Hydration Effects in a Guanidine Hydrochloride-Pyruvic Acid Low-Transition-Temperature Mixture: Physicochemical Insights

Abstract

A low-transition-temperature mixture (LTTM) composed of guanidine hydrochloride (GUC) and pyruvic acid (PA) in a 1:2 molar ratio was prepared, and the effect of water on its physicochemical, electrochemical, and structural properties was investigated. To the best of our knowledge, this is the first report of a GUC-PA-based LTTM. Density (ρ), speed of sound (u), and conductivity (κ) were measured over the temperature range of 293.15–323.15 K. The excess molar volume (V ^E) of the pseudo-binary GUC-PA/water system was calculated and correlated using the third-order Redlich-Kister equation. Density decreased with increasing temperature, whereas the thermal expansion coefficient (α _P ) increased. The variation of conductivity with temperature and water composition was correlated using the VFT and Casteel–Amis equations. Water addition significantly enhanced conductivity, reaching a maximum at an LTTM mole fraction of 0.1170 ± 0.002, indicating improved ionic transport upon dilution. Cyclic voltammetry revealed that the neat GUC-PA LTTM and the system containing an LTTM mole fraction of 0.2314 ± 0.002 exhibited comparable electrochemical behavior. The neat LTTM displayed an electrochemical stability window of −0.45 to +1.35 V (vs Ag). Nile Red solvatochromic studies revealed that the neat LTTM provides a highly polar microenvironment than water, and water addition induces only a small hypsochromic shift (∼6 nm). SAXS measurements showed that water progressively weakens intermolecular correlations while preserving the local liquid structure. Density functional theory (DFT) analysis was performed to evaluate the interaction energies and frontier molecular orbitals (FMOs), providing valuable insights into the intermolecular interactions within the GUC-PA. Collectively, these results demonstrate that the GUC-PA LTTM retains its characteristic hydrogen-bonded structure up to an LTTM mole fraction of 0.2298 ± 0.002, beyond which the network progressively weakens. These findings provide fundamental insight into the effect of water on the structure and physicochemical properties of LTTM.

Introduction

Ionic liquids (ILs) have received considerable attention because of their negligible vapor pressure, wide electrochemical windows, and tunable physicochemical properties. − These unique characteristics have enabled their application in catalysis, separation processes, electrochemistry, and biomass processing. , However, their widespread industrial use remains limited due to factors such as high cost, toxicity concerns, and high viscosity. These limitations have motivated the search for alternative solvent systems that retain the desirable properties of ILs while being simpler, less expensive, and environmentally benign. In this context, deep eutectic solvents (DESs) have emerged as promising alternatives − and often referred as the modern counterpart of ILs.

DESs are hydrogen-bonded liquid mixtures formed by mixing of Lewis or Brønsted acids and bases, where strong intermolecular interactions result in a significant depression of the melting point, giving rise to thermodynamically non-ideal eutectic mixtures. , Typically, DESs are prepared by mixing of a hydrogen-bond acceptor (HBA), such as choline chloride, guanidine hydrochloride, or betaine, with a hydrogen-bond donor (HBD), including ethylene glycol or carboxylic acids, in a specified ratio at ambient temperature. Their simple preparation and tunable physicochemical properties have led to applications in CO₂ capture, extraction processes, catalysis, biomass dissolution, electrochemical applications, etc. Despite these advantages, the relatively high viscosity of DESs limits their practical applications. A common strategy to overcome this limitation is the addition of co-solvents such as water, ethanol, or DMSO. , Among these, water is the most preferred choice because of its low cost, low toxicity, and environmental friendliness. Water-added systems were treated as pseudo-binary mixture instead of a ternary mixture. In addition to reducing viscosity, water influences the hydrogen-bonding network, phase behavior, thermophysical properties, and solvation environment of these systems. Therefore, understanding the effect of water is important for designing DES-based systems with suitable physicochemical properties.

Previous studies have shown that the addition of small amounts of water modifies the hydrogen-bond network of DESs without significantly altering their overall nanostructure. For example, Hammond et al. reported that the nanostructure of choline chloride-malic acid DES remains intact up to approximately 50 wt % water, after which the system gradually transforms into a water-dominated phase. Similarly, Chenthamara et al. investigated the effect of water on choline chloride-pyruvic acid (1:1) DES using molecular dynamics simulations. Their results showed that the DES structure remained stable up to 25 wt % water. Above this concentration, structural rearrangement occurred with increased water mobility and converging diffusion coefficients, indicating a gradual transition toward aqueous like behavior. Above 50 wt % water, the mixture becomes fully solvated and exhibited the characteristics of an aqueous soluation.

Water also has a strong influence on the physicochemical properties of DESs, including density, sound velocity, viscosity, and electrical conductivity. − These properties provide valuable information about intermolecular interactions and the structural changes occurring in the liquid. Li et al. showed that the addition of water to choline chloride-monoethanolamine (MEA) DES changes its density, viscosity, and volumetric properties due to progressive restructuring of the hydrogen-bond network. At lower water concentrations, water molecules are incorporated into the DES structure, whereas at higher concentrations water disrupts the HBA-HBD interaction and leads to higher mobility and, eventually, the system begins to exhibit properties similar to those of aqueous solutions. Similarly, Zuo et al. reported that increasing water content decreases the density and viscosity but increases the electrical conductivity of quaternary ammonium salt-(DL)-lactic acid DESs. Their results also showed negative excess molar volumes, indicating strong intermolecular interactions. Furthermore, the DES structure remained intact at low water contents, becomes partially disrupted at intermediate water contents, and gradually transformed into an aqueous-like solution beyond a critical water composition.

The growing interest in DESs has also led to the preparation of many low-melting hydrogen-bonded mixtures. However, many of these systems do not exhibit a distinct melting transition, and reliable solid-liquid equilibrium data are often unavailable. Francisco and coworkers reported that some mixtures of HBA-HBD exhibit a glass transition rather than a clear melting transition. These liquids were designated as Low-Transition-Temperature Mixtures (LTTMs). LTTMs share many physicochemical properties with DESs and have shown potential for biomass dissolution, , CO₂ capture and other applications. Guanidine hydrochloride-based DESs/LTTMs are of particular interest because of their strong hydrogen-bonding ability. Nevertheless, systematic studies addressing their physicochemical properties and, more importantly, the influence of water on their thermophysical behavior, nanoscale structure, and solvation dynamics remain limited. Therefore, a detailed understanding of these systems is essential for their effective application and industrial utilization and represents an important research gap that must be further addressed.

