Formulating with cyclomethicone replacements

The INCI name, cyclomethicone, refers to a family of cyclic dimethyl siloxanes that includes cyclotetrasiloxane (D4), cyclopentasiloxane (D5), and cyclohexasiloxane (D6), a family that has come under increased environmental scrutiny in recent years. One of the main reasons for this scrutiny is the volatility of cyclomethicone. With the problems and concerns of cyclomethicone, the quest for a replacement material that provides a dry feel but is not volatile becomes paramount. Key areas where this family of materials is used include antiperspirants, colour cosmetics and as a base solvent to blend with fragrance oils and perfume oils. Cyclomethicone is a clear, odourless silicone. It leaves a silky-smooth dry feel when applied to the skin. As shown in Figure 3, cyclomethicone has a cyclic structure with alternating silicone-oxygen atoms. Cyclomethicone provides a low heat of vaporisation and low vapour pressure, which leads to their use as cosmetic vehicles. In other words, the feel correctly or incorrectly has been associated with volatility. However, the physical chemistry that results in the dry feel of cyclomethicone is a complex one. Volatility is but one aspect of the complex phenomenon that contributes to a dry feel in a solvent used in cosmetics. Typically volatile organic compounds are flammable and dangerous to incorporate into a cosmetic formulation. Along with volatility, other considerations that impact upon selection of a D5 replacement include viscosity, surface tension reduction (which affects spreadability), flammability of the solvent, effects of the solvent on skin, and cost. Clearly, a D5 replacement that is flammable, defatting and expensive is unacceptable. The assumptions we used in evaluating the suitability of materials as D5 replacements:

•  Non-volatile.
•  Not flammable.
•  Dry feel.

 Volatility

Volatility is the ability of the compound being tested to evaporate under the temperatures at which the compound is used in formulation. For cosmetic products, this temperature is ambient (~25°C). It has generally been accepted that cyclomethicone compounds provided this feel because they evaporate quickly after helping to carry oils into the top layer of the epidermis. The assumption that volatility is required for dry feel is due to the fact that historically D4 has a dry feel and is volatile. D5 has replaced D4 in formulations and consequently has been found acceptable in many cosmetic formulations as a D4 replacement, but is it volatility that is primarily responsible for the dry feel? Personal care products are unique in that they are not applied at elevated temperatures as are many industrial products. It is therefore unreasonable to talk about volatility of a cosmetic solvent at elevated temperatures. The temperature of application is more typically 20°C than 200°C, yet we measure volatility at elevated temperatures. If one looks at a temperature of 20°C, the volatility more closely resembles what might occur in topical cosmetic products. Table 1 shows the percentage of material remaining over a 24-hour period at 20°C for several silicone compounds commonly evaluated as replacements for D5 (see also Fig. 2). In recent history, silicone polymers or silicone fluid has been used as a nonvolatile replacement for cyclomethicone. Silicone fluid is synthesised by the equilibration reaction of D5 with MM. The structure is shown in Figure 1. The amount of repeat units (DP) of the polymer is represented by (b). The higher the amount of D units (b), the higher the molecular weight and viscosity. Silicone fluids are typically sold by their viscosity; the higher the viscosity, and the higher the molecular weight. When compared to cyclomethicone, low molecular weight, or viscosity, fluids are volatile but as molecular weight is increased they become nonvolatile. To compare their volatility to their cyclic counter parts, the evaporation time of D4 and D5 were compared to three different silicone fluids. Table 1 shows the results of the evaporation experiment. Low viscosity silicone fluid (0.65 cst fluid (INCI: Disiloxane) and 1 cst fluid (INCI: Trisiloxane) are volatile while D4, D5 and 2 cst fluids are not volatile. However, it is also clear that D4>>D5>2 cst. When plotted, it is easy to see how the volatility changes in these materials. The cosmetic industry has accepted D5 as a replacement for D4 and it is clearly seen from this data that it is non-volatile at temperatures that the consumer is likely to encounter, so the dry feel of D5 formulations must be caused by other factors than volatility alone.

