Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Sep 29;27(19):e70581. doi: 10.1002/cphc.70581

Structural and Spectroscopic Evidence of Hydrogen Bonded Anionic Wires in Protic Ionic Liquids

Maham Azhar 1, Lasse Hunger 1, Loai Al‐Sheakh 1, Andreas Appelhagen 1, Lennart Kruse 1, Ralf Ludwig 1,2,3,✉
PMCID: PMC13624368  PMID: 42811607

Abstract

Hydrogen Bonding (H‐bonds) between ions of like charge challenges the fundamental principle of electrostatics that opposite charges attract and like charges repel. This study provides direct structural and spectroscopic evidence of H‐bonded anionic wires in protic ionic liquids (PILs), which consist of triethylammonium cations and hydrogen dicarboxylate anions. The anionic wires were identified by X‐ray crystallography in the solid state, wherein the anions are linked by −OH···O− type H‐bonds. The H‐bonds within the anionic wire are significantly stronger than those between cation and anion despite repulsive Coulomb interaction in the first and attractive ones in the latter case. Analysis by nuclear magnetic resonance (NMR) and infrared (IR) spectroscopy suggests that the binding motifs of the solid structure are mainly preserved in the liquid state of the supercooled PILs. This conclusion is drawn from density functional theory (DFT) calculations of the same structural motifs as observed in the solid state. The calculated 1H NMR chemical shifts and IR C═O frequencies for these H‐bond structures correspond well with the measured spectroscopic signatures. The anionic chains remain intact even when gradually removing the counterions. Our findings highlight the fundamental role of hydrogen bonding for the unusual attractive interaction between ions of like charge.

Keywords: hydrogen bonds, infrared spectroscopy, nuclear magnetic resonance spectroscopy, quantum chemistry, X‐ray crystallography

Anionic wires of deprotonated carboxylic acids challenge the general wisdom that like charges repel. The hydrogen bond motif observed in the X‐ray structure is even preserved in the liquid state of the material as shown by δ 1H NMR spectroscopy. The experimental and theoretical findings provide deep insight into the nature of hydrogen bonding. The highly delocalized protons in the wires hold promise for materials with excellent proton conductivity.

graphic file with name CPHC-27-e70581-g006.webp

1. Introduction

Hydrogen bonds (H‐bonds) are ubiquitous in nature and essential for life [1]. In liquid water, a flickering three‐dimensional network of H‐bonds is responsible for its unique properties [2, 3]. In proteins, H‐bonds stabilize helices and beta sheets and control the fold into functional shapes [4, 5]. The base pairs in double‐stranded DNA are held together by double and triple H‐bonds, allowing them to unzip for replication without falling apart [6]. H‐bonds play an important role even in ionic liquids (ILs), although they are an order of magnitude weaker than the ionic interactions between cations and anions in this liquid material [7, 8, 9, 10]. This may not be surprising for H‐bonds between ions of opposite charges, which further strengthen the attractive interaction between cation and anion. In contrast, H‐bonds between ions of like charge seem to be unusual, challenging the general wisdom that opposite charges attract but like charges repel. Nevertheless, H‐bonds between cations have been observed in hydroxy‐functionalized ILs [11, 12]. In linear and cyclic clusters, the interaction between ions of like charge is additionally strengthened by cooperative H‐bonds. The recipe for enhanced formation of cationic clusters includes weakly interacting anions, polarizable cations, and longer alkyl chains, which increases the distance between the hydroxy functional groups and the positively charged ring system of the cations, thereby reducing Coulomb repulsion [13, 14]. More recently, we replaced the hydroxy‐ by carboxy‐functionalized cations in similar ILs. We observed doubly H‐bonded cationic dimers in the solid, liquid, and gas phases [15, 16, 17]. The carboxy‐functionalized ILs exhibit the same structural motif as that known for formic and acetic acid in the gas phase [18, 19]. Obviously, the double H‐bonds are strong enough to overcome the repulsive Coulomb forces between the cations in all phases. If H‐bonds are strong and directional, microscopic systems display the H‐bond motifs present in the bulk.

In this work, we take on the challenge of finding similar structural motifs for anionic clusters in protic ionic liquids (PILs). However, the synthesis of PILs comprising anionic dimers or even larger like‐charged clusters is much more difficult to realize and thus rarely discussed in the literature [20, 21, 22]. Herein, we show that the syntheses from triethylamine (TEA) and dicarboxylic acids lead to PILs exhibiting such anionic structural motifs. This kind of PILs have been synthesized because herein both ions provide single donor functional groups NH and OH, which allow studying the competition between well‐defined cation–anion +NH···O− and anion–anion −OH···O− types of H‐bonds. Here, fundamental understanding of interionic interaction, in particular the nature of inter anionic interaction and the important role of H‐bonds, is required. Instead of doubly H‐bonded anions, we observed infinite chains of singly H‐bonded anions. The H‐bonded anionic wires are flanked by additional H‐bonds between cation and anion. X‐ray crystallography provide evidence that the anion–anion H‐bonds are stronger than the cation‐anion H‐bonds despite repulsive Coulomb interaction in the first and attractive ones in the latter case. The structural motifs observed in the solid state survive the transition to the liquid phase, as demonstrated by 1H nuclear magnetic resonance (NMR) and infrared (IR) spectroscopy. Density functional theory (DFT) calculations on differently charged cluster species clarify the role of H‐bonding between like‐charged ions and whether they can be regarded as “antielectrostatic” H‐bonds (AEHB) or rather as simple interanionic H‐bonds (IAHB) [23, 24, 25, 26, 27, 28, 29].

2. Results and Discussion

The starting point of our investigation has been the synthesis of the ILs or low melting salts via a simple acid–base reaction of dicarboxylic acids and TEA as the base. For studying the effect of adding one methylene group CH2 between the two carboxylic groups, we used two different alkanedicarboxylic acids (HOOC(CH2) n COOH) with n = 0,1, namely, ethanedionic (oxalic) and propanedionic (malonic) acid (Scheme 1) [30, 31]. The general synthesis procedure is described in the Supporting Information.

SCHEME 1.

