Zeyu
Zhao
ad,
Tengda
Wang
bc,
Jiling
Yue
a,
Yaxun
Fan
*abc and
Yilin
Wang
*abcd
aBeijing National Laboratory for Molecular Sciences and Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China. E-mail: yxfan@iccas.ac.cn; yilinwang@iccas.ac.cn
bSuzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, Jiangsu 215123, P. R. China
cUniversity of Science and Technology of China, Hefei, Anhui 230026, P. R. China
dUniversity of Chinese Academy of Sciences, Beijing 100049, P. R. China
First published on 28th September 2023
Excessive usage of surfactants in daily life and industry and their undesirable high foamability have caused serious environmental pollution and economic loss. Improving cleaning efficiency and reducing foam stability concurrently is a delicate strategy but a challenging task. Herein, we mixed the most widely used surfactant sodium dodecyl sulfate (SDS) with cyclic amines (CnN, n = 6, 8, 12), by which the self-assembly ability of SDS at the air/water interface and in bulk is significantly enhanced, while spherical micelles, vesicles and wormlike micelles are formed at appropriate total surfactant concentration (CT) and molar fraction of SDS (XSDS). Especially around XSDS = 0.50 and above critical micellar concentration (CMC), the stronger self-assembly ability leads to a higher contact angle of machine oil on stainless-steel plates and lower oil–water interfacial tension in CnN/SDS solution, thus the oil-fouling removal efficiency of CnN/SDS solutions is remarkably improved. Meanwhile, the foamability and foam stability dramatically decline at smaller XSDS and slightly above CMC, attributed to the rapid molecular migration from liquid film of foams to the bulk between the films when the limited surfactant molecules in the films prefer to aggregate in bulk. As a result, C8N/SDS exhibits the best oil cleaning and lowest foaming simultaneously at low XSDS and just above the CMC. This study opens an efficient avenue to eliminate the contradiction between cleaning ability and foamability, thereby obtaining a high-efficiency and low-foam detergent.
The generation of liquid foams is at the heart of numerous natural, technical, or scientific processes related to surfactants. In many cases, they have very useful properties for practical applications, such as cleaning, foam flotation, food processing and firefighting. However, foams are not always desirable. The unwanted foams can obstruct gas transport and render the process of interest ineffective with significant cost implications. Kister et al.50,51 made a survey on causes of malfunctions in industry, and they found that the existence of foam in surfactant solutions in distillation processes was a major reason for 900 investigated cases of column malfunctions. The accumulation of impurities that stabilize the foam on the sea also produces foams and destroys the normal ecological environment.52–54 Coincidently, those aforementioned physicochemical properties facilitating the cleaning efficiency, such as low surface tension, strong aggregation ability and fast adsorption kinetics, often result in the strong foamability and foam stability in surfactant solution.55–58 Therefore, the development of surfactant systems with high cleaning efficiency but adjustable foaming ability in different conditions encounters great technological hurdles although it is urgently desired in multiple fields.
The length and topological structure of alkyl chains play important roles in physicochemical properties of surfactants at surface/interface and in solution. Cyclic amines with bulky alkyl chains may form a larger hydrophobic domain in bulk but create looser arrangement at the air/water interface.59,60 Their special structure makes it greatly possible for cyclic amine-containing systems to achieve high cleaning efficiency and low foam concurrently. Bearing in mind the structural advantages, we designed mixed systems composed of cyclic amines with various sizes of the cycloalkane ring (CnN, n = 6, 8, 12) and widely used sodium dodecyl sulfate (SDS) at pH = 7.0. We found that the mixed solutions (CnN/SDS) show lower CMC and surface tension and form various assemblies at different molar fractions of SDS (XSDS) and total concentrations (CT), driven by the electrostatic and hydrophobic interactions between SDS and cyclic amines. With the balance of the self-assembly of surfactants at the liquid film of foam and in bulk between the films, the oil cleaning efficiency of C8N/SDS is much higher than that of the other two systems, whereas its foamability and foam stability are remarkably weakened. This work offers a feasible method for effectively removing oil with low foamability and foam stability at low surfactant concentrations, thereby potentially reducing surfactant usage and promoting sustainable development of environments.
