- Email: [email protected]
The synergistic effect between hydrophobic and electrostatic interactions in the uptake of amino acids by strongly acidic cation-exchange resins
Journal of Chromatography A, 1108 (2006) 43–49
The synergistic effect between hydrophobic and electrostatic interactions in the uptake of amino acids by strongly acidic cation-exchange resins Shaoling Cheng, Husheng Yan ∗ , Changqing Zhao Key Laboratory of Functional Polymer Materials, Ministry of Education, Institute of Polymer Chemistry, Nankai University, Tianjin 300071, China Received 7 October 2005; received in revised form 15 December 2005; accepted 23 December 2005 Available online 26 January 2006
Abstract The goal of this work was to investigate the synergistic effect between the electrostatic and hydrophobic interactions upon the uptake of organic ions with hydrophobic moieties by ion-exchange resins with hydrophobic matrixes. The uptake of neutral amino acids by a macroporous polystyrene-based strongly acidic cation-exchange resin (D001) and two strongly acidic cation-exchange resins (poly(2-acrylamido-2-methyl propanesulfonic acid) and poly(vinylsulfonic acid)) with much less hydrophobic matrixes essentially follow an ion exchange stoichiometry. However, the thermodynamic parameters of the uptakes indicate that besides electrostatic interaction, hydrophobic interaction also contributes to the affinity of the amino acids with hydrophobic side chains for D001. No detectable uptake capacities for the amino acids by D001AM, which was obtained by amidation of the sulfonic acid groups of D001, can be determined. Thus, it is deduced that the hydrophobic interaction alone contributes little to the uptake of these amino acids by D001, of which hydrophobicity is the same with or lower than that of D001AM. These results indicate that synergistic effect exists between the electrostatic and hydrophobic interactions when the two interactions exist in a chelate manner and the hydrophobic interaction contributes to the uptake even if the hydrophobic interaction is so weak that it contributes little to the uptake when it acts alone. © 2006 Elsevier B.V. All rights reserved. Keywords: Synergistic effect; Ion exchange; Hydrophobic interaction; Ion-exchange resin; Amino acid
1. Introduction Utilization of synthetic ion exchange resins is the most prominent method employed in modern industrial processes to accomplish the task of separation and purification. Ionexchange resins are also used in analytical chemistry, e.g. as packings for ion exchange chromatography. Ion exchange processes of organic ions, such as biomolecules, drugs, etc., with ion-exchange resins have been widely used in their separations, purifications, removals from plasma or wastewater, analytical or clinical determinations, and in controlled drug deliveries [1–8]. It is considerable interest to find out the nature of the interactions in the ion exchange processes. The knowledge concerning the binding of organic ions by ion-exchange resins will aid in the
∗
Corresponding author. Tel.: +86 22 23503935; fax: +86 22 23503935. E-mail address: [email protected] (H. Yan).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2005.12.101
proper development and selection of ion-exchange resins, and selection of conditions of the ion exchange processes for these applications mentioned above. The electrostatic interaction between the organic ions and the oppositely charged ions on the resin is undoubtedly the prominent interaction. Besides the electrostatic interaction, hydrophobic interaction also exists between the hydrophobic moieties of the organic ions and the hydrophobic matrixes of the ion exchange resins (e.g. poly(styrene-divinylbenzene)-based ion-exchange resins) or other hydrophobic groups on the resin in some ion exchange processes [2,7,9–16]. However, the uptake of some hydrophobic organic ions by poly(styrene-divinylbenzene)-based ion-exchange resins follows an ion exchange stoichiometry and the hydrophobic interaction contributes little to the maximum uptake capacity, while the ion exchange selectivity is greatly controlled by the hydrophobic interaction in some ion exchange processes [13–16]. The apparent inconsistent effects of the hydrophobic interaction on uptake capacity and selectivity
44
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
have not yet been satisfactorily explained in the literature. For example, through studying the uptake capacity and selectivity coefficient, respectively, of Leu, Ile, Val, Ala and Phe by a polystyrene-based strongly acidic cation-exchange resin, two inconsistent conclusions were reached in a paper [13]: the uptake process depends only the ionic fraction of amino acid cations and the occurrence of nonionic adsorption can be discounted; the hydrophobic interactions of the amino acid side chain with the organic backbone of the resin play an important role in determining the affinity of amino acid cations for the resin’s functional group, without, however, significantly contributing to the maximum uptake capacity. In order to better understand these apparent inconsistent conclusions, we have investigated theoretically and experimentally the cooperativity between ion-exchanges and hydrophobic interactions in uptake of neutral amino acids by strongly acidic cation exchange resins. Based on our results, these inconsistent conclusions mentioned above can be satisfactorily explained. 2. Experimental section and thermodynamic analysis 2.1. Materials D001, a macroporous poly(styrene-8-wt.%-divinylbenzene)based sulfonic cation-exchange resin, was purchased from Hecheng Co. Ltd (Tianjin, China). The cation exchange capacity was determined to be 5.04 mmol/g dry resin. 2-Acrylamido2-methyl propanesulfonic acid (AMPS) was purchased from San Wing International Ltd (Shanghai, China). Sodium vinylsulfonate (VSS) (25% aqueous solution) was purchased from Shuang Jian Chemical Co. (Guangzhou, China). Acrylamide was purchased from Tianjin Medical Co. (Tianjin, China). N,N -Methylenebisacrylamide was purchased from Tianjin Institute of Chemical Reagent (Tianjin, China). Span-60 was purchased from Shanghai Chemical Reagent Co. (Shanghai, China). Glycine (Gly), leucine (Leu) and phenylalanine (Phe) were purchased from Shanghai Changjiang Biochemical Plant (Shanghai, China). 2.2. Synthesis of crosslinked PAMPS resin PAMPS, an AMPS-N,N -methylenebisacrylamide copolymer, was prepared as following. Twenty-five grams of AMPS, 6.25 g of N,N -methylenebisacrylamide and 0.25 g of K2 S2 O8 were dissolved in 125 ml of water. After the pH was adjusted to 7–8 with 20% NaOH, the aqueous solution was suspended in an oil phase which was composed of 311.5 ml of dichloroenthane, 138.5 ml of petroleum ether and 5.5 g of span-60 in a three-neck round-bottom flask with a mechanical stirrer and a nitrogen gas inlet. The mixture was stirred while bubbling nitrogen gas over for 30 min, making sure that there is no oxygen existing in the solution. Then a few drops of N,N,N ,N tetramethylethlenediamine was added into the flask to initiate the polymerization and the mixture was stirred for further 2 h at room temperature. The resulting copolymer was washed thoroughly with ethanol, water, 2N HCl and water. The cation