SCHEME 1

Open in a new tab

Reaction scheme for the synthesis of the protic ionic liquids of type triethylammonium hydrogen dicarboxylate [TEA][OOC(CH2) n COOH] with n = 0,1 obtained from triethyl amine and oxalic acid (n = 0) and malonic acid (n = 1), respectively.

2.1. Crystal Structures Exhibit Hydrogen Bonded Anionic Wires

The crystal structures of triethylammonium hydrogen oxalate and triethylammonium hydrogen malonate have been determined by X‐ray crystallography [32, 33, 34, 35]. The CDS deposition numbers are 2386822 and 2386827, respectively. Figure 1 shows the X‐ray structure of triethylammonium hydrogen oxalate. We observe hydrogen‐bonded wires consisting of alternating dicarboxylic acid molecules and doubly negatively charged oxalate anions. Thereby, the two OH groups of the molecule serve as H‐bond donors to the neighboring oxygen acceptors of the oxalate anions. The OH···O─ H‐bond between molecule and anion is very strong, suggested by the O···O distances of 2.479 Å and the elongated OH‐bond length of about 1.081 Å (see Table 1 for all geometry values). The H‐bond angle ∠(O,H,O) is 179.4°, thus perfectly linear. The triethylammonium cation exclusively forms H‐bonds to the other oxygen of the oxalate anion. The N···O distance is 2.900 Å and the NH···O distance is 1.877 Å. The NH‐bond is only 0.96 Å long, and the H‐bond angle ∠(N,H,O) is 160.5°, both suggesting moderate H‐bonds despite the Coulomb attraction between the opposite charged ions. Because we were aiming for singly H‐bonded anionic wires, we switched to triethylammonium malonate synthesized from malonic acid, which has a significantly larger pKa1 value than oxalic acid (pKa1 = 2.83 vs. pKa1 = 1.23) [36]. The oxalate and malonate anions only differ by introducing a single methylene group between the charged entities.

FIGURE 1.

FIGURE 1

Open in a new tab

Crystal structure of triethylammonium hydrogen oxalate [TEA][OOCCOOH] showing H‐bonded wires alternately built by dicarboxylate acid molecules and doubly negative charged oxalate anions. Thereby, the two OH groups of the molecule serve as H‐bond donors to the neighbored oxygen acceptors of the oxalate anions. The H‐bonds characterized by anion–anion −OH···OC− interaction and flanked by cation‐anion +NH···O− interaction. The H‐bonds are denoted by dashed lines. In the ORTEP representation, the atoms are represented by ellipsoids at 50% probability levels.

TABLE 1.

Hydrogen bond donor–acceptor distances (in Å) and angles in the crystalline structures of triethylammonium hydrogen oxalate and triethylammonium malonate.

[TEA][OOCCOOH] [TEA][OOC(CH2)COOH]
R(NH···O)/Å 1.877 1.875
R(OH···O)/Å 1.398 1.387
R(N···O)/Å 2.800 2.735
R(O···O)/Å 2.479 2.479
R(N─H)/Å 0.960 0.886
R(O─H)/Å 1.081 1.094
R(C─O)/Åa 1.246 1.242
R(C─O)/Åb 1.267 1.265
∠(O,H,O) 179.4 174.8
∠(N,H,O) 160.4 163.5
Open in a new tab
a

CO bond involved in the moderately strong H‐bonds +NH···OC−.

b

CO bond involved in the significantly stronger H‐bonds −OH···OC−.

Figure 2a shows the X‐ray structure of triethylammonium hydrogen malonate. We observe that the hydrogen malonate anions are arranged in infinite single‐H‐bonded wires. Herein, the OH group of each anion is oriented at a carboxylate O atom of the next anion in the wire. The H‐bonds −OH···O− between the anions are again flanked by H‐bonds +NH···O− between the triethylammonium cation and hydrogen malonate anion. These H‐bonds show typical H‐bond behavior, as known for molecular systems. The N···O distance is 2.735 Å and the NH···O distance is 1.875 Å. The H‐bond angle ∠(N,H,O) is 163.5°, also indicating that the H‐bonds are moderate despite the Coulomb attraction between the opposite charged ions. In stark contrast, the −OH···O− type H‐bonds between the anions in the anionic wires are very strong despite the repulsive Coulomb interaction. The O···O distance is only 2.479 Å and the OH···O distance is 1.388 Å. The H‐bond angle ∠(O,H,O) is 174.8°, thus close to being linear. Within this strong −OH···O− type H‐bonds, the OH bond length is R(OH) = 1.09 Å and thus elongated more than 10 pm compared to the typical gas‐phase value of monomeric carboxylic acids, being 0.972 Å in formic acid and 0.971 Å in acetic acid [37, 38, 39]. The typical OH bond lengths for water are 0.945 Å in the gas phase, 0.98 Å for liquid water, and 1.01 Å in ice [3]. Thus, the elongation of the OH bond between two anions exceeds the elongation of the OH bond in water and ice due to the H‐bond network. Such lengthening is inherent to strong H‐bonds and is a consequence of weakening of the covalent OH bond by the −OH···O− interaction. Further evidence for only moderate H‐bonds +NH···O− between cation and anion but strong −OH···O− between the anions is given by the CO bond lengths of the carboxylate group (see Figure 2b). The CO bond involved in the moderately strong H‐bonds +NH···OC− is 1.242 Å, whereas the CO bond involved in the significantly stronger H‐bonds −OH···OC− is elongated to 1.265 Å. This means that the CO bond involved in the anion–anion H‐bonds is 0.023 Å longer than the CO bond involved in the cation–anion H‐bonds, clearly indicating much stronger H‐bonds between two anions compared to those between cation and anion. Recently, Steiner and Saenger correlated the OH and H···O distance of 125 OH···O interactions found in neutron diffraction crystal structures extracted from the Cambridge Structural Database [39, 40]. This data included different kinds of H‐bonds: OH···O, OH···O─, ─OH···O, and −OH···O−. Using their relation, our measured intermolecular H···O distance of about 1.388 Å predicts an OH bond length of about 1.1 Å, which is exactly the OH bond length we measured. Here, we conclude that the H‐bonds between like charged ions are surprisingly strong but do not differ from classical H‐bonds. Considering that the covalent OH bond distances are underestimated by about 0.1 Å in X‐ray crystallography, the proton can be assumed to be almost equally shared between the two O‐atoms, resulting in the extreme case of the centered H‐bonds O···H···O. This suggests that the PILs may be efficient proton‐conducting materials [41, 42].