In this work, a low-transition-temperature mixture (LTTM) was prepared from guanidine hydrochloride (GUC) as the HBA and pyruvic acid (PA) as the HBD in a 1:2 molar ratios. To the best of our knowledge, the physicochemical properties of this LTTM and its corresponding aqueous pseudo-binary mixtures have not been reported previously. Since water strongly influences the structure and properties of hydrogen-bonded liquids, identifying the hydration composition at which the physicochemical properties begin to change is important for understanding the transition from an LTTM-rich environment to a water-rich environment. This information is also useful for selecting the appropriate water content for applications such as extraction, separation, and electrolyte systems. To achieve this, the thermal behavior and stability of the neat LTTM and its aqueous mixtures were investigated using differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). Density (ρ), specific conductivity (κ), and electrochemical properties were measured over the temperature range of 293.15–323.15 K, and the electrochemical window was determined at 303.15 K. Nile Red (NR) solvatochromic studies were performed to investigate changes in the local polarity and microenvironment upon water addition, while small-angle X-ray scattering (SAXS) was used to examine the nanoscale structure of the GUC-PA LTTM. In addition, interaction energies and Frontier Molecular Orbital (FMO) analysis were calculated using Density Functional Theory (DFT). The combined results provide a comprehensive understanding of the effect of water on the physicochemical properties, structure, and intermolecular interactions of the GUC-PA LTTM.

Experimental Section

Materials

Guanidine hydrochloride (purity 99%) was purchased from Loba Chemie Private Limited. Pyruvic acid was obtained from Spectrochem. Nile Red (NR) was purchased from Sigma Chemical Co. (Bangalore, India). All chemicals were used as received without further purification. Details of the chemicals and their purities are provided in Table .

1. Description of the Chemicals Used in the Present Work.

chemical name	CAS number	molar mass (g mol–1)	source	purity (mass fraction)
guanidine hydrochloride	50-01-1	95.53	Loba Chemie Private Limited	≥0.99
pyruvic acid	127-17-3	88.06	Spectrochem	≥0.98

^aDeclared by the supplier and used as received.

Preparation of Low-Transition-Temperature Mixture

Prior to the preparation, guanidine hydrochloride was kept in an oven at 80 °C overnight. The LTTM was prepared by mixing guanidine hydrochloride and pyruvic acid in a 1:2 molar ratio under a nitrogen atmosphere. The mixture was heated at 50 °C in an oil bath with continuous stirring until a clear and homogeneous liquid was obtained. The prepared mixture was kept for 48 h in order to conform there was no phase transition occurred. The aqueous pseudo-binary mixtures were prepared by adding deionized water at the desired concentrations to the stock LTTM and stirring at room temperature for 15 min until a homogeneous solution was obtained (Figure S1). All samples were stored in sealed containers prior to further measurements. The formed LTTM was characterized using FT-IR spectroscopy (Figure S2). Scheme illustrates the proposed hydrogen-bonding interactions in the GUC-PA (1:2) low-transition-temperature mixture (LTTM). Dashed lines indicate possible hydrogen-bond interactions between chloride ions, guanidinium, and pyruvic acid molecules that contribute to the formation of the hydrogen-bond network.

1. Low Transition Temperature Mixtures (LTTMs) Preparation.

Thermal Analysis

The phase transition behavior and thermal stability of the prepared LTTM were investigated using differential scanning calorimetry (DSC, Q200, V24.11 Build 124) and thermogravimetric analysis (TGA 2 SF/1100/1824). TGA analysis was performed under a nitrogen atmosphere using a platinum pan over the temperature range of 30–500 °C at a heating rate of 10 °C·min^–1. DSC measurements were carried out under a nitrogen atmosphere over the temperature range of −90 to 100 °C at a heating rate of 5 °C·min^–1.

Density and Sound Velocity

The density and sound velocity of the mixture were concurrently determined using an Anton Paar DSA 5000M vibrating-tube digital density and sound velocity meter at atmospheric pressure. The density meter was calibrated using dry air and Millipore water at atmospheric pressure according to the recommended procedure. The instrument has a measurement range of up to 3000 kg·m^–3 for density and 1000–2000 m·s^–1 for sound velocity. The standard uncertainty with density was ±0.005 kg·m^–3 and in sound velocity with ±0.01 m·s^–1. All measurements were performed over the temperature range 293.15–323.15 K with 5 K increments. Each sample was measured in replicate, and the mean value was reported. The standard uncertainty in experimental values for density and sound velocity was evaluated to be 0.5 kg·m^–3 and 0.5 m·s^–1.

Conductivity

Conductivity measurements were carried out using a Systronics conductivity meter (Model 306). The instrument was calibrated using a 0.1 N KCl standard solution at 25 °C, and the calibration was performed in triplicate. The specific conductance that can be measured using this instrument was in the range 1 × 10^–6 S·m^–1 to 20 S·m^–1. Temperature was controlled using a JULABO circulating thermostatic bath (Germany), which maintains a temperature stability of ±0.03 °C over the operating range of −5 to 100 °C.

Cyclic Voltammetry (CV)

The electrochemical test was carried out using cyclic voltammetry (CV) on a computer-controlled electrochemical workstation (model: Origalys 500), utilizing a conventional three-electrode cell setup: glassy carbon (working electrode), Ag wire (reference electrode) and Pt wire (counter electrode). The glassy carbon, Pt and Ag electrodes were washed with distilled water and acetone. The CV of the respective samples was then examined at varying scan rates, with the potential range between −0.5 V and +1.5 V.

Steady-State UV–Visible Absorption Measurement

UV–visible absorption spectra were recorded using a UV-2600 spectrophotometer (Shimadzu, Japan). All samples were freshly prepared prior to measurement, and the absorption spectra were collected at room temperature using quartz cuvettes with a 1 cm path length.

