Polymer-surfactant stabilised foams are of growing interest in a wide range of industries-paper, foodstuffs, home, personal care and pharmaceutical-either because the foam is an end-product or encountered during the manufacturing process. The ability to control the interactions between the polymers and the surfactants provides new approaches to control the foaming properties of these systems, and eventually, optimizing the performance of the formulation. Foams are thermodynamically unstable, and therefore surface active species like proteins, particles, polymers, surfactants and their mixtures are commonly used to stabilise the foam by slowing the drainage, coalescence and coarsening of the foam structures [1
]. How this stability is achieved is still not fully understood.
Mixtures of polymers and surfactants are ubiquitous and their bulk and equilibrium interfacial behaviours have been investigated at length. Generally, the systems may be differentiated by the strength of the interactions between the surfactant and the polymer chains, these being hydrophobic and/or electrostatic in nature, depending on the chemical composition of the system. The key surfactant structure in these complexes could be of monomeric or micellar nature depending on several factors, but notably the surfactant/polymer concentration, the presence of any additives and the conditions of the solution being studied [2
]. Multi-layer structures are often observed, especially in the context of oppositely charged polymers and surfactants, but less so with non-ionic surface active polymers and surfactants [3
Few studies have focused on the relationship between adsorbed layers and the foam stability (time taken for the foam to collapse) and/or “foaminess” (measured height of a column of foam generated under controlled conditions) from a detailed structural analysis of the interfacial layers. One notable study is that of Petkova et al., who investigated foams stabilised by blends of non-surface active
polymers (poly(vinylamine), poly(N
-vinyl-formamide)) and small molecule surfactants (SMS) (sodium dodecylsulfate (SDS), C12
TAB (dodecyltrimethylammonium bromide) and Brij 35 (C12
)) that show strong and weak solution interactions. In these studies, less foamabililty, but higher foam stability was recorded from polymer-surfactant mixtures showing strong synergistic interactions [12
Recent neutron reflectivity (NR) studies have concluded that the “equilibrium” interfacial structures of surface active species such as surfactants and polymers are rather more complex than historically modelled, and often, experimental findings are difficult to deconvolute. The origin of this is thought to be the formation of multilayer structures [13
], though not all experiments provided unequivocal evidence for this (such as Bragg peaks). However, some studies especially on oppositely charged polymer/surfactant complexes do exhibit these features e.g., Campbell et al. [3
] and others [8
]. Further, for oppositely charged systems, characteristics that impact the kinetics of interaction such as the order of mixing, are shown to be dominating factors in defining the structures that ultimately form [10
]. Therefore, it is hypothesized that there is an as-yet, an ill-defined relationship between the surface and bulk structures in these slowly equilibrating systems.
We have previously deployed small-angle neutron scattering (SANS) to study foams stabilised by single component solutions of non-ionic polymers of the Pluronic family [21
] and small molecule surfactants [22
], since neutron techniques have a proven ability to probe the adsorption of molecules at interfaces. Such experimental approaches have contributed significantly to the understanding of the structure activity relationships of interfacial bound species. Of key interest in that work were observations of (Bragg) peaks in the scattering data suggesting the presence of polymer and/or surfactant multilayer structures [23
] at the air-water interface present in wet foams. It is therefore hypothesized that the multilayer structure is induced by the non-equilibrium nature of the foam, and as such these observations resonate with “equilibrium” reflectivity studies on the more slowly equilibrating oppositely charged polymer/surfactant systems. Our data were successfully described by a small number (M
) of discrete layers of thickness (L
) and spacing (D
] though it must be said, that multilayer structures at dynamic interfaces is not a universally accepted view [28
] and further research is warranted.
Herein, we extend our previous SANS studies to include investigation of the interfacial structure of foams stabilised by two surface active species (and thus, in contrast to Petkova [12
]); non-ionic triblock copolymers Pluronic and SMS. To the best of our knowledge, this is the first time that foams stabilised by mixtures of surface active polymers and surfactants has been investigated by SANS, and should complement the reflectively studies on oppositely charged polymer/surfactant systems. Significant changes in the foam stability measurements were also observed as the strength of interactions varied from weak to strong, evident by more stable foams for the systems showing strong or “synergistic” solution interactions (Supplementary Information
Wet (continuously generated) foams consisting of bubbles ranging in size from a few to tens of millimetres in diameter, with film thicknesses of microns, were prepared using the following SMS; anionic SDS, cationic C12
TAB, and non-ionic polymeric surfactants Pluronic P123 (EO20
) and L62 (EO6
). Further, mixtures of Pluronic P123 and SMS at concentrations significantly below the respective critical micelle concentration (CMC) or mixed CMC as measured by surface tensiometry (Figures S5 and S6
) to avoid the presence of any solution-like micelles have also been explored.
