• Bursting the bubble: A molecular understanding of surfactant-water interfaces

The importance of amphiphilic surfactants is difficult to overstate. Hygiene, cosmetics, foods, oil recovery, and, increasingly, organic synthesis are some of the industries where detergents are being used as emulsifiers and thickening agents. It is not surprising that the fundamental chemistry of amphiphile behavior—solubilizing polar and nonpolar molecules—is taught as early as high school and that the mechanisms of saponification reactions are standard fare in undergraduate organic chemistry textbooks.^1,21. S. Sutheimer, J. M. Caster, and S. H. Smith, J. Chem. Educ. 92, 1763 (2015). https://doi.org/10.1021/acs.jchemed.5b001882. M. C. S. de Mattos and D. E. Nicodem, J. Chem. Educ. 79, 94 (2002). https://doi.org/10.1021/ed079p94 However, as is often the case in our understanding of molecular behavior, this ubiquity belies a staggering complexity, which presents many opportunities for study.

Addition/elimination saponification reactions can be straightforward to carry out in the laboratory, but real-world reaction conditions are more complicated. Fatty acid components are commonly extracted from natural sources, such as palm oil, which leads to variability in the surfactant tail length. Palm oil, for example, is primarily palmitic acid, but this component only accounts for ∼44% by weight of the total mixture.³3. Y. B. C. Man, T. Haryati, H. M. Ghazali, and B. A. Asbi, J. Am. Oil Chem. Soc. 76, 237 (1999). https://doi.org/10.1007/s11746-999-0224-y The oil of choice matters greatly: cocoa butter, as another example, contains roughly equal amounts of 16:0, 18:0, and 18:1 fatty acids.⁴4. A. S. Bhatnagar, P. K. P. Kumar, J. Hemavathy, and A. G. G. Krishna, J. Am. Oil Chem. Soc. 86, 991 (2009). https://doi.org/10.1007/s11746-009-1435-y While the natural oils can be purified, many industrial surfactants are, nevertheless, synthesized with a heterogeneous mixture of reactants. Furthermore, the bulk synthesis of detergents is done in a one-pot reaction where the degree of headgroup substitution is controlled by tuning the stoichiometric ratios of the fatty acid to the headgroup.^5,65. D. B. Tripathy, A. Mishra, J. Clark, and T. Farmer, C. R. Chim. 21, 112 (2018). https://doi.org/10.1016/j.crci.2017.11.0056. K. Hill and O. Rhode, Lipid-Fett 101, 25 (1999). https://doi.org/10.1002/(sici)1521-4133(19991)101:1<25::aid-lipi25>3.0.co;2-n For headgroups capable of forming only a single ester linkage, this is not a problem. However, for polyol headgroups, there may be as many addition reactions as there are hydroxyl groups, leading to a wide array of structures present in a given sample. We have shown in a previous study of sorbitan surfactants that an uncontrolled synthesis can generate over 5000 individual surfactant structures, all of which may be present in the final sample.⁷7. C. P. Baryiames, M. Teel, and C. R. Baiz, Langmuir 35, 11463 (2019). https://doi.org/10.1021/acs.langmuir.9b01693 Even if a truly homogeneous surfactant was made, emulsifiers are often employed in high-temperature, acidic conditions—the exact kind required to reverse the saponification reaction and separate the fatty acid from the headgroup. This new equilibrium may scramble fatty acid substituents, generating new surfactant species not present in the initial sample.

The consequences of structural heterogeneity are beginning to be uncovered. Most studies of industrially-relevant surfactants have been performed on bulk solutions, which have been robustly characterized.^8–118. A. M. Bellocq, J. Biais, P. Bothorel, B. Clin, G. Fourche, P. Lalanne, B. Lemaire, B. Lemanceau, and D. Roux, Adv. Colloid Interface Sci. 20, 167 (1984). https://doi.org/10.1016/0001-8686(84)80005-69. D. Levitt, A. Jackson, C. Heinson, L. N. Britton, T. Malik, V. Dwarakanath, and G. A. Pope, SPE Res Eval & Eng 12, 243–253 (2009). https://doi.org/10.2118/100089-PA10. H. Pawignya, A. Prasetyaningrum, E. R. Dyartanti, T. D. Kusworo, and B. Pramudono, AIP Conf. Proc. 1710, 030055 (2016). https://doi.org/10.1063/1.494152111. M. Korhonen, J. Hirvonen, L. Peltonen, O. Antikainen, L. Yrjänäinen, and J. Yliruusi, Int. J. Pharm. 269, 227 (2004). https://doi.org/10.1016/j.ijpharm.2003.09.020 Indeed, surfactant science has largely relied on bulk experiments to characterize the properties of microemulsions such as phase stability, interfacial tension, and solute encapsulation. Such function-forward, thermodynamic studies have generated useful data that allow for detergents to be used in a host of applications.^5,12,135. D. B. Tripathy, A. Mishra, J. Clark, and T. Farmer, C. R. Chim. 21, 112 (2018). https://doi.org/10.1016/j.crci.2017.11.00512. L. J. Peltonen and J. Yliruusi, J. Colloid Interface Sci. 227, 1 (2000). https://doi.org/10.1006/jcis.2000.681013. M. S. Kamal, I. A. Hussein, and A. S. Sultan, Energy Fuels 31, 7701 (2017). https://doi.org/10.1021/acs.energyfuels.7b00353 However, the intense focus on bulk behavior presents a rich vein of potential study to uncover the relationship between the surfactant structure, intermolecular interactions, interfacial properties, and bulk behavior. Consider, for example, predicting the microemulsion phase of a surfactant. The most common formulas assessing this fundamental characteristic are based on geometric arguments, which use the ratio of the polar headgroup’s size to the volume occupied by hydrophobic moieties to predict the preferred packing arrangement: water-in-oil if the hydrophobic region dominates, oil-in-water if the surfactant is hydrophilic, and bicontinuous phases if the hydrophilic and hydrophobic regions are balanced.^8,138. A. M. Bellocq, J. Biais, P. Bothorel, B. Clin, G. Fourche, P. Lalanne, B. Lemaire, B. Lemanceau, and D. Roux, Adv. Colloid Interface Sci. 20, 167 (1984). https://doi.org/10.1016/0001-8686(84)80005-613. M. S. Kamal, I. A. Hussein, and A. S. Sultan, Energy Fuels 31, 7701 (2017). https://doi.org/10.1021/acs.energyfuels.7b00353 This overly-simplistic prediction identifies a single preferred phase, but there is a large body of work showing that the molar ratios of oil:surfactant:water in the system, along with environmental factors such as temperature, can result in the surfactant adopting many phases.^11,14,1511. M. Korhonen, J. Hirvonen, L. Peltonen, O. Antikainen, L. Yrjänäinen, and J. Yliruusi, Int. J. Pharm. 269, 227 (2004). https://doi.org/10.1016/j.ijpharm.2003.09.02014. A. P. Carpenter, E. Tran, R. M. Altman, and G. L. Richmond, Proc. Natl. Acad. Sci. U. S. A. 116, 9214 (2019). https://doi.org/10.1073/pnas.190080211615. G. Garcia-Olvera, T. M. Reilly, T. E. Lehmann, and V. Alvarado, Energy Fuels 31, 95 (2017). https://doi.org/10.1021/acs.energyfuels.6b01413 What, then, are the structure/phase relationships for a given detergent molecule? Furthermore, what is the effect of structural entropy on this phase behavior, and how do mixtures of dissimilar surfactants interact to give rise to the bulk properties that have been identified as beneficial for industrial processes?

