---------------------------- Forwarded with Changes --------------------------- From: Henson, Sam::(HENSNS) at ~FABRIK Date: 9/3/96 4:40AM To: Andrew P Magee at Rogers-MCD Subject: Re: FAB:Cupric Chloride Etchant ---------------------------------------------------------------------- --------- From: Henson, Sam [log in to unmask] (Henson, Sam) Date: Tue, Sep 3, 1996 4:40 AM Subject: Re: FAB:Cupric Chloride Etchant To: Magee, Andrew P There is a design problem with some equipment. There is not enough contact time between the chlorine gas and the liquid. If you add the gas in excess of 500 Pounds per day flow rate then it liberates out. I have some methods dealing with this but I would need to know which model machine. ---------- ---------------------------- Forwarded with Changes --------------------------- From: RSedlak::(RSEDLK) at ~FABRIK Date: 8/13/96 7:44PM *To: 27=TechNet::[log in to unmask]::25=SMTP at ~FABRIK Subject: Re: FAB:Cupric Chloride Etchant --------------------------------------------------------------------- - --------- An extract from the IPC TechNet. From: [log in to unmask] Date: Tue, Aug 13, 1996 7:44 PM Subject: Re: FAB:Cupric Chloride Etchant To: TechNet Chlorine is a gas, which is marginally soluble in water. The more salts, all other things being equal, such as pH, the less soluble the Chlorine. So, you could be creating Chlorine when the Copper is low, and it is soluble enough to not be given off. One key question is do you monitor your ORP, that is to say, actually keep a continuous record of it?? If you are getting serious spikes in an ORP chart, you are creating Chlorine, whether you smell it or not. Another factor, that could be contributing is the viscosity, and thus mixing speed of the etchant. If the Copper is low, the mixing speed of the components is higher, and this will allow the Cuprous ion to come into contact with the Chlorine gas, and react with it, and make it go away, faster. One of the factors that I have seen personally that can contribute seriously to this problem is the type (and horsepower) of the pump making the additions. If the pump is too powerful, and ANY centrifugal pump I have ever seen is too powerful, It will seriously over-add before the ORP sensor tells it to shut off. Best type of pump is a bellows pump, and they should be sized so that they are running a lot of the time that panels are going down the line. Never less than 20% of the time, and more is better, up to even as high as 70% of the time. This ensures slow addition, so that the reactants can circulate around enough to find the Cuprous ion, and not have time to say, "Hey, can't find no Cuprous, I'm history!", and escape to the atmosphere. Also, speaking of mixing, if you are running more than one chamber, you should have some serious horsepower mixing the two sumps. This is best done by pumping from the last chamber to the first, and cascading back. I am going to make a rough estimate that you want one HP of pump circulating for every etcher sump in a line. And that is an absolute minimum. Also, if you have more than one etcher sump, you may want to add Chlorate and HCl at more than one point. If you would like more free opinions, which some have said are worth almost what you pay for them, e-mail, or call. Rudy Sedlak RD Chemical Company 415-962-8004 From: Gurian, Marshall Date: Tue, Aug 13, 1996 3:03 AM Subject: Re: FAB:Cupric Chloride Etchant To: TechNet I have two guesses Re: fuming with high copper levels. Guess 1: ORP reading becomes less sensitive as copper levels get higher, requiring more Cu(I) to trip ORP adder point. This may encourage a slow control response, allowing overdosing of oxidizer. Guess 2: Conductivity controllers are set up on a narrow response to conductance in a very specific mixture where the only significant variable is HCl. High Cu may add to the conductivity and cause acid to fail to add. This will definitely throw the system out of balance, either causing overadd of oxidizer (liberating gas when some HCl is added) or some other phenomenon. I may be able to be more complete with more detailed operation info. Marshall Gurian Coates ASI e-mail: [log in to unmask] Date: Tue, Aug 13, 1996 3:40 PM Subject: RE: FAB:Cupric Chloride Etchant To: "[log in to unmask]" Look in Coombs, Handbook of Printed Circuits 3 rd edition on page 14.15 at paragraph 14.4.3 for the chemical reactions. I think you will find that the amount of copper is not as important as the state of the copper, if there is no more cuprous ions to be converted to cupric then the chemistry you are still adding is there to react. Or if you are really putting through the volume you could be overwhelming your cooling system allowing the temperature to rise and liberating gasses from solution. In short you have excess chlorate and excess hydrochloric acid and should anticipate chlorine gas and hydrogen chloride gas as a minimum as a result. If the solution is bright green the correct response is in fact to add copper clad material. This converts some of the cupric back to cuprous and that regenerates knocking down the excess chemistry. ---------- From: [log in to unmask] To: [log in to unmask] Subject: FAB:Cupric Chloride Etchant Date: Tuesday, August 13, 1996 11:01AM We etch inner layers using Cupric Chloride Etchant. We control the chemistry with ORP and Conductivity probes. The ORP adds Sodium Chlorate and the Conductivity controls HCl adds. We have experienced gas generation (chlorine or HCl fumes) when both the ORP and Conductivity are in control. The gas/fume release occurs when the copper gets as high as 33 - 35 ounces per gallon. Standard procedure is to feed copper into the system when a chemical out of balance situation results in gas generation. Clearly in this case this would be the wrong thing to do. Can anyone explain the chemical reaction behind a release of gas/fumes when the copper is high? From: [log in to unmask] (Henson, Sam) *************************************************************************** * TechNet mail list is provided as a service by IPC using SmartList v3.05 * *************************************************************************** * To unsubscribe from this list at any time, send a message to: * * [log in to unmask] with <subject: unsubscribe> and no text. * ***************************************************************************