Introduction

The benefits of asphalt emulsion are very relevant when they are used as on-site techniques, because the energy and gas emissions are lower than in the conventional ones.

The balance of intermolecular forces drives the stability of bituminous emulsions. Although this kind of emulsions are being used for many paving applications, further improvement will be possible when the whole principles of physical chemistry that ruled the colloidal behavior are included to the applied engineering.

The interfacial area between the asphalt and the water is about 500 m2 per liter of bituminous emulsion. In order to prevent the coalescence and providing stability to the emulsion, surfactants are required.

Surfactants are materials that concentrate at interfaces—oil and water, for example—and surfaces. This particular behavior let them to lower interfacial tension; consequently, they are involved in several applications such as detergency, cosmetics (Ontiverosa et al. 2014), water purification (Gonzalez-Perez and Persson 2016), production of silica-based mesoporous materials (Yang et al. 2017), and phase transfer catalysis (Gang et al. 2017).

A typical surfactant molecule is made up of two different parts: polar or hydrophilic head group and non-polar or hydrophobic chain group. The first one makes them soluble in water, and this group is very important for the aqueous surfactant solution properties.

Taking into account the charge carried by the head group, surfactants can be classified as anionic, cationic, non-ionic, and zwitterionic ones.

One of the most important properties of cationic surfactants is the emulsification, which is relevant in many applications, such as secondary oil recovery and emulsifiers in asphalt emulsions. In addition, their properties can be modified easily by the substitution of the counterion. Another important property is its resistance to bacteria, fungus, and other microorganisms.

As it was said, surfactants play an important role as emulsifiers in asphalt emulsions. Some of them, such as dodecylbenzene and p-toluene sulfonic acids, were used as additives in crumb-rubber-modified asphalt binders (Zhou et al. 2015). More recently, Schulz et al. studied the properties of sodium oleate and hexadecyltrimethylammonium bromide mixtures in order to establish their applicability in the asphalt emulsification for pavement production (Schulz et al. 2016). Rodriguez-Valverde et al. used a commercial cationic surfactant—a blend of N-alkyl propylendiamines and alkylamidoamines, both derived from tallow—in order to characterize bituminous emulsions using imaging techniques (Rodríguez-Valverde et al. 2008). Urbina-Villalba et al. carried out an interesting theoretical study about the surfactant structure and the stability of their oil/water emulsion (Urbina-Villalba and García-Sucre 2000). On the other hand, Gómez-Meijide et al. studied the mechanical properties of cold mix (Gómez-Meijide and Pérez 2014) and the effect of using demolition waste in cold asphalt mixtures (Gómez-Meijide and Pérez 2013).

Bitumen emulsions provide an alternative approach in which the bitumen is liquefied by dispersing in water. Bitumen or asphalt emulsions have a great advantage upon asphalt, because the processes with asphalt emulsions need less energy, because their viscosity is lower than the asphalt one. In road construction applications, emulsions are environmentally friendlier system than hot bitumen mix. It does mean less risks of fire; burns are avoided; and, it was already mentioned, the process uses less energy. About 7 L fuel/t (270 MJ/t) is consumed in a typical hot mix asphalt at 150 °C, while a half-warm mix consumes just only around 1 L fuel/t (46.45 MJ/t) (Almeida-Costa and Benta 2016). Furthermore, warm mix releases less ozone-generating hydrocarbons. In other words, asphalt emulsions are a good alternative to the conventional asphalt paving process, because the lower the energy needed is, the lower the emissions of carbon dioxide to the atmosphere are. In addition, the amount of bitumen used is considerably reduced. Amount about of 0.01 to 0.2 M of surfactant is needed for a rapid-setting emulsion, and even larger quantities should be used for cationic quick-setting emulsions (Takamura and James 2015). Critical micelle concentration (cmc), solubility of ionic surfactants, and the capacity to be ionized are very important aspects of the emulsifier. Thermodynamic properties of a micellar system are of major importance to obtain relevant information on the behavior of it. Particularly, enthalpy, entropy, or standard free Gibbs energy change upon the micellization process (Perger and Bešter-Rogac 2007); hence, the changes in hydrophobic interactions leading to micellization can be better evaluated. These main thermodynamics quantities can be derived from the critical micelle concentration with temperature, and all of them undergo changes upon micellization (Gonzalez-Perez and Sanchez-Dominguez 2013).

