pubs.acs.org/Langmuir © 2009 American Chemical Society
Characterization of Copolymer Latexes by Capillary Electrophoresis Nadia Anik,† Marc Airiau,‡ Marie-Pierre Labeau,§ Wojciech Bzducha,‡ and Herve Cottet*,† †
Institut des Biomol ecules Max Mousseron (IBMM, UMR 5247 CNRS - Universit e de Montpellier 1- Universit e de Montpellier 2), place Eug ene Bataillon, case courrier 1706, 34095 Montpellier Cedex 5, France, ‡ Rhodia Op erations, 52 rue de la Haie Coq, 93308 Aubervilliers, France, and §Rhodia Inc., 350 G. Patterson Bvd, Bristol, Pennsylvania 19007 Received July 21, 2009. Revised Manuscript Received September 25, 2009 Latexes are widely used for industrial applications, including decorative paints, binders for the papermaking industry, and drilling fluids for oil-field applications. In this work, the interest of capillary zone electrophoresis (CE) for the characterization of hydrophobic block copolymer latexes obtained by the conventional emulsion polymerization technique consisting of a core of polystyrene (PS) surrounded by a layer of poly(ethyl acrylate) (PEA) has been investigated. The PEA part of the copolymer can be partially hydrolyzed in poly(acrylic acid) (PAA) leading to PS-PEA-AA water-soluble amphiphilic copolymer having high viscosifying properties. The main purpose of this work was to evaluate the potential of CE for the characterization of the latexes at the different stages of the synthesis (PS core, PS-PEA diblock latex, and hydrolyzed PS-PEA-AA gel). The main analytical issues were to state (i) if there was free PS or PEA homopolymer latexes in the PS-PEA latex sample and (ii) if there was free PS, PEA, PS-PEA latexes, or free PAA chains in the PS-PEA-AA gel. Within this scope, this work describes the optimization of the selectivity of the separation between the different species (PS, PEA particles in the not hydrolyzed diblock latex and PS, PEA, PS-PEA particles as well as the polymer PAA chains in the PS-PEA-AA diblock gel sample obtained by latter latex hydrolysis). For that purpose, several experimental parameters were investigated such as pH and ionic strength of the background electrolyte (BGE) or the concentration of neutral surfactant added in the BGE. A challenging issue was to overcome the high viscosity of the PS-PEA-AA gel. This was resolved by the addition of 10 mM neutral surfactant in the gel sample and in the BGE. Finally, it is demonstrated that, within the detection limits, CE is a suitable analytical tool for controlling and monitoring the syntheses of these latexes and for intrinsically characterizing the distribution in charge density of the final PS-PEA-AA gel at different hydrolysis rates.
1. Introduction Initially developed to satisfy the requirement of the rubber industry, synthetic latexes represent today an extremely wide area of research with numerous applications. Many industrial sectors, such as painting, coating, adhesives, papermaking industry, cosmetics, and, more recently, the biomedical and pharmaceutical industry, develop applications based on synthetic latexes.1 Latexes are related to a colloid solid/liquid medium since they are submicrometer polymeric particles dispersed in aqueous media as a suspension. Most present latexes in the industry are families of vinyl acetate,2,3 carboxylated styrene-butadiene copolymers,4-6 polyurethanes, and styrene acrylic copolymers. The characterization of latexes is commonly carried out by electron microscopy7 (TEM or SEM), scattering techniques (DLS, SAXS, SANS),8 or NMR.9,10 Among the separation techniques used, gel electrophoresis provides information on *Corresponding author: Tel +33 4 6714 3427, Fax +33 4 6763 1046, e-mail
[email protected]. (1) Daniel, J. C.; Pichot, C. les latex synthetiques; Paris, 2006. (2) Bouchard, M. Gulf Oil Canada Limited, Canada, 1970. (3) Inskip, H. K. Du Pont de Nemours, 1971. (4) Brown, H. P. B.F. Goodrich Company, 1955. (5) Eilbeck, G. E.; Urig, E. R. B.F. Goodrich Company, 1959. (6) Bolze, J.; Ballauf, M.; Kijlstra, J.; Rudhardt, D. Macromol. Mater. Eng. 2003, 288, 495–502. (7) Woodwards, R. C.; Heeris, J.; St. Pierre, T. G.; Saunders, M. E. A. J. Appl. Crystallogr. 2007, 40, 495–500. (8) Holoubek, J. J. Quant. Spectrosc. Radiat. Transf. 2007, 106, 104–121. (9) Soula, O.; Petiaud, R.; Llauro, M. F.; Guyot, A. Macromolecules 1999, 32, 6938–6943. (10) Tarcha, P. J.; Fitch, R. M.; Dumais, J. J.; Jelinski, L. W. J. Polym. Sci., Polym. Phys. 1983, 21, 2389–2402. (11) Tietz, D. J. Chromatogr.: Biomed. Appl. 1987, 418, 305–344. (12) Hanauer, M.; Pierrat, S.; Zins, I.; Lotz, A.; Sonnichsen, C. Nano Lett. 2007, 7, 2881–2885.