The SANS data will be presented in the following manner; (a) foams stabilised by a weakly interacting system; Pluronic P123 and C12
TAB system [33
]; (b) foams stabilised by a strongly interacting system; Pluronic P123 and SDS and finally; (c) to show temperature induced micellisation, foams stabilised by Pluronic L62 as a function of temperature. It is hoped that this work would highlight how the structural variations of the commonly used temperature sensitive surface active polymers (Pluronic) and the interactions between these polymers and SMS in bulk affect the surfactant structures at the foam air-water interface.
3. Results and Discussion
The measured SANS data from these systems, the insets in Figure 1
(and also in Figure 3), show a pronounced Q−4
dependence, as expected from the intense Porod surface scattering from the large smooth surface of the bubbles. A visualisation strategy in which the data is plotted in a I
representation was developed, (n
= 4 ± 0.05), Figure 1
and Figure 3, to reveal the several subtle inflexion(s) in these data observed across the Q
range. The Q
position of these inflexions was found to be sensitive to the surfactant and/or polymer structure, and to the level of the interactions between the components. The data were fitted to a multilayer model, as shown in the materials and methods section and the fit is also presented in an I
SANS from pure P123 at 0.025 wt% (half of the measured CMC), Figure 1
, show that these inflexions corresponds to d-spacings
) of ≈ 370 Å (Q
≈ 0.017 Å−1
) and ≈ 180 Å (Q
≈ 0.0347 Å−1
), Table S2
. These d-spacing
values were consistent in both data sets acquired from the two different diffractometers used in this study (SANS2d
The peak at mid Q (≈ 0.03 Å−1) is usually the most discernible. The presence of what could be interpreted as a higher order peak at the lower Q value can be related to the regular arrangement of the surfactant multilayers, albeit within fractured or heterogeneous lamellar structures. For a perfectly lamellar structure, one would expect to see regular reflections (n = 1, n = 2, n = 3) however, the subtle differences observed in the peak positions implies that the structure is not perfectly lamellar (in the direction normal to the interface).
Both P123 datasets (SANS2d
) have been successfully fitted to the multilayer model, Table 1
. The fit revealed a Pluronic layer thickness of 140 Å and a D
of 180 Å, however for the SANS2d
data set, the fit was able to capture both low and mid Q
peaks with a D
value of 390 Å, in good agreement with our previous work [21
]. Clearly, the model captures the gross features in the data very well, namely the peak position (especially that of the main peak), but it also captures subtleties in the data, such as the weaker shoulders evident in the Porod plots.
Upon introducing C12
TAB to the system, no significant change in the peak position at mid Q
is observed. In the foam stability studies, the stability of the foam formed from these systems is (only) slightly reduced (supplemental information
). The non-changing position of this inflexion from the foam stabilised by the mixture of P123 (0.025 wt%) and C12
TAB (0.1 mM), when compared with foams from both the pure systems of P123 (0.0347Å−1
) and C12
TAB (0.0370 Å−1
), indicates the conclusion proposed by Petkova et al. [13
] regarding the correlation between the weak interactions observed in bulk and dynamic interfacial structures is more general i.e., it pertains to other non-surface active polymer/surfactant mixtures as well as surface active polymer/surfactant mixtures. The data fitting results are also in agreement with these conclusions, Table 1
One can postulate that the thinner C12
TAB (40 Å; 1.5 + (1.26 × Nc
) × 2) layers are “coexisting” between the segregated thicker P123 layers (140 Å), without a significant change in the dimension and/or the separation of the P123 layers. Such structures would seem logical assuming distinct populations of the two species in bulk solution [33
] associated with weak interactions between the two components.