To begin unraveling these questions and others, biologically occurring lipids present an analogy to detergents. Both are amphiphilic molecules (indeed, some synthetic surfactants are designed using lipids as a template) that exist in highly heterogeneous environments.^5,6,16–185. D. B. Tripathy, A. Mishra, J. Clark, and T. Farmer, C. R. Chim. 21, 112 (2018). https://doi.org/10.1016/j.crci.2017.11.0056. K. Hill and O. Rhode, Lipid-Fett 101, 25 (1999). https://doi.org/10.1002/(sici)1521-4133(19991)101:1<25::aid-lipi25>3.0.co;2-n16. A. Carmona-Ribeiro, Curr. Med. Chem. 10, 2425 (2003). https://doi.org/10.2174/092986703345661117. F. Schmid, Biochim. Biophys. Acta, Biomembr. 1859, 509 (2017). https://doi.org/10.1016/j.bbamem.2016.10.02118. Y. Suzuki, Y. Fujita, and K. Kogishi, Am. Rev. Respir. Dis. 140, 75 (1989). https://doi.org/10.1164/ajrccm/140.1.75 Especially interesting is the increasing application of ultrafast spectroscopies, molecular dynamics simulations, and probe molecule syntheses, which have empowered the biophysics community to make insights into the interfacial properties of lipid bilayers that were hitherto inaccessible using conventional techniques.^19–2519. M. L. Valentine, A. E. Cardenas, R. Elber, and C. R. Baiz, Biophys. J. 115, 1541 (2018). https://doi.org/10.1016/j.bpj.2018.08.04420. M. L. Valentine, A. E. Cardenas, R. Elber, and C. R. Baiz, Biophys. J. 118, 2694 (2020). https://doi.org/10.1016/j.bpj.2020.04.01321. P. C. Singh, K. Inoue, S. Nihonyanagi, S. Yamaguchi, and T. Tahara, Angew. Chem., Int. Ed. 55, 10621 (2016). https://doi.org/10.1002/anie.20160367622. X. Chen, J. Wang, A. P. Boughton, C. B. Kristalyn, and Z. Chen, J. Am. Chem. Soc. 129, 1420 (2007). https://doi.org/10.1021/ja067446l23. Y. Nagata and S. Mukamel, J. Am. Chem. Soc. 132, 6434 (2010). https://doi.org/10.1021/ja100508n24. R. K. Venkatraman and C. R. Baiz, Langmuir 36, 6502 (2020). https://doi.org/10.1021/acs.langmuir.0c0087025. J. C. Flanagan, M. L. Valentine, and C. R. Baiz, Acc. Chem. Res. 53, 1860 (2020). https://doi.org/10.1021/acs.accounts.0c00302 These studies have helped develop an image of the bilayer–water interface that contains molecular-scale structural and dynamical information and has begun uncovering the importance of lipid diversity in the biological function.²⁶26. H. I. Okur, O. B. Tarun, and S. Roke, J. Am. Chem. Soc. 141, 12168 (2019). https://doi.org/10.1021/jacs.9b02820 Indeed, recently, ultrafast methods have been applied to measure interfacial dynamics within complex multicomponent surfactant interfaces that approach, or arguably exceed, the complexity of biological membranes.