FIGURE 2.

FIGURE 2

Open in a new tab

(a) Crystal structure of triethylammonium hydrogen malonate [TEA][OOC(CH2)COOH] showing anionic wires characterized by anion–anion −OH···OC− H‐bonds and flanked by cation–anion +NH···O− H‐bonds. The H‐bonds are denoted by dashed lines. In the ORTEP representation, the atoms are represented by ellipsoids at 50% probability levels. (b) Enlarged section of the X‐ray structure of triethyl ammonium hydrogen malonate showing the H‐bond distances d(O2H···O3) between anion and anion and the H‐bond distances d(N1···O4) between cation and anion. Additionally, the covalent CO bond distances d(C7‐O1) of the carbonyl group and d(C9‐O2) and d(C9‐O4) of the carboxylate group are shown. Distances are given in Å.

2.2. H‐Bonded Anionic Wires Preserved in the Liquid State Shown by NMR and IR Spectroscopy

Now we address the question, whether the anionic wires and the H‐bonds, −OH···O− and +NH···O−, survive the melting from the solid into the liquid state of the triethylammonium hydrogen malonate. For finding structural evidence, we used NMR and IR spectroscopy. While 1H NMR is commonly used to probe interionic H‐bonds, it is often limited by proton exchange and line broadening due to chemical exchange [29]. However, for the viscous triethylammonium hydrogen malonate, the proton exchange is slowed down at 298 K, so that we observed two separate signals for the OH and the NH protons around 14 and 10 ppm, respectively. Strongly downfield‐shifted OH protons of the carboxy group ‐ typically ranging between 12–15 ppm ‐ indicate very strong H‐bonds, as shown for carboxylic acids [43]. In Figure 3, we show the 1H NMR chemical shift range from 7 to 16 ppm as a function of temperature. The full 1H and 13C spectra can be found in the Supporting Information. The recorded spectra at 298 and 303 K clearly show two signals, which are referred to the downfield chemical shifts of the OH protons (red area) and NH protons (blue area). Additional evidence for this assignment of the signals is provided by DFT calculations at the B3LYP‐D3/6‐31+G* level of theory on structures showing the H‐bond patterns as observed in the X‐ray structures [44, 45, 46, 47]. Calculated proton chemical shifts were derived against the reference tetramethylsilane (TMS), similar to the experiment, and reflect our assignment of the signals in the measured 1H NMR spectra (Figure 3). The fact that the signals of OH and NH protons of the solid‐like calculated clusters are found in the same ppm range as those of the measured liquid state structures suggests that the structural features observed in the X‐ray patterns are mainly preserved in the liquid state of the supercooled PILs. Figure 3 also shows that above 303 K, the two signals coalesce to one due to fast exchange of the NH and OH protons.

FIGURE 3.

FIGURE 3

Open in a new tab

NMR proton chemical shifts δ 1H of the COOH and the NH groups observed for triethylammonium hydrogen malonate [TEA][OOC(CH2)COOH] as a function of temperature. At 298 K the proton exchange is obviously slow and we observe two broad signals at around 10 ppm and slightly above 14 ppm. They are assigned to chemical shifts δ 1H of the NH and the COOH protons, lying in the blue and red areas of the enlarged spectra. The assignment is supported by the DFT calculated proton chemical shifts (vs. TMS) of stable H‐bonds clusters [TEA+]4[−OOC(CH2)COOH]4 as indicated by the squares. At high temperatures the two signals coalesce due to fast proton exchange.

A valuable method for investigating the structural features present in the liquid triethylammonium hydrogen malonate is Fourier‐ transform infrared (FT‐IR) spectroscopy. We measured the IR absorption spectra in the liquid state for a broad range of temperatures between 298 and 353 K in 10 K steps. It has turned out that this PIL can be supercooled and measured below the melting point. Although covering wavenumbers between 900 and 4000 cm−1, we mainly focus on the CO stretching vibrational bands, which are usually observed between 1600 and 1800 cm−1. As discussed for the X‐ray structure of solid triethylammonium hydrogen malonate, the CO bond distances in the carboxy and carboxylate groups depend on the intermolecular interaction strength. Thus, the CO vibrational bands ν CO are sensitive probes of H‐bonding, namely, in the manner of shifting to lower wavenumbers with increasing H‐bond strength. The X‐ray structure of the crystalline PIL showed that the two carboxylate oxygens are involved in strong H‐bonds −OH···OC− and moderate H‐bonds +NH···OC−, as indicated by the differently elongated CO bond distances R(CO) = 1.265 Å and R(CO) = 1.242 Å (see Table 1). In contrast, the oxygen of the carboxyl group is not involved in H‐bonds, as expressed by the shortest CO bond distance R(CO) = 1.214 Å. The H‐bond dependence of the CO bond lengths should be reflected in the corresponding CO stretching frequencies ν CO. That is what we observe in the IR spectra shown in Figure 4 . We assign the CO stretching band with the highest frequency at 1720 cm−1 to the non‐H‐bonded CO in the carbonyl group. The redshifted CO stretches around 1600 cm−1 are referred to the CO bonds of the carboxylate group that are heavily involved in H‐bonds. This clear assignment is supported by DFT frequency calculations at the B3LYP‐D3/6‐31+G* level of theory on structures resembling the anionic wires observed in the X‐ray structures [44, 45, 46, 47]. In Figure 4, we show that the calculated frequencies of the non‐H‐bonded CO fall into the red area and the H‐bonded CO into the blue area in accordance with our assignment. Obviously, the structural H‐bond motifs observed in the crystal structure of hydrogen triethyl ammonium malonate are still present in the bulk liquid phase. The fact that the −OH···OC− interaction is stronger than the +NH···OC− interaction suggests that the interanion −OH…O− interactions are definitely based on H‐bonding between ions of like charge.

FIGURE 4.

