Yes, in Edge, you can capture the PDF correctly after complete the human-machine verification. But in Google Chrome, the human-machine verification will keep looping (circling), specifically, the human-machine verification will circle, there is no confirmation button or after clicking the confirmation successfully, the page will refresh and ask you to reconfirm. It can't be completed, and so you can't get the PDF. But the most troublesome thing is that when these two browsers capture PDF, each human-machine verification is too cumbersome. This situation only occurs in Elsevier (ScienceDirect). Elsevier is almost the largest publisher in the world, which undoubtedly kills the acquisition of a large number of papers.

I am very sorry that you misunderstood my meaning. It is unethical to obtain documents in bulk at once. But what I want to emphasize is the necessity of Elsevier's translator optimization. As the largest supplier, if Elsevier cannot easily obtain PDFs, it will affect a large number of documents (because a large number of documents are published by it).

Sorry for the partial misunderstanding, and glad to hear that the Chrome fix will be deployed soon :-)