1700 DOI: 10.1021/la902661m
isoelectric point (IEP)11 and are employed with the purpose of profiling12 or purification. Size exclusion chromatography (SEC) gives access to the molecular weight distribution and allows the separation of polymer chains according to their hydrodynamic volume. However, analytical techniques without stationary phase are preferred to SEC such as field-flow fractionation13-17 (FFF), which give access to particles size distribution. Capillary electrophoresis (CE) has been also used for the investigation of colloidal systems, in particular for the separation of latex particles according to their size and charge.18-20 The main body of the early work on the CE of colloidal solutions was focused on carboxylated and sulfated polystyrene particles with a stress on the separation according to the size.21-23 For colloidal particles of similar chemical composition, CE separation according to the size was possible with larger particles having generally the highest electrophoretic mobilities. Radko et al.24 pointed out the physical mechanisms at the origin of the separation. It was demonstrated that the size-based separation of sub-micrometer polystyrene (13) Jensen, K. D.; Williams, S. K. R.; Calvin Giddings, J. J. Chromatogr. A 1996, 746, 137–145. (14) Tri, N.; Caldwell, K.; Beckett, R. Anal. Chem. 2000, 72, 1823–1829. (15) Lee, W. J.; Min, B. R.; Moon, M. H. Anal. Chem. 1999, 71, 3446–3452. (16) Lee, S.; Rao, S. P.; Moon, M. H.; Giddings, J. C. Anal. Chem. 1996, 68, 1545–1549. (17) Shirley Carro, J. H.-O. Macromol. Rapid Commun. 2006, 27, 274–278. (18) Janca, J.; Le Hen, S.; Scaronpı´ rkova, M.; Stejskal, J. J. Microcolumn Sep. 1997, 9, 303–306. (19) Radko, S. P.; Chrambach, A. Electrophoresis 2002, 23, 1957–1972. (20) Rodriguez, M. A.; Armstrong, D. W. J. Chromatogr. B 2004, 800, 7–25. (21) VanOrman, B. B.; McIntire, G. L. J. Microcolumn Sep. 1989, 1, 289–293. (22) Jones, H. K.; Ballou, N. E. Anal. Chem. 1990, 62, 2484–2490. (23) Petersen, S. L.; Ballou, N. E. Anal. Chem. 1992, 64, 1676–1681. (24) Radko, S. P.; Stastna, M.; Chrambach, A. Electrophoresis 2000, 21, 3583– 3592.
Published on Web 10/29/2009
Langmuir 2010, 26(3), 1700–1706
Anik et al.
latexes in capillary zone electrophoresis (CZE) is a function of the ionic strength according to the electrokinetic theory described by Overbeek-Booth.25 Furthermore, distribution in electrophoretic mobility due to sample heterogeneity was determined to be the prevailing factor contributing to band broadening. This effect was confirmed by Stokes et al.26 on submicrometer commercial particles with fluorescent dye. Vanhoenacker et al.27 studied the separation of a series of acrylic styrene copolymer particles. Particles similar in size (65, 68, 69 nm) but of different chemical compositions were separated due to different charge densities and thus different zeta potential. Particles of different sizes (50, 68, and 270 nm) but of similar compositions (similar charge density) were also separated since the relaxation effect depends on the size of the particle. Therefore, electrophoretic separation depends on both the size and the chemical composition of the particles. In commonly employed electrophoretic conditions,21,22,24,28 the higher the particles size is or the charge density is, the higher the zeta potential is and the higher the electrophoretic mobility is (in absolute value). Others groups performed CE experiments on inorganic particles. Successful separations of silica29,30 and metal oxide particles31-33 have been described. Morneau et al.34 described the application of CE to the analysis of magnetic particles with stabilizing surface groups. More recently, d’Orlye et al.35 reported a comprehensive study on magnetic particles by CZE. The influence of particles size and concentration, applied voltage, and ionic strength is discussed. Recently, Rhodia has developed latexes of block copolymers obtained by MADIX technology composed of a hydrophobic core of polystyrene (PS) surrounded by a layer of poly(ethyl acrylate) (PEA).36,37 The outer PEA layer can be partially hydrolyzed into poly(acrylic acid) (PAA). These PS-PEA-AA amphiphilic copolymers have strong viscosifying properties at low polymer concentrations in water and present interesting rheological behaviors for many applications, involving rheology control in aqueous industrial formulations. The characterization of these products, either original latexes or gels, must address the chemical homogeneity, since two unwanted situations may occur: first, an incomplete extension of PS homopolymer cores by PEA second blocks, leaving a certain fraction of uncoated PS particles, and