FIGURE 4

Open in a new tab

Attenuated total reflection (ATR) IR spectra of triethylammonium hydrogen malonate [TEA][OOC(CH2)COOH] showing two main vibrational bands in the CO stretching region. The vibrational band at around 1720 cm−1 is assigned to the nonbonded CO group of the COOH carboxyl group (red area), whereas the strongly redshifted bands at around 1580 cm−1 are assigned to H‐bonded CO in the carboxylate group (blue area), as shown in the structural motif and assigned by the arrows. The assignment is supported by the DFT calculated CO vibrational bands obtained from stable H‐bonds clusters [TEA+]4[−OOC(CH2)COOH]4.

2.3. DFT Calculated Neutral and Charged Clusters Mimic the Condensed Phase H‐Bond Motifs

Recently, this interpretation of interanion −OH···O− interactions as H‐bonds has been seriously challenged. Braga and coworkers argued that the −OH···O− interaction in potassium hydrogen oxalate cannot be regarded as a H‐bond despite the short O···O distance of 2.52 Å and the linearity of the geometry [20]. They questioned the anion–anion interaction from molecular electrostatic potential considerations and concluded that it is instead a “tugboat interaction,” controlling anion aggregation and minimizing anion–anion repulsions. They claimed that the anionic dimer would fall apart if the counter cations are removed. For demonstrating that we are dealing with real strong and directional H‐bonds within the anionic wires, we calculated neutral and charged stable H‐bond clusters [TEA+] m [−OOC(CH2)COOH]4 with m = 0–4 consisting of m triethylammonium cations and four hydrogen malonate anions as calculated at the B3LYP–D3/6–31+G* level of theory considering additional dispersion interaction as provided by Grimme's D3 method [44, 45, 46, 47]. In Figure 5, we show the neutral tetrameric cluster [TEA+]4[−OOC(CH2)COOH]4 followed by charged species, wherein successively triethylammonium cations are removed, resulting in overall charged species Q = 0, Q = 1−, Q = 2−, and Q = 3−. The clusters show the tetrameric anionic wires and the H‐bond patterns as observed in the X‐ray structures. The calculated binding energies ΔE per ion of the neutral and charged stable H‐bond clusters are −280.8 kJ mol−1 for Q = 0, −257.7 kJ mol−1 for Q = 1−, −201.8 kJ mol−1 for Q = 2−, and −73.5 kJ mol−1 for Q = 3−, as shown in Figure 6. Thus, even for the threefold negatively charged cluster [TEA+][−OOC(CH2)COOH]4, the binding energies ΔE per ion are still negative. The overall binding energy for the Q = 3− cluster sums up to −367.5 kJ mol−1, which is similar to the calculated binding energy of the ion pair [TEA+][−OOC(CH2)COOH] as isolated from this structure, indicating attractive H‐bonding within the anionic wire despite the repulsive Coulomb interaction.

FIGURE 5.

FIGURE 5

Open in a new tab

Neutral and charged stable H‐bonds clusters [TEA+] m [−OOC(CH2)COOH]4 with m = 0–4 consisting of m triethylammonium cations and four hydrogen malonate anions as calculated at the B3LYP–D3/6–31+G* level of theory. The neutral tetrameric cluster [TEA+]4[−OOC(CH2)COOH]4 (Q = 0) is followed by charged species wherein successively triethylammonium cations are removed, resulting in charged species Q = 1−, Q = 2−, Q = 3−, and finally the pure linear anionic tetramer (Q = 4−).

FIGURE 6.

FIGURE 6

Open in a new tab

The calculated binding energies ΔE per ion of the neutral and charged stable H‐bonds clusters [TEA+] m [−OOC(CH2)COOH]4 with m = 0–4. The ΔE values are −280 kJ mol−1 for Q = 0, −250 kJ mol−1 for Q = 1−, −200 kJ mol−1 for Q = 2−, and −75 kJ mol−1 for Q = 3−. Thus, even for the threefold negatively charged cluster the binding energies ΔE per ion is still negative. The overall binding energy for the Q = 3− cluster sums up to −367.5 kJ mol−1, which is even higher than the calculated binding energy of an ion pair [TEA+][−OOC(CH2)COOH] in indicating attractive H‐bonding within the anionic wire despite the repulsive Coulomb interaction.

2.4. Double‐Charged Hydrogen Malonate Dimer Shows Robust Kinetic Stability

In Figure 6, we also show the energies per ion for the pure anionic clusters [HOOC(CH2)COO−] m for m = 2,3,4. The positive ΔE values show that the repulsive Coulomb are stronger than the attractive H‐bonds between the anions. However, none of the anionic clusters falls apart during the geometry optimization and showed substantial kinetic stability, as observed for hydroxy‐ and carboxy‐functionalized cations of pyridinium‐based ILs [11, 12, 13, 14, 15, 16, 17]. This is demonstrated for the double‐charged hydrogen malonate dimer [HOOC(CH2)COO−]2 in more detail. The potential energy curve for the hydrogen‐bonded anionic dimer [HOOC(CH2)COO−]2, as calculated at the B3LYP–D3/6–31+G* level of theory, shows a minimum with a clear dissociation barrier preventing “Coulomb explosion” into separated anions. We achieved robust kinetic stabilities while dissociating the anionic dimer along the R(COH···OC) H‐bond stretching coordinate with respect to the energy of the isolated cations at ΔE = 0. For each intermolecular distance R(COH…OC), all other geometrical variables were optimized at each 0.5 pm step of the scan. The relaxed potential energy curve for the hydrogen malonate dimer [HOOC(CH2)COO−]2 is shown in Figure 7. We indicate the minimum energy ΔE = 123.89 kJ mol−1 at the equilibrium H‐bond distance R(H…O) = 1.701 Å. The distance R(O…O) = 2.688 Å is extremely short too despite the strong repulsive Coulomb interaction present due to only one methylene group spacer between the carboxyl and the carboxylate groups. The intermolecular oxygen–oxygen distance in this anionic dimer is even shorter than that in ice (2.72 Å), exhibiting strong hydrogen bonds and even 0.19 Å shorter than that of water dimers calculated at the same level of theory [40]. For the anionic hydrogen malonate dimer [HOOC(CH2)COO−]2, we could calculate energy potential curves well beyond the activation barrier height. In Figure 7 we show the effective well depths ΔE* with respect to the transition state at ΔR*(H…O) that signals descent towards separated cations. For this anionic dimer, a distance of about 3.1 Å corresponds to an activation barrier height of about ΔE* = 18.3 kJ mol−1, showing the enormous strength of the H‐bonds.