Small-Angle X-ray Scattering

Small-angle X-ray scattering (SAXS) measurements were performed on a Xeuss 3.0 SAXS/WAXS/USAXS system, beamline equipped with Cu–Kα radiation and a single Dectris Eiger2 SAXS detector positioned directly before the sample to ensure broad Q-coverage. LTTM and its water mole fractions were measured with an exposure time of 15 min, with the sample temperature maintained at 25 °C using a Peltier stage. The sample to detector distance was set to 45 cm for measurements of the neat LTTM and the LTTM/water mixtures.

Computational Details

Density Functional Theory (DFT) calculations were performed using the Gaussian 16 package. All the possible molecular structures were optimized using the B3LYP − functional combined with Grimme’s D3 empirical Dispersion correction and 6–311++G­(d,p) basis set. All the optimized minima were confirmed by the absence of imaginary vibrational frequencies. The free energies were calculated at 298.15 K, and thermal corrections to enthalpy, entropy and zero-point energy were included. Basis set superposition error (BSSE)-corrected interaction energies were calculated using the counterpoise method at the same level of theory.

Results and Discussion

Thermogravimetric Analysis

Thermal stability is one of the major factors determining industrial applications. The thermal stability of the prepared LTTM was determined using thermogravimetric (TGA) and differential thermogravimetric (DTG) analysis. Thermal stability analysis of the mixture also gives an idea about the strength of the hydrogen bonding interaction and the decomposition of the mixture. Generally, LTTM decompositions are two steps, in which HBD will be decomposed first, followed by HBA. Therefore, the strength of hydrogen bonding mainly depends on the thermal stability of HBD. The overall thermal stability was determined from TGA and DTG curves, which are shown in Figure a. T _Peak is the temperature at which the maximum decomposition rate occurs in the sample, as determined from the DTG curve.

1.

(a) TGA curve and (b) DTG curves of GUC-PA and GUC-PA/H₂O mixtures.

The TGA curve depicts a two-step mass loss for the LTTM. The first-stage decomposition for neat LTTM takes place between 74 to 139 °C, which is responsible for the decomposition of pyruvic acid. The second-stage decomposition is between 150 to 310 °C, which corresponds to the decomposition of guanidinium hydrochloride. Table shows the data for the temperature corresponding to the weight loss at 5% (T _5%), 10% (T _10%) and 50% (T _50%) decomposition. According to the data, mixtures with x _LTTM values up to 0.6415 exhibit higher thermal stability, with T _5% values of 74, 76, and 67 °C, compared to mixtures with x _LTTM values ranging from 0.5297 to 0.3170. Additionally, the DTG curve of LTTM Figure b also suggests two-stage mass loss. The mass loss occurs around 139 and 310 °C, which corresponds to the maximum decomposition temperature of PA and GUC, respectively. A small shoulder peak in the first mass loss at 107 °C which corresponds to the elimination of trace moisture content in the mixture. The incorporation of water into the mixture leads to the extensive degradation of thermal stability. Figure a reveals there is a gradual decrease in thermal stability up to x _LTTM = 0.6401; subsequently, a rapid decrease in thermal stability was observed from 0.5789. The decline signifies the disruption of the hydrogen-bonding network.

2. Decomposition Temperature at 5%, 10% and 50% Weight Loss Determined from TGA and Glass Transition Temperature Determined from DSC for the Neat GUC-PA LTTM and Pseudo-Binary Mixtures with Different Water Concentrations.

x LTTM	T 5% (°C)	T 10% (°C)	T 50% (°C)	T g (°C)
1.0000	74	88	164	–54
0.7822	76	91	160	–66
0.6415	67	80	149	–68
0.5297	44	52	109	–79
0.4411	40	46	99	–
0.3737	41	46	91	–
0.3170	39	44	82	–

Differential Scanning Calorimetry

Determining the phase transition behavior of the mixture is crucial for industrial applications. Differential scanning calorimetry (DSC) is employed to determine whether a phase transition is occurring in LTTM. The heating curve of the DSC plot reveals that a glass transition (T _g) as occurring instead of a melting peak. Several studies indicate that a glass transition occurs in these kinds of mixtures. Francisco et al, named these kinds of mixtures “low-transition-temperature mixtures” (LTTMs) instead of DESs due to the absence of a sharp melting peak in DSC. The same group subsequently reported various LTTMs formed by the mixing of HBDs like oxalic acid, lactic acid, malic acid, etc., with HBAs such as choline chloride, amino acids and betaine. The investigation reveals the formation of a hydrogen-bonding network with the components that leads to the significant lowering in the transition temperature of the mixtures, formed from high melting solids. Additionally, Dai et al, reported the presence of a glass transition in DSC analysis of certain NADES.

In Figure for the pure GUC-PA mixture, a T _g peak at −54 °C was observed. The addition of water to the mixture shows a decline in the T _g temperature peak Table . Here, water behaves as a plasticizer, which disrupts the hydrogen-bonding network and increases the mobility of molecules and thereby lowers the glass transition temperature. Due to the limitations of the instrument, we were unable to analyse combinations beyond x _LTTM = 0.5297 (Figure S3).

2.

DSC curves of the neat GUC-PA LTTM and its GUC-PA/H₂O pseudo-binary mixtures.

Density Measurements

The density and speed of sound of the neat GUC-PA LTTM and its aqueous pseudo-binary mixtures were measured over the temperature range of 293.15–323.15 K, and the results are summarized in Table . The neat LTTM exhibited a higher density than the corresponding aqueous mixtures. As shown in Figure a, the density decreased linearly with increasing temperature and water content. The decrease in density with temperature is attributed to thermal expansion, which increases the average intermolecular separation and weakens the hydrogen-bond network. The reduction in density with increasing water content arises from the gradual disruption of the compact hydrogen-bonded network of the GUC-PA LTTM by water molecules, leading to a less densely packed liquid structure. A more pronounced decrease was observed below an LTTM mole fraction x _LTTM = 0.2298, where the density approached that of pure water, indicating the gradual transition from an LTTM-dominated liquid to a water-rich system. The temperature dependence of the density was fitted using eq :

$$\rho (x_{\mathrm{LTTM}},T)=A+\mathrm{{\rm B}}T$$ 1

where ρ is the density in kg·m^–3 and A and B are fitting parameters. The fitted parameters are listed in Table S1 together with the corresponding regression coefficient R ². Within the studied temperature range, the experimental density data showed a good correlation with high R ² values. The speed of sound (u) of the GUC-PA/H₂O mixtures was measured over the temperature range of 293.15–323.15 K, and the results are presented in Table S2 and Figure S4. For the neat LTTM and mixtures with x _LTTM ≥ 0.2298, the speed of sound decreased nearly linearly with increasing temperature. The decrease in the speed of sound with increasing temperature may be attributed to thermal expansion and increased molecular motion, which increase the distance between molecules and slow the propagation of sound waves through the liquid. With increasing water content, the speed of sound gradually approached that of pure water. In addition, mixtures with x _LTTM ≤ 0.1171), exhibited a slight increase in the speed of sound with temperature, consistent with the behavior of pure water (Figure S4). These observations indicate a gradual change in the acoustic properties of the mixtures from those of the neat LTTM to those of water, with the transition occurring below an LTTM mole fraction of approximately 0.2298.

