Metal Embrittlement in galvanized bolts

[Ductmate]

We use 1/4"-20 Grade 5 galvanized bolts in one of our products. I don't know if they're electroplated or hot-dipped. Recently we've had problems with the heads breaking off during installation. When I do a pull test on them, (at 2"/min) they are breaking around 2500-2800lbs - sometimes even less. Normally my grade 5 bolts break around 4500lbs or more. The problem bolts do not seem to elongate before breaking, but snap off clean.

Doing some net research of my own, it seems that the problem could be coming from either liquid metal embrittlement or hydrogen embrittlement, but i don't know how to tell for sure.

My bolt supplier is ISO9002 and tests all incoming raw mat'ls to make sure they are the proper composition. I don't know if they do any testing after the galvanizing is applied.

My questions are:
What is the most likely cause of this?
How can I tell for sure if it is from the galvanizing?
Would a zinc chromate finish be less likely to cause problems?
Any other advice?

Thanks,
Vince Bloom

 

[Metalguy]

Very briefly, this sounds like it could well be H related cracking from electroplating.

First I'd check the hardness, then determine if its electroplated. You can also try to bake out H for 24-48 hrs. at 375 def F.

 

[Paskenell]

Vince

I would like to address a common misconception about ISO 9002 certified bolt suppliers. You stated your supplier "tests all incoming raw mat'ls to make sure they are the proper composition". In my 20 years in the fastener trade, I have found this usually means they look at the chemical certs that have been sent from the metal factory to the manufacturer.

ISO 9002 means your supplier has certain inspection procedures in place. The actual quality of the fastener will only be as good as the procedures require them to determine. Having worked for both ISO 9002 suppliers and those that are not, I have never been impressed with ISO 9002, except from an administrative point of view.

To actually test the proper composition would involve an extreme expense and time involved with sending the parts to a lab that is usually done at the customers request and expense. While the composition is probably correct, papers that would be more helpful here might be the heat treatment papers. Also did your bolt supplier re-test these parts after they performed a secondary operation (plating) on a thru hardened fastener?

My sugestion would be to return the parts to the bolt supplier and if they are failing outside their requirement, tell them to fix "their" problem. If I am wrong and they are truly testing the composition of the material, then obviously they have a qualified metallurgist on staff and available to access your problem.

As to possible solutions, mechanical galvanized might be a consideration for the future, depending on your application. My hunch is Metalguy has put you in the right track.

Just my penny's worth.

 

[Ductmate]

The supplier knows it's their problem and they're replacing them at their cost, so determining who's at fault isn't the issue. I don't know if they're sending their raw mat'l out for testing, or just going on the certificate of analysis.

I also don't know if what sort of QC is done on the heat treating.

My concern here is that I've never dealt with a problem like this and i want know how to determine what the cause is and any ways to test for it in case it comes up again.

As for ISO 9002, i wasn't implying that clears them of all QC sins, only that they have a system in place for it.

Vince

 

[kenvlach]

Hi Ductmate,
Metalguy and Paskenell made some good points. I just want to clarify a few items. Agree that the problem could be in the initial material, the initial heat treatment (probably a quench & temper) or hydrogen embrittlement (HE).

The HE would be the result of acid pickling cleaning prior to hot-dip galvanizing. The hydrogen then gets ‘sealed in’ by the zinc coating, and thus requires a special ‘bake-out.’ Electroplating of zinc on high strength steel bolts is not permitted by ASTM A325 (~ SAE Grade 5) to avoid intensifying the HE problem, only hot-dip and mechanical plating are allowed. ASTM A490 (~Grade 8) prohibits both electroplated & hot-dip galvanizing; only mechanical plating of zinc is allowed. [don’t have the SAE specs, but presume similar to ASTM. Manufacturers generally try to accommodate both specs. so can cross-list their products.]

You inquired as to the effect of zinc chromate on this problem. Chromating is used to create an anti-corrosion coating. It does not affect the bolt’s strength. However, if the bolts were hot-dip (or electro)galvanized and subsequently given a hydrogen bake-out, a brief acid cleaning would be necessary to remove the oxide formed during the bake-out prior to chromating. It is not permitted to heat the chromate, so this would give a bit of HE effect [sounds like a Catch-22 situation, doesn’t it?].

Mechanical plating seems to be more and more common, as it totally avoids the HE problem and gives a more uniform thickness than hot-dip. (And it certainly avoids the formation of a brittle intermetallic at the zinc-steel interface.) It is also readily chromated without need for a pre-acid cleaning.

