Rheological Characterization of Mixed Surfactant ... - ACS Publications

1. Rheological Characterization of Mixed Surfactant. Films at Droplet Interfaces .... hydrochloride concentration of 0.1-0.2 mM, a phase change can be...
0 downloads 0 Views 1MB Size
Subscriber access provided by University of Winnipeg Library

Interface Components: Nanoparticles, Colloids, Emulsions, Surfactants, Proteins, Polymers

Rheological Characterization of Mixed Surfactant Films at Droplet Interfaces via Micropipette Aspiration Benjamin Locke Micklavzina, Kostyantyn Luferov, and Marjorie L. Longo Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00800 • Publication Date (Web): 28 Jun 2018 Downloaded from http://pubs.acs.org on June 30, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

1

Rheological Characterization of Mixed Surfactant Films at Droplet Interfaces via Micropipette Aspiration Benjamin L. Micklavzina,† Kostyantyn Luferov§, and Marjorie L. Longo§,* †

Department of Materials Science and Engineering, University of California Davis, Davis, California, 95616

§

Department of Chemical Engineering, University of California Davis, Davis, California, 95616 *[email protected]

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

2 Abstract Cationic and anionic surfactant mixtures can form viscous films that dominate the rheology and stability of micrometer sized droplet suspensions. In this work, we use micropipette aspiration to study the mechanical properties of mixed surfactant surface films of anionic sodium dodecyl sulfate (SDS) and cationic dodecylamine hydrochloride (DAH) on alkane and lipid droplets. For octane droplets, SDS was found to decrease surface tension until a minimum of 5±1 mJ/m2 was reached after the CMC. The surface viscosity of the droplets was found to be on the order of 10-3 mNs/m at an SDS concentration of 10 mM. Addition of 0.2 mM of DAH was found to increase this viscosity to a peak of 0.24 ± 0.01 mNs/m. Similar to octane, dodecane’s surface tension

decreased to a value of 7.7 ± 0.4 mJ/m2 at SDS concentrations above the CMC. Unlike with

octane, however, the dodecane droplets had a significant surface viscosity of 0.37 ± 0.01 mNs/m when only the 10 mM SDS film was present. Addition of DAH caused a decrease in this

viscosity initially, before rising to a peak viscosity of 0.45 ± 0.01 mNs/m at a DAH concentration of 0.15 mM. We speculate that the peaks in viscosities were the result of the completions of a phase change associated with microcrystalline SDS/DAH grains growing in the film at the surface of the droplets. Fluorescence microscopy and visual observations provided further evidence that these films can show rigid microcrystalline-like structure. Further work done with soybean oil in the same conditions and with a lipid film, simulating biological lipid droplets, confirmed that lipid droplets behave rheologically similar to alkanes in the presence of these mixed surfactant and lipid films. These results imply that droplet mechanics may be heavily influenced by the presence of microcrystalline grains in oil-water systems with complex surfactant mixtures. Keywords: oil, lipid, monolayer, mechanics, surface tension, surface viscosity

ACS Paragon Plus Environment

Page 3 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3 Introduction The study of the rheology of cationic and anionic surfactant (i.e. catanionic) mixtures is a topic of interest in colloid science, where emulsion stability can be dramatically affected by the presence of mixed surfactants.1-3 It is known that these mixtures form a wide array of lamellar structures and phases, and that these structures give oil-water interfaces differing rheological properties depending on the film density and the surfactant concentration.4,

5

The mechanical

properties of the interfacial films are often dictated by the composition of surfactants used and the phases present at the interface.6 While significant work has been done to study how components in these surfactant mixtures affect bulk emulsion properties,7-9 less has been done to directly probe the mechanics of mixed films at the droplet interface. Our goal then, is to directly study the mechanics of oil-water emulsion droplets and the films that form on them to better determine how these films affect their mechanical stability. There are several means of characterizing rheology for emulsion droplets in the presence of these surfactant mixtures, and several measures by which to do so. The most common measures used to characterize a droplet are its surface tension, viscosity, and yield shear. Surface tension is one of the simpler measurements that can be made at an interface; measurements have been done for oil droplets in the presence of surfactant films using the hanging drop method,

