed in most compounds are complex. To fully describe the trends and the impact of the phase transition behaviors on barocaloric effects, detailed investigations into the structural changes and microscopic motions associated with each transition are required, both at ambient and applied pressures.
Comparison with melting of n-alkane: For the melting transition of n-alkane and solid-solid transitions of 2-D metal-halide perovskites and related systems, the relationships among transition entropy (ΔStr), chain length (n), and temperature of transition (Ttr) have been investigated, where ΔStotal normalized by chain length n is fit with Ttr. Although this approach, often referred to as SnT analysis, can provide insights into the trends in series of 2-D perovskites and related compounds, we did not include the temperature of transition as a fitting parameter because most 2-D perovskites undergo stepwise transitions and the nature of molecular motions associated with each transition is not fully understood. However, both approaches provide similar insights into the conformational degrees of freedom of alkylammonium chains of 2-D perovskites during the phase transition. For example, the SnT analysis indicates that the total transition entropy increases by 9.1 J K−1 mol−1 per carbon in (Cn)2MnCl4 (n=7, 11-17) and 13.5 J K−1 mol−1 per carbon in n-alkane. The chain length dependence of 9.1 J K−1 mol−1 in 2-D Mn—Cl perovskites can be translated to the flexibility number of 3, which agrees with our analysis. As previously pointed out, the ratio between the chain length dependences of ΔStotal in (Cn)2MnCl4 and n-alkane is 2:3 and correlates to the ratio between their dimensionalities. This interesting relationship is supported by a theoretical prediction based on kink-block transitions, where the melting entropy of alkyl chains confined in two-dimensional layers was shown to exhibit similar chain length dependence of 9.1 J K−1 mol−1. We also note that this value is close to the pressure dependences of melting transitions of n-decane (21 K kbar−1) and n-nonane (20 K kbar−1) (29). (nonane, with a dTdP solid-solid transition also similar).
Notes on dynamics: In addition to the kink formation (gauche defect), the cooperative torsion along the chains is coupled to the overall molecular motion at the HT phase. According to incoherent neutron scattering experiments, molecular motion corresponding to kink formation and cooperative torsion occurs over 1-5 ps timescale. However, due to the relatively low Q-values explored in the measurements, overall molecular motions, cooperative torsions, and kink formations were not accurately distinguishable. We also note that further studies are needed to fully model the microscopic details of the chain melting processes, both for fast motions (e.g., kink motion within a chain) and slow processes (e.g., the formation of “clusters” with similar chain conformations), at ambient and applied pressures.
Differences in phase transition behaviors between (DA)2MnCl4 and (DA)2CdCl4: First of all, their transition behaviors are qualitatively similar, with the chains at HT phase adopting approximately one kink conformation per chain. Neutron scattering experiments indicates, however, that the diffusion of kink, which was previously observed in (C10)2CdCl4 with ˜1000 ps time scale by proton NMR and 35Cl and 14N quadrupole resonance spectroscopies, is unlikely to occur in (C10)2MnCl4, because no cooperative conformational interconversion within a chain with a time scale greater than 20 ps was observed. This indicates that, unlike in (C10)2CdCl4, in (C10)2MnCl4, some kink formations are energetically more dominant. For the diffusion of the kinks to occur, several conformers with similar energies need to be in equilibrium. From the spectroscopic studies, it was revealed that (DA)2CdCl4 does have multiple conformers with similar energy levels. We also note that vibrational studies revealed that {tgtttg′tt} is the most probable conformer in (C10)2CdCl4, which is more flexible in the middle of the chain and does not have an end-gauche conformation.
Note that the IR signals used for conformational analysis are summarized in Tables 7 and 8.