exchange capacity of the resulting resin, PAMPS, was determined to be 4.90 mmol/g dry resin. 2.3. Synthesis of crosslinked PVSS resin PVSS, a VSS-N,N -methylenebisacrylamide copolymer, was prepared as following. An aqueous solution which was composed of 4 ml of VSS, 3 g of acrylamide, 2.16 g of N,N methylenebisacrylamide, 0.096 g of K2 S2 O8 and 36 ml of water was suspended in an oil phase which was composed of 195 ml of chlorobenzene, 10 ml of toluene and 2.6 g of span-60 in a three-neck round-bottom flask with a mechanical stirrer and a nitrogen gas inlet. The mixture was stirred at while bubbling nitrogen gas over for 30 min, making sure that the oxygen was replaced completely. Then a few drops of N,N,N ,N -tetramethylethlenediamine was added into the mixture to initiate the polymerization and the mixture was stirred for further 2 h at room temperature. The resulting copolymer was washed thoroughly with ethanol, water, 2N HCl and water. The cation exchange capacity of the resulting resin, PVSS, was determined to be 2.50 mmol/g dry resin. 2.4. Synthesis of polystyrenesulfonamide resin (D001AM) from D001 D001AM, sulfonamide of D001, was prepared as following. Sulfonyl chloride of D001 was prepared according to a literature method [17]. Briefly, D001 was treated with pyridine in ethanol at room temperature over night to convert the sulfonic acid to pyridine salt. Then the resulting resin was treated with SOCl2 at 0 ◦ C to form sulfonyl chloride resin. The transformation was confirmed by the disappearance of a peak at 3442 cm−1 and the appearance of peaks at 1374 and 1173 cm−1 of the IR spectrum of the resin after the reaction. The sulfonyl chloride resin (3 g) was mixed with 5 ml of aqueous dimethylamine solution (containing 35% of dimethylamine) at 0 ◦ C, and then another 20 ml of aqueous dimethylamine solution was added dropwise into the mixture at 0 ◦ C while stirring. After all the dimethylamine solution was added, the mixture was stirred at room temperature over night. The resulting resin was washing thoroughly with water, and dried under vacuum. No peak at ∼3440 cm−1 appeared and two strong absorptions at 1337 and 1161 cm−1 appeared in the IR spectrum of the resulting resin, confirming that sulfonyl chloride groups have been converted to sulfonamide groups. Nitrogen content was determined by elementary analysis to be 4.76 mmol/g. Cation exchange capacity was determined to be 0.01 mmol/g dry resin. 2.5. Determination of cation exchange capacity A dry resin sample (∼0.1 g) was suspended in 25 ml of standard aqueous NaOH solution (∼0.1N) in an Erlenmeyer flask and the mixture was shaken for 8 h. 10 ml of the supernatant was taken from the flask and was titrated with standard aqueous HCl solution (∼0.1N) with methyl orange as the indicator. The cation exchange capacity (Qie in mmol/g) of the resin was
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
45
calculated by Qie =
25NNaOH − 2.5VHCl × NHCl W
where NNaOH and NHCl are the concentrations of the standard aqueous NaOH solution and the standard aqueous HCl solution, respectively. VHCl is the volume (ml) of the standard aqueous HCl used in the titration, W is the sample weight (g). 2.6. Equilibrium uptake of amino acids by resins Resin samples (∼0.2 g) were added into Erlenmeyer flasks containing 25 ml of amino acid solutions with certain concentrations. The flasks were sealed and shaken for 20 h in a constanttemperature bath. Then the concentration of amino acids in the supernatant was determined by photometry using Cu2+ as the chromogenic reagent [18]. The uptake capacity was calculated from the initial and equilibrium concentrations, the volume of the solution used and the sample weight. For determination of the uptake equilibrium constant, the adsorption temperature was 25 ◦ C. For the van’t Hoff plot (see Section 3), 25, 30, 40 and 50 ◦ C were chosen. The amino acid concentration ranges were 0–1.2 mM for D001, 0–2.0 mM for PAMPS and 0–3.0 mM for PVSS, where the adsorption isotherms are linear. 2.7. Thermodynamic analysis of uptake of organic ions by ion-exchange resin Poly(styrene-divinylbenzene)-based ion-exchange resins are mostly common ion exchange resins in practical applications. It has been noticed that hydrophobic interactions contributed to the adsorption of some hydrophobic organic ions by poly(styrenedivinylbenzene)-based ion-exchange resins. We first consider the electrostatic and hydrophobic interactions independently. For the electrostatic interaction alone in the ion exchange process, the free energy change at the standard state is given by G◦ el = H ◦ el − T S ◦ el
(1)
We divide the entropy change S◦ el into two components, i.e. negative entropy change caused by the loss of translational and rotational entropy when the solute was immobilized onto the resin by the electrostatic interaction (ion exchange), S◦ elt+r , and the other entropy changes, S◦ elo . Then Eq. (1) can be rewritten as G◦ el = H ◦ el − T S ◦ elo − T S ◦ elt+r
(2)
For the hydrophobic interaction without ion exchange, similar equation can be written as: G◦ h = H ◦ h − T S ◦ ho − T S ◦ ht+r
Fig. 1. Scheme of chelate manner of the electrostatic and hydrophobic interactions.
the loss of translational and rotational entropy for the electrostatic interaction which occurs alone is the same with that for the hydrophobic interaction which occurs alone. G◦ overal = H ◦ el + H ◦ h − T S ◦ elo − T S ◦ ho − T S ◦ t+r − T ∆S ◦ br
(4)
S◦ br is the entropy change caused by restriction of the free rotation of the single bonds between the ion group and the hydrophobic moiety of both the resin and the organic ion upon the chelate binding. The value of S◦ br should be negative. Eq. (4) can be re-written as: G◦ overal = (H ◦ el − T S ◦ elo − T S ◦ t+r ) + (H ◦ h − T S ◦ ho − T S ◦ t+r ) + T S ◦ t+r − T S ◦ br = G◦ el + G◦ h + T (S ◦ t+r − S ◦ br )
(5)
It can be concluded from Eq. (5) that the overall free energy change (G◦ overal ) of the chelate binding would be more negative than the sum of the free energy changes (G◦ el and G◦ h ) of the two interactions occurred independently when (S◦ t+r − S◦ br ) < 0, or when the loss of the translational and rotational entropy is greater than the loss of the bond rotation entropy upon binding. We can say there is cooperativity or synergistic effect between the two interactions in this case. 3. Experimental results and discussion Neutral amino acids such as Gly, Ala, Leu and Phe in neutral aqueous solution should exist in their zwitterions. Their isoelectric pH values are 5.97, 6.00, 5.98 and 5.48, respectively [19]. Thus, uptake of neutral amino acids by cation-exchange resins in aqueous medium can be expressed in the following equilibrium:
(3)
If the electrostatic and hydrophobic interactions occur simultaneously in a chelate manner, as shown in Fig. 1, the binding free energy change (G◦ overal ) can be described by Eq. (4) upon the assumption of (1) the natures of the two interactions do not influence each other (in other words, the binding sites have good complementary), and (2) S◦ ht+r = S◦ elt+r = S◦ t+r , i.e.,
(6) The equilibrium constant, K6 , can be expressed as: K6 =
qe ce (qmax − qe )
(7)
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
46
qmax and the correlation coefficients (r2 ) by the fitting. It can be seen from Table 1 that the maximum uptake capacity, qmax , for all amino acids tested approaches the total exchange capacity of the resin, indicating a stoichiometric ion exchange process similar to the result of Dye et al. [13]. The free energy change (G◦ 6 ) of the equilibrium expressed by Eq. (6) can be obtained from the data in Table 1 according to Eq. (9) and was listed in Table 2. G◦ = −RT ln K
(9)
The enthalpy change of the uptake can be determined from van’t Hoff equation [20]: ln K = − Fig. 2. Isotherms of uptake of amino acids by D001 at 25 ◦ C.