10

where the

buoyant force of a budding oil droplet is matched against the tension preventing the droplet’s break-off. This method is an established means of studying surface tensions to high accuracy, but it is not useful for measuring the rheology of droplets while they are flowing. Another way to study surface tensions and fluid properties of surfactant films is through use of a Langmuir trough.11,

12

While this a well-known method for measuring interfacial tension in biofilms, a

specialized apparatus would be required to measure how tension changes at an oil-water

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

4 interface.13 Similar to the hanging drop method, Langmuir troughs are not typically used to measure dynamic film properties of oil-water interfaces such as surface viscosity. It is for these reasons that we turned to micropipette aspiration. Micropipette aspiration was first established in the biological sciences as a means of studying the mechanics of cells.14, 15 In this technique, a pressure is exerted on the droplet directly via a micropipette, while the structure rests within a chamber in constant conditions. It was the work of Kwok and Evans, however, which brought this technique to the study of thin organic films such as lipid vesicles.14,

16, 17

Since their initial work, it has been shown that micropipette

aspiration can accurately determine yield tension, surface viscosity, and bending moduli of lipid films for a variety of structures.1, 18-20 Some work has been done with this method to study the mechanical properties of water-oil emulsions,21, 22 though these studies have focused more on characterizing complex crude oil mixtures than on droplets in the presence of lamellar aggregates. The ability of this technique to measure both surface tension and surface viscosity is of particular interest in our case, as it eliminates the need for multiple apparatuses to characterize droplet rheology. In this work, we explore how the presence of these mixed surfactant films affects the rheological properties of octane, dodecane, and soybean oil droplets via micropipette aspiration. We will explore how surface tension can be measured using the micropipette apparatus, as well as how differing surfactant concentrations and the presence of lipid affect these and surface shear viscosity results. Octane and dodecane were selected because a good deal of quantitative data exists for their surface tensions in the presence of surfactants such as SDS. Our results will indicate that, at a sodium dodecyl sulfate (SDS) concentration of 10 mM and a dodecylamine hydrochloride concentration of 0.1-0.2 mM, a phase change can be observed at our droplet

ACS Paragon Plus Environment

Page 5 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

5 surfaces. We speculate that the phase change and surface properties are tied to the presence of microcrystalline grains at the surface. Fluorescence microscopy of collapsed films provides some support for this theory, showing evidence for the presence of microcrystalline grains in conditions similar to our experimental setup.

Materials and Methods Micropipette Preparation For creating our pipettes, we used borosilicate glass microcapillary tubes from Sutter Instruments with length 7.5 cm, inner diameter 0.58 mm, and outer diameter 1 mm. Pipettes were bisected with a diamond edge then pulled using a P-97 Flaming/Brown pipette puller also from Sutter Instruments. The pipettes were subsequently forged with a MF-900 Microforge from Narishige to an outer diameter of 10-15 µm before cleaning and treatment. Pipette cleaning consisted of rinses with chloroform and methanol. First, the pipette was submerged in roughly 1 cm of chloroform, and the liquid was aspirated 5-10 mm into the pipette before being flushed out with air from a syringe. This procedure was repeated 3-5 times before removing the pipette from the chloroform. This same washing procedure was repeated with methanol immediately afterwards for another 3-5 times. Micropipettes were then filled with purified water using a Microfil Nonmetallic Syringe Needle (34 gauge/ 67mm) before being transferred to the micropipette holder for use in the apparatus. All water used in the work described was purified using a Barnstead Nanopure System (Barnstead Thermolyne, Dubuque, IA) with a minimum resistivity of 17.9 MΩ•cm. Micropipette Apparatus The micropipette apparatus consisted of a water reservoir attached to an Acu-Rite Senc150