C—H stretching: In the compounds containing long alkyl chains, shifts in C—H stretching peaks to higher wavenumbers are correlated with an increase in the number of gauche C—C bonds, a change in chain packing, and presence of order-disorder transitions. For example, in the melting transitions of n-alkanes, vsymmetric(C—H) and vanti-symmetric(C—H) shift from 2920 to 2928 cm−1 and 2850 to 2856 cm−1, respectively. 2-D perovskites also display similar trend, with the shifts Δv of 2-3 cm−1. The temperature dependence of these peaks can be either abrupt or gradual depending on metal and chain packing. Overall, this feature provides indirect evidence on the changes in the disorder in 2-D perovskites.
CH2 rocking and bending: For well-ordered chains in a monoclinic or orthorhombic lattice, CH2 rocking and bending bands split into doublets near 720 and 1470 cm−1, respectively, because of the factor group splitting that arises from directional intermolecular interactions between the chains. Generally, these features have been observed across materials with long hydrocarbon chains, including n-alkanes, layered silver thiolates, and 2-D metal-halide perovskites. The frequencies, shapes, and separations of these signals are correlated with orientations between neighboring chains and conformational disorder within the chain. In 2-D perovskites, the disappearance of the splitting can be used to determine if the chain undergoes conformational disordering, because it is correlated with the emergence of CH2 wagging bands specific to defect conformations (e.g., gtg′ kink) and indicative of the re-orientational motion of whole chain. However, we note that understanding specific microscopic motions associated with these features requires further investigations.
CH2 wagging: CH2 wagging bands, which provide insights into vibrational modes localized on a few CH2 units pinned in specific conformation sequences, are weakly coupled with the host lattice and independent of chain length. Thus, they provide characteristic signals of specific conformational defects. In 2-D perovskites, these modes are readily mixed with internal modes of chain end (CH3) and head (NH3+), and this coupling enhances the intensity of some CH2 wagging modes. Through normal-mode calculations on (C10)2CdCl4 and related compounds, the key peaks have been assigned. These signals appear at 1310 cm−1 (in-plane, two CH2 units, kink), a shoulder near 1370 cm−1 (out-of-plane, two CH2 units, kink), 1350 cm−1 (gg conformation), 1340 cm−1 (tg conformation near the chain end). The kink refers to gt2n+1g′-type conformational defects.
NH3 and CH3 bending: In 2-D M-Cl perovskites, the signal from NH3 symmetric bending modes appears around 1490 cm−1 at the LT phase and are often split into doublet due to the crystal field effect. The degree of the peak splitting highly depends on the identity of metals, with Mn—Cl and Cd—Cl perovskites displaying very small and often negligible splitting and Cu—Cl perovskites showing clear doublet (1492-1480 cm−1). At the HT phase, the peak splitting disappears, often accompanied by noticeable broadening, and this indicates the reorientational motion of the polar head within the halide pocket. The NH3 antisymmetric bending mode is associated with a strong singlet peak near 1580 cm−1 and red-shifts by 6-8 cm−1, as a compound undergoes a structural transition. This feature can be correlated with the decrease in the strength of N—H . . . Cl hydrogen bond. As these features are associated with the onset of chain melting transitions, vibrational spectra from NH3 bending can provide insights into the motions of chains within the halide pocket. The CH3 symmetrical bending mode, which appears near 1376 cm−1 at the LT phase, blue-shifts (Δv˜3 cm−1) as the chains undergo structural transitions. This feature is correlated with the change in inter-lamellar interactions within the organic bilayers.