The main goal of this work was to analyze the influence of temperature on the micelle formation for a non-commercial cationic emulsifier: dodecylpyridinium thiocyanate (C12PCNS) and compare our results with literature data.

Experiment

Material and methods

Dodecylpyridinium thiocyanate is not a commercial emulsifier; it was synthesized in the following way:

A solution of 0.1 mol of dodecylpyridinium chloride (Aldrich) in 200 cm3 of water was treated with a fivefold excess of aqueous solution of potassium thiocyanate.

The precipitate was filtered off and washed through by rinsing it with the KCNS solution.

The raw material was dissolved in warm water and re-precipitated by adding an excess of KCNS solution. The crude product was then purified by several recrystallizations from cold water. Thus, purified dodecylpyridinium thiocyanate was air-dried.

Conductivity measurements were made with a Wheastone bridge conductometer type CM-177 from Kyoto Electronics and a cell type K-212 from Kyoto Electronics too. The cell constant was determined by calibration with several dilute concentrations of potassium chloride solutions (Barthel et al. 1980). All measurements were carried out in a thermostat bath (Polysciencie 9010), maintaining the temperature constant to within ± 0.05 °C.

A special measuring cell was designed in order to optimize material. As Fig. 1 shows, the cell has a smaller base radius than at the central and upper sections. This way, we used a small amount of C12PCNS. The initial concentration of the water solution of C12PCNS inside of the cell was equal to twice the expected cmc. Then, it was being diluted by means of a calibrated dispenser.

Fig. 1
figure 1

Measuring cell

Full size image

Results and discussion

The great temperature dependency of solubility in many ionic surfactants is one of the most remarkable effects in this kind of materials. Solubility may be very low at low temperatures and then increase by orders of magnitude in a narrow temperature range (Jönsson et al. 1998). This is a well-known phenomenon; the temperature at which the solubility increases strongly is called Krafft temperature or Krafft point.

Krafft point depends not only the length of alkyl chain but also on the head group and counterion.

Krafft point was first estimated for this non-commercial emulsifier in order to establish the onset temperature of the measurements.

For this purpose, we have used the same method that we reported in a previous paper (Galan and Rodriguez 2010); that is, an oversaturated water solution of C12PCNS was prepared at temperature below 6 °C. Afterwards, the specific conductivity of this solution against temperature was measured without changing the concentration of the emulsifier (Fig. 1).

As Fig. 2 shows, the Krafft temperature, TK, is around 20 °C. TK varies as a function of both the nature of the hydrophobic chain and the head group with its counterion. However, there is no a general trend for the counterion dependence; in some cases, the Krafft temperature increases as the atomic number of counterion decreases, and in another, the situation is the opposite (Galan et al. 2003). The most of dodecylpiridinum halides have a very low TK. In a previous paper, we reported a TK around 16 °C for pentadecypyridinium bromide C15PBr (Galan and Rodriguez 2010), that is, three more methylene groups in the hydrophobic chain. Therefore, these results show that the interaction between the head group and the CNS− is very strong. Conventional asphalt hot mixtures (“hot mix”) are generally produced at more than 150 °C. The asphalt emulsions are used in half-warm, warm, and in cold mixes. In the first case, the range of the temperature mixes is from 40 to 60 °C; the second from 60 to 120 °C; and in the third one, the materials are unheated and mixed at ambient temperatures. Due to its high Krafft temperature, this emulsifier is more suitable for half-warm or warm mixes than the cold ones, except when the ambient temperature is much higher than 20 °C, that is, in warm climates.

