Unique Mechanisms of Colloidal Stability Probed by Surface-Specific Vibrational Spectroscopy

by 

Ashley N. Mapile 

A dissertation accepted and approved in partial fulfillment of the  

requirements for the degree of 

Doctor of Philosophy  

in Chemistry

Dissertation Committee: 

Jeffrey Cina, Chair 

Lawrence Scatena, Advisor  

Carl Brozek, Advisor

Cathy Wong, Core Member 

Daniel Steck, Institutional Representative 

University of Oregon 

Summer 2024



 

2 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

© 2024 Ashley N. Mapile  

 This work is openly licensed via CC BY-NC 4.0 



 

 

 

DISSERTATION ABSTRACT 

 

Ashley N. Mapile 

 

Doctor of Philosophy in Chemistry 

 

Title: Unique Mechanisms of Colloidal Stability Probed by Surface-Specific Vibrational 

Spectroscopy 

 

 

The stability of nanoparticles suspended in a solution, known as colloids, is crucial for their 

application in drug delivery systems, the solution processibility required for drop-casting films, 

and the long-term storage or transport of sensitive chemical materials. While current mechanisms 

for colloidal stability include implicit models of solvation – namely Derjaguin-Landau-Verwey-

Overbeek (DLVO) and Flory-Huggins theories – these classical approaches neglect the role of 

specific solvent-surface interactions. Analyzing these surface-specific interactions in colloidal 

stability becomes increasingly relevant for nanosized particles, which have a highly accessible 

surface area compared to their bulk counterparts.  

This dissertation seeks to understand unconventional mechanisms of colloidal stability that 

are not explained by traditional theories alone, with oil-in-water emulsions and nanoparticles of 

metal-organic frameworks (nanoMOFs) as model materials. Leveraging the surface-specific 

spectroscopic technique, vibrational sum frequency scattering spectroscopy (VSFSS), this work 

provides a molecular-level understanding of the specific surface interactions that contribute to 

colloidal stability. In particular, emulsions can be stabilized by a steric layer of polymer alone, 

with colloidal behavior tunable by pH, electrolyte concentration, molecular weight, and 

temperature. These sterically-stabilized emulsions find applications in drug delivery systems that 

must withstand extreme physiological conditions. For bare nanoMOFs, an ordered solvation shell 

and solvent-metal surface binding contribute to unforeseen long-term stability in common 

solvents. Additionally, nanoMOFs coated with a polymeric binding agent – similar to those used 

in the paint industry – yield ultra-strong mixed-matrix membranes for gas separation technologies. 

Ultimately, this work bridges molecular interfacial chemistry with material properties, 

emphasizing the importance of understanding mechanisms of colloidal stability.  

This dissertation includes previously published co-authored material. 

  



 

2 

 

 

CURRICULUM VITAE 

 

 

NAME OF AUTHOR:  Ashley N. Mapile 

 

GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: 

 

 University of Oregon, Eugene, OR 

 University of Puget Sound, Tacoma, WA 

 

DEGREES AWARDED: 

 

 Doctor of Philosophy, Physical Chemistry, 2024, University of Oregon 

 Master of Science, Physical Chemistry, 2021, University of Oregon 

 Bachelor of Science, Chemistry, 2019, University of Puget Sound 

 

AREAS OF SPECIAL INTEREST: 

 

 Vibrational Spectroscopy 

 Colloids and Soft Matter 

 Interfacial Chemistry 

 

PROFESSIONAL EXPERIENCE: 

 

 Co-Chair, Gordon Research Seminar in Vibrational Spectroscopy, August 2024 

 Associate Editor, Journal of Emerging Investigators, 2023-Present  

 Newsletter Complier, Society for Applied Spectroscopy, 2023-Present 

Graduate Research Assistant, University of Oregon Department of Chemistry and 

Biochemistry, 2019-2024 

Graduate Teaching Assistant, University of Oregon Department of Chemistry and 

Biochemistry, 2019-2024  

Undergraduate Researcher, National Science Foundation Research Experience for 

Undergraduates (NSF-REU), University of Oregon, Summer 2018 

 Undergraduate Researcher, NSF-REU, University of Washington, Summer 2017 

 

GRANTS, AWARDS, AND HONORS: 

 

Honorable Mention, National Science Foundation Graduate Research Fellowship Program, 2021 

 

PUBLICATIONS: 

 

1. Mapile, A. N.; Scatena, L.F. Bulking Up: The Impact of Polymer Sterics on Emulsion 

Stability. In Review. 2024  

 

2. Mapile, A. N.; Svensson Grape, E.; Brozek, C. K. Solvation of Nanoscale Materials. 

Chem. Mater. 2024. https://doi.org/10.1021/acs.chemmater.4c01518. 

https://doi.org/10.1021/acs.chemmater.4c01518


 

3 

 

 

 

3. Mapile, A. N.; LeRoy, M. A.; Fabrizio, K.; Scatena, L. F.; Brozek, C. K. The Surface of 

Colloidal Metal–Organic Framework Nanoparticles Revealed by Vibrational Sum 

Frequency Scattering Spectroscopy. ACS Nano 2024, 18 (20), 13406–13414. 

https://doi.org/10.1021/acsnano.4c03758. 

 

4. Mapile, A. N.; Scatena, L. F. Stabilizing Strands: Exploring the Kinetic Stability of 

Polymer-Coated Emulsions with Surface Specific Spectroscopy. Colloids Surfaces A 

2024, 697, 134414. https://doi.org/10.1016/j.colsurfa.2024.134414. 

 

5. LeRoy, M. A.; Perera, A. S.; Lamichhane, S.; Mapile, A. N.; Khaliq, F.; Kadota, K.; 

Zhang, X.; Ha, S.; Fisher, R.; Wu, D.; Risko, C.; Brozek, C. K. Colloidal Stability and 

Solubility of Metal–Organic Framework Particles. Chem. Mater. 2024. 

https://doi.org/10.1021/acs.chemmater.3c03191. 

 

6. Huang, J.; Marshall, C. R.; Ojha, K.; Shen, M.; Golledge, S.; Kadota, K.; McKenzie, J.; 

Fabrizio, K.; Mitchell, J. B.; Khaliq, F.; Davenport, A. M.; LeRoy, M. A.; Mapile, A. N.; 

Debela, T. T.; Twight, L. P.; Hendon, C. H.; Brozek, C. K. Giant Redox Entropy in the 

Intercalation vs Surface Chemistry of Nanocrystal Frameworks with Confined Pores. J. 

Am. Chem. Soc. 2023, 145 (11), 6257–6269. https://doi.org/10.1021/jacs.2c12846. 

 

7. Carpenter, A. P.; Christoffersen, E. L.; Mapile, A. N.; Richmond, G. L. Assessing the 

Impact of Solvent Selection on Vibrational Sum-Frequency Scattering Spectroscopy 

Experiments. J. Phys. Chem. B 2021, 125 (12), 3216–3229. 

https://doi.org/10.1021/acs.jpcb.1c00188. 

 

8. Tran, E.; Mapile, A. N.; Richmond, G. L. Peeling Back the Layers: Investigating the 

Effects of Polyelectrolyte Layering on Surface Structure and Stability of Oil-in-Water 

Nanoemulsions. J. Colloid Interface Sci 2021, 599, 706–716. 

https://doi.org/10.1016/j.jcis.2021.04.115. 

 

9. Wilson, K. S.; Mapile, A. N.; Wong, C. Y. Broadband Single-Shot Transient Absorption 

Spectroscopy. Opt. Express, OE 2020, 28 (8), 11339–11355. 

https://doi.org/10.1364/OE.390938. 

 

 

https://doi.org/10.1021/acsnano.4c03758
https://doi.org/10.1016/j.colsurfa.2024.134414
https://doi.org/10.1021/acs.chemmater.3c03191
https://doi.org/10.1021/jacs.2c12846
https://doi.org/10.1021/acs.jpcb.1c00188
https://doi.org/10.1016/j.jcis.2021.04.115
https://doi.org/10.1364/OE.390938


 

4 

 

 

ACKNOWLEDGMENTS 

 

 

I am extremely grateful to receive the support of so many passionate and dedicated individuals 

over the course of my chemistry career. From starting as a music major to finishing a Ph.D. it’s 

been quite a journey, and I couldn’t have done it without help. 

 First and foremost, thank you to Geri Richmond for inspiring me to study at the University 

of Oregon and providing me with a home in the Richmond Lab. I still remember when I introduced 

myself to Geri at my first ACS conference in New Orleans as a scared little undergraduate unaware 

of my future. After an REU in the Richmond lab (and meeting Geri’s chickens), I was sold on 

Eugene and pursuing a graduate degree. I truly wouldn’t be here if it wasn’t for her. Geri’s efforts 

to promote equality in science and her ability to craft engaging presentations are two aspects that 

I hope to carry on in my own career. 

 Second, thank you to my advisors Larry Scatena and Carl Brozek who took me under their 

wing and directed me through the difficult, winding path of graduate school. Larry, thank you for 

your tough conversations that challenged me to think deeper about my results and push myself as 

a researcher. I had a great time writing these manuscripts and flushing out the ideas with you (even 

despite the tough peer review process!). I’m honored to have learned from one of the best laser 

brains around. Carl, to think that we’ve only worked together for three years! We’ve cranked out 

a lot of impressive work (and more on the way!). Thank you for making the scientific process fun 

in a time when I was feeling discouraged and drained. Brainstorming in your office, drawing out 

figures, making two-week plans, analyzing data...to the regular person, these might seem like 

boring tasks, but our discussions helped me realize what I loved most about science: the 

investigative process and collaborating with others. Diamonds are made under pressure, right? Just 

like the perspective we wrote at the last minute! We make a great scientific team and I hope we 

get to work together again. 

 Thank you to my committee members Jeff Cina, Cathy Wong, and Daniel Steck who asked 

difficult questions and pushed me to be the best researcher I could be. I also appreciated the advice 

and teachings of the UO faculty, particularly Marina Guenza, Jim Prell, Andy Marcus, Scott 

Hansen, Julia Widom, Mark Lonergan, Chris Hendon, Karl Reasoner, Mike Haley (and Rocky), 

and Amanda Cook (and Pyewacket). Thank you to the chemistry office, especially Leah O’Brien 

and Christi Mabinuori whose helpful smiles always brightened my day. Thank you to Priscilla 



 

5 

 

 

Lewis (and Shiloh) who made the logistics of graduate school so much easier and helped me be 

involved in the PURS and REU program. I am also grateful to my professors at the University of 

Puget Sound who inspired me to pursue graduate school and enriched my education in more ways 

than they will ever know. Thank you especially to Dan Burgard, Amanda Mifflin, Steven Neshyba, 

Eric Scharrer, Jo Crane, Luc Boisvert, Gerard Morris, and Jennifer Nelson. Hack hack chop chop! 