FIGURE 7.

FIGURE 7

Open in a new tab

Potential energy curve for the hydrogen‐bonded anionic dimer [−OOC(CH2)COOH]2, as calculated at the B3LYP–D3/6–31+G* level of theory, relaxing the structure for each hydrogen bond length R(COH···OC). The curve progression is characterized by a clear minimum energy ΔE of about 123.99 kJ mol−1 at the equilibrium bond distance R(COH···OC) = 1.70 Å. The dissociative energy profile shows a pronounced binding well with respect to the dissociated anions and effective equilibrium well depths ΔE* = 18.3 kJ mol−1 at ΔR* = 3.10 Å with respect to the transition state.

2.5. Natural bond orbital (NBO) Descriptors Strongly Correlate With NMR Proton Chemical Shifts and Vibrational Frequencies

The kinetically stable anionic dimers, trimers, and tetramers show that the short‐range donor–acceptor covalency forces successfully compete with the powerful long‐range electrostatic repulsions [15, 16, 17]. The characteristic covalency features of the H‐bonds −OH···O− can be readily recognized in the framework of the NBO analysis as distinctive n O → σ*OH donor–acceptor interactions expressed by the second‐order stabilization energies ΔE (2) n → σ* (see Supporting Information) [48, 49, 50]. The charge from the oxygen lone‐pair orbitals of one of the carboxylate groups CO of a malonate anion is donated into the OH antibonding orbital of the carboxyl group of the next anion and so on. This way, the short‐range donor–acceptor covalency forces overcome the strong long‐range electrostatic repulsive forces, as expected for ions of like charge. Here, we show that the NBO approach is also suitable for describing the energetic and spectroscopic features observed for the hydrogen‐bonded anionic wires in hydrogen malonate.

In Figure 8, we show the NBO energy versus the spectroscopic descriptors NMR proton chemical shifts δ 1H and infrared vibrational frequencies ν CO for the anionic dimers, trimers, and tetramers. As shown, the second‐order stabilization energies ΔE (2) n → σ* are almost linearly correlated with the spectral descriptors calculated for the anionic clusters. The increasing ΔE (2) n → σ* values are reflected by downfield chemical shifts δ 1H of about 2 ppm and redshifted vibrational frequencies ν CO of about 43 cm−1. Despite the increasingly positive values for ΔE with increasing cluster size due to the strong repulsive Coulomb forces between the anions, the spectral signatures show typical behavior of enhanced H‐bonding ranging from dimer to trimer to tetramer. Obviously, the second‐order stabilization energies ΔE (2) n → σ* seems to be a suitable measure for the strength of H‐bonding. We demonstrate the usefulness of NBO analysis and highlight that NMR and IR properties are sensitive probes of hydrogen bonding and suitable for characterizing the interaction strengths between ions of like charge. Overall, the NBO parameters also indicate the covalent nature of hydrogen bonding in anionic wires.

FIGURE 8.

FIGURE 8

Open in a new tab

NMR proton chemical shifts, δ 1H (relative to the standard tetramethylsilane, TMS) (a) and IR vibrational frequencies, ν CO (b) of hydrogen malonate dimers (blue), trimers (green) and tetramers (red) plotted versus NBO‐calculated second‐order stabilization energies ΔE (2) n → σ*. The almost linear dependences indicate the strong relation between NBO stabilization energies with geometric and spectroscopic properties of the anionic clusters.

Our combined experimental and theoretical approach shed some light into the vivid debate about so‐called “AEHB” [23, 24, 25, 26, 27]. This study suggests that the term IAHB is more suitable to describe the strong noncovalent interaction between ions of like charge [28, 29]. Nevertheless, our findings also challenge traditional electrostatic expectations and emphasize the important role of charge transfer and polarization for interionic H‐bonding, as expressed by spectral NMR and IR signatures. We found convincing structural and spectroscopic evidence that strong IAHB can even overcome the electrostatic repulsion. Thus, the claim of AEHB at least inspired the discussion about the nature of H‐bonding between ion of like charge and provoked experimental evidence for this kind of H‐bonds.

3. Conclusion

To summarize, we have studied the synthesis and solid‐state structural properties of triethylammonium hydrogen oxalate [TEA][OOCCOOH] and triethylammonium hydrogen malonate [TEA][OOC(CH2)COOH] PILs. X‐ray crystallography showed that the OH proton of the carboxy group of the anion forms single H‐bonds to one of the accepting oxygens of the carboxylate group of another anion −COH···OC−. The NH proton of the triethylammonium cation forms single H‐bonds to the other accepting oxygen of carboxylate group of each anion +NH···O− type H‐bonds. Surprisingly, the H‐bonds −OH···O− between the anions are stronger than the H‐bonds +NH···O− between cation and anion, finally resulting in the formation of anionic wires despite the repulsive Coulomb interaction between the ions of like charge. 1H NMR and ATR IR spectroscopy provided strong evidence that the structural motifs observed in the solid, including the anionic wires, are still present in the bulk liquid phase. DFT‐calculated charged clusters up to Q = 4− exhibit appreciable kinetic stability and show characteristic structural and spectroscopic signatures of H‐bonds, despite the powerful Coulomb repulsion. Overall, our integrated experimental and computational investigation provides compelling evidence for the formation of anionic wires, challenging the traditional electrostatic expectations. The highly delocalized protons in the wires hold promise for materials with excellent proton conductivity. This strategy for probing interionic H‐bonding can be extended to a wide range of ionic environments such as promising “water‐in‐salt” and “solvate‐ionic‐liquids” electrolytes [51, 52, 53, 54, 55].

Author Contributions

Maham Azhar: investigation (lead), data curation (equal), formal analysis (lead), writing – original draft (equal), writing – review and editing (equal). Lasse Hunger: data curation (equal), formal analysis (supporting). Loai Al‐Sheakh: data curation (supporting), formal analysis (supporting). Andreas Appelhagen: formal analysis (supporting). Lennart Kruse: formal analysis (supporting), writing – review and editing (equal). Ralf Ludwig: conceptualization (lead), project administration (lead), funding acquisition (lead), writing – original draft (equal), writing – review and editing (equal), resources (lead).

