Monodisperse Micro-Oil Droplets Stabilized by Polymerizable

Subscriber access provided by FLORIDA ATLANTIC UNIV

Article

Monodisperse Micro-Oil Droplets Stabilized by Polymerizable Phospholipid Coatings as Potential Drug Carriers Yoonjee C Park, Tuan Pham, Carl Beigie, Mario Cabodi, Robin Cleveland, Jon O Nagy, and Joyce Y Wong Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.5b02747 • Publication Date (Web): 24 Aug 2015 Downloaded from http://pubs.acs.org on August 29, 2015

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 free 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 accessible to all readers and 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.

Langmuir 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 32

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

369x160mm (150 x 150 DPI)

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 32

Monodisperse Micro-Oil Droplets Stabilized by Polymerizable Phospholipid Coatings as Potential Drug Carriers Yoonjee Park‡1#, Tuan A. Pham‡1, Carl Beigie1, Mario Cabodi1,3, Robin O. Cleveland4, Jon O. Nagy5, and Joyce Y. Wong1,2,3* 1

Department of Biomedical Engineering and 2Division of Materials Science & Engineering, and 3

Center for Nanoscience and Nanobiotechnology, Boston University, 44 Cummington Mall, Boston, MA 02215, USA

4

Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Old Road Campus Research Building, Oxford, OX3 7DQ, UK 5

NanoValent Pharmaceuticals, Inc., 910 Technology Blvd. STE G, Bozeman, MT 59718, USA ‡

#

These authors contributed equally

Current Address: Department of Biomedical, Chemical & Environmental Engineering, University of Cincinnati, OH 45221 *

Author to whom correspondence should be addressed (TEL: 617-353-2374; FAX: 617-353-6766; email: [email protected])

KEYWORDS: Microfluidic Flow-focusing, Oil droplets, Emulsion, Polymerizable Lipid


1

Page 3 of 32

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

Abstract There is a critical need to formulate stable micron-sized oil droplets as hydrophobic drug carriers for efficient drug encapsulation, long-term storage, and sustained drug release. Microfluidic methods were developed to maximize the stability of micron-sized, oil-in-water (o/w) emulsions for potential use in drug delivery, using doxorubicin-loaded triacetin oil as a model hydrophobic drug formulation. Initial experiments examined multiple flow conditions for the dispersed (oil) and continuous (liposome aqueous) phases in a microfluidic device to establish the parameters that influenced droplet size. These data were fit to a mathematical model from the literature and indicate that the droplet sizes formed are controlled by the ratio of flow rates and the height of the device channel, rather than the orifice size. Next, we investigated effects of o/w emulsion production methods on the stability of the droplets. The stability of o/w emulsion produced by microfluidic flow-focusing techniques was found to be much greater (5 hours vs 1 hour) than for emulsions produced by mechanical agitation (vortexing). The increased droplet stability was attributed to the uniform size and lipid distribution of droplets generated by flow-focusing. In contrast, vortexed populations consisted of a wide size distribution that resulted in a higher prevalence of Ostwald ripening. Finally, the effects of shell polymerization on stability were investigated by comparing oil droplets encapsulated by a photopolymerizable diacetylene lipid shell to those with a non-polymerizable lipid shell. Shell polymerization was found to significantly enhance stability against dissolution for flow-focused oil droplets but did not significantly affect the stability of vortexed droplets. Overall, results of these experiments show that flow-focusing is a promising technique for generating tunable, stable, monodisperse oil droplet emulsions, with potential applications for controlled delivery of hydrophobic drug formulations.


2

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 32

1. Introduction Delivery of hydrophobic drugs through blood vessels remains a persistent challenge as their poor solubility and permeability results in poor biodistribution. Micron-sized oil droplets dispersed in an aqueous phase, i.e. oil-in-water (o/w) emulsions, are attractive because they are easy to produce; however they are thermodynamically unstable, making both efficient drug encapsulation and sustained drug release difficult. It is possible to generate thermodynamically stable o/w colloidal systems known as ‘microemulsions’, but this process often requires large amounts of surfactants and co-surfactants at high concentrations that can be toxic to cells or tissues1,2. Moreover, excessively high stability of surfactant- and co-surfactant-stabilized microemulsions makes it difficult to control diffusion and partitioning of drugs between dispersed and continuous phases. While the toxicity issue can be overcome by using various pharmacopoeia-approved phospholipids as the emulsifier, emulsions made in this way generally lack stability1,2. Several studies have shown that either the mole fraction of oil to lipids (or emulsifier) or the chemical potential of lipids is a critical determinant of emulsion instability3-6. For example, a reduction of lipids at the interface is likely to cause diffusion of oil into the carrier phase because of the decreased physical barrier against leakage combined with high interfacial tension. Other reports have demonstrated stabilization of micron-sized drug carriers by using photopolymerizable materials in the dispersed phase7,8, but these systems are not easily degradable, making them inefficient at drug delivery.