Sum-frequency-generation (SFG) spectroscopy and two-dimensional infrared (2D IR) spectroscopy are both proven methods for studying the surfactant–water interface and will be the focus of this Perspective.^25,27–2925. J. C. Flanagan, M. L. Valentine, and C. R. Baiz, Acc. Chem. Res. 53, 1860 (2020). https://doi.org/10.1021/acs.accounts.0c0030227. A. Ghosh, J. S. Ostrander, and M. T. Zanni, Chem. Rev. 117, 10726 (2017). https://doi.org/10.1021/acs.chemrev.6b0058228. P. Hamm and M. Zanni, Concepts and Methods of 2D Infrared Spectroscopy (Cambridge University Press, Cambridge, 2011).29. S. Yamaguchi and T. Tahara, J. Phys. Chem. C 119, 14815 (2015). https://doi.org/10.1021/acs.jpcc.5b02375 SFG reports inherently on the interfacial environment, generating a signal where the inversion symmetry is broken.^30–3430. X. Chen and Z. Chen, Biochim. Biophys. Acta, Biomembr. 1758, 1257 (2006). https://doi.org/10.1016/j.bbamem.2006.01.01731. C. Chen, M. A. Even, J. Wang, and Z. Chen, Macromolecules 35, 9130 (2002). https://doi.org/10.1021/ma020614j32. S. Kataoka and P. S. Cremer, J. Am. Chem. Soc. 128, 5516 (2006). https://doi.org/10.1021/ja060156k33. A. G. F. de Beer, J.-S. Samson, W. Hua, Z. Huang, X. Chen, H. C. Allen, and S. Roke, J. Chem. Phys. 135, 224701 (2011). https://doi.org/10.1063/1.366246934. D. Verreault, W. Hua, and H. C. Allen, J. Phys. Chem. Lett. 3, 3012 (2012). https://doi.org/10.1021/jz301179g It is used to determine the structures and orientations of molecules at the interface. In the case of vibrational SFG, the hydroxyl, methyl, and methylene regions of the IR spectrum are often studied because SFG is limited to functional groups with high molar absorptivity or high concentration in the region of interest. This is due to the inherently low signal of the technique as the number of oscillators is limited to a single interface. 2D IR, by contrast, provides time-resolved spectra that allow the fs-ns dynamics of molecular motions to be quantified. It lacks inherent interfacial selectivity, though careful experimental design does report on these factors. 2D IR typically provides improved signal-to-noise ratios compared to SFG, which allows it to interrogate a wider selection of functional groups. These include phosphates and carbonyls, functional groups commonly used in surfactants.^35,3635. R. Costard, T. Tyborski, B. P. Fingerhut, and T. Elsaesser, J. Chem. Phys. 142, 212406 (2015). https://doi.org/10.1063/1.491415236. B. Guchhait, Y. Liu, T. Siebert, and T. Elsaesser, Struct. Dyn. 3, 043202 (2015). https://doi.org/10.1063/1.4936567 These methods are highly complementary, jointly providing the structural and dynamic information needed to understand the interfacial environment (Scheme 1). In addition to their established strengths, developments in both spectroscopies have continued to expand the range of systems accessible to ultrafast methods. 2D SFG combines the structural and temporal resolutions of SFG and 2D IR, enabling the direct measurement of structural dynamics.^37,3837. W. Xiong, J. E. Laaser, R. D. Mehlenbacher, and M. T. Zanni, Proc. Natl. Acad. Sci. U. S. A. 108, 20902 (2011). https://doi.org/10.1073/pnas.111505510838. E. H. G. Backus, J. D. Cyran, M. Grechko, Y. Nagata, and M. Bonn, J. Phys. Chem. A 122, 2401 (2018). https://doi.org/10.1021/acs.jpca.7b12303 Similarly, 2D IR has recently been used to obtain spectra from Langmuir monolayers, previously inaccessible due to limitations in signal detection.^39,4039. M. K. Petti, J. S. Ostrander, V. Saraswat, E. R. Birdsall, K. L. Rich, J. P. Lomont, M. S. Arnold, and M. T. Zanni, J. Chem. Phys. 150, 024707 (2019). https://doi.org/10.1063/1.506551140. C. Yan, J. E. Thomaz, Y.-L. Wang, J. Nishida, R. Yuan, J. P. Breen, and M. D. Fayer, J. Am. Chem. Soc. 139, 16518 (2017). https://doi.org/10.1021/jacs.7b06602

Additionally, both methods have been complemented by advances in molecular dynamics (MD) simulations, which provide atomistic structural resolution and fs time resolution.^41,4241. J. Faeder and B. M. Ladanyi, J. Phys. Chem. B 104, 1033 (2000). https://doi.org/10.1021/jp993076u42. I. Nezbeda, F. Moučka, and W. R. Smith, Mol. Phys. 114, 1665 (2016). https://doi.org/10.1080/00268976.2016.1165296 For example, molecular dynamics simulations can visualize heterogeneous mixtures that are too complex to be readily assessed through experiments. Spectroscopic signals arise from whichever molecules happen to reside in the beampath and cannot distinguish one detergent from another without exacting the experimental design. Through the use of electrostatic maps, which calculate the vibrational frequency of a given oscillator in a simulation based on the bond’s solvation environment, spectra of complex systems may be calculated and compared to experimental measurements.⁴³43. S. C. Edington, J. C. Flanagan, and C. R. Baiz, J. Phys. Chem. A 120, 3888 (2016). https://doi.org/10.1021/acs.jpca.6b02887 When simulations accurately recreate experimental lineshapes, they can be used to interpret these lineshapes by providing information that the experiment alone cannot provide. Specific interactions, arrangements, and dynamics experienced by each atom in the system, regardless of complexity, may all be quantified. Together, ultrafast spectroscopies and MD simulations provide more information than either method apart.

Biophysical methods are beginning to be applied to non-biological amphiphilic self-assemblies, including surfactant microemulsions. The structure–function relationships of detergents, the nature of the surfactant–water interface, and the archetypal interactions that govern industrially-relevant bulk behaviors have been the focus of recent studies. Despite these advances, many opportunities to explore the fundamental properties of detergents remain. Understanding the relation between intermolecular interactions, surface properties, and surfactant thermodynamics could enable a “rational design” approach to identifying surfactant compositions for industrial and scientific applications. In this Perspective, we catalog some of the advances made in unraveling the interactions between the surfactant structure, interfacial composition, and interfacial environments. Reverse micelles (RMs)—microemulsions where surfactants solubilize a small amount of water into bulk oil—have been used extensively as model systems for probing interfacial dynamics.^41,4441. J. Faeder and B. M. Ladanyi, J. Phys. Chem. B 104, 1033 (2000). https://doi.org/10.1021/jp993076u44. T. H. van der Loop, M. R. Panman, S. Lotze, J. Zhang, T. Vad, H. J. Bakker, W. F. C. Sager, and S. Woutersen, J. Chem. Phys. 137, 044503 (2012). https://doi.org/10.1063/1.4736562 We present a selection of recent advances with a focus on ultrafast spectroscopy techniques. We also present preliminary results demonstrating that an improved understanding of the reverse micelle interfaces is essential to understanding chemical reactions carried out in these complex environments.