 To my lab mates past, present, and future: Andrew Carpenter, Rebecca Altman, Brandon 

Schabes, Emma Tran, Marc Foster, Konnor Jones, Checkers Marshall, Kevin Fabrizio, Jiawei 

Huang, Erik Svensson Grape, Michael LeRoy, Jacob McKenzie, Audrey Davenport, Faiqa Khaliq, 

Quinn Valentine, Golnaz Navidi, Racheal Fisher, Rachel Galfo, Dario Nunez, and Adam Mather, 

thank you for being there every day for a good laugh, a tough conversation, or just a coffee walk. 

For the Richmond Lab members, thank you for providing me with the foundations for graduate 

school and inspiring me to be just like you! To the Brozek group, thank you for the fun, wild times, 

and for welcoming a physical chemistry person like me into your materials world. 

 Finally, the biggest thank you to the four most important beings in my life. A huge thank 

you and all the love to my partner, Cassidi Howard. Thank you for being there at the exciting and 

stressful times, the times when I was grumpy and grumbling with Adobe Illustrator, and for all the 

times that I’ve practiced my talks for you. Thank you to my parents, Kim Mapile and Todd Mapile, 

without a doubt, I wouldn’t be finishing this crazy thing if it wasn’t for your unwavering support. 

And to Pepper, thank you for your kisses and cuddles! I love you all so much. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

6 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

For Nibbler. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

7 

 

 

TABLE OF CONTENTS 

Chapter      Page 

 

 

CHAPTER I: INTRODUCTION .................................................................................................. 15 

I.A. SOLVATION AT NON-POROUS INTERFACES .......................................................... 16 

I.A.1. Colloidal Stability and Solvation ................................................................................ 16 

I.A.2. Capped Colloidal Nanoparticles .................................................................................. 18 

I.A.3. Bare Nanoparticle Solvation ....................................................................................... 19 

I.B. SOLVATION AT PSEUDO-POROUS INTERFACES .................................................... 20 

I.B.1. Polymer Solubility ....................................................................................................... 20 

I.B.2. Hydration of Proteins................................................................................................... 21 

I.C. SOLVATION OF OIL-IN-WATER EMULSIONS .......................................................... 23 

I.C.1. Emulsion Kinetic Stability........................................................................................... 23 

I.C.2. Steric Stabilization of Emulsions ................................................................................ 25 

I.D. SOLVATION AT POROUS INTERFACES .................................................................... 26 

I.D.1. Simple Porous Structures ............................................................................................ 27 

I.D.2. Porous Frameworks ..................................................................................................... 30 

CHAPTER II: VIBRATIONAL SUM FREQUENCY SCATTERING SPECTROSCOPY 

(VSFSS) ........................................................................................................................................ 33 

II.A.VSFSS THEORY .............................................................................................................. 33 

II.B. LASER SETUP ................................................................................................................. 35 

CHAPTER III: EMULSIONS STABILIZED BY POLY(ACRYLIC ACID) ............................. 38 

III.A. AN INTRODUCTION TO POLY(ACRYLIC ACID) ................................................... 38 

III.B. MATERIALS AND METHODS .................................................................................... 40 

III.B.1. Materials ................................................................................................................... 40 



 

8 

 

 

III.B.2. Emulsion Formation.................................................................................................. 40 

III.B.3. Dynamic Light Scattering and Zeta Potential ........................................................... 41 

III.C. RESULTS AND DISCUSSION ...................................................................................... 41 

III.C.1. Impact of PAA Concentration .................................................................................. 41 

III.C.2. Emulsions Stabilized by PAA Alone ........................................................................ 42 

III.C.3. Addition of Salt to PAA-stabilized Emulsions ......................................................... 48 

III.D. CONCLUSIONS ............................................................................................................. 53 

CHAPTER IV: IMPACT OF POLYMER STERICS ON EMULSION STABILITY ................ 54 

IV.A. AN INTRODUCTION TO POLYMER STERIC HINDRANCE .................................. 54 

IV.B. MATERIALS AND METHODS .................................................................................... 57 

IV.B.1. Materials ................................................................................................................... 57 

IV.B.2. Emulsion Formation ................................................................................................. 57 

IV.B.3. Dynamic Light Scattering and Zeta Potential ........................................................... 57 

IV.C. RESULTS AND DISCUSSION - EMULSIONS COATED WITH PAA OF VARYING 

MOLECULAR WEIGHT ......................................................................................................... 58 

IV.C.1. Emulsion Characteristics .......................................................................................... 58 

IV.C.2. Entropy and Enthalpy Favored Adsorption .............................................................. 59 

IV.C.3. VSFSS of Varying Molecular Weight PAA Emulsions at pH 4 .............................. 60 

IV.C.4. VSFSS of Varying Molecular Weight PAA Emulsions at pH 2 .............................. 64 

IV.C.5. C=O Region and Summary ....................................................................................... 66 

IV.D. RESULTS AND DISCUSSION - EMULSIONS COATED WITH PMAA .................. 67 

IV.D.1. Emulsion Characteristics .......................................................................................... 67 

IV.D.2. VSFSS of PMAA Coated Emulsions ....................................................................... 67 

IV.E. RESULTS AND DISCUSSION - EMULSION THERMAL STABILITY .................... 70 

IV.F. CONCLUSIONS ............................................................................................................. 71 



 

9 

 

 

CHAPTER V: COLLOIDAL STABILITY OF ZIF-8 ................................................................. 73 

V.A. AN INTRODUCTION TO BARE MOF NANOPARTICLES ....................................... 73 

V.B. MATERIALS AND METHODS ..................................................................................... 76 

V.B.1. Materials..................................................................................................................... 76 

V.B.2. Synthesis of nanoZIF-8 Particles ............................................................................... 76 

V.B.3. nanoZIF-8 Sample Preparation for VSFSS ................................................................ 76 

V.B.4. Powder X-Ray Diffraction ......................................................................................... 77 

V.B.5. FT-IR Spectroscopy ................................................................................................... 77 

V.B.6. Dynamic Light Scattering and Zeta Potential Measurements .................................... 77 

V.C. RESULTS AND DISCUSSION ....................................................................................... 78 

V.C.1. Ordered Solvation Shells Revealed by VSFSS .......................................................... 78 

V.C.2. Spontaneous Ordering within the ZIF-8 Nanoparticles ............................................. 82 

V.D. CONCLUSIONS .............................................................................................................. 85 

CHAPTER VI: CHARACTERIZING THE MOF-POLYMER INTERFACE FOR IMPROVED 

MEMBRANE FORMULATION ................................................................................................. 86 

VI.A. AN INTRODUCTION TO MIXED MATRIX MEMBRANES .................................... 86 

VI.B. MATERIALS AND METHODS .................................................................................... 89 

VI.B.1. Materials ................................................................................................................... 89 

VI.B.2. Synthesis of nanoZIF-8 ............................................................................................. 89 

VI.B.3. Formation of MMMs ................................................................................................ 89 

VI.B.4. Dynamic Light Scattering and Zeta Potential ........................................................... 90 

VI.B.5. FT-IR Spectroscopy .................................................................................................. 90 

VI.B.6. Scanning Electron Microscopy ................................................................................. 90 

VI.C. RESULTS AND DISCUSSION ..................................................................................... 90 

VI.C.1. Water Stability of ZIF-8 with Binding Agent ........................................................... 90 



 

10 

 

 

VI.C.2. Characterization of Colloids ..................................................................................... 92 

VI.C.3. Ordered Polymer Layers Measured by VSFSS ........................................................ 94 

Poly(acrylic acid) .................................................................................................................. 94 

Poly(acrylamide) ................................................................................................................... 95 

Poly(N-vinyl acetamide) ........................................................................................................ 97 

VI.C.4. Characterization of MMMs ...................................................................................... 98 

VI.C.5. Tensile Strength Testing ........................................................................................... 99 

VI.C.6. Force-Probe AFM ..................................................................................................... 99 

VI.D. CONCLUSIONS ........................................................................................................... 100 

CHAPTER VII: ORDERED SOLVATION SHELLS AND THE STABILITY OF POROUS 

AND NON-POROUS COLLOIDS ............................................................................................ 101 

CHAPTER VIII: THE FUTURE OF VSFSS FOR PROBING UNIQUE MECHANISMS OF 

COLLOIDAL STABILITY ........................................................................................................ 104 

APPENDIX A: SUPPLEMENTARY INFORMATION FOR CHAPTER III .......................... 108 

APPENDIX B: SUPPLEMENTARY INFORMATION FOR CHAPTER IV........................... 123 

APPENDIX C: SUPPLEMENTARY INFORMATION FOR CHAPTER V ............................ 131 

APPENDIX D: SUPPLEMENTARY INFORMATION FOR CHAPTER VI .......................... 139 

APPENDIX E: SUPPLEMENTARY INFORMATION FOR CHAPTER VII ......................... 143 

APPENDIX F: CALCULATION OF NANOPARTICLE SURFACE DENSITY .................... 144 

APPENDIX G: VSFSS NORMALIZATION PROCEDURE .................................................... 147 

APPENDIX H: DETERMINATION OF PARTICLE CONCENTRATION FOR VSFSS 

DETECTION LIMIT .................................................................................................................. 151 

REFERENCES ........................................................................................................................... 153 

 

 

  



 

11 

 

 

LIST OF FIGURES 

 

Figure      Page 

 

Figure I. 1 A) Typical DLVO graph ............................................................................................ 17 

Figure I. 2. Illustration of emulsion formation ............................................................................ 24 

Figure I. 3 Illustration of A) 1-dimensional (polymers, nanotubes). ........................................... 28 

Figure I. 4 Representation of A) framework materials. ............................................................... 31 

Figure II. 1 A) Schematic of VSFSS............................................................................................33 

Figure II. 2 Schematic of the VSFSS experiment ....................................................................... 36 

Figure III. 1 Chemical structure of poly(acrylic acid)..................................................................38 

Figure III. 2 Characteristics A) hydrodynamic diameter ............................................................ 41 

Figure III. 3 VSFSS measurements. ............................................................................................ 44 

Figure III. 4 Characteristics (top panel: hydrodynamic diameter ............................................... 48 

Figure III. 5 VSFSS measurements (ssp polarization) ................................................................ 50 

Figure III. 6 Illustration depicting the behavior of PAA ............................................................. 52 

Figure IV. 1 A) Hydrodynamic diameter......................................................................................58 

Figure IV. 2 Critical adsorption parameter (δc)........................................................................... 60 

Figure IV. 3 VSFSS measurements (ssp polarization) ................................................................ 61 

Figure IV. 4 Molecular picture of the hexadecane/water droplet interface ................................. 63 

Figure IV. 5 VSFSS measurements (ssp polarization). ............................................................... 64 

Figure IV. 6 Molecular picture of the hexadecane/water droplet interface ................................. 65 

Figure IV. 7 VSFSS measurements (ssp polarization) ................................................................ 68 

Figure IV. 8 Temperature-dependent diameter. ........................................................................... 70 

Figure V. 1 A) VSFSS measurements of nanoZIF-8 particle.......................................................79 