Oil cleaning efficiency% = [(M1 − M2)/(M1 − M0)] × 100 |
Obviously, the surface tension curves exhibit broadly similar profiles for all the CnN/SDS mixtures regardless of the sizes of the cycloalkane ring and the XSDS values. The CMC value of SDS without CnN derived from the breakpoint is 8.0 mM and the surface tension at CMC (γCMC) is 38 mN m−1, which are consistent with previously reported values. For CnN solution alone, the surface tension curves of C6N and C8N display no inflection point in the concentration range studied (<50 mM) and just show a slight decline to ∼55 mN m−1, whereas the CMC and γCMC values of C12N are 28 mM and 47 mN m−1, indicating that the three cyclic amines possess a very weak surface activity. As expected, the molecular arrangement at the air/water interface of cyclic amines with a bulky alkyl chain is much looser than that of the alkyl amine with the same carbon number, e.g., CMC and γCMC values of N-dodecylamine are 2.6 mM and 30 mN m−1 (Fig. S1†).
Upon mixing the two components, the CMC and γCMC values decrease significantly to lower than those of individual components, attributed to the strong electrostatic attraction between the cationic headgroup of protonated CnN and the anionic sulfate of SDS and the hydrophobic interaction of the alkyl ring of amines with the alkyl chain of SDS. Even at XSDS = 0.01, the CMC values have already reduced to 27.8, 11.6 and 1.1 mM for C6N/SDS, C8N/SDS and C12N/SDS mixtures, respectively, whereas the γCMC values declined to ∼30 mN m−1, nearly 8.0 mN m−1 lower than that of the pure SDS solution. When XSDS approaches 0.5 from 0 or 1.0, the CMC value is steadily decreased to the lowest one, i.e., 4.0, 2.0 and 0.4 mM for C6N/SDS, C8N/SDS and C12N/SDS, respectively, while the γCMC values are maintained at low values (∼30 mN m−1) and there is almost no further decrease with varying XSDS. Thus, the variation of CMC and γCMC values displays similar trends to XSDS changing from 0 to 1.0, i.e., decreasing first and then reaching a plateau and increasing rapidly again. It is worth noting that the surface tension curves of the mixed solutions at high XSDS show an upward end with increasing CT, which is attributed to the significant excess SDS, thus causing the increase in surface tension to reach the γCMC value of SDS itself. These results show that the binding affinity becomes stronger when the molar ratio of CnN and SDS is close to 1:1, and the strong binding between the two components possibly induces the remarkable deformation or the transformation of the molecular configuration of CnN,60–62 thereby the molecular packing is significantly compacted at the air/water interface. However, the surface tension curves of the C12N/SDS mixtures at 0.20 < XSDS < 0.70 are excluded because the precipitation takes place in such a wide range of XSDS, which is different from the C6N/SDS and C8N/SDS mixtures with smaller cycloalkane rings. Despite these, we can still observe that the ability to reduce CMC is enhanced from C6N/SDS to C12N/SDS, demonstrating that the larger alkyl ring leads to the stronger hydrophobic interaction of the cyclic amines with SDS.
For C6N/SDS, the turbidity value remains extremely low no matter which value XSDS or CT is. Taking the situation at CT = 20 mM as a representative, the mean particle size remains below 20 nm. The broadly invariable turbidity and size as well as the lower CMC values (Fig. 1B1) demonstrate that the addition of C6N promotes the micellization of SDS, but only induces the formation of small spherical or rodlike micelles, which is confirmed by the cryo-TEM micrographs of C6N/SDS aggregates at XSDS = 0.20 and CT = 20 mM (Fig. S2†).
In contrast, the self-assembly behavior of SDS is significantly affected by C8N and C12N, and undergoes different processes at various XSDS and CT. For the C8N/SDS mixture, there is almost no change in the turbidity with CT as XSDS > 0.5, indicating that the small fraction of C8N has a strong ability to accelerate the micellization of SDS but not enough to affect the molecular packing and the aggregate structures. At XSDS > 0.5, the turbidity values increase with increasing CT, and the largest value appears at XSDS = 0.40. Fixing CT = 20 mM, a weakly bluish solution was observed, and the C8N/SDS mixed solution forms vesicles of a few tens of nanometers at XSDS = 0.20, while the solution becomes transparent and forms wormlike micelles with a few millimeters long at XSDS = 0.60, as evidenced by the results of DLS (Fig. 2C) and cryo-TEM (Fig. 2D1 and D2). Compared with the phase boundaries of C8N/SDS and C12N/SDS mixtures, they go through similar changing processes with increasing XSDS at the same CT, but the big differences take place at much lower XSDS and CT for the aggregate formation, the larger vesicles (>100 nm, Fig. 2D3) and the wider precipitate region over the CT range in C12N/SDS.
