You need to use the BIGINT datatype. Recommend you lookup BIGINT in Books Online to see what it is and, possibly, lookup the other datatypes and read about them, as well. Datatypes are almost as important as the data itself.

So far as bringing in NULL's instead of blanks, I know virtually nothing about DTS or SSIS because I don't use them... I do use BCP and BULK INSERT extensively. Nulls will only occur if the data has two delimiters adjacent to each other with nothing inbetween. The -k parameter may help in that area but no guarantee's because I've not used it either. The reason is that I never import directly into a final table... too big a risk for too many reasons. I always import to staging tables and process from there. It's actually a pretty good habit to get into.