3. Experimental Values of Density ρ (kg·m^–3) of GUC-PA (1:2)/H₂O Binary Mixtures at Atmospheric Pressure P = 0.1 MPa and Temperature Ranges from 293.15 to 323.15 K.

		ρ (kg·m–3)
x LTTM	x H2O	293.15 K	298.15 K	303.15 K	308.15 K	313.15 K	318.15 K	323.15 K
1.0000	0.0000	1308.21	1303.75	1299.15	1294.36	1289.44	1284.26	1277.63
0.7886	0.2114	1296.53	1291.80	1286.96	1281.98	1276.93	1271.94	1267.07
0.6401	0.3599	1283.12	1278.24	1273.28	1268.20	1263.08	1257.99	1252.92
0.5289	0.4711	1270.98	1265.75	1260.47	1255.17	1249.90	1244.67	1239.38
0.4427	0.5573	1257.27	1251.96	1246.63	1241.27	1235.85	1229.97	1223.53
0.3720	0.6280	1244.09	1238.67	1233.28	1227.92	1222.61	1217.37	1212.19
0.3148	0.6852	1228.26	1222.97	1217.68	1212.41	1207.18	1202.02	1196.91
0.2298	0.7702	1200.57	1195.57	1190.52	1185.46	1180.45	1175.48	1170.55
0.1171	0.8829	1134.16	1130.30	1126.27	1122.15	1117.94	1113.47	1108.60
0.0474	0.9526	1065.93	1063.44	1060.76	1057.90	1054.90	1051.78	1046.81
0.0000	1.0000	998.66	997.50	996.11	994.50	992.68	990.64	988.06

3.

Density of the GUC-PA/H₂O pseudo-binary mixtures: (a) variation with temperature and (b) variation with the LTTM mole fraction (x _LTTM). Measurements were carried out at atmospheric pressure (P=0.1 MPa) over the temperature range of 293.15–323.15 K.

The variation of ρ with LTTM mole fraction (x _LTTM) is shown in Figure b. The density value increased with increasing LTTM mole fraction because the neat LTTM possesses a higher density than water. A noticeable change in the slope was observed in the composition range of x _LTTM ≈ 0.2–0.3, with the transition occurring near x_LTTM = 0.2298. The change in slope near x _LTTM = 0.2298 suggest a structural cross over in the mixture which is consistent with the transition observed with the sound velocity measurements. Furthermore, the mixture shows a positive deviation from the ideal behavior, which is shown in Figure a, such deviations are widely studied using excess molar volume studies. ,

4.

(a) Density of the GUC-PA/H₂O pseudo-binary mixtures as a function of the LTTM mole fraction (x _LTTM) at 313.15 K. (b) Variation of the excess molar volume (V ^E/m³·mol^–1) with the LTTM mole fraction over the temperature range of 293.15–323.15 K. The solid curves represent the fits obtained using the Redlich–Kister equation.

Excess Molar Volume (V ^E)

The non-ideality of the pseudo-binary mixture was analyzed by calculating the excess molar volume (V ^E). The V ^E for the pseudo-binary mixture was calculated between the temperature range 293.15–323.15 K, given in Table . V ^E was calculated using eq :

$$V^{\mathrm{E}}=\frac{x_1{M}_1+x_2{M}_2}{{\rho }_{\mathrm{mixture}}}-(\frac{x_1{M}_1}{{\rho }_1}+\frac{x_2{M}_2}{{\rho }_2})$$ 2

where x ₁ and x ₂ are the mole fractions of LTTM and water, respectively, and ρ₁, ρ₂, and ρ_mixture are the densities of pure LTTM, water, and the mixture, respectively. The variation of V ^E with temperature and composition is shown in Figure b. The solid line in the figure represents the values fitted using the Redlich–Kister model (eq ):

$$V^{\mathrm{E}}=x_{\mathrm{LTTM}}{x}_{{\mathrm{H}}_2\mathrm{O}}\sum _{j=0}^k{A}_j{(x_{\mathrm{LTTM}}-x_{{\mathrm{H}}_2\mathrm{O}})}^j$$ 3

where A _j represents the equation coefficient and k is the degree of polynomial expansion, which can be adjusted according to the experimental data to a precise mathematical model. Here we have fitted the model where k = 2. The fitting parameters are given in Table S3, which shows a good fit with the experimental and predicted data, and also shows a good relation with high R ² values.

4. Estimated V ^E (m³·mol^–1) Values of GUC-PA (1:2)/ H₂O Pseudo-Binary Mixtures at Atmospheric Pressure P = 0.1 MPa and Temperature Ranges from 293.15 to 323.15 K.

	V E (10–7 m3·mol–1)
x LTTM	293.15 K	298.15 K	303.15 K	308.15 K	313.15 K	318.15 K	323.15 K
0.7886	–4.317	–3.974	–3.703	–3.463	–3.244	–3.092	–3.097
0.6401	–6.113	–5.682	–5.321	–4.997	–4.705	–4.464	–4.328
0.5289	–7.801	–7.229	–6.733	–6.304	–5.929	–5.606	–5.352
0.4427	–8.527	–7.928	–7.403	–6.935	–6.49	–5.925	–5.266
0.3720	–9.297	–8.667	–8.115	–7.636	–7.213	–6.842	–6.589
0.3148	–9.107	–8.507	–7.973	–7.500	–7.08	–6.712	–6.457
0.2298	–9.238	–8.698	–8.205	–7.762	–7.367	–7.019	–6.775
0.1171	–6.657	–6.300	–5.962	–5.648	–5.354	–5.041	–4.736
0.0474	–3.413	–3.247	–3.088	–2.940	–2.803	–2.681	–2.294

The excess molar volume of the mixture was negative. The magnitude of V ^E relies on the interaction between the solute and solvent particles. The negative V ^E indicates that the mixture exhibits non-ideal behavior; upon mixing, the total volume contracts, resulting in a decrease in available free volume. Furthermore, the negative values of V ^E indicate favorable interactions between the LTTM molecules and water during mixing, arising from strong hydrogen bonding and efficient molecular packing. Such structural changes may also influence mass-transfer processes in extraction and separation applications. Although closer molecular packing can reduce the free volume available for molecular diffusion, the stronger interactions between the LTTM and solute molecules may improve solvation and stabilization of the extracted species. Therefore, the overall extraction performance is expected to depend on the balance between molecular packing, diffusion, and solute-solvent interactions.