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

6 sensor (1 µm), which was in turn attached to a 30 inch vertical slide. The sensor output was read using an Acu-Rite 200S readout, and the reservoir was attached to a Narishige HIK-7 micropipette holder via tygon tubing. Pressures were exerted at the mouth of the pipette by raising and lowering the reservoir along the slide. A Narishige MC35A Micromanipulator was used to move the pipette once it was inserted into the pipette holder. The chamber in which micropipette aspiration was performed consisted of a 25x75 mm glass slide (1 mm thick, Thermo Fischer Scientific) with a 24x50 mm cover slip (No. 1 ½, Gold Seal) with a sheet of soft silicone (1/32’’ thickness, MSC Direct) sandwiched between them. Vacuum grease was used to make a seal between the silicone and glass pieces, and small chambers were cut from the silicone at the edges of the glass so that the micropipette would be able to enter. The microscope used was a Nikon Eclipse TS100 inverted microscope with Hoffman modulated optics purchased from Modulation Optics Inc. An Amscope MU500 camera, along with the Amscope 3.7 program, were used to record videos and images, as well as take dimensional measurements during experiments. Emulsion Preparation Sodium dodecyl sulfate (ACS Reagent, ≥99%) purchased from Sigma-Aldrich was used to prepare solutions at concentrations ranging from 1 mM to 15 mM and volumes of 2 mL. To these solutions were added 30 µL of either octane (anhydrous, ≥99%), dodecane (anhydrous, ≥99%), or soybean oil (long chain triglycerides) all purchased from Sigma-Aldrich. Solutions were mixed with a vortexer for 1 minute directly before experiments in order to produce a large range of droplet radii. In some cases, dodecylamine HCl (>98%) from TCI America was added to the solution at concentrations between 0 and 0.5 mM. For some trials, soybean oil was used with 10 µM 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) purchased from Avanti Polar Lipids,

ACS Paragon Plus Environment

Page 7 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

7 instead of the regular surfactant solution. These solutions used 1 mM SDS as an emulsifying agent, and 2 g% bovine serum albumen (fatty acid free) purchased from Sigma-Aldrich to prevent adhesion to the pipette. In all cases, samples were given at least 30 minutes to equilibrate in a separate cuvette before any trial would begin. For solutions where fluorescence microscopy was performed, the solutions were prepared in a slightly different way. 30 µL hexane (mixture of isomers) from Sigma-Aldrich, octane or dodecane was added to a 10 mM SDS and 0-0.3 mM DAH solution with total volume 2 mL. This solution was first sonicated to create a solution of small droplets. Another 30 µL of the same alkane was added after this, and the solution was vortexed for 30-60 seconds to create a mixture of medium sized droplets in the smaller droplet emulsion. Nile Red from Thermo Fischer Scientific was added to solution at concentrations between 5-50 µM, and the solution was refrigerated at 2 °C for 3 minutes before imaging to increase the rate of surface condensation. In some cases, the solution included 2 mM NaCl to further increase the condensation rate. Surface Tension/Viscosity Experiments After the mixed emulsions were pipetted into the micropipette aspiration chamber, the pressure of the micropipette was calibrated. This was done by taking the pipette to the top of the chamber (where droplets were present) and changing the height of the reservoir until the zero pressure, at which droplets were neither attracted to nor repelled from the mouth of the pipette. Once calibration was completed, a droplet with radius >3 times larger than the pipette radius was aspirated into the pipette and moved away from the top surface by translation of the micromanipulator. The radius of the droplet and the inner radius of the pipette were measured visually using the Amscope 3.7 program, and the length deformation of the droplet into the pipette was measured as the pressure was changed. The pressure at which the droplet began to

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

8 flow into the pipette was recorded as the critical pressure and used to calculate surface tension. During viscosity experiments, the position of the front was measured from the furthest point of protrusion into the pipette as shown in Figure 1 and Video 1 (Supporting Information). Video 1 shows three viscosity experiments taken for a dodecane droplet in 10 mM SDS and 0.1 mM DAH. Data was only taken after pressures had stabilized (at that moment the pressures appear on the video), and the pressures in the video denote the pressure applied above the critical pressure. At the end of the video, the pressures were lowered below the critical pressure causing the projections to retract. Our apparatus was capable of applying pressures up to 7.8 kPa on any given droplet.