Comparison of vibrational spectra related to chain conformations: IR spectra of (DA)2MnCl4 and (NA)2CuBr4 are summarized in
For symmetric C—H stretching peaks, both compounds display blue-shifts (˜2 cm−1) after the transitions, which supports that the phase transition introduces disorder in the alkylammonium chains (
Both compounds show pronounced differences in the progression of CH2 wagging bands, as shown in
Taken together, these results indicates that (i) the differences in conformations of and local environments around the chains in (NA)2CuBr4 is smaller than those in (DA)2MnCl4 and (ii) a noticeable degree of conformational disorder is present in (NA)2CuBr4 at the LT phase. These results, qualitatively, agree with our interpretations of single-crystal structures at the HT phase. Although the trends we observed from both compounds are consistent with those measured in other 2-D metal-halide perovskites as shown in Table 8, we note that accurate assignments and quantitative interpretation of IR spectra, particularly of CH2 wagging bands, are challenging, because the signals are relatively weak, coupled with the internal modes of chain end and head, and sensitive to the positions and diffusion of the defects, with the relationships among these factors not well understood. For accurate assignments of chain conformations, particularly those for (NA)2CuBr4, further investigations using normal-mode calculations and other complementary spectroscopic techniques, such as Raman or sum frequency generation vibrational spectroscopy, will be required.
Chemical origin of the difference in the solid-state disorder between (NA)2CuBr4 and (DA)2MnCl4: The difference in the disorder between the two compounds may arise from the difference in the size of metal-halide pocket and the strength of N—H . . . X (X=Cl, Br) hydrogen-bond interactions within the pocket. In particular, comparisons of IR spectra associated with NH3 bending modes provide useful insights into how the charge-assisted H-bond interactions differ between the two compounds. As shown in
Collectively, these results illustrate that, for (NA)2CuBr4, the weaker N—H . . . Br hydrogen bonds, in combination with a larger area provided for each chain (from incorporation of larger Br ions and presence of Jahn-Teller distortion), result in the increased degrees of freedom of chains in (NA)2CuBr4 at the LT phase, which contributes to the difference in the chain disorder between LT and HT phase smaller than the difference in (DA)2MnCl4. We also hypothesize that the smaller thermal hysteresis observed in (NA)2CuBr4 may arise from this enhancement in the conformational degrees of freedom, as it may lower the energy barrier associated with the formation of nucleation sites (i.e., “clusters” with similar chain conformations). The structure-property relationship between the two compounds provides insights into how conformational disorder in the 2-D perovskites can be controlled through chemical manipulations of the organic-inorganic interfaces.
aCn = CnH2n+1NH3.
bWhen a compound displays multiple transitions, the transition with the highest ΔS was labeled as a major transition.
cThe temperature of transition measured during the first heating scans are tabulated here.
dThis compound was synthesized and characterized for the completeness of the series.
eClosely spaced transitions were not resolved.
fThese compounds was synthesized and characterized by DSC at a slow scan rate (0.5 K/min) to resolve major and minor transitions.
gThe difference between the previously reported value are within experimental uncertainties; however, the higher Ttr reported here may represent the higher purity of the sample, because some of the long-chain amines used in the previous literature have been shown to contain impurities, such as amines with different chain lengths (n ± 2), which typically give rise to lowering of Ttr (up to 4-5 K), ΔH and ΔS.
317e
333e
aCn = CnH2n+1NH3.
bWhen a compound displays multiple transitions, the transition with the highest ΔS was labeled as a major transition.
cThe temperature of transition measured during the first heating scans are tabulated here. These minor transitions were not resolved in the initial reports. Note that reported full DSC traces of (Cn)2CuCl4 (n = 2-14), are reported without integrated thermodynamic values.
aCn = CnH2n+1NH3.
bWhen a compound displays multiple transitions, the transition with the highest ΔS was labeled as a major transition.
cThe temperature of transition measured during the first heating scans are tabulated here. Note that 2-D Pb-I perovskites have partially interdigitated organic bilayers.
aCn = CnH2n+1NH3.
bNote that the specific volume of each phase was estimated using the relationship, V = dAc × Mw/NA, where Ac is the area of metal-halide sheet per chain, d is the interlayer distance, Mw is the molecular weight, and NA is Avogadro's number. We used this estimation because the unit cell parameters of intermediate phases were often not available. The reported mean values of Ac are 26.5 Å2 and 27.5 Å2 for Mn—Cl and Cu—Cl perovskites, respectively. Note that this approach is likely to overestimate the volume change.
cBarocaloric coefficients were calculated through the Clausius-Clapeyron equation (dTtr/dP = ΔVtr/ΔStr).
aCn = CnH2n+1NH3.
avs and vas refer to symmetric and anti-symmetric modes, respectively.
bprevious reports on Cd—Cl and Mn—Cl analogs revealed that the peak near 1337 cm−1 does not depend on chain conformation (20, 37).