The hydrogen-bonding interactions in the mixture reorganize to form new hydrogen-bonds with the water molecules during mixting, leading to a reduction in the free volume of the mixture. The volume contraction is mainly attributed to the strong hydrogen-bonding interactions between the LTTM and water molecules, while the tiny water molecules occupy the vacant spaces between the LTTM molecules. In general, upon the addition of water, the hydrogen-bonding interactions within the LTTM are disrupted. This disruption leads to the formation of new hydrogen-bonding interactions between the water molecules and the Cl^– ion present in the guanidine hydrochloride. Owing to their small size, the water molecule can readily occupy in the intermolecular spaces with in the larger LTTM structure, resulting in more efficient molecular packing and a reduction in the free volume of the mixture. This enhances the packing efficiency of the liquid, leading to volume contraction and, consequently, a negative V ^E. The addition of water increases the magnitude of V ^E up to a specific composition, after which the trend reverses. The increase in the magnitude of V ^E is attributed to the disruptions of the intermolecular interactions between LTTM molecules. Beyond a certain water content, the LTTM molecules become increasingly solvated, leading to a weakening of both LTTM–LTTM and LTTM–water interactions. Consequently, the packing efficiency decreases, resulting in a reduction in the magnitude of V ^E. ,

For better understanding, we can classify the entire region into three (i) water-enriched region x _LTTM < 0.5, (ii) equimolar region x _LTTM = 0.5, and (iii) LTTM-rich region x _LTTM > 0.5. The maximum V ^E value for the LTTM/H₂O mixture was −9.297 × 10^–7, obtained at x _LTTM = 0.3720 and 293.15K, which was a water-enriched area. The solute-solvent interaction was very strong at water rich region than in the LTTM-rich region. This is due to the stronger interaction created between the LTTM molecules and the water molecules, and this also leads to the incorporation of tiny water molecules into the voids created by larger LTTM molecules. In the LTTM-rich region, the LTTM–LTTM interactions were more dominant than those of LTTM-water interactions. Additionally, the high α _P ( Table ) value in this region encourages the incorporation of water molecules.

5. Estimated Thermal Expansion Coefficient Values (α _P ) as a Function of Temperature for GUC-PA (2:1)/H₂O Mixtures.

	α P  (10–4 K–1)
T (K)	x LTTM = 1.000	0.7886	0.6401	0.5289	0.4427	0.372	0.3148	0.2298	0.1171	0.0474
293.15	7.663	7.608	7.865	8.291	8.858	8.553	8.517	8.347	7.471	5.795
298.15	7.693	7.637	7.896	8.325	8.898	8.590	8.553	8.382	7.499	5.812
303.15	7.722	7.667	7.927	8.360	8.938	8.627	8.590	8.418	7.527	5.829
308.15	7.752	7.696	7.958	8.395	8.978	8.664	8.627	8.453	7.555	5.846
313.15	7.782	7.726	7.990	8.430	9.018	8.702	8.664	8.489	7.584	5.863
318.15	7.813	7.756	8.022	8.466	9.059	8.740	8.702	8.525	7.613	5.880
323.15	7.843	7.786	8.055	8.502	9.100	8.779	8.740	8.562	7.642	5.897

Isobaric Thermal Expansion Coefficient (α _P )

The linear decline in density as temperature rises also affects the isobaric thermal expansion coefficient (α _P ). The isobaric thermal expansion coefficient determines the expansion of the volume as the temperature varies at a given pressure. α _P was calculated using eq :

$${\alpha }_P=\frac{1}{V_{\mathrm{m}}}{\left(\frac{\partial V_{\mathrm{m}}}{\partial T}\right)}_{P,x}=-\frac{1}{\rho }{\left(\frac{\partial \rho }{\partial T}\right)}_{P,x}=-\frac{B}{A+BT}$$ 4

where ρ is the density and A and B are the empirical fitting parameters obtained from fitting the density data (Table S1).

The calculated isothermal expansion coefficient values are given in Table . The rise in temperature shows an increment in the α _P values. The lower α _P values for the neat mixture show that the volume expansion was lower than that of the water-added mixtures. The α _P values for the neat mixture were lower when compared to those of the water-added mixtures, which shows strong molecular interaction in mixtures. A gradual increase in the α _P value was observed up to x _{LTTM =} 0.4427; afterward, there was a decline. Initially, the LTTM interaction was dominating. Afterward, the decline indicates the destruction of the structure of the eutectic mixture and the dominance of water behavior in the system.

Specific Conductivity

Specific conductivity is an important property that reflects the mobility of ions in the LTTM and is essential for applications in electrolytes and energy storage systems. The specific conductivities of the neat mixture and its aqueous mixtures were measured over a temperature range of 293.15 to 323.15 K at 5 K intervals. The measured data were summarized in Table . The Figure S5 represents the variation of specific conductivity (κ) versus temperature. The conductivity of the neat LTTM increased from 0.044 to 0.52 S·m^–1 over the investigated temperature range. A similar increase in conductivity with temperature was observed for all LTTM/H₂O mixtures. This increase is attributed to the higher mobility of ions at elevated temperatures. The temperature dependence of κ in Figure a was explained by the Vogel–Fulcher–Tammann (VFT) equation (eq ):

$$\ln \left(\frac{Y}{Y_0}\right)=\ln \left(\frac{A_Y}{Y^0}\right)+\frac{B_Y}{T-T_{0,Y}}$$ 5

where Y = κ (S·m^–1) and Y ⁰ = 1 S·m^–1. The fitted parameters are summarized in Table S4.

