Introduction: Functional silicones are needed in our daily life

Polysiloxanes, commonly known as silicones, are used in industrial products that are essential to our daily lives. Silicones have several characteristics that cannot be replaced by carbon-based polymers, such as high thermal stability, low surface free energy, and excellent biocompatibility, and are used in a variety of product forms such as oils, rubbers, and resins. In particular, in the cosmetic industry, silicone oils are widely used owing to their good repellency and resistance to water, leading to excellent lubricity, a nonadhesive and smooth feel, and improved sensoriality in cosmetics. By introducing substituents to polysiloxanes, it is also possible to control the interfacial activity and adsorption of powders and pigments. In addition, well-defined polysiloxanes have the potential to improve these functions of silicone materials for cosmetic applications.

On the other hand, polysiloxanes have generally been synthesized through equilibrium polymerization using strong acids, but this methodology results in a broadening of dispersity (Mw/ Mn = Ð). Alternatively, catalyst-free ring-opening polymerization (ROP) of cyclic siloxane affords polysiloxane with a narrow Ð. However, an initiating species (alkoxide or silanolate) generally has an ionic form and thus tends to become insoluble with increasing number of initiating points. Hence, the synthesis of well-defined polysiloxanes is important for the fabrication of functional materials. Thus, the organocatalytic ROP of cyclic siloxane reported in the seminal work by Hedrick and Waymouth would be very attractive to the cosmetic industry. Encouraged by this seminal work and recent progress by our collaborator, Dr. Honda, we initiated studies aimed at the preparation of well-defined star-shaped polydimethylsiloxanes (PDMSs) based on binary organocatalytic ROP (BOROP).

The challenge: a highly efficient catalyst system and a solubilizing initiating system

In the model experiment, we used 1,5,7-triazabicyclo[4.4.0]dec-5- ene (TBD), which is a common organocatalyst for ROP. However, a mixture of TBD and the trifunctional silanol initiator (I₃) showed poor solubility in typical ROP solvents, likely due to the mismatch of solubility between highly polar silanols and nonpolar monomers and polymers (Figure 1).

To address this, a previous study reported by Dr. Honda and Prof. Waymouth used a mixture of urea anions and I₃ as an initiating system. Although urea anions solubilized the initiating system, and ROP using urea anions afforded polysiloxanes with a narrow Ð when the conversion of D3 was low, the ROP reached equilibrium when the conversion of D3 exceeded approximately 50%, and the Ð of the resulting PDMS became broad at higher conversions.

For cosmetic applications pursued under the aforementioned expectations, we would like to achieve precision synthesis of polysiloxanes even at high monomer conversions (to prepare high molecular weight and well-defined silicone oil, resins, etc.). Thus, our challenge was to develop a more efficient catalyst system for the ROP of D3.

Fig. 1. Unitaty organocatalytic ROP for the synthesis of star-shaped PDMSs. ROP of D3 with TBD initiated from I3.

Results and discussion – Two types of developed binary organocatalytic ROP systems

We investigated the BOROP of D3 using TBD as an organic base and (thio)ureas as cocatalysts. The synergetic use of TBD and (thio)ureas solubilized I₃ in THF and afforded PDMSs with narrow Ðs until the monomer conversion reached approximately 90%. Moreover, we investigated the potential of proton sponges for the ROP of D3 as an organocatalyst. A proton sponge, 1,8-bis(tetramethylguanidino)naphthalene (TMGN), has similar basicity to TBD, yet TMGN itself was inactive for ROP (Figure 2a). In contrast, the BOROP of D3 proceeded by using TMGN and urea. I would like you to read the paper for more details, but after various investigations, we have proposed a novel activation–deactivation equilibrium between dormant and propagating species in the BOROP system using TMGN and urea (Figure 2b).

Surprisingly, the combination of TMGN and urea resulted in an exceptionally narrow Ð (~ 1.10) even at a polymerization time of 360 min and at a monomer conversion of more than 80%!

Fig. 2. Binary organocatalytic ROP for the synthesis of star-shaped PDMSs. a  Organocatalysts and their pK_as. Chemical structures of TBD, and TMGN and their　reported pK_as in MeCN. b Proposed activation–deactivation mechanism for the present synergistic BOROP catalysis with TMGN and urea.

When we investigated the ability of TMGN to control the molecular weight, which we examined as an additional condition for comparison with TBD, we were surprised to find that the exceptional controllability of Ð. In addition, the fact that we had coincidentally selected urea with only the right acidity for BOROP using TMGN also contributed to the exciting results. Interestingly, the steric hindrance of TMGN around the protons derived from silanols could be the reason for both the inactivity of the unitary organocatalytic ROP system and the reactivity of the BOROP system with acidic urea.

Conclusion

This study aimed to identify an efficient catalytic system for the precision synthesis of star-shaped branched silicones. We found that the combination of urea and either TBD or TMGN can be used to control the MW and Ð even at high monomer conversions. We expect that this research will be applied in various fields and will be the starting point for developing functionalized silicones for cosmetic applications. We are planning to use our PDMS synthesized based on the present BOROP system to develop competitively advantageous functionalized silicone materials.

The full article (DOI:10.1038/s42004-024-01140-3) is available at https://doi.org/10.1038/s42004-024-01140-3