Unless you get something completely ancient, just about any scanner will do the trick.
You'll probably end up scanning to either PDFs or TIFF images.
What else you need will depend on where you go from there. If all you want to do is get the book readable on your Sony, I'd use
PDFLRF to take the images from the scan, trim the margins and cut it up into manageable size chunks. That will keep it in an image-based format. It'll look more or less exactly like a photocopy of the book. I've read scanned books on my sony that way many times.
The next option would be to try use OCR software (optical character recognition) to try to "read" the words of the books. How successful this will be will vary depending on a lot of factors: the quality of the scans, the quality of your OCR software, the complexity of the book (e.g., is it plain text, or is there a lot of mathematics and charts, etc.?) Even with good software, there will be mistakes you'll have to correct by hand. At the end of the process, however, you'll have something you can export to any format, and use things such as the zoom/reflow feature of your reader (and pick the fonts, etc.)
Most scanners will come with OCR software, but they may not be very good. If you have money to spend, you can get ABBYY fine reader, which is quality OCR software. If not, there are free alternatives like tesseract. (sp?)