Surface tension

One of the key properties that make silicone interesting in personal care products is its ability to lower surface tension. Silicones lower surface tension so effectively because of their unique solubility. Figure 3 shows an optical image of a three-phase separation of water, oil and silicone fluid. As seen from the photo, silicones are hydrophobic and oleophobic. This means when silicone is placed into the same container or formulation as oil or water, it will minimise the interaction with that solvent and migrate to the interface. Fatty compounds have a surface tension of about 30 dynes/cm. Silicone compounds have a surface tension of about 20 dynes/cm. When silicones, D4 and D5 or silicone fluids are added into a formulation, they can lower the surface tension as low as 20 dynes/cm under very low concentrations. The major problem with silicones, as seen in Figure 3, is solubility. Silicones can be modified to be compatible with many different solvents.

Silicone fluid modification

There are several types of silicone polymers that can be used to increase the solubility of silicones in different solvents. Alkyl silicone polymers are high molecular weight silicone polymers that have alkyl pendent groups covalently bonded to a silicone backbone. Alkyl silicones are soluble (clear) in oils, but they are surface active due to the silicone backbone of the polymer. Alkyl silicones when added to oils, including triglycerides, lower surface tension at very low concentrations. This lower surface tension together with low viscosity, results in a dry skin feel because of efficient spreading on the skin and the fact that these silicone polymers will lower the surface tension of the oils on the skin. Figure 4 shows the effect of adding a C22 alkyl silicone to soybean oil. As the concentration of alkyl silicone is increased, the oil forms a gel. The dry effect achieved on the skin by the efficient reduction of surface tension of a natural oil using low concentrations of alkyl silicone is a patent pending, and is an extremely valuable formulation tool for the formulator. Not only has the approach of replacing D5 with combination of natural oils and a small amount of alkyl silicone been used to replace D5, the replacement is more green than D5, (due to the fact that natural oils are present), soluble in different solvents that D5 is not, it is not flammable, is effective in many formulations, and is cost effective because the majority of the material used is a relatively inexpensive natural oil.

Problems with silicone fluids as a replacement for D4 and D5

While silicone fluids and polymers have a great upside, they feel dry like D5 and they are non-volatile where D4 is, they still have one major disadvantage that must be considered before using them in formulation. The problem arises from the reaction that they are produced from. Silicone fluid is made by an equilibration reaction and, under the correct conditions, the silicone fluid can be converted back to D4 and D5. This means that finished silicone fluids in formulation can produce D4 and D5.

Silicone equilibration reaction

Silicone fluids are most often made by the equilibration of D4 and MM. The equilibration refers to a catalytic polymerisation process that works by a combination of ring-opening and condensation of siloxane bonds. The products of the reaction always include raw materials that are reacted. The excess raw materials can be stripped off in a subsequent step after catalyst is removed. The reaction has been shown to be an equilibrium that can be established with acid, base or other catalyst.1 The reaction pathways are shown in Figure 5. Silicone fluids are stripped of residual D4 before sale, to low non-zero per cent concentrations. However, the fact that the reaction is reversible means that in formulation it does not matter how well the fluid is stripped, there is always a possibility for D4 to be formed if acid, base or other catalysts are present. This observation should make the formulator cautious when using low molecular weight silicone polymers as replacements for D4 or D5. Dimethicone compounds are often made using D4 as a reactant. They have two methyl groups on each Si, as shown in Figure 6. Methicone compounds are not made using D4 as a reactant. They are made by hydrosilylation, whereby methicone compounds conform to the structure in Figure 7 and have only one methyl group on each repeating siloxane unit. Our patent pending approach to replacing D5 is achieved using a silicone that is a methicone, not a dimethicone. The mechanism of action is similar to what was illustrated in Figure 4: the silicone lowers the surface tension of natural oils at low concentrations making the blend have similar properties to D5. Ethyl methicone (Silwax D-02) conforms to the generic structure shown in Figure 8. Since it is a methicone, not a dimethicone, it is not made using D4, so it cannot re-equilibrate into D4 regardless of the formulation. Additionally, ethyl methicones, chosen for low viscosity, spreadability and low surface tension in the organic chosen, have the following properties making them of interest when used in combination with esters or triglycerides as D5 replacements:

•  Low surface tension.
•  Low concentration needed in oils to make silicone like (siliphilic)
•  Does not gel oil.
•  Not flammable.
•  Can be used alone or in properly selected ester.
•  Effective at less than 5% in many oils.
•  Cost effective.

Reference 1 O’Lenick AJ et al. Equilibration of Silicone Fluid. Cosmetics and Toiletries 2004; 119 (5): 89-98