Table partitioning does not require a clustered index but I suggest one create a clustered index on SQL Server tables unless there is a specific reason to organize the table as a heap.
The clustered index does not need to be a primary key nor does it need to be unique. SQL Server will implicitly add the partitioning column as a clustered index key for non-unique clustered indexes if it's not already an explicit key column.
A non-unique clustered on your date column will be especially appropriate If data are incrementally inserted into the table by date and queries specify date ranges in WHERE
clauses without additional criteria. Additional non-clustered indexes may be needed to optimize queries for other criteria.