Figure V. 2 A) VSFSS measurements (ssp polarization)............................................................. 81 

Figure V. 3 A) VSFSS measurements (ssp polarization)............................................................. 83 

Figure VI. 1 Representation of the ZIF-8 nanocrystal surface.....................................................87 

Figure VI. 2 PXRD patterns. ....................................................................................................... 91 

Figure VI. 3 A) Hydrodynamic diameter ..................................................................................... 92 

Figure VI. 4 VSFSS measurements (ssp polarization) ................................................................ 96 

Figure VI. 5 A) Image of MMMs ................................................................................................ 98 



 

12 

 

 

Figure VII. 1 VSFSS measurements (ssp polarization)..............................................................102 

Figure VII. 2 Dynamic light scattering distribution curves ....................................................... 103 

Figure VIII. 1 Temperature dependent hydrodynamic diameter................................................105 

Figure VIII. 2 a) Zeta potential of Cu(TA)2 nanoparticles ....................................................... 106 

Figure A. 1 Zeta potential of 500 ppm PAA-stabilized..............................................................108 

Figure A. 2 Visual representation of the time-dependent colloidal stability ............................. 109 

Figure A. 3 Equilibrium surface pressure. ................................................................................. 110 

Figure A. 4 Visual representation of the time-dependent colloidal stability. ............................ 111 

Figure A. 5 Hydrodynamic diameter distribution curves of colloids ........................................ 111 

Figure A. 6 Dynamic light scattering and Zeta Potential measurements ................................... 112 

Figure A. 7 VSFSS measurements (ssp polarization) ................................................................ 116 

Figure A. 8 VSFSS measurements (ssp polarization) ................................................................ 116 

Figure B. 1 VSFSS measurements (ssp polarization).................................................................123 

Figure B. 2 Fit parameter trends for VSFSS measurements ...................................................... 124 

Figure B. 3 Equilibrium surface pressure. ................................................................................. 125 

Figure B. 4 Polydispersity index (PDI) of nanoemulsions ........................................................ 126 

Figure C. 1 SEM images of ZIF-8 nanoparticles........................................................................131 

Figure C. 2 PXRD patterns of powder nanoZIF-8..................................................................... 131 

Figure C. 3 FT-IR spectra. ......................................................................................................... 131 

Figure C. 4 Impact of DMF absorption on VSFSS spectra. ...................................................... 132 

Figure C. 5 Size distribution curves of nanoZIF-8 particles...................................................... 133 

Figure C. 6 Acid digestion 1H NMR ......................................................................................... 133 

Figure C. 7 TGA trace of nanoZIF-8. ........................................................................................ 134 

Figure C. 8 Surface zeta potential of nanoZIF-8 ....................................................................... 134 

Figure C. 9 VSFSS measurements (ssp polarization) ................................................................ 135 

Figure D. 1 pH measurements of ZIF-8......................................................................................139 

Figure D. 2 Pictures of ZIF-8 colloids ....................................................................................... 139 

Figure D. 3 A) Amplitude, B) Gaussian linewidth, and C) peak position trends ...................... 140 

Figure D. 4 FT-IR spectra of polymeric membrane films ......................................................... 140 

Figure E. 1 Dynamic light scattering curves...............................................................................143 

Figure G. 1 Non-resonant response from KNbO3 crystal...........................................................147 



 

13 

 

 

Figure G. 2 Raw signal obtained from one scan. ....................................................................... 147 

Figure G. 3 Average signal trace ............................................................................................... 148 

Figure G. 4 Averaged signal trace. ............................................................................................ 148 

Figure G. 5 Averaged signal trace. ............................................................................................ 149 

Figure G. 6 Signal trace from colloidal ZIF-8 in DMF normalized .......................................... 149 

Figure G. 7 Completely normalized signal trace. ...................................................................... 150 

Figure H. 1 Signal-to-noise ratio................................................................................................152 



 

14 

 

 

LIST OF TABLES 

 

Table      Page 

 

Table IV. 1 Integrated area of VSFSS spectra ............................................................................. 69 

Table A. 1 Emulsion zeta potential.............................................................................................109 

Table A. 2 Polydispersity index (PDI) values ............................................................................ 113 

Table A. 3 Fitting parameters for CH spectra ............................................................................ 117 

Table A. 4 Fitting parameters for C=O spectra .......................................................................... 118 

Table A. 5 Fitting parameters for spectra ................................................................................... 119 

Table A. 6 Fitting parameters for spectra ................................................................................... 120 

Table A. 7 Fitting parameters for spectra ................................................................................... 121 

Table A. 8 Fitting parameters for spectra. .................................................................................. 122 

Table B. 1 Characteristics (diameter, polydispersity index, and zeta potential).........................125 

Table B. 2 Fitting parameters for C-H spectra ........................................................................... 127 

Table B. 3 Fitting parameters for C=O spectra. ......................................................................... 128 

Table B. 4. Fitting parameters for C-H spectra .......................................................................... 130 

Table B. 5 Fitting parameters for C=O spectra .......................................................................... 130 

Table C. 1 Fitting parameters for spectra of ZIF-8.....................................................................136 

Table C. 2 Fitting parameters for spectra of ZIF-8 .................................................................... 137 

Table C. 3 Fitting parameters for spectra of ZIF-8. ................................................................... 138 

Table D. 1 Fitting parameters for spectra of ZIF-8.....................................................................141 

Table D. 2 Fitting parameters for spectra of ZIF-8 .................................................................... 141 

Table D. 3 Fitting parameters for spectra of ZIF-8 nanoparticles. ............................................. 142 

  



 

15 

 

 

CHAPTER I: INTRODUCTION 

This introduction provides a background on the traditional theories used to describe solvation and 

stability of common colloids (nanoparticles, polymers, and proteins) whose mechanistic details are 

important for application to our nano-sized emulsions and MOFs. Next, this introduction specifies 

the unique characteristics of the systems studied here – namely the kinetic stability of emulsions 

and the porosity of nanoMOFs – and presents avenues for measuring colloidal behavior.  

The introduction contains material published in Chemistry of Materials as a Perspective 

written with co-authors Erik Svensson Grape and Carl K. Brozek in 2024. Chapter II provides a 

technical background on vibrational sum frequency scattering spectroscopy (VSFSS) and details 

the experimental setup used in this work. Chapters III and IV describe two research projects that 

study the role of steric bulk from a polymer layers as a mechanism for stability of hexadecane 

emulsions. The role of pH- and salt-sensitivity is introduced in Chapter III and molecular weight 

dependence is probed in Chapter IV. Chapter III was published in Colloids and Surfaces A: 

Physiochemical and Engineering Aspects in 2024 and Chapter IV is currently submitted to Soft 

Matter, with both manuscripts supervised and written with Lawrence F. Scatena. Chapter V was 

published in ACS Nano in 2024 in collaboration with Michael A. Leroy, Kevin Fabrizio, Lawrence 

F. Scatena, and Carl K. Brozek and introduces VSFSS as an emerging technique to measure the 

surface of bare, colloidally stable MOF nanoparticles. Chapter VI contains unpublished work in 

preparation for submission co-authored with Michael A. Leroy, Audrey M. Davenport, Lawrence 

F. Scatena, and Carl K. Brozek that combines the two main aspects of this dissertation, nanoMOF 

particles and polymer coatings, where the surface structure of carbonyl polymers at very low 

concentrations is measured and proposed as a solution to homogenously disperse nanoMOF 

particles in a polymeric membrane. Finally, Chapter VII details unpublished work in collaboration 

with Adam P. Mather and Carl K. Brozek that connect the fundamentals of colloidal stability 

discovered in this work, namely solvation shells, to the relevant parameter of critical aggregation 

concentration of both porous and non-porous inorganic colloids. Chapter VIII will conclude this 

dissertation with remarks on the exciting and broad future of VSFSS for studying buried, colloidal 

surfaces. 

 

 

 



 

16 

 

 

I.A. SOLVATION AT NON-POROUS INTERFACES 

 

I.A.1. Colloidal Stability and Solvation 

In the 1940s, two teams of researchers independently studied the forces contributing to particle 

colloidal stability. Boris Derjaguin and Lev Landau in the Soviet Union presented their theory of 

colloidal stability that invoked short-range van der Waals attractions between particles overcome 

by electrostatic surface repulsions.1 While the electrical double layer had previously been 

introduced by Hermann von Helmholtz (1853),2 Louis Georges Gouy (1910),3 David Leonard 

Chapman (1913),4 and Otto Stern (1924),5 Derjaguin and Landau pioneered the notion of 

electrostatic potentials at curved interfaces where the surface electric field decays as a function of 

the particle radius. Meanwhile, in the Netherlands, Evert Verwey and Theodoor (Jan) Overbeek 

developed the now-ubiquitous potential energy curves of two interacting spherical particles as a 

function of interparticle distance and electrolyte concentration.6,7 Ultimately, a model of the forces 

governing nanoparticle stability in solution was named DLVO theory for the four authors 

involved.8 DLVO theory has been thoroughly derived to describe the colloidal stability of hard-

shell nanoparticles in a variety of electrolyte concentrations,9,10 surface charge composition,11,12 

solvent environments,13 and many other scenarios.13–18 DLVO theory can be summarized by Eq. 

1: 

𝑊(𝐷) = 𝑊𝑣𝑑𝑤 + 𝑊𝑒𝑙𝑒𝑐 (1) 

where W(D) represents the interaction energy between neighboring particles, Wvdw is the attractive 

energy due to van der Waals interactions and Welec is the repulsive electrostatic energy (Figure 

I.1A). The entropy and steric pressures of the system have been incorporated into extended DLVO 

theory (XDLVO)19,20  as described by Eq. 2: 

𝑊(𝐷) =  𝑊𝑣𝑑𝑤 + 𝑊𝑒𝑙𝑒𝑐 + 𝑊𝑜𝑠𝑚𝑜𝑡𝑖𝑐 + 𝑊𝑒𝑛𝑡𝑟𝑜𝑝𝑖𝑐 (2) 

where the addition of the Wosmotic term accounts for the repulsive energy caused by neighboring 

nanoparticles with an increased overlap region (typically due to increased surface sterics), while 

the Wentropic term accounts for the attractive energy arising from an increase in solvent entropy 

upon particle coalescence. This entropy-driven aggregation is often termed the hydrophobic effect. 