Funding

This study was supported by the Deutsche Forschungsgemeinschaft (SPECTRE Research Training Group (507189291/GRK 2943), 470038970/LU‐506/17‐1, 517661181/LU‐506/18‐1).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

A detailed synthesis procedure, corresponding 1H and 13C NMR spectra, experimental details of infrared and X‐ray spectroscopic measurements, as well as information on the theoretical calculations are available in the Supporting Material. The authors have cited additional references within the Supporting Information [56, 57].

CPHC-27-e70581-s001.pdf (402.5KB, pdf)

Acknowledgments

Support from the DFG through the SPECTRE Research Training Group (507189291/GRK 2943) and projects (470038970/LU‐506/17‐1) and (517661181/LU‐506/18‐1) is gratefully acknowledged.

Open Access funding enabled and organized by Projekt DEAL.

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article and from the corresponding author upon reasonable request.

References

  • 1. Jeffrey G. A. and Saenger W., Hydrogen Bonding in Biological Structures, (Springer Science & Business Media, 1991). [Google Scholar]
  • 2. Keutsch F. N. and Saykally R. J., “Water Clusters: Untangling the Mysteries of the Liquid, One Molecule at a Time,” Proceedings of the National Academy of Sciences 98, no. 19 (2001): 10533, 10.1073/pnas.191266498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Ludwig R., “Water: From Clusters to the Bulk,” Angewandte Chemie International Edition 40, no. 10 (2001): 1808, 10.1002/1521-3773(20010518)40:10<1808::AID-ANIE1808>3.0.CO;2-1. [DOI] [PubMed] [Google Scholar]
  • 4. Hubbard R. E. and Haider M. K., Hydrogen Bonds in Proteins: Role and Strength, (Encyclopedia of Life Sciences, 2010). [Google Scholar]
  • 5. Sheu S.‐Y., Yang D.‐Y., Selzle H. L., and Schlag E. W., “Energetics of Hydrogen Bonds in Peptides,“ Proceedings of the National Academy of Sciences 100, no. 22 (2003): 12683, 10.1073/pnas.2133366100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Watson J. D. and Crick F. H. C., “Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid,” Nature 171, no. 4356 (1953): 737, 10.1038/171737a0. [DOI] [PubMed] [Google Scholar]
  • 7. Weingärtner H., “Understanding Ionic Liquids at the Molecular Level: Facts, Problems, and Controversies,” Angewandte Chemie International Edition 47, no. 4 (2008): 654, 10.1002/anie.200604951. [DOI] [PubMed] [Google Scholar]
  • 8. Plechkova N. V. and Seddon K. R., “Applications of Ionic Liquids in the Chemical Industry,” Chemical Society Reviews 37, no. 1 (2008): 123, 10.1039/B006677J. [DOI] [PubMed] [Google Scholar]
  • 9. Hunt P. A., Ashworth C. R., and Matthews R. P., “Hydrogen Bonding in Ionic Liquids,” Chemical Society Reviews 44, no. 5 (2015): 1257, 10.1039/C4CS00278D. [DOI] [PubMed] [Google Scholar]
  • 10. Fumino K., Reimann S., and Ludwig R., “Probing Molecular Interaction in Ionic Liquids by Low Frequency Spectroscopy: Coulomb Energy, Hydrogen Bonding and Dispersion Forces,” Physical Chemistry Chemical Physics 16, no. 40 (2014): 21903, 10.1039/C4CP01476F. [DOI] [PubMed] [Google Scholar]
  • 11. Knorr A., Stange P., Fumino K., Weinhold F., and Ludwig R., “Spectroscopic Evidence for Clusters of Like‐Charged Ions in Ionic Liquids Stabilized by Cooperative Hydrogen Bonding,” ChemPhysChem 17, no. 4 (2016): 458, 10.1002/cphc.201501134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Menges F. S., Zeng H. J., Kelleher P. J., et al., “Structural Motifs in Cold Ternary Ion Complexes of Hydroxyl‐Functionalized Ionic Liquids: Isolating the Role of Cation–Cation Interactions,” The Journal of Physical Chemistry Letters 9, no. 11 (2018): 2979, 10.1021/acs.jpclett.8b01130. [DOI] [PubMed] [Google Scholar]
  • 13. Niemann T., Neumann J., Stange P., et al., “The Double‐Faced Nature of Hydrogen Bonding in Hydroxy‐Functionalized Ionic Liquids Shown by Neutron Diffraction and Molecular Dynamics Simulations,” Angewandte Chemie International Edition 58, no. 37 (2019): 12887, 10.1002/anie.201904712. [DOI] [PubMed] [Google Scholar]
  • 14. Strate A., Paschek D., and Ludwig R., “Hydrogen‐Bonding Motifs in Hydroxy‐Functionalized Ionic Liquids,” Annual Review of Physical Chemistry 76, no. 1 (2025): 589‐614, 10.1146/annurev-physchem-082423-020307. [DOI] [PubMed] [Google Scholar]
  • 15. Hunger L., Al‐Sheakh L., Zaitsau D. H., et al., “Dissecting Noncovalent Interactions in Carboxyl‐Functionalized Ionic Liquids Exhibiting Double and Single Hydrogens Bonds Between Ions of Like Charge,” Chemistry – A European Journal 28, no. 46 (2022): e202200949, 10.1002/chem.202200949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Hunger L., Al Sheakh L., Fritsch S., et al., “Spectroscopic Evidence for Doubly Hydrogen‐Bonded Cationic Dimers in the Solid, Liquid, and Gaseous Phases of Carboxyl‐Functionalized Ionic Liquids,” The Journal of Physical Chemistry B 128, no. 22 (2024): 5463, 10.1021/acs.jpcb.4c02094. [DOI] [PubMed] [Google Scholar]
  • 17. Harville P. A., Moss O. C., Hassan Y., et al., “Intramolecular Polarization Contributions to the pKa's of Carboxylic Acids Through the Chain Length Dependence of Vibrational Tag‐Shifts in Cryogenically Cooled Pyridinium‐(CH2)n‐COOH (n = 1‐7) Cations,” The Journal of Physical Chemistry A 128, no. 47 (2024): 10159, 10.1021/acs.jpca.4c05241. [DOI] [PubMed] [Google Scholar]