3

Page 5 of 32

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

In addition to physiologically relevant release kinetics, effective emulsion-assisted drug dosing must be both reproducible and predictable. Many applications (e.g., foods, cosmetics, pharmaceuticals) utilize dispersions produced with bulk emulsification methods that use homogenizers or sonicators. While relatively simple to carry out on a large scale, these methods also result in wide particle size distributions, batch-to-batch variations, and unstable dispersions, all of which make accurate and reproducible drug dosing difficult.9 Thus, design of emulsion formulations for drug delivery would ideally entail the generation of droplets of uniform size and composition. Microfluidic techniques allow for accurate control of droplet size and composition at the micrometer scale and in situ visualization of droplet production to enable optimization10. Previous studies11-13 showed that droplet size can be controlled by modulating microfluidic device design, physical properties of the liquids, and relative flow rates of the two immiscible phases. The microfluidic method is also attractive for creating formulations that employ expensive materials (e.g. drugs) because of low volumes required and reduction of material waste during purification steps. In this study, microscale phospholipid-encapsulated oil droplets of uniform size and composition were produced using a custom microfluidic flow-focusing device, where the dependence of quantitative features (e.g. droplet size, and distance from the orifice at which the droplet forms) is related to the ratio of flow rates of the two immiscible fluids. Moreover, to the best of our knowledge, here we reveal for the first time the relationship between that the smallest droplet size and the device height. There are no reports in the literature investigating effects of production methods of o/w emulsion on micron-sized droplet stability.

Bulk emulsification methods, such as


4

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 32

homogenization or sonication, require surfactant concentration greater than 10~ 20 mol% to make the emulsion stable.12,13 In contrast, this study investigates the stability of the oil droplets with surfactant concentration of just 1 mol% and compares two different production methods, vortexing and microfluidics. The use of low mol% of lipids is particularly attractive in pharmaceutical applications because high emulsifier concentration levels are costly and can often be toxic, causing side effects. Building on our previous work showing that coating microbubbles with polymerizable diacetylene lipids enhances stability against gas dissolution,14 we used the same lipids to allow stabilization of the droplet shell by UV-crosslinking and 1,2-distearoyl-sn-glycero-3phosphocholine (DSPC) as a component of the phospholipid emulsifier.

2. Materials and Methods 2.1. Fabrication of Microfluidic Devices Soft lithography techniques were used to generate patterned silicon wafers and microfluidic polydimethylsiloxane (PDMS) devices.15,16 The single emulsion microfluidic device design was modified from Hettiarachchi et al.17 as previously described3 to fit our specific needs. Briefly, an AutoCAD design of the fluid channels was printed onto a chrome mask and transferred onto a silicon wafer. The wafer was then used as a PDMS mold for the microfluidic devices. The molded PDMS was cut, port holes were punched, and the device was bonded to a glass slide via plasma ashing. The PDMS devices (Figure 1) were plasma-treated for 5 min immediately prior to use to ensure that the device channels were hydrophilic. The hydrophilicity of the surface of the PDMS device was verified by means of a contact angle goniometer (Kruss DSA100, Hamburg, Germany).


5

Page 7 of 32

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

2.2. Liposome Dispersions The lipid mixture containing 25 mol% of polymerizable diacetylene lipids, N-(5’hydroxy-3’-oxypentyl)-10-12-pentacosadiynamide

(“h-Peg1-PCDA”)

and

N-

[methoxy(polyethylene glycol)-5000] -10-12-pentacosadiynamide (“m-Peg5000-PCDA”) (10 mol% h-PEG1-PCDA and 15 mol% m-PEG5000-PCDA) (NanoValent Pharmaceuticals, Inc., Bozeman, MT), and 75% of L-α-phosphatidylcholine, hydrogenated Soy (hydro soy PC) (Avanti Polar Lipids, Alabaster, AL) were used for the 25 DA lipid mixture (Figure 1, inset). The lipids were dissolved in chloroform and mixed to form solutions of various mole fractions, with a total lipid concentration of 5.32 µM. Desired amounts of this blend were then placed under vacuum overnight in a UV opaque vial to evaporate off chloroform to form a thin lipid film. The drieddown lipid mixture was then resuspended into a mixture of 10/10/80 (10% glycerol, 10% propylene glycol, 80% DI water) at a 5.32 µM (total lipid) concentration and tip-sonicated for 10 min at 25% power in a Branson Digital Sonifier Model 450 (Branson Ultrasonics, Danbury, CT) with a 3-mm tip attachment. The temperature was not controlled during the sonication process, and the solution reached approximately 80°C. The solution was then capped and heated in an 80°C oven for 10 min to remove any foaming caused by tip sonication. Finally, the mixture was cooled to approximately 5°C for 10 min in a -20°C freezer, taking care that the solution did not freeze, then was allowed to return to room temperature and stored in a 37°C oven until used. The non-polymerizable lipid formulation was made in a similar manner, but with a molar ratio of 85% of 1,2-distearoyl-sn-glycero-3-phosphatidylcholine (DSPC) (Avanti, Alabaster,

AL)

and

15%

1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-

[methoxy(polyethylene glycol)-5000] (DSPE-PEG5K) (Avanti, Alabaster, AL). 2.3. Doxorubicin in Oil


6

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 32

Doxorubicin hydrochloride (Sigma-Aldrich, St. Louis, MO) was added to DI water and then neutralized with the addition of NaOH to form a doxorubicin precipitate. The solution was then centrifuged at 2,500 rpm for 5 min, and the supernatant was removed. The remaining solution was dried under vacuum, and the doxorubicin powder was then weighed and dissolved into filtered (0.2 µm syringe filter) glyceryl triacetate (Sigma-Aldrich, St. Louis, MO) at 0.57 mg/ml. Note: doxorubicin is extremely cytotoxic and in powdered form must be handled in a properly ventilated fume hood to avoid accidental exposure. 2.4. Oil-in-water Emulsions Prepared by Vortexing and Microfluidic Flow-Focusing Two different methods were used to achieve drug-loaded oil droplets: vortexing and microfluidic flow-focusing. The ratio of doxorubicin-infused oil to the liposome dispersion was the same for both methods of droplet formation. For vortexed oil droplets, a known ratio of liposome dispersion to doxorubicin-loaded oil was placed into an Eppendorf tube. The mixture was then vortexed at level 5 for 30 sec on a Baxter Scientific Vortex Mixer S8223-1 (Deerfield, IL). In order to produce the flow-focused oil droplets, a microfluidic device was used as previously described3. Briefly, polyethylene tubing was cut and placed into the device at the two inlet positions. The oil and liposome dispersion were pumped through a syringe filter and into the device via syringe pumps to form oil droplets. Oil droplet size and flow regimes were controlled by varying flow rates of the liposome dispersion and oil. Once the oil droplets formed, a pipet tip was used to collect droplets, and droplets were polymerized as needed in a Spectrolinker Xl-1000 (Spectroline, Westbury, NY) at 0.5 J/cm2. The polydispersity, σ, was calculated as σ = δ/davg × 100%, where δ is the standard deviation and davg the average droplet diameter.


