Tim

Sorry, but I cannot believe that there is sufficient energy available to break down any
of the covalent bonds within an epoxy molecule at a hundred and odd degrees. Let us
analyse what happens when an epoxy "cures". In the prepolymerisation process, taking
the simplest epoxy resin chemistry, the epoxy group in the epichlorohydrin is replaced
by a hydroxyl group sticking out to the side roughly half-way between bisphenol groups.
It is this reactive group which allows crosslinking during curing and is, I believe
released, to accelerate lateral, three-dimensional, polymerisation in a chain reaction.
This starts, more or less at random, throughout the prepolymer/linking chemistry. The
"lumps" of molecules grow until either there is no more chemical to react with or it
meets a neigbouring molecule competing for what is left. As an analogy, it is rather
like a well fitting dry-stone wall with a certain amount of physical interlocking
between the stones. When the temperature rises, each molecule expands as well it may
and some of the interlocks reversibly spring open, as the "stones in the wall" are
pushed apart, weakening the whole stucture and opening it up to chemical
absorption/adsorption (cf. Zado effect). It is therefore a purely physical and not a
chemical effect, if my understanding is right. The important point to remember is that
the Tg is not like a eutectic melting point where there is a phase change. If the Tg of
a resin is, say, 130 deg C, it is not vitreous at 129.9 and rubbery at 130.1. It is a
gradual process over +/- 10 or 15 deg C around the published figure and full
"rubberiness" will not occur below, say, 142 deg C in my example. This can be shown by
plotting the coefficient of linear expansion against temperature: it increases rapidly
over the range and does not form a sharp elbow, so that it may be 20 times higher above
the Tg than below it (no more "interlocking stones in the wall"). This is why barrel
cracking of PTHs may occur, as there is no mechanical restraint in the Z axis of epoxy
laminates other than the relatively non ductile copper which has a much lower TCE than
the epoxy over the Tg.

Also, I have known for many years that the Tg of a given resin varies with humidity: it
can drop by as much as 20 deg C with an extra 1% absorbed moisture. A recent posting in
this forum also stated that it can drop with absorbed organic solvents, as well, which
I did not know.

I am not an epoxy chemist - or a chemist at all, I hasten to add - but this simplistic
explanation may help you to understand what happens and that the chemists will forgive
me trespassing clumsily into their field with it.

Brian

Timothy Reeves wrote:

> My understanding is that heating FR-4 above its Tg does permanently change
> the material (unlike the same process for a thermoplastic, where the
> hydrogen bonds re-form, some covalent bonds in a thermoset like epoxy are
> irreversibly broken upon heating above Tg), but my own experience is that
> the change for any one excursion is small, and that you can do it a few
> times without significantly changing the properties of the laminate.
> Tim Reeves
>
> -----Original Message-----
> From: michael [mailto:[log in to unmask]]
> Sent: Wednesday, July 12, 2000 7:15 PM
> Subject: Re: Glass Transition Temperature (Tg) Questions
>
> Epoxies are thermosetting. Surely this process is not indefinitely two-way?
> Or
> should I say: beyond what temperature is the process no longer two-way (and
> what
> is the technical term for this temp.)?
>
> Michael Fennema
>
> ##############################################################
> TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
> ##############################################################
> To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
> the body:
> To subscribe:   SUBSCRIBE TECHNET <your full name>
> To unsubscribe:   SIGNOFF TECHNET
> ##############################################################
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
> information.
> If you need assistance - contact Keach Sasamori at [log in to unmask] or
> 847-509-9700 ext.5315
> ##############################################################

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################