Figure 1: Shown is an example of a viscosity experiment done with a dodecane droplet in the presence of 10 mM SDS and 0.1 mM DAH. The images above show a change in initial length (A) from L= 97µm to a final length of L= 116 µm (B) after an elapsed time of 4.5 s. The excess pressure above the critical pressure for this experiment was 535 Pa, and the measured surface shear viscosity was 0.3 mNs/m.

ACS Paragon Plus Environment

Page 9 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

9 Measurements of surface and bulk viscosities were often performed within the same trials as those for surface tension. Pressure was increased above the measured critical pressure and flow rates were recorded along with the pressure associated with that flow rate. Flow rates between 1 to 5 µm/s were recorded. The results were used to calculate a true critical pressure for our experiments assuming a linear relationship between the flow rate and pressure in excess of the critical pressure (see Supporting Information). The changes in critical pressure with surfactant concentration were used to calculate interfacial tension. Data analysis for recorded pressures, deformation lengths, and flow rates was done in Microsoft Excel. Fluorescence Microscopy Imaging was done using a Nikon Eclipse E400 microscope with a 75 W Xenon source. Images were taken with a combination of oil and water immersion lenses. Images and videos were taken using Micro Manager 1.3, and image analysis was performed using ImageJ. Samples were imaged in the same chambers described for micropipette aspiration, consisting of a glass slide and coverslip separated by a silicone layer.

Results and Discussion Alkane Droplet Characterization Our first aim was to verify our ability to measure the surface tension of droplets of octane and dodecane with our apparatus. The surface tension of droplets was measured in one of two ways: via the critical pressure or the length of protrusion into the pipette. When the pressure in the pipette becomes sufficiently negative, the selected droplet enters the mouth. As pressure continues to decrease, the droplet begins to deform until a critical pressure is reached. At the

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

10 critical pressure, the droplet begins to flow into the pipette. This pressure can be defined by equation 1,14 (1) ∆ = 2 











.

Where ∆ is the critical pressure, is the surface tension of the droplet,  is the radius of

the droplet, and  is the pipette’s inner radius. It should be noted that, so long as the plane of the pipette entrance is flat, the effects of the mouth’s curvature can be ignored. Alternatively, the pressure below the critical pressure can be related to the surface tension using equation 2, (2) ∆ = 2 











−   = 2  −  



 

.

Where ∆ is a measured pressure below ∆ ,  is the radius of curvature of the protrusion, and

is the measured length of the projection into the pipette. A derivation for equation 2 can be

found in the Supporting Information section. A detailed diagram of these parameters and what they mean can be found in Figure 2. While many rheological models exist in literature for biological systems, the relative homogeneity of our system compared to that of a cell means that simpler Young-Laplace relations can still accurately characterize droplet mechanics. It was found that both of these approaches to measuring surface tension gave similar values (within 20%), and as such the results presented in subsequent figures use a combination of both methods to determine surface tensions. Specifically, each data point represents the average of 3-4 data points at lower pressures using the deformation length and a single measurement using the critical tension.

ACS Paragon Plus Environment

Page 11 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

11

Figure 2: Shown are (A) example images of the micropipette apparatus and the meaning of specific variables Rd, Rz, L, and Rp and (B) a similar image taken at the critical pressure, where Rz and L both approach Rp. Images were taken with dodecane droplets at an SDS concentration of 10 mM.