H ◦ + k0 RT
(10)
where k0 is a constant. In the low-concentration region where the isotherm was linear, equilibrium constant (K) is correlated to adsorption affinity (qe /ce ),
Table 1 Langmuir adsorption parameters and correlation coefficients (T = 298 K) Amino acid
K6 (mM−1 )
qmax (mmol/g)
r2
qe = ψK ce
D001
Gly Ala Leu Phe
0.333 0.350 0.656 1.10
5.02 5.01 4.96 4.84
0.9997 0.9979 0.9976 0.9967
PAMPS
Gly Phe
0.153 0.107
4.63 4.76
0.9987 0.9987
where ψ is a constant. From Eqs. (10) and (11), we can get the following equation: H ◦ qe =− ln +c (12) ce RT
PVSS
Gly Phe
0.145 0.105
2.38 2.47
0.9984 0.9990
Ion-exchange resin
where qe is the uptake capacity under the equilibrium concentration, ce . qmax is the maximum uptake capacity. Eq. (7) can be re-written as: qe = qmax
K 6 ce 1 + K 6 ce
(11)
where c is a constant. A plot of ln(qe /ce ) versus 1/T yields a straight line with a slope of −H◦ /R, from which H◦ 6 can be obtained, as shown in Table 2. The uptake equilibrium expressed by Eq. (6) involves two equilibriums: ion exchange and protonation, as shown by Eqs. (13) and (14), respectively.
(8) (13)
Eq. (8) has the same form with Langmuir adsorption equation. Fig. 2 shows the uptake isotherms of Gly, Ala, Leu and Phe by D001. These data fit Eq. (8) (Langmuir equation) well, as shown by the curves in Fig. 2. Table 1 shows the values of K6 ,
(14)
Table 2 Thermodynamic data of the uptake of amino acid by cation-exchange resins Amino acid
G◦ 6 (kJ/mol)
H◦ 6 (kJ/mol)
G◦ 14 a (kJ/mol)
H◦ 14 b (kJ/mol)
G◦ 13 (kJ/mol)
H◦ 13 (KJ/mol)
TS◦ 13 (KJ/mol)
D001
Gly Ala Leu Phe
−14.6 −14.8 −16.3 −17.6
−5.13 −3.68 −2.83 −5.80
−13.4 −13.4 −13.3 −12.6
−5.15 −3.75 −2.98 −2.99
−1.2 −1.4 −3.0 −5.0
0.02 0.07 0.15 −2.84
1.2 1.5 3.2 2.2
PAMPS
Gly Phe
−12.7 −11.8
−5.14 −2.75
−13.4 −12.6
−5.15 −2.99
0.7 0.8
0.01 0.24
−0.7 −0.6
PVSS
Gly Phe
−12.5 −11.7
−5.10 −2.82
−13.4 −12.6
−5.15 −2.99
0.9 0.9
0.05 0.17
−0.8 −0.7
Resin
a b
Calculated from the pKa values taken from Ref. [21]. Taken from Ref. [22].
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
In order to study the ion exchange processes of the amino acid zwitterions on the ion exchange resin, the thermodynamic parameters of the equilibrium shown by Eq. (13) are needed. We assume the thermodynamic parameters of the equilibrium shown by Eq. (14) are the same with that of the corresponding amino acid zwitterion protonation, which are taken from the literature [21,22] and are listed in Table 2 (G◦ 14 and H◦ 14 ). Then G◦ 13 can be obtained from G◦ 13 = G◦ 6 − G◦ 14 and H◦ 13 can be obtained from H◦ 13 = H◦ 6 − H◦ 14 . T S◦ 13 was then obtained from G◦ 13 and H◦ 13 . All These thermodynamic parameters of the ion exchange equilibrium (Eq. (13)) are listed in Table 2. It can be seen that, for the exchange on the polystyrene-based strongly acidic cation exchange resin (D001), the values of H◦ 13 for Gly, Ala and Leu are almost the same and approximately to be zero, indicating their electrostatic interactions are almost the same, and the ammonium ion of the amino acids and the proton ion on the resin have the same order of ion exchange selectivity. The values of T S◦ 13 increase in the order of Gly, Ala, Phe and Leu, which is consistent with the hydrophobicity order of these amino acids [23,24]. This may be explained by different hydrophobic interactions between the side chain of the amino acids and the resin matrix while the electrostatic interactions remain almost the same. When the hydrophobic side chain of a amino acid binds to the hydrophobic resin matrix, highly ordered “icebergs” water molecules [25–28] surrounding the binding faces of both the side chain and the resin matrix would collapse and be released into the bulk aqueous solution, which represents a less ordered state, causing an increase of the entropy. The larger the binding face, the larger the entropy increase is. The value of H◦ 13 of Phe is, however, much more negative than those of the other amino acids. This may be attributed to additional – interaction between the phenyl ring of the amino acid and the phenyl ring of the resin matrix. In order to further verify the role of the hydrophobic interaction in the ion exchange process, another two strongly acidic cation-exchange resins (PAMPS and PVSS) with much less hydrophobic matrixes were synthesized and used to adsorb the amino acids. The structures of these resins as well as D001 and D001AM are shown in Fig. 3. Figs. 4 and 5 show the uptake isotherms of Gly and Phe by PAMPS and PVSS, respectively. The thermodynamic parameters of the uptakes were obtained similarly to those for D001 and are listed in Table 2. It can be seen that the thermodynamic parameters of the exchange of hydrophilic Gly and hydrophobic Phe are almost the same with negative entropy changes, indicating the exchanges of both Gly and Phe are controlled by the electrostatic interactions, while the hydrophobic interactions, if any, play a minor role in the ion exchange processes. The negative entropy changes indicate the loss of translational and rotational entropy upon binding. The loss of translational and rotational entropy for the Phe binding should be greater than that for Gly binding, as the former has higher molecular weight than the latter. The almost same entropy changes for the uptake of Gly and Phe by PAMPS or PVSS shown in Table 2 indicate that a weak hydrophobic interaction may exist between the side chain of Phe and the resin matrixes. These results further indicate that the hydrophobic interaction
Fig. 3. Structures of four resins used in this paper.
Fig. 4. Isotherms of uptake of amino acids by PAMPS at 25 ◦ C.
Fig. 5. Isotherms of uptake of amino acids by PVSS at 25 ◦ C.
47
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
48
play a role in the uptake of the amino acids with the hydrophobic side chains by polystyrene-based cation-exchange resins. In order to elucidate why the hydrophobic interaction contributes to the affinity of the amino acids with hydrophobic side chains for polystyrene-based cation-exchange resin but does not contribute to the maximum uptake capacity, a resin, D001AM, which possesses similar hydrophobicity with D001 but without ion exchange groups was prepared by transforming the sulfonic acid groups of D001 to sulfonamide groups. As D001AM has no ion exchange groups, the driving forces for the uptake of amino acids, if any, by D001AM may be hydrophobic interaction or hydrogen bonding or the both. The experimental results indicate that the uptake capacity of Leu and Phe by D001AM in the initial amino acid concentration up to 2 mM is not detectable. Thus, the hydrophobic interaction contributes little for the adsorption of Leu and Phe by D001AM. As the hydrophobicity of D001 is the same with or slightly lower than that of D001AM, it is deduced that the hydrophobic interaction contributes little too to the adsorption of Leu and Phe by D001 if the hydrophobic interaction occurs alone, i.e. G◦ h > 0. Then from Eq. (3) we can get G◦ h = H ◦ h − T S ◦ ho − T S ◦ t+r > 0
4. Conclusion
Then it is rewritten as: −(H
◦
h
− T S
◦
ho )
The above consideration of the synergistic effect has interesting implications for adsorptions driven by multiple weak interactions. Some weak interactions that contribute little to an adsorption when they act alone may significantly contribute to the adsorption when they act synergistically with other interactions. For example, the hydrophobic interaction, which is neglectable when it acts alone in the uptake of amino acids by polystyrene-based strongly acidic ion-exchange resin, as shown in this article, plays a key role in the selectivity in separation of the amino acids of the same group by ion exchange process using polystyrene-based strongly acidic ion-exchange resin, as in the post-column technique of amino acid analysis [11]. This kind of synergistic effect can also occur in the adsorptions driven by other multiple weak interactions [29–31]. An unexpected result is the greater entropy increase of uptake of Gly by D001 than those by PAMPS and PVSS. The expected driving force for these uptakes is the electrostatic interaction and they should be the same. The reason for this difference is not clear. This may be caused by the structural difference between the arylsulfonic acid groups of D001 and the alkylsulfonic acid groups of PAMPS or PVSS.
