Surface Potential of DPPC Monolayers on Concentrated Aqueous Salt

Article pubs.acs.org/JPCB

Surface Potential of DPPC Monolayers on Concentrated Aqueous Salt Solutions Clayton B. Casper,† Dominique Verreault, Ellen M. Adams, Wei Hua, and Heather C. Allen* Department of Chemistry and Biochemistry, The Ohio State University, 100 West 18th Avenue, Columbus, Ohio 43210, United States S Supporting Information *

ABSTRACT: The presence and exchange of electrical charges on the surfaces of marine aerosols influence their ability to act as cloud condensation nuclei and play a role in thundercloud electrification. Although interactions exist between surfaceactive inorganic ions and organic compounds, their role in surface charging of marine aerosols is not well understood. In this study, the surface potential of dipalmitoylphosphatidylcholine (DPPC) monolayers, a zwitterionic phospholipid found in the sea surface microlayer, is measured on concentrated (0.3−2.0 M) chloride salt solutions containing marine-relevant cations (Na+, K+, Mg2+, Ca2+) to model and elucidate the electrical properties of organic-covered marine aerosols. Monovalent cations show only a weak effect on the surface potential of DPPC monolayers in the condensed phase compared to water. In contrast, Mg2+ and Ca2+ increase the surface potential, indicating different cation binding modes and affinities for the PC headgroup. Moreover, it is found that for divalent chloride salt solutions, the PC headgroup and interfacial water molecules make the largest dipolar contribution to the surface potential. This study shows that for equal charge concentrations, divalent cations impact surface potential of DPPC monolayers more strongly than monovalents likely through changes in the PC headgroup orientation induced by their complexation along with the lesser ordering of interfacial water molecules caused by phosphate group charge screening.

■

the surface.7−9 These compounds include, among others, simple species such as phospholipids, fatty acids, amino acids, and sterols but also more complex macromolecules such as polysaccharides, polypeptides, and proteins, as well as humiclike substances. Due to their amphiphilic character, some of these compounds like phospholipids are surface-active, i.e., readily adsorb at the aqueous surface to form monolayers. Comprising a large portion of cellular membranes of marine organisms, saturated phospholipids with 14−18 carbons alkyl chains like dipalmitoylphosphatidylcholine (DPPC) are effectively prevalent in the SSML10,11 and are believed to be present in the organic layer of marine aerosols.12 The physicochemical characterization of native and reconstructed samples from the SSML on pure water and seawater has traditionally been done using standard monolayer techniques (surface pressure− and surface potential−area isotherms) as well as Brewster angle microscopy (BAM) and electrochemical and nonlinear spectroscopy methods.13−17 However, to obtain a fundamental understanding of the surface properties including intermolecular interactions and molecular organization of the SSML, it has been customary to provide

INTRODUCTION Marine aerosols, which make up the largest portion of the global tropospheric aerosol production, have been found to play critical roles in atmospheric chemistry, including the biogeochemical cycling of nutrients (e.g., C, N, and S), and in global climate change through their direct and indirect effects on cloud condensation nuclei (CCN), radiative balance, and levels of precipitation.1 In addition, through the development and exchange of electrical charges at their surfaces, which also influence their ability to act as CCN, aerosols have been suggested to play some role in thundercloud electrification.2 Primary marine aerosols are produced from wind-driven wave action at the sea surface through the generation of sea sprays.3,4 Through this process newly formed marine aerosols become enriched in organic matter found in the sea surface microlayer (SSML), the thin (1−1000 μm) organic layer present at the surface of seawater where many bio- and physicochemical processes as well as gas exchange take place.5 Depending on the extent of these processes, marine aerosols can have a vast range of salt and organic concentrations, with larger aerosols primarily composed of sea salts while smaller ones are highly enriched in organics.6 Although there is some variability as to the composition of the SSML, it is generally accepted that it is a complex mixture of organic compounds coming from living and decaying marine microorganisms (phytoplankton, bacteria) that have risen to © 2016 American Chemical Society