Fig. 2
figure 2

Krafft temperature for C12PCNS

Full size image

Specific conductivity against molality concentration plot, in the temperature range from 25 to 50 °C, is shown in Fig. 3. As it is well known, the break in the slopes is due to the micelle formation. The cmc was determined from the intersection of two linear fits above and below on the critical point in the specific conductivity vs molality plot.

Fig. 3
figure 3

Specific conductivity vs molality from 25 to 50 °C

Full size image

In order to establish a comparison, Table 1 shows the results of the cmc for several alkylpyridinium studied previously and determined with the same procedure (Galan et al. 2002, Galan et al. 2003, Galan et al. 2005; Galan and Rodriguez 2010). The cmc values for C12PCNS are more similar to C14PBr and C14PCl than the homologous with the same chain length. This result is consistent with the lyotropic series established by Hofmeinster (Mukerjee 1967); that is, the cmc decreases in the order Cl− > BrO3− > NO2− > N3+ > Br− > NO3− > ClO3− > I− > SCN−.

Table 1 cmc (mol kg−1) values for C12PCNS and other pyridinium homologs
Full size table

The degree of micelle ionization, β, was estimated as the ratio of the slopes above and below of cmc of the specific conductivity-molality plot (Bouchal et al. 2016). β displays a linear dependence with temperature. This behavior was expected, because the higher the temperature is, the higher the thermal agitation is; hence, the counterion is easily dissociated from the aggregate.

Figure 4 shows a comparison of the micelle ionization degree of C12PCNS with two homologs with the same chain length, previously reported (Galan et al. 2002). In the present case, β follows the Hofmeinster series too. For that reason, the degree of ionization for C12PCNS is lower than of the other homologs of the same chain length. This fact limits its range of application at low temperatures. However, since beta grows linearly with temperature, at temperatures above 30 °C, its behavior can be considered optimal to achieve the flocculation of the emulsion. In fact, the lower temperature was already limited, as stated above, by Krafft temperature. For cold mixtures, Cl− is more suitable than CNS− because its degree of ionization is higher and its Krafft temperature is lower.

Fig. 4
figure 4

Ionization degree for C12PCNS and other homologs with the same chain length

Full size image

Figure 5 shows the behavior between the cmc vs temperature. Unlike other homologs with the same chain length (Galan et al. 2002, Galan et al. 2003), in this case, the typical U-shaped curve between cmc against temperature does not appear; it is because the Krafft temperature is high enough to prevent the presence of a minimum. As the temperature becomes higher, the hydrophobicity of the emulsifier decreases; therefore, the critical micellar concentration increases (Chen et al. 1998).

Fig. 5
figure 5

cmc for C12PCNS vs temperature

Full size image

The more remarkable thermodynamic parameters of micellization such as the standard enthalpies, ΔHm0; entropy, ΔSm0; and Gibbs free energy, ΔGm0, of micellization for C12PCNS were estimated by means of the pseudo-phase separation model (Shinoda and Hutchinson 1962; Hunter 1993; Blandamer et al. 1995; Kim and Lim 2004). According to this model, the standard Gibbs free energy of micellization can be calculated from the relation

$$ \Delta {G}_m^0=\left(2-\beta (T)\right) RT\ln {x}_{cmc}(T) $$
(1)

where Xcmc is the mole fraction of the surfactant at the cmc, and it is a temperature function, the same as β.

On the other hand, ΔHm0 can be estimated from the Gibbs-Helmholtz relation

$$ \Delta {H}_m^0=\left[\frac{\partial \left(\raisebox{1ex}{$\Delta {G}_m^0$}\!\left/ \!\raisebox{-1ex}{$T$}\right.\right)}{\partial \left(\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$T$}\right.\right)}\right] $$
(2)

and ΔSm0 from the relation

$$ \Delta S=\frac{\Delta {H}_m^0-\Delta {G}_m^0}{T} $$
(3)

Table 2 shows the results obtained for C12PCNS.

