Homologue Distribution
Derivatives of 2-ethylhexyl alcohol were synthesized by the polyoxyalkylation reaction. The structure of the prepared compounds is presented in Scheme 1. Methyloxirane was added to 2-ethylhexanol in order to increase its hydrophobic character in the first step, followed by polyethoxylation, to obtain amphiphilic compounds (Scheme 2).
Scheme 2Scheme of the sequential propoxylation 2-ethylhexanol followed by polyethoxylation
Results on GC quantitative determination for the first series of 2-ethylhexanol ethoxylates (EHE
n
) are presented in figure. It can be seen from Fig. 1 that by application of the DMC type catalyst, the generated product distributions are narrower, as compared to KOH. It was widely reported that narrowing of homologue distribution improves physical and chemical properties, as well as environmental interactions, widening the application possibilities of ethoxylated products [7–9]. Moreover, significantly higher conversion of the starter alcohol is evidenced; this is due to very high activity of DMC-type catalysts, which is their major advantage in the polyaddition of oxiranes. This enables us to use them in a very small concentration (down to ppm). The formula of DMC catalysts can be described as follows:
$${\text{Me}}_{x}^{\text{I}} \left[ {{\text{Me}}^{\text{II}} \left( {\text{CN}} \right)_{6} } \right]_{y} \cdot L_{{a_{1} }}^{1} \ldots L_{{a_{n} }}^{n} , ,$$
(1)
where MeI and MeII denote transition metals (mostly Zn and Co), \(L_{{a_{1} }}^{1} \ldots L_{{a_{n} }}^{n}\) mean suitable ligands (alcohols or ethers), and x, y, a
1…a
n
are integers. Generally, DMC catalysts are prepared by the reaction of a selected metal salt with a cyanide derivative of another metal (mostly alkaline) in the presence of an organic ligand. Regardless of the appropriate selection of substrates, preparation of the DMC catalysts is highly complex, with many critical factors, which determine the obtained catalytic activity and the quality of the polyadducts. The precise characteristic of DMC is described in one of our previous papers [10].
However, in spite of the differences of homologue distributions generated by the two compared catalysts, an additional question arises with regard to the quantities of byproducts. It had already been reported that parallel to the desired polyaddition, polymerization occurs and polyoxyethylene and polyoxypropylene diols are formed, especially at high molecular weights [11]. Therefore, size exclusion chromatography was applied to compare the products, depending on the average polyaddition degrees and the applied catalyst. It can be seen from Fig. 2, that the average molecular weights determined by GPC for the DMC-based products are close to the theoretical ones, as calculated from the synthesis weight balance. For that purpose the M
w, M
n
and M
z
parameters were calculated from the GPC chromatograms by the following equations:
$$M_{n} = \frac{{\sum N_{i} M_{i} }}{{\sum N_{i} }};\,M_{w} = \frac{{\sum N_{i} M_{i}^{2} }}{{\sum N_{i} M_{i} }};\,M_{z} = \frac{{\sum N_{i} M_{i}^{3} }}{{\sum N_{i} M_{i}^{2} }};$$
(2)
where M
i
is molar mass of an ith individual molecule and N
i
is the number of particles i in the sample. More significant differences among the M
w, M
n
and M
z
parameters were found in the case of the KOH based products.
Larger gaps between the average molecular weights determined by GPC and the theoretical weight balance were noted at the average polyaddition degree (N
av) equal to 12. Even at N
av = 9, KOH yields already remarkable discrepancy. It means that along the increasing molecular weight, the polymerization rate overcomes that of polyaddition. It may be explained by formation of polymers (diols) with two reactive centers at each chain and double consumption of oxirane, relative to the monofunctional polyadducts. Furthermore, the alkaline catalyst was found to be less selective for the products with higher average propoxylation degree, which seems neutral in the case of DMC, where the calculated M
w, M
n
and M
z
values were mostly close each to the other (Fig. 2).
Additionally, the homologue distribution can also be compared by the polydispersity parameter MWD = M
w/M
n
. The calculated MWD distribution parameters for the studied series of EHP
m
E
n
obtained with KOH and DMC as the catalysts are presented in Table 1.
Table 1 Physical and chemical properties of synthesized surfactants in comparison to nonionic C12–C14 alcohol surfactants
It can be observed that the determined MWD values for products synthesized in the presence of DMC catalyst are lower, as compared to their equivalents produced with KOH. This can also confirm a more selective reaction towards the aimed molecular weights with DMC type catalyst, where the concentration of homologues near the average polyaddition degree is higher.
The much more complex homologue distributions of the EHP
m
E
n
series are difficult to determine quantitatively by the GC technique. Generally, polydispersed products of narrower distributions should contain lower amounts of the unconverted starter, as well as lower concentrations of the low molecular fractions. As the result, they must be assessed by relevant differences determined by thermogravimetric (TG) measurements, where the lower fractions evaporate at lower temperatures. Such an approach had already been reported in the literature [12].
Thermal analysis shows that the decomposition temperatures (T
onset50) for EHP
m
E
n
based on KOH are in the range 209–411 and 220–403 °C for DMC ethoxylates. Note that the greater average propoxylation grade (P1–P3) is, the higher the temperature of mass loss. The same can be observed with the polyethylene ether chain (E3–E12) length. However, the role of propoxylation was found to be negligible, because temperatures of decomposition for EHE
n
were in a similar range to EHP
m
E
n
. The EHP3E12 products at 50 % mass reduction already fall as a whole within the temperature range where thermal decomposition dominates, so the observed differences in mass decrease between the KOH and DMC equivalents are reduced. The discussed relationship for all series of the studied products is presented in Table 1.
The same was also confirmed by determination of hydroxyl numbers (L
OH) for the discussed series of products, where the experimental L
OH values indicate much higher molecular weight then the theoretical values calculated from the synthesis weight balance, especially for the products with n = 9 and n = 12 (Fig. 3).
Concentrations of DMC and KOH catalyst at 130 °C were 0.05 and 0.3 wt% per product weight, respectively. Therefore, the average reactivity parameter, Rp, was calculated to compare the unit conversion rates in each case:
$$Rp = \frac{{g_{A} }}{{g_{\text{C}} \times g_{\text{EH}} \times h}}$$
(3)
where g means mass (grams), A is methyloxirane or oxirane, EH states for 2-ethylhexanol, h denotes hour and C means catalyst.
It can be seen from Table 1, that the DMC type catalyst is much more reactive than KOH. Both, propoxylation and ethoxylation commenced many times faster, which is desired from economical point of view, especially if the product quality is equivalent or better.
Physicochemical Properties
Refractive Index and Hydroxyl Number
The refractive index (Rf) was determined for the studied series of products to investigate their sensitivity towards the average polyaddition degree and homologue distribution. Values were determined at 60 °C, where all of the investigated samples were liquid (Table 1). Values of Rf index increase with the average polyaddition degree. Therefore, not a big difference was found between the series of products generated with DMC and KOH as catalysts. However, it is interesting that the addition of the first propoxy-mere decreases the values of Rf index, as compared to the EHE
n
series. It may be a steric effect of the methyl branched chain. At higher average polyaddition degrees (n > 6) the discussed influence of the first P block is dominated by the larger number of the additive ether bounds of the ethoxy blocks.
The complex analytical investigation by GC, TG, GPC techniques supplemented by determination of L
OH numbers and Rf indexes gives a clear picture of the composition of the studied products. It was proven that they are distinguished by a much narrower homologue distribution and selectivity where the DMC catalyst is applied. The differences are significant, especially at lower average polyaddition degrees (n < 9). At the higher range of molecular weight (n > 9) the polyaddition reaction seems dominated by parallel polymerization yielding larger amounts of the polymer diols.
Cloud Point
One of the key features of surfactants is their solubility in polar media. Values of the cloud point in 25 % butyldiglycol (BDG) solution were determined (Table 1). Again, the recorded solubility of the studied surfactants in polar solutions (BDG) indicates similar behavior of the investigated EHP
m
E
n
series compared to those of C12–14 alcohol. The temperatures of the cloud point rise in the series of homologues for both catalysts, i.e. values for EHP2E
n
are in the ranges 38.1–94.5 and 38.1–94.5 °C for KOH and DMC derivatives, respectively. The temperatures increase with the length of polyoxyethylene ether chain in both groups. Among P1–P3 blocks there are some differences, but not so significant, i.e. 80.0 °C for EHP1E9, 82.1 °C for EHP2E9 and 88.1 °C for EHP3E9 in the DMC group. The same minor differences are observed for compounds derived using KOH.
The results from Table 1 show that with the elongation of the polyoxyethylene chain the physical state changes. Compounds EHP
n
E3, EHP
n
E6 and EHP1E9 based on both catalysts are liquids. However, DMC-derived EHP2E9 is a liquid, whereas the KOH-derived one is a solid. Physical states of compounds from the EHP3E
n
group for both catalysts do not differ much. Ethoxylates with higher polyaddition degree, where n = 12, are solid. However EHE
n
DMC products remain less viscous and liquid at higher polyaddition degrees in comparison to KOH based ethoxylates. The liquid state is much more convenient from the technological point of view, because the material does not require melting for transport and discharge. Generally, the EHP
m
E
n
products appear very similar to their C12–C14 alcohol equivalents, in this aspect. A higher average propoxylation grade (P1–P3) tends to favour solidification at room temperature of the products of higher molecular weights. Physical and chemical results presented in Table 1 confirm that the EHP
m
E
n
surfactants show a behavior similar to conventional C12–14 alcohol ethoxylates, which is a positive prerequisite for the market.