Special Issue: Bruce C. Garrett Festschrift Received: October 26, 2015 Revised: December 30, 2015 Published: January 13, 2016 2043

DOI: 10.1021/acs.jpcb.5b10483 J. Phys. Chem. B 2016, 120, 2043−2052

Article

The Journal of Physical Chemistry B

obtained from Fisher Scientific (Waltham, MA). 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) was purchased from Avanti Polar Lipids (Alabaster, AL). Ultrapure water with a resistivity of 18.2−18.3 MΩ·cm and pH ≈ 5.6 (acidic due to dissolved atmospheric CO2) was obtained from a Barnstead Nanopure system (D4741, Thermolyne Corporation, Dubuque, IA) with additional organic removing cartridges (D5026 type I ORGANICfree cartridge kit; pretreat feed). Lipid and Salt Solutions Preparation. DPPC stock solutions were prepared by dissolution in chloroform. Salts were dissolved in ultrapure water and then filtered 2−4 times using activated carbon filters (Whatman Carbon Cap 75, Fisher Scientific) in order to remove any organic impurities. The filtration of salt solutions was required, as it has been shown that solutions made with ACS grade salts (≥99% purity) still contain trace amounts of surface-active organic impurities.38 The filtered solutions were then checked for impurities using vibrational sum frequency generation (VSFG) spectroscopy in the C−H stretching region (2800−3000 cm−1). The VSFG spectra revealed no peaks associated with C−H stretching modes of organic impurities. Concentrations of these salt solutions were then standardized using the Mohr titration method.39 Finally, each salt stock solution was diluted to the following concentrations: 0.6 (0.3 for divalent cations), 1.0, and 2.0 M. The lower concentration was chosen in order to emulate the typical salt concentration in seawater.40 Brewster Angle Microscopy. BAM images were recorded simultaneously with surface pressure (Π)−molecular area (A) compression isotherms using a custom-built Brewster angle microscope that has been previously described.41,42 Briefly, the output of a 543 nm p-polarized laser (Research Electro-Optics, Boulder, CO) is reflected off the aqueous surface at an appropriate Brewster angle for each solution. The reflected beam is collected and collimated by a 10× infinity-corrected superlong working distance objective (CFI 60 TU Plan EPI, Nikon, Melville, NY) and tube lens (MXA22018 CFI, Nikon). Images were recorded by a back-illuminated EM-CCD (iXon DV887-BV, Andor Technology, South Windsor, CT) using the Andor Solis software. A movie of the monolayer during compression was recorded, in which an image was taken approximately every 5 s. Still frames were then extracted from the movie for further analysis. Due to the inclination of the collection optics, only a narrow region in the image is focused. All BAM images were therefore cropped to show the most resolved region, which may differ from image to image. Scale bars are accurate for each specific image. BAM images and compression isotherms were recorded at room temperature (23 ± 1 °C) in a climate-controlled environment and were repeated at least three times to ensure reproducibility. Standard deviations of the measured mean molecular area (MMA) and surface pressure were ±0.01 nm2 and ±1 mN/m, respectively. Thickness (d) of the DPPC monolayer was determined using the relation43

