[Table of Contents]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ARSCLIST] Ampex 456 - tip of the iceberg...chemistry

Hi, Tom,

The problem I'm having is connecting the dots. I'm talking with chemists who understand these things far better than I do and there is some indication that some of the preliminary conclusions in Richardson's patent are perhaps closer to what really happens than what Ampex says in their patent or Bertram/Cuddihy in their paper. It's all very complex and even the measurement techniques are open to interpretation.

What we know from a large body of knowledge is that baking/incubation works for the vast majority of back-coated tapes. I am less sure we know how/why it works precisely. We also know that with one or two exceptions, it doesn't solve any problems on non-back-coated tapes.

We can create a hypothesis that back-coating in contact with binder has a catalytic effect on the degradation of the binder. Additionally we have a preliminary, back of the envelope level, study that suggests that there is truth in this, thanks to John Chester: http://stuff.jkc-lab.com/sticky-shed/

We do see gooey deposits of shorter molecular weights than the original binder -- but longer chains than the typical lubricants. In a filled matrix, I am led to believe that the shorter molecules do not recombine. Now what exactly happens is harder to determine. There are many factors going on here.

One mental model that might work (according to some) is thinking of the tape binder as a sponge. The sponge is the cross-linked polyethelene-polyurethane binder with the oxide carried in it. The holes in the sponge hold the lubricant. When the lubricant is thrown into the blender before coating, it doesn't really bond with the binder chemistry, but is trapped in the matrix. One theory has the lubricant being pushed out momentarily under the pressure of contact.

But, what happens to the shorter fragments during baking/incubation? I don't know yet. Richardson has stated that when he sent tape to an analytical lab, they said they hadn't seen evidence of hydrolysis. We pooh-poohed that at the time, but if we look at a paper from 1984 that Ric Bradshaw did where he used the lubricant extraction to calibrate the binder extraction, the results are different from the 1982 Cuddihy and Bertram paper. Also, the typical lubricants that permeate the binder matrix would repel water.

Remember, Bradshaw is the one who salvaged the Challenger tapes after others said they were non-salvageable.

There's a 1000+ page book, "The Tribology and Mechanics of Magnetic Storage Devices" 2nd edition by Dr. Bharat Bhushan who is at Ohio State University and is a former colleague of Bradshaw's at IBM. Bradshaw/Bhushan et al are the team behind LTO and it's proprietary predecessor (3480) data tape formulation. This has had an excellent track record.

I don't think (m)any media-centric tape manufacturers analyzed the tape issues as closely as the IBM lab. There is a paper at the IBM Research Journal by Bradshaw that is a help.
http://www.research.ibm.com/journal/rd/302/ibmrd3002H.pdf

Another article that is of a bit of interest (with some minor audio-related historical errors up front) is www.research.ibm.com/journal/rd/474/bradshaw.pdf and it provides the history of data tape.

As I said, I'm learning about this and I don't have answers. I know others who don't have answers as well. The thought is methodical measurements and lots of questioning and brainstorming will help us create a better picture.

Imagine this scenario:
Tape is made with proprietary formulations
These formulations were well understood in terms of the then-current state of the art
The production lines have some variations
The variations are now studied deeply and certainly not to the level of today's polymer chemistry
Today's scientists can only evaluate snapshotted degrading tape as even NOS sealed
tape is degrading
We can't get back to what really new virgin tape looked like to compare and we
don't really know the formulations of the tape nor how these were really
applied/modified in production. Lots of production aspects can modify chemical behaviour

In some circles, including people who made the tapes, they were not considered archival. 10-15 years was a typical life expectancy for a tape--at least in some circles. Have many of them done better?Yes! Absolutely! But is it harder to get high quality off aging tapes?Yes, for some.

Analog magnetic recording was a rather imperfect science that took the world by storm. I think we all have our work cut out for us. I suspect we'll know more in a couple of years, including whether it's safe to freeze tapes to preserve them as is done with film and paper. Yes, I know what the standards say, but at this point it is necessary to consider other options to what we're doing. Remember when Corbis bought out some of the NY Stock Photo Agencies (Black Star?)? Henry Wilhelm was behind the move to the Pennsylvania mine and freezing conditions. Why? Even if they had started digitizing the entire collection right then, if the material had been kept under the then-current conditions some of it would be gone before its turn would come up for digitization.

