Solubility of DNP-amino acids and their partitioning in biodegradable ATPS: Experimental and ePC-SAFT modeling
Graphical abstract
Introduction
Understanding the behavior of organic molecules in aqueous solutions is of importance for process design with special attention on separation processes. Separating a compound of interest between the two aqueous phases has been considered in the literature as an appropriate approach to evaluate interactions in water systems [1], [2], [3], [4], [5], [6], [7]. Willingness to unravel the phenomena which are behind the partitioning of very complex and large molecules such as proteins was the reason for introducing to research smaller compounds that share the same functional groups as macromolecules [1], [2], [3], [6]. Thus, dinitrophenyl derivatives of amino acids (DNP-AA) have been selected to study interactions of the amino acid side-chain groups in Aqueous Two-Phase Systems (ATPS) and the properties of ATPS phases [1], [5], [6], [8], [9], [10], [11].
In the literature, DNP-AA or their sodium salts were partitioned in ATPS which have usually contained two different polymers [6], [10], [11], or a combination of one polymer with an inorganic salt [5], [10], [12]. Partition coefficients (K) of reasonable magnitude measured for DNP-AA in polymer – inorganic salt systems have shown that these media can be very appropriate for efficient separation of ionizable organic components such as DNP-AA. However, high amounts of inorganic salt (e.g. phosphates, sulfates) needed to form two immiscible phases with polymer (e.g. polyethylene glycol – PEG), might be very inconvenient for the environment if such ATPS were applied at the industrial scale. For this reason, in recent studies, the replacement of inorganic salts by biodegradable organic compounds (e.g. citrates, tartrates) has been proposed [9], [13]. Such modification turned out to be a successful solution as very similar K values for DNP-AA could be obtained in the alternative PEG – sodium citrate ATPS if comparing with former results in PEG – sodium sulfate system [8], [9].
The partitioning in PEG – biodegradable salt ATPS can be still optimized by the change of several factors, e.g. the two-phase composition (tie-line length - TLL) or the molecular weight of PEG used [13]. However, measuring the partitioning at these moderating conditions is usually time- and cost-intensive. Thermodynamic predictive models that can describe the interactions in aqueous systems can reduce the number of experiments needed to determine K values. Thus, in literature at least several models have been proposed to correlate and describe the partitioning in PEG – organic salt ATPS, e.g. Pitzer model [14], extended NRTL [15], modified UNIQUAC [15]. Also predictive Statistical Association Fluid Theory (SAFT)-based models showed the ability to model the partitioning in both PEG – inorganic salt and PEG – organic salt ATPS [16], [17]. The quantitative modeling of K for five water-soluble vitamins in PEG (PEG 4000, PEG 6000) – organic sodium salt (Na3Citrate, Na2Tartrate) ATPS was achieved with electrolyte PC-SAFT (ePC-SAFT) [17] through activity coefficients. The success of ePC-SAFT with implemented Debye-Hückel term for this application relied on the precise description of interactions in ATPS, as well as on considering the presence of different ionic species at pH of ATPS. To the best of our knowledge, there has not been any work yet in which K for DNP-AA in PEG – organic salt ATPS were modeled.
In this work, ePC-SAFT has been at first used for predicting the liquid densities and solubilities of four N-(2,4-dinitrophenyl) amino acids, namely DNP-glycine, DNP-alanine, DNP-valine, DNP-leucine. The predictions were performed at T=298.15 – 318.15 K, p=1 bar (density), and T=298.15 K and p=1 bar (solubility). The ePC-SAFT pure-component parameters and the binary interaction parameters (kij) for DNP-AA were estimated with the joint -parameter method (J-PM). The J-PM that has been introduced in this work specifically for DNP-AA, origins from a strategy published for proteins [45]. The conception adopted in original method from literature was also used by Greinert et al. [3] for modeling solutions of 1,3-bisphospho-D-glycerate [44]. ePC-SAFT parameters for m-dinitrobenzene (DNB) fitted to vapor pressures and densities (data from the literature [18], [19]) have been joint with parameters for aliphatic amino acids (AA) established in the previous study [20]. Experimental data required for validation of density and solubility predictions of DNP-AA were not available in literature, so they needed to be measured in this work. The measurements were performed at the same conditions of pressure and temperature as the predictions. Since the solubility of ionizable DNP-AA strongly depends on the pH of a solution, solubility data were obtained at different pH values. Then, the intrinsic solubilities (neutral DNP-AA) were determined from pH-solubility profiles estimated using the Henderson-Hasselbalch (H-H) equation and dissociation constants (pKa). Afterward, these solubilities were compared with ePC-SAFT modeling results. Solubilities measured at higher pH (pH≈6) in which DNP-AA are solely present as negatively charged species were used to access kij parameters between water and DNP-AA−. The method of fitting Ka with pH-dependent solubilities is established [21].