Although developed as a general model, DLVO theory applies best to “hard” colloidal 

systems, such as colloidal quantum dots, core-shell metallic nanoparticles, or metal oxides with  



 

17 

 

 

well-defined boundaries between the nanoscale object and the suspending solution. Whereas hard-

shelled colloids maintain their rigidity in solution, “soft-shelled” particles stretch and adapt to 

nearby fluid interfaces (Figure I.1B).21–24 Hard-shell particles are convenient systems to study as 

short-range van der Waals interactions decay exponentially as a function of distance from the 

particle without the need to consider interfacial charge-screening. On the other hand, soft-shell 

particles, such as oil droplets, metal nanoparticles coated with a polymer layer, proteins clusters, 

and even living cells,25 present complications to DLVO theory as solvent intercalation, ion 

adsorption,26 spontaneously ordered solvation shells , and surface roughness,15 together 

comprising so-called “non-DLVO forces”. Experimental attempts to bridge these two types of 

systems, such as core-shell colloids with soft exteriors and hard interiors,21,27 impart improved 

thermodynamic stability and, in some cases, unique optical behavior.28,29 Theoretical models of 

 
Figure I. 1 A) Typical DLVO graph showing the interaction energy as a sum of electrostatic 

repulsion and attractive van der Waals interactions. B) Illustration of the soft-to-hard range of 

nanoscale materials ranging from polymers/proteins to hard spheres, such as metal oxide 

nanoparticles. C) The lattice model of polymer solvation as described by Flory-Huggins 

theory. D) Illustration and ion examples in relation to the Hofmeister series of solvation. 

 

Figure I.1 A) Typical DLVO graph showing the interaction energy as a sum of electrostatic 

repulsion and attractive van der Waals interactions. B) Illustration of the soft-to-hard range of 

nanoscale materials ranging from polymers/proteins to hard spheres, such as metal oxide 

nanoparticles. C) The lattice model of polymer solvation as described by Flory-Huggins 

theory. D) Illustration and ion examples in relation to the Hofmeister series of solvation. 



 

18 

 

 

such systems account only for the hard-soft interface, while neglecting the possibility of low 

density and heterogeneous surfaces and their associated energetics.   

According to the IUPAC definition, a colloid involves a “molecule or polymolecular 

particle dispersed in a medium” with “at least in one direction a dimension roughly between 1 nm 

and 1 µm”.30 Examples of colloids include a solid in a gas (smoke), a liquid in a gas (aerosol), or 

a liquid in a liquid (emulsion), in addition to many other dispersion types that do not require solvent 

as a medium. Understanding the stability of heterogeneous mixtures in solution is critical to their 

implementation in displays, coatings, or membranes, but the clear phase boundary creates 

significant challenges because of the difficulty in studying the chemistry that emerges at the 

colloidal interface. The process of dissolution, by contrast, involves the formation of a single, 

homogeneous phase.31 Unlike solutions, the heterogeneity of colloidal dispersions creates 

thermodynamic instability resulting in their eventual phase separation. Suspended particles can be 

filtered out or mechanically separated, whereas separating dissolved particles requires additional 

chemical transformations.32,33 In the following sections, we provide a brief overview  of the 

chemistry underlying the colloidal stability of non-porous inorganic nanoparticles with and 

without functionalized surfaces, and how it critically involves solvent. 

 

I.A.2. Capped Colloidal Nanoparticles 

Nearly all non-porous inorganic nanoparticles require surface functionalization for long-term 

stability. The strong van der Waals attractions between bare metallic surfaces results in nearly 

instantaneous aggregation and sedimentation. Semiconductor nanocrystals, specifically quantum 

dots, the topic of the Nobel Prize in Chemistry in 2023,34,35 represent another key example of 

nanoparticles that find utility only by surface functionalization. Attaching surfactant ligands to 

prevent Ostwald ripening36,37 allows semiconductor nanoparticles to remain stable at precise sizes 

that dictate their tunable optical properties.38–40 Functionalizing quantum dots with either an 

inorganic shell also diversifies the available solvents for dispersion.41,42 Most notably, the addition 

of water-soluble ligands, such as citrates,43 polymers,44,45 or other small molecules46,47 allows for 

the use of metallic nanoparticles within cellular media. A wide variety of ligands have been studied 

for their interaction with various solvent media.48,49 By contrast, nearly all porous colloidal 

materials, as described below, require no surface functionalization for colloidal stability. 



 

19 

 

 

Therefore, for context, we must understand the mechanism of colloidal stability for nanoparticles 

prepared without capping agents. 

 

I.A.3. Bare Nanoparticle Solvation 

Few reports have documented bare nanoparticles exhibiting long-term colloidal stability. The 

examples described below are restricted in their industrial applications due to highly defective 

surfaces, the necessity of a stabilizing supporting material, and the need for post-synthetic cleaning 

or filtration. For example, while it has been shown that metallic nanoparticles can be stabilized by 

a partially oxidized surface,50 this chemical decomposition generates a heterogeneous and rough 

surface with mixed valence sites.51,52 An alternate approach that avoids surface oxidation was 

achieved by synthesizing bare copper nanoparticles with an electron-donating gadolinium 

support.53 Implementing nanoparticles in biological sensing applications cannot be achieved, 

however, with a bulky and potentially toxic supporting material. Another approach involves 

“sterilization” of bare gold nanoparticles by autoclave.54 Studies suggest that < 5-nm particles 

assemble into larger 10-30 nm particles during this process. This size-focusing serves as a form of 

filtration to achieve a narrow size dispersity, but size selection does not always ensure colloidal 

stability.55 Because these few examples of bare nanoparticles are short-lived colloids, they are 

often quickly drop-casted back to their solid state.56–59  

In the few studies of bare metallic nanoparticles in solution without a stabilizing support, 

the ionic structure of the supporting electrolyte and the resulting electrical double layer at the 

surface has proven to be a key, but poorly understood feature contributing to colloidal stability. 

Simulations suggest that facet-specific ion adsorption and an ordered primary solvation shell 

promote gold nanoparticle stability in water.52,60 Experimental studies also suggest water stability 

of bare nanoparticles arises from ion adsorption dependent on electrolyte concentration and 

following the so-called Hofmeister series for anions.54,61–63 The colloidal stability of nanoparticles 

in solvents without electrolyte, however, have only been studied as computational simulations of 

bare particles64 or through weakly ligated experimental systems where a portion of the surface 

remains deliberately exposed.65 Thus, achieving a metal nanoparticle colloid stabilized by 

solvation forces alone remains an open challenge and would open an avenue for tuning the opto-

electronic and electrochemical electrical properties of metallic nanoparticles in a wide variety of 



 

20 

 

 

solutions. To understand the colloidal stability of bare nanomaterials, therefore, we instead turn to 

the supramolecular solvation chemistry of polymers and proteins. 

 

I.B. SOLVATION AT PSEUDO-POROUS INTERFACES 

 

I.B.1. Polymer Solubility 

At nearly the same time that Derjaguin, Landau, Verwey, and Overbeek developed their model of 

colloidal stability, Paul Flory and Maurice Huggins independently considered how polymers could 

dissolve despite distinct differences in molecular size from the surrounding solvent.66,67 While the 

entropy of mixing a simple molecule in solution can be described by Gibbs energy of mixing: 

∆𝐺𝑚𝑖𝑥 = ∆𝐻𝑚𝑖𝑥 − 𝑇∆𝑆𝑚𝑖𝑥 (3) 

where H is enthalpy, T is temperature, and S is entropy, this relationship only considers the gross 

interaction between a molecule and its surroundings. It neglects synergistic behavior between 

individual units within a chain, such as in a polymer (Figure I.1C). Flory and Huggins adapted 

the Gibbs energy of mixing for polymers as: 

∆𝐺𝑚𝑖𝑥 = 𝑅𝑇(𝑛1𝑙𝑛𝜙1 + 𝑛2𝑙𝑛𝜙2 + 𝑛1𝜙2𝜒12) (4) 

which now considers the number of moles (𝑛) and the volume fraction (𝜙) of the solvent 

(component 1) and polymer (component 2). 𝜒12 is material-specific and describes the synergistic 

interaction between the polymer and the solvent, allowing for specific descriptions of polymer 

solubility. 

We introduce the term “pseudo-porosity” to describe the solvation of polymers and related 

systems. While polymers lack permanent porosity (the ability to maintain a rigid, porous structure 

when suspended in solution), their solvation mechanism depends on the large solvent accessible 

surface areas carved out by the wrapping of individual monomer units. In Flory-Huggins theory, a 

polymer solution is modeled as a lattice of cells containing either polymer monomer units or a 

solvent molecule. Based on this image, the individual mixing components of a monomer unit with 

the polymer is well described by equation 4. However, realistic intermolecular forces from 

hydrophobicity,68 ionic screening,69 and other polymer or solvent-specific interactions70,71 integral 

to solvation and polymer conformation are absent from eq. 4. 



 

21 

 

 

In addition to the pseudo-porosity of conventional polymers, intrinsically porous polymers 

and porous polymer membranes also exist. Typical examples of porous polymer membranes 

include polycarbonate, polyester, or cellulose with pore diameters ranging from nanometers to 10s 

of µms. The narrow, one-dimensional confinement of these pores within a membrane have led to 

a versatile platform for studying solvent-controlled ion-transport under spatial confinement akin 

to the ion-transport across cellular membranes and carbon nanotubes.72,73 Chemical 

functionalization of these materials leads to specific ion effects, such as appended carbonate groups 

showing preferential binding of metal cations.74 Polymers of intrinsic microporosity (PIMs), on 

the other hand, contain voids of 2 nm or less.75 Controlling the solubility of PIMs is critical for 

creating robust films with consistent pore sizes.76 The commonly used PIM-1, however, dissolves 

only in tetrahydrofuran (THF) and chloroform (CHCl3), leading to reduced solution processability 

and restricting the incorporation of fillers or other composites. Recent work has introduced post-

synthetic modifications of PIMs to improve their solubility,77 with the key result that specific 

solvent interactions with appended moieties improves solubility. In understanding the solvation of 

porous materials discussed in Section 3, these reports of porous polymers offer the insight that 

solution stability of nanoscale materials benefits from monomer units accessible to solvent. 

 

I.B.2. Hydration of Proteins 

Proteins comprise another class of macromolecules whose solvation depends on solvent interacting 

with accessible building blocks, yet with the added compositional diversity of amino acid mixtures 

and H-bonding networks. For proteins, solvation can be described as monomer-by-monomer 

solubility or by secondary structure solvation via protein folding to bury hydrophobic moieties. As 

a result, proteins arrange into secondary structures, like helices or sheets, and fold into native states 

to satisfy solvent interactions with their respective amino acid chains. Numerous studies have 

quantified surface interactions between proteins and water in solution,78,79 with one notable 

computational investigation into the hydrogen bonding of a protein in water as a function of solvent 

accessible surface area and protein conformation.80 These authors report that the solvation free 

energy of a protein decreases linearly as a function of the solvent accessible surface area of the 

protein. This finding suggests that an increase in the surfaces available for solvation promote 

stability in solution. As the basis of Flory-Huggins theory, leveraging enthalpically favorable 



 

22 

 

 

solvent-material interactions is critical for solubility and stability. In a permanently porous 

material, as explained below, the unparalleled solvent accessible surface areas provide increased 

sites for solvation interactions via internal and external surfaces. 