7

Page 9 of 32

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

Figure 1. Microfluidic Device Design and Flow Directions. The lipid formulation (chemical structures shown in lower left panel) and the drug-loaded oil are focused at the orifice. The lower right panel shows the orifice and a representative image of oil droplet formation. Device dimensions at the orifice are: oil inlet: 20µm (A); liposome dispersion inlets: 30µm (B); orifice: 6µm (C); outlet: 93µm (D) and the channel height is 21µm.

2.4. Flow regimes The viscosity of the liquids was measured using an Advanced Rheometer AR2000 (TA Instruments, New Castle, DE) with a rotor Peltier cylinder system and a concentric cylinder system at different shear rates from 0 to 150 s-1. The flow regimes of the two immiscible fluids in the microchannels were observed using a phase contrast microscope (Axiovert S25, Zeiss, Oberkochen, Germany) with a 10X objective lens and recorded with a high-speed camera


8

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 32

(Photron, San Diego, CA). All dimensions, such as droplet sizes or thread length, were measured

using ImageJ software. 2.5. Stability Tests Triacetin oil droplet stability was measured in buffered saline to compare the effects of fabrication method and lipid shell polymerization. Droplets were prepared using either microfluidic flow-focusing or direct vortexing in the manner described above. Polymerizable 25 DA lipids and the non-polymerizable 85/15 DSPC/DSPE PEG-5K formulations were used in parallel to compare effects of polymerization on droplet stability. Each liposome dispersion was prepared as described above except for buffering of the 10/10/80 solution to a pH of 7.0 prior to dissolving dried-down lipids. A total of three distinct groups of oil droplets were used to compare stability: 25 DA polymerized, non-polymerized 25 DA, and 85/15 DSPC/DSPE PEG5K as a control. Each oil droplet formulation was held constant at a ratio of 1:10 (v:v) oil:liposome dispersion. Polymerization of 25 DA polymerized droplets was performed in the Spectrolinker Xl-1000 with a cumulative UV energy of 0.5 J/cm2 at a wavelength of 254 nm. Stability for each group was observed in suspension over the course of five hours. After fabrication, the droplet-containing liposome dispersion was allowed to settle for 10 min, and a loose pellet of oil droplets accumulated at the bottom. The now-translucent supernatant above the collected droplets was removed. A total of 25 µl of remaining solution was then diluted in 175 µl of 150mM NaCl at pH 7 and mixed for a total volume of 200 µl. The stability of each oil droplet formulation was measured by observing the percent of oil droplets remaining over time. The number of droplets over a constant area was counted using a hemocytometer and normalized by the number of droplets over the same area at time zero. At each time point, the solution was manually shaken, from which 15 µl was placed onto a hemocytometer. Three random sections


9

Page 11 of 32

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

were imaged at 10X using a phase contrast microscope (Axiovert S25, Zeiss, Oberkochen, Germany) and a high-speed camera. Images were taken at 0 and 30 min, and 1, 2, 3, 4, and 5 hr, and droplets were counted manually using ImageJ. Statistical analysis was performed on the percentages calculated from images (n = 3) taken at a given time point. A two-tailed unpaired Student’s t-test, with α= 0.05, was used to compare differences between fabrication methods for a given liposome formulation (flow-focused vs. vortexed). Likewise, the percentage of oil droplets remaining was compared among liposome formulations for a given fabrication method using ANOVA and a subsequent post hoc Tukey’s range test.

3. Results 3.1. Flow regimes In order to compare directly with the mathematical model presented by Nie et al.3, we adopt their nomenclature of labeling flow of the inner channel to be Qi (oil) and flow of the outer channels to be Qo (10/10/80 liposome dispersion). We studied the emulsification process by varying the flow rate of the 10/10/80 liposome dispersion (continuous phase), Qo, for two different constant flow rates of triacetin oil (dispersed phase), Qi, to determine the effect of flow rate ratio of the two phases on droplet size. The flow rates tested were Qi = 0.05 and 0.1 µl/min for 0.5 ≤ Qo ≤ 7 and 0.5 ≤ Qo ≤ 9 µl/min, respectively, until Qo dominates the flow to such a point that it interrupts droplet production by dominating the dispersed phase flow. The viscosity of triacetin was ηi = 16.8 cP, and of 10/10/80 solution was ηo = 2.03 cP. Both fluids showed constant viscosities at different shear rates, suggesting they are Newtonian fluids. For the range of flow rates considered, flow was laminar (as defined by the Reynolds number, Re = ρQ/ηh,


