Optimization of ethane recovery using the CRR process shows that, except for the case of lean gas at low demethanizer pressure, the CRR process reduces to GSP, in which there is no reflux stream and therefore no added cryogenic compression and heat exchange equipment. Adding a second cold separator, operating at lower temperature, in GSP is found to lead to more or less recovery depending on the NGL content of the feed gas and the demethanizer pressure. GSP is also compared with the conventional turboexpander process. Optimization shows that adding more equipment or even flow splitting may lead to less ethane recovery.

To minimize the adverse side effects of conventional chemotherapy, a targeted micellar drug carrier was investigated that retains hydrophobic drugs in its core and then releases the drug via ultrasonic activation. This paper compares the percent drug release from folated versus non-folated micelles by insonation at 70 kHz and different acoustic power densities. The encapsulated drug is Doxoru- bicin (Dox). A physical model of zero-order release with first-order re-encapsulation was used to fit the experimental kinetic data. Additionally, the acoustic activation power density and Gibbs free energy were introduced and calculated for folated and non-targeted micelles. The data suggests an important role of inertial cavitation in drug release and the presence of a power density threshold for inertial cavitation.

SUNMMARY To minimize the adverse side effects of conventional chemotherapy, a targeted micellar drug carrier was investigated that retains hydrophobic drugs in its core and then releases the drug via ultrasound. Here we report on the amount of drug release from folated versus non-folated micelles at 70 kHz and different acoustic power densities. The encapsulated drug is Doxorubicin (Dox). The data suggests an important role of inertial cavitation in ultrasonic drug release due to the Observation of a power density threshold for drug delivery. INTRODUCTION Conventional chemotherapy has always had debilitating drawbacks; therefore upon discovery that chemotherapy drugs could be sequestered inside a nanocarrier that would release them only when and where needed, cancer research was presumed to have taken a giant leap forward. Hence, it is important to synthesize drug carriers capable of minimizing the adverse side effects of chemotherapy by preferentially targeting tumor cells both actively (e.g. folate receptor) and using external stimulus (ultrasound). The purpose of this study is to quantify the percent of acoustic release of Dox from folate-targeted micelles (P105-FA) and compare their release behavior to that of non-targeted polymeric micelles. EXPERIMENTAL METHODS The chamber built to measure the change of fluorescence 1 (and hence the amount released in the presence and absence of ultrasound) employed an argon ion laser (Ion Laser Technology, Model 5500 A) mounted on an optical bench. The laser beam was directed to a beam splitter attenuator (metal film neutral density attenuator). The intensity of the split portion of the beam was measured by a photo detector and is used to monitor the laser power. The drug concentration was quantified by measuring the fluorescence emissions produced by an excitation wavelength of 488 nm. A fiber optic probe (100 bundled multinode fibers, approximately 40 cm in length) was used to collect fluorescence emissions. The emitted light was directed through a multinode dielectric band and filter (Omega Optical Model 535DF35) to a silicon detector (Model EGSG). The filter was used to cut off any emissions with a wavelength below 535 nm. The detector signal was digitized with an A/D converter (National Instruments) and sent to a computer for storage and processing. To mimic physiological conditions, the temperature of the ultrasonic exposure chamber was maintained at 37 o C using a thermostated bath. The chamber described above was used to measure the kinetics of acoustically activated drug release from micelles. Dox exhibits a large decrease in fluorescence when transferred from the hydrophobic core of the micelle to the surrounding aqueous solution. Therefore, the release can be determined by measuring the decrease in fluorescence intensity. P105-FA was synthesized using 1,1carbonyldiimidazole (CDI, Sigma Aldrich). 