< −T S
◦
(15)
t+r
Eq. (15) indicates that, if the hydrophobic interaction occurs alone without ion exchange in the adsorption of Leu and Phe by D001, the positive contribution caused by the entropy increase from the collapse of the “icebergs” water molecules and enthalpy decrease from – interaction (for Phe) of the hydrophobic interaction could not offset the negative contribution caused by the loss of translational and rotational entropy upon binding. These results explain the reason why the maximum uptake capacity of hydrophobic organic ions by polystyrene-based ion exchange resins cannot excess the maximum ion-exchange capacity of the resins, as shown in this paper or by Dye et al. [13], and Li and SenGupta [2,15,16]. If the electrostatic and hydrophobic interactions occur in a chelate manner, the free energy change should be expressed by Eq. (4), which can be rewritten as Eq. (16). G◦ overal = H ◦ el + H ◦ h − T S ◦ elo − T S ◦ ho − T S ◦ t+r − T S ◦ br
(4)
G◦ overal = G◦ el + (H ◦ h − T S ◦ ho ) − T S ◦ br −(H◦
− T S◦
(16)
◦ Eq. (16) indicates that if h ho ) ≥ − T S br , the hydrophobic interaction has a positive contribution to the affinity of the amino acids for the resin upon the chelate binding. This means that if the positive contribution caused by the entropy increase from the collapse of the “icebergs” water molecules and enthalpy decrease from – interaction (for Phe) from the hydrophobic interaction is greater than the negative contribution caused by the loss of the restriction of the bond rotation upon forming the chelate form, the hydrophobic interaction has a positive contribution to the binding.
The uptake of Gly, Ala, Leu and Phe by a macroporous polystyrene-based strongly acidic cation-exchange resin (D001) essentially follows an ion exchange stoichiometry. The negative free energy changes for the ion exchange of the zwitterions of these amino acids by D001 increase in the order of Gly, Ala, Leu and Phe; the enthalpy changes the ion exchange for Gly, Ala and Leu are all proximately zero while the enthalpy change for Phe is much more negative; the entropy contributions (T S◦ ) are all positive values and increase in order of Gly, Ala, Phe and Leu. These results indicate that hydrophobic interaction (including – interaction for Phe) between the side chains of the amino acids and resin matrix may exist. In contrast, the binding affinities of Gly and Phe zwitterions for another two strongly acidic cation-exchange resins, PAMPS and PVSS, with much less hydrophobic matrixes are almost the same with negative entropy changes, indicating the hydrophobic interaction plays a minor role in the uptakes. No detectable uptake capacities for Leu and Phe by D001-sulfonamide, of which the hydrophobicity is similar to that of D001, deducing that the hydrophobic interaction alone contribute to little to the uptake of these two amino acids. These results indicate that synergistic effect exists between the electrostatic and hydrophobic interactions if the two interactions occur in a chelate manner. When the positive contribution of the hydrophobic interaction cannot offset the negative contribution caused by the loss of the translational and rotational entropy, but the energy gaining of the positive contribution is greater than the energy loss caused by restriction of bond rotation upon binding, the hydrophobic interaction contributes to the uptake affinity, but does not contribute to the maximum uptake capacity, or the uptake capacity cannot excess the maximum ion exchange capacity of the resin.
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
Acknowledgement This work was supported by the National Natural Science Foundation of China (Grant No. 20374028). References [1] A.L. Cinquina, A. Cali, F. Longo, L. De Santis, A. Severoni, F. Abballe, J. Chromatogr. A 1032 (2004) 73. [2] P. Li, A.K. SenGupta, Environ. Sci. Technol. 32 (1998) 3756. [3] M. Knochen, J. Giglio, Talanta 64 (2004) 1226. [4] P. Gluszcz, T. Jamroz, B. Sencio, S. Ledakowicz, Bioproc. Biosyst. Eng. 26 (2004) 185. [5] P. DePhillips, I. Lagerlund, J. Farenmark, A.M. Lenhoff, Anal. Chem. 76 (2004) 5816. [6] V. Anand, R. Kandarapu, S. Garg, Drug Discov. Today 6 (2001) 905. [7] M. Nichifor, X.X. Zhu, D. Cristea, A. Carpov, J. Phys. Chem. B 105 (2001) 2314. [8] X.X. Zhu, F. Brizard, J. Piche, C.T. Yim, G.R. Brown, J. Colloid Interface Sci. 232 (2000) 282. [9] G.E. Boyd, Q.V. Larson, J. Am. Chem. Soc. 89 (1967) 6038. [10] R.L. Gustafson, J.A. Lirio, Ind. Eng. Chem. Prod. Res. Dev. 7 (1968) 116. [11] G.W. Robinson, in: S. Blackburn (Ed.), Amino Acid Determination, second ed., Marcel Dekker Inc., New York, 1978 (Chapter 4).
49
[12] M.S. Saunders, J.B. Vierow, G. Garta, AIChE J. 35 (1989) 53. [13] S.R. Dye, J.P. DeCarli II, G. Carta, Ind. Eng. Chem. Res. 29 (1990) 849. [14] I.L. Jones, G. Carta, Ind. Eng. Chem. Res. 32 (1993) 107. [15] P. Li, A.K. SenGupta, Colloids Surf. A 60 (2004) 27. [16] P. Li, A.K. SenGupta, React. Funct. Polym. 60 (2004) 27. [17] W. Huang, C. Li, A. Men, T. Pan, B. He, Chem. J. Chin. Univ. 14 (1993) 432. [18] J.R. Spies, J. Biol. Chem. 195 (1952) 62. [19] T. Shen, Y.J. Wang, Biochemistry, Part One, Advanced Education Publisher, Beijing, 1980, p. 77. [20] G.F. Payne, N. Maity, Ind. Eng. Chem. Res. 31 (1992) 2024. [21] D. Voet, J.G. Voet, Biochemistry, third ed., John Wiley & Sons Inc., Hoboken, NJ, 2004, p. 66. [22] Z. Tao, H. Zhang, B. Zhang, React. Funct. Polym. 27 (1995) 29. [23] J. Kyte, R.F. Doolittle, J. Mol. Biol. 157 (1982) 105. [24] A. Radzicka, R. Wolfenden, Biochemistry 27 (1988) 1664. [25] H.S. Frank, M.W. Evans, J. Chem. Phys. 13 (1945) 507. [26] G. N´emethy, H.A. Scheraga, J. Chem. Phys. 36 (1962) 3382. [27] N. Muller, Acc. Chem. Res. 23 (1990) 23. [28] M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 7022. [29] G. Liu, H. Yu, H. Yan, Z. Shi, B. He, J. Chromatogr. A 952 (2002) 71. [30] S.D. Alexandratos, M.J. Hong, Sep. Sci. Technol. 37 (2002) 2587. [31] S.D. Alexandratos, S.D. Smith, J. Appl. Polym. Sci. 91 (2004) 463.