In summary, the addition of CnN only promotes the micellization of SDS for C6N, while leads to an aggregate transition from small micelles to vesicles and then to wormlike or spherical micelles at a fixed CT for C8N without precipitation and for C12N with a large range of precipitation. In all the three CnN/SDS mixtures, the electrostatic interaction between the protonated amine and sulfate groups and the hydrophobic interaction among the alkyl chains of CnN and SDS are the main driving forces for their self-assembly. However, taking a variety of conformations of cycloalkylamine in aqueous solution into account,60–62 we speculate that both the size and conformation of the cycloalkane ring of CnN may play a decisive role in the different aggregate formation and transition. Fig. 3 summarizes a simplified schematic diagram of aggregation behavior for the CnN/SDS mixture in bulk with changing in XSDS at a fixed CT.
The C6N molecule has the smallest cycloalkane ring. It is not long enough to enlarge the hydrophobic area in the aggregate relative to the SDS itself, thereby the C6N/SDS mixture only forms small micelles. In comparison, the C8N molecule with a larger cycloalkane ring can produce stronger hydrophobic interaction with SDS and form a larger hydrophobic area. As such, the spontaneous curvature of the C8N/SDS aggregates becomes slightly smaller, forming the wormlike micelles rich in SDS (XSDS > 0.5) and vesicles rich in C8N (XSDS < 0.5). As for C12N with a much larger cycloalkane ring, it forms precipitate with SDS in a large range, which could be attributed to the remarkably enhanced hydrophobic interaction by the unique C12N ring. It was reported that the conformers of cyclooctane can be classified into three families, including boat–chair, chair–chair or crown, and boat–boat, while cyclododecane shows a square topology and has at least 100 possible conformations.62 Thus, the large hydrocarbon ring is flexible enough to optimize the molecular conformation and nest together, in which case the ion pairs formed by C12N and SDS may be drawn closer with each other due to the tight entanglement between the C12N rings, thereby the precipitate with compact microstructure is formed, especially with the excessive C12N. For the same reason, a small amount of C12N can lead to a significant change in the molecular packing of C12N/SDS mixtures, thus the vesicles are formed at a large XSDS (e.g., 0.65 < XSDS < 0.85 at CT = 20 mM). Hence, the cycloalkylamines with different sizes of rings lead to diverse aggregate structures, and the large chair conformation of C12N displays the most obvious effect on the association with SDS.
With the visual approach to observe the oil-fouling efficiency, we took the photos of residual oil on stainless-steel plates under ultraviolet light (UV) based on the blue light character of oil-foiling under UV light (Fig. 4A). For both the C6N/SDS and C8N/SDS mixtures, the area of blue light on plates becomes smaller first and then enlarges again with the increase of XSDS, indicating that the optimum oil-fouling removal result appears around XSDS = 0.5. The C12N/SDS mixture follows the same changing pattern except for the precipitate region. In parallel, the quantitative evaluation by the weighting method is summarized in Fig. 4B, also indicating that the XSDS value for the highest oil-fouling removal efficiency is nearly 0.5. In addition, the C8N/SDS system shows the wider XSDS region (0.1 < XSDS < 0.5) to reach high cleaning efficiency and the efficiency is better than that of C6N/SDS. For C12N/SDS, in the region without precipitation at larger XSDS, it displays better cleaning performance than either C6N/SDS or C8N/SDS. That is, the CnN/SDS mixture with a larger cycloalkylamine and stronger aggregation ability exhibits higher oil-fouling removal efficiency.
To further understand the role of cycloalkylamine size in the oil-fouling removal efficiency of CnN/SDS, the cleaning process is investigated by oil–water interfacial tension and contact angles (θ) of oil on plate under solution (Fig. 4C and D). Normally, cleaning by surfactants is realized through the de-wetting mechanism,63 for which the contact angles of oil drops on substrates under surfactant solutions are very large. Herein, the contact angles of machine oil drops on the plates in the CnN/SDS solutions at the representative XSDS show a similar changing trend with the oil-fouling cleaning performance, i.e., good cleaning performance corresponds to a large contact angle. The oil-fouling tends to separate from the stainless-steel substrates at the medium XSDS (Fig. 4C). Therefore, the de-wetting occurs around XSDS = 0.5. Meanwhile, the interface tension between the oil and CnN/SDS solutions becomes smaller than that of either CnN or SDS, which is beneficial to the formation of O/W emulsion (Fig. 4D). And the O/W interfacial tension curves as a function of XSDS display a concave edge, showing a contrary changing trend of cleaning efficiency curves. It means that both the de-wetting and emulsion are enhanced at the medium XSDS, which facilitates the oil-fouling removal and impedes the separated oil to stick back on the substrates. Moreover, C8N/SDS shows 0.5 mN m−1 lower interfacial tension than C6N/SDS at XSDS < 0.50, whereas C12N/SDS shows the minimum interfacial tension at XSDS > 0.50. As a result, the oil-fouling is readily dispersed into the CnN/SDS solutions due to de-wetting and emulsion, and the highly efficient cleaning performance is obtained for C8N/SDS at XSDS < 0.50 and C12N/SDS at XSDS > 0.50.