10

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 32

where ρ is the density of the liquid and h the height of the channel; Rei = 0.0027 and 0.0055 for the two oil flow rates, and 0.2 ≤ Reo ≤ 3.62 for the range of liposome dispersion flow rates). In order to reproducibly produce droplets of predictable size, it is helpful to determine quantitative relationships between production input and output parameters. The snapshots of droplet formation (Figure 2) show flow regimes of the two immiscible fluids and resulting droplet size where Qi was held constant at 0.1 µl/min while Qo, was varied from 0.5 to 9 µl/min. The images show that the droplets decrease in size as Qo increases. The volume of the droplets as a function of normalized flow rate (Qo/Qi) is shown on a log-log scale in Figure 3. Due to the height restriction in the microfluidic device channel (h =21 µm), droplets with diameters greater than h were not spherical, but discoid (vertically), and their volume was calculated as = (/12)[2 − ( − ℎ) (2 + ℎ)](Nie et al. 2008), where D is the droplet diameter measured in the snapshot image. For both inner flow rates, Vd appeared to be inversely proportional to the ratio of flow rates: Vd ∝ (Qo/Qi)-1. This correlation suggests that the production conditions are in the break-up regime controlled by the rate of flow18, which will be addressed in more detail in the Discussion section.


11

Page 13 of 32

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

Figure 2. Snapshots of micro-oil droplet formation at Qi=0.1ul/min with Qo varying from 0.5 to 9 ul/min. Figures 3 and 4 quantify drop volume and formation distance.


12

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 32

Figure 3. A log-log plot of Vd vs. Qo/Qi for Qi = 0.05 ul/min (square) and Qi = 0.1 µl/min (triangle). The data points are the average of two or three sets of data and the error bars are smaller than the data points. The solid lines are trendlines of each set with slopes -1.104 (R2=0.9169) for Qi = 0.05 µl/min and -1.067

(R2=0.8038), for Qi =0.1 µl/min. The dashed line represents a threshold, above which droplets are discoid and below which droplets are spherical. The arrows indicate points where the inner flow started to be retracted by continuous flow as shown in the inset snapshot images.

Depending on the flow rates of the two immiscible liquids and the properties of the dispersed phase, emulsification was observed in two different modes: a jetting mode and a dripping mode. In the jetting mode (shown in Figure 2), a continuous thread of the inner phase extends into the outlet channel and breaks several orifice widths downstream of the orifice where the jet becomes unstable with respect to the Rayleigh-Plateau instability.11 In the dripping regime,


13

Page 15 of 32

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

the thread of the inner phase retracts upstream of the orifice and breaks either in the orifice or within close proximity of the orifice, after which the whole process repeats to continuously produce droplets. The jetting mode was observed most frequently for the production of oil droplets, which was in contrast to when we have used the device to form micron-sized gas bubbles (data not shown) in which case the dripping mode was more common. When Qo/Qi was 60 and 80 at Qo = 0.05 and 0.1 µl/min, respectively, the oil phase (inner flow) started to retract due to the dominance of the continuous phase (outer flow), and a continuous jet stream was no longer observed, as shown in the inset images in Figure 3. Even though Vd was inversely proportional to (Qo/Qi) as mentioned above, Vd increased sharply when the inner phase started to be retracted (arrows in Figure 2). This increase corresponded with the droplet having a diameter

the same as the height of the channel, that is, for h = 21 µm, = = 4849 µm3, which corresponds to the dashed line in Figure 2. As (Qo/Qi) increased further, Vd gradually decreased again until droplet production stopped at Qo = 7 µl/min for Qi = 0.05 µl/min and Qo = 9 µl/min for Qi = 0.1 µl/min, with droplet volumes of 1767 µm3 and 4173 µm3, respectively. Droplet production stopped in both cases because the continuous phase completely retracted the inner phase and pushed the flow backwards into the oil inlet. The smallest droplet size when production stopped was D = 15 µm (Qi = 0.05 µl/min) and 20 µm (Qi = 0.1 µl/min) which are much larger than the 6 µm orifice. When different channel heights were tested, 6 and 12.5 µm, the minimal droplet size which can be produced by each height was 6 and 13 µm, respectively (data not shown). These results indicate that droplet size is dependent not on orifice width but on channel height and the ratio of flow rates in the dripping regime for this device geometry.


14

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 32

On the other hand, the width of the jet thread varied from 4 µm (at low Qo) to 6 µm (at high Qo) and therefore was governed by the orifice size, w = 6 µm. It was observed that the width right before droplet break-up was always 6 µm, which is the same as orifice size, i.e. it is independent of flow rate or the ratio of the flow rates. Figure 4 shows the relationship between the distance from the orifice at which the droplet forms d normalized by the channel height h, and Qo/Qi. For Qo/Qi < 80 droplets are formed in the jetting regime and the normalized distance is constant, but the magnitude depends on Qi. When normalized height is divided by Qi (Figure 4, inset), most of the values collapse to a constant value, indicating that d is proportional to Qi. For Qo/Qi >80, the flow regime changes from jetting to dripping, by the retraction process described earlier, and d/h converges to 8 (dashed line in Figure 4) until droplet formation is halted. These results indicate that in the dripping regime, the distance from the orifice where the droplet forms depends on the device geometry, rather than the flow rates.


15

Page 17 of 32

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

Figure 4. The distance from the orifice at which a droplet forms d, normalized by the channel height h, as a function of Qo/Qi at constant Qi = 0.05 µl/min (square) and 0.1 µl/min (triangle). The dashed line indicates the distance where the droplets consistently formed before the formation stopped due to retraction. The inset shows a plot of d/(hQi) vs. Qo/Qi.

