- Email: [email protected]
Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media
Fluid Phase Equilibria xxx (2015) 1e9
Contents lists available at ScienceDirect
Fluid Phase Equilibria j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d
Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media I.M.B. Felix a, b, *, O. Chiavone-Filho c, S. Mattedi a Chemical Engineering Graduate Program, Polythechnic School, Federal University of Bahia, Rua Aristides Novis, n 2, 40210-630, Salvador, BA, Brazil Rural Federal University of Semi-Arido, RN 233, KM 01, Sítio Nova Esperança II, 59700-000, Caraúbas, RN, Brazil c Chemical Engineering Department, Federal University of Rio Grande do Norte, Av. Senador Salgado Filho, n 3000, NUPEG-NTI, 59066-800, Natal, RN, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 August 2015 Received in revised form 2 October 2015 Accepted 6 November 2015 Available online xxx
Ampicillin is one of the most consumed antibiotics. Due to environmental issues the production of ampicillin has been forbidden by the conventional chemical route. An alternative is the enzymatic route followed by crystallization of the antibiotic. Therefore, this work aims the determination of a series of ampicillin and phenylglycine solubility data in aqueous media using an analytical method. The measurements were carried out at 283.15 and 298.15 K, varying the pH between 3 and 8, and ethanol composition up to 70 wt%. Dissociation constants (pKa's) have also been measured at the studied temperatures and ethanol compositions. It is demonstrated that ideal thermodynamic model with the predetermined pKa's is able to describe satisfactorily solubility profiles. Solubility measurements of ampicillin and phenylglycine have been determined at different conditions of industrial interest. These data may be applied for the evaluation of the crystallization step of ampicillin in the enzymatic synthesis. © 2015 Elsevier B.V. All rights reserved.
Keywords: Ampicillin Solubility Dissociation constant Enzymatic synthesis
1. Introduction The discovery of ampicillin represented a landmark in the era of antibiotics, being still the most commonly prescribed class of antibiotics in human and veterinary medicine. Amoxicillin and ampicillin are in the list of the most consumed antibiotics. Both are semi-synthetic penicillin belonging to the class of b-lactams, whose structure presents a b lactam heteroatomic ring responsible for the antibacterial activity. Brazil, for instance, in 2014 closed its trade balance of medicines containing ampicillin (AMP) and its salts in 3.99 million US$ FOB negative, corresponding to 80.8 tons. This year the importations already reached 2.24 million US$ FOB [1]. Traditionally these antibiotics are produced on industrial scale by chemical synthesis in a complex process requiring extreme conditions of temperature and the use of toxics solvents, generating non-biodegradable wastes [2,3]. Due to environmental laws increasingly restrict, this route is not allowed in many countries. The enzymatic synthesis appears as an alternative, where the
* Corresponding author. Chemical Engineering Graduate Program, Polythechnic School, Federal University of Bahia, Rua Aristides Novis, n 2, 40210-630, Salvador, BA, Brazil. E-mail address: [email protected] (I.M.B. Felix).
antibiotic is produced at moderate temperatures and without the use toxic solvents. However, this process presents a series of side reactions. That is, beyond the antibiotic synthesis, there is hydrolysis of the substrate and antibiotic itself, yielding 6aminopenicillanic acid (6-APA), phenylglycine (PG) and alcohol, as byproducts. These undesirable reactions promote low yield to the process, when compared to the conventional process. When the antibiotic is removed from liquid phase, it becomes unavailable for hydrolysis, making the crystallization a promising technique. Several studies have been conducted to determine the most effective method for obtaining these antibiotics in its pure form right after production. Examples of these techniques are extraction involving volatile solvents or ionic liquids [4,5], liquideliquid extraction [6], and introduction of cosolvents and polymeric resins [7e10]. However, the technique that stands out is the crystallization, as suggested by Youshko et al. [11] for the synthesis of ampicillin. Therefore, the knowledge of the antibiotic solubility is essential for the process. Reports of ampicillin's solubility are scarce in the literature at different operating conditions. Santana et al. [12] obtained AMP and PG data in pure water varying pH and temperature, i.e., 5.5 to 7.5 and 283.15e298.15 K, respectively. Rudolph et al. [13] presented solubility data in pure water varying pH at 298.15 K. They also
http://dx.doi.org/10.1016/j.fluid.2015.11.008 0378-3812/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
2
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
studied AMP and PG solubilities at the isoelectric point (pI) in pure water as function of the temperature and for waterþ1-butanol mixtures at 298.15 K. Youshko et al. [11] determined also the solubility of AMP and PG in water at 298.15 K varying pH. Ampicillin and phenylglycine present in their structure both amino and carboxyl groups, i.e., acidic and basic characters coexist in the same molecule. As usual chemical groups have different affinities for protons. Thus, in solution these molecules can coexist in three different ionic forms, i.e., cationic, zwitterionic and anionic, depending on the pH of the solution. Different ionic forms present different solubilities, being the minimal when the molecules are in their zwitterionic form (no net charge), or so called isoelectric point. The knowledge of the dissociation constant (pKa) is essential for predicting the ionization state of a molecule with respect to pH, since this parameter indicates the tendency of an acid donate proton to a solvent [14]. There are several available methods to determine pKa values, such as potentiometry, conductivity, calorimetry, UVevisible and NMR spectroscopy, mass spectroscopy, liquid chromatography, capillary electrophoresis, and predictions from computational tools [14e17]. Besides the scarce aqueous solubility data involving semisynthetics b-lactam antibiotics and their precursors, the available thermodynamic models are not able to describe properly the behavior found for these systems [13]. This is mainly due to the complexity of the mixtures containing charges, dissociation, organic chains, demanding a more rigorous thermodynamic approach. Nevertheless experimental database and reliable thermodynamic models are indispensable tools for optimizing separation of ampicillin processes. Therefore, in the framework of enzymatic synthesis of b-lactam antibiotics, the aim of this work is to report ampicillin and phenylglycine solubility data in different values of pH (3e8), temperature (283.15 and 298.15 K) and ethanol composition (0e70 wt%). Furthermore, dissociation constants were also determined for each set of conditions, i.e., temperature and solvent composition. Finally,