Inspired by the Young's equation, we try to figure out the reasons for the various cleaning ability of CnN/SDS solutions in the view of equilibrium of forces. Herein, the interfacial tensions of solid/water, oil/water and solid/oil are defined as γ1, γ2 and γ3, respectively. The three parameters fit the equation, γ1 = γ2cosθ + γ3 (Fig. 4E). The contact angle of liquid on solid is proportional to the interfacial tension and energy, so according to the contact angles of CnN/SDS (n = 6, 8, 12) solutions on the stainless-steel plates (Fig. S3†), the γ1 values for pure SDS (γ10), C6N/SDS (γ1a) and C8N/SDS (γ1b) at XSDS < 0.5 are in order of γ10 > γ1a ≈ γ1b, while for C6N/SDS (γ1A), C8N/SDS (γ1B) and C8N/SDS (γ1C) when XSDS > 0.5, the order is γ1A > γ1B > γ1C. Meanwhile, γ20 > γ2a > γ2b at XSDS < 0.5 and γ2A > γ2B > γ2C at XSDS > 0.5 are obtained from Fig. 4D. Combining with the order of contact angle of machine oil droplets on the plates in the CnN/SDS solution (Fig. 4C), i.e., θ0 < θa < θb at XSDS < 0.5 and θA < θB < θC at XSDS > 0.5, we can observe that γ30 ≤ γ3a < γ3b at XSDS < 0.5 and γ3A < γ2B < γ2C at XSDS > 0.5. The results mean that the oil droplet has a stronger trend to retract inward and escape from the plate in C8N/SDS at XSDS < 0.5 or in C12N/SDS at XSDS > 0.5. All the experimental results and theoretical derivations provide insights into the cleaning efficiency of CnN/SDS, from which the oil-foiling removal performance could be predicted by the oil/water interfacial tension and solid/liquid contact angle of oil droplets in surfactant solutions.
Herein, to intuitively observe the foaming and defoaming ability of CnN/SDS systems with great oil cleaning ability, we firstly shake 10 mL CnN/SDS solutions of 5 mM for 30 s and record the change in foam height with aging (Fig. S4†). In parallel, we gain the variation of foam volume (Vfoam max), foam state, foam height, and bubble counts with aging by using a dynamic foam analyzer (Fig. 5 and S5†). We compare foam state against aging for the C6N/SDS and C8N/SDS mixtures at fixed XSDS = 0.10 mM and CT = 5.0 mM with the visual inspection method (inset) and air blowing method using a dynamic foam analyzer (Fig. S6†), and apparently the visual and quantitative results are consistent with each other. CnN (n = 6, 8, 12) alone almost has no ability to form liquid foams. With the addition of CnN, the foamability characterized by the maximum foam volume (Vfoam max, Fig. 5B) gradually becomes weaker with decreasing XSDS, and the C8N/SDS system displays the smallest foam volume when XSDS < 0.50, especially at XSDS = 0.10. Although the values of the maximum foam volume do not show significant differences, the differences in the foamability and foam stability are manifested in the changes of the bubble size, the foam heights and bubble count with time. Obviously, the low foamability of C8N/SDS is further verified by the bigger bubbles formed at XSDS = 0.10 and 30 s (Fig. 5A and S5†), following which the bubbles begin to break at 1500 s when the foams in pure SDS, C6N/SDS (XSDS = 0.10, 0.40, 0.50, 0.80), C8N/SDS (XSDS = 0.30, 0.40, 0.80) and C12N/SDS (C12N/SDS = 0.80) are still very stable. Correspondingly, the C6N/SDS system shows the similar changing speed in foam height against aging compared with SDS, the C12N/SDS system defoams also quite slowly, but the foam height of C8N/SDS decreases 3–5 times faster than that of SDS itself (Fig. 5C). The bubble count of CnN/SDS also confirms that the C8N/SDS system shows the fastest defoaming speed (Fig. 5D). All the above results demonstrate that the C8N/SDS mixture at XSDS < 0.50 can effectively reduce the foamability and enhance the defoam ability, perfectly corresponding to the condition for the high oil-fouling removal efficiency.
