316 vs 304 stainless steel rock anchors

For those interested in the corrosion resistance of stainless steel rock anchors I have a new post up on my blog.

I am finally in a position to start taking a serious look at the 316 vs 304 (A4 vs A2) debate. Here I describe an anomalously corrosion-resistant specimen of 304 and reveal post-manufacture annealing as the likely reason for this improvement.

https://cragchemistry.com/2023/08/02/now-this-is-different/

https://cragchemistry.com/2023/08/02/now-this-is-different/ 

I have no idