I don't think it has anything to do with '@' -- if you sniff while you try to download the flie in PB -- you get:
Regardless of using '@' or %40, the download fails.
HTTP/1.1 403 Forbidden
and
... but your computer or network may be sending automated queries. To protect our users, we can't process your request right now.
Perhaps it has something to do with the client browser is identified as (or lack thereof of one...)
The url is not a url to an image -- it's a url that is processed by the server and then image
data is sent back. Since no browser is identified in the stream, google's servers may be declining the request (just a guess, who really knows what google does?)
ANyway; if you 'hand code it' and include a browser declaration ,I bet it will work. (FWIW: wget fetches the file fine, wget is available on most platforms)