The synergistic effect between hydrophobic and electrostatic interactions in the uptake of amino acids by strongly acidic cation-exchange resins Shaoling Cheng, Husheng Yan ∗ , Changqing Zhao Key Laboratory of Functional Polymer Materials, Ministry of Education, Institute of Polymer Chemistry, Nankai University, Tianjin 300071, China Received 7 October 2005; received in revised form 15 December 2005; accepted 23 December 2005 Available online 26 January 2006
Abstract The goal of this work was to investigate the synergistic effect between the electrostatic and hydrophobic interactions upon the uptake of organic ions with hydrophobic moieties by ion-exchange resins with hydrophobic matrixes. The uptake of neutral amino acids by a macroporous polystyrene-based strongly acidic cation-exchange resin (D001) and two strongly acidic cation-exchange resins (poly(2-acrylamido-2-methyl propanesulfonic acid) and poly(vinylsulfonic acid)) with much less hydrophobic matrixes essentially follow an ion exchange stoichiometry. However, the thermodynamic parameters of the uptakes indicate that besides electrostatic interaction, hydrophobic interaction also contributes to the affinity of the amino acids with hydrophobic side chains for D001. No detectable uptake capacities for the amino acids by D001AM, which was obtained by amidation of the sulfonic acid groups of D001, can be determined. Thus, it is deduced that the hydrophobic interaction alone contributes little to the uptake of these amino acids by D001, of which hydrophobicity is the same with or lower than that of D001AM. These results indicate that synergistic effect exists between the electrostatic and hydrophobic interactions when the two interactions exist in a chelate manner and the hydrophobic interaction contributes to the uptake even if the hydrophobic interaction is so weak that it contributes little to the uptake when it acts alone. © 2006 Elsevier B.V. All rights reserved. Keywords: Synergistic effect; Ion exchange; Hydrophobic interaction; Ion-exchange resin; Amino acid
1. Introduction Utilization of synthetic ion exchange resins is the most prominent method employed in modern industrial processes to accomplish the task of separation and purification. Ionexchange resins are also used in analytical chemistry, e.g. as packings for ion exchange chromatography. Ion exchange processes of organic ions, such as biomolecules, drugs, etc., with ion-exchange resins have been widely used in their separations, purifications, removals from plasma or wastewater, analytical or clinical determinations, and in controlled drug deliveries [1–8]. It is considerable interest to find out the nature of the interactions in the ion exchange processes. The knowledge concerning the binding of organic ions by ion-exchange resins will aid in the
∗
Corresponding author. Tel.: +86 22 23503935; fax: +86 22 23503935. E-mail address: [email protected] (H. Yan).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2005.12.101
proper development and selection of ion-exchange resins, and selection of conditions of the ion exchange processes for these applications mentioned above. The electrostatic interaction between the organic ions and the oppositely charged ions on the resin is undoubtedly the prominent interaction. Besides the electrostatic interaction, hydrophobic interaction also exists between the hydrophobic moieties of the organic ions and the hydrophobic matrixes of the ion exchange resins (e.g. poly(styrene-divinylbenzene)-based ion-exchange resins) or other hydrophobic groups on the resin in some ion exchange processes [2,7,9–16]. However, the uptake of some hydrophobic organic ions by poly(styrene-divinylbenzene)-based ion-exchange resins follows an ion exchange stoichiometry and the hydrophobic interaction contributes little to the maximum uptake capacity, while the ion exchange selectivity is greatly controlled by the hydrophobic interaction in some ion exchange processes [13–16]. The apparent inconsistent effects of the hydrophobic interaction on uptake capacity and selectivity
44
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
have not yet been satisfactorily explained in the literature. For example, through studying the uptake capacity and selectivity coefficient, respectively, of Leu, Ile, Val, Ala and Phe by a polystyrene-based strongly acidic cation-exchange resin, two inconsistent conclusions were reached in a paper [13]: the uptake process depends only the ionic fraction of amino acid cations and the occurrence of nonionic adsorption can be discounted; the hydrophobic interactions of the amino acid side chain with the organic backbone of the resin play an important role in determining the affinity of amino acid cations for the resin’s functional group, without, however, significantly contributing to the maximum uptake capacity. In order to better understand these apparent inconsistent conclusions, we have investigated theoretically and experimentally the cooperativity between ion-exchanges and hydrophobic interactions in uptake of neutral amino acids by strongly acidic cation exchange resins. Based on our results, these inconsistent conclusions mentioned above can be satisfactorily explained. 2. Experimental section and thermodynamic analysis 2.1. Materials D001, a macroporous poly(styrene-8-wt.%-divinylbenzene)based sulfonic cation-exchange resin, was purchased from Hecheng Co. Ltd (Tianjin, China). The cation exchange capacity was determined to be 5.04 mmol/g dry resin. 2-Acrylamido2-methyl propanesulfonic acid (AMPS) was purchased from San Wing International Ltd (Shanghai, China). Sodium vinylsulfonate (VSS) (25% aqueous solution) was purchased from Shuang Jian Chemical Co. (Guangzhou, China). Acrylamide was purchased from Tianjin Medical Co. (Tianjin, China). N,N -Methylenebisacrylamide was purchased from Tianjin Institute of Chemical Reagent (Tianjin, China). Span-60 was purchased from Shanghai Chemical Reagent Co. (Shanghai, China). Glycine (Gly), leucine (Leu) and phenylalanine (Phe) were purchased from Shanghai Changjiang Biochemical Plant (Shanghai, China). 2.2. Synthesis of crosslinked PAMPS resin PAMPS, an AMPS-N,N -methylenebisacrylamide copolymer, was prepared as following. Twenty-five grams of AMPS, 6.25 g of N,N -methylenebisacrylamide and 0.25 g of K2 S2 O8 were dissolved in 125 ml of water. After the pH was adjusted to 7–8 with 20% NaOH, the aqueous solution was suspended in an oil phase which was composed of 311.5 ml of dichloroenthane, 138.5 ml of petroleum ether and 5.5 g of span-60 in a three-neck round-bottom flask with a mechanical stirrer and a nitrogen gas inlet. The mixture was stirred while bubbling nitrogen gas over for 30 min, making sure that there is no oxygen existing in the solution. Then a few drops of N,N,N ,N tetramethylethlenediamine was added into the flask to initiate the polymerization and the mixture was stirred for further 2 h at room temperature. The resulting copolymer was washed thoroughly with ethanol, water, 2N HCl and water. The cation
