Hi Alan,
I think it's a great idea!
You have few options here:
Archive:
AFF --> Cloud Tiering --> BLOB
Cloud Tiering can tier cold blocks from the AFF directly to Blob, but the meta data and hot data will be kept on the AFF performance tier. It will enable you to expand your AFF capacity without any need for additional hardware.
DR / Secondary copy:
AFF --> SnapMirror --> CVO --> FabricPool --> BLOB
There's no need to use it as cloud tier, you can SnapMirror the data from the AFF to CVO and then enable FabricPool with "All" /"Auto" tiering policy and the cold data will be tiered directly to BLOB.
Backup:
2) AFF --> Cloud Backup Service --> BLOB
It will create an independent copy on BLOB for backup