The unique folding ability of a protein allows it to adapt to an otherwise thermodynamically 

unstable environment. In contrast, porous framework materials, such as metal-organic frameworks 

(MOFs), cannot protect select components by geometrical reconfiguration. Protein folding or 

unfolding can be induced by addition of ions that “salt-in” or “salt-out” the macromolecule of 

interest (protein, polymer, nanoparticle, etc.) vis-à-vis the Hofmeister series (Figure I.1D). Ion 

specific effects in protein precipitation was introduced by Franz Hofmeister in the late 1880s to 

describe how different salts impact the solubility of proteins despite possessing the same net 

charges.81,82 Kosmotropes, or “structure-makers” are ions that interact more strongly with water 

than with the protein themselves, causing the protein to remain in its native folded state and salt-

out of solution. Common examples of kosmotropes include citrate, sulfate, and phosphate anions 

or magnesium, calcium, or lithium cations. On the other hand, chaotropes, or “structure-makers” 

interact closely with individual protein units, promoting an unfolded, solubilized state of the 

protein. Chaotropes include iodide, nitrate, and tetrafluoroborate anions or calcium, magnesium, 

and aluminum cations. Polyoxometalates (POMs), for example, are super-chaotropes due to their 

large, delocalized charge and low surface densities.83 While Hofmeister’s initial experiments 

provide a basis for harnessing specific-ion effects, recent work has produced a more detailed 

understanding of salts and their impact on colloidal stability and solubility, especially cooperative 

contributions from ion pairs.84–87 Vibrational spectroscopy and molecular dynamic simulations 

reveal that cations follow the Hofmeister series through strong backbone-salt interactions and 

weaker interactions with negatively charged side chains.88 Anions, however, despite following the 

Hofmeister series at the backbone, exhibit a reversed trend on positively charged amino acid 

residues. For this reason, the Hofmeister series for anions holds only when the backbone-salt 

interactions outweigh the salt interactions with side chains. Tuning the colloidal stability of metal-

based nanoparticles requires understanding the specific-ion effect in solvents beyond water.89,90 A 

comprehensive study of the ion effects in non-aqueous solvents found that water is not unique in 

its role in the Hofmeister series.91 In fact, aprotic solvents also show an ion specificity due to the 

inherent molar volume and electrostriction of the ion, or the ability of the ion to slightly deform. 



 

23 

 

 

These results confirm that a Hofmeister trend persists regardless of solvent identity, thereby 

providing a basis for tailoring the colloidal stability of non-aqueous materials.  

Specific ion effects are critical to the colloidal stability of bare nanoparticles. In one study, 

gold nanoparticles without any organic ligands at the surface were stabilized by as little as 10 µM 

of a chaotropic anion.61 The electrolyte solutions were added during nanoparticle synthesis, 

thereby acting as a non-organic capping ligand. Additional electrolyte may destabilize colloids, 

however. Reports indicate that increasing salt concentration to the mM regime often leads to 

nanoparticle aggregation, as exhibited by metallic nanosheets titrated with potassium salts of 

varying valency anions92 and other metal nanoparticles.93,94 A likely causes is that with increased 

salt concentrations electrolyte ions screen the effective charge density that normally prevents 

aggregation. Nevertheless, tuning salt concentrations could serve as a promising strategy for 

controlling the solvation of porous nanoparticles with bare surfaces. 

 

I.C. SOLVATION OF OIL-IN-WATER EMULSIONS 

 

I.C.1. Emulsion Kinetic Stability 

Formulating stable emulsions and understanding the mechanism of oil droplet stability has far 

reaching implications, from the development and enhancement of new cosmetic products, to the 

efficacy of drug delivery systems and oil spill remediation technologies.95–97 Traditionally, oil 

droplets with diameters of 20-200 nm are classified as nanoemulsions while larger droplets with 

diameters of 200-1000 nm in diameter can be referred to as miniemulsions.98 Regardless of the 

size, surfactant-coated emulsions are useful laboratory analogues for modeling commercially-used 

dispersants that reduce the surface tension of a large oil patch and enable the formation of stable, 

small oil droplets that are continuously distributed by waves and eventually biodegraded by marine 

microbes.99 However, these oil droplets require high energy mixing for formation and are 

thermodynamically unstable leading to eventual phase separation given enough time.100–103 In 

practice, ultrasonication is necessary to input the energy required for the system to enter the 

kinetically trapped state (Figure I.2). The addition of surfactants, electrolyte, polymers, and/or 

proteins can provide a protective layer that prevents particle coalescence over time. DLVO theory 

explains that particle-particle repulsion in emulsions is commonly accounted for by balancing 



 

24 

 

 

attractive van der Waals interactions with repulsive electrostatic forces, such as those typically 

achieved by an ionic surfactant. A zeta potential, or surface charge, of about ± 30 mV is typically 

necessary to a stabilize a colloid. Extended DLVO theory further accounts for steric effects with 

the addition of a destabilizing osmotic pressure term and a stabilizing entropic term. In most 

applications, emulsion stabilization is ensured by surface functionalization that provides 

electrostatic or steric repulsion, or a combination of the two. 

Beyond oil remediation, the curved interface of emulsions have proven to be of interest to 

measure the molecular packing behavior as compared to planar oil/water interfaces. The 

Richmond/Scatena lab and others have postulated that the geometry of the interface and the 

thermodynamic instability of emulsions are responsible for the observed differences in molecular 

orientation and conformation at the two interfaces.104–106 In one regard, ultrasonication forces both 

the formation of oil into small droplets and the adsorption of surfactants or polymers to the 

interface. The addition of ultrasonicating energy can cause unexpected interfacial behavior as 

compared to what has been exhibited at a planar oil/water interface where small molecules 

 
Figure I. 2. Illustration of emulsion formation from a phase separated state to a dispersion via 

ultrasonication. The dispersed emulsion is kinetically trapped and will eventually return to the 

phase separate state. 

 

Figure I.2  



 

25 

 

 

thermodynamically adsorb to the interface at an equilibrium time scale. Not only do emulsions 

differ from their planar counterparts in terms of kinetic stability, but they exhibit a curved 

geometry, instead of flat. While the droplet geometry increases overall interfacial surface area for 

chemical adsorption, it also enhances steric and electrostatic crowding effects which can reduce 

interfacial coverage of surfactants, polymers, or ions.107 Keeping these aspects of thermodynamic 

and kinetic stability in mind for the planar and droplet oil/water interfaces, respectively, are 

important when interpreting VSFS spectra from each chemical environment. 

 

I.C.2. Steric Stabilization of Emulsions 

While non-toxic surfactants are useful for environmental applications, the introduction of 

proteins as colloid stabilizers have the potential to encapsulate and selectively deliver hydrophobic 

drugs.108–112 It is hypothesized that colloids are stabilized by protein coronas through steric 

hinderance or, in other words, a thick protein shell prevents droplets from coalescing. While recent 

work using 2D-IR spectroscopy and sum frequency scattering spectroscopy have investigated the 

newfound importance of protein-stabilized colloids, the molecular conformation and detailed 

mechanism of stability remain difficult to probe due to the complicated nature of long-chain 

proteins and their resulting coronas.113,114 

Polyelectrolytes are useful analogues for studying protein coronas as both macromolecules 

have varying functional groups along a repeating backbone, a nuanced folding or coiling structure 

with the ability to hydrogen bond or form secondary structures, and charge localization along the 

chain dictated by environmental pH. Yet, polyelectrolytes are often easier to study than proteins 

due to their known monomer units and simple tunability of molecular weight, percent protonation, 

and secondary structure. In addition to their pH-dependent behavior in bulk solution, acids and 

polyelectrolytes adsorbed at interfaces are known to have a surface pKa ~1 unit more alkaline than 

the bulk pKa due to interfacial solvent dielectrics reducing the solvation energy of the acid 

functional groups.115–120 This pH-dependence is critical for materials such as polyelectrolyte 

coarcervates, where stabilization across a variety of conditions can be a challenge.121,122 The 

adsorption behavior of polymers has generated both theoretical123 and experimental124,125 studies 

that seek to understand interfacial conformation. In particular, a “train, loop, tail” representation 

has been developed that describes how polymers lie flat at the surface (train), coil beyond the 

surface into the bulk (loop) and extend at the ends of the polymer chains (tail) to emphasize 



 

26 

 

 

favorable surface-polymer interactions or bulk-polymer interactions, dependent on the polymer 

identity and interface.126–128 Here, VSFSS is a powerful tool to probe polymer conformation at  

interfaces.  

Previous work from Marc Foster in the Richmond Laboratory investigated the role of 

carboxylic acid surfactants at the curved oil/water interface.106 These experiments were 

fundamental in developing a rigorous experimental procedure to measure the carbonyl and 

carboxylate modes via VSFSS. A non-resonant contribution from the CaF2 windows used in 

VSFSS contribute to a large background that requires de-timing of the visible pulse to deconvolute 

the resonant vibrational modes of interest.129 Additionally, Marc found that the carbonyl and 

carboxylate modes at the oil/water droplet interface differed in both  spectral line shape and peak 

position then the same modes measured at a planar oil/water interface or the planar air/water 

interface.104,130 These differences were attributed to the difference in solubility of the hydrophobic 

phases with the alkyl chains of the surfactant. The improved solubility of the surfactant in the CH-

rich oil phase as compared to air, for example, would increase the ordering of the head group at 

the droplet interface. At the same time, it was proposed that the curved geometry of the droplet 

resulted in charge repulsion across the droplet that caused a lower population of surfactants at the 

droplet surface. With this increased surfactant head group area, the carbonyl and carboxylate 

modes exhibited improved ordering at the oil/water droplet interface. 

In this dissertation, the ability of a low-charge polymer to stabilize emulsions without the 

assistance of a surfactant is explored. The colloidal stability of this polymer system was further 

probed by adjusting the pH, salt identity and concentration, molecular weight of the polymer, 

backbone substitution, and temperature. The adjustment of these variables provide insight into the 

physiological robustness of polymer-stabilized emulsions with tunable parameters that can 

improve drug delivery platforms. 

 

I.D. SOLVATION AT POROUS INTERFACES 

To understand the solubility and stability of porous colloids, we suggest using the aforementioned 

models of solvation, namely electrostatic or steric repulsion, the hydrophobic effect, monomer-by-

monomer solubility, and specific ions interactions. By possessing both internal and external 

surfaces, porous colloids exhibit surface area-to-volume ratios and accessible void spaces far 



 

27 

 

 

exceeding any class of non-porous or pseudo-porous material.131 For example, the surface density 

of the commonly studied zeolitic imidazolate framework ZIF-8 (Zn(2-methylimidazolate)2), has a 

surface density of 7 atoms/nm2 while non-porous ZnO has a surface density of 78 atoms/nm2 

(Appendix F). Additionally, as illustrated by the example above, the internal and external surfaces 

of porous materials resemble the heterogeneous surfaces of proteins. We expect the surface of 

porous materials to be susceptible to specific solvent-surface interactions as opposed to 

generalized hard-shell interactions between a homogeneous nanoparticle surface or surfactant 

ligands and the supporting solvent. In the following sections, we describe common permanently 

porous materials (i.e., materials that retain porosity in solution) and propose mechanisms for their 

colloidal stability and solubility. This research field is nascent, without consensus around theory 

that explains the colloidal stability of porous materials. We propose, given the examples below, 

that specific solvent-surface interactions, such as through ordered solvation shells and chemical 

interactions between the material building blocks and the solvent provide the colloidal stability of 

porous materials. 