Ultrafast vibrational spectroscopy probes molecular interactions at the surfactant–water interface by measuring structural dynamics with sub-picosecond time resolution.^45–4745. Y. S. Kim and R. M. Hochstrasser, J. Phys. Chem. B 113, 8231 (2009). https://doi.org/10.1021/jp811397846. R. A. Nicodemus, K. Ramasesha, S. T. Roberts, and A. Tokmakoff, J. Phys. Chem. Lett. 1, 1068 (2010). https://doi.org/10.1021/jz100138z47. Q. Du, R. Superfine, E. Freysz, and Y. R. Shen, Phys. Rev. Lett. 70, 2313 (1993). https://doi.org/10.1103/physrevlett.70.2313 Hydrogen bond populations and lifetimes can be directly measured using these methods. Much effort across synthetic, analytical, and computational disciplines has resulted in a library of vibrational chromophores providing rich molecular information on the interface.^48–5048. S. D. Fried, S. Bagchi, and S. G. Boxer, J. Am. Chem. Soc. 135, 11181 (2013). https://doi.org/10.1021/ja403917z49. R. J. Xu, B. Blasiak, M. Cho, J. P. Layfield, and C. H. Londergan, J. Phys. Chem. Lett. 9, 2560 (2018). https://doi.org/10.1021/acs.jpclett.8b0096950. H. Lee, J.-H. Choi, and M. Cho, Phys. Chem. Chem. Phys. 12, 12658 (2010). https://doi.org/10.1039/c0cp00214c There are several considerations when selecting a probe: localization within the system, responsiveness to the local environment, and ease of interpretation of the resulting spectra. These criteria vary depending on the method being used. In the case of SFG spectroscopy, only molecules located within a few nanometers of an interface contribute to the measured signal, therefore allowing to probe solvent dynamics at the interface. For techniques that lack this intrinsic selectivity, probe selection becomes a key consideration. Metal carbonyls are often used because they can be solubilized in a variety of media and the M—C≡O stretch generates a very strong signal in an uncluttered region of the spectrum.^51,5251. C. R. Baiz, P. L. McRobbie, J. M. Anna, E. Geva, and K. J. Kubarych, Acc. Chem. Res. 42, 1395 (2009). https://doi.org/10.1021/ar900026352. J. T. King, M. R. Ross, and K. J. Kubarych, J. Phys. Chem. B 116, 3754 (2012). https://doi.org/10.1021/jp2125747 There has, however, been a push to use intrinsic probes—functional groups naturally present in the system—which are non-perturbative. Regardless of whether an intrinsic or extrinsic probe is used, lineshapes must be sensitive to local environments. This may manifest as, for example, a Stark-effect-like peak shift, a change in intensity, or a change in full-width-half-max, among others. Carbonyl spectra, for example, redshift by ∼15 cm⁻¹ when the oxygen atom participates in a hydrogen bond, enabling the number of hydrogen bond populations and their relative populations to be quantified.^48,5348. S. D. Fried, S. Bagchi, and S. G. Boxer, J. Am. Chem. Soc. 135, 11181 (2013). https://doi.org/10.1021/ja403917z53. S. D. Fried and S. G. Boxer, Acc. Chem. Res. 48, 998 (2015). https://doi.org/10.1021/ar500464j A strong spectral dependence on the probe’s local electrostatic environment is desirable.

Vibrational probes can report on both the interfacial structure and dynamics. Understanding the structure facilitates visualizing the system—surfactant geometries, packing efficiencies, and preferential localization are essential factors to consider when describing the interface. Using water O—H stretching modes, for example, reports on the structure and dynamics of water. CH₂ stretching vibrations report on the surfactant ordering and packing within the alkyl tail region. Interfacial dynamics necessarily differ from bulk dynamics as molecules due to a disruption in the extended H-bond networks present in water.^54,5554. R. A. Nicodemus, S. A. Corcelli, J. L. Skinner, and A. Tokmakoff, J. Phys. Chem. B 115, 5604 (2011). https://doi.org/10.1021/jp111434u55. D. E. Moilanen, E. E. Fenn, D. Wong, and M. D. Fayer, J. Chem. Phys. 131, 014704 (2009). https://doi.org/10.1063/1.3159779 A water molecule in the plane of the interface is spatially isolated from other water molecules, preventing it from participating in the hydrogen bond network. However, by virtue of this isolation, interfacial water molecules are influenced by the interface. Changes in the interfacial structure manifest as changes in the absorption frequency of these water molecules, allowing the timescales and degree of frequency shifting to be correlated with the timescales and degree of structural rearrangement of the interface. Understanding the rapid changes taking place at the interface, therefore, is equally as important as being able to describe the structure of the interface itself (Fig. 1).

Interfacial environments are the result of a balance between surfactant–surfactant and surfactant–water interactions, and therefore, characterizing the structure and dynamics of local water molecules is essential in describing the interface. Accordingly, there has been much carried out to describe the aqueous interfaces in microemulsions. One of the most foundational theories to arise from these studies is the core/shell hypothesis of confined water. The hypothesis was developed using Aerosol OT (AOT) reverse micelles, which were commonly used in early spectroscopic studies of microemulsion properties.^56,5756. J. K. Hensel, A. P. Carpenter, R. K. Ciszewski, B. K. Schabes, C. T. Kittredge, F. G. Moore, and G. L. Richmond, Proc. Natl. Acad. Sci. U. S. A. 114, 13351 (2017). https://doi.org/10.1073/pnas.170009911457. D. E. Moilanen, E. E. Fenn, D. Wong, and M. D. Fayer, J. Phys. Chem. B 113, 8560 (2009). https://doi.org/10.1021/jp902004r A correlation exists between the AOT reverse micelle size and the water dynamics of the internal nanopool: larger micelles showed faster dynamics, while small micelles exhibited slower water motions. According to the core/shell model, the dynamics of water motion and hydrogen bond switching are slowest near the surfactant interface and bulk-like within the core.^58–6058. I. R. Piletic, D. E. Moilanen, D. B. Spry, N. E. Levinger, and M. D. Fayer, J. Phys. Chem. A 110(15), 4985 (2006). https://doi.org/10.1021/jp061065c59. D. E. Moilanen, N. E. Levinger, D. B. Spry, and M. D. Fayer, J. Am. Chem. Soc. 129, 14311 (2007). https://doi.org/10.1021/ja073977d60. A. Baksi, P. Kr. Ghorai, and R. Biswas, J. Phys. Chem. B 124, 2848 (2020). https://doi.org/10.1021/acs.jpcb.9b11895 Figure 2 shows a representation of a reverse micelle and illustrates the core/shell model.