 

[CoryPad]

The cause of fracture seems to be due to either Hydrogen Embrittlement or Liquid Metal Embrittlement. One way to distinguish between these is to use a Scanning Electron Microscope with Energy Dispersive X-ray Spectrometry to view the fracture surface. If significant zinc is on the surface, Liquid Metal Embrittlement is likely. If the screws are electroplated, hydrogen is generated during the plating process, and the hydrogen is absorbed by the steel screw. Acid cleaning can add more hydrogen, but usually in much lower quantities. Some recommened practices that you may want to include on your drawings and purchasing documents include:

SAE/USCAR-5 Avoidance of Hydrogen Embrittlement of Steel
SAE/USCAR-7 Deembrittlement Verification Test
ASTM F1940-01 Standard Test Method for Process Control Verification to Prevent Hydrogen Embrittlement in Plated or Coated Fasteners
ASTM F1624-00 Standard Test Method for Measurement of Hydrogen Embrittlement Threshold in Steel by the Incremental Step Loading Technique
ASTM F519-97 Standard Test Method for Mechanical Hydrogen Embrittlement Evaluation of Plating Processes and Service Environments
ASTM F1459-93(1998) Standard Test Method for Determination of the Susceptibility of Metallic Materials to Gaseous Hydrogen Embrittlement
ASTM B850-98 Standard Guide for Post-Coating Treatments Steel for Reducing the Risk of Hydrogen Embrittlement
ASTM B839-94 Standard Test Method for Residual embrittlement in Metallic Coated, Externally Threaded Articles, Fasteners, and Rod-Inclined Wedge Method
ASTM B849-02 Standard Specification for Pre-Treatments of Iron or Steel for Reducing Risk of Hydrogen Embrittlement
ASTM F2078-01 Standard Termninology Relating to Hydrogen Embrittlement Testing
ASTM G142-98 Standard Test Method for Determination of Susceptibility of Metals to Embrittlement in Hydrogen Containing Environments at High Pressure, High Temperature, or Both

And one ASTM specification for a coating that reduces or eliminates hydrogen embrittlement concerns is:

F1428-92(1999) Standard Specification for Aluminum Particle-Filled Basecoat/Organic or Inorganic Topcoat, Corrosion Protective Coatings for Fasteners

The SAE/USCAR documents are available from SAE International at:

http://www.sae.org

The ASTM documents are available from ASTM International at:

http://www.astm.org

 

[rmp319]

Kenvalch,

You mentioned ASTM does not allow for galvanizing of bolts. I have always thought A325 were allowed. ASD Ninth Ed. 5-291 states "ASTM specifications provide for the galvanizing of A325 bolts but not A490. Galvanizing of A490 bolts is not permitted.". I am aware of the problems with galvanizing HSB's but still see it done frequently.

Is the AISC overlooking something? Please help.

RP

 

[Metalguy]

rpoche,

Kenvlach said no electroplating, but hot-dipped zinc is allowed!

 

[rmp319]

Metalguy,

I ran the lines together, sorry. Thanks for the answer.

 

[kenvlach]

Zinc is commonly applied by 3 processes: electroplating, hot-dip galvanizing and mechanical plating. Mechanical plating is done at room temperature by tumbling parts and zinc powder together with high kinetic energy. There is no chance of hydrogen embrittlement.

Electroplated zinc is not allowed for ASTM A325 or A490.
Hot-dipped 'galvanized' zinc is allowed for A325, but not for A490.
Mechanically-plated zinc is (I believe) allowed for all grades. SAE Grade 8 bolts are commonly mechanically-plated with zinc (and often, chromated to a yellow-gold color). I presume likewise for A490 from the last sentence of the following:

From Manual of Steel Construction, LRFD, 2nd Edn., Vol.1, p. 6-402 to 6-403 (1998), re A490:
“if the bolt is hot-dipped galvanized, a hazard of delayed brittle fracture in service exists because of the real possibility of introduction of hydrogen into the steel during the pickling operation of the hot-dip galvanizing process…Galvanizing of A490 bolts is not permitted. Because pickling and immersion in molten zinc is not involved, galvanizing by the mechanical process esssentially avoids potential for hydrogen embrittlement.”

The AISC Commentary is not totally consistent in their usage of ‘galvanizing.’
There is a new version of ASTM A490:
“A490-02 Standard Specification for Structural Bolts, Alloy Steel, Heat Treated, 150 ksi Minimum Tensile Strength.” Anyone with a copy please verify if mechanical plated zinc is allowed.

I do not believe that liquid metal embrittlement is involved, if the only liquid Zn exposure was during a brief hot dip. Zinc LME of steel requires long term exposure at elevated temperature – ASM Handbook Vol. 13, Corrosion, p. 184 (1987). If Zn is found near a crack tip, it is likely that the Zn entered a pre-existing crack.

 

[Rich2001]

Recently, (past 9 months) some of the Galvanizers in the United States have begun to add a small amount of TIN to the molten zinc bath. The tin make the spandles brighter, i.e, more architecturally pleasing. However, this appears to be causing Liquid Metal Embrittlement.

Wire, chains and other small diameter parts seem to be most affected. Because of this reasearch is now being done in Canada on the affects of TIN on the material properties of galvanized steel. I have witnessed cracks that have formed in ASTM A572 steel in the Zinc (with 0.17% tin) bath. The steel was 100% inpected with Magnetic particle prior to galvanizing, Cracks were then examined with SEM. High concentration of tin was found at the grain boundaries.