second, the polymerization of the ethyl acrylate in uncontrolled manner generates a population of core-free homopolymer PEA latex. Capillary electrophoresis can provide a separation process, where surface charge density is the key parameter and can hopefully distinguish latexes with different chemical compositions and similar sizes. However, a particular issue for this kind of separation is brought by the high viscosity of partially hydrolyzed latex (PS-PEA/AA). We shall see how the optimization of (25) Overbeek, J. T. G.; Wiersema, P. H. Electrophoresis: Theory, Methods, and Applications; Academic Press: New York, 1967; pp 1-52. (26) Stokes, J. C.; Johnson, M. E. Microchem. J. 2004, 76, 121–129. (27) Vanhoenacker, G.; Goris, L.; Sandra, P. Electrophoresis 2001, 22, 2490– 2494. (28) Vanifatova, N. G.; Spivakov, B. Y.; Mattusch, J.; Wennrich, R. J. Chromatogr. A 2000, 898, 257–263. (29) McCormick, R. M. J. Liq. Chromatogr. 1991, 14, 939–952. (30) Vanifatova, N. G.; Spivakov, B. Y.; Mattusch, J.; Wennrich, R. Talanta 2003, 59, 345–353. (31) Ducatte, G. R.; Ballou, N. E.; Quang, C. Y.; Petersen, S. L. J. Microcolumn Sep. 1996, 8, 403–412. (32) Quang, C.; Petersen, S. L.; Ducatte, G. R.; Ballou, N. E. J. Chromatogr. A 1996, 732, 377–384. (33) Petersen, S. L.; Ballou, N. E. J. Chromatogr. A 1999, 834, 445–452. (34) Morneau, A.; Pillai, V.; Nigam, S.; Winnik, F. M.; Ziolo, R. F. Colloids Surf., A 1999, 154, 295–301. (35) D’Orlye, F.; Varenne, A.; Gareil, P. Electrophoresis 2008, 29, 1–11. (36) Destarac, M.; Joanicot, M.; Reeb, R. Rhodia Chimie, US 6 506 837 B2, 2003. (37) Bendejacq, D.; Ponsinet, V.; Joanicot, M. Eur. Phys. J. E 2004, 13, 3–13.
Langmuir 2010, 26(3), 1700–1706
Article
experimental parameters can be worked out in a CE analysis based on interactions between surfactants and gels of PS-PEA/ AA copolymers. The purpose of the present work is to separate and to quantify residual homopolymers and/or latexes at each step of the synthesis of block copolymer via MADIX (Macromolecular Design by the Interchange of Xanthates) approach leading to PS-PEA-AA gels.
2. Experimental Part 2.1. Chemicals. Sodium phosphate monobasic (NaH2PO4) was purchased from Aldrich (Steinheim, Germany). Sodium tetraborate decahydrate was from Fluka (Buchs, Switzerland). Boric acid (HBO2) was from Avocado (Heysham, England). Sodium hydroxide (NaOH) was from VWR (Leuven, Belgium). Poly(ethylene glycol) hexadecyl ether (HO(CH2CH2O)20C16H33, Brij 58) and poly(ethylene glycol) octadecyl ether (Brij 78) were from Aldrich (Milwaukee, WI). Deionized water was further purified with a Milli-Q system from Millipore (Molsheim, France). 2.2. Polymers. Materials. We used the xanthate named Rhodixan A1 (2-mercaptopropionic acid, methyl ester, o-ethyl dithiocarbonate) provided by Rhodia (99%, NMR purity) as a MADIX (Macromolecular Design by Interchange of Xanthates) control agent. Ethyl acrylate (EA), styrene (S), methacrylic acid (MAA), sodium dodecyl sulfate surfactant (SDS), sodium peroxodisulfate initiator (Na2S2O8), and sodium carbonate (Na2CO3) were from Aldrich (99%, purity) and were used as received. Water was purified using a Milli-Q plus water purification system. Polymerization. Typical emulsion polymerization reactions were performed in a 2500 mL jacketed reactor. Reactor was equipped with a cold water condenser and mechanical stirring system. The Rhodixan A1 (xanthate) amount was adapted according to the targeted molecular weight Mn of polymer by eq 1: mxanthate ¼
mmonomer Mw;xanthate Mn;polymer
ð1Þ
where mxanthate and mmonomer are respectively the mass of the xanthate and the monomer in grams; M w,xanthate and M n,polymer are respectively the molecular weight of the xanthate (208 g/mol) and the target polymer molecular weight. A typical reaction time was 2-5 h for each block (styrene and ethyl acrylate) to achieve ∼99% conversion. The detailed synthesis procedure is described in ref 36. The following polymers were synthesized and used for the study: (i) A polystyrene (first block) of theoretical molecular weight of Mn =2000 g/mol (actually it is a statistical copolymer of styrene (S) and methacrylic acid (MAA), with mass ratio S/MAA=98/2). The molecular weight of the produced polymer was analyzed by size exclusion chromatography (SEC) in THF. The molecular weight was found equal to 2100 g/mol. The polydispersity index Mw/Mn was equal to 2.0. The styrene conversion was determined from gas chromatography (GC) and was determined to be 99%. This sample of the first block was further used within the study as a reference point. (ii) A diblock copolymer polystyrene-block-poly(ethyl acrylate) of molecular weight (M n) target 2000-block-42 000 g/mol through the above first block extension with the second block of poly(ethyl acrylate) of theoretical molecular weight Mn=42 000 g/mol (actually the second block is a statistical copolymer of ethyl acrylate (EA) and methacrylic acid (MAA) with mass ratio EA/MAA = 98/2). The product obtained is a white dispersion in water (latex) of dry extract of about 40%. The molecular weight was found equal to 41 000 g/mol. The polydispersity index Mw/Mn was equal to 3.2. The ethyl acrylate conversion was determined to be 99.5%. This sample of the diblock copolymer was further used within the study as a reference point. DOI: 10.1021/la902661m