 

I.D.1. Simple Porous Structures 

As a starting point, nanotubes serve as a basis for understanding the solvation of a one-dimensional 

porous material (Figure I.3A). The pores, or tunnels, of nanotubes can range from 0.5-2 nm in 

diameter and are typically grown from a metal catalyst via chemical vapor deposition. One might 

assume that water should be excluded from the nanotube interior due to its hydrophobicity and 

because water absorption is entropically disfavored.132 Surprisingly, water adsorbs inside the pores 

of  carbon nanotubes, leading to the phenomenon of “water wires”.133 One explanation proposes 

that water wires and carbon nanotube solubility results from the entropy of water flowing through 

the pores and from free rotation of water molecules. Studies also suggest dissolution in water does 

not involve favorable enthalpic interactions between water and the nanotube, suggesting water 

solubility of other hydrophobic materials through entropy-favored interactions. Changing the 

nanotube polarity, via simulations or experimental crystal engineering, allows for tuning of the 

water wire mobility, specifically in terms of water migration from one opening of the tube to the 

other, functioning as a nanosized garden hose. Carbon nanotubes also dissolve in ionic 

liquids,134,135 where both solvation shells and internal hydrogen bonding are postulated to stabilize 

the particles. Similar solvation behavior has been observed for carbon nanotubes in benzene,136 



 

28 

 

 

alcohols,137 and polymer solutions,138 however nearly all studies are computational. Although 

challenging, experimental studies of nanomaterials solvation will lay the foundation for designing 

their application in solution state applications, such as drug delivery or membranes. 

Whereas nearly all nanotubes are homogenous and comprised of carbon, the solvation of 

2D materials introduces the probability of surface heterogeneity (Figure I.3B). In addition to the 

atomically homogeneous example of graphene, 2D materials include the heterogeneous boron 

nitride, metal chalcogenides, or metal oxides. In these materials, solvent accessible surface areas 

exist between sheets (interlayer), introducing another unique solvation environment in addition to 

interior and exterior pores. Despite their enhanced surface areas, most nanosheets require a 

surfactant or post-synthetic modification for long-term colloidal stability, similar to 3D, non-

porous inorganic analogs.139–141 As illustrated by the examples below, electrical double layers are 

thought to spontaneously assemble at the surface of nanosheets, providing a large electrostatic 

surface repulsion (as measured by zeta potential) and site-specific water interactions at the surface. 

These results suggest that in high-surface area materials without deliberately added surface 

 
Figure I. 3 Illustration of A) 1-dimensional (polymers, nanotubes), B) 2-dimensional (layered 

or sheet-like materials), and C) 3-dimensional structures (porous materials) and their possible 

solvent interactions. 

 

Figure I.3  



 

29 

 

 

capping agents, specific solvent interactions are important for colloidal stability. Surprisingly, 

hydrophobic 2D materials, including graphene and MoS2, disperse in aqueous solutions via 

exfoliation-induced oxidation.142,143 Studies suggest that ultrasonication causes etching of edge-

groups that improves the solubility in water. In a notable study, graphene ultrasonicated at high-

temperatures exhibited long-term colloidal stability in water while the sample sonicated at low-

temperatures maintained a pristine morphology and was unstable in water.144 The introduction of 

edge-site functionalization to graphene oxide in the form of hydroxyl and carboxyl groups also 

contributes to the increased water stability. However, for heterogeneous nanosheets of hexagonal 

boron nitride, MoS2, WS2, and MoSe2, colloidal stability in water was achieved after sonication at 

both high- and low-temperatures and no surface functionalization was observed.  

3D porous materials with relatively simple compositions include microporous 

nanoparticles such as silica and zeolites (Figure I.3C). Mesoporous silica (SiO2) has pore 

diameters of 2-50 nm, while microporous silica has much smaller pores below 2 nm. Zeolites are 

aluminosilicates with even smaller pore diameters ranging from 0.3-0.8 nm. As with colloidal 

nanoparticles, polymers, and proteins detailed above, the colloidal stability of 3D porous 

nanomaterials depends on proper electrolyte concentration,145 surfactant surface coverage,146,147 

and polymer coatings.148 Studies remain largely empirical and an underlying mechanism of 

colloidal stability remains unclear. Representative studies include the finding that mesoporous 

silica is typically synthesized with a surfactant like hexadecyltrimethylammonium bromide 

(CTAB) acting as both a structure-directing molecule (to synthesize a specific shape) and a 

stabilizing capping ligand.149–151 These surfactant-capped materials can be stable for upwards of a 

year in water.152 Mesoporous silica nanoparticles can also be surface functionalized as 

hydrophobic, which renders them useful drug delivery agents.153,154 Little is known about the 

solvation and colloidal stability of mesoporous silica, although recent work suggests that 

performing dialysis solvent exchange or coating the nanoparticles with proteins improves 

dispersability.155,156  

A potential method for probing mechanisms of colloidal stability in porous materials would 

rely on surface functionalization. Although porous, silica and zeolites lack chemical tunability 

beyond their typical inorganic compositions (Figure I.4B). Instead, organic-inorganic framework 

materials assemble from a wide variety of metallic and organic-compounds, resulting in a diverse 



 

30 

 

 

range of pore diameters, aperture shapes, solvent-accessible surface area, morphology, and well-

developed methods for post-synthetic modulation of surface compositions. 

 
I.D.2. Porous Frameworks 

Understanding the colloidal stability of porous frameworks attracts intense recent attention, in part 

because 3D porosity challenges conventional mechanisms of solvation, such as the notion of 

electrostatic forces between smooth, uniform, hard spheres. Porous framework materials metal-

organic frameworks (MOFs) are pursued for a wide range of applications due to their unparalleled 

performance at selective gas separation, water filtration, carbon sequestration, and other areas 

leveraging tunable guest-host interactions (Figure I.4A). Practical implementation as solution 

processible and reusable materials, such as thin film membranes, demands their ability to suspend 

as uniform, stable colloids in a range of solvents. Stability in water would facilitate their utility in 

biological applications, for example, while compatibility with low-boiling solvents would render 

them amenable to spraying coating and other forms of industrial production at-scale. In addition 

to a lack of knowledge of solvation structure, few studies exist for any form of surface 

functionalization. Reported surface ligands deviate from those well studied with conventional 

quantum dots.157 Solvent interactions likely dictate the interaction of porous materials with 

polymers in the so-called mixed matrix membranes envisioned for chemical separation 

technologies. Little is known about microscopic aspects of the polymer-porous colloid interface, 

except that great care must be taken to prevent polymers from intercalating and clogging 

pores,158,159 which can be detrimental to gas sorption and chemical separation applications. In this 

section, we will describe the current theories of colloidal stability for porous frameworks and the 

challenges in studying them. 

Unlike typical, nonporous nanoparticles, MOFs exhibit colloidal stability without the need 

for conventional capping ligands.160  In fact, while traditional surfactants such as dodecanoic acid 

or cetyltrimethylammonium bromide can be included in a MOF synthesis to impart size or shape 

control,161,162 these ligands do not remain with the MOF nanoparticles after washing, as we have 

observed by nuclear magnetic resonance spectroscopy, thermogravimentric analysis, and other 

analytical techniques.160,163 Instead, while post-synthetic addition of ligand dyes157 or DNA/protein 

coronas164 leads to functionalized surfaces, few if any studies detail how they impact colloidal 

stability. Instead, we recently demonstrated that the solvent identity plays a greater role in 



 

31 

 

 

stabilizing the surface as only solvents that can dissolve the organic linker can suspend the MOF 

nanoparticle.160 This result strongly suggests that the interaction of MOF nanoparticles resembles 

the dissolution of cage molecules, polymers, and other macromolecules systems where solubility 

depends solvent-monomer energetics and accessible void spaces (Figure I.4C,D). 

In addition to specific solvent-surface interactions, classical electrostatic arguments of 

DLVO theory could explain the stability of uncapped MOF nanoparticles, where surface charges 

would arise from deprotonated linker molecules or open metal sites. Electrostatics alone is unlikely 

to account for the full mechanism of colloidal stability, however.165 Recent reports of zeta 

potentials—indicators of net surface charge— typically at the border (±30 mV) of values required 

for colloidal stability by DLVO theory.160,166 These reduced zeta potentials can be attributed to the 

low surface density of porous nanoparticles and the weak, short-range electrostatic fields produced 

by surface defects that quickly decayed through porous channels.167 Binding of solvent or 

polymers to open metal sites at the surface of MOF nanoparticles improves colloidal stability, 

 
Figure I. 4 Representation of A) framework materials, comprised of inorganic/organic 

building units that are assembled into porous frameworks and B) inorganic porous materials, 

which are made up of secondary building units. In C) both material classes show distinct 

internal vs. external surfaces due to their high solvent-accessible surface areas. Direct 

functional groups or open metal sites within the materials D) may induce different solvent 

arrangements. 

 



 

32 

 

 

while also reducing the apparent surface charge.159,163 In other words, zeta potentials may not relate 

directly to MOF colloidal stability. 

Instead of traditional electrostatics, we suggest the intrinsic porosity of nanoMOFs 

improves colloidal stability through entropic effects. For nonporous colloids, entropy favors 

aggregation because it removes solvation shells and disorders excluded solvent. This process is 

termed the “hydrophobic effect” and was introduced as a mechanism to describe porous 

nanoparticles,168 where the total amount of ordered solvent must be lower than nonporous colloids 

due to the low density surfaces. As a result, the potential energy favoring aggregation decreases as 

well. To examine the interaction of solvent with porous interfaces, we use VSFSS to measure the 

orientation of N,N-dimethylformamide (DMF)  or water solvation shells at the exterior of Zn(2-

methylimidazolate)2 (ZIF-8) colloids. Although predicted to exist, solvation shells of colloids were 

previously documented only by  atomic force microscopy and x-ray studies.169–173 In addition to 

providing a protective shell of steric repulsion, we propose that solvent interactions with open 

metal sites and linker monomer units improves solvation energetics akin to Flory-Huggins theory. 

We also observed that the bridging 2-methylimidazolate linkers spontaneously align at the solvent 

interface in a manner that expands the internal pore volumes. This lattice flexibility, like a protein, 

may further enhance favorable solvent interactions and colloidal stability. Taken together, these 

recent reports provide a roadmap for preparing and stabilizing porous materials in solution. 

This thesis describes research which probes the surface structure of colloidal materials that 

are stabilized by non-DLVO forces, including factors like steric stabilization, low electrostatic 

potential, sparse surface density, particle porosity, and surface heterogeneity. A background on the 

surface-specific technique of vibrational sum frequency scattering spectroscopy is provided in the 

following chapter, laying the groundwork for using this powerful tool to study the interfaces of 

uniquely stable colloidal materials.  