We speculate that there are two factors affecting the foam stability and defoaming ability of CnN/SDS mixed systems: surface tension and dynamic molecular exchange between the bulk phase and air/water interface. The low surface tension manifests that the surfactant molecules prefer packing at the air/water interface, which contributes to the formation and stabilization of foams. However, all the CnN/SDS systems reduce the surface tension of SDS from 38 mN m−1 to 30 mN m−1, thus the foamability and foam stability should be improved in principle. Actually, only the C12N/SDS system obeys this prediction, while the C6N/SDS and C8N/SDS systems exhibit similar or even weaker foamability and foaming stability than SDS. We turn back to analyze the CT value we selected. In order to compare the foams, we selected the same CT (5 mM), which is below the CMC of SDS (∼8.0 mM) and C6N/SDS (∼6.0 mM), or much larger than that of C12N/SDS (∼0.40 mM), but only slightly higher than the CMC value of C8N/SDS (∼3.0 mM). In general, there are three different mechanisms governing the lifetime of foam: (i) foam drainage caused by gravity, (ii) coarsening caused by the transfer of gas between bubbles generated by the capillary pressure differences, and (iii) bubble coalescence caused by the rupture of liquid films between neighbouring bubbles.64 For the present systems, it is proposed that the dynamic molecular exchange between the bulk phase and air/water interface can accelerate the enlargement and destruction of bubbles, which can perfectly explain the anomalous phenomenon in the foam (Fig. 5E). Given that the selected CT is close to but below the CMC for SDS and C6N/SDS, the surfactant molecules are enough to promote the adsorption saturation at the air/water interface but still have no ability to form micelles, in which case the molecules locating in the liquid film are relatively stable, resulting in the low defoaming ability. However, this CT is much higher than the CMC of C12N/SDS, which indicates that the absorption saturation has been completely reached at the air/water interface, and the number of molecules in the C12N/SDS aqueous solution is large enough to form stable micelles, thereby the molecules at the air/water interface do not exhibit the tendency to enter the bulk solution, leading to the stronger foam stability. In comparison, the CT is slightly higher than the CMC value of C8N/SDS. In the foaming process of C8N/SDS, more surfactant molecules join the air/water interface, while the real surfactant concentration in bulk is just high enough for micellization. The unstable bubble films promote the fusion of neighbour bubbles, and the surfactant molecules prefer to return to the micelles in bulk solution from the bubble films, i.e., the air/water interface, leading to the enlargement and destruction of bubbles under this condition.
In brief, it is concluded that high oil cleaning efficiency and low foam stability of surfactant mixtures are achieved at an appropriate concentration, i.e., above CMC, but not too large at a proper XSDS. Inspired by the present results, we tested oil-fouling removal efficiency and foam height with the C6N/SDS and C8N/SDS mixtures after vortexing for 30 s and resting for 2 h at XSDS = 0.10 and various CT. For the given XSDS at 0.10, we found that the cleaning efficiency becomes higher with the increase of CT and is significantly enhanced especially above CMC. However, there is a minimum foam height just above CMC, after which the height becomes larger again (Fig. 6A and B, S7 and S8†). The CMC of C6N/SDS is larger than that of C8N/SDS, so the minimum appears at the larger concentration for C6N/SDS, i.e., at ∼7.5 mM for C6N/SDS and ∼6.0 mM for C8N/SDS. Combining with the continuous enhancement of cleaning efficiency and the minimum foam height above CMC, it can be speculated that selecting the concentration slightly above CMC at low XSDS should be beneficial to obtain a high-efficiency and low-foam cleaning system. Therefore, the concentration of 5 mM used above meets the requirements for C8N/SDS, resulting in good cleaning and deforming performance at XSDS < 0.50. Due to the weaker self-assembly ability of C6N/SDS at the air/water interface and in bulk, its performance in cleaning and defoaming efficiency cannot be better than that of C8N/SDS.
Footnote |
† Electronic supplementary information (ESI) available: Fig. S1–S7. See DOI: https://doi.org/10.1039/d3lf00145h |
This journal is © The Royal Society of Chemistry 2024 |