These “slow shell” fluctuations persist for several Angstroms into the bulk. The different timescales of core and shell rearrangements arise from the collective motions of water molecules: when a hydrogen bond is broken, another rapidly forms to maintain the dynamic equilibrium. This causes one hydrogen bond switching event to propagate a series of switches through the entire water network. Confined, interfacial water molecules have fewer hydrogen bonding partners, which slows the switching process. This effect has been studied rigorously using ultrafast IR spectroscopies [Fig. 2(b)], which can measure the timescales of the fast and slow rearrangements of interfacial and bulk water.^61–6361. V. P. Roy and K. J. Kubarych, J. Phys. Chem. B 121, 9621 (2017). https://doi.org/10.1021/acs.jpcb.7b0822562. A. K. Mora, P. K. Singh, S. A. Nadkarni, and S. Nath, J. Mol. Liq. 327, 114819 (2020). https://doi.org/10.1016/j.molliq.2020.11481963. E. E. Fenn and M. D. Fayer, J. Chem. Phys. 135, 074502 (2011). https://doi.org/10.1063/1.3625278 While the degree of slowing depends on the surfactant:water molar ratio, small reverse micelles containing exclusively interfacial water have dynamics ∼3× slower than the bulk measurements.⁵⁸58. I. R. Piletic, D. E. Moilanen, D. B. Spry, N. E. Levinger, and M. D. Fayer, J. Phys. Chem. A 110(15), 4985 (2006). https://doi.org/10.1021/jp061065c The thickness of the interfacial shell has been calculated to be ∼0.4 nm.⁵⁷57. D. E. Moilanen, E. E. Fenn, D. Wong, and M. D. Fayer, J. Phys. Chem. B 113, 8560 (2009). https://doi.org/10.1021/jp902004r

The core/shell hypothesis connects the solvation dynamics to the interfacial structure. Given a set of experimental conditions, the systems with slower water dynamics can be said to have more confined, interfacial water. Modifying the headgroup structure of octylphenoxypolyethoxyethanol (IGEPAL), a nonionic surfactant, modulates the dynamics of the water pool in the center of the micelle. Specifically, the IGEPAL headgroup contains a PEG chain. Lengthening this chain has been shown, using vanadium NMR, to slow the dynamics of the bulk water at the center of the micelle nanopool.⁶⁴64. M. A. Sedgwick, D. C. Crans, and N. E. Levinger, Langmuir 25, 5496 (2009). https://doi.org/10.1021/la8035067 The longer chain can interact with a greater portion of the water molecules, including those deeper in the micelle, thereby reducing the amount of water in the micelle core.

The core/shell model does not, however, inherently describe the interfacial H-bond structure; for this, techniques such as sum-frequency generation spectroscopy and MD simulations, which by their nature, provide structural information. While the water interface is a highly dynamic environment, water molecules adopt an ordered structure within ∼3 nm of the interface.^21,6521. P. C. Singh, K. Inoue, S. Nihonyanagi, S. Yamaguchi, and T. Tahara, Angew. Chem., Int. Ed. 55, 10621 (2016). https://doi.org/10.1002/anie.20160367665. D. E. Gragson, B. M. McCarty, and G. L. Richmond, J. Phys. Chem. 100, 14272 (1996). https://doi.org/10.1021/jp961034p Competing factors drive interfacial water ordering. Short interactions, such as van der Waal’s forces and hydrogen bonding, can influence shell water molecules. Electrostatic interactions, including the formation of electric double layers, drive long-range water ordering.^66,6766. A. Melcrová, S. Pokorna, S. Pullanchery, M. Kohagen, P. Jurkiewicz, M. Hof, P. Jungwirth, P. S. Cremer, and L. Cwiklik, Sci. Rep. 6, 38035 (2016). https://doi.org/10.1038/srep3803567. O. Carrier, E. H. G. Backus, N. Shahidzadeh, J. Franz, M. Wagner, Y. Nagata, M. Bonn, and D. Bonn, J. Phys. Chem. Lett. 7, 825 (2016). https://doi.org/10.1021/acs.jpclett.5b02646 Negatively charged interfaces order water molecules with their dipoles pointed toward the bulk while positively charged surfactants induce a new water dipole toward the interface, although local interactions with the headgroup are important contributions to determining the orientational ensembles of water molecules at the interface.^68–7068. Md. R. Khan, U. I. Premadasa, and K. L. A. Cimatu, J. Colloid Interface Sci. 568, 221 (2020). https://doi.org/10.1016/j.jcis.2020.02.05669. C. Dutta, M. Mammetkuliyev, and A. V. Benderskii, J. Chem. Phys. 151, 034703 (2019). https://doi.org/10.1063/1.506659770. M. Schleeger, Y. Nagata, and M. Bonn, J. Phys. Chem. Lett. 5, 3737 (2014). https://doi.org/10.1021/jz5019724 Nonionic surfactants also strongly orient water dipoles, but the direction is dependent, in part, on the nature of the headgroup—HB-donating headgroups induce cationic-like water ordering, while HB-accepting headgroups orient the net dipole similarly to anionic interfaces.^71–7371. E. Tyrode, C. M. Johnson, A. Kumpulainen, M. W. Rutland, and P. M. Claesson, J. Am. Chem. Soc. 127, 16848 (2005). https://doi.org/10.1021/ja053289z72. D. Hu, A. Mafi, and K. C. Chou, J. Phys. Chem. B 120, 2257 (2016). https://doi.org/10.1021/acs.jpcb.5b1171773. A. Mafi, D. Hu, and K. C. Chou, Surf. Sci. 648, 366 (2016). https://doi.org/10.1016/j.susc.2015.10.010 These results, shown in Fig. 3, are in agreement with similar work done on biological surfactants.²¹21. P. C. Singh, K. Inoue, S. Nihonyanagi, S. Yamaguchi, and T. Tahara, Angew. Chem., Int. Ed. 55, 10621 (2016). https://doi.org/10.1002/anie.201603676 Further underscoring the importance of electrostatic forces in water ordering, the range of the interfacial influence can be tuned by the concentration of the surfactant at the interface due to the formation of electrostatic double layers.