2 Folic acid (Sigma Aldrich) was dissolved in dried DMSO. CDI was added and allowed to react for 4 hours under dark conditions at room temperature. After this activation of the FA, Pluronic P105 (dried overnight under vacuum) was added. The activated CDI and Pluronic P105 were reacted for 20 hours at room temperature in darkness. Next, the product was dialyzed (Spectra Millipore MWCO 3500) against DMSO for 2 days and then against double distilled water for 2 days. The purified product was then lyophilized and stored at -20°C. The formation of P105-FA was confirmed using NMR. For release experiments, P105-FA was dissolved in DD-water to make a final concentration of 4.5 wt. %. Dox (Sigma Aldrich) was introduced into the micelles by mixing at room temperature. Using dynamic light scattering, the size of these micelles was measured to be 10.2 + 2.2 nm. RESULTS AND DISCUSSION Fig. 1 summarizes the percent of drug release from folate-targeted and non-targeted micelles as a function of acoustic power density. At low power densities, the measured drug release is very small and cannot be distinguished from the release in conditions of no ultrasound. This trend continues up till about 0.5 W/cm 2 , after which the release from folated micelles follows an almost linear increase with the power densities up until about 3 W/cm 2 which constitutes the second region on the plot. Meanwhile, the release from non-targeted micelles is not quite linear in this power density region. After 1.03 W/cm 2 , however, the release from untargeted micelles is always lower than the release from the folated micelles. In the third region, above 3 W/cm 2 , both types of micelles demonstrate a fairly constant amount of release. For folated micelles, this amounts to approximately 13%, while for the untargeted micelles it is only about 10%. Figure 1. Percent release of Dox from targeted and un-targeted micelels as a function of acoustic power density at 70 kHz. The data show the existence of two thresholds, 0.5 W/cm 2 and 3 W/cm 2 , which suggest the role of inertial cavitation in the drug release. The lower threshold value is believed to signify the onset of inertial cavitation in water. While there is yet no explanation for the upper threshold, it has been reported in other studies. 3,4 At the lower threshold, a possible explanation is that onset of inertial cavitation produces shock waves that rupture the micelles, resulting in the release of the encapsulated drug into the aqueous environment. It can be hypothesized that the higher release from folated micelles can be attributed to the compromised integrity of the micelle structure due to the attached folic acid; i.e, the attached moiety has made the micelles more leaky when subjected to cavitation pressure waves and shear stresses from lowfrequency ultrasound. The non-parametric Mann-Whitney U test was conducted on the folated and non-targeted micelle release data for each power density in order to assess the statistical significance of experimental data. The Mann-Whitney test was chosen because the data were from two unpaired groups and did not follow a Gaussian distribution. The lower power densities did not have enough data points to give statistically reliable results, but power densities of 1.03 W/cm 2 and higher all gave statistically significant levels (α = 0.05). Therefore, drug release from targeted micelles is statistically significantly higher than release from non-targeted carriers above 1.03 W/cm 2 . CONCLUSION The ultrasound-triggered drug release from folate-targeted Pluronic P105 micelles loaded with Doxorubicin (Dox) has been investigated and the results compared to the release from non-targeted micelles under the same conditions of low-intensity ultrasound at different acoustic power densities. It was found that folated micelles exhibit greater percent of drug release. Furthermore, the presence of a power density threshold indicates that inertial cavitation plays a role in acoustic drug release from micelles. In vivo work is currently being conducted to test the feasibility of using folated micelles in combination with ultrasound as a drug delivery system. REFERENCES 1. G.A. Husseini, G. D. Myrup, W. G. Pitt, D.A. Christensen, N. Y. Rapoport, Journal of Controlled Release, 2000, 69, p. 43-52. 2. G.A. Husseini, D. Velluto, L. Kherbeck, W.G. Pitt, J.A. Hubbell, D.A. Christensen, Colloids and Surfaces B, 2012, 101(1), p. 153-155. 3. G.A. Husseini, M.A. Diaz, E.S. Richardson, D.A. Christensen, W.G. Pitt, J. Controlled Release, 2005, 107, p. 253-261. 4. G.A. Husseini, M.A. Diaz, Y. Zeng, D.A. Christensen, W.G. Pitt, Journal of Nanoscience and Nanotechnology, 2007, 7 p. 1-6. ACKNOWLEDGEMENTS We thank the American University of Sharjah sabbatical funds, the AUS Faculty Research Grant (FRG-AUS11) and the Pope Fellowship of Brigham Young University for funding this research. 0 2 4 6 8 10 12 14 0 1 2 3 4 5 6 % R el ea se Power Density (W/cm2) PF