the ideal thermodynamic model was applied to describe the solubility profiles, demonstrating to be a tool for the estimation of the solubility behavior in the conditions of industrial interest. 2. Materials and methods 2.1. Chemicals PG and AMP were obtained from SigmaeAldrich. PG has 99% of purity grade. The purity of AMP was checked by Total Organic Carbon (TOC) analysis with a 100 ppm solution, prepared gravimetrically, estimating a value of 90%. Furthermore, a differential thermal analysis (DTG) of the AMP sample determined a melting point within 1% of the literature value. NaOH and HCl, both of analytical grade according ACS, were obtained from Neon and Vetec, respectively. Absolute ethanol 99% was provided from Merck. All the reagents were used without further purification. In all measurements deionized water milli-Q were used. Conductivity and pH measurements were used to check the quality of the water, and to avoid carbon dioxide interference nitrogen bubbling was provided during the potentiometric titration. 2.2. Experimental procedure for solubility The jacketed glass cell was charged with a mixture of water and ethanol prepared gravimetrically in the desired concentration. The temperature of the cell was controlled by circulating thermostated water (Tecnal thermostated bath, model TE-184) in the jacket. The temperature stability was determined by the sensor connected to a Metrohm pHmeter, model 827 pH-lab, and is considered to be accurate within 0.1 K. Then, an excess of solute was added to ensure saturation, and the pH was adjusted to the desired value by adding 0.1 molal NaOH or 0.1 molal HCl aqueous solution. Under these conditions, the mixture was kept at constant magnetic stirring for at least three hours when PG was used as solute, and seven hours,
Table 1 Solubility data of phenylglycine in aqueous media at different solvent compositions of ethanol and 298.15 K. pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
3.07 3.52 4.22 4.52 5.04 5.61 6.05 6.49 7.00 7.39 7.94 3.12 3.69 4.12 4.56 5.00 5.59 5.88 6.52 6.93 7.34 7.91 3.13 3.53 4.06 4.51 4.87 5.54
0 0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30 30
33.72 32.59 32.14 31.95 31.82 31.92 31.79 32.02 32.27 32.88 34.05 23.66 23.16 23.04 22.95 22.99 23.06 23.12 23.26 23.35 23.52 24.94 15.94 14.51 13.88 13.77 13.71 13.69
0.08 0.08 0.04 0.12 0.08 0.08 0.14 0.02 0.05 0.12 0.06 0.23 0.20 0.10 0.08 0.10 0.02 0.13 0.19 0.04 0.10 0.07 0.12 0.16 0.10 0.10 0.07 0.07
6.16 6.68 7.02 7.33 7.86 3.12 3.54 4.08 4.49 4.94 5.36 5.92 6.55 7.07 7.47 7.90 3.21 3.61 4.09 4.63 4.93 5.57 6.20 7.05 7.25 7.56 8.08
30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70 70
13.68 13.69 13.71 13.88 14.61 16.26 12.39 11.10 10.93 10.62 10.54 10.34 10.47 10.54 10.78 11.12 12.31 7.62 6.11 5.56 5.27 5.20 5.23 5.24 5.27 5.26 5.20
0.16 0.04 0.09 0.07 0.15 0.12 0.10 0.09 0.04 0.04 0.07 0.08 0.05 0.14 0.09 0.11 0.09 0.06 0.04 0.04 0.01 0.06 0.01 0.07 0.05 0.03 0.02
a
[EtOH]’ ¼ solute free ethanol composition.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
3
Table 2 Solubility data of phenylglycine in aqueous media at different ethanol compositions and 283.15 K. pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
2.80 3.29 3.77 4.48 5.06 5.94 6.31 7.21 7.45 8.17 3.02 3.64 4.14 4.65 5.10 5.60 5.95 6.54 6.92 7.42 7.96 3.20 3.56 4.00 4.71 4.99 4.99
0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30 30
28.89 27.61 26.76 26.04 26.05 26.11 26.22 26.39 26.64 27.75 18.89 17.43 17.22 17.32 17.34 17.42 17.33 17.36 17.55 17.60 17.59 10.80 9.75 9.11 8.76 8.63 8.65
0.18 0.26 0.15 0.16 0.33 0.15 0.05 0.06 0.01 0.05 0.12 0.06 0.08 0.22 0.09 0.14 0.05 0.08 0.22 0.05 0.14 0.14 0.06 0.11 0.10 0.17 0.06
5.23 6.03 6.61 7.34 7.76 3.19 3.58 3.98 4.45 4.84 5.40 6.12 6.78 7.42 8.23 3.36 3.67 4.21 4.66 4.93 5.52 6.07 6.49 7.04 7.56 8.32 8.08
30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70 70 70
8.59 8.59 8.61 8.59 8.93 10.47 8.74 7.24 6.90 6.80 6.75 6.70 6.84 6.71 6.71 7.89 5.83 3.74 3.59 3.71 3.65 3.77 3.65 3.82 3.73 3.79 5.20
0.04 0.05 0.08 0.07 0.09 0.05 0.02 0.16 0.19 0.18 0.01 0.04 0.07 0.13 0.06 0.22 0.05 0.11 0.14 0.20 0.15 0.12 0.07 0.10 0.03 0.15 0.02
a
[EtOH]’ ¼ solute free ethanol composition.
when AMP was used as solute, to ensure complete dissolution and solideliquid equilibrium. After that, a small aliquot (c.a. 3 mL) was withdrawn from the liquid phase using a 0.20 mm filter attached to a syringe to ensure a total absence of solid particles in the sample. Then, the aliquot was diluted in deionized water and analyzed in a
Varian ultraviolet spectrophotometer, model Cary 50, for determination of solute concentration. To confirm that equilibrium, or solubility, has been reached during stirring, the remaining mixture in the cell was again stirred for one hour more, and then a second sample retrieved for analysis. The solubility was then determined
Table 3 Solubility data of ampicillin in aqueous media at different ethanol compositions and 298.15 K. pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
4.58 4.73 4.74 5.56 5.65 6.21 6.84 7.27 7.44 3.45 3.65 3.75 3.97 4.04 4.24 4.45 4.57 4.78 5.36 5.66 5.87 6.46 6.59 6.78 7.13 7.37 7.42
0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
17.70 16.84 17.84 15.57 15.51 16.00 18.93 26.36 39.47 28.46 26.54 25.26 24.62 24.24 23.91 23.45 23.12 22.75 16.59 15.12 14.81 14.75 16.56 17.52 20.52 25.85 29.89
0.01 0.16 0.05 0.11 0.13 0.17 0.15 0.13 0.15 0.27 0.05 0.22 0.15 0.04 0.10 0.06 0.14 0.11 0.08 0.07 0.15 0.06 0.06 0.16 0.15 0.12 0.23
3.47 3.61 3.98 4.58 5.05 5.53 5.92 6.45 7.01 7.38 3.63 4.15 4.57 4.57 5.03 5.39 5.96 6.67 7.02 7.46 3.45 3.82 4.31 4.98 5.71 6.27 6.81 7.02 7.21 7.61
30 30 30 30 30 30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70
29.78 25.45 20.73 17.67 17.04 17.81 18.48 20.87 29.68 47.60 49.68 26.48 20.71 20.39 17.79 17.69 17.98 23.75 31.17 76.39 108.33 55.26 23.41 13.78 10.90 11.35 14.57 17.27 20.34 75.04
0.15 0.15 0.23 0.15 0.15 0.09 0.13 0.13 0.06 0.18 0.34 0.14 0.08 0.10 0.13 0.16 0.13 0.17 0.12 0.15 0.19 0.12 0.11 0.14 0.11 0.04 0.20 0.05 0.15 0.28
a
[EtOH]’ ¼ solute free ethanol composition.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
4
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
Table 4 Solubility data of ampicillin in aqueous media at different ethanol compositions and 283.15 K. pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
3.14 3.32 3.86 4.47 4.87 5.17 6.03 6.26 6.48 6.78 7.08 7.40 7.56 2.88 3.11 3.14 3.66 3.77 3.87 4.26 4.38 5.02 5.30 5.91 6.55 7.06 7.49 7.70 3.11 3.65 3.81 3.87 4.13
0 0 0 0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30
16.94 14.26 12.14 11.92 11.42 11.65 11.69 11.47 11.97 12.12 12.98 14.91 17.09 14.31 11.97 11.79 9.90 9.56 9.68 9.44 9.18 8.69 8.69 8.63 8.69 9.41 11.70 14.49 11.97 16.30 15.24 9.68 12.92
0.12 0.15 0.11 0.27 0.19 0.14 0.16 0.16 0.07 0.12 0.12 0.05 0.19 0.11 0.18 0.17 0.17 0.00 0.13 0.06 0.29 0.08 0.20 0.27 0.19 0.11 0.10 0.19 0.18 0.15 0.05 0.13 0.11
4.15 4.52 4.63 4.96 5.01 5.43 5.74 6.22 6.27 6.37 6.69 6.86 7.30 7.53 3.76 4.25 4.60 4.99 5.61 5.98 6.33 6.95 7.40 3.69 4.21 4.56 4.90 5.28 5.90 6.32 6.81 7.38
30 30 30 30 30 30 30 30 30 30 30 30 30 30 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70
12.84 12.56 11.78 11.72 11.38 11.23 10.85 7.60 7.52 7.59 7.07 6.96 8.60 10.06 32.70 20.28 16.09 13.65 13.46 13.37 13.83 16.40 22.92 54.39 25.68 16.43 12.40 10.69 9.74 9.70 11.05 14.01
0.29 0.16 0.14 0.09 0.02 0.18 0.19 0.10 0.12 0.07 0.08 0.15 0.11 0.01 0.22 0.06 0.04 0.16 0.07 0.10 0.11 0.15 0.09 0.22 0.16 0.07 0.03 0.12 0.06 0.20 0.09 0.13
a
[EtOH]’ ¼ solute free ethanol composition.