1701
Article
Anik et al. Table 1. Composition of the Electrolytes Used for the pH Study at Constant Ionic Strength
buffer 1 2 3 4 5 6 7 8 9 10 11
NaH2PO4 (mM)
Na2HPO4 (mM)
38 10.6 1.3
3.8 10.6 13
B4Na2O7 BORAX (mM)
13.25 13.25 7.25 1.3 2.7 8.2 10.9 13.6
HBO2 (mM)
NaOH (mM)
ionic strength (mM)
pH
12 2.3 4.5 13.6 18.2 22.7
26.6 26.5 26.6 26.5 26.5 26.5 5 10 30 40 50
6.02 7.02 7.9 8.33 9.42 10.1 10.1 10.1 10.1 10.1 10.1
119.25
Scheme 1. Schematic Representation of the Structure of the Studied Latexes
(iii) The diblocks of poly(styrene)-block-poly(ethyl acrylate-r-acrylic acid sodium salt) were obtained through a partial hydrolysis with sodium hydroxide of the poly(ethyl acrylate) block of the diblock copolymer described above (target 90%, 75%, 60%, and 40% hydrolysis degree of ethyl ester groups). The products recovered at the end of the reactions were translucent gels in water/isopropanol/ethanol solution of dry extract of ∼20%. (iv) A polymer poly(EA-r-2% MAA) (mass %) of target Mn= 42 000 g/mol was synthesized in order to be able to have a reference point. The molecular weight of the synthesized polymer was found equal to 39 500 g/mol. The polydispersity index Mw/Mn was equal to 2.1. The ethyl acrylate conversion was determined from GC and was determined to be 99%. Conversion Determination. The global conversion rate was followed by percent solids measurements using a moisture balance (Mettler PM 480/LP 16). The conversion rates of controlling agent Rhodixan A1, S, and EA monomers have been measured with a GC system (Hewlett-Packard GC 5890 device equipped with head space sampler 7694 and a FID detection system). The GC column was a ZorBax-1 (60 m 0.53 mm) with a flow of helium of 1 mL/ min. The temperature of the sample was fixed at 105 C. The temperature of the oven was put at 60 C for 5 min, then increased with a ramp of 15 C/min, and kept at 250 C for 5 min more. 1702 DOI: 10.1021/la902661m
Average Molecular Weight Determination. Average molecular weight distributions of PS and PS-PEA were determined by SEC (Agilent 1100 series). A pump, an autosampler, a refractive index detector, and a UV detector at 290 nm were mounted in series. A set of three columns from Polymer Laboratories PLgel Mixed-B (particle size 8-10 μm, range of Mw=500-10 000 000 g/ mol) were used in series with THF as a mobile phase at a temperature of 40 C and a flow rate of 0.75 mL/min. Polystyrene standards (Polymer Laboratories EasiCal PS-2) were used for calibration. 2.3. Capillary Electrophoresis. CE experiments were carried out with an 3D-CE Agilent Technologies system (Waldbronn, Germany) equipped with a diode array detector. Separation capillaries prepared from bare silica tubing were purchased from Composite Metal Services (Worcester, UK). Capillary lengths were 33.5 cm (25 cm to the detector) 50 μm for all analysis unless otherwise specified. New capillaries were conditioned by performing the following washes: 1 M NaOH for 20 min, 0.1 M NaOH for 15 min, and water for 10 min. Between two runs, the capillary was flushed with electrolyte (5 min, 930 mbar). Table 1 describes the composition of the different electrolytes (phosphate, borate buffers) used in this work. The temperature of the capillary cassette was maintained constant at 25 C. The polymer samples were prepared in deionized water. Langmuir 2010, 26(3), 1700–1706
Anik et al.
Article
Figure 1. Electropherograms in effective mobility scale (A) and variation of the electrophoretic mobility (B) of PS, PEA, and PS-PEA
latexes. Electrophoretic conditions: fused-silica capillary, 50 μm i.d. 33.5 cm (effective length 25 cm). Electrolytes: see Table 1; pH is indicated in the figure. Applied voltage: +25 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Samples: 1, PS latex at 0.1% in water; 2, PS-PEA latex at 0.5% in water; 3, PEA latex at 0.5% in water. Direct UV detection at 194 nm. The bars in (B) represent ( the standard deviation of the electrophoretic mobility centered on the average value.