Yes, I'm speculating, although I prefer to say formulating hypothesis that need testing and evaluation <smile>. I am also standing on the shoulders of people who have/are researching this material. A lot of very good work has been done into arcane and minute aspects of the tribology and mechanics of magnetic media--at least 1000 pages of it IN SUMMARY! Connecting the dots is a very hard part of this.

Some interesting snippets.

A mixture of fatty acid ester lubricants that was a common lubricant in tape freezes at about 20C.

Other charts show the modulus of elasticity as a two-horned curve with peaks around 0C and 20C.

Pressure in the tape pack contributes to SSS severity (and other items like pinning of the tape to the backing below in non-back-coated tapes). This varies throughout the wind.

What we've been calling "Loss of Lubricant" that causes squealing may not be, it may be a flattening of the asperities (or peaks) of the oxide of the tape increasing head-to-tape contact area and exacerbating stiction.

Archivists want help predicting the lifetime of specific tapes. This is an almost impossible task. I know it sounds self-serving, but the more I learn the more I really think that the "tape on the shelf" model is too high a risk and all tapes we care about should be migrated to managed digital stores within the next few decades. That is the safest approach. Anything short of that is gambling.

I used to think SSS would be a non-issue and we could bake anytime, but in the seven years I've been doing this seriously, it seems the required baking times are slowly increasing.

Anyway, Tom, I wish I had simple answers for you and everyone on this list. I think the best answer to the problem is what you and I are doing. We have to hope that once we create digitial files, that the clients' IT departments will preserve them along with the email and the financial records.

I don't think that there is a foaming reaction, but one of the chemists is on this list and perhaps he has a more specific answer to your question. The molecular weights of the molecules as shown by Fourier Transform Infrared Spectroscopy typically show degradation products in roughly the 5,000 g/mol region, with lubricants being in the under 500 g/mol region and the binder being in the 50,000 g/mol region (I'm simplifying, but this is illustrative).

I don't know if the behaviour of these 5,000 g/mol particles is well understood during the baking cycle. I also left my shorter message appended to this thread, but snipped a bunch of the middle stuff.

Cheers,

Richard

At 06:28 PM 6/16/2006, Tom Fine wrote:
Another interesting point in this thread is the gentleman from Norway saying that despite constant temp/very low humidity storage, SSS rear its head reliably. Richard Hess has speculated and researched a lot about this topic. As I understand it, it's not clear that SSS is a purely ambient-humidity mechanism and may indeed have nothing to do with moisture. I'd like to know, has anyone ever done detailed analysis of the "goo" on a sticky tape and compared its makeup to the binder material on unstuck or baked tape? I'd like to know if this is some sort of gas reaction with some chemical in the binder, leading to a foaming action, rather than having to do with H2O levels in the air. Hey, I don't pretend to know chemistry beyond 10th grade regents course, so please let me know if I'm speculatin' out my butt. 

arclists@xxxxxxxxxxxxxxx 6/16/2006 8:42 AM >>>
At 06:51 AM 6/16/2006, Lars Gaustad wrote:

Hi,
I have got a 2" Apmex 456 that is really sticky.
I has been stored at 8 C 35% RH for 8 years,
which makes me believe that the stickyness is not related to
hydrolysis
(SS),
as such storage should rejuvenate the tape just as well as baking
will.

Any suggestions? 

Baking should still rejuvenate it if past history is any guide. I'm
looking for a good explanation of precisely why baking works, but
I've been told by people who understand these things that the
generally accepted explanation is not the whole story.

There does appear to be mounting evidence that there is interaction
between the back coat and the oxide binder system. Again, no answers
at this time.

I'm not being mysterious, I'm slowly studying this.

Thanks for a really useful data point.

Many of the people working on this are not on this list and I'm
taking the liberty of passing on your observations to them.

Cheers,

Richard


lars gaustad
preservation adviser
national library of norway
www.nb.no

Richard L. Hess email: richard@xxxxxxxxxxxxxxx
Aurora, Ontario, Canada (905) 713 6733 1-877-TAPE-FIX
Detailed contact information: http://www.richardhess.com/tape/contact.htm
Quality tape transfers -- even from hard-to-play tapes.

[Subject index] [Index for current month] [Table of Contents]