Get the historical Bath concentrations. Look at the concentration of the Tin, Lead and Aluminium. The galvanizer should make this information available to you.

Contact Thomas J. Langill, Ph.D., Technical Director - American Galvanizers Association, tlangill@galvanizeit.org

American Galvanizers Association

email: aga@galvanizeit.org

Toll-Free Technical Support for Specifiers: 1.800.HOT.SPEC [800.468.7732]

Local Phone: 720.554.0900

Fax: 720.554.0909

Address:
American Galvanizers Association
6881 South Holly Circle, Suite 108
Centennial, Colorado 80112

http://www.galvanizeit.org/

 

[Ductmate]

Thanks to everyone for their input.

I tried baking them at 375F over the weekend, but they've shown no increase in strength.

I haven't heard back from the supplier yet as to what they plan to do to fix the problem on future products. I'll leave an update if I discover any further info.

Thanks again,
Vince

 

[Metalguy]

Is it possible to take a good quality closeup photo of a fracture surface and post it here?

 

[Ductmate]

Ok, I took some pictures of the breaks and you can see the grain/fractures pretty well. I don't know how to post pics on this forum, so I can mail the pics to anyone interested, or if someone tells me how to post the pics, i'll put them here.

If you want them mailed directly to you, send me your address at vbloom@ductmate.com

Thanks,
Vince

 

[Rich2001]

Picture and Comments are Vince's

pictures 61, 62 & 65 are of a bolt that was baked at 375F for app. 60 hours. Note how the metal seems to flake off in pieces.

pics 66, 67, 72, & 73 are of a bolt that broke in two places. First, it sheared the threads off, but left them intact inside the (undamaged) female threads. when I removed that piece, there was another piece snapped off underneath that! I can understand it shearing the threads off, but the female threads should have prevented it from breaking that far inside. this bolt was also baked.

pics 75 & 77 are of an unbaked bolt. this one broke around 2000lbs. nice, clean break with no signs of elongation.

 

[Metalguy]

Ductmate,

Your photos are pretty high quality. Unfortunately *I* can't really say what is wrong here. But we can form a few conclusions. It appears that baking these bolts makes them worse, if anything. The unbaked bolts appear to have a single plane of fracture, which is typical of "hydrogen embrittlement" and LME (liquid-metal embrittlement). BUT, the baking at 375 deg F. MAY have caused SME (solid-metal embrittlement), which typically has multiple planes of fracture.

I can't tell from the photos what the actual fracture mode is--intergranular, transgranular/shear, or trans. gran. cleavage, but hydrogen can be involved with the 1st and 3rd.

At this point you need a good met. lab. to perform a full failure analysis. They should be able to find traces of an embrittling metal (if any) on the fracture faces, and can even bake out and measure the hydrogen from unbaked bolts.

One lab. I have used in the past (I'm not connected with them) is in Phoenix--It's METL, and they did excellent work for me. There are others-the answers shouldn't be that difficult for a good lab.

 

[kenvlach]

Ductmate,
Pretty brittle. Have you checked the hardness yet, as Metalguy originally suggested? Maybe the bolts weren’t tempered.

Are these photos representative of those that failed earlier w/o the 375 deg F treatment?

Any info yet whether
a) hot-dipped, mechanically plated or electroplated,
b) acid descaling was used anywhere in process,
c) hydrogen bake-out was conducted where/if required?

Maybe a furnace operator used a reducing atmosphere to avoid scaling (rather than a neutral or air atm.). How was scale removed after heat treatment? Salt bath or mechanical methods are acceptable, acid pickling is not unless quickly hydrogen baked out.

Rich, thanks for posts. I knew that a little Al was sometimes used to minimize spangle size (for more uniform painting), but not about the tin problem.

 

[Ductmate]

I have not checked the hardness as I don't have the means to do so. No info on your other questions. We'll be meeting with the supplier today to discuss the problem. I'll get some more info out of them today, as well as discussing q.c. they should be doing to prevent this.

I'll post an update after I get all the info.

Again, thanks to everyone for their help.

Vince

 

[Ductmate]

After meeting with the supplier, I've gathered some more info.
The bolts are electroplated.
They are acid pickled.
There is no bake out period.
The composition of the steel is C45 (I think.)

Another question, does hydrogen embrittlement occur immediately after processing, or does it take a few days for it to happen?

Vince

 

[Paskenell]

Vince

Not sure if this is helpful or not but thought I would share it.

Fastener Standards Sixth Edition B-161 under Electrodeposited Coatings of Iron and Steel"
"6.5 - All steel parts having an ultimate strength of 1,200 MPa (174 ksi) or higher, shall be baked at a minimum of 190 degrees for 3 hours or more within 4 hours after electroplating to provide hydrogen embrittlement relief."

Some suggest this process be done immediatley and later than one hour after plating.