1376a
1589d
avs and vas refer to symmetric and anti-symmetric modes, respectively.
bkink generally refers to gt2n+19′-type conformational defects.
cthe splitting in Mn—Cl and Mn—Cl analogs tends to be small and often not discernible.
dtemperature-dependent band progression is not discussed
aPXRD data was obtained during cooling and the volume change was obtained through linear fit away from the transition.
acalculated as the distance between mean plane of four N atoms and mean plane of [MnCl4]2− layer
bcalculated from V/(Zd), where V is the unit cell volume, Z is the number of molecules in the unit cell, and d is the interlayer distance.
cestimated by d2, where d is the nearest metal-metal distance.
acalculated as the distance between mean plane of four N atoms and mean plane of [MnCl4]2− layer
bcalculated from V/(Zd), where V is the unit cell volume, Z is the number of molecules in the unit cell, and d is the interlayer distance.
ccalculated by d1 × d2, where di and d2 are metal-metal distances.
athe tilt angle is defined as the angle between a line connecting the atoms N and C and a plane through the metal atoms of the inorganic layers.
70 (2)
aPart 1 and part 2 refer to the sets of disordered positions of the alkylammonium chains conformer B (C2-C3 gauche bond)
aPart 1 and part 2 refer to the sets of disordered positions of the alkylammonium chains conformer conformers A (C1-C2 gauche) and B (C2-C3 gauche bond)
aR1 = Σ||Fo| − |Fc||/Σ|Fo|.
bwR2 = {Σ[w(Fo2 − Fc2)2]/Σ[w(Fo2)2]}1/2.
aR1 = Σ||Fo| − |Fc||/Σ|Fo|.
bwR2 = {Σ[w(Fo2 − Fc2)2]/Σ[w(Fo2)2]}1/2.
23.1g
38.2g
aDA = decylammonium; NA = nonylammonium; TPrA = tetrapropylammonium; bntrz = 4-(benzyl)-1,2,4-triazole; tcnset = 1,1,3,3-tetracyano-2-thioethylepropenide; L = 2,6-di(pyrazol-1-yl)pyridine.
bTransition temperatures measured during heating are tabulated here.
cEntropy of transition ΔStr measured at ambient pressure are tabulated here.
dΔThys refers to the difference between Ttr, heating and Ttr, cooling at ambient pressure.
ePrev is calculated through Prev = ΔThys/|dTtr/dP|, and dTtr/dP values for exothermic and endothermic transitions are used for conventional and inverse barocaloric materials, respectively. Note that inverse barocaloric materials refer to the compounds with dTtr/dP < 0.
fThe reversible isothermal entropy changes, ΔSit, rev, at the driving pressure ΔP are tabulated here. Note that these values were derived from quasi-direct measurements. At high pressure (typically above 1 kbar), the additional entropy change outside of the transition, ΔS+, plays a role and contributes to ΔSit values, leading to ΔSit to higher than ΔStr. The ΔSit measured at higher pressures are shown in the parentheses.
gdTtr/dP values were averaged from heating and cooling data.
hOnly irreversible ΔSit value is reported (Ni0.85Fe0.15S, 53 J kg−1 K−1 at 1000 bar; [FeL2](BF4)2, 68 J kg−1 K−1 at 430 bar).
ian irreversible phase transitions occur at the pressure above 4900 bar.