Figure 2. Electropherograms of PS, PEA, and PS-PEA latexes obtained at pH 10.1 at different ionic strengths. Other conditions as in Figure 1.
Sample volumes of ∼4 nL were injected hydrodynamically (17 mbar, 5 s). The prepunchers and electrodes were cleaned every 2 days to remove solid deposits. Data were collected at 194 nm. Electroosmotic mobility was calculated from the migration time of mesityl oxide (neutral marker). Electropherograms were plotted in effective mobility scale (μep) using the following equation: μep ¼ μapp - μeo ¼
lL lL Vtapp Vteo
ð2Þ
where l is the effective length up to the detection point, L the total capillary length, V the applied voltage, tapp the apparent detection time, and teo the detection time of the neutral marker. 2.4. Dynamic Light Scattering. Dynamic light scattering experiments were carried out on a multiangle static and dynamic light scattering setup from ALV Gmbh, Germany (ALV CGS-3). Autocorrelation functions of diluted latex dispersions were acquired with a multitau ALV 5003 correlator. Dilution was performed in deionized water (in two steps). Intensity autocorrelation functions were recorded at 90 scattering angle and 298.2 K temperature in three runs of 15 s each. Data were treated according to the cumulants methods at the second order implemented on the DLS ALV 3.0 software. This software allows the determination of a polydispersity index (IP) which should be distinguished from the polydispersity index calculated by SEC. Langmuir 2010, 26(3), 1700–1706
Table 2. Values of Average Electrophoretic Mobility, Reduced Mobility, Reduced Zeta Potential, KR Product, and Zeta Potential for the Three Latexes at Different pHa
sample
pH
Æμepæ (10-5 cm2 V-1 s-1)
E
ζ-
κR
ζ (mV)
PS -43.42 3.25 3.48 10.18 -89.5 PS-PEA 6.02 -42.34 3.17 2.57 26.78 -66.1 PEA -43.30 3.24 2.59 29.46 -66.6 PS -42.84 3.20 3.52 10.16 -90.5 PS-PEA 7.02 -53.68 4.01 3.52 26.73 -90.5 PEA -52.82 3.95 3.39 29.40 -87.2 PS -48.79 3.65 4.71 10.16 -121 PS-PEA 7.9 -59.49 4.45 4.55 26.73 -117 PEA -56.58 4.23 3.87 29.40 -99.5 PS -44.49 3.33 3.94 10.18 -101 PS-PEA 8.33 -52.72 3.94 3.37 26.78 -86.7 PEA -50.19 3.75 3.16 29.46 -81.3 PS -45.64 3.41 4.44 10.16 -114 PS-PEA 9.42 -51.74 3.87 3.33 26.73 -85.6 PEA -50.81 3.80 3.26 29.40 -83.8 PS -45.07 3.37 3.87 10.16 -99.5 PS-PEA 10.1 -52.85 3.95 3.44 26.73 -88.5 PEA -49.60 3.71 3.2 29.40 -82.3 a Electrophoretic conditions: as in Figure 1. Physical data used: kB = 1.38 10-23 J K-1, T = 298 K, ε0 = 8.85 10-12 F/m, εwater = 78.5; e = 1.60 10-19 C.
DOI: 10.1021/la902661m
1703
Article
Anik et al.
Figure 3. Influence of neutral surfactant concentration on latex electrophoretic mobilities. Electrophoretic conditions: fused-silica capillary, 50 μm i.d. 50.5 cm (effective length, 41.5 cm). Electrolyte: 53 mM borate buffer þ Brij 58 at the indicated concentration, pH = 10.1. Applied voltage: þ20 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Samples: 1: PS latex at 0.1% in water; 2: PS-PEA latex at 0.5% in water; 3: PEA latex at 0.5% in water. UV detection at 194 nm. This IP is derived from the cumulant expansion of the field correlation function. It is zero for monodisperse particle size distributions and increases to values typically higher than 0.3 for polydisperse distributions.