This hypothesis was further explored by a contrast variation SANS approach with deuterated C12
TAB, where the foam scattering is dominated by the P123 (supplemental information
). This experiment, Figure S10
, shows that for the P123 + h-C12
TAB system, two peaks are observed at d-spacing
≈ 380 Å and 195 Å respectively, but for the P123 + d-C12
TAB, there is a shift in both peak positions towards lower Q
≈ 400 Å and 200 Å). The observation of these larger spacings-with dimensions much more akin to the pure P123 foam—is consistent with the fact that the C12
TAB is rendered invisible.
Moving to the strongly interacting system, P123 (EO20
) and SDS, several published works have shown the formation of mixed micelles at concentrations above the mixed CMC. These micelles were found to be smaller in size (≈28 Å) when compared with pure P123 micelles (≈70 Å). [34
] Further, the interfacial structure of the triblock copolymer EO23
in solution was also studied by neutron reflection at different concentrations, where a total layer thickness of 72 Å was noted. Upon addition of SDS, the layer thickness was found to be between 46 and 49 Å [13
Foams stabilised by the mixture of P123 and SDS showed the most significant change in foam stability, Figure 2
, and in the SANS data, Figure 3
. For P123 at concentrations equal to half of its measured CMC (0.025 wt%), the decay in the stability profile is rapid with a half-life of ≈4250 s. Upon the addition of a small concentration of SDS, 0.1 mM, the strong synergy between both components leads to a significantly enhanced foam stability, with a half-life that is now almost double that of the P123 only (≈8000 s).
Recall the SANS data from foams stabilised by 4 mM SDS (only) showed one peak at mid Q consistent with a surfactant layer thickness of 35 Å. On addition of SDS to the P123 solution, even at concentrations as low as 0.1 mM, there is a significant impact on the position of the P123 peaks. For example, the peak at low Q, now corresponds to a d-spacing value of ≈ 380 Å (from 370 Å), whereas the peak at mid Q corresponds to a d-spacing value of ≈ 205 Å (from 180 Å). This was also evident in the data modelling where the layer thickness was found to be 120 Å each (M ≈ 5) with a D value of 410 Å.
The increase in both values of the spacing for the P123/SDS case suggests the formation of a new polymer-surfactant structure at the air-water interface, in which the SDS is absorbed or laterally interdigitated within the P123 layers, forming thinner, mixed surfactant layers, resulting in a larger spacing between these layers. This is consistent with the smaller micelles seen in solution, albeit at higher concentrations.
The same approach for data presentation has been followed as before and Porod plots have been used to highlight the peaks at low and mid Q
, Figure 3
. The observations from the SANS contrast variation (d-SDS and P123) measurements, Figure S9
, from these systems also reiterate the hypothesis that the peak positions of the d-SDS/P123 system show no significant change from the h-SDS/P123 system, however, both peak positions differ from the pure P123 system.
To further illustrate that these Bragg features arise from interfacial species, foams stabilised by a different molecular weight Pluronic, L62, were also investigated by SANS as a function of temperature (Pluronic P123 has a low critical micelle temperature, CMT; 16 °C), since on passing through the CMT the aqueous phase comprising the interstitial volumes will now contain micelles.
shows the scattering behaviour from foams stabilised by 0.05 wt % L62 (below its 25 °C CMC) at different temperature ranges (below and above the CMT ≈ 28 °C). Below the CMT (and CMC), a similar SANS pattern to the P123 is observed. The data fitting to the multilayer model has revealed a surfactant layer thickness (L
) of 90 Å, spacing between layers (D
) of 195 Å and that 6 layers (M
) of the polymer are stacked at the interface. The change in the fit parameters are consistent with the change in the molecular weight and EO:PO ratio as we move from the bulkier P123 to the smaller L62.
As the temperature approaches the CMT, 25 °C, a broadening of the peak at mid Q
≈ 0.035 Å−1
can be observed. The temperature increase, namely between 37 °C and 45 °C, smears the mid Q
peak rendering the fine features related to the air-water layer structures hard to define. However, these broad peaks indicate that micellar structures have been induced in the system and are present in the aqueous solution comprising the foam cell walls and further, that they dominate the scattering intensity. As an aside, the temperature induced micellisation also had a significant effect on the L62 foamability, as the temperature increased, the maximum foam height reached increased, Figure S8