measurements of model compounds relevant to the SSML as comparison. One proven experimental approach is to use Langmuir monolayers of pure phospholipids like DPPC on pure water or on simple aqueous salt solutions as simplified model systems for the SSML and the organic layer of marine aerosols, on which environmental conditions can be easily varied.18−20 For instance, the influence of a salt like NaCl on the alkyl chains ordering and orientation, headgroup orientation, phase behavior, compressibility, and domain morphology of DPPC monolayers has previously been extensively studied using surface pressure−area isotherms.21−27 More recently, surface-sensitive vibrational sum frequency generation (VSFG) spectroscopy has provided further molecular details on the effect of Na+ cation binding on the PC headgroup orientation.28 Yet, to the authors’ knowledge, there have been very few experimental studies examining the impact of Na+ and other important cations found in the marine environment such as K+, Mg2+, and Ca2+, at moderate and high concentrations (>0.3 M).28−32 Surface potentiometry has also been used as a complementary technique to study the electrical properties of bare and lipid-covered aqueous interfaces. The surface potential, most often measured using an ionizing electrode (IE)33,34 or a vibrating plate (VP),35,36 provides a direct measurement of the electric field across the interface, through which electrostatic and structural behavior at the surface can be inferred. Surface potential measurements of SSML samples in a compressed state on water have been previously reported;13−16 however, the magnitude of the observed surface potentials could not be easily resolved in terms of known surface potentials of model compounds. Nakahara and co-workers have recently measured the surface potential of DPPC monolayers on 0.5 M NaCl solution using IE surface potentiometry, but the effects of the metal cation binding to the lipid headgroup were not explored. 37 A comparative surface potential study of phospholipids commonly found in the SSML on different marine-relevant aqueous salt solutions at moderate to high concentrations could give some valuable insight into the electrical properties of marine aerosols. In addition, differences in measured surface potential could help gain information concerning the cation interactions with the DPPC monolayer and surrounding water molecules. In this work, the surface potentials of model zwitterionic DPPC monolayers on concentrated aqueous chloride solutions of alkali (Na+, K+) and alkaline earth (Mg2+, Ca2+) cations prevalent in the marine environment are investigated using VP surface potentiometry. The combined contributions of the DPPC polar headgroup and oriented interfacial water were estimated by subtracting from the overall measured surface potential both the EDL contribution predicted by the Gouy− Chapman model and the DPPC chain contribution experimentally determined from BAM. We show that monovalent and divalent cations contribute to different extents to the overall surface potential, indicating that they interact differently with the DPPC monolayer. This difference can have a significant impact on the electrical properties of the organic boundary between marine aerosols and the atmosphere.

d=

■

λ Rp π sin(2θB − 90)

nm 2(na 2 − ns 2) na 2 + ns 2 (na 2 − nm 2)(ns 2 − nm 2) (1)

EXPERIMENTAL DETAILS Materials. NaCl (≥99%, ACS certified crystalline), KCl (≥99%, ACS certified crystalline), MgCl2·6H2O (≥99%, ACS certified crystalline), CaCl2·2H2O (≥99%, ACS certified crystalline), and chloroform (HPLC grade, ≥99%) were

where Rp is the reflectance of the p-polarized light at the Brewster angle (θB), λ is the laser wavelength, and na, ns, and nm are the refractive indices of air, the aqueous solution, and the monolayer, respectively. The value of ns was determined from BAM measurements on the bare aqueous solutions, whereas 2044

DOI: 10.1021/acs.jpcb.5b10483 J. Phys. Chem. B 2016, 120, 2043−2052

Article

The Journal of Physical Chemistry B

Figure 1. BAM images of DPPC monolayers in the LC phase (45 mN/m) spread on NaCl and CaCl2 aqueous solutions at different concentrations. An image of DPPC on water is given as a reference. BAM images were cropped to show the most resolved region, which may differ from image to image. Scale bars (50 μm) are accurate for each specific image.

m); and another 4.6 μL for the LC phase (0.41 nm2, ∼41 mN/ m). After each spreading, a period of 10 min was allowed to let the chloroform evaporate. For each DPPC phase, the surface potential was recorded for 5 min before proceeding to the next. After each set of experiments, the cell and counter electrode were cleaned thoroughly twice with ethanol (reagent alcohol, Fisher Scientific) and 8 times with ultrapure water. All measurements were recorded at ambient temperature and were repeated at least three times to ensure reproducibility.