  



 

33 

 

 

CHAPTER II: VIBRATIONAL SUM FREQUENCY SCATTERING SPECTROSCOPY 

(VSFSS) 

 

II.A.VSFSS THEORY 

Developed by Roke et. al.,174 vibrational sum frequency scattering spectroscopy (VSFSS) 

is a scattering adaptation of the reflection-based vibrational sum frequency spectroscopy 

introduced by Shen et. al..175 While reflection VSFS has been commonly used to probe planar 

air/water176, oil/water177, oxide/water 178, and silica/water179 interfaces, the scattering variation of 

this technique enables investigation into the interface of nano-sized colloids.180,181 In the scattering 

experiment, a visible and tunable infrared (IR) pulse are overlapped spatially and temporally at the 

 
Figure II. 1 A) Schematic of VSFSS, in which a scattered sum frequency response (blue) is 

generated by overlapping a visible (green) and IR (red) pulse spatially and temporally at a 

curved interface. The energy level diagram of VSFSS is shown in the bottom right. B) 

Cartoon representation of the possible scenarios in which a sum frequency signal could be 

expected. IR transition dipoles at anisotropic surfaces (top) are sum frequency active while 

mismatched or directly opposing transition dipoles (bottom) are sum frequency inactive. C) 

Schematic of the ssp polarization scheme in which the sum frequency, visible beams, and IR 

beams are s, s, and p polarized, respectively. This polarization combination probes vibrational 

dipoles perpendicular to an interface. 



 

34 

 

 

curved interface to generate a scattered sum frequency response that is the sum of the two incoming 

beams (Figure II.1A). VSFSS directly probes vibrational modes at interfaces due to selection rules 

under the electric dipole approximation prohibiting contributions from a centrosymmetric 

environment such as a bulk liquid phase. The signal indicates two main attributes of the chemical 

environment: the population of molecules at the interface and the average molecular orientation of 

vibrational modes, also referred to as net molecular ordering. Only highly ordered resonant 

vibrations generate a sum frequency response (Figure II.1B).  

In VSFSS, the intensity of the scattered SF response is proportional to the intensities of the 

incoming visible and IR beams (𝐼𝐼𝑅𝐼𝑣𝑖𝑠) and the square modulus of the second-order nonlinear 

susceptibility (𝛸(2)) as shown in Equation 5.  

𝐼𝑆𝐹 ∝  𝐼𝐼𝑅𝐼𝑉𝑖𝑠|𝜒(2)|
2

 (5) 

In a centrosymmetric environment where all directions are equivalent, 𝜒𝑖𝑗𝑘
(2)

 is identical in two 

opposite directions such as: 𝜒𝑖𝑗𝑘
(2)

= 𝜒−𝑖−𝑗−𝑘
(2)

. Additionally, 𝜒𝑖𝑗𝑘
(2)

 is a third rank tensor and obeys the 

property that changing the sign of the three subscripts is the same as reversing the axis system like:  

𝜒𝑖𝑗𝑘
(2)

= −𝜒−𝑖−𝑗−𝑘
(2)

. To satisfy both of the above symmetry properties, 𝜒𝑖𝑗𝑘
(2)

 must equal 0, and thus a 

SF response in a centrosymmetric environment is forbidden. Therefore, VSFS is a surface specific 

technique as the interface of two materials in non-centrosymmetric and results in non-zero tensor 

elements. VSFS involve 3 electric fields (visible, IR, and SF), thus 𝜒(2) is a 3x3x3 tensor with 27 

elements. These elements are reduced through symmetry considerations to result in seven non-

zero elements that contribute to a SF response. Only four of these elements are unique. These 

contributions physically represent different orientations of the vibrational mode dipoles relative to 

the interface and can be probed through a variety of polarization combinations. The possible 

polarization combinations and the respective 𝜒𝑖𝑗𝑘
(2)

 element(s) that contribute to the spectrum are: 

pss (𝜒𝑧𝑦𝑦
(2)

), sps (𝜒𝑦𝑧𝑦
(2)

), ssp (𝜒𝑦𝑦𝑧
(2)

), and ppp (𝜒𝑧𝑧𝑧
(2)

, 𝜒𝑧𝑥𝑥
(2)

, 𝜒𝑥𝑧𝑥
(2)

, 𝜒𝑥𝑥𝑧
(2)

), where the polarizations are 

listed in the order of SF, visible, and infrared. The resonant portion of the second-order 

susceptibility is dependent on both the net orientation and population of molecules at the interface 

(Equation 6).  



 

35 

 

 

𝜒𝑅
(2)

=  
𝑁

𝜀0

〈𝛽𝑉〉 (6) 

Where 𝑁 is the number of molecules at the interface,  𝜀0 is the vacuum permittivity, and 〈𝛽𝑉〉 is 

the average macroscopic hyperpolarizability which accounts for the orientation of molecules at the 

interface. 

The 𝑋(2) has non-resonant and resonant terms which are accounted for in the fitting 

equation (7) as developed by Bain et al. 

|𝛸(𝜔)(2)|
2

=  |𝛸𝑁𝑅
(2)

𝑒𝑖𝜙 +  ∑ ∫
𝐴𝑣𝑒𝑖𝜙𝑣𝑒

−(
𝜔𝐿−𝜔𝑣

Г𝑣
)

2

𝜔𝐿 − 𝜔𝐼𝑅 + 𝑖Г𝐿
𝑑𝜔𝐿

+∞

−∞𝑣

|

2

 (7) 

where the amplitude of the non-resonant susceptibility is described by 𝛸𝑁𝑅
(2)

 with a phase 𝜙.182 The 

summation of all vibrational transitions that are SFG active describes the resonant susceptibility, 

where 𝐴𝑣 is the peak amplitude, 𝜙 is the phase, Г𝐿 is the Lorentzian linewidth describing 

homogenous broadening, and Г𝑣 is the Gaussian linewidth describing inhomogeneous broadening. 

Equation 5 is used to fit all experimental spectra. 

A practical aspect of VSFSS is the ability to change the polarization of the incoming and 

collected beams to isolate vibrational responses from unique molecular orientations relative  

to the interface. A three-letter nomenclature denotes the polarization of the beams in the sum-

frequency, visible, and IR beams, respectively (Figure II.1C). In this dissertation, ssp, sps, and 

ppp polarization schemes are used. The ssp polarization scheme probes IR transition moments 

perpendicular to the interface while the sps polarization probes IR transition moments aligned 

parallel to the interface. The ppp polarization combination probes a combination of transition 

moments in both the perpendicular and parallel contributions.  

 

II.B. LASER SETUP 

The VSFSS system used to collect the data presented in this dissertation is engineered as follows 

and in Figure II.2. A Ti:Sapphire regenerative amplifier laser (Coherent Libra) generates an 800 

nm, ~80 fs fundamental pulses with a 1 kHz repetition rate. A portion of that beam is used as the 

visible pulse while the remaining is sent through an optical parametric amplifier (Coherent OPerA 

Solo) to generate a broadband infrared (IR) beam through difference frequency  



 

36 

 

 

generation. The visible and IR pulses are then directed along a series of mirrors, lenses, and 

polarizers to become spatially and temporally overlapped at the sample stage. The sample cell is 

composed of a CaF2 window in the front and a quartz cuvette in the back (Helma QS) with an 

optical path length of 200 µm. The IR beam is focused at the sample to a spot size of ~80 µm with 

a parabolic gold mirror, while the visible beam is focused right after the sample cell to a ~500 µm 

spot size. The scattered SFG response is collected at an angle of ~60°, collimated with a plano-

convex lens, and focused into a spectrograph and accompanying charge-coupled device intensifier 

(Princeton Instruments IsoPlane SCT320 and PI-MAX4). For all experiments, the visible pulse 

energy was 25 µJ while the IR pulse energies were 2-3 µJ and 5 µJ for the C=O and C-H stretching 

regions, respectively. 

To account for daily fluctuations in laser power, a single trace in a figure is the result of at 

least 3 averaged trials across different days. One trial consists of 2 signal and 2 background 

measurements. Each trial is averaged, background subtracted, and normalized by a non-resonant 

response from KNbO3 to account for changes in IR spectral shape. In the C-H region, spectra are 

further normalized by the integrated SFG intensity (from 2800-3000 cm-1) generated from a d-

hexadecane emulsion in D2O stabilized with 1 mM AOT. Each trial is further normalized by the 

size of the droplet (determined by a monomodal distribution from DLS) through a scattering 

pattern developed by Roke et. al. that accounts for the percentage of scattered signal collected at 

60° which is dependent on droplet size and beam polarizations.183,184 More details on the 

 
Figure II. 2 Schematic of the VSFSS experiment utilized in this dissertation with visible 

(green), IR (red), and SF (blue) beam paths. Mirrors are omitted for clarity. 

 

 

  



 

37 

 

 

normalization procedure for VSFSS utilized in the Richmond/Scatena laboratory can be seen in a 

in Appendix G.185 Additionally, a calculation of the particle concentration necessary to overcome 

the signal to noise threshold for VSFSS is provided in Appendix H. 

The above procedure was used to collect all of the VSFS spectra presented in this 

dissertation and are not repeated in the experimental sections of each chapter for clarity. Chapter 

III begins the experimental section of this thesis by using VSFSS to study the surface structure of 

emulsions stabilized with a steric layer of polymer. 

  



 

38 

 

 

CHAPTER III: EMULSIONS STABILIZED BY POLY(ACRYLIC ACID) 

This work was published in volume 697 of the journal Colloids and Surfaces A: Physiochemical 

and Engineering Aspects in June 2024. Ashley N. Mapile is credited with conceptualization, 

formal analysis, investigation, writing – original draft, writing – review and editing,  and 

visualization. Lawrence F. Scatena assisted in conceptualization, writing – review and editing, 

supervision, and funding acquisition. Supplementary information for this chapter is provided in 

Appendix A 

 

III.A. AN INTRODUCTION TO POLY(ACRYLIC ACID) 

As explained in the introduction to this thesis, polyelectrolytes are useful analogues to studying 

protein coronas that are known to stabilize colloids due to the tunable nature of polymers. Many 

characteristics of polymers, such as the chain length (via molecular weight), chemical 

functionality, % protonation, and morphology can all be adjusted synthetically or by environmental 

conditions. Poly(acrylic acid) (PAA, Figure III.1) is a simple polyelectrolyte that can mimic 

protein behavior at interfaces. PAA, which contains a single carboxylic acid repeating group, can 

expand in solution when deprotonated or hydrogen bond with itself to form coils when protonated 

and is a useful material in forming zwitterionic polymer coarcervates.186–190 Through pH-

dependent zeta potential measurements (Appendix A, Figure A1) the interfacial pKa of PAA at 

the hexadecane/water emulsion interface was determined to be ~5.3, about 0.8 units more alkaline 

 
Figure III. 1 Chemical structure of poly(acrylic acid) and percent protonation curve as a 

function of pH for surface (solid, orange line) and bulk (dotted, purple line). Percent 

protonation was calculated with the Henderson-Hasselbalch equation, the known bulk pKa 

(4.5), and the experimentally derived interfacial pKa (5.3). 