Table 2 Thermodynamics parameters for C12PCNS at several temperatures
Full size table

These values are in good agreement with our previous works for pyridinium surfactants (Galan et al. 2002; Galan and Rodriguez 2010) and with other homologs of ammonium quaternary salts (Perger and Bešter-Rogac 2007). At a given chain length, as the value of the standard Gibbs free energy is more negative, the larger the counterion.

The micelle formation is an exothermic process in the whole temperature range studied.

Several processes in aqueous solution with small solutes exhibit a linear relationship between enthalpy change and entropy change. This relationship is given by Eq. 4, and it is called compensation rule (Zielinsky et al. 1987; Moroi 1992; Mehrian et al. 1993; Muller 1993; Chen et al. 1998; Sugihara and Hisatomi 1999; Gonzalez-Perez et al. 2003; Galan et al. 2005).

$$ \Delta {H}_m^{{}^{\circ}}=\Delta {H}_m^{\ast }+{T}_c\Delta {S}_m^{{}^{\circ}} $$
(4)

Although this phenomenon is unresolved yet (Pan et al. 2016), we can explain it as follows: the mechanism of micellization is the following: firstly, the hydrophobic tails are dehydrated; we called it the desolvation part. Finally, the aggregation of the hydrophobic tails happen; this is so-called the chemical part. According to several authors (Chen et al. 1998), the slope of Eq. (4), Tc, is a measure of the first part, desolvation, and it is known as compensation temperature and the intercept is considered as an index of the second part, the chemical one.

On the other hand, taking into account that in the intercept, ΔH*, there is no any entropy change, ΔH* has the same value as the standard Gibbs free energy change (Eq. (2)); then, the larger the ΔH* is, the less stable the micelle is. So, the value of the intercept can be considered as a measure of the stability of the aggregate.

Figure 6 shows the compensation rule plot for C12PCNS. In this case, the value for Tc is around 310 K and 45.63 kJ for the intercept. Comparing these values with other homologs (Table 3), we observed that the compensation temperature for C12PCNS is higher than the other ones for homologs of the same chain and it is similar to C16PBr.

Fig. 6
figure 6

Compensation rule for C12PCNS

Full size image
Table 3 ΔH*m and TC values for several alkylpyridinium halides
Full size table

The value for ΔH* reveals that the aggregates of C12PCNS are very stable, because it reaches a value around −45.63 kJ, which is very close to the C16PBr one, which has two more methylene groups.

Summary

In present work, we have used conductivity measurements to study the micellization process in aqueous solution of a non-commercial emulsifier, dodecylpyridinium thiocyanate (C12PCNS). It has been shown that the inflection point in the specific conductivity against molality plots corresponds to the position of cmc. This process was carried out at several temperatures. The shape of the curve of dependence of cmc on temperature shows only a branch of U-shaped plot. That is due to the high value of Krafft point.

The analysis of the thermodynamic parameters of the micellization for this compound shows that ΔHm0 becomes more exothermic upon increasing the temperature. The similar ΔGm0 values for the whole range of temperature show that the micellization process occurs spontaneously. On the other hand, the ΔSm0 values of the emulsifier solution decrease when temperature rises; it reveals that the order of the system increases with the temperature.

The compensation rule phenomenon was studied for this compound, and the values of Δ H*m and TC were compared with other alkylpyridinium halides.

According to the results, this emulsifier is useful for warm asphalt mixes or half-warm mixes, mainly due to high Krafft temperature and low degree of ionization. The latter increases with temperature; therefore, the stability of the mixture will be higher for temperatures above 50 °C. Although taking into account that the range of temperatures of the cold mixes extends from 0 to 40 °C, C12PCNS can be used as emulsifier in cold mixes in countries with warm climates.