Table 5 Dissociation constants of phenylglycine and ampicillin at different ethanol compositions and temperature. [EtOH]’ wt%
Phenylglycine 0 10 30 50 70 Ampicillin 0 10 30 50 70
298.15 K
283.15 K
based on the reproducibility and average of the two successive samples. 2.3. Dissociation constants determination
pKa1
SD
pKa2
SD
pKa1
SD
pKa2
SD
2.15 1.71 2.50 2.97 3.44
0.15 0.40 0.82 0.18 0.24
9.09 9.05 9.06 9.14 9.24
0.32 0.01 0.01 0.03 0.07
1.96 2.11 2.75 3.21 3.28
0.40 0.24 1.06 0.86 0.33
9.60 9.52 9.50 9.59 9.76
0.01 0.03 0.00 0.02 0.05
2.81 2.66 3.26 3.90 4.53
0.22 0.36 0.08 0.03 0.08
7.20 7.13 6.99 6.91 6.94
0.11 0.04 0.01 0.01 0.04
2.26 3.70 3.01 3.78 4.27
0.03 0.56 0.02 0.08 0.18
7.65 7.68 7.42 7.39 7.54
0.01 0.01 0.02 0.07 0.05
The determination of the dissociation constants was based on the potentiometric curves. Stock aqueous solutions of 6 mmolal of PG and 3 mmolal of AMP were used. 15 mL of stock solution was added in the jacketed glass cell under constant magnetic stirring. The temperature was adjusted to the value of interest, which was controlled and monitored by the same set-up used for the solubility measurements. In resume, the solution was first acidified to pH around 2.6 with 0.1 molal HCl aqueous solution. Then it was subjected to alkaline titration dropwise through a Metrohm dosimat, model 776, with the value pH being measured after each addition (approximately 0.01 mL). The titrant used was 1.0 molal NaOH and
Fig. 1. Solubility of AMP and PG in pure water at 298.15 K; line represents the smoothed values of Youshko et al. [11] data.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9 Table 6 pKa data in pure water and 298.15 K from the literature. Reference
Phenylglycine
Ampicillin
pKa1
pKa2
pKa1
pKa2
[12] [13] [14],a
2.08 1.96 e
9.14 9.02 e
[19] [20]
2.20 2.0
9.30 9.0
2.14 2.66 3.24 2.57 2.66 3.97 e e
7.31 7.24 7.44 7.04 7.12 7.54 e e
a
5
the solution to exclude the eventual presence of CO2 dissolved. Furthermore, the NaOH solutions were also prepared right before the titration, to avoid the presence of CO2 in the medium. All measurements were carried out in duplicate. Afterwards, the dissociation constants were obtained from the second derivative of the potentiometric curve with aid of a computational tool. The first inflection point corresponds to the isoelectric point (pI) of the compound. The pH equivalent to the average volume of the first and second inflection points corresponds to pKa2. From the definition, pKa1 is obtained as: 2 * pI-pKa2. 2.4. Mathematical modeling
Different methodologies.
0.5 molal NaOH for PG and AMP, respectively. This process occurred until a stop criterion determined (DpH 0.05 after each alkaline solution drop added). During the titration, N2(g) was bubbled into
For the mathematical modeling of AMP and PG solubility curves, a thermodynamic approach described by Franco and Pessoa Filho [18] was used. The ideal solution approach using pka values was found to be adequate to describe the experimental data. Neglecting
Fig. 2. Phenylglycine solubility for different ethanol compositions at 298.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
6
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
Fig. 3. Phenylglycine solubility for different ethanol compositions at 283.15 K.
the influence of pressure, the authors started from the condition of solideliquid equilibrium, which can be represented by equality between the chemical potential of the solute in both phases. As a consequence of this equilibrium condition, the chemical potential of the solute in the liquid phase may be expressed.
mL2 ¼ m*2 þ RT ln
m 2
m*
consequence of Equation (1), the molality of neutral molecules must remain constant, independently of pH in the liquid phase. Considering that the solubility is the summation of the concentration of all species, i.e., neutral and charged, the molality of neutral molecules can be obtained by multiplying the solubility by the fraction of neutral molecules. As the molality is constant whatever the pH, the following relation may be obtained.
(1)
mL2 is the chemical potential of the solute in the liquid phase, m2 is the molality of the solute in the liquid phase and m*2 is the standard chemical potential of the solute, i.e., it is of a hypothetical solution in which the solute has concentration m* ¼ 1.0 mol kg1. The equilibrium between the solid and liquid phases is established only among electrically neutral molecules in accordance with the principle of electroneutrality of the solid phase. Thus equilibrium condition is valid only for neutral molecules. As a
SðpHÞ ¼
f0 ðpIÞSðpIÞ f0 ðpHÞ
(2)
f0 is the fraction of neutral molecules in the liquid phase and S is the solubility. After several mathematical operations for calculating the fraction of the electrically neutral molecules, the following solubility equation may be obtained.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
SðpHÞ 1 þ 10pHpKa1 ¼ ðpI pHÞ þ log log SðpIÞ 1 þ 10pIpKa1 ! 1 þ 10pHpKa2 þ log 1 þ 10pIpKa2
!
(3)
pka is the dissociation constant determined experimentally. 3. Results and discussions 3.1. Solubility of compounds Solubility measurements of PG and AMP at 283.15 and 298.15 K are reported in Tables 1e4, with their respective standard deviations (SD). Low values of standard deviations between successive samples indicate reproducibility and reliability of the solubility
7
data. Solubility isotherms with respect to pH showed U-shape profiles, which is consistent with the expected behavior for molecules that contain amino and carboxylic acid radicals [12]. This is due to the predominance of electric charges on the molecules when they are in highly alkaline or acid medium, causing repulsion between them and favoring their solubility. The solubility behavior of both compounds was directly proportional to temperature, due to the fact that heat causes an increase in kinetic energy of their molecules, facilitating the interaction solute-solvent and dissolution power. It is easily observed the decrease of PG solubility when ethanol is added up to 70 wt%. This phenomenon is called salting out and can be attributed to the decrease in the dielectric constant of the medium. However, for AMP solubilities it was observed a different behavior. For ethanol concentrations of 30 and 50 wt% a salting in behavior was obtained. Further, the U-shape at 70 wt% of ethanol is
Fig. 4. Ampicillin solubility for different ethanol compositions at 298.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
8
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
narrower and in accordance to the ideal solubility description from measured values of pka, reported at Table 5. In general all curves showed the same order of magnitude in the region where zwitterionic molecules are predominant. The AMP curve with 10 wt% of ethanol at 298.15 K presented a singular behavior among the other solvent concentrations. There was a change in the trend of the solubilities from the isoelectric point. For pH values below the isoelectric point, increasing solubility values were observed. By the other hand, at pH values higher than pI, AMP solubilities decreased significantly. This behavior was also observed at 293.15 K with ethanol content of 30 wt%. Various solubility experiments were performed in similar conditions of interest to confirm this unexpected behavior. Analyses of the solid phase composition may explain the unexpected solubilities values observed. It is noteworthy that it was found no variation of antibiotic absorption profile in the UVevisible spectrum of the liquid samples, indicating that no other substances were formed.