3. Results and Discussion 3.1. Presentation of Studied Latexes. Scheme 1 presents the different steps of the synthesis of poly(styrene-co-ethyl acrylate/acrylic acid) gel (PS-PEA/AA) as well as the schematic representation of this structure in solution. Starting from a polystyrene latex (Mn =∼2 103 g/mol, particles polydispersity index IP=0.23), the copolymerization with ethyl acrylate leads to a poly(styrene-b-ethyl acrylate) latex, with a particle polydispersity index of 0.22. Poly(ethyl acrylate) block chains are then partially converted into poly(acrylic acid) (PAA) sodium salt by a base hydrolysis reaction. This PS-PEA/AA block copolymer self-assembles in aqueous solution to form a system where the core is composed of the short PS block (diameter ∼ 40 nm) and the corona consists of the poly(acrylate ethyl/acrylic acid sodium salt) water-soluble polymer chains. 3.2. Influence of the pH and the Ionic Strength on the Latex Separation. The first part of this work deals with the study of the electrophoretic behavior of PS, PEA, and PS-PEA latexes that are likely to be present in the synthesized of PS-PEA/ AA gels. For that, the influence of the pH and the ionic strength on the separation was studied to find the optimal conditions for the separation of these different latexes. Six different electrolytes (see Table 1) at constant ionic strength with pH ranging from 6.02 to 10.1 were compared. For pH ranging from 6.02 to 7.9, the background electrolyte (BGE) is a phosphate buffer, whereas for pH ranging from 8.33 to 10.1, BGE was a borate buffer. For a better clarity, Figure 1A only displays the effective mobilityscaled electropherograms obtained for the three latexes (PS, PS-PEA, and PEA) at pH 6.0, 7.9, and 10.1. The variation of the effective mobility with pH is shown in Figure 1B. For pH < 8, the electrophoretic mobility increases with the pH due to higher ionization of the carboxylate functions. Above pH 8, the mobility was found to be almost constant. Differences in electrophoretic mobility observed between borate and phosphate buffers can be explained by a preferential adsorption of phosphate ions onto the surface of the latexes. The best separation between the three latexes was obtained at pH 7.9 or 10.1. Apart from pH 6, where the ionization is not complete, the electrophoretic mobilities (in absolute values) are in the range μPS < μPEA< μPS-PEA. 1704 DOI: 10.1021/la902661m
Figure 4. Electropherograms of PS, PS-PEA, and PEA. Electro-
phoretic conditions: fused-silica capillary, 50 μm i.d. 50.5 cm (effective length, 41.5 cm). Electrolyte: 53 mM borate buffer, pH = 10.1 þ 1 mM Brij 58. Applied voltage: þ20 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Samples: 1: PS latex at 0.1% in water; 2: PS-PEA latex at 0.5% in water; 3: PEA latex at 0.5% in water. Direct UV detection at 194 nm.
The peak dispersion in Figure 1A should be representative of the latex polydispersity (in size or in charge density). This peak dispersion is reported in Figure 1B via the error bars that represent (1 standard deviation centered on the average mobility. Better resolution between the latexes also leads to broader peaks due to the increase in selectivity of the separation. Figure 2A displays the superposition of the electropherograms obtained for the three latexes at pH = 10.1 and for different ionic strengths ranging from 5 to 50 mM (see Table 1, electrolytes 6-11). For ionic strength higher than 50 mM, a decrease of the current intensity during electrophoresis was observed and was likely due to a flocculation of the latexes inside the capillary. Figure 2A,B indicates that effective mobility (in absolute value) increases with the ionic strength as generally observed for latexes separated by CE at classical ionic strengths (5-100 mM).24,38 As discussed for the influence of the pH, better selectivity at higher ionic strength leads to higher peak broadening. The separation between the latexes is optimal for an ionic strength of 25-30 mM. (38) Kobayashi, M. Colloid Polym. Sci. 2008, 286, 935–940.
Langmuir 2010, 26(3), 1700–1706
Anik et al.
Article
Table 3. Comparison of Sensitivity of Detection and Limits of Detection for the Three Latexes in Different Experimental Conditionsa PS electrolyte
dc (μm)
ΔP0 t0 (mbar s)
b
sensitivity (mAu L g-1)
PS-PEA LOD (g/L)
b
sensitivity (mAu L g-1)
PEA LOD (g/L)
sensitivityb (mAu L g-1)
LOD (g/L)
A. BGE þ Brij 58 1 mM 50 17 5 360 0.01 24 0.10 6.6 0.37 B. BGE þ Brij 58 10 mM 50 17 5 90 0.34 19 0.71 5.7 1.22 C. BGE þ Brij 78 10 mM 100 17 5 965 0.07 84 0.56 28 0.26 a Electrophoretic conditions: fused-silica capillary, 50 μm i.d. 50 cm (effective length, 41.5 cm) (A and B); 100 μm i.d. 70 cm (effective length, 61.5 cm) (C). Electrolyte: 53 mM borate buffer þ 1 mM Brij 58 (A), 10 mM Brij 58 (B), 10 mM Brij 78 (C). Applied voltage: þ20 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Samples: 1: latex PS, C = 0.05% to 0.25% in water (A and B), C = 0.01% to 0.09%, (C) 2: latex PS-PEA at 0.5% in water, 3: PEA latex at 0.5% in water. Direct UV detection at 194 nm. b The sensitivity of detection is defined as the ratio of the time-corrected area to the particle mass concentration.