the value nm = 1.49 was taken from ellipsometry measurements of DPPC monolayer on water.44 Calibration curves were created to relate the reflectance to the gray level of the camera following methods previously described.45,46 From the cropped BAM images, gray levels were averaged across three different horizontal lines spanning the entire image. A final gray level for each image was determined by averaging the gray levels of the lines and then converted to a reflectance value based on the calibration curves. Surface Potential Measurements. Static surface potential measurements of the DPPC/aqueous interfaces were performed using a surface potential sensor (SPOT I, Biolin Scientific USA, Linthicum Heights, MD), which is based on the VP (Kelvin) method.33 The surface potentiometer setup was shielded from electrostatic noise by a custom-built Faraday cage, which was enclosed in a Plexiglas box in order to minimize disturbances from air currents and dust particles. The relative humidity of the enclosure was controlled by a large Petri dish containing hot water and varied from 40% to 70%. The aqueous solutions were contained in a clean Teflon dish (88 mm × 44 mm × 10 mm) placed on top of a vertical translation stage to ensure reproducible air gap distances between the aqueous surface and the VP electrode. In each experiment, the sample cell was filled with ultrapure water and the surface potential was recorded for 10 min. This surface potential value was taken as the reference (ΔV0). In the case of aqueous salt solutions, the water reference was removed, the surface potential was zeroed, and then the water was replaced with the salt solution. The surface potential of the bare salt aqueous interface was then recorded for 10 min. The monolayer was added to the interface by spreading a definite volume of DPPC solution using a clean 10 μL glass microsyringe (701N, Hamilton, Reno, NV). In order to measure the surface potential over each DPPC phase, sequential additions of 0.96 mM DPPC were made: 7.9 μL for the liquid-expanded phase (LE, 0.81 nm2, ∼5 mN/m); an additional 2.4 μL aliquot for the liquid-expanded−liquidcondensed coexistence region (LE-LC, 0.62 nm2, ∼8 mN/

■

RESULTS AND DISCUSSION BAM Images of DPPC Monolayers on Chloride Salt Solutions. BAM images of DPPC monolayers in the LC phase (45 mN/m) on water, and NaCl and CaCl2 aqueous solutions are shown in Figure 1. In the LC phase DPPC molecules interact strongly and are highly oriented, causing formation of a fully condensed film. A bright image with uniform intensity is observed in the LC phase for DPPC on water, consistent with previous studies.47 Relative to water, no appreciable change is observed in intensity or morphology for 0.6 M NaCl. However, a decrease in intensity is observed for 2.0 M NaCl. A similar trend is observed for CaCl2, where the intensity of images decreases with increasing salt concentration. Darkening of the images cannot be attributed to changes in reflectance due to deviations from the Brewster angle, as the incident angle was adjusted to the Brewster angle of each respective aqueous solution. Therefore, the lower image intensity comes from changes in monolayer thickness and/or refractive index. Monolayer thickness and refractive index may be determined from BAM images but not simultaneously. Here, to determine the change in monolayer thickness, it was assumed that the refractive index of the DPPC monolayer remained constant (nm = 1.49). The thickness of the DPPC monolayer in the LC phase (45 mN/m) on various chloride salt solutions is given in Table 1. On water DPPC was found to have a thickness of 2.48 ± 0.37 nm, consistent with a reported value (2.53 ± 0.04 nm) from X-ray reflectivity experiments.48 As for salt solutions, 2045