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

12 Figure 3: The measured interfacial tensions for octane and dodecane droplets with varying SDS concentration. Error bars represent the standard deviation over n=4-5 trials. Our findings for how the surface tension of octane and dodecane vary with SDS concentration can be found in Figure 3. In prior work studying how the interfacial tension of alkanes in water changed in the presence of SDS, the relationship between surface tension and surfactant concentration was not an exactly exponential relationship.10 We see for our results that the surface tension decreases nearly exponentially with concentration for both alkanes as SDS is added, which is consistent with this prior result.10, 23 It should be noted that interfacial tensions at concentrations of SDS below 1 mM could not be obtained with this method, as relatively strong adhesion between the glass pipette and the droplet appeared at SDS concentrations below this limit. While experiments were possible at lower SDS concentrations with BSA passivation, the addition of BSA to solution could have significantly affected the magnitude of results.24 Since the purpose of these trials was to compare with values in literature, we avoided testing under these conditions for our alkanes. The highest observed surface tensions for these alkanes was between 25-30 mJ/m2 at an SDS concentration of 1 mM, which is lower than the values found in literature for the interfacial tensions of octane (~50 mJ/m2)25 and dodecane (~55 mJ/m2)25 in the absence of added surfactant. Since we expect the addition of SDS to decrease the interfacial tension, this result is in line with expectations. As SDS concentration increases, the saturation of molecules at the surface reaches a peak. At this peak, the critical micelle concentration (CMC), the surface tension is no longer strongly affected by surfactant concentration. Our results indicate that the measured CMC for SDS is similar for both alkanes at a value of 7 ± 1 mM for octane and 8 ± 1 mM for dodecane. These values are consistent with those previously reported for these surfaces in the presence of SDS,10,

ACS Paragon Plus Environment

Page 13 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

13 26

and serve as evidence of our method’s numerical accuracy. In addition, we found that the

surface tension of these substances stabilized at values of 5 ± 1 mJ/m2 for octane and 7.7 ± 0.1 mJ/m2 for dodecane. That the interfacial tension between an alkane-water interface increases with the length of the alkane is known, and the values acquired are reasonably close to those found in other work.10, 25 With our ability to measure surface tension established we looked next to measurements of droplet viscosity. Measuring droplet viscosity involves determining the rate of flow of the droplet into the pipette at pressures above the critical pressure. From here, we must make an assumption as to whether the flow of the droplet is a function of bulk or surface viscosity. If we assume that the resistance to viscous flow largely comes from the bulk modulus of the fluid within the droplet, this modulus can be determined using equation 3,14 (3) !" =

∆#$ 

.

( %& '   (

Where !" is the bulk viscosity, ∆) is the pressure above ∆ , and & is the velocity of the droplet front during viscous flow. In cases where a thin surface layer is present, however, the observed viscosity is also function of this layer’s properties. In the scenario where the surface viscosity determines flow characteristics, viscous flow is defined by the relation in equation 4,20 (4) !* =

 ∆#$ 



.

( +& ,'  (

Where !* is the surface shear viscosity and all other variables are identical to those in equation (3). In the case of octane, for SDS concentrations between 1 to 15 mM, the measured value of & at

relatively small values of ∆) (30-50 Pa) was so large (100-400 µm/s) that it was difficult to

determine a value for viscosity using either of equations (3) or (4). We can say, however, that the

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

14 bulk viscosity calculated via equation (3) for octane in the presence of SDS would be on the order of magnitude of 1 Ns/m2. This is far larger than the literature value27 for octane, which is 5x10-4 Ns/m2. From this stark discrepancy, we can conclude that the observed droplet viscosity is not a bulk value but is primarily a property of the surface layer. Using equation (4), we calculated a surface shear viscosity, !* , for octane in the presence of SDS on the order of 10-3 mNs/m. While the surface shear viscosity may have been affected by changing SDS concentrations, our experimental setup was not sufficiently sensitive to detect or describe such changes in the case of octane. For dodecane, at SDS concentrations between 1 to 15 mM, the flow velocity was much slower than what was observed for octane at similar pressures. Using equation (3) again, the bulk viscosity of dodecane would have to be on the order of 103 Ns/m2 to explain our observed & values. This is several orders of magnitude larger than the literature value28 of 10-3 Ns/m2, again forcing us to conclude that the primary resistance to flow comes from the surface layer rather than the fluid bulk. If we instead apply equation (4) as was done for octane, we arrive at a value for !* of 0.36 ± 0.01 mNs/m at an SDS concentration of 10mM, which is comparable to the viscosity observed for DPPC monolayers coating gas filled microbubbles20 and lamellar phases for double-tailed surfactants.1 The reason for the observed difference between the surface viscosities of the two alkanes is not entirely understood. In general, it is known that the shear viscosity of any thin layer is a function of its thickness, and that a thicker film will usually be more viscous than a thinner one.1 Our results imply that the thickness and, perhaps, the microstructure of the surface films is different between the two alkanes.20 It should also be noted that we were unable to observe any elastic properties for the SDS films present on either alkane, which would have manifested as an increase in critical pressures with