Measurements of solubility in pure water and 298.15 K at various pH were compared with literature values, as shown in Fig. 1. Thus, it is possible to observe that the results presented in this work are in good agreement with the values reported by Youshko et al. [11], validating also the experimental procedure.
3.2. Dissociation constants Values of pKa were determined in the studied temperatures and ethanol compositions. They are reported in Table 5. The addition of ethanol increased significantly the pKa1 values and decreased, generally, the value of pKa2. The temperature presented similar effect, since the increase in this variable has resulted in larger pKa1 values and smaller pKa2 values. For a coherence test, the results obtained in this work were compared with the values reported in the literature for water and 298.15 K (Table 6).
Fig. 5. Ampicillin solubility for different ethanol compositions at 283.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9 Table 7 Global average deviation between experimental and calculated data. [EtOH]’ wt%
0 10 30 50 70
Phenylglycine
Ampicillin
298.15 K
283.15 K
298.15 K
283.15 K
1.0% 0.5% 1.1% 2.2% 4.1%
1.1% 0.9% 1.3% 5.6% 4.3%
9.5% 55.0% 6.3% 4.6% 14.7%
8.9% 22.8% 417.8% 7.5% 9.0%
9
described satisfactorily the solubility data. Although the average deviations obtained in the ideal modeling of AMP are larger, it does not make the model unable to describe the trend of the data. In this case, it is essential to apply an activity coefficient model to consider non idealities of this system. In addition, future studies are needed to elucidate the phenomenon presented in AMP curves at 10 wt% of ethanol and 298.15 K, and also at 30 wt% of ethanol and 283.15 K. Solubility information is useful for the crystallization step of the enzymatic route.
3.3. Mathematical modeling
Acknowledgments
Franco and Pessoa Filho [18] proposed a thermodynamic model based on physical and chemical fundamentals which consider the liquid phase as ideal, relating the protein solubility with pH. Considering structural similarity between proteins and the species studied here, this model was applied to describe PG and AMP solubilities. In this model, a reasonable simplification was made using the solubility at the isoelectric point as an experimental reference. In this work, pI was measured without pH correction, i.e. without addition of alkaline or acid solution at the beginning of the experiment. Figs. 2e5 show the experimental data in the form of solubility versus pH diagram. The experimental data are plotted by error bars, indicating the corresponding experimental uncertainties. In order to evaluate the validity of the model used for predicting the experimental data, it was used the global average deviations for each temperature and ethanol composition studied. These results are shown in Table 7. The systems containing PG were satisfactorily predicted from the model, even considering the ideal behavior of the liquid phase. However, it is possible to notice the deviation of ideality obtained with the increase of ethanol composition. The global average deviation between experimental and calculated data is remarkably larger for AMP, especially in conditions where there is the discontinuous behavior of the curve (10 wt% EtOH and 298.15 K, and 30 wt % EtOH and 283.15 K). It is also noteworthy that the need of a correction of the non-idealities by an activity coefficient model is demonstrated, especially for the system with AMP.
The authors acknowledge the Brazilian agencies CAPES (Coor~o de Aperfeiçoamento de Pessoal de Nível Superior), INCTdenaça ^ncia e Tecnologia de Estudos do EMA (Institutos Nacionais de Cie ~o de Apoio Meio Ambiente) and FUSP (Fundaça a Universidade de ~o Paulo) by indispensable financial support for the realization of Sa this work.
4. Conclusions New PG and AMP aqueous solubility data were presented with reliability, and in the presence of ethanol. The dissociation constants of both solutes were also obtained experimentally at the temperatures and ethanol compositions of the solubility measurements. Although the validity of pka results is difficult because of the large oscillation of the available data in the literature, the measured values demonstrated to be consistent. The ideal model
References [1] Sistema Alice e MDIC, Brazil, available in www.aliceweb.mdic.gov.br, accessed on 7.20.15. pez-Munguía, Enzyme Microb. [2] S. Ospina, E. Barzana, O.T. Ramírez, A. Lo Technol. 19 (1996) 462e469. n, R.L.C. Giordano, Appl. [3] L.R.B. Gonçalves, R. Fern andez-Lafuente, J.M. Guisa Biochem. Biotechnol. 84e86 (2000) 931e945. [4] M. van Berlo, M.T. Gude, L.A.M. van der Wielen, K.ChA.M. Luyben, Ind. Eng. Chem. Res. 36 (1997) 2474e2482. [5] M.S. Manic, V. Najdanovic-Visak, J. Chem. Thermodyn. 44 (2012) 102e106. ndez-Justiz, R. Fern [6] O. Herna andez-Lafuente, M. Terreni, J.M. Guisan, Enzyme Microb. Technol. 59 (1998) 73e79. s, A. Fajardo, J. Mol. Catal. B Enzym 11 (2001) 587e595. [7] A. Illane [8] A. Illanes, S. Anjari, R. Arrieta, C. Aguirre, Appl. Biochem. Biotechnol. 97 (2002) 165e179. [9] C. Aguirre, M. Toledo, V. Medina, A. Illanes, Process Biochem. 38 (2002) 351e360. [10] M. Dutta, N.N. Dutta, K.G. Bhattacharya, Sep. Purif. Technol. 16 (1999) 213e224. [11] M.I. Youshko, L.M. van Langen, E. Vroom, H.M. Moody, F.V. Rantwijk, R.A. Sheldon, V.K. Svedas, J. Mol. Catal. B Enzym 10 (2000) 509e515. [12] M. Santana, M.P.A. Ribeiro, G.A. Leite, R.L.C. Giordano, R.C. Giordano, S. Mattedi, Alche J. 56 (2010) 1578e1583. [13] E.S.J. Rudolph, M. Zomerdijk, K.ChA.M. Luyben, L.A.M. van der Wielen, Fluid Phase Equilib. 158e160 (1999) 903e912. [14] E.Ç. Demiralay, D. Koç, Y.D. Daldal, C. Çakir, J. Pharm. Biomed. Anal. 71 (2012) 139e143. [15] Y. Mrestani, R. Neubert, A. Munk, M. Wiese, J. Chromatogr. A 803 (1998) 273e278. [16] Z. Qiang, C. Adams, Water Res. 38 (2004) 2874e2890. [17] A.V. Pereira, A.A. Garabeli, G.D. Schunemann, P.C. Borck, Quim. Nova 34 (2011) 1656e1660. [18] L.F.M. Franco, P.A. Pessoa Filho, Fluid Phase Equilib. 25 (2011) 242e250. n, Enz. Microb. [19] R. Fern andez-Lafuente, C.M. Rosell, B. Piatkowska, J.M. Guisa Technol. 19 (1996) 9e14. [20] R.V. Ulinj, A.E.M. Janssen, B.D. Moore, P.J. Halling, J. Chem. 7 (2001) 2089e2098.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
Contents lists available at ScienceDirect
Fluid Phase Equilibria j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d
Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media I.M.B. Felix a, b, *, O. Chiavone-Filho c, S. Mattedi a Chemical Engineering Graduate Program, Polythechnic School, Federal University of Bahia, Rua Aristides Novis, n 2, 40210-630, Salvador, BA, Brazil Rural Federal University of Semi-Arido, RN 233, KM 01, Sítio Nova Esperança II, 59700-000, Caraúbas, RN, Brazil c Chemical Engineering Department, Federal University of Rio Grande do Norte, Av. Senador Salgado Filho, n 3000, NUPEG-NTI, 59066-800, Natal, RN, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 August 2015 Received in revised form 2 October 2015 Accepted 6 November 2015 Available online xxx
Ampicillin is one of the most consumed antibiotics. Due to environmental issues the production of ampicillin has been forbidden by the conventional chemical route. An alternative is the enzymatic route followed by crystallization of the antibiotic. Therefore, this work aims the determination of a series of ampicillin and phenylglycine solubility data in aqueous media using an analytical method. The measurements were carried out at 283.15 and 298.15 K, varying the pH between 3 and 8, and ethanol composition up to 70 wt%. Dissociation constants (pKa's) have also been measured at the studied temperatures and ethanol compositions. It is demonstrated that ideal thermodynamic model with the predetermined pKa's is able to describe satisfactorily solubility profiles. Solubility measurements of ampicillin and phenylglycine have been determined at different conditions of industrial interest. These data may be applied for the evaluation of the crystallization step of ampicillin in the enzymatic synthesis. © 2015 Elsevier B.V. All rights reserved.
