I was told by the owner of a lens company that the glass was badly fitted and that the cement was used to 'fill' too much, over time the strain of the element finally pulling apart caused the crystallization.
I'm more familiar with LF optics and Goerz lenses which are very well made rarely rarely have separation while many other makers lenses do.
Although I cannot dispute the comments you attributed on a general basis, I must admit I have a bit of difficulty translating them directly to the Tele Rollei issues which are well documented. Firstly, they're a Zeiss lens, and, whilst nobody's perfect, Zeiss quality control was usually pretty good at the time if you take a look at many of the other lenses they were making including those fitted to 2.8 Rolleis. Secondly, a good sample of the 135mm Sonnar from the Tele is capable of stunning image quality. I don't pretend to have any special knowledge of or expertise in optical cementing, however these two points don't really gel with problems emanating from poor assembly or substandard lens grinding do they? Which doesn't mean it's not possible, of course.
It's worth noting that some other Zeiss lenses from around the same time have the same type of fault. Occasionally a 2.8 Planar lens fitted to the TLRs will be found with separation (nowhere near as regularly as the Teles Sonnar). I'm a keen Contaflex SLR user, and it's not unknown for the optional Pro Tessar lenses to have separation problems also. The 85mm versions, in particular, are extremely prone to it, and it is the exception rather than the rule to find one with no separation at all, but the 35mm and 115mm Pro Tessars are not immune, either. My own 85mm f/4 Pro Tessar has a few small bubbles on the edge of the glass, but still performs very well despite this.
So if the root cause of the element separation that manifests itself in these older Zeiss lenses is as per your suggestion, Zeiss must have been guilty of sub standard assembly or loose tolerances in relation to several lens types, not just the 135mm Sonnar.
I was under the impression that deterioration of the synthetic cement used to join the bonded groups over decades was the root problem. This stems from discussions I've had with a motion picture lens specialist who is also a Contaflex user. As well as being a very competent lens specialist, he also separated and rejoined a Contaflex lens, and thus has first hand experience with some of the affected lenses. I recall him telling me it took weeks of soaking in Milsolve, before the Zeiss cement finally gave up its grip on the joint involved.
Of course some of the above is conjecture, but sevo's previous comments also echo my understanding that Zeiss were at the forefront in adopting synthetic alternatives to canada balsam (although I wasn't aware of the patent details). So I was simply of the understanding that the issues some of the tele lenses have now, is a consequence of Zeiss being the first in the field to move to synthetics. Naturally, I'd find any solid information about the issue to be of great interest, but I hope my comments assist.
Regards
Brett