ACS Paragon Plus Environment

Page 15 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

15 added surfactant.19,

20

This is consistent with results for other surfactant films at hydrophobic

interfaces, and implies that the surface layer is not as tightly bound or formed as a lipid layer, for which elastic moduli would be easily observable at our exerted pressures.10 With baseline trials of alkanes with added SDS established for our apparatus, we decided to move forward exploring the effect of aggregates on surface properties. When SDS is in the presence of dodecylamine HCl (DAH), it is known to form lamellar aggregates in solution rather than micelles formed by the individual components. This is a common occurrence when cationic and anionic surfactants are mixed, but how the presence of such a film affects droplet properties has not been thoroughly probed. For each alkane, we kept the SDS concentration fixed at 10 mM while varying DAH concentration between 0 mM and 0.5 mM. The SDS concentration was selected at 10 mM because it was safely above the measured CMC indicated by the surface tension measurements. The DAH concentration limit of 0.5 mM was selected both due to turbidity issues and problems with aggregate masses clogging the pipette at higher concentrations. Our goal with these trials was twofold: to see whether any phase change in the surface film could be observed when lamellar aggregates are present in solution, and to determine the effect of such aggregates on surface viscosity.

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

16 Figure 4: The measured surface shear viscosity of octane and dodecane droplets with changing DAH concentration at a fixed SDS concentration of 10 mM. The regions noted at the top of the graph (I-III) represent hypothesized regions of phase stability, with the dotted lines marking the approximate location of suspected phase transitions. Error bars shown represent the standard deviation of n=3-5 measurements. Shown in Figure 4 are the results of surface viscosity measurements for both octane and dodecane with varying DAH concentration at a fixed SDS concentration of 10 mM. In the case of octane, we see the apparent !* rise from ~10-3 mNs/m when no DAH is present to 0.24 ± 0.01 mNs/m at a DAH concentration of 0.2 mM. Higher DAH concentrations appeared to cause a small decrease in film viscosity. The relatively high peak viscosity observed is similar to those observed for short chain lipid monolayer shells such as DPPC coating gas filled microbubbles,20 which implies that the surface structure approaches the microcrystalline nature of lipid monolayer shells as shown in the schematic (Figure 5). A clear phase change can be observed with the naked eye for solutions of octane droplets containing 10 mM SDS at DAH concentrations above 0.2 mM, with the solution shifting from relatively clear to an opaque white. The suspected DAH concentrations at which phase changes may occur have been marked in Figure 4 by dotted lines between regions I through III. It is known, in general, that cationic and anionic surfactant mixtures undergo phase changes as surfactant composition shifts, with lamellar phases of varying packing geometries comprising the majority of phases observed.1, 29 Moreover, it has been shown for a lipid monolayer surrounding the gas core of a bubble that ordered regions of microcrystalline grains can self-assemble, and that the size of these grains can have a major impact on yield shear and shear viscosity.20 Kim et al. observed that, as grain size decreased the shear viscosity decreased dramatically, and that faster cooling rates produced