Nguyen et al. have shown that increasing the amount of cationic hexadecyltrimethylammonium bromide (CTAB) present in the system correlates with a decrease in the amount of signal generated by ordered, interfacial water, indicating a greater amount of screening.⁷⁴74. K. T. Nguyen, A. V. Nguyen, and G. M. Evans, J. Phys. Chem. C 119, 15477 (2015). https://doi.org/10.1021/acs.jpcc.5b04416 Interestingly, the region of ordered water is much thicker than the ∼0.4-nm-thick shell described in the core–shell model, highlighting the importance of cooperative dynamic and structural studies of the interface. The dynamic measurements might lead one to underestimate the structural influence of the interface, while the thick layer of ordered water could lead to an overestimation of the effect of the interface on dynamics. Interestingly, this water organization is disrupted when ions are added, but only if the surfactants are charged. The charged ions interact more strongly with the charged interface, resulting in greater disruptions to the interfacial double layer. Nonionic surfactants, lacking an inherently charged interface, interact more weakly with ions, making non-electrostatic interactions more important for determining how water is ordered by the interface.^75,7675. B. J. Park, J. P. Pantina, E. M. Furst, M. Oettel, S. Reynaert, and J. Vermant, Langmuir 24, 1686 (2008). https://doi.org/10.1021/la700880476. M. J. Qazi, S. J. Schlegel, E. H. G. Backus, M. Bonn, D. Bonn, and N. Shahidzadeh, Langmuir 36, 7956 (2020). https://doi.org/10.1021/acs.langmuir.0c01211 Within CTAB interfaces, the addition of NaCl to the aqueous phase dramatically disrupts water ordering, resulting in a tenfold drop in signal as the cation flips the dipoles of interfacial water molecules toward the bulk.⁷⁶76. M. J. Qazi, S. J. Schlegel, E. H. G. Backus, M. Bonn, D. Bonn, and N. Shahidzadeh, Langmuir 36, 7956 (2020). https://doi.org/10.1021/acs.langmuir.0c01211 The hydration shells of nonionic surfactants are largely unaffected by monovalent and divalent salts—anions slightly perturb the net water dipole at the surfactant:water interface. The changes in spectral intensity due to these perturbations are, however, on the order of approximately 20%, roughly one-fifth of the effect described at charged interfaces.⁷⁷77. M. Hishida, Y. Kaneko, M. Okuno, Y. Yamamura, T. Ishibashi, and K. Saito, J. Chem. Phys. 142, 171101 (2015). https://doi.org/10.1063/1.4919664

Water dynamics and ordering are important for understanding the aqueous region of the interface and several studies, described above, have probed the surfactant behavior from the perspective of water dynamics. Similarly, the surfactants themselves can be probed directly using the same spectroscopic methods. For example, interest in understanding the unique behaviors of individual ions has driven studies looking to describe the molecular origins of cosmotropic and chaotropic behavior, including systems where the Hoffmeister Series breaks down.^77,7877. M. Hishida, Y. Kaneko, M. Okuno, Y. Yamamura, T. Ishibashi, and K. Saito, J. Chem. Phys. 142, 171101 (2015). https://doi.org/10.1063/1.491966478. H. I. Okur, J. Hladílková, K. B. Rembert, Y. Cho, J. Heyda, J. Dzubiella, P. S. Cremer, and P. Jungwirth, J. Phys. Chem. B 121, 1997 (2017). https://doi.org/10.1021/acs.jpcb.6b10797 The ion-dependent structural and dynamic perturbations to surfactants themselves have been studied as part of these efforts. In the case of nonionic surfactants, described above, interfacial water orientations were unperturbed by the addition of salts.⁷⁶76. M. J. Qazi, S. J. Schlegel, E. H. G. Backus, M. Bonn, D. Bonn, and N. Shahidzadeh, Langmuir 36, 7956 (2020). https://doi.org/10.1021/acs.langmuir.0c01211 However, while the net water dipole is unaffected, recent MD simulations of nonionic sorbitan surfactants revealed that the orientations of the surfactant headgroups were altered by the presence of salts, as shown in Fig. 4.⁷⁹79. C. P. Baryiames, E. Ma, and C. R. Baiz, J. Phys. Chem. B 124, 11895 (2020). https://doi.org/10.1021/acs.jpcb.0c09086 NaCl showed a highly localized disordering effect, limited to disrupting the surfactants within 2 nm of the cation. In contrast, CaCl₂ disrupted headgroup packing throughout the interface. Water ordering, as measured by hydrogen bond populations and thermodynamics, is unperturbed by the presence of ions. The studies are supported by SFG experiments of structurally-similar Tween surfactants, which show that ions have little effect on the interfacial water structure.⁷⁶76. M. J. Qazi, S. J. Schlegel, E. H. G. Backus, M. Bonn, D. Bonn, and N. Shahidzadeh, Langmuir 36, 7956 (2020). https://doi.org/10.1021/acs.langmuir.0c01211 Dynamic measurements, by contrast, show that dynamics in the surfactant layer slow by up to ∼300% when ions are added to the aqueous phase.⁷⁹79. C. P. Baryiames, E. Ma, and C. R. Baiz, J. Phys. Chem. B 124, 11895 (2020). https://doi.org/10.1021/acs.jpcb.0c09086 Only by including surfactant dynamics, the interface effect of ions on nonionic interfaces can be described. This observation again reinforces the need for cooperative structural and dynamic studies of the interface, as well as the synergistic applications of experimental measurements together with MD simulations.

Surfactant packing can also be disrupted by introducing heterogeneity in the form of structural heterogeneity. The compositions of commercially-obtained samples of sorbitan monostearate (Span-60) vary dramatically by manufacturer and batch, as evidenced from recent chromatography and mass spectrometry analysis of two commercial sources of the same surfactant.⁷7. C. P. Baryiames, M. Teel, and C. R. Baiz, Langmuir 35, 11463 (2019). https://doi.org/10.1021/acs.langmuir.9b01693 In addition to sorbitan monostearate, samples contained sorbitan di-, tri-, and tetrastearates. Comparing purified Span-60 to a heterogeneous sample, surfactant–surfactant interactions decrease by 27%. Figure 5(a) shows that as additional fatty acid tails are added to the sorbitan headgroup, the number of interfacial hydroxyl groups decreases, preventing the formation of surfactant–surfactant hydrogen bonds. The loss of these attractive forces leads to reduced packing efficiency, as determined by area-per-surfactant measurements obtained from simulations. Larger intermolecular spaces lead to increased water penetration into the surfactant layer, resulting in a 10% increase in interfacial hydration, shown in Fig. 5(b). The alkyl tails also become more disordered. Interestingly, these effects are attributable to heterogeneity in composition alone, which reduces the packing efficiency between surfactants. A homogeneous mixture of mono and tristereate (three acyl tails), in contrast, showed comparable hydrogen bond populations, areas-per-surfactant, and interfacial dynamics, indicating that the average degree of headgroup substitution was not the origin of the observed differences.