Ultrasound is an ideal trigger for site-actuated drug delivery because it can be focused through the skin to internal targets without surgery. Thermal or mechanical energy can be delivered via tissue heating or bubble cavitation, respectively. Bubble cavitation, which concentrates energy that can trigger drug release from carriers, occurs more readily at low frequencies and at bubble resonant frequencies. Other mechanical and physical consequences of cavitation are reviewed. Micelles are nanosized molecular assemblies of amphiphilic molecules that spontaneously form in aqueous solution and possess a hydrophobic core capable of sequestering hydrophobic drugs. Micelles have traditionally been used to increase the solubility of hydrophobic therapeutics for oral and intravenous administration. For ultrasonic drug delivery, polymeric micelles containing polyethylene oxide blocks are preferred because they have longer circulation time in vivo. Passive delivery occurs when micelles accumulate in tumor tissues that have malformed capillaries with porous walls. In active delivery targeting ligands are attached to the micelles, which directs their binding to specific cells. Actuated delivery occurs when ultrasound causes drug release from micelles and is attributed to bubble cavitation since the amount released correlates with acoustic signatures of cavitation. The mechanisms of ultrasonic drug release are discussed, including the prevalent theory that gas bubble cavitation events create high shear stress and shock waves that transiently perturb the structure of the micelles and allow drug to escape from the hydrophobic core. Ultrasound also perturbs cell membranes, rendering them more permeable to drug uptake. Tumors in rats and mice have been successfully treated using low-frequency ultrasound and chemotherapeutics in polymeric micelles. Ultrasonically activated drug delivery has great clinical potential.

The main problem associated with the administration of anti-cancer medication is that the drug is delivered throughout the body causing undesirable side effects. Therefore, it is important to synthesize drug carriers capable of minimizing the adverse side effects of chemotherapy by preferentially targeting tumor cells both actively (e.g. a folate receptor) and using external stimulus (e.g. ultrasound). In this paper, we report the synthesis of Pluronic P105 micelles with a folate targeting moiety (with a yield of 48%) containing doxorubicin (Dox). We applied low frequency ultrasound as an external stimulus and measured the amount of release of Dox from these folated micelles. The results showed that the percent drug release increases as the power intensity of ultrasound increases. The maximum amount of release (14%) was measured at 5.4 W/cm2. A power density threshold at approximately 0.55 W/cm2 exists below which no statistically significant release was observed. This lower threshold suggests that cavitation plays an important role in triggering drug release from targeted micelles.

.......................................................................................................................................... 5 List of Tables .................................................................................................................................. 7 List of Figures ................................................................................................................................. 8 Chapter 1: Introduction ................................................................................................................... 9 Chapter 2: Literature Review ........................................................................................................ 12 2.1 Micelles ................................................................................................................... 12 2.2 Drug Targeting ........................................................................................................ 14 2.3 Triggered Targeting................................................................................................. 15 2.4 In Vitro .................................................................................................................... 19 2.5 In Vivo..................................................................................................................... 20 Chapter 3: Objectives .................................................................................................................... 25 Chapter 4: Data Analysis .............................................................................................................. 26 4.1 Measuring Release from Targeted and Non-Targeted Micelles ............................. 26 4.2 Materials and Methods ............................................................................................ 28 4.2.1 Acoustic Measurements .................................................................................... 29 4.3 Results and Discussion ............................................................................................ 31 4.3.1 Comparison of release from folated and non-folated micelles ......................... 31 4.3.2 Testing statistical significance .......................................................................... 33 Chapter 5: Mathematical Modeling .............................................................................................. 35 5.1 Model 1: first-order release and first-order re-encapsulation.................................. 35 5.2 Model 2: zero-order release and first-order re-encapsulation ................................. 38 5.3 Data Fitting .............................................................................................................. 40 5.4 Acoustic Activation Power Density and Gibbs Free Energy .................................. 46 Chapter 6: Modeling Dox Release Using an Artificial Neural Network ...................................... 50 Chapter 7: Conclusion................................................................................................................... 54 Appendix ....................................................................................................................................... 61 Vita ................................................................................................................................................ 62