Is there any way we can cutdowm the time by dividing to multiple queries and so on? because we can't afford this much of down time in PROD the max is 1 Hr.
How did you move them?
For operating on big amounts of data, it is often a good idea to split up the work in batches to keep down the growth of the transaction log. But it is important that you follow the clustered index for the batching, else performance will be horrible. I would try with a batch size of 5 million rows, but it is a number that is subject to fine-tuning.
But to be honest, I am not sure that this alone will cut down the execution time with a factor of four.
A more aggressive idea is to insert the batches in parallel. Not all in parallel, but a few of them. This certainly will go as advanced.
And even more advanced is to do a trickle load to the new table while keeping the old table live - so you need to take care of the updates.
Then again, maybe you can get away with CREATE INDEX WITH (DROP_EXISTING = ON, ONLINE = ON)?