Keywords: Ampicillin Solubility Dissociation constant Enzymatic synthesis
1. Introduction The discovery of ampicillin represented a landmark in the era of antibiotics, being still the most commonly prescribed class of antibiotics in human and veterinary medicine. Amoxicillin and ampicillin are in the list of the most consumed antibiotics. Both are semi-synthetic penicillin belonging to the class of b-lactams, whose structure presents a b lactam heteroatomic ring responsible for the antibacterial activity. Brazil, for instance, in 2014 closed its trade balance of medicines containing ampicillin (AMP) and its salts in 3.99 million US$ FOB negative, corresponding to 80.8 tons. This year the importations already reached 2.24 million US$ FOB [1]. Traditionally these antibiotics are produced on industrial scale by chemical synthesis in a complex process requiring extreme conditions of temperature and the use of toxics solvents, generating non-biodegradable wastes [2,3]. Due to environmental laws increasingly restrict, this route is not allowed in many countries. The enzymatic synthesis appears as an alternative, where the
* Corresponding author. Chemical Engineering Graduate Program, Polythechnic School, Federal University of Bahia, Rua Aristides Novis, n 2, 40210-630, Salvador, BA, Brazil. E-mail address: [email protected] (I.M.B. Felix).
antibiotic is produced at moderate temperatures and without the use toxic solvents. However, this process presents a series of side reactions. That is, beyond the antibiotic synthesis, there is hydrolysis of the substrate and antibiotic itself, yielding 6aminopenicillanic acid (6-APA), phenylglycine (PG) and alcohol, as byproducts. These undesirable reactions promote low yield to the process, when compared to the conventional process. When the antibiotic is removed from liquid phase, it becomes unavailable for hydrolysis, making the crystallization a promising technique. Several studies have been conducted to determine the most effective method for obtaining these antibiotics in its pure form right after production. Examples of these techniques are extraction involving volatile solvents or ionic liquids [4,5], liquideliquid extraction [6], and introduction of cosolvents and polymeric resins [7e10]. However, the technique that stands out is the crystallization, as suggested by Youshko et al. [11] for the synthesis of ampicillin. Therefore, the knowledge of the antibiotic solubility is essential for the process. Reports of ampicillin's solubility are scarce in the literature at different operating conditions. Santana et al. [12] obtained AMP and PG data in pure water varying pH and temperature, i.e., 5.5 to 7.5 and 283.15e298.15 K, respectively. Rudolph et al. [13] presented solubility data in pure water varying pH at 298.15 K. They also
http://dx.doi.org/10.1016/j.fluid.2015.11.008 0378-3812/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
2
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
studied AMP and PG solubilities at the isoelectric point (pI) in pure water as function of the temperature and for waterþ1-butanol mixtures at 298.15 K. Youshko et al. [11] determined also the solubility of AMP and PG in water at 298.15 K varying pH. Ampicillin and phenylglycine present in their structure both amino and carboxyl groups, i.e., acidic and basic characters coexist in the same molecule. As usual chemical groups have different affinities for protons. Thus, in solution these molecules can coexist in three different ionic forms, i.e., cationic, zwitterionic and anionic, depending on the pH of the solution. Different ionic forms present different solubilities, being the minimal when the molecules are in their zwitterionic form (no net charge), or so called isoelectric point. The knowledge of the dissociation constant (pKa) is essential for predicting the ionization state of a molecule with respect to pH, since this parameter indicates the tendency of an acid donate proton to a solvent [14]. There are several available methods to determine pKa values, such as potentiometry, conductivity, calorimetry, UVevisible and NMR spectroscopy, mass spectroscopy, liquid chromatography, capillary electrophoresis, and predictions from computational tools [14e17]. Besides the scarce aqueous solubility data involving semisynthetics b-lactam antibiotics and their precursors, the available thermodynamic models are not able to describe properly the behavior found for these systems [13]. This is mainly due to the complexity of the mixtures containing charges, dissociation, organic chains, demanding a more rigorous thermodynamic approach. Nevertheless experimental database and reliable thermodynamic models are indispensable tools for optimizing separation of ampicillin processes. Therefore, in the framework of enzymatic synthesis of b-lactam antibiotics, the aim of this work is to report ampicillin and phenylglycine solubility data in different values of pH (3e8), temperature (283.15 and 298.15 K) and ethanol composition (0e70 wt%). Furthermore, dissociation constants were also determined for each set of conditions, i.e., temperature and solvent composition. Finally,
the ideal thermodynamic model was applied to describe the solubility profiles, demonstrating to be a tool for the estimation of the solubility behavior in the conditions of industrial interest. 2. Materials and methods 2.1. Chemicals PG and AMP were obtained from SigmaeAldrich. PG has 99% of purity grade. The purity of AMP was checked by Total Organic Carbon (TOC) analysis with a 100 ppm solution, prepared gravimetrically, estimating a value of 90%. Furthermore, a differential thermal analysis (DTG) of the AMP sample determined a melting point within 1% of the literature value. NaOH and HCl, both of analytical grade according ACS, were obtained from Neon and Vetec, respectively. Absolute ethanol 99% was provided from Merck. All the reagents were used without further purification. In all measurements deionized water milli-Q were used. Conductivity and pH measurements were used to check the quality of the water, and to avoid carbon dioxide interference nitrogen bubbling was provided during the potentiometric titration. 2.2. Experimental procedure for solubility The jacketed glass cell was charged with a mixture of water and ethanol prepared gravimetrically in the desired concentration. The temperature of the cell was controlled by circulating thermostated water (Tecnal thermostated bath, model TE-184) in the jacket. The temperature stability was determined by the sensor connected to a Metrohm pHmeter, model 827 pH-lab, and is considered to be accurate within 0.1 K. Then, an excess of solute was added to ensure saturation, and the pH was adjusted to the desired value by adding 0.1 molal NaOH or 0.1 molal HCl aqueous solution. Under these conditions, the mixture was kept at constant magnetic stirring for at least three hours when PG was used as solute, and seven hours,