exchange capacity of the resulting resin, PAMPS, was determined to be 4.90 mmol/g dry resin. 2.3. Synthesis of crosslinked PVSS resin PVSS, a VSS-N,N -methylenebisacrylamide copolymer, was prepared as following. An aqueous solution which was composed of 4 ml of VSS, 3 g of acrylamide, 2.16 g of N,N methylenebisacrylamide, 0.096 g of K2 S2 O8 and 36 ml of water was suspended in an oil phase which was composed of 195 ml of chlorobenzene, 10 ml of toluene and 2.6 g of span-60 in a three-neck round-bottom flask with a mechanical stirrer and a nitrogen gas inlet. The mixture was stirred at while bubbling nitrogen gas over for 30 min, making sure that the oxygen was replaced completely. Then a few drops of N,N,N ,N -tetramethylethlenediamine was added into the mixture to initiate the polymerization and the mixture was stirred for further 2 h at room temperature. The resulting copolymer was washed thoroughly with ethanol, water, 2N HCl and water. The cation exchange capacity of the resulting resin, PVSS, was determined to be 2.50 mmol/g dry resin. 2.4. Synthesis of polystyrenesulfonamide resin (D001AM) from D001 D001AM, sulfonamide of D001, was prepared as following. Sulfonyl chloride of D001 was prepared according to a literature method [17]. Briefly, D001 was treated with pyridine in ethanol at room temperature over night to convert the sulfonic acid to pyridine salt. Then the resulting resin was treated with SOCl2 at 0 ◦ C to form sulfonyl chloride resin. The transformation was confirmed by the disappearance of a peak at 3442 cm−1 and the appearance of peaks at 1374 and 1173 cm−1 of the IR spectrum of the resin after the reaction. The sulfonyl chloride resin (3 g) was mixed with 5 ml of aqueous dimethylamine solution (containing 35% of dimethylamine) at 0 ◦ C, and then another 20 ml of aqueous dimethylamine solution was added dropwise into the mixture at 0 ◦ C while stirring. After all the dimethylamine solution was added, the mixture was stirred at room temperature over night. The resulting resin was washing thoroughly with water, and dried under vacuum. No peak at ∼3440 cm−1 appeared and two strong absorptions at 1337 and 1161 cm−1 appeared in the IR spectrum of the resulting resin, confirming that sulfonyl chloride groups have been converted to sulfonamide groups. Nitrogen content was determined by elementary analysis to be 4.76 mmol/g. Cation exchange capacity was determined to be 0.01 mmol/g dry resin. 2.5. Determination of cation exchange capacity A dry resin sample (∼0.1 g) was suspended in 25 ml of standard aqueous NaOH solution (∼0.1N) in an Erlenmeyer flask and the mixture was shaken for 8 h. 10 ml of the supernatant was taken from the flask and was titrated with standard aqueous HCl solution (∼0.1N) with methyl orange as the indicator. The cation exchange capacity (Qie in mmol/g) of the resin was
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
45
calculated by Qie =
25NNaOH − 2.5VHCl × NHCl W
where NNaOH and NHCl are the concentrations of the standard aqueous NaOH solution and the standard aqueous HCl solution, respectively. VHCl is the volume (ml) of the standard aqueous HCl used in the titration, W is the sample weight (g). 2.6. Equilibrium uptake of amino acids by resins Resin samples (∼0.2 g) were added into Erlenmeyer flasks containing 25 ml of amino acid solutions with certain concentrations. The flasks were sealed and shaken for 20 h in a constanttemperature bath. Then the concentration of amino acids in the supernatant was determined by photometry using Cu2+ as the chromogenic reagent [18]. The uptake capacity was calculated from the initial and equilibrium concentrations, the volume of the solution used and the sample weight. For determination of the uptake equilibrium constant, the adsorption temperature was 25 ◦ C. For the van’t Hoff plot (see Section 3), 25, 30, 40 and 50 ◦ C were chosen. The amino acid concentration ranges were 0–1.2 mM for D001, 0–2.0 mM for PAMPS and 0–3.0 mM for PVSS, where the adsorption isotherms are linear. 2.7. Thermodynamic analysis of uptake of organic ions by ion-exchange resin Poly(styrene-divinylbenzene)-based ion-exchange resins are mostly common ion exchange resins in practical applications. It has been noticed that hydrophobic interactions contributed to the adsorption of some hydrophobic organic ions by poly(styrenedivinylbenzene)-based ion-exchange resins. We first consider the electrostatic and hydrophobic interactions independently. For the electrostatic interaction alone in the ion exchange process, the free energy change at the standard state is given by G◦ el = H ◦ el − T S ◦ el
(1)
We divide the entropy change S◦ el into two components, i.e. negative entropy change caused by the loss of translational and rotational entropy when the solute was immobilized onto the resin by the electrostatic interaction (ion exchange), S◦ elt+r , and the other entropy changes, S◦ elo . Then Eq. (1) can be rewritten as G◦ el = H ◦ el − T S ◦ elo − T S ◦ elt+r
(2)
For the hydrophobic interaction without ion exchange, similar equation can be written as: G◦ h = H ◦ h − T S ◦ ho − T S ◦ ht+r
Fig. 1. Scheme of chelate manner of the electrostatic and hydrophobic interactions.
the loss of translational and rotational entropy for the electrostatic interaction which occurs alone is the same with that for the hydrophobic interaction which occurs alone. G◦ overal = H ◦ el + H ◦ h − T S ◦ elo − T S ◦ ho − T S ◦ t+r − T ∆S ◦ br
(4)
S◦ br is the entropy change caused by restriction of the free rotation of the single bonds between the ion group and the hydrophobic moiety of both the resin and the organic ion upon the chelate binding. The value of S◦ br should be negative. Eq. (4) can be re-written as: G◦ overal = (H ◦ el − T S ◦ elo − T S ◦ t+r ) + (H ◦ h − T S ◦ ho − T S ◦ t+r ) + T S ◦ t+r − T S ◦ br = G◦ el + G◦ h + T (S ◦ t+r − S ◦ br )
(5)
It can be concluded from Eq. (5) that the overall free energy change (G◦ overal ) of the chelate binding would be more negative than the sum of the free energy changes (G◦ el and G◦ h ) of the two interactions occurred independently when (S◦ t+r − S◦ br ) < 0, or when the loss of the translational and rotational entropy is greater than the loss of the bond rotation entropy upon binding. We can say there is cooperativity or synergistic effect between the two interactions in this case. 3. Experimental results and discussion Neutral amino acids such as Gly, Ala, Leu and Phe in neutral aqueous solution should exist in their zwitterions. Their isoelectric pH values are 5.97, 6.00, 5.98 and 5.48, respectively [19]. Thus, uptake of neutral amino acids by cation-exchange resins in aqueous medium can be expressed in the following equilibrium:
(3)
If the electrostatic and hydrophobic interactions occur simultaneously in a chelate manner, as shown in Fig. 1, the binding free energy change (G◦ overal ) can be described by Eq. (4) upon the assumption of (1) the natures of the two interactions do not influence each other (in other words, the binding sites have good complementary), and (2) S◦ ht+r = S◦ elt+r = S◦ t+r , i.e.,
(6) The equilibrium constant, K6 , can be expressed as: K6 =
qe ce (qmax − qe )
(7)
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
46
qmax and the correlation coefficients (r2 ) by the fitting. It can be seen from Table 1 that the maximum uptake capacity, qmax , for all amino acids tested approaches the total exchange capacity of the resin, indicating a stoichiometric ion exchange process similar to the result of Dye et al. [13]. The free energy change (G◦ 6 ) of the equilibrium expressed by Eq. (6) can be obtained from the data in Table 1 according to Eq. (9) and was listed in Table 2. G◦ = −RT ln K
(9)
The enthalpy change of the uptake can be determined from van’t Hoff equation [20]: ln K = − Fig. 2. Isotherms of uptake of amino acids by D001 at 25 ◦ C.