3.3. Determination of Latex Zeta Potentials. Zeta potential of the latexes were determined from the average effective mobilities using the numerical values given by the E(ζh,kR) plot from O’Brien and White theory (see Figure 4 in ref 39). Knowing the effective mobility of the latex, it is possible to determine the reduced mobility defined as E ¼
3ηe μ 2ε0 εr kB T ep
ð3Þ
where η is the solvent viscosity, e is the elementary charge, ε0 is the vacuum permittivity, εr is the dielectric constant of water, kB is the Boltzmann constant, T is the temperature, and μep is the electrophoretic mobility. Knowing the hydrodynamic radius and the ionic strength, one can determine the κR value. Finally, using the E(ζh,kR) plot from O’Brien and White theory, the reduced zeta potential is obtained. Zeta potential values are derived from eq 4: ζ ¼
eζ kB T
ð4Þ
The zeta potential values of the latexes are gathered in Table 2 with the average effective mobility, the reduced mobility, the reduced zeta potential, and the κR values. Zeta potentials vary between -66 and -120 mV and are in the order of ζPEA < ζPS-PEA < ζPS in absolute value. Zeta potential values of PS are in agreement with values usually reported in the literature.24 It is worth noting that this order does not strictly follow the order of the effective mobility. This is due to the much lower size of the PS latex that leads to different κR values (and different curves in the E(ζh,κR) plot from O’Brien and White theory). The zeta potential of a latex is generally controlled by the adsorption of surfactant40 (sodium dodecyl sulfate, SDS), the methacrylate ion content incorporated during emulsion polymerization,41 and the preferential adsorption of ions onto the surface.42 3.4. Influence of the Addition of Surfactant on the Separation. To improve the selectivity of separation of the three latexes, the influence of the addition of a neutral surfactant in the BGE was studied. The surfactant used was the poly(ethylene glycol hexadecyl ether) (Brij 58). The influence of surfactant concentration (from 1 10-3 to 10 mM) was studied at 26.5 mM ionic strength and pH 10.1. For a better clarity, only three electropherograms obtained at 2.5 10-3, 5 10-2, and (39) O’Brien, R. W.; White, L. R. J. Chem. Soc., Faraday Trans. 1978, 274, 1607. (40) Sefcik, J.; Verduyn, M.; Storti, G.; Morbidelli, M. Langmuir 2003, 19, 4778– 4783. (41) Davies, M. C.; Lynn, R. A. P.; Hearn, J.; Paul, A. J.; Vickermann, J. C.; Watts, J. F. Langmuir 1996, 12, 3866–3875. (42) Elimelech, M.; O’Melia, C. R. Colloids Surf. 1990, 44, 165–178.
Langmuir 2010, 26(3), 1700–1706
Figure 5. PS-PEA-AA gel characterization in optimal experimental conditions using a 100 μm i.d. capillary. Electrophoretic conditions: fused-silica capillary, 100 μm i.d. 50 cm (effective length, 41.5 cm). Electrolyte: 53 mM borate buffer þ 10 mM Brij 78, pH = 10.1. Applied voltage: þ20 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Samples: 1: PS latex at 0.1% in water; 2: PS-PEA latex at 0.5% in water; 3: PEA latex at 0.5% in water; 4: PS-PEA-AA gel at 0.5% in water. Direct UV detection at 194 nm.
1 mM (cmc ∼ 10-2 mM43) are displayed in Figure 3A. The addition of surfactant had a great influence on the separation even below the cmc (see Figure 3A). For example, the electrophoretic mobility of the PS latex measured at the apex of the peak is 45.76 10-5 cm2 V-1 s-1 in free surfactant borate buffer and decreases (in absolute value) compared with -9 10-5 cm2 V-1 s-1 after addition of 1 10-3 mM Brij 58. This decrease of the electrophoretic mobility is due to the adsorption of surfactant unimers which tend to reduce the particle charge density. Figure 3B displays the variation of electrophoretic mobility of the three latexes as a function of the concentration in neutral surfactant. The rapid decrease of the mobility below the cmc followed by a plateau confirms the interaction between latexes and surfactant unimers. The bar on each data point represents the dispersion in mobility of the peak ((1 standard deviation centered on the average mobility value). Good separations of the three latexes are obtained by the addition of Brij 58 at a concentration ranging from 5 10-2 to 10 mM. The optimal electrolyte was set at 53 mM borate buffer, pH 10.1, and 1 mM Brij 58. The corresponding separation is given in Figure 4 for the three latexes injected individually and for the mixture. The injection of the mixture confirms the absence of interaction between latexes. The limits of detection, defined as the concentration leading to a signal (peak height) that is 3 times the noise (43) Helenius, A.; Simons, K. Biochim. Biophys. Acta 1975, 415, 29–79.
DOI: 10.1021/la902661m
1705
Article
Figure 6. Electropherograms obtained for gels at different hydrolysis rate. Electrophoretic conditions: fused-silica capillary, 100 μm i.d. 50 cm (effective length, 41.5 cm). Electrolyte: 53 mM borate buffer þ 10 mM Brij 78, pH = 10.1. Applied voltage: þ20 kV. Hydrodynamic injection: 17 mbar, 5 s. Temperature: 25 C. Direct UV detection at 194 nm.