Table 1 Solubility data of phenylglycine in aqueous media at different solvent compositions of ethanol and 298.15 K. pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
3.07 3.52 4.22 4.52 5.04 5.61 6.05 6.49 7.00 7.39 7.94 3.12 3.69 4.12 4.56 5.00 5.59 5.88 6.52 6.93 7.34 7.91 3.13 3.53 4.06 4.51 4.87 5.54
0 0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30 30
33.72 32.59 32.14 31.95 31.82 31.92 31.79 32.02 32.27 32.88 34.05 23.66 23.16 23.04 22.95 22.99 23.06 23.12 23.26 23.35 23.52 24.94 15.94 14.51 13.88 13.77 13.71 13.69
0.08 0.08 0.04 0.12 0.08 0.08 0.14 0.02 0.05 0.12 0.06 0.23 0.20 0.10 0.08 0.10 0.02 0.13 0.19 0.04 0.10 0.07 0.12 0.16 0.10 0.10 0.07 0.07
6.16 6.68 7.02 7.33 7.86 3.12 3.54 4.08 4.49 4.94 5.36 5.92 6.55 7.07 7.47 7.90 3.21 3.61 4.09 4.63 4.93 5.57 6.20 7.05 7.25 7.56 8.08
30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70 70
13.68 13.69 13.71 13.88 14.61 16.26 12.39 11.10 10.93 10.62 10.54 10.34 10.47 10.54 10.78 11.12 12.31 7.62 6.11 5.56 5.27 5.20 5.23 5.24 5.27 5.26 5.20
0.16 0.04 0.09 0.07 0.15 0.12 0.10 0.09 0.04 0.04 0.07 0.08 0.05 0.14 0.09 0.11 0.09 0.06 0.04 0.04 0.01 0.06 0.01 0.07 0.05 0.03 0.02
a
[EtOH]’ ¼ solute free ethanol composition.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
3
Table 2 Solubility data of phenylglycine in aqueous media at different ethanol compositions and 283.15 K. pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
pH
[EtOH]’a wt%
[PG] mmolal
SD mmolal
2.80 3.29 3.77 4.48 5.06 5.94 6.31 7.21 7.45 8.17 3.02 3.64 4.14 4.65 5.10 5.60 5.95 6.54 6.92 7.42 7.96 3.20 3.56 4.00 4.71 4.99 4.99
0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30 30
28.89 27.61 26.76 26.04 26.05 26.11 26.22 26.39 26.64 27.75 18.89 17.43 17.22 17.32 17.34 17.42 17.33 17.36 17.55 17.60 17.59 10.80 9.75 9.11 8.76 8.63 8.65
0.18 0.26 0.15 0.16 0.33 0.15 0.05 0.06 0.01 0.05 0.12 0.06 0.08 0.22 0.09 0.14 0.05 0.08 0.22 0.05 0.14 0.14 0.06 0.11 0.10 0.17 0.06
5.23 6.03 6.61 7.34 7.76 3.19 3.58 3.98 4.45 4.84 5.40 6.12 6.78 7.42 8.23 3.36 3.67 4.21 4.66 4.93 5.52 6.07 6.49 7.04 7.56 8.32 8.08
30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70 70 70
8.59 8.59 8.61 8.59 8.93 10.47 8.74 7.24 6.90 6.80 6.75 6.70 6.84 6.71 6.71 7.89 5.83 3.74 3.59 3.71 3.65 3.77 3.65 3.82 3.73 3.79 5.20
0.04 0.05 0.08 0.07 0.09 0.05 0.02 0.16 0.19 0.18 0.01 0.04 0.07 0.13 0.06 0.22 0.05 0.11 0.14 0.20 0.15 0.12 0.07 0.10 0.03 0.15 0.02
a
[EtOH]’ ¼ solute free ethanol composition.
when AMP was used as solute, to ensure complete dissolution and solideliquid equilibrium. After that, a small aliquot (c.a. 3 mL) was withdrawn from the liquid phase using a 0.20 mm filter attached to a syringe to ensure a total absence of solid particles in the sample. Then, the aliquot was diluted in deionized water and analyzed in a
Varian ultraviolet spectrophotometer, model Cary 50, for determination of solute concentration. To confirm that equilibrium, or solubility, has been reached during stirring, the remaining mixture in the cell was again stirred for one hour more, and then a second sample retrieved for analysis. The solubility was then determined
Table 3 Solubility data of ampicillin in aqueous media at different ethanol compositions and 298.15 K. pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
4.58 4.73 4.74 5.56 5.65 6.21 6.84 7.27 7.44 3.45 3.65 3.75 3.97 4.04 4.24 4.45 4.57 4.78 5.36 5.66 5.87 6.46 6.59 6.78 7.13 7.37 7.42
0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10
17.70 16.84 17.84 15.57 15.51 16.00 18.93 26.36 39.47 28.46 26.54 25.26 24.62 24.24 23.91 23.45 23.12 22.75 16.59 15.12 14.81 14.75 16.56 17.52 20.52 25.85 29.89
0.01 0.16 0.05 0.11 0.13 0.17 0.15 0.13 0.15 0.27 0.05 0.22 0.15 0.04 0.10 0.06 0.14 0.11 0.08 0.07 0.15 0.06 0.06 0.16 0.15 0.12 0.23
3.47 3.61 3.98 4.58 5.05 5.53 5.92 6.45 7.01 7.38 3.63 4.15 4.57 4.57 5.03 5.39 5.96 6.67 7.02 7.46 3.45 3.82 4.31 4.98 5.71 6.27 6.81 7.02 7.21 7.61
30 30 30 30 30 30 30 30 30 30 50 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70 70
29.78 25.45 20.73 17.67 17.04 17.81 18.48 20.87 29.68 47.60 49.68 26.48 20.71 20.39 17.79 17.69 17.98 23.75 31.17 76.39 108.33 55.26 23.41 13.78 10.90 11.35 14.57 17.27 20.34 75.04
0.15 0.15 0.23 0.15 0.15 0.09 0.13 0.13 0.06 0.18 0.34 0.14 0.08 0.10 0.13 0.16 0.13 0.17 0.12 0.15 0.19 0.12 0.11 0.14 0.11 0.04 0.20 0.05 0.15 0.28
a
[EtOH]’ ¼ solute free ethanol composition.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
4
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
Table 4 Solubility data of ampicillin in aqueous media at different ethanol compositions and 283.15 K. pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
pH
[EtOH]’a wt%
[AMP] mmolal
SD mmolal
3.14 3.32 3.86 4.47 4.87 5.17 6.03 6.26 6.48 6.78 7.08 7.40 7.56 2.88 3.11 3.14 3.66 3.77 3.87 4.26 4.38 5.02 5.30 5.91 6.55 7.06 7.49 7.70 3.11 3.65 3.81 3.87 4.13
0 0 0 0 0 0 0 0 0 0 0 0 0 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 30 30 30 30 30
16.94 14.26 12.14 11.92 11.42 11.65 11.69 11.47 11.97 12.12 12.98 14.91 17.09 14.31 11.97 11.79 9.90 9.56 9.68 9.44 9.18 8.69 8.69 8.63 8.69 9.41 11.70 14.49 11.97 16.30 15.24 9.68 12.92
0.12 0.15 0.11 0.27 0.19 0.14 0.16 0.16 0.07 0.12 0.12 0.05 0.19 0.11 0.18 0.17 0.17 0.00 0.13 0.06 0.29 0.08 0.20 0.27 0.19 0.11 0.10 0.19 0.18 0.15 0.05 0.13 0.11
4.15 4.52 4.63 4.96 5.01 5.43 5.74 6.22 6.27 6.37 6.69 6.86 7.30 7.53 3.76 4.25 4.60 4.99 5.61 5.98 6.33 6.95 7.40 3.69 4.21 4.56 4.90 5.28 5.90 6.32 6.81 7.38
30 30 30 30 30 30 30 30 30 30 30 30 30 30 50 50 50 50 50 50 50 50 50 70 70 70 70 70 70 70 70 70
12.84 12.56 11.78 11.72 11.38 11.23 10.85 7.60 7.52 7.59 7.07 6.96 8.60 10.06 32.70 20.28 16.09 13.65 13.46 13.37 13.83 16.40 22.92 54.39 25.68 16.43 12.40 10.69 9.74 9.70 11.05 14.01
0.29 0.16 0.14 0.09 0.02 0.18 0.19 0.10 0.12 0.07 0.08 0.15 0.11 0.01 0.22 0.06 0.04 0.16 0.07 0.10 0.11 0.15 0.09 0.22 0.16 0.07 0.03 0.12 0.06 0.20 0.09 0.13
a
[EtOH]’ ¼ solute free ethanol composition.