ACS Paragon Plus Environment

Page 17 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

17 smaller grains.20 Noting the structural similarity of ordered lipid shells to ordered surfactant shells, we can expect the presence of these microcrystalline structures at our surface. The structure at the surface of the octane droplet is likely a liquid monolayer of SDS surfactant molecules before the addition of DAH, as the shear viscosity is negligible at low DAH concentrations. As the concentration of DAH increases, we speculate that the rise of surface viscosity takes place because of the appearance of a mixed phase region, where liquid–like and microcrystalline phases compete for area at the surface. At a critical DAH concentration around 0.2 mM, the magnitude of the surface viscosity indicates that the low viscosity liquid phase of surfactant disappears in favor of a high viscosity and high order microcrystalline structure, i.e. the completion of the phase transition.

Figure 5: Schematic of the structure of a lipid-like shell comprised of a mixture of SDS and DAH. Assuming our system resembles those seen for lipid shells, the liquid-like region acts as a grain boundary between microcrystalline regions rich in the dominant surfactant species. For dodecane at a fixed SDS concentration of 10 mM we see a similar trend with an important difference. Notably, as mentioned earlier, the apparent !* without added DAH is significantly high at 0.37 ± 0.01 mNs/m. This value drops to 0.10 ± 0.01 mNs/m at DAH concentrations of

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

18 0.025 mM before rising to a peak of 0.45 ± 0.05 mNs/m at a concentration of 0.15 mM. Concentrations higher than this show a ~50% decrease in observed viscosity, similar to that seen in octane. A similar change in opacity is observed for dodecane solutions containing 10 mM SDS at DAH concentrations near 0.15 mM, with the solution changing from being clear in the absence of DAH to opaque white at higher concentrations. The initial drop in surface viscosity lends support to the hypothesis that a phase change begins at the surface of the dodecane droplets at very low DAH concentrations. The addition of small amounts of DAH could be disrupting the packing order at the dodecane interface, as low molecular weight surfactants have been shown to decrease the surface viscosity of protein films for similar reasons.30 Given the relatively high viscosity of the droplet surface before any DAH is added, we conclude that the SDS film is in an ordered microcrystalline state at the beginning of the trial. This is in contrast with our result from octane and highlights the effect of the “substrate” on phase stability at the surface. The drop in viscosity between 0 and 0.025 mM DAH may be an indication of shrinking grain sizes as two microcrystalline phases compete at the surface. As the concentration of DAH increases further, the subsequent rise of surface viscosity infers a transition from mixed microcrystalline phases to a single microcrystalline phase, presumably the same phase observed at high DAH concentrations in the octane. Interestingly, the phase transition in 10 mM SDS at the dodecane interface appears to complete at a lower DAH concentration (0.15 mM) than for octane (0.2 mM). This is somewhat consistent with data gathered in literature which suggests that phase transitions involving surfactants are dependent on the chemical composition at the interface.10, 26 The decrease in surface viscosity after the transition seen for SDS-DAH films on octane and dodecane is likely tied to the size of microcrystalline grains, which has been shown to be very influential in film strength in other

ACS Paragon Plus Environment

Page 19 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

19 work.20 More information about the size and ordering of these inferred surface domains is necessary to make strong conclusions about the rise and fall of viscosities near the phase transition. Shear Thinning and Droplet Breakoff While the data gathered provides compelling evidence that the surface viscosity of these droplets is dominated by surfactant structures at the surface, we encountered some important issues that should be addressed before continuing. First, we should note that for droplets in 10 mM SDS, at concentrations of DAH greater than 0.2 mM there were large variations in the recorded viscosity depending upon the shear rate applied. In theory, the measured viscosity of these films should be roughly constant at low shear rates.8 However, shear thinning is somewhat expected for lamellar aggregates when under significant strain; thicker multilayer films generally thin through the rupture of outer layers or presence of phase transitions,7-9 whereas monolayers can thin due to the presence of surface tension and surfactant gradients along the surface.2, 31, 32 As such, all results shown in Figure 4 were taken within a similarly small range of low shear rates ( &