(standard deviation of the background in the absence of signal), for the PS, PS-PEA, and PEA latexes are respectively of 0.01, 0.1, and 0.37 g/L (see Table 3). The sensitivity of detection follows the order PS > PS-PEA > PEA since the styrene moieties absorb much more than the ethyl acrylate groups at 194 nm. 3.5. Analysis of the PS-PEA-AA Gel. The analysis of the PS-PEA/AA gel obtained after partial hydrolysis (f = 75% in mol) of PS-PEA latexes was tried without success (not shown) under the previously described optimal conditions. Despite the presence of 1 mM Brij 58 in both sample and electrolyte, the viscosity of the sample was too high (40 10-2 Pa s). The sample viscosity is due to hydrophobic interactions of the PS-PEA/AA copolymers in water. Indeed, the Brij 58 concentration had to be increased in the electrolyte and in the sample up to 10 mM to reduce the viscosity (11.2 10-2 Pa s) and permit the analysis (not shown). The drawback of the increase in surfactant concentration is the drop in signal-to-noise ratio and an increase of the limits of detection (0.34, 0.71, and 1.22 g/L respectively for the PS, PS-PEA, and PEA latexes; see Table 3). An attempt to reduce the surfactant concentration keeping low viscosity was performed by replacing the Brij 58 by the Brij 78 (Brij 78 has a C18 chain with the same number of oxide ethylene units as compared to Brij 58). However, the same concentration (10 mM) was required to reduce the viscosity (15 10-2 Pa s). Therefore, similar results were obtained using both Brij 58 or 78. PS-PEA-AA gels present higher mobilities than those of the three latexes. This can be explained since the PS-PEA/AA gels are obtained after hydrolysis of the PS-PEA latexes increasing the content of ionized moieties via the formation of ionized AA groups.
1706 DOI: 10.1021/la902661m
Anik et al.
To improve the sensitivity of detection for the quantification of PS, PS-PEA, and PEA latexes in PS-PEA/AA gel samples, a 100 μm inner diameter capillary was used with the electrolyte composed of 53 mM borate buffer and 10 mM Brij 78 and keeping constant the injection conditions (ΔP tinj). Separations realized under these conditions are presented in Figure 5 . If one compares the sensitivity of detection of the 50 and 100 μm capillaries, one would expect an increase by a factor 8 (factor 2 on the absorbance-capillary diameter and factor 4 on the injected length at constant ΔP tinj). Experimentally, the improvement of the sensitivity of detection is in the same order of magnitude (11 for PS, 4 for PS-PEA, and 5 for PEA). The corresponding limits of detection are respectively 0.07, 0.56, and 0.26 g/L for the PS, PS-PEA, and PEA. Within these limits of detection, the PS-PEA/AA gel does not contain any PS, PS-PEA, and PEA latexes. Represented in mass concentration relative to the PS-PEA/AA gel, the maximum latexes concentrations in the gel could be 1.5% of PS, 5.2% of PEA, and 11.2% of PS-PEA. It is likely that PS-PEA/AA gels are out-of-equilibrium systems, the rheological properties of which depend on the process. We did not investigate the ability of CE to analyze different systems with different rheological properties; however the drastic decrease of viscosity brought by surfactant addition suggests that this strategy is powerful enough to handle different rheologies. 3.6. Separation of PS-PEA/AA Gels According to the Hydrolysis Rate. With the aim to study the selectivity of separation according to hydrolysis rate (f), different PS-PEA/ AA gels have been synthesized at different values of f ranging from 40 to 90%. As expected, experimental results obtained in a 100 μm i.d. capillary and with 10 mM Brij 78 in the BGE indicate that the electrophoretic mobility increases with the hydrolysis rate (see Figure 6 and the inset). Interestingly, due to this selectivity on the hydrolysis rate, the peak dispersion should be directly related to the heterogeneity of AA content in the gel. For instance, the 90% hydrolyzed gel does not contain gels with hydrolysis rate below 75%. In addition, the large peak indicated with an arrow detected in front of the main 90% hydrolyzed peak could be attributed to the presence of completely hydrolyzed PEA chains.
4. Conclusion This work demonstrates the interest of CE for the characterization of block copolymer latexes. It has been possible to separate the moieties formed at the different stages of the synthesis (polymer particles and gels). Within the limits of detection, no PS, PS-PEA, and PEA latexes were detected in the gel samples of PS-PEA/AA. Because of high viscosity of the gels (hydrolyzed latexes), the addition of surfactant was necessary for their analysis. One of the main interests of this separation technique is the possibility to obtain information on the charge density distribution of the latexes/gels and the absence of stationary phase that limit undesirable interactions.
Langmuir 2010, 26(3), 1700–1706