3.2. Size Distribution and Encapsulation of Doxorubicin Lipid-coated doxorubicin-encapsulated oil droplet populations were formed via both vortexing and microfluidic flow-focusing methods. Figure 5 shows oil droplets formed by each method using bright-field and fluorescent imaging. The flow-focusing method produced droplets with Davg = 11.1 µm and σ = 8.81%, while vortexing produced droplets with Davg = 15.0 µm and σ = 83.2%. Doxorubicin has fluorescent properties.19 The TRITC-filtered images (Figure 5)


16

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 32

indicate that fluorescently active doxorubicin was successfully encapsulated within the oil droplets for both cases. For the vortexed TRITC image, pixels were oversaturated in order to reveal the existence of smaller vortexed droplets, which have their signals overshadowed by the more intense fluorescence given off by larger droplets.

Figure 5. Bright-field and TRITC filter fluorescent images of doxorubicin-encapsulated micro-oil droplets produced by microfluidic flow-focusing (left) and vortexing (right) methods.


17

Page 19 of 32

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

3.3. Stability of Micro-oil Droplets

3.3.1. Effect of Preparation Method on Stability For each lipid mixture and fabrication method, oil droplet stability was studied over the course of 300 min, while suspended in 150 mM NaCl buffered to a pH of 7, by measuring the percentage of oil droplets remaining after the initial time point. In general, the percentage of oil droplets remaining decreased because oil droplet size decreased, and the particles coalesced over time. Gravity separation occurred over time in all samples because the density of oil (ρ =1.1562 g/cm3) is heavier than liposome solutions (ρ≈ 1 g/cm3). Particles smaller than 1 µm dispersed in the supernatant were removed, and particles larger than 1 µm were counted. Microfluidic flow-focusing produced oil droplets with greater stability than those produced via vortexing. Polymerized 25 DA (P=0.004) and unpolymerized 25 DA (P=0.037) were both significantly more stable at the five hour mark when fabricated using microfluidic flow-focusing compared to vortexed droplets comprised of the same lipid shell (Figure 6). The vortexing method created a wide size distribution of droplets as seen in the bright-field image (Figure 5), which causes instability via Ostwald ripening or bridging. The analysis of Ostwald ripening is addressed further in the Discussion section.

3.3.2. Effect of Polymerizable Lipid Coating on Stability Polymerization of the lipid coating for the 25 DA formulations was achieved through ultraviolet irradiation, which results in cross-linking of diacetylene residues present in lipid acyl chains, leading to highly colored blue polymers.20 Polymerization of the lipid coating for the 25 DA formulations resulted in enhanced stability of the droplets with the greatest stability


18

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 20 of 32

exhibited by droplets produced through flow-focusing. Comparing only among droplets fabricated by microfluidics, ANOVA revealed a significant difference (P = 0.0001) on droplet stability as a result of lipid formulation. Both polymerized 25 DA (84.7±1.2%) and nonpolymerized 25 DA shelled (77.7±2.7%) oil droplet formulations, with less than 25% population reduction over five hours, were significantly more stable than non-polymerizable DSPC/DPSEPEG (64.9±2.7%) shelled oil droplets, which suffered a 35% reduction in droplet population over the course of five hours. Additionally, polymerization of the 25 DA lipids further enhanced droplet stability, indicated by the statistically significant increase in the percentage of droplets remaining over non-polymerized 25 DA from the 180 min time point onwards. While 25 DA formulations maintained a higher percentage of droplets remaining among flow-focusing fabricated oil droplets, there was no statistical difference in the percentage of droplets remaining among lipid formulations when vortexed (ANOVA, P=0.847). The difference in the remaining percentage of polymerized 25 DA droplets (62.1±2.2%) was not significantly higher than either unpolymerized 25 DA (57.9±3.3%) or non-polymerizable DSPC/DSPE-PEG (56.5±5.8%) formulations. In other words, the effect of lipid formulation was not significant among the vortexed samples. Overall, this study demonstrates that the stability of micron-sized oil droplet is dependent both on production method and shell materials. For the polymerized-shell droplets, the percent remaining after 5 hr of flow-focusing vs. vortex is 85% vs. 62%. Also, when the best (flowfocusing with polymerized) is compared with the worst (vortex with non-polymerizable), it improved by 30% (85% vs 55%).


19

Page 21 of 32

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

Figure 6. Comparison of the effect of fabrication method and polymerization on droplet stability in 150 mM NaCl pH 7. Triacetin oil droplets were fabricated using either microfluidic flow focusing (FF solid lines) or by vortexing (V dotted lines). Error bars represent standard deviation in percentage of oil droplets remaining at each time point and an asterisk (*) represents p < 0.05 when comparing FF to V for each lipid formulation.

4. Discussion 4.1. Flow regimes In general, for all flow rates tested, it was found that the relationship between Vd and the ratio of the flow rate of continuous (Qo) and disperse phases (Qi) was of the form Vd ∝ (Qo/Qi)-1. This finding is consistent with the results from Nie et al. where the droplet break-up regime was