DOI: 10.1021/acs.jpcb.5b10483 J. Phys. Chem. B 2016, 120, 2043−2052

Article

The Journal of Physical Chemistry B

determined at each concentration. Molecular dynamics (MD) simulations have found that Na+ ions bind with DPPC molecules while K+ does not.49 Compression isotherms and BAM measurements, however, suggest that these monovalent ions interact similarly with DPPC. In contrast, a comparison of thickness values for MgCl2 and CaCl2 solutions at the same concentration shows that above 0.3 M the thickness is dependent upon the cation identity. Surface Potential of DPPC Monolayers on Monovalent Chloride Salt Solutions. The measured surface potentials of DPPC monolayers in different phases on pure water and on selected aqueous alkali (NaCl, KCl) and alkaline earth (MgCl2, CaCl2) chloride salt solutions are shown in Figure 2. Whether spread on water or on salt solutions, DPPC monolayers show a monotonic increase of surface potential with decreasing mean molecular area. The surface potential value obtained for DPPC monolayers in the LC phase on water (595 ± 13 mV; see Table S1 of the Supporting Information) agrees well with those from literature (∼525−585 mV).50−53 In the cases of NaCl and KCl solutions, the surface potential moderately increases (∼20−80 mV) with salt concentration for the noncondensed phases (LE and LE-LC) but drops by 30− 60 mV for the LC phase when compared to DPPC monolayers on water (Figure 2a,b). Further comparison of surface potential values between these solutions also indicates that Na+ and K+ cations have comparable effect on the DPPC monolayers. Because the surface potential is a measure of the vertical component of the electrostatic field across the lipid/aqueous interface, the variation in surface potential observed on water is typically associated with a change in DPPC molecular orientation (and, to a lesser extent, of conformation) upon compression. DPPC monolayers in the LC phase have large surface potential values because their molecules are closely

Table 1. Thickness of DPPC Monolayers on Aqueous Chloride Salt Solutions solution H2O NaCl KCl MgCl2

CaCl2

salt concentration (M) 0.6 2.0 0.6 2.0 0.3 1.0 2.0 0.3 1.0 2.0

monolayer thickness (nm) 2.48 2.45 2.22 2.61 2.21 2.58 2.40 1.66 2.39 2.01 1.71

± ± ± ± ± ± ± ± ± ± ±

0.37 0.46 0.60 0.45 0.53 0.47 0.47 1.14 0.43 0.50 0.96

monolayer thickness decreases with increasing salt concentration. This indicates changes in DPPC molecules conformation and/or orientation occurring, respectively, through an increase in gauche defects within the alkyl chain and an increase in the molecular tilt angle with respect to the surface normal. However, as the DPPC monolayer reaches surface pressures of ∼55 mN/m or greater upon compression, indicating that a highly structured and organized film forms on all salt solutions at all concentrations investigated, it is more likely that the decrease in thickness occurs through a change of the tilt angle of the alkyl chains and phosphate headgroup as they become oriented more toward the aqueous surface. In addition, the DPPC monolayer thickness differs little compared to water for aqueous solutions with 0.6 M Cl−, indicating that the cations have minimal influence on the molecular organization of DPPC at this concentration. In the cases of NaCl and KCl similar thickness values were

Figure 2. Measured surface potential of DPPC monolayers spread on (a) NaCl, (b) KCl, (c) MgCl2, and (d) CaCl2 salt solutions at different concentrations. The surface potential of DPPC monolayer on water is also given as a reference. 2046

DOI: 10.1021/acs.jpcb.5b10483 J. Phys. Chem. B 2016, 120, 2043−2052

Article

The Journal of Physical Chemistry B

× 104 C·mol−1), T is the absolute temperature (in K), εr is the dielectric constant of the solvent (78.38 for water at 298 K), and Cb is the salt solution bulk concentration (M). Equation 4 is valid for 1:1 salt solutions like NaCl; however, for 2:1 salts like MgCl2 and CaCl2, a more complicated form of the Grahame equation must be solved (see Supporting Information).58 Using a pKa ≈ 3.8 for the PC headgroup59 and considering that at pH = 5.6 more than 99% of the phosphate moieties are deprotonated, the surface charge density of DPPC monolayers is estimated to be −0.35 C·m−2 at a mean molecular area of 0.45 nm2 (see Supporting Information). However, this value may be overestimated, as MD simulations have shown that PC headgroups are in fact partially negatively charged due to the small excess negative charge of the phosphate moiety relative to choline (−1.2e vs +0.78e).60 Taking into account this correction, the surface charge density of DPPC monolayers is reduced to about −0.15 C·m−2. It has been argued that the GC model should be valid only for lipid monolayers with low surface charge densities (