Arguments passed to the classification or regression routines are included in the dots parameter.

As you want to include minbucket, parameter control should be included inside train.

As example:

library("caret")

train(Kyphosis ~ Age + Number + Start, data = kyphosis, 
      method = "rpart",
      tuneGrid = data.frame(cp = c(0.01, 0.05)),
      control = rpart.control(minsplit = 10, minbucket = 5))

If you want to tune minbucket you will have to do it manually.