6. Specific Conductivities (κ) of GUC-PA (1:2)/H₂O Binary Mixtures at Atmospheric Pressure P = 0.1 MPa and Temperature Ranges from 293.15 to 323.15 K.

	κ (S·m–1)
T (K)	x LTTM = 1.000	0.7822	0.6415	0.5297	0.4411	0.4411	0.317	0.2291	0.1170	0.0473
293.15	0.04	0.21	0.52	0.97	1.66	2.26	2.99	4.67	6.88	6.31
298.15	0.07	0.31	0.73	1.30	2.12	2.84	3.67	5.54	7.91	7.17
303.15	0.12	0.43	0.95	1.62	2.54	3.41	4.39	6.16	8.80	7.90
308.15	0.18	0.59	1.23	1.98	3.12	4.04	5.04	7.26	9.72	8.65
313.15	0.26	0.79	1.54	2.43	3.62	4.62	5.71	8.09	10.80	9.52
318.15	0.38	0.99	1.86	2.86	4.16	5.21	6.41	8.91	11.69	10.21
323.15	0.52	1.28	2.28	3.40	4.81	5.94	7.20	9.86	12.76	11.04

5.

(a) Temperature dependence of the experimental conductivity (κ; symbols) of the GUC-PA/H₂O mixtures at different LTTM mole fractions. The solid lines represent the VFT fits (eq ). (b) Specific conductivity (κ; symbols) as a function of the LTTM mole fraction. The solid lines represent the Casteel–Amis fits (eq ).

The variation of κ with the composition of the mixture was explained by the modified Casteel–Amis equation (eq ):

$$\kappa ={\kappa }_{\max }{{\left(\frac{x_1}{x_1^{\max }}\right)}_{\kappa }}^a\exp [-a_{\kappa }(\frac{x_1}{x_1^{\max }}-1)+b_{\kappa }{(x_1-x_1^{\max })}^2+c_{\kappa }{(x_1-x_1^{\max })}^3]$$ 6

where κ_max represents the maximum conductivity, x ₁ is the corresponding point of maximum conductivity, and a _κ, b _κ, and c _κ are empirical fitting parameters without any physical significance. The fitted parameters are listed in Table S5. Based on the presented data in Figure b, initially, the magnitude of κ for the mixtures rises and then declines with the increase in the LTTM molar concentration. This variation was observed due to the collective effects of ion concentration, ion interactions, including solvation or association and the viscosity. Lin et al, suggested that prior, to reaching x ₁ the molecules present in the LTTM are completely dissociated and get hydrated with the water molecule. The magnitude of κ of the system increases significantly with increasing LTTM mole fraction (x ₁) until x ₁ is reached. After this x ₁ , the interaction between the LTTM ions dominates over that with the water molecules. This leads to a reduction in charge carriers and thereby a reduction in the κ. The higher specific conductivity observed in the optimum hydration region indicates enhanced ion transport in the GUC-PA/H₂O mixtures. Such behavior is desirable for electrolyte systems used in electrochemical energy-storage devices and electrochemical sensors, where efficient ion transport is important. However, the performance of an electrolyte also depends on factors such as electrochemical stability, compatibility with electrode materials, and long-term cycling stability, which were not investigated in the present study. Therefore, further studies are needed to evaluate the suitability of the hydrated GUC-PA LTTM for these applications.

Nile Red Solvatochromism and Microenvironmental Polarity in GUC-PA (1:2) and GUC-PA/H₂O Mixtures

Polarity describes the ability of a solvent to stabilize charged, dipolar, or nonpolar species. Nile Red (NR) is a well-established solvatochromic probe for evaluating the polarity of deep eutectic solvents, ,, ionic liquids and complex systems. Figure a shows the UV–vis absorption spectra of NR in water and in GUC-PA mixtures with different water mole fractions. In pure water (LTTM mole fraction = 0), Nile Red exhibits a broad absorption band centered at approximately 583 nm, whereas in the neat GUC-PA mixture (LTTM mole fraction = 1.0000), the absorption maximum shifts to approximately 645 nm. The normalized absorption spectra shown in Figure b reveal that increasing the water mole fraction causes a small hypsochromic shift of about 6 nm. At the same time, the calculated E _NR values increase by approximately 0.4 kcal mol^–1. According to the Nile Red solvatochromic scale, these changes indicate that the polarity experienced by the probe decreases slightly with increasing water content. The relatively small spectral shift suggests that the local environment surrounding Nile Red changes only modestly over the investigated composition range. Overall, these results indicate that the addition of water modifies the local solvation environment of the probe while largely preserving the characteristic microenvironment of the GUC-PA mixtures.

6.

(a) Absorption spectra of Nile Red in GUC-PA (1:2) mixture (LTTMs) and GUC-PA/H₂O mixtures of varying LTTM mole fractions. (b) Corresponding normalized absorption spectra highlighting the shift in the absorption maximum with increasing water content. The inset in (b) shows the small hypsochromic shift of the long-wavelength absorption band upon water addition, (c) SAXS profiles of the neat GUC-PA (1:2) mixture and the GUC-PA (1:2) mixture containing a LTTM mole fraction of 0.3173.

The UV–vis absorption spectrum of Nile Red in the neat GUC-PA (1:2) mixture exhibits two vibrational bands centered at approximately 600 and 645 nm (Figure a). The long-wavelength absorption maximum at 645 nm is considerably red-shifted compared to that in water (583 nm), indicating that Nile Red experiences a different and more polar local environment in the GUC-PA mixture. The appearance of two absorption bands further suggests that the probe is distributed between at least two distinct solvation environments. A similar behavior has been reported for the choline chloride/malonic acid (1:1) deep eutectic solvent, where the two bands were assigned to a more polar (MP) and a less polar (LP) microenvironment. By analogy, the band at 645 nm may be attributed to Nile Red molecules associated with the hydrogen-bonded GUC-PA network, whereas the band near 600 nm likely arises from molecules located in a relatively less polar environment. These observations indicate that the GUC-PA mixture contains microdomains with different local polarities, reflecting its heterogeneous solvation structure.