Binary amphiphile mixtures can modify the morphology of the microemulsion formed by a mixture of oil, water, and detergent. Mixtures of Aerosol OT (AOT) and cholesterol, for example, have been shown to form two separate water-in-oil microemulsion phases: one set of reverse micelles containing only AOT and another containing mostly cholesterol and a small amount of AOT.⁸⁰80. M. A. Sedgwick, A. M. Trujillo, N. Hendricks, N. E. Levinger, and D. C. Crans, Langmuir 27, 948 (2011). https://doi.org/10.1021/la103875w The two coexistent phases were characterized by dynamic light scattering, which showed that the two phases formed reverse micelles of different sizes. The AOT-only micelles have a diameter of ∼4–8 nm, while the mixed micelles are much smaller, which are only ∼1–3 nm in diameter.

While the importance of the surfactant–water interface is well-established, the role the surfactant–oil interface plays in microemulsion properties is less understood. The above studies and others have suggested that hydrophobic regions of microemulsions and biomolecules may influence interfacial water, including the work that shows that altering a protein’s buried residues can affect water dynamics in its enzyme binding pocket. This effect has been measured directly, by tuning the viscosity of the oleic phase by mixing n-octane and squalene, sampling viscosities from 0.3 to 10.5 cP.⁸¹81. C. P. Baryiames and C. R. Baiz, J. Am. Chem. Soc. 142, 8063 (2020). https://doi.org/10.1021/jacs.0c00817 Figure 6(a) shows that interfacial dynamics slow linearly with the increase in viscosity, resulting in a twofold slowdown in water motions when squalene (1.5 ps) is used as the oil phase instead of octane (0.9 ps). Changing the viscosity of the oleic phase changes the dynamics of water ∼2.6 nm away by “transmitting” low-frequency molecular motions through the surfactant tail region, but the structural origins of this effect could not be determined from the experiment alone. MD simulations, represented in Fig. 6(b), modeled the reverse micelle with octane and squalene oil phases. Under the hypothesis that intercalations between the hydrocarbon and the alkyl surfactant tails were responsible for transmitting the slow oil dynamics to the water, a third trajectory was run with a soft (50 kJ mol⁻¹ nm⁻¹) positional constraint applied to the three carbon atoms at the end of the alkyl tail. The dynamics measured in the simulations with constrained surfactant tails matched those of the squalene containing simulations, indicating that slowing the motion of the tails slows the interfacial water.

Recent studies have shown that the surfactant tail can change the aqueous interface. Smolentsev et al. showed, for example, that Span-80 surfactants lie with their headgroups parallel to the surfactant–water interface (Fig. 7).⁸²82. N. Smolentsev, W. J. Smit, H. J. Bakker, and S. Roke, Nat. Commun. 8, 15548 (2017). https://doi.org/10.1038/ncomms15548 This results in approximately half of the hydrophobic tail being exposed to an aqueous environment. This is not the case with Span-60 surfactants, which are identical to Span-80 in structure except the acyl chains, which are fully saturated. The only difference between these structures is the kink in the Span-80 tailgroup introduced by the cis double bond at position 9. This difference in interfacial conformation results in a larger area-per-surfactant: 46 and 35 Ų/molecule for Span-80 and Span-60, respectively.

Chemical reactions in small (nano-length scale) have gained attention for their ability to precisely control reaction rates and product composition and morphology through controlling local reactant concentrations, the vesicle size, and temperature uniformity in exothermic reactions.^83,8483. J. Eastoe, M. J. Hollamby, and L. Hudson, Adv. Colloid Interface Sci. 128-130, 5 (2006). https://doi.org/10.1016/j.cis.2006.11.00984. V. Uskoković and M. Drofenik, Adv. Colloid Interface Sci. 133, 23 (2007). https://doi.org/10.1016/j.cis.2007.02.002 This is accomplished by manipulation of the water loading ratio and surfactant composition. However, whether these beneficial effects arise from interactions with the interface or as a result of nanoconfined reactants is not understood. The nature of the container when performing bulk reactions is rarely considered, however, when a reaction is occurring in a confined environment, the surface area to volume ratio is greatly increased, and thus, one must consider the effects of the interface. For example, small ∼3–5 nm reverse micelles support a large fraction of interfacial water, which is characterized by slower dynamics compared to bulk water in the core, as described above. Understanding how interfacial dynamics interact with encapsulated reactants and products is the first step in understanding the effects the reverse micellular environment has on chemical reactions.

Uncovering the chemistry that arises from reactions in reverse micelles requires disentangling the local environments, dynamics, and interactions between the reactants and products with the interface. Polymers offer a tractable model system to study the effect of confinement on inverse emulsion polymerization reactions. These radical-polymerization reactions are diffusion-limited: the rate at which reactants move through the water pool is the deciding factor in the overall reaction kinetics, making them ideal to investigate the effects of fast and slow environments on local diffusion rates, and how the monomer reactants and growing polymer chain interact directly with the surfactant interface.⁸⁵85. I. P. Kim and V. A. Benderskii, Russ. J. Phys. Chem. A 88, 1954 (2014). https://doi.org/10.1134/s0036024414110090