20

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 22 of 32

identified as being controlled by the rate of flow which occurs for low-to-moderate values of the viscosity of the inner phase; ηi/ηo = 10 for Nie et al., and ηi/ηo = 8.3 in our experiments. On the other hand, when the viscosity of inner phase is high3, the droplet break-up is due to shearing, and the correlation becomes Vd ∝ (Qo/Qi)-3. In this mode, the shear stress of the outer fluid overcomes interfacial tension; therefore, there is a stronger dependence of Vd on the capillary number, = /ℎ, where is the width of the orifice, is the flow rate, ℎ is the height of the channel, and γ is interfacial tension where for triacetin and water, = 35.8 mN/m.21 It is instructive to compare the results from this study to our previous study3 with gasfilled microbubbles produced by microfluidic flow-focusing. For monodisperse microbubbles, the size was controlled by the flow rate and gas pressure or the ratio of the two, but the break-up always occurred at the orifice. In other words, the jetting regime, where the droplet forms downstream of the orifice, never occurred for microbubble production. Also microbubble size decreased as a function of the continuous phase flow rate or the ratio of the flow rate and the gas pressure until the size became the same or smaller than the orifice size. In contrast, here it was found that the oil droplet size produced by the same microfluidic flow-focusing device in the dripping regime was related to the channel height and appeared to be independent of the orifice size. Moreover, an oil thread was observed even in the dripping regime, which suggests that the dripping is governed by a Rayleigh-Plateau instability after the orifice – not by pinching at the orifice. This difference in break-up regimes between gas bubbles and oil droplets is likely because the gas is more compressible than the oil.22 For every system used in the present study, we determined the range of flow rates and the flow rate ratios of the two immiscible liquids within which a single population of droplets with a narrow size distribution was formed. In contrast to Nie et al. who reported high polydispersity


21

Page 23 of 32

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

for droplets produced in the jetting regime, we observed that droplets had relatively low polydispersity (around 10%). We speculate that this is because for our device the drop size was governed by the channel height, which was constant, and so any fluctuations in flow parameters had little impact on the droplet size.

4.2. Stability of Micro-oil Droplets

The results reported here demonstrated that the use of flow-focusing (instead of vortexing) combined with polymerization of the shell resulted in oil droplets with longer life times. The microfluidic flow-focusing likely improved stability by producing oil droplets with greater monodispersity that enabled uniform lipid packing on the surface of the droplets, effectively stabilizing them from dissolution. The uniformity of the flow-focused droplets and the better consistency of the lipid coating would also have resulted in improved polymerization and hence, enhanced life-time. When volume of liquids, not the number of droplets, is a controlling factor, vortexing produced a higher number of droplets than flow-focusing because it created a large number of small droplets, ranging from sub-micron to 5 µm. Thus, vortexing creates a droplet population with a higher surface to volume ratio, a larger total surface area, than flow-focusing does. This may lead to a lower amount of surfactant molecules that stabilize each particle. This could be the reason why conventional microemulsification process often requires large amounts of surfactants and co-surfactants. Furthermore, the broad size distribution produced by vortexing encourages Ostwald ripening, a phenomenon by which smaller droplets dissolve and redeposit on the surface of large


22

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 24 of 32

droplets. Triacetin oil has relatively great solubility in water (70 g/L); therefore, it is relatively easy for small droplets to dissolve and re-deposit on the surfaces of large droplets if the surface tension is sufficiently large. This process will reduce the apparent number of particles that are larger than 1 µm over time because droplets smaller than 1 µm could not be detected by the imaging system. The evolution of the average droplet radius 3 from Ostwald ripening when diffusion is limited is given by:

(1) where, t is time, 03 is the initial mean number droplet radius, γ is the surface tension of the oil-water interface, c∞ is the molecular solubility of the solute, υ is the molar volume of the solute, D is the diffusion coefficient of the solute, T is the absolute temperature, and Rg is the molar gas constant.23 The average radius was determined from images of the droplets and fit to Eq 1 (Figure 7). Micro-oil droplets produced by flow-focusing did not show size changes over time, that is, 3 - 03 for the 5 hr of observation. In contrast, droplets produced by vortexing increased bilinearly with time; the initial linear slope was steeper in the first hour, and then the slope decreased to a lower value. These observations correspond well qualitatively to the slopes of Percentage Remaining (%), which is a steep drop within 1 hr, followed by a gentler drop (Figure 6). Of note, the R2 (coefficient of determination) value for 85% DSPC + 15% DSPE-PEG 5K oil droplets vortexed was equal to 1 within 1 hr, and 0.96 after 1 hr. A comparison of the rates of size increment among all vortexed conditions reveals that 85% DSPC + 15% DSPE-PEG 5K oil droplets exhibited the greatest slope within 1 hr and that 25 DA polymerized the least. In terms of stability data (Figure 6), there seems to be no difference among the vortexed conditions.


23

Page 25 of 32

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

Figure 7. Size changes of micro-oil droplets produced by vortexing over time: 3 - 03 for the 5 hr of observation. Micro-oil droplets produced by flow-focusing did not show size changes over time, which was near zero for 3 - 03. Although the effect of polymerization on Percent Remaining was not strongly observed over time in vortexed droplets, the effect of 25 DA Polymerized on suppressing size increment via Ostwald ripening was significant. This implies that the polymerization on the surface of big droplets prevents the oil molecules from adsorbing. In addition, the reason why Percent Remaining for 25 DA Polymerized vortexed still showed a similar value to the other conditions may be due to instability of small droplets. The small droplets are less stable than large ones because a large surface pressure across the oil-water interface is applied due to their small radius according to Young-Laplace equation. To the best of our knowledge, there are no reports in the literature investigating effects of production methods of o/w emulsion on droplet stability. Assessing stability is a very important