To compare the polarity of the GUC-PA LTTM with other systems, the E _NR values were compared with those reported for representative deep eutectic solvents in the literature (Table ). The UV–vis absorption spectrum of Nile Red is recorded in DES, and the longest-wavelength absorption maximum (λ_max) is used to calculate the transition energy (E _NR, kcal·mol^–1), which is commonly used to describe the local polarity of deep eutectic solvents, using eq : ,,

$$E_{\mathrm{NR}}=28591/{\lambda }_{\max }$$ 7

7. Comparison of the Nile Red Polarity Values (E _NR) of the GUC-PA LTTM with Representative Deep Eutectic Solvents (DESs) Reported in the Literature; The Table Lists the HBA, HBD, Water Content, Molar Ratio, and Corresponding E _NR Values.

HBA	HBD	water content	molar ratio	Nile Red polarity scale E NR (kcal/mol)	ref
GUC	PA	–	(1:2)	44.32	This study
ChCl	citric acid	20 wt %	(1:1)	44.53
ChCl	lactic acid	20 wt %	(1:1)	44.88
ChCl	tataric acid	–	(1:1)	44.56
ChCl	oxalic acid	–	(1:1)	44.67
ChCl	malonic acid	–	(1:1)	44.67
citric acid	glucose	–	(1:1)	47.81
glucose	tartaric acid	–	(1:1)	47.73
ChCl	citric acid	5.2 wt %	(1:1)	47.94
ChCl	citric acid	5.9 wt %	(1:1)	47.89
ChCl	citric acid	–	(1:1)	47.97
ChCl	citric acid	4.8 wt %	(1:1)	48.3
ChCl	citric acid	–	(2:1)	48.3
ChCl	tartaric acid	–	(2:1)	48.13
Citric acid	sucrose	–	(1:1)	49.72
ChCl	sucrose	–	(1:1)	49.72
ChCl	xylose	2.2 wt %	(2:1)	50.78
ChCl	xylose	3.8 wt %	(2:1)	50.69
ChCl	xylose	6 wt %	(2:1)	50.16
menthol	oleic acid	–	(1:1)	53.8
menthol	lauric acid	–	(1:1)	50.14
thymol	oleic acid	–	(1:1)	51.16
menthol	thymol	0.157%	(1:1)	50.90
thymol	borneol	–	(1:1)	50.84

^aChCl refers to choline chloride.

By measuring changes in λ_max of Nile Red relative to the value it exhibits in a reference solvent, it is possible to calculate the relative polarity of the solvent of interest. As the solvent polarity increases, the absorption maximum shifts to longer wavelengths and the E _NR value decreases. Thus, lower E _NR values indicate higher polarity. The GUC-PA LTTM shows an E _NR value of 44.32 kcal·mol^–1, which is among the lowest values reported for DESs. Since lower E _NR values correspond to a more polar environment around the Nile Red probe, this result indicates that the GUC-PA LTTM provides a highly polar local environment. Its polarity is comparable to that of choline chloride-organic acid DESs and higher than that of many sugar-based and hydrophobic DESs reported in the literature. ,

X-ray Scattering Experiments

SAXS measurements were carried out to examine the nanoscale structure of the GUC-PA (1:2) LTTM and to evaluate the effect of water addition (Figure c). Both the neat GUC-PA (1:2) LTTM and the sample with an LTTM mole fraction of 0.3173 show a broad scattering maximum at q = 1.5–2.0 Å^–1, which is typical of disordered liquids and has been reported for several deep eutectic solvents. , No sharp Bragg peaks are observed, indicating that the system remains lack of long-range crystalline ordering. The broad peak corresponds to a real-space correlation distance of approximately 3–4 Å (d = 2π/q), consistent with short-range intermolecular packing and hydrogen-bonding interactions between the constituent molecules. In addition to the main scattering peak, the neat GUC-PA LTTM shows a weak shoulder at low q ≈ 0.4 Å^–1, corresponding to a correlation distance of about 16 Å. Similar low-q features have been reported in deep eutectic solvents and are attributed to intermediate-range structural heterogeneity arising from hydrogen-bonded domains. Upon the addition of water (x _LTTM = 0.3173), this shoulder becomes much weaker, while the main scattering peak shows a slight decrease in intensity and becomes slightly broader. These changes suggest that water weakens the intermediate- and short-range intermolecular correlations by partially disrupting the hydrogen-bond network, consistent with the structural changes reported for other water-containing eutectic systems. However, the overall scattering profile changes only modestly, indicating that the basic nanoscale organization of the GUC-PA LTTM is largely preserved.

Cyclic Voltammetry

Cyclic voltammetry is a powerful technique for evaluating the electrochemical stability and redox processes of DESs. Before envisioning any practical electrochemical application, it is important to determine the electrochemical potential window (EPW) of the solvents. EPW is defined as the voltage gap between the cathodic limiting current and anodic limiting current. The electrochemical stability of DESs depends on several factors, including the intrinsic stability of the hydrogen-bond acceptor (HBA), the hydrogen-bond donor (HBD), and the strength of the intermolecular interactions between them, such as hydrogen bonding and van der Waals interactions.

The effect of scan rate on the electrochemical behavior of the GUC:PA mixture and its aqueous mixtures was investigated at 298.15 K, and the corresponding cyclic voltammograms are shown in Figure . No distinct redox peaks were observed for the neat GUC:PA mixture or its aqueous mixtures (x _LTTM = 0.4427 and 0.2314), indicating the absence of significant electrochemical reactions within the investigated potential range. As shown in Figure , the cathodic and anodic current densities increased with increasing scan rate, while the cathodic and anodic peak potentials shifted slightly towards negative and positive potentials, respectively. The neat GUC:PA mixture and its aqueous mixtures exhibited similar voltammetric profiles, suggesting that the characteristic electrochemical behavior of the LTTM is largely preserved upon water mole fraction addition.

7.

Cyclic voltammograms of (a) neat GUC-PA (1:2) mixture (1:2) (mole fraction = 1.0000), (b) GUC-PA/H₂0 mixture (mole fraction = 0.4427), (c) GUC-PA/H₂O mixture (mole fraction = 0.2314), and (d) pure pyruvic acid with different scan rates.