The second part of this work relied on ePC-SAFT modeling of DNP-AA partitioning in ATPS formed by PEG (PEG 1500, PEG 4000, PEG 6000) and organic salt (Na3Citrate, K3Citrate, KNaTartrate). The reliability of the modeling results was determined using experimental K values from the literature [9], [13]. For the partition-coefficient modeling, the pure-component parameters for phase forming components: PEG, Na+, K+, and anions (Tartrate2−, Citrate3−) were obtained from formerly reported studies [17], [22], [23], [24]. Almost all binary interaction parameters between ATPS phase formers could be found in the literature as well [17], [22], [24], [25]. In one case, the kij between PEG and K+ was fitted to tie-line data from the literature [26]. The interactions between K+ and salt anions (Tartrate2−, Citrate3−) also needed to be estimated in this study. This was done by fitting kij parameters to experimental osmotic coefficients [26], [51]. The kij between DNP-AA− and ATPS phase formers (Na+, K+, PEG) were fitted just to one experimental K point for each DNP-AA− partitioned in ATPS: PEG 6000 – Na3Citrate, and PEG 6000 – K3Citrate [9], [13].
Section snippets
Dissociation equilibrium of DNP-AA and solubility
DNP-AA act as acids, so they will be present as neutral molecular species at low pH but dissociated into anion and proton at high pH. At the pH of ATPS studied in this work (pH≈8) DNP-AA considered in the present study are solely present as negatively charged species (charge q=-1). Studies on the dissociation behavior of DNP-AA in ATPS has to be preceded by the analysis of equilibria expressions for ionizable components in their water solutions [27].
A solution of low soluble weak monoprotic
Materials and experimental methods
The experimental data obtained in this work are collected in tables in the Supplementary Information.
pH-solubility profiles of DNP-AA
Experimental pH-dependent aqueous solubilities of DNP-AA were measured at T=298.15 K, p=1 bar (SI, Table S1). These results are presented in Fig. 4 together with the pH-solubility curves calculated using Henderson-Hasselbalch (H-H) solubility equation (SI, S4) and dissociation constants pKa (Table 1). The pKa values were estimated (DNP-valine, DNP-leucine) in this work or taken from the literature (DNP-alanine, DNP-glycine) [32]. The method of fitting Ka with pH-dependent solubilities has been
Conclusions
Different phenomena rule the solubilization and partitioning of amino-acid derivatives in aqueous systems. Typically, the effort related to understanding the behavior during these processes requires numerous experiments. In this work, the pH-dependent DNP-AA solubilities (T=298.15 K, p=1 bar) and liquid densities of solutions containing DNP-AA (T=298.15 – 318.15 K, p=1 bar) were determined experimentally. Intrinsic DNP-AA solubilities, liquid densities of DNP-AA solutions, and partitioning of
CRediT authorship contribution statement
Kamila Wysoczanska: Investigation, Validation, Formal analysis, Writing - original draft, Conceptualization, Methodology. Birte Nierhauve: Investigation, Validation, Formal analysis. Gabriele Sadowski: Project administration, Supervision. Eugénia A. Macedo: Project administration, Supervision. Christoph Held: Conceptualization, Methodology, Project administration, Writing - review & editing, Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work is a result of Project “AIProcMat@N2020 - Advanced Industrial Processes and Materials for a Sustainable Northern Region of Portugal 2020”, with the reference NORTE-01-0145-FEDER-000006, supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the Portugal 2020 Partnership Agreement, through the European Regional Development Fund (ERDF); Associate Laboratory LSRE-LCM - UID/EQU/50020/2019 - funded by national funds through FCT/MCTES (PIDDAC). K. Wysoczanska and E.
References (63)
- et al.
Effect of salt additives on protein partition in polyethylene glycol–sodium sulfate aqueous two-phase systems
Biochimica et Biophysica Acta (BBA) - Proteins and Proteomics
(2013) - et al.
Analysis of amino acid–water interactions by partitioning in aqueous two-phase systems. I—amino acids with non-polar side-chains
J. Chromatogr. A
(2013) - et al.
Phase separation in aqueous poly (ethylene glycol)-(NH4) 2SO4 systems and some physicochemical properties of the phases
J. Colloid Interface Sci.
(1990) - et al.
Measurement of relative hydrophobicity of amino acid side-chains by partition in an aqueous two-phase polymeric system: hydrophobicity scale for non-polar and ionogenic side-chains
J. Chromatogr. A
(1982) - et al.
Cation effect on the (PEG 8000+ sodium sulfate) and (PEG 8000+ magnesium sulfate) aqueous two-phase system: Relative hydrophobicity of the equilibrium phases
J. Chem. Thermodyn.
(2015) - et al.
Effect of different organic salts on amino acids partition behaviour in PEG-salt ATPS
Fluid Phase Equilib.
(2018) - et al.
Effect of molecular weight of polyethylene glycol on the partitioning of DNP-amino acids: PEG (4000, 6000) with sodium citrate at 298.15 K
Fluid Phase Equilib.
(2016) - et al.
Experimental determination and thermodynamic modeling of phase equilibrium and protein partitioning in aqueous two-phase systems containing biodegradable salts
J. Chem. Thermodyn.
(2013) - et al.
Modeling aqueous electrolyte solutions: part 1. Fully dissociated electrolytes
Fluid Phase Equilib.
(2008) - et al.
Predicting the high concentration co-solvent influence on the reaction equilibria of the ADH-catalyzed reduction of acetophenone
J. Chem. Thermodyn.
(2019)