Here, we present some preliminary results on acrylamide and polyacrylamide in Span-60 reverse micelles, which were used as the model system for studying the effects due to encapsulation of both reagents and polymers [see Fig. 8(a)]. Experimental details are provided in the supplementary material. This polymer was selected not only for its wide array of applications but also for its compatibility with aqueous environments and ability to both accept and donate hydrogen bonds. A first principles understanding of the effects of monomers and polymers on the surfactant interface is the initial step in understanding how the reverse micelle dictates reaction rates and product ratios. Here, we present 2D-IR measurements of the ester carbonyl modes in ∼120 nm sorbitan monostearate reverse micelles, with pure water, 5 wt. % acrylamide monomers, and polyacrylamide. Figure 8 displays the centerline slopes extracted from the waiting-time 2D IR series of reverse micelles with the encapsulated 5 wt. % acrylamide monomer and polymer. The centerline slope of 2D IR spectra shows monoexponential behavior. This can be interpreted as the frequency-fluctuation correlation function of the carbonyl modes. The relaxation rates are in good agreement with the previous results. The smaller time constants translate to faster dynamics at the interface. Figure 8(b) shows that the monomer introduces a slight slowdown compared to pure water. The polymer, however, slows down the interface nearly threefold compared to pure water or the monomer. These results strongly suggest that the polymer localizes to the interface and interacts more strongly with the surfactant compared to the monomer. Previously, anisotropy measurements reported a slowdown of bulk water dynamics with the addition of 5 wt. % polyacrylamide.⁸⁶86. C. Yan, P. L. Kramer, R. Yuan, and M. D. Fayer, J. Am. Chem. Soc. 140, 9466 (2018). https://doi.org/10.1021/jacs.8b03547 However, the water orientational relaxation rates slowed from 1.7 ± 0.2 ps in pure water to 2.7 ± 0.4 ps with the addition of a polymer when compared to our interfacial slow down from 1.3 ± 0.1 ps in pure water to 1.5 ± 0.1 ps with the addition of a monomer, and 3.2 ± 0.4 ps with a polymer as demonstrated in this study. This comparison strongly suggests that polymers interact with the interface causing a slowdown that is greater than the bulk, therefore, demonstrating that descriptions of chemical reactions in reverse micelles must include interfacial effects.

Starting from Agnes Pockels’ early studies of detergent interfaces in the late 1800s, surfactant science has been largely concerned with bulk, thermodynamic properties.^87,8887. A. Pockels, Ann. Phys. 313, 854 (1902). https://doi.org/10.1002/andp.1902313081188. A. Pockels, Nature 48, 152 (1893). https://doi.org/10.1038/048152a0 Measuring interfacial tensions of different phases with varying surfactant compositions remains a valuable study, as does exploring microemulsion phase diagrams and studying the effects of nonconfinement on bulk reaction processes. Recent work on the specific, molecular interactions occurring in microemulsions is building on this foundation by connecting those interactions, measured with ultrafast spectroscopies and cutting-edge simulations, to the bulk behaviors that have been so robustly characterized. Uncovering these connections opens new avenues for more advanced surfactant studies and rational design of compositions for a wide range of applications. Understanding amphiphile structure–function relationships is key to designing surfactants for future high-performance applications in industrial and healthcare applications. For example, bioinspired surfactants are the default route to deliver the mRNA payload of cutting-edge pharmaceuticals, including COVID-19 vaccines, by promoting endocytosis.^89–9289. L. A. Jackson, E. J. Anderson, N. G. Rouphael, P. C. Roberts, M. Makhene, R. N. Coler, M. P. McCullough, J. D. Chappell, M. R. Denison, L. J. Stevens, A. J. Pruijssers, A. McDermott, B. Flach, N. A. Doria-Rose, K. S. Corbett, K. M. Morabito, S. O’Dell, S. D. Schmidt, P. A. Swanson, M. Padilla, J. R. Mascola, K. M. Neuzil, H. Bennett, W. Sun, E. Peters, M. Makowski, J. Albert, K. Cross, W. Buchanan, R. Pikaart-Tautges, J. E. Ledgerwood, B. S. Graham, and J. H. Beigel, N. Engl. J. Med. 383, 1920 (2020). https://doi.org/10.1056/nejmoa202248390. P. Midoux and C. Pichon, Expert Rev. Vaccines 14, 221 (2015). https://doi.org/10.1586/14760584.2015.98610491. A. M. Reichmuth, M. A. Oberli, A. Jaklenec, R. Langer, and D. Blankschtein, Ther. Delivery 7, 319 (2016). https://doi.org/10.4155/tde-2016-000692. H. Zhang, X. You, X. Wang, L. Cui, Z. Wang, F. Xu, M. Li, Z. Yang, J. Liu, P. Huang, Y. Kang, J. Wu, and X. Xia, Proc. Natl. Acad. Sci. U. S. A. 118, e2005191118 (2021). https://doi.org/10.1073/pnas.2005191118 However, the design of these surfactants remains largely a trial-and-error endeavor.

The position of surfactant physical chemistry as a growing field opens up many opportunities to make important discoveries. However, the nature of an emergent field is that these opportunities are themselves challenges—while the ultrafast communities, for example, have made inroads on surfactant problems, there are factors these methods are unable to assess and necessitate additional inputs from other fields. Spectroscopy, as we have shown, can report on the interfacial environment and dynamics but cannot describe the molecular composition of a mixture of detergents in the manner of a lipidomics study. The success of lipid bilayer biophysics can point toward useful studies to complement our current understandings of surfactant behavior. For example, the surfactant tail region is a difficult target due to overlap between the CH_x signals from the surfactant and oil molecules, which requires the use of complementary techniques, such as MD, to robustly interpret the data.⁹³93. L. S. Vermeer, B. L. de Groot, V. Réat, A. Milon, and J. Czaplicki, Eur. Biophys. J. 36, 919 (2007). https://doi.org/10.1007/s00249-007-0192-9 One common way of verifying the validity of MD simulations of lipids is to compare the calculated order parameters of the tail groups to those determined experimentally from NMR measurements.⁹⁴94. N. O. Petersen and S. I. Chan, Biochemistry 16, 2657 (1977). https://doi.org/10.1021/bi00631a012 Unfortunately, to our knowledge, no such study has been performed on surfactants.

As interest in “saponophysics” grows, we are optimistic that the outstanding questions of surfactant behavior will be answered. Their relevance to industrial, synthetic, biological, and pharmacological problems is too great for this not to be the case. Until the structure–function relationships of individual and mixed surfactants are understood, applied microemulsion science will be limited by the necessity of high-throughput, empirical guess-and-check methodologies. However, the tools needed to begin addressing this problem exist and have been proven valuable in studying natural and synthetic detergents. We predict that the field will continue to grow, address the challenges we present in this Perspective, and enable further development of advanced surfactant applications.

The data that support the findings of this study are available from the corresponding author upon reasonable request.