  • 18. Xue Z. and Suhm M. A., “Adding More Weight to a Molecular Recognition Unit: The Low‐Frequency Modes of Carboxylic Acid Dimers,” Molecular Physics 108, no. 17 (2010): 2279, 10.1080/00268976.2010.508600. [DOI] [Google Scholar]
  • 19. Kollipost F., Larsen R. W., Domanskaya A. V., Nörenberg M., and Suhm M. A., “Communication: The Highest Frequency Hydrogen Bond Vibration and an Experimental Value for the Dissociation Energy of Formic Acid Dimer,” The Journal of Chemical Physics 136, no. 15 (2012): 151101, 10.1063/1.4704827. [DOI] [PubMed] [Google Scholar]
  • 20. Braga D., Grepioni F., and Novoa J. J., “Inter‐Anion O–H−···O− Hydrogen Bond Like Interactions: The Breakdown of the Strength–Length Analogy,” Chemical Communications 34, no. 18 (1998): 1959, 10.1039/a804392b. [DOI] [Google Scholar]
  • 21. Steiner T., “Inter‐Anion O–H..O Interactions Are Classical Hydrogen Bonds,” Chemical Communications 35, no. 22 (1999): 2299, 10.1039/a905966k. [DOI] [Google Scholar]
  • 22. Macchi P., Iversen B. B., Sironi A., Chakoumakos B. C., and Larsen F. K., “Interanionic O−H⋅⋅⋅O Interactions: The Charge Density Point of View,” Angewandte Chemie International Edition 39, no. 15 (2000): 2719, 10.1002/1521-3773(20000804)39:15<2719::AID-ANIE2719>3.0.CO;2-8. [DOI] [PubMed] [Google Scholar]
  • 23. Weinhold F. and Klein R. A., “Anti‐Electrostatic Hydrogen Bonds,” Angewandte Chemie International Edition 53, no. 42 (2014): 11214, 10.1002/anie.201405812. [DOI] [PubMed] [Google Scholar]
  • 24. Weinhold F. and Klein R. A., “Addendum: Anti‐Electrostatic Hydrogen Bonds,” Angewandte Chemie International Edition 53, no. 48 (2014): 12992, 10.1002/anie.201410097. [DOI] [PubMed] [Google Scholar]
  • 25. Frenking G. and Caramori G. F., “No Need for a Re‐Examination of the Electrostatic Notation of the Hydrogen Bonding: A Comment,” Angewandte Chemie International Edition 54, no. 9 (2015): 2596, 10.1002/anie.201411374. [DOI] [PubMed] [Google Scholar]
  • 26. Novák M., Marek R., and Foroutan‐Nejad C., “Anti‐Electrostatic CH‐Ion Bonding in Decorated Graphanes,” Chemistry – A European Journal 23, no. 59 (2017): 14931, 10.1002/chem.201703459. [DOI] [PubMed] [Google Scholar]
  • 27. Mata I., Alkorta I., Molins E., and Espinosa E., “Electrostatics at the Origin of the Stability of Phosphate‐Phosphate Complexes Locked by Hydrogen Bonds,” ChemPhysChem 13, no. 6 (2012): 1421, 10.1002/cphc.201200068. [DOI] [PubMed] [Google Scholar]
  • 28. Martín‐Fernández C., Montero‐Campillo M. M., and Alkorta I., “Hydrogen Bonds Are Never of an “Anti‐Electrostatic” Nature: A Brief Tour of a Misleading Nomenclature,” The Journal of Physical Chemistry Letters 15, no. 15 (2024): 4105, 10.1021/acs.jpclett.4c00779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Kim P., Kang R., Carter‐Fenk K., Wilson K. R., Head‐Gordon M., and Ahmed M., “Structural Evidence of Interanionic Hydrogen Bonding in Phosphoric Acid Solutions,” Journal of the American Chemical Society 147, no. 49 (2025): 44916, 10.1021/jacs.5c12699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Zhao C., Burrell G., Torriero A. A. J., et al., “Electrochemistry of Room Temperature Protic Ionic Liquids,” The Journal of Physical Chemistry B 112, no. 23 (2008): 6923, 10.1021/jp711804j. [DOI] [PubMed] [Google Scholar]
  • 31. Kohler F., Atrops H., Kalali H., et al., “Molecular Interactions in Mixtures of Carboxylic Acids with Amines. 1. Melting Curves and Viscosities,” The Journal of Physical Chemistry 85, no. 17 (1981): 2520, 10.1021/j150617a021. [DOI] [Google Scholar]
  • 32. Sheldrick G. M., “A Short History of SHELX,” Acta Crystallographica Section A Foundations of Crystallography 64, no. 1 (2008): 112, 10.1107/S0108767307043930. [DOI] [PubMed] [Google Scholar]
  • 33. Sheldrick G. M., “Crystal Structure Refinement with SHELXL,” Acta Crystallographica Section C Structural Chemistry 71, no. 1 (2015): 3–8, 10.1107/S2053229614024218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Krause L., Herbst‐Irmer R., Sheldrick G. M., and Stalke D., “Comparison of Silver and Molybdenum Microfocus X‐Ray Sources for Single‐Crystal Structure Determination,” Journal of Applied Crystallography 48, no. 1 (2015): 3–10, 10.1107/S1600576714022985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Deposition numbers 2386822 (for [TEA][OOCCOOH]) and 2386827 (for [TEA][OOC(CH2)COOH]) contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service.
  • 36. Brown H. C., McDaniel D. H., and Häfliger O., in Determination of Organic Structures by Physical Methods, eds. Braude E. A. and Nachod F. C. (Academic Press, 1955), 567–662. [Google Scholar]
  • 37. Kwei G. H. and Curl R. F., “Microwave Spectrum of O18 Formic Acid and Structure of Formic Acid,” The Journal of Chemical Physics 32, no. 5 (1960): 1592, 10.1063/1.1730986. [DOI] [Google Scholar]
  • 38. van Eijck B. P. and van Zoeren E., “Microwave Spectra of Partially Deuterated Acetic Acid: Internal Rotation, Potential Energy Function, and Methyl Group Geometry,” Journal of Molecular Spectroscopy 111, no. 1 (1985): 138, 10.1016/0022-2852(85)90076-1. [DOI] [Google Scholar]