Surface Activity
The surface-active properties of the series of 2-ethylhexyl alcohol derivatives are summarized in Table 2. One of the criteria of surface activity, characteristic for surfactants is the critical micelle concentration (CMC), defining their concentration in water solution, at which monomers start to aggregate into micelles. The CMC was determined for the water solutions of studied surfactant series and the results are shown in Table 2. The addition of methyloxirane decreases values of the CMC, while the lowest CMC is observed for the EHP
m
E3–6 systems. For these surfactants, CMC values are usually one order lower in comparison to EHE
n
. This supports the concept of enhancement of surface activity of the EH-based surfactants by the addition of the P1–P3 blocks into the molecule. Moreover, they appear close to those of the reference LaE
n
surfactant series.
Table 2 Surface activity of synthesized surfactants with comparison to nonionic C12–C14 alcohol surfactants
Furthermore, the values of the surface tension γ
CMC of the investigated surfactants were determined (Table 2).
The surface tension of aqueous solutions γ
CMC of the studied surfactants decreased from water value (72.8 mN m−1) to a minimum located from 26.9 to 40.07 mN m−1, where it reached a plateau. The determined surface tensions for P1–P3 blocks appear at the similar level of common values. The influence of homologue distribution in EHP3E
n
group, catalyzed with DMC is presented in Fig. 4. Generally, values of surface tension γ
CMC increase with the length of polyethylene ether chain. However, the curves represented compounds with low polyaddition degree, where n = 3 and n = 6, show that values of γ
CMC and CMC do not differ significantly.
Calculations of surface excess concentrations Γ
max, Gibbs free energy of the adsorption layer ΔG
0ads
and the minimum surface area occupied by a molecule at the interface A
min were described earlier [13]. It was observed that values of these parameters do not differ significantly. However, with the elongation of polyethylene ether chain, the values of A
min for EHP1E
n
increase from 6.99 to 11.3 × 10−19 m2. This may be caused by hydration of the polyethylene ether chain. It was found that the length of the polyethylene ether chain determines the size of the surface area occupied by a molecule. The negative values of ΔG
0ads
for all studied surfactants indicate that the process proceeds spontaneously [14].
There are two additional surface activity parameters, the adsorption efficiency, pC20, and the effectiveness of surface tension reduction, Π
CMC, which are calculated from surface tension measurements. The pC20 is defined as the negative logarithm of the surfactant concentration in the bulk phase required to reduce the surface tension of the water by 20 mN m−1, which represents efficiency of surface adsorption on an air–water interface. Then, the greater the pC20 value is, the higher is the adsorption efficiency of the surfactant. The other parameter, Π
CMC is the surface pressure at the CMC, being defined by \(\varPi_{\text{CMC}} = \gamma_{0} - \gamma_{\text{CMC}} ,\) where γ
0 is the surface tension of pure solvent and γ
CMC is the surface tension of the solution at the CMC. Surface pressure at the CMC provides information on how significant is the ability of the surfactant to reduce the surface tension of the solvent and hence, what is the effectiveness of this phenomenon. It can be observed that the highest values of pC20 were obtained for EHP2E3 KOH and EHP3E3–6 DMC, which is equal to 3.81 and 4.11, respectively. The highest effectiveness of surface tension reduction Π
CMC was determined for EHE6 KOH and EHE3 DMC, 45.5 and 44.9 mN m−1, respectively. Moreover, these values are comparable to those from LaE
n
(Table 2).
Wettability was also investigated for the studied surfactant series and the results of contact angle measurements (CA) are presented in Table 2. Values of CA increase with the length of polyoxyethylene chain. There are no significant differences in CA measurements between EHEn and their nonionic equivalents LaE
n
. This similarity is best illustrated in Fig. 5, where the drops of EHE12 (57.2°) and its nonionic equivalent LaE12 (56.9°) are compared. Reduction of CA is essential for the cleaning or washing effect, because of better wetting of hydrophobic surfaces.
Generally, there are no significant differences in surface-active parameters for EHP
m
E
n
and EHE
n
in comparison to their nonionic equivalents LaE
n
.
A practically important issue of surfactant solutions is their foaming performance, which is required at a high or minimum level depending on application. Many reports on foamability of polyglycerol fatty acid ester are available [15–18]. The most common detergent applications include the household automotive washing or industrial Clean in Place systems (CIP), which require minimum or not foaming at all. The other bulk applications like lubrication, flotation and many others do also limit foaming of the applied surfactants. The example of foaming performance for the studied series of surfactants and their nonionic equivalents is presented in Fig. 6. It might be noted, that while the initial foaming (after 1 min observation) is comparable for most of the studied nonionic series, the stability of foam (after 10 min) of the EHE
n
is visibly lower than LaE
n
. It may point to the application of EHE
n
(especially those synthesized with DMC catalyst) as low-foaming surfactants.