Table 5 Dissociation constants of phenylglycine and ampicillin at different ethanol compositions and temperature. [EtOH]’ wt%
Phenylglycine 0 10 30 50 70 Ampicillin 0 10 30 50 70
298.15 K
283.15 K
based on the reproducibility and average of the two successive samples. 2.3. Dissociation constants determination
pKa1
SD
pKa2
SD
pKa1
SD
pKa2
SD
2.15 1.71 2.50 2.97 3.44
0.15 0.40 0.82 0.18 0.24
9.09 9.05 9.06 9.14 9.24
0.32 0.01 0.01 0.03 0.07
1.96 2.11 2.75 3.21 3.28
0.40 0.24 1.06 0.86 0.33
9.60 9.52 9.50 9.59 9.76
0.01 0.03 0.00 0.02 0.05
2.81 2.66 3.26 3.90 4.53
0.22 0.36 0.08 0.03 0.08
7.20 7.13 6.99 6.91 6.94
0.11 0.04 0.01 0.01 0.04
2.26 3.70 3.01 3.78 4.27
0.03 0.56 0.02 0.08 0.18
7.65 7.68 7.42 7.39 7.54
0.01 0.01 0.02 0.07 0.05
The determination of the dissociation constants was based on the potentiometric curves. Stock aqueous solutions of 6 mmolal of PG and 3 mmolal of AMP were used. 15 mL of stock solution was added in the jacketed glass cell under constant magnetic stirring. The temperature was adjusted to the value of interest, which was controlled and monitored by the same set-up used for the solubility measurements. In resume, the solution was first acidified to pH around 2.6 with 0.1 molal HCl aqueous solution. Then it was subjected to alkaline titration dropwise through a Metrohm dosimat, model 776, with the value pH being measured after each addition (approximately 0.01 mL). The titrant used was 1.0 molal NaOH and
Fig. 1. Solubility of AMP and PG in pure water at 298.15 K; line represents the smoothed values of Youshko et al. [11] data.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9 Table 6 pKa data in pure water and 298.15 K from the literature. Reference
Phenylglycine
Ampicillin
pKa1
pKa2
pKa1
pKa2
[12] [13] [14],a
2.08 1.96 e
9.14 9.02 e
[19] [20]
2.20 2.0
9.30 9.0
2.14 2.66 3.24 2.57 2.66 3.97 e e
7.31 7.24 7.44 7.04 7.12 7.54 e e
a
5
the solution to exclude the eventual presence of CO2 dissolved. Furthermore, the NaOH solutions were also prepared right before the titration, to avoid the presence of CO2 in the medium. All measurements were carried out in duplicate. Afterwards, the dissociation constants were obtained from the second derivative of the potentiometric curve with aid of a computational tool. The first inflection point corresponds to the isoelectric point (pI) of the compound. The pH equivalent to the average volume of the first and second inflection points corresponds to pKa2. From the definition, pKa1 is obtained as: 2 * pI-pKa2. 2.4. Mathematical modeling
Different methodologies.
0.5 molal NaOH for PG and AMP, respectively. This process occurred until a stop criterion determined (DpH 0.05 after each alkaline solution drop added). During the titration, N2(g) was bubbled into
For the mathematical modeling of AMP and PG solubility curves, a thermodynamic approach described by Franco and Pessoa Filho [18] was used. The ideal solution approach using pka values was found to be adequate to describe the experimental data. Neglecting
Fig. 2. Phenylglycine solubility for different ethanol compositions at 298.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
6
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
Fig. 3. Phenylglycine solubility for different ethanol compositions at 283.15 K.
the influence of pressure, the authors started from the condition of solideliquid equilibrium, which can be represented by equality between the chemical potential of the solute in both phases. As a consequence of this equilibrium condition, the chemical potential of the solute in the liquid phase may be expressed.
mL2 ¼ m*2 þ RT ln
m 2
m*
consequence of Equation (1), the molality of neutral molecules must remain constant, independently of pH in the liquid phase. Considering that the solubility is the summation of the concentration of all species, i.e., neutral and charged, the molality of neutral molecules can be obtained by multiplying the solubility by the fraction of neutral molecules. As the molality is constant whatever the pH, the following relation may be obtained.
(1)
mL2 is the chemical potential of the solute in the liquid phase, m2 is the molality of the solute in the liquid phase and m*2 is the standard chemical potential of the solute, i.e., it is of a hypothetical solution in which the solute has concentration m* ¼ 1.0 mol kg1. The equilibrium between the solid and liquid phases is established only among electrically neutral molecules in accordance with the principle of electroneutrality of the solid phase. Thus equilibrium condition is valid only for neutral molecules. As a
SðpHÞ ¼
f0 ðpIÞSðpIÞ f0 ðpHÞ
(2)
f0 is the fraction of neutral molecules in the liquid phase and S is the solubility. After several mathematical operations for calculating the fraction of the electrically neutral molecules, the following solubility equation may be obtained.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
SðpHÞ 1 þ 10pHpKa1 ¼ ðpI pHÞ þ log log SðpIÞ 1 þ 10pIpKa1 ! 1 þ 10pHpKa2 þ log 1 þ 10pIpKa2
!
(3)
pka is the dissociation constant determined experimentally. 3. Results and discussions 3.1. Solubility of compounds Solubility measurements of PG and AMP at 283.15 and 298.15 K are reported in Tables 1e4, with their respective standard deviations (SD). Low values of standard deviations between successive samples indicate reproducibility and reliability of the solubility
7
data. Solubility isotherms with respect to pH showed U-shape profiles, which is consistent with the expected behavior for molecules that contain amino and carboxylic acid radicals [12]. This is due to the predominance of electric charges on the molecules when they are in highly alkaline or acid medium, causing repulsion between them and favoring their solubility. The solubility behavior of both compounds was directly proportional to temperature, due to the fact that heat causes an increase in kinetic energy of their molecules, facilitating the interaction solute-solvent and dissolution power. It is easily observed the decrease of PG solubility when ethanol is added up to 70 wt%. This phenomenon is called salting out and can be attributed to the decrease in the dielectric constant of the medium. However, for AMP solubilities it was observed a different behavior. For ethanol concentrations of 30 and 50 wt% a salting in behavior was obtained. Further, the U-shape at 70 wt% of ethanol is
Fig. 4. Ampicillin solubility for different ethanol compositions at 298.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
8
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9
narrower and in accordance to the ideal solubility description from measured values of pka, reported at Table 5. In general all curves showed the same order of magnitude in the region where zwitterionic molecules are predominant. The AMP curve with 10 wt% of ethanol at 298.15 K presented a singular behavior among the other solvent concentrations. There was a change in the trend of the solubilities from the isoelectric point. For pH values below the isoelectric point, increasing solubility values were observed. By the other hand, at pH values higher than pI, AMP solubilities decreased significantly. This behavior was also observed at 293.15 K with ethanol content of 30 wt%. Various solubility experiments were performed in similar conditions of interest to confirm this unexpected behavior. Analyses of the solid phase composition may explain the unexpected solubilities values observed. It is noteworthy that it was found no variation of antibiotic absorption profile in the UVevisible spectrum of the liquid samples, indicating that no other substances were formed.
