It might be worth it to see if you can rename one of the columns temporarily to something different, then run your ADF pipeline and see if the issue still persists. If it doesn't, then rename it back.
Check for views, custom fields, or other configurations in Salesforce that might cause the data to be treated in this manner.
Some platforms treat column names as case-sensitive, while others do not. If ADF is pulling Salesforce data in a case-sensitive manner, but Salesforce is storing it in a case-insensitive manner, this could cause conflicts. Make sure to account for case discrepancies.
Instead of using Lookup directly on the Salesforce data, you can create a Data Flow. In the Data Flow, you can rename columns, handle transformations, and even manage case issues. Once the columns are properly renamed or managed in the Data Flow, use the Lookup on this corrected dataset.