H ◦ + k0 RT
(10)
where k0 is a constant. In the low-concentration region where the isotherm was linear, equilibrium constant (K) is correlated to adsorption affinity (qe /ce ),
Table 1 Langmuir adsorption parameters and correlation coefficients (T = 298 K) Amino acid
K6 (mM−1 )
qmax (mmol/g)
r2
qe = ψK ce
D001
Gly Ala Leu Phe
0.333 0.350 0.656 1.10
5.02 5.01 4.96 4.84
0.9997 0.9979 0.9976 0.9967
PAMPS
Gly Phe
0.153 0.107
4.63 4.76
0.9987 0.9987
where ψ is a constant. From Eqs. (10) and (11), we can get the following equation: H ◦ qe =− ln +c (12) ce RT
PVSS
Gly Phe
0.145 0.105
2.38 2.47
0.9984 0.9990
Ion-exchange resin
where qe is the uptake capacity under the equilibrium concentration, ce . qmax is the maximum uptake capacity. Eq. (7) can be re-written as: qe = qmax
K 6 ce 1 + K 6 ce
(11)
where c is a constant. A plot of ln(qe /ce ) versus 1/T yields a straight line with a slope of −H◦ /R, from which H◦ 6 can be obtained, as shown in Table 2. The uptake equilibrium expressed by Eq. (6) involves two equilibriums: ion exchange and protonation, as shown by Eqs. (13) and (14), respectively.
(8) (13)
Eq. (8) has the same form with Langmuir adsorption equation. Fig. 2 shows the uptake isotherms of Gly, Ala, Leu and Phe by D001. These data fit Eq. (8) (Langmuir equation) well, as shown by the curves in Fig. 2. Table 1 shows the values of K6 ,
(14)
Table 2 Thermodynamic data of the uptake of amino acid by cation-exchange resins Amino acid
G◦ 6 (kJ/mol)
H◦ 6 (kJ/mol)
G◦ 14 a (kJ/mol)
H◦ 14 b (kJ/mol)
G◦ 13 (kJ/mol)
H◦ 13 (KJ/mol)
TS◦ 13 (KJ/mol)
D001
Gly Ala Leu Phe
−14.6 −14.8 −16.3 −17.6
−5.13 −3.68 −2.83 −5.80
−13.4 −13.4 −13.3 −12.6
−5.15 −3.75 −2.98 −2.99
−1.2 −1.4 −3.0 −5.0
0.02 0.07 0.15 −2.84
1.2 1.5 3.2 2.2
PAMPS
Gly Phe
−12.7 −11.8
−5.14 −2.75
−13.4 −12.6
−5.15 −2.99
0.7 0.8
0.01 0.24
−0.7 −0.6
PVSS
Gly Phe
−12.5 −11.7
−5.10 −2.82
−13.4 −12.6
−5.15 −2.99
0.9 0.9
0.05 0.17
−0.8 −0.7
Resin
a b
Calculated from the pKa values taken from Ref. [21]. Taken from Ref. [22].
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
In order to study the ion exchange processes of the amino acid zwitterions on the ion exchange resin, the thermodynamic parameters of the equilibrium shown by Eq. (13) are needed. We assume the thermodynamic parameters of the equilibrium shown by Eq. (14) are the same with that of the corresponding amino acid zwitterion protonation, which are taken from the literature [21,22] and are listed in Table 2 (G◦ 14 and H◦ 14 ). Then G◦ 13 can be obtained from G◦ 13 = G◦ 6 − G◦ 14 and H◦ 13 can be obtained from H◦ 13 = H◦ 6 − H◦ 14 . T S◦ 13 was then obtained from G◦ 13 and H◦ 13 . All These thermodynamic parameters of the ion exchange equilibrium (Eq. (13)) are listed in Table 2. It can be seen that, for the exchange on the polystyrene-based strongly acidic cation exchange resin (D001), the values of H◦ 13 for Gly, Ala and Leu are almost the same and approximately to be zero, indicating their electrostatic interactions are almost the same, and the ammonium ion of the amino acids and the proton ion on the resin have the same order of ion exchange selectivity. The values of T S◦ 13 increase in the order of Gly, Ala, Phe and Leu, which is consistent with the hydrophobicity order of these amino acids [23,24]. This may be explained by different hydrophobic interactions between the side chain of the amino acids and the resin matrix while the electrostatic interactions remain almost the same. When the hydrophobic side chain of a amino acid binds to the hydrophobic resin matrix, highly ordered “icebergs” water molecules [25–28] surrounding the binding faces of both the side chain and the resin matrix would collapse and be released into the bulk aqueous solution, which represents a less ordered state, causing an increase of the entropy. The larger the binding face, the larger the entropy increase is. The value of H◦ 13 of Phe is, however, much more negative than those of the other amino acids. This may be attributed to additional – interaction between the phenyl ring of the amino acid and the phenyl ring of the resin matrix. In order to further verify the role of the hydrophobic interaction in the ion exchange process, another two strongly acidic cation-exchange resins (PAMPS and PVSS) with much less hydrophobic matrixes were synthesized and used to adsorb the amino acids. The structures of these resins as well as D001 and D001AM are shown in Fig. 3. Figs. 4 and 5 show the uptake isotherms of Gly and Phe by PAMPS and PVSS, respectively. The thermodynamic parameters of the uptakes were obtained similarly to those for D001 and are listed in Table 2. It can be seen that the thermodynamic parameters of the exchange of hydrophilic Gly and hydrophobic Phe are almost the same with negative entropy changes, indicating the exchanges of both Gly and Phe are controlled by the electrostatic interactions, while the hydrophobic interactions, if any, play a minor role in the ion exchange processes. The negative entropy changes indicate the loss of translational and rotational entropy upon binding. The loss of translational and rotational entropy for the Phe binding should be greater than that for Gly binding, as the former has higher molecular weight than the latter. The almost same entropy changes for the uptake of Gly and Phe by PAMPS or PVSS shown in Table 2 indicate that a weak hydrophobic interaction may exist between the side chain of Phe and the resin matrixes. These results further indicate that the hydrophobic interaction
Fig. 3. Structures of four resins used in this paper.
Fig. 4. Isotherms of uptake of amino acids by PAMPS at 25 ◦ C.
Fig. 5. Isotherms of uptake of amino acids by PVSS at 25 ◦ C.
47
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
48
play a role in the uptake of the amino acids with the hydrophobic side chains by polystyrene-based cation-exchange resins. In order to elucidate why the hydrophobic interaction contributes to the affinity of the amino acids with hydrophobic side chains for polystyrene-based cation-exchange resin but does not contribute to the maximum uptake capacity, a resin, D001AM, which possesses similar hydrophobicity with D001 but without ion exchange groups was prepared by transforming the sulfonic acid groups of D001 to sulfonamide groups. As D001AM has no ion exchange groups, the driving forces for the uptake of amino acids, if any, by D001AM may be hydrophobic interaction or hydrogen bonding or the both. The experimental results indicate that the uptake capacity of Leu and Phe by D001AM in the initial amino acid concentration up to 2 mM is not detectable. Thus, the hydrophobic interaction contributes little for the adsorption of Leu and Phe by D001AM. As the hydrophobicity of D001 is the same with or slightly lower than that of D001AM, it is deduced that the hydrophobic interaction contributes little too to the adsorption of Leu and Phe by D001 if the hydrophobic interaction occurs alone, i.e. G◦ h > 0. Then from Eq. (3) we can get G◦ h = H ◦ h − T S ◦ ho − T S ◦ t+r > 0
4. Conclusion
Then it is rewritten as: −(H
◦
h
− T S
◦
ho )
The above consideration of the synergistic effect has interesting implications for adsorptions driven by multiple weak interactions. Some weak interactions that contribute little to an adsorption when they act alone may significantly contribute to the adsorption when they act synergistically with other interactions. For example, the hydrophobic interaction, which is neglectable when it acts alone in the uptake of amino acids by polystyrene-based strongly acidic ion-exchange resin, as shown in this article, plays a key role in the selectivity in separation of the amino acids of the same group by ion exchange process using polystyrene-based strongly acidic ion-exchange resin, as in the post-column technique of amino acid analysis [11]. This kind of synergistic effect can also occur in the adsorptions driven by other multiple weak interactions [29–31]. An unexpected result is the greater entropy increase of uptake of Gly by D001 than those by PAMPS and PVSS. The expected driving force for these uptakes is the electrostatic interaction and they should be the same. The reason for this difference is not clear. This may be caused by the structural difference between the arylsulfonic acid groups of D001 and the alkylsulfonic acid groups of PAMPS or PVSS.
