Scaling predictions for bottle‐brush polymers with a rigid backbone and flexible side chains under good solvent conditions are discussed and their validity is assessed by a comparison with Monte Carlo simulations of a simple lattice model. It is shown that typically only a rather weak stretching of the side chains is realized, and then the scaling predictions are not applicable. Also two‐component bottle brush polymers are considered, where two types (A,B) of side chains are grafted, assuming that monomers of different kind repel each other. In this case, variable solvent quality is allowed. Theories predict “Janus cylinder”‐type phase separation along the backbone in this case. The Monte Carlo simulations, using the pruned‐enriched Rosenbluth method (PERM) give evidence that the phase separation between an A‐rich part of the cylindrical molecule and a B‐rich part can only occur locally. The correlation length of this microphase separation can be controlled by the solvent quality. This lack of a phase transition is interpreted by an analogy with models for ferromagnets in one space dimension.