 

 

 



 

39 

 

 

than the bulk pKa (Figure III.1). Therefore, studying the pH-dependent conformational behavior 

of PAA adsorbed at the emulsion interface can provide insight into droplets stabilized by pH-

sensitive protein coronas as well as a detailed understanding of the mechanisms of colloidal 

stability. 

The adsorption behavior of PAA without the addition of a surfactant, has been studied at 

both the planar CCl4/water and air/water interfaces using vibrational sum frequency spectroscopy 

(VSFS). These studies found that in conditions above the pKa (pH >4.5), the polymer is 

interfacially inactive, but the addition of charge-screening cations promotes adsorption of the 

polymer to the interface.105,191–194 Below pH 4.5, the polymer is interfacially active and exhibits a 

well-ordered structure as evident by strong CH and C=O VSFS spectral features. At a planar 

interface, molecular dynamics are thermodynamically determined by equilibrium surface activity 

and surface-active molecules develop a stable configuration given time. On the other hand, 

emulsions require the input of external energy to form droplets resulting in a kinetically trapped 

interface.98,101 During the formation of emulsions, surface-active molecules initially adsorb to the 

interface via ultrasonication but subsequently adsorb/desorb through equilibrium thermodynamics 

after formation of the curved interface. Thus, the conformational ordering of interfacial molecules 

adsorbed to emulsions is dictated by the thermodynamic equilibrium of molecular adsorption and 

the kinetic energy encouraging the platform of these colloids to destabilize and coalesce. 

Stabilization of emulsions is best described by the theory developed by Derjaguin, Landau, 

Verwey, and Overbeek (DLVO) which states that both electrostatic and steric repulsion prevents 

droplet coalescence and keeps the emulsions in a kinetically trapped state.8,98,101,195 Reports of 

emulsions coated with solid polymer particles (such as Pickering emulsions),196 nanoparticles with 

a polymeric coating,197 and emulsions formed with surfactant/polymer mixtures198 suggest unique 

behavior at curved hydrophobic interfaces as opposed to their planar counterparts. While there 

have been reports of emulsions stabilized with polymer alone, observed through Janus particles199 

or magnetic nanoemulsions,200 and specifically emulsions coated with diblock copolymers,201,202 

they lack insight into the detailed mechanism of stability promoted by polymers or the molecular 

level information to provide a structural understanding of the droplet interface. 

Here, we demonstrate the stability of emulsions coated with PAA and provide a picture of 

polymer molecular conformation with the surface-specific vibrational sum frequency scattering 

spectroscopy (VSFSS). Despite the lack of surface charge thought to be necessary for droplet 



 

40 

 

 

stabilization, PAA-coated emulsions prepared at pH 2 are stable and exhibit interfacial polymer 

ordering. Surprisingly, the highly charged pH 6 polymer does not stabilize emulsions and exhibits 

no molecular organization. With the addition of salt, we observe charge screening at the interface 

and a disruption to polymer ordering because of localized electrostatics. We propose that 

emulsions are stabilized through the steric hindrance of an adsorbed PAA layer, whose molecular 

conformation and charge-dependent behavior can provide insight to protein corona-stabilized 

droplets and their application as drug delivery vehicles. 

 

III.B. MATERIALS AND METHODS 

III.B.1. Materials 

All materials were used as delivered without further purification. Hexadecane (≥99%), dioctyl 

sodium sulfosuccinate (AOT, ≥97%), poly(acrylic acid) (PAA, average Mv ~ 450,000), calcium 

chloride (CaCl2, anhydrous, ≥97%), magnesium chloride (MgCl2, anhydrous, ≥98%), sodium 

chloride (NaCl, BioXtra, ≥99.5%), and ammonium chloride (NH4Cl, ACS reagent, ≥99.5%) were 

purchased from Sigma-Adrich. Deuterated hexadecane (n-hexadecane-d34, 98.6% D), sodium 

deuteroxide (NaOD, 99.5% D), and deuterium chloride (DCl, 99.8% D) were purchased from CDN 

Isotopes. Deuterium oxide (D2O, 99.9% D) was purchased from Cambridge Isotope Labs. All 

glassware was copiously cleaned in a bath containing sulfuric acid (98% Sigma-Aldrich) and 

AlNOCHROMIX oxidizer from Godax Laboratories Inc. After sitting in the acid bath for at least 

24 hours, glassware was rinsed for at least 2 min. with 18.2 MΩ-cm water and dried in an oven. 

 

III.B.2. Emulsion Formation 

Emulsions consisting of hexadecane (or d-hexadecane) suspended in water (or D2O) were prepared 

by ultrasonication (Branson Sonifier 250) of the sample at 5% output power (~12.5%) at 20 kHz 

for 5 minutes at a constant duty cycle. The ultrasonication probe tip was placed at the interface of 

the aqueous solution and oil layer to ensure homogenous mixing. The AOT-coated emulsion 

standard was formed by ultrasonicating AOT stock solution with oil to give a 1mM AOT emulsion 

with 2.5% v/v oil in water. PAA-stabilized emulsions were formed in the same fashion with DCl 

or NaOD added to reach the desired pD (pH). Emulsion pH was measured using MilliporeSigma 

MColorpHast pH strips and an Oakton Instruments Portable Meter Kit. The detection of pH values 

can differ from pD values by 0.43,203 however for ease of communication, pH is used in this paper.  



 

41 

 

 

III.B.3. Dynamic Light Scattering and Zeta Potential 

Emulsion hydrodynamic diameter (Z-average), polydispersity index (PDI), and zeta potential were 

measured by a Malvern Zetasizer Nano ZS. Details regarding the theory and practice of dynamic 

light scattering (DLS) and zeta potential can be found elsewhere.204,205 From DLS measurements, 

hydrodynamic diameter and PDI was reported from an average of at least three measurements. 

DLS and zeta potential measurements were collected by pipetting 1 mL of emulsion solution into 

a Malvern folded capillary zeta cell, therefore maintaining consistent v/v oil concentrations. Zeta 

potential values were reported from an average of at least five measurements. Determination of 

the surface pKa using zeta potential measurements is detailed in the supporting information (Figure 

S1) and follows a procedure developed by Haes et. al.206 

 

III.C. RESULTS AND DISCUSSION 

III.C.1. Impact of PAA Concentration 

Emulsions were formed with concentrations of PAA ranging from 1 ppm to 4000 ppm in the 

natural pH conditions of PAA (pH 4) and measured by dynamic light scattering and zeta potential 

(Figure III.2). For context, previous experiments at the CCl4/water interface used 5 ppm PAA 

while air/water interface experiments used 4500 ppm.105,191 Figure A2 and Table A1 demonstrate 

that emulsions prepared at or below 250 ppm PAA were unstable after 1 week (Appendix A). Bare 

emulsions (those without deliberately added surface active agents) are known to have a negative 

zeta potential due to miniscule amounts of surface active contaminants.180,207 The zeta potential of 

our hexadecane/water emulsions (-32 mV) suggest minor surface active impurities. The presence 

 
Figure III. 2 Characteristics A) hydrodynamic diameter, B) polydispersity index, and C) zeta 

potential) of emulsions coated with varying amounts of PAA at pH 4. Points shaded orange 

indicate 500 ppm PAA. 

 
 



 

42 

 

 

of these impurities were identified in previous emulsion studies from our laboratory that employed 

meticulous and extensive cleaning procedures to minimize their presence.180 After PAA addition, 

the polymer is much more surface active than the impurities and displaces any contaminants that 

contributed to the negative surface charge. As additional PAA is added, charge screening and 

hydrogen bonding causes a reduction in the magnitude of the zeta potential.  

At a concentration of 500 ppm, PAA-stabilized emulsions exhibit good colloidal stability, 

moderately polydisperse size distributions, and a highly charged surface layer as indicated by the 

orange shaded points in Figure III.2. Additionally, as confirmed by pendant drop surface 

tensiometery, discussed in Appendix A and Figure A3, PAA adsorbs to the hexadecane/water 

interface at a concentration of 500 ppm. Emulsions formed using 250 ppm PAA and below resulted 

in unstable droplets presumably due to a lack of surface population. Despite a remarkably high 

magnitude zeta potential (≥30 mV) emulsions stabilized with 100 ppm PAA or below are unstable 

and crash out quickly after formation, likely due to a thin or completely absent polymer layer. 

While the addition of PAA at concentrations above 500 ppm leads to more stable emulsions, as 

indicated by the long-term stability, significant charge screening occurs making them difficult 

targets for measuring the interplay between electrostatics and steric repulsion when salt ions are 

added. The slight increase in emulsion size and reduction in zeta potential with increasing 

concentration of PAA is due to polymer layering at the surface which extends the hydrodynamic 

diameter and charge screens the surface as is consistent with other polymer-layering 

studies.16,201,208,209 Emulsions stabilized with 1500 ppm PAA or higher have a low zeta potential 

(~-7 mV) and are colloidally stable up to a week, solidifying the observation that sterics alone are 

enough to stabilize emulsions. Due to the desire to study the impact of surface charge screening, 

the emulsions used in this paper are formed with 500 ppm PAA as they have long-term colloidal 

stability, are  moderately polydisperse and have appreciable interfacial charge. 

 

III.C.2. Emulsions Stabilized by PAA Alone 

Emulsions coated with 500 ppm PAA were formed at pH 2, pH 4, and pH 6 via ultrasonication of 

2.5% v/v hexadecane in water. Pendant drop surface tensiometery measurements indicate that high 

amounts of PAA adsorb to the hexadecane/water interface at pH 2 and pH 4 while little adsorption 

is observed at pH 6 (Figure A4). Emulsions at pH 2 and pH 4 maintained colloidal stability after 

one week, as evident by their visual turbidity, while pH 6 oil droplets phase separated (Figure 



 

43 

 

 

A4). Dynamic light scattering of pH 2, pH 4, and pH 6 colloids measured 20 minutes after 

formation revealed hydrodynamic diameters of 477.7, 363.2, and 761.0 nm, respectively, 

demonstrating that the average size of emulsions prepared at pH 2 and 4 are in the nanosized 

regime while pH 6 emulsions are inherently larger. Narrow, monodisperse colloids have a PDI < 

0.1, while polydisperse systems have a PDI > 0.1.210 Moderate polydispersity is defined by a PDI 

from 0.1-0.4 and describes the pH 2 and pH 4 emulsions studied `here (PDIs of 0.247 and 0.272, 

respectively), while the pH 6 emulsions have broad polydispersity (PDI of 0.628). Similarly, size 

distribution plots from DLS measurements are monomodal for emulsions at pH 2 and pH 4, but 

multimodal for pH 6 (Figure A5). Table S2 lists the PDI of each emulsions system studied by 

VSFSS which are in line with recent studies on emulsion formulations.201