  • 39. Steiner T. and Saenger W., “Lengthening of the Covalent O–H Bond in O–H … O Hydrogen Bonds Re‐Examined from Low‐Temperature Neutron Diffraction Data of Organic Compounds,” Acta Crystallographica Section B Structural Science 50, no. 3 (1994): 348, 10.1107/S0108768193011966. [DOI] [Google Scholar]
  • 40. Allen F. H. and Kennard O., “Cambridge Structural Database (CSD),” Chemical Design Automation News 8 (1993): 1. [Google Scholar]
  • 41. Ke C., Li J., Li X., Shao Z., and Yi B., “Protic Ionic Liquids: An Alternative Proton‐Conducting Electrolyte for High Temperature Proton Exchange Membrane Fuel Cells,” RSC Advances 2, no. 24 (2012): 8953, 10.1039/C2RA21378H. [DOI] [Google Scholar]
  • 42. Johansson K. M., Izgorodina E. I., Forsyth M., MacFarlane D. R., and Seddon K. R., “Protic Ionic Liquids Based on the Dimeric and Oligomeric Anions: (AcO)xH(x‐1)‐,” Physical Chemistry Chemical Physics 10, no. 20 (2008): 2972, 10.1039/B801405A. [DOI] [PubMed] [Google Scholar]
  • 43. Brück A., McCoy L. L., and Kilway K. V., “Hydrogen Bonds in Carboxylic Acid−Carboxylate Systems in Solution. 1. in Anhydrous, Aprotic Media,” Organic Letters 2, no. 14 (2000): 2007, 10.1021/ol005776j. [DOI] [PubMed] [Google Scholar]
  • 44. Frisch M. J., Trucks G. W., Schlegel H. B., et al., Gaussian 16, Revision C.01, (Gaussian, Inc., 2016). [Google Scholar]
  • 45. Grimme S., Antony J., Ehrlich S., and Krieg H., “A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT‐D) for the 94 elements H‐Pu,” The Journal of Chemical Physics 132, no. 15 (2010): 154104, 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]
  • 46. Ehrlich S., Moellmann J., Reckien W., Bredow T., and Grimme S., “System‐Dependent Dispersion Coefficients for the DFT‐D3 Treatment of Adsorption Processes on Ionic Surfaces,” ChemPhysChem 12, no. 17 (2011): 3414, 10.1002/cphc.201100521. [DOI] [PubMed] [Google Scholar]
  • 47. Grimme S., Hansen A., Brandenburg J. G., and Bannwarth C., “Dispersion‐Corrected Mean‐Field Electronic Structure Methods,” Chemical Reviews 116, no. 9 (2016): 5105, 10.1021/acs.chemrev.5b00533. [DOI] [PubMed] [Google Scholar]
  • 48. Glendening E. D., Badenhoop J. K., Reed A. E., et al., Natural Bond Orbital 7.0., (Theoretical Chemistry Institute, University of Wisconsin, 2018). [Google Scholar]
  • 49. Weinhold F. and Landis C. R., Valency and Bonding. A Natural Bond Orbital Donor‐Acceptor Perspective, (Cambridge University Press, 2005). [Google Scholar]
  • 50. Weinhold F., Advances in Protein Chemistry, eds. Baldwin R. L. and Baker D. (Academic, 2005). [Google Scholar]
  • 51. Tamura T., Yoshida K., Hachida T., et al., “Physicochemical Properties of Glyme–Li Salt Complexes as a New Family of Room‐Temperature Ionic Liquids,” Chemistry Letters 39, no. 7 (2010): 753, 10.1246/cl.2010.753. [DOI] [Google Scholar]
  • 52. Tamura T., Hachida T., Yoshida K., Tachikawa N., Dokko K., and Watanabe M., “New Glyme–Cyclic Imide Lithium Salt Complexes as Thermally Stable Electrolytes for Lithium Batteries,” Journal of Power Sources 195, no. 18 (2010): 6095, 10.1016/j.jpowsour.2009.11.061. [DOI] [Google Scholar]
  • 53. Philipp J. K., Fumino K., Appelhagen A., Paschek D., and Ludwig R., “The Competition Between Cation‐Anion and Cation‐Triglyme Interaction in Solvate Ionic Liquids Probed by Far Infrared Spectroscopy and Molecular Dynamics Simulations,” ChemPhysChem 26, no. 5 (2025): e202400991, 10.1002/cphc.202400991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Philipp J. K., Kruse L., Paschek D., and Ludwig R., “Structural Transformations within the Solvate Ionic Liquid [Li(Triglyme)][NTf2]: Implications for Self‐Diffusion, Viscosity, and Ionic Conductivity,” The Journal of Physical Chemistry B 129, no. 22 (2025): 5561, 10.1021/acs.jpcb.5c01772. [DOI] [PubMed] [Google Scholar]
  • 55. Ueno K., “Categorizing Molten Salt Complexes as Ionic Liquids and Their Applications to Battery Electrolytes,” Electrochemistry 84, no. 9 (2016): 674, 10.5796/electrochemistry.84.674. [DOI] [Google Scholar]
  • 56. Boys S. F. and Bernardi F., “The Calculation of Small Molecular Interactions by the Differences of Separate Total Energies. Some Procedures with Reduced Errors,” Molecular Physics 19, no. 4 (2006): 553, 10.1080/00268977000101561. [DOI] [Google Scholar]
  • 57. Mentel L. M. and Baerends E. J., “Can the Counterpoise Correction for Basis Set Superposition Effect Be Justified?,” Journal of Chemical Theory and Computation 10, no. 1 (2014): 252, 10.1021/ct400990u. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

A detailed synthesis procedure, corresponding 1H and 13C NMR spectra, experimental details of infrared and X‐ray spectroscopic measurements, as well as information on the theoretical calculations are available in the Supporting Material. The authors have cited additional references within the Supporting Information [56, 57].

CPHC-27-e70581-s001.pdf (402.5KB, pdf)

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article and from the corresponding author upon reasonable request.

Articles from Chemphyschem are provided here courtesy of Wiley

RESOURCES