Compared with the GUC:PA LTTM, pure pyruvic acid displayed a markedly different voltammetric response. At higher scan rates, the voltammograms showed a nearly rectangular profile, indicating predominantly capacitive behavior. The cathodic and anodic current densities of pyruvic acid also increased with increasing scan rate. The neat GUC:PA mixture showed cathodic and anodic decomposition potentials of −0.45 and +1.35 V (vs Ag), respectively, with an electrochemical potential window (EPW) of 1.80 V. These results indicate that the GUC:PA mixture possesses good electrochemical stability and retains its electrochemical behavior even after the addition of water.

DFT Analysis

The optimized structure exhibited a well-defined hydrogen-bonding between the O–H of the pyruvic acid and the Cl^–of guanidine hydrochloride Figure . LTTM formation leads to the elongation of the O–H bond in PA from 0.97 Å to 1.03 Å, which indicates the weakening of the covalent O–H bond due to the charge redistribution associated with the hydrogen bond formation between the Cl^– ion. The C–O bond length of isolated PA decreases from 1.35 Å to 1.30 Å in LTTM, which indicates enhanced electron delocalization and the partial double-bond character after forming the mixture. Furthermore, the optimized structure revealed the presence of hydrogen bonding interactions between the N–H hydrogen atoms of the guanidinium cation and the carbonyl oxygen atoms of PA, which contribute to the stabilization of the LTTM. All these collective variations indicate the formation of a hydrogen bonding interaction between the HBA and HBD molecules.

8.

Optimized geometries of (a) GUC, (b) PA, and (c) GUC-PA LTTM with the lowest interaction energies at B3LYP/6–311++G­(d,p) level of theory.

To explore different hydrogen bonding possibilities, different possible orientations of PA molecule around GUC were optimized. The thermodynamic stability of each possible structure was evaluated using the free energy equation for complex formation (eq ):

$$\mathrm{\Delta }G=G_{\mathrm{LTTM}}-(G_{\mathrm{GUC}}+G_{\mathrm{PA}})$$ 8

BSSE corrections were incorporated using the counterpoise method to obtain more accurate interaction energies.

The various optimized configurations of GUC-PA LTTM were given in Figure S6. The BSSE corrected interaction energy varies from −10.35 to −27.32 kcal·mol^–1, along with Cartesian coordinates of the optimized structures presented in Table S6 and Table S7, respectively, which indicates all the structures were thermodynamically stable; among them, the most stable structure (with interaction energy −27.32 kcal·mol^–1) was taken for the further analysis. The negative interaction energy indicates that the LTTM is thermodynamically favorable, and the main driving force for the formation of LTTM is a strong hydrogen bonding interaction taking place between the HBA and HBD molecules. The O–H···Cl^– hydrogen bonding interactions and N–H···O–C hydrogen bonding interactions primarily contribute to the interaction energy, along with other electrostatic interactions with the guanidinium cation and carboxylate anion. Additionally, Frontier Molecular Orbital (FMO) analysis was employed to characterize the electronic properties of GUC-PA LTTM. The energy of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) are −6.18 eV and −2.92 eV, respectively. The HOMO-LUMO energy gap is 3.26 eV; this moderate energy gap suggests that the system is stable under ambient conditions (Figure S7).

Conclusion

In this study, a new low-transition-temperature mixture (LTTM), GUC-PA (1:2), was prepared and the effect of water mole fraction on its physicochemical properties was systematically evaluated. To the best of our knowledge, this is the first report of a guanidine hydrochloride-pyruvic acid based LTTM. Density (ρ), speed of sound (u), ionic conductivity, excess molar volume (V ^E), and thermal expansion coefficient (α _P ) were determined for the neat mixture and its pseudo-binary mixtures with different water mole fractions over the temperature range of 293.15–323.15 K. The results showed that the properties of the system are strongly influenced by both temperature and water content. The V ^E values were successfully correlated using the third-order Redlich-Kister equation, and negative V ^E values were observed over the entire composition and temperature range studied. The density decreased with increasing temperature, whereas the thermal expansion coefficient increased. The specific conductivity increased with the water concentration and temperature. The variation of κ as a function of temperature and composition was correlated using the VFT equation and the Casteel–Amis equation. Thermal studies revealed a two-stage decomposition process for both the neat LTTM and its pseudo-binary mixtures. The neat GUC mixture exhibited an electrochemical stability window of approximately −0.45 to +1.35 V (vs Ag), indicating good electrochemical stability. The combined thermophysical results identified a critical hydration composition at which the physicochemical properties begin to change significantly. This information provides useful guidance for selecting the water content to balance the characteristic properties of the LTTM with improved transport properties for applications such as extraction, separation, and electrolyte systems.

Nile Red solvatochromic measurements showed that the GUC-PA mixture is more polar (λ_abs = 643 nm) than water (λ_abs = 583 nm), while the small blue shift observed with increasing water content suggests only moderate changes in the local solvation environment. SAXS analysis revealed that water addition weakens intermediate-range structural correlations and perturbs the hydrogen-bond network of the GUC-PA LTTM, while preserving its nanoscale organization. The combined physicochemical, electrochemical, and solvatochromic results show that the characteristic properties of the GUC-PA LTTM are retained up to a LTTM mole fraction of 0.2314 ± 0.002. Further addition of water gradually shifts the system toward a water-dominated state with properties increasingly similar to those of aqueous solutions. In addition, DFT analysis revealed the intermolecular interactions in the pure LTTM. The interaction energy calculated for the most stable configuration was −27.32 kcal·mol^–1, and the HOMO–LUMO energy gap was 3.26 eV. The findings of this study show that controlling the water content is important for tuning the properties of the GUC-PA LTTM. These findings may be useful for future studies on metal dissolution, CO₂ capture, and electrolyte systems. Further studies are needed to evaluate the performance of the GUC-PA LTTM in these applications.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsphyschemau.6c00114.

• Image of the prepared GUC-PA LTTM, FT-IR and DSC characterization; sound velocity and fitting parameters for ρ, V ^E, the VFT equation, and the Casteel–Amis equation; possible optimized configuration and Cartesian coordinates for LTTM; and HOMO-LUMO orbitals of the most stable configuration (PDF)

S.N.S.K.: Writingoriginal draft, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. B.R.D.: Writingoriginal draft, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. R.L.G.: Writingreview and editing, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization.

The authors declare no competing financial interest.

Published as part of ACS Physical Chemistry Au special issue “Visions for the Future of Physical Chemistry 2026”.