24

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 26 of 32

issue if these droplets are to be used as drug carriers as long-term storage is a critical factor. For example there have been many studies on the stability of gas bubbles (produced by microfluidics) against gas dissolution24-26. We hypothesize that the stabilization mechanism of oil droplets against dissolution can be attributed to a uniform packing of lipids on the boundary—this is likely the same mechanism for gas bubbles. We note that there are studies that report photopolymerization on the lipid membrane facilitates release of liposomal contents.27,28 In this case, these findings are attributed to the formation of polymerized lipid domains, which causes enhanced permeability. However, the lipid compositions in these papers are different from the lipid composition used in this study. Our lipid composition may form more homogenous shells in comparison. This is supported by our previous study showing that the presence of a polymerized shell increased shell resistance against gas dissolution across the lipid membrane; the same lipid composition was used as in this study.14 Another important point to note is that “microemulsion” is very different from ‘micronsized emulsion’ that we report in this study in terms of stability, size, and composition. It is possible to generate thermodynamically stable o/w colloidal systems known as ‘microemulsions’, but this process often requires large amounts of surfactants and co-surfactants at high concentrations that can be toxic to cells or tissues. Our oil droplet sizes are 11.1 µm (σ = 8.81%) with flow-focusing, and 15.0 µm (σ = 83.2%) with vortexing. On the other hand, “microemulsion” droplet size varies from 1 to 100 nm, usually 10 to 50 nm (IUPAC definition). In this study, we excluded particles smaller than 1 µm for the stability test. When emulsions are produced by bulk emulsification methods, such as homogenization or sonication, several studies mentioned that emulsions with oil > 80 ~ 90 mol% are not stable;


25

Page 27 of 32

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

therefore, it is necessary to employ solutions where the phospholipid concentration is greater than 10~ 20 mol% to make the emulsion stable12,13. However, in this study, the lipid concentration was just 1 mol% and yet stable oil droplets were produced by microfluidics. The use of low mol% of lipids is particularly attractive in pharmaceutical applications because high emulsifier concentration levels are costly and can often be toxic, causing side effects. There are several other factors to consider with respect to the emulsion stability such as size, the solubility of dispersed phase in the continuous phase29 and the degree of lipid adsorption. The triacetin used for this study has relatively high solubility in water. Therefore, the effect of the type of dispersed phase, pH, and temperature on emulsion stability should be further investigated. The long-term goal of this study is to develop a stable emulsion that consists of monodisperse micro-oil droplets with polymerized lipid coatings as potential drug carriers. To produce monodisperse oil droplets, we have used microfluidic flow-focusing devices to quantitatively study droplet sizes and flow regimes as a function of the ratio of flow rates of continuous and dispersed phases. The results presented here provide a framework for engineering vehicles for future drug release studies.

5. Conclusions Our work shows that, for the liquids tested, microfluidic flow-focusing is an efficient method for producing micro-oil droplets with narrow size distributions. Narrow size distributions are important for developing drug carriers with predictable drug release profiles and better stability. The droplet production was shown to occur in two regimes: a jetting regime and a dripping regime. The design and protocols here resulted in droplets with a size governed by the


26

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

channel height that makes production robust to variations in flow properties.

Page 28 of 32

The results of this

study can be helpful for a rational design of microfluidic devices and selection of liquids with suitable viscosity and interfacial tension for controlled formation of droplets with a predetermined size and a narrow size distribution. We have verified the advantage of using microfluidic flow-focusing for stable oil droplets by comparing stability with oil droplets produced by vortexing. This result provides the potential of using flow-focusing as a method for encapsulating hydrophobic and relatively expensive pharmaceutical agents for precise dosages. Finally, microfluidics combined with polymerizable lipids show promise for development of controlled drug release vehicles.

Acknowledgments We acknowledge the Optoelectronic Processing Facility in the Photonics Center and the Biomedical Engineering Core Facilities at Boston University. J.Y.W. acknowledges support from the Boston University College of Engineering Distinguished Faculty Fellowship. We thank Ms. Anna Cristina Shivers and Ms. Diana Miniovich, for assistance in device and sample preparation, and the Boston University UROP program for partial funding of Ms. Anna Cristina Shivers.

References


27

Page 29 of 32

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) Bagwe, R. P.; Kanickly, J. R.; Palla, B. J.; Patanjali, P. K.; Shah, D. O. Improved Drug Delivery Using Microemulsions: Rationale, Recent Progress, and New Horizons. Crit. Rev. Ther. Drug Carrier Syst. 2001, 18, 77-140. (2) Trotta, M.; Pattarino, F.; Ignoni, T. Stability of Drug-carrier Emulsions Containing. Eur. J. Pharm. Biopharm. 2002, 53, 203–20. (3) Bibette, J.; Morse, D. C.; Witten, T. A.; Weitz, D. A. Stability Criteria for Emulsions, Phys. Rev. Lett. 1992, 69, 2439-42. (4) Lu, F. S. H.; Nielsen, N. S.; Baron, C. P.; Jensen, L. H. S.; Jacobsen, C. Physico-chemical Properties of Marine Phospholipid Emulsions. J. Am. Oil Chem. Soc. 2012, 89, 2011-24. (5) Asai, Y.; Watanabe, S. Interaction of sesame oil with soybean phosphatidylcholine and their formation of small dispersed particles. J. Microencapsulation, 1999, 16, 705-13. (6) Sznitowska, M.; Janicki, S.; Dabrowska, E.; Zurowska-Pryczkowska, K. Submicron Emulsions as Drug Carriers Studies on Destabilization Potential of Various Drugs. Euro. J. Pham. Sci. 2001, 12, 175-9. (7) De Geest, B. G.; Urbanski, J. P.; Thorsen, T.; Demeester, J.; De Smedt, S. C. Synthesis of Monodisperse Biodegradable Microgels in Microfluidic Devices, Langmuir 2005, 21, 1027510279. (8) Ikkai, F.; Iwamoto, S.; Adachi, E.; Nakajima, M. New Method of Producing Mono-Sized Polymer Gel Particles Using Microchannel Emulsification and UV Irradiation. Colloid Polym. Sci. 2005, 283, 1149-1153.