< −T S
◦
(15)
t+r
Eq. (15) indicates that, if the hydrophobic interaction occurs alone without ion exchange in the adsorption of Leu and Phe by D001, the positive contribution caused by the entropy increase from the collapse of the “icebergs” water molecules and enthalpy decrease from – interaction (for Phe) of the hydrophobic interaction could not offset the negative contribution caused by the loss of translational and rotational entropy upon binding. These results explain the reason why the maximum uptake capacity of hydrophobic organic ions by polystyrene-based ion exchange resins cannot excess the maximum ion-exchange capacity of the resins, as shown in this paper or by Dye et al. [13], and Li and SenGupta [2,15,16]. If the electrostatic and hydrophobic interactions occur in a chelate manner, the free energy change should be expressed by Eq. (4), which can be rewritten as Eq. (16). G◦ overal = H ◦ el + H ◦ h − T S ◦ elo − T S ◦ ho − T S ◦ t+r − T S ◦ br
(4)
G◦ overal = G◦ el + (H ◦ h − T S ◦ ho ) − T S ◦ br −(H◦
− T S◦
(16)
◦ Eq. (16) indicates that if h ho ) ≥ − T S br , the hydrophobic interaction has a positive contribution to the affinity of the amino acids for the resin upon the chelate binding. This means that if the positive contribution caused by the entropy increase from the collapse of the “icebergs” water molecules and enthalpy decrease from – interaction (for Phe) from the hydrophobic interaction is greater than the negative contribution caused by the loss of the restriction of the bond rotation upon forming the chelate form, the hydrophobic interaction has a positive contribution to the binding.
The uptake of Gly, Ala, Leu and Phe by a macroporous polystyrene-based strongly acidic cation-exchange resin (D001) essentially follows an ion exchange stoichiometry. The negative free energy changes for the ion exchange of the zwitterions of these amino acids by D001 increase in the order of Gly, Ala, Leu and Phe; the enthalpy changes the ion exchange for Gly, Ala and Leu are all proximately zero while the enthalpy change for Phe is much more negative; the entropy contributions (T S◦ ) are all positive values and increase in order of Gly, Ala, Phe and Leu. These results indicate that hydrophobic interaction (including – interaction for Phe) between the side chains of the amino acids and resin matrix may exist. In contrast, the binding affinities of Gly and Phe zwitterions for another two strongly acidic cation-exchange resins, PAMPS and PVSS, with much less hydrophobic matrixes are almost the same with negative entropy changes, indicating the hydrophobic interaction plays a minor role in the uptakes. No detectable uptake capacities for Leu and Phe by D001-sulfonamide, of which the hydrophobicity is similar to that of D001, deducing that the hydrophobic interaction alone contribute to little to the uptake of these two amino acids. These results indicate that synergistic effect exists between the electrostatic and hydrophobic interactions if the two interactions occur in a chelate manner. When the positive contribution of the hydrophobic interaction cannot offset the negative contribution caused by the loss of the translational and rotational entropy, but the energy gaining of the positive contribution is greater than the energy loss caused by restriction of bond rotation upon binding, the hydrophobic interaction contributes to the uptake affinity, but does not contribute to the maximum uptake capacity, or the uptake capacity cannot excess the maximum ion exchange capacity of the resin.
S. Cheng et al. / J. Chromatogr. A 1108 (2006) 43–49
Acknowledgement This work was supported by the National Natural Science Foundation of China (Grant No. 20374028). References [1] A.L. Cinquina, A. Cali, F. Longo, L. De Santis, A. Severoni, F. Abballe, J. Chromatogr. A 1032 (2004) 73. [2] P. Li, A.K. SenGupta, Environ. Sci. Technol. 32 (1998) 3756. [3] M. Knochen, J. Giglio, Talanta 64 (2004) 1226. [4] P. Gluszcz, T. Jamroz, B. Sencio, S. Ledakowicz, Bioproc. Biosyst. Eng. 26 (2004) 185. [5] P. DePhillips, I. Lagerlund, J. Farenmark, A.M. Lenhoff, Anal. Chem. 76 (2004) 5816. [6] V. Anand, R. Kandarapu, S. Garg, Drug Discov. Today 6 (2001) 905. [7] M. Nichifor, X.X. Zhu, D. Cristea, A. Carpov, J. Phys. Chem. B 105 (2001) 2314. [8] X.X. Zhu, F. Brizard, J. Piche, C.T. Yim, G.R. Brown, J. Colloid Interface Sci. 232 (2000) 282. [9] G.E. Boyd, Q.V. Larson, J. Am. Chem. Soc. 89 (1967) 6038. [10] R.L. Gustafson, J.A. Lirio, Ind. Eng. Chem. Prod. Res. Dev. 7 (1968) 116. [11] G.W. Robinson, in: S. Blackburn (Ed.), Amino Acid Determination, second ed., Marcel Dekker Inc., New York, 1978 (Chapter 4).
49
[12] M.S. Saunders, J.B. Vierow, G. Garta, AIChE J. 35 (1989) 53. [13] S.R. Dye, J.P. DeCarli II, G. Carta, Ind. Eng. Chem. Res. 29 (1990) 849. [14] I.L. Jones, G. Carta, Ind. Eng. Chem. Res. 32 (1993) 107. [15] P. Li, A.K. SenGupta, Colloids Surf. A 60 (2004) 27. [16] P. Li, A.K. SenGupta, React. Funct. Polym. 60 (2004) 27. [17] W. Huang, C. Li, A. Men, T. Pan, B. He, Chem. J. Chin. Univ. 14 (1993) 432. [18] J.R. Spies, J. Biol. Chem. 195 (1952) 62. [19] T. Shen, Y.J. Wang, Biochemistry, Part One, Advanced Education Publisher, Beijing, 1980, p. 77. [20] G.F. Payne, N. Maity, Ind. Eng. Chem. Res. 31 (1992) 2024. [21] D. Voet, J.G. Voet, Biochemistry, third ed., John Wiley & Sons Inc., Hoboken, NJ, 2004, p. 66. [22] Z. Tao, H. Zhang, B. Zhang, React. Funct. Polym. 27 (1995) 29. [23] J. Kyte, R.F. Doolittle, J. Mol. Biol. 157 (1982) 105. [24] A. Radzicka, R. Wolfenden, Biochemistry 27 (1988) 1664. [25] H.S. Frank, M.W. Evans, J. Chem. Phys. 13 (1945) 507. [26] G. N´emethy, H.A. Scheraga, J. Chem. Phys. 36 (1962) 3382. [27] N. Muller, Acc. Chem. Res. 23 (1990) 23. [28] M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 7022. [29] G. Liu, H. Yu, H. Yan, Z. Shi, B. He, J. Chromatogr. A 952 (2002) 71. [30] S.D. Alexandratos, M.J. Hong, Sep. Sci. Technol. 37 (2002) 2587. [31] S.D. Alexandratos, S.D. Smith, J. Appl. Polym. Sci. 91 (2004) 463.