there's no option in database to change characterset , if the database is empty recreate with what you want otherwise check the below :
788156.1 - AL32UTF8 / UTF8 (Unicode) Database Character Set Implications
260192.1 - Changing the NLS_CHARACTERSET to AL32UTF8 / UTF8 (Unicode)
select * from v$nls_parameters where parameter like '%CHARACTERSET%';
to see what characterset you are using ?
also check the below to define characterset for column :
Yes you can change the character set of a database. This is not a straight forward issue.
See [url http://docs.oracle.com/cd/E11882_01/server.112/e10729/ch11charsetmig.htm#g1011430]11 Character Set Migration chapter in Globalization Support Guide
to change collating sequence for character values in session level see [url http://docs.oracle.com/cd/E11882_01/server.112/e25513/initparams153.htm#REFRN10127]NLS_SORT.
For character set change to AL32UTF8, use Database Migration Assistant for Unicode (http://www.oracle.com/technetwork/products/globalization/dmu/overview/index.html). Its documentation has interesting information about the migration process.
Oracle does not currently have collations assigned to data (columns, database). Collation is either specified dynamically through the NLS_SORT/NLS_COMP session parameters, or specified explicitly in calls to selected collation-sensitive functions: NLSSORT, NLS_UPPER, NLS_LOWER, and NLS_INITCAP.