28

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 30 of 32

(9) McClements, D. J. Nanoemulsions versus Microemulsions: Terminology, Differences, and Similarities, Soft Matter, 2012, 8, 1719-29. (10) Teh, S.-Y.; Lin, R.; Hung, L.-H.; Lee, A. P. Droplet Microfluidics, Lab Chip, 2008, 8, 198220. (11) Nie, Z.; Seo, M. S.; Xu, S.; Lewis, P. C.; Mok, M.; Kumacheva, E.; Whitesides, G. M.; Garstecki, P.; Stone, H. A. Emulsification in a Microfluidic Flow-Focusing Device: Effect of the Viscosities of the Liquids. Microfluid .Nanofluid. 2008, 5, 585–594. (12) Cubaud, T.; Mason, T. G. Capillary Threads and Viscous Droplets in Square Microchannels Phys. Fluids. 2008, 20, 053302. (13) Yobas, L.; Martens, S.; Ong, W.-L.; Ranganathan, N. High-performance Flow-focusing Geometry for Spontaneous Generation of Monodispersed Droplets. Lab Chip, 2006, 6, 1073– 1079. (14) Park, Y.; Luce, A. C.; Whitaker, R. D.; Amin, B.; Cabodi, M.; Nap, R. J.; Szleifer, I.; Cleveland, R. O.; Nagy, J. O.; Wong, J. Y. Tunable Diacetylene Polymerized Shell Microbubbles as Ultrasound Contrast Agents, Langmuir 2012, 28, 3766−3772. (15) Xia, Y. N.; Whitesides, G. M. Soft Lithography. Angew. Chem. Int. Ed. 1998, 37, 551-575. (16) Hettiarachchi, K.; Lee, A. P.; Zhang, S.; Feingold, S.; Dayton, P. A. Controllable Microfluidic Synthesis of Multiphase Drug-Carrying Lipospheres for Site-Targeted Therapy. Biotechnol. Prog. 2009. 25, 938-945.


29

Page 31 of 32

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) Hettiarachchi, K.; Talu, E.; Longo, M. L.; Dayton, P. A.; Lee, A. P. On-chip Generation of Microbubbles as a Practical Technology for Manufacturing Contrast Agents for Ultrasonic Imaging. Lap Chip, 2007, 7, 463-8. (18) Garstecki, P.; Stone, H. A.; Whitesides, G. M. Mechanism for Flow-Rate Controlled Breakup in Confined Geometries:A Route to Monodisperse Emulsions. Phys. Rev. Lett. 2005, 94, 164501. (19) Changenet-Barret, P.; Gustavsson, T.; Markovitsi, D.; Manet, I.; Montib, S. Unravelling molecular mechanisms in the fluorescence spectra of doxorubicin in aqueous solution by femtosecond fluorescence spectroscopy. Phys. Chem. Chem. Phys. 2013, 15, 2937-2944. (20) Hupfer, B.; Ringsdorf, H.; Schupp, H. Liposomes from Polymerizable Phospolipids, Chem. Phys. Lipids. 1983, 33, 355-374. (21) Mulligan, M. K. Morphology and Development of Droplet Deformation Under Flow Within Microfluidic Devices. 2012, Open Access Dissertations, Paper 519 (22) Meriani, F.; Coceani, N.; Sirotti, C.; Voinovich, D.; Grassi, M. Characterization of a Quaternary Liquid System Improving the Bioavailability of Poorly Water Soluble Drugs. J. Colloid Interface Sci. 2003, 263, 590-596. (23) Baldan, A. Review Progress in Ostwald ripening theories and their applications to nickel base superalloys Part 1: Ostwald ripening theories. J. Mater. Sci. 2002, 37, 2171-2202. (24) Talu, E.; Lozano, M. M.; Powell, R. L.; Dayton, P. A.; Longo, M. L. Long-Term Stability by Lipid Coating Monodisperse Microbubbles Formed by a Flow-Focusing Device. Langmuir. 2006, 22, 9487-90.


30

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 32 of 32

(25) Borden, M. A.; Longo, M. L. Dissolution Behavior of Lipid Monolayer-Coated, Air-Filled Microbubbles: Effect of Lipid Hydrophobic Chain Length. Langmuir, 2002, 18, 9225-33. (26) Mohamedi, G.; Azmin, M.; Pastoriza-Santos, I.; Huang, V.; Pérez-Juste, J.; Liz-Marzán, L. M.; Edirisinghe, M.; Stride, E. Effects of Gold Nanoparticles on the Stability of Microbubbles. Langmuir, 2012, 28, 13808-15. (27) Spratt, T.; Bondurant, B.; O'Brien, D. F. Rapid release of liposomal contents upon photoinitiated destabilization with UV exposure. Biochim Biophys Acta. 2003, 1611, 35-43. (28) Punnamaraju, S.; You, H.; Steckl, A.J. Triggered Release of Molecules across Droplet Interface Bilayer Lipid Membranes Using Photopolymerizable Lipids, Langmuir, 2012, 28, 7657–7664. (29) Tadros, T. F. 2005 Applied Surfactants Principles and Applications, Weinheim Germany: Wiley-VCH.


31

Monodisperse Micro-Oil Droplets Stabilized by Polymerizable

Recommend Documents