Measurements of solubility in pure water and 298.15 K at various pH were compared with literature values, as shown in Fig. 1. Thus, it is possible to observe that the results presented in this work are in good agreement with the values reported by Youshko et al. [11], validating also the experimental procedure.
3.2. Dissociation constants Values of pKa were determined in the studied temperatures and ethanol compositions. They are reported in Table 5. The addition of ethanol increased significantly the pKa1 values and decreased, generally, the value of pKa2. The temperature presented similar effect, since the increase in this variable has resulted in larger pKa1 values and smaller pKa2 values. For a coherence test, the results obtained in this work were compared with the values reported in the literature for water and 298.15 K (Table 6).
Fig. 5. Ampicillin solubility for different ethanol compositions at 283.15 K.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008
I.M.B. Felix et al. / Fluid Phase Equilibria xxx (2015) 1e9 Table 7 Global average deviation between experimental and calculated data. [EtOH]’ wt%
0 10 30 50 70
Phenylglycine
Ampicillin
298.15 K
283.15 K
298.15 K
283.15 K
1.0% 0.5% 1.1% 2.2% 4.1%
1.1% 0.9% 1.3% 5.6% 4.3%
9.5% 55.0% 6.3% 4.6% 14.7%
8.9% 22.8% 417.8% 7.5% 9.0%
9
described satisfactorily the solubility data. Although the average deviations obtained in the ideal modeling of AMP are larger, it does not make the model unable to describe the trend of the data. In this case, it is essential to apply an activity coefficient model to consider non idealities of this system. In addition, future studies are needed to elucidate the phenomenon presented in AMP curves at 10 wt% of ethanol and 298.15 K, and also at 30 wt% of ethanol and 283.15 K. Solubility information is useful for the crystallization step of the enzymatic route.
3.3. Mathematical modeling
Acknowledgments
Franco and Pessoa Filho [18] proposed a thermodynamic model based on physical and chemical fundamentals which consider the liquid phase as ideal, relating the protein solubility with pH. Considering structural similarity between proteins and the species studied here, this model was applied to describe PG and AMP solubilities. In this model, a reasonable simplification was made using the solubility at the isoelectric point as an experimental reference. In this work, pI was measured without pH correction, i.e. without addition of alkaline or acid solution at the beginning of the experiment. Figs. 2e5 show the experimental data in the form of solubility versus pH diagram. The experimental data are plotted by error bars, indicating the corresponding experimental uncertainties. In order to evaluate the validity of the model used for predicting the experimental data, it was used the global average deviations for each temperature and ethanol composition studied. These results are shown in Table 7. The systems containing PG were satisfactorily predicted from the model, even considering the ideal behavior of the liquid phase. However, it is possible to notice the deviation of ideality obtained with the increase of ethanol composition. The global average deviation between experimental and calculated data is remarkably larger for AMP, especially in conditions where there is the discontinuous behavior of the curve (10 wt% EtOH and 298.15 K, and 30 wt % EtOH and 283.15 K). It is also noteworthy that the need of a correction of the non-idealities by an activity coefficient model is demonstrated, especially for the system with AMP.
The authors acknowledge the Brazilian agencies CAPES (Coor~o de Aperfeiçoamento de Pessoal de Nível Superior), INCTdenaça ^ncia e Tecnologia de Estudos do EMA (Institutos Nacionais de Cie ~o de Apoio Meio Ambiente) and FUSP (Fundaça a Universidade de ~o Paulo) by indispensable financial support for the realization of Sa this work.
4. Conclusions New PG and AMP aqueous solubility data were presented with reliability, and in the presence of ethanol. The dissociation constants of both solutes were also obtained experimentally at the temperatures and ethanol compositions of the solubility measurements. Although the validity of pka results is difficult because of the large oscillation of the available data in the literature, the measured values demonstrated to be consistent. The ideal model
References [1] Sistema Alice e MDIC, Brazil, available in www.aliceweb.mdic.gov.br, accessed on 7.20.15. pez-Munguía, Enzyme Microb. [2] S. Ospina, E. Barzana, O.T. Ramírez, A. Lo Technol. 19 (1996) 462e469. n, R.L.C. Giordano, Appl. [3] L.R.B. Gonçalves, R. Fern andez-Lafuente, J.M. Guisa Biochem. Biotechnol. 84e86 (2000) 931e945. [4] M. van Berlo, M.T. Gude, L.A.M. van der Wielen, K.ChA.M. Luyben, Ind. Eng. Chem. Res. 36 (1997) 2474e2482. [5] M.S. Manic, V. Najdanovic-Visak, J. Chem. Thermodyn. 44 (2012) 102e106. ndez-Justiz, R. Fern [6] O. Herna andez-Lafuente, M. Terreni, J.M. Guisan, Enzyme Microb. Technol. 59 (1998) 73e79. s, A. Fajardo, J. Mol. Catal. B Enzym 11 (2001) 587e595. [7] A. Illane [8] A. Illanes, S. Anjari, R. Arrieta, C. Aguirre, Appl. Biochem. Biotechnol. 97 (2002) 165e179. [9] C. Aguirre, M. Toledo, V. Medina, A. Illanes, Process Biochem. 38 (2002) 351e360. [10] M. Dutta, N.N. Dutta, K.G. Bhattacharya, Sep. Purif. Technol. 16 (1999) 213e224. [11] M.I. Youshko, L.M. van Langen, E. Vroom, H.M. Moody, F.V. Rantwijk, R.A. Sheldon, V.K. Svedas, J. Mol. Catal. B Enzym 10 (2000) 509e515. [12] M. Santana, M.P.A. Ribeiro, G.A. Leite, R.L.C. Giordano, R.C. Giordano, S. Mattedi, Alche J. 56 (2010) 1578e1583. [13] E.S.J. Rudolph, M. Zomerdijk, K.ChA.M. Luyben, L.A.M. van der Wielen, Fluid Phase Equilib. 158e160 (1999) 903e912. [14] E.Ç. Demiralay, D. Koç, Y.D. Daldal, C. Çakir, J. Pharm. Biomed. Anal. 71 (2012) 139e143. [15] Y. Mrestani, R. Neubert, A. Munk, M. Wiese, J. Chromatogr. A 803 (1998) 273e278. [16] Z. Qiang, C. Adams, Water Res. 38 (2004) 2874e2890. [17] A.V. Pereira, A.A. Garabeli, G.D. Schunemann, P.C. Borck, Quim. Nova 34 (2011) 1656e1660. [18] L.F.M. Franco, P.A. Pessoa Filho, Fluid Phase Equilib. 25 (2011) 242e250. n, Enz. Microb. [19] R. Fern andez-Lafuente, C.M. Rosell, B. Piatkowska, J.M. Guisa Technol. 19 (1996) 9e14. [20] R.V. Ulinj, A.E.M. Janssen, B.D. Moore, P.J. Halling, J. Chem. 7 (2001) 2089e2098.
Please cite this article in press as: I.M.B. Felix, et al., Effect of ethanol on the solubility of ampicillin and phenylglycine in aqueous media, Fluid Phase Equilibria (2015), http://dx.doi.org/10.1016/j.fluid.2015.11.008