jdTtr/dP values obtained at a pressure range < 2 kbar values are shown. Note that (dTtr/dP)heating and (dTtr/dP)cooling were obtained from calorimetry and SQUID magnetometry, respectively.
aDA = decylammonium; NA = nonylammonium; TPrA = tetrapropylammonium; bntrz = 4-(benzyl)-1,2,4-triazole; tcnset = 1,1,3,3-tetracyano-2-thioethylepropenide.
bTransition temperatures measured during heating are tabulated here.
cThe reversible isothermal entropy change ΔSit, rev normalized by the driving pressure, often referred to as barocaloric strength, are tabulated here. Note that the barocaloric strength values were maximized by choosing the smallest ΔP values that can capture the full entropy of the transition.
dMaximum adiabatic temperature changes tabulated here were predicted by indirect methods, with ΔTad, max = TΔSit/cp, or, by quasi-direct methods.
eΔThys values measured at ambient pressure were used.
fThe second-law efficiency η, which corresponds to coefficient of performance (COP) of material with hysteresis normalized by COP of Carnot cycle, is estimated using the equation
Note that this relation is derived from a phenomenological model that integrates the dissipative losses due to hysteresis in a Carnot-like cycle and provides insights into how thermal hysteresis of a material reduces the efficiency.
gEstimated through the indirect method, with ΔSit, rev of 210 J kg−1 K−1 predicted to occur at the driving pressure of 270 bar with T = 312 K and cp = 1550 J kg−1 K−1.
hEstimated through the indirect method, with ΔSit, rev of 68 J kg−1 K−1 from the pressure change of 150 bar with T = 306 K and cp = 800 J kg−1 K−1.
iEstimated through the indirect method, with ΔSit, rev of 68 J kg−1 K−1 from the pressure change of 150 bar with T = 332 K and cp = 2450 J kg−1 K−1.
jEstimated to be around 50 K and later confirmed through quasi-direct measurement to be 45 K at the driving pressure of 5700 bar.
kquasi-direct measurements at the driving pressure of 5.9 kbar.
lEstimated through the indirect method, with ΔSit, rev of 60 J kg−1 K−1 predicted to occur at the driving pressure of 1000 bar with cp = 1700 J kg−1 K−1.
mEstimated through the indirect method, with the irreversible ΔSit value of 53 J kg−1 K−1 at the driving pressure of 1000 bar.
mQuasi-direct measurements at the driving pressure of 2600 bar.
34e
aDA = decylammonium; NA = nonylammonium.
bTransition temperatures measured during heating are tabulated here.
cEntropy of transition ΔStr measured at ambient pressure are tabulated here. The literature values are associated with large uncertainty.
dBarocaloric coefficients tabulated here were measured through high-pressure differential calorimetry under Helium gas environment. Thermal hysteresis were measured at ambient pressure.
ePhase-change properties, including the change in volume, was previously reported.
fCompounds listed here are predicted to display large barocaloric effects, due to high ΔStr and large volume change (ΔVtr)of ~7%.
gPredicted to display large inverse barocaloric effects due to ΔVtr, = −7%.
hPredicted to display large barocaloric effects due to large ΔVtr ~11%. However, the reversibility of the major transition requires further investigation.
iPredicted to display large barocaloric effects due to large ΔVtr of 2% and 5% for (C18H37)3NH+ and (C18H37)4N+, respectively.
jFour successive transitions occur at a temperature range between 290 K and 330 K, with a major transition around 305 K. The total entropy change is tabulated here. To probe reversibility and volume change, further investigations are required.
Table 27. Phase-change properties of newly synthesized asymmetric dialkylammonium salts. Preliminary characterizations through high-pressure differential scanning calorimetry (HP-DSC) indicate that these candidate compounds all display high pressure sensitivity (dTtr/dP) between 20-30 K kbar−1.
Other embodiments are in the claims.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2022/023891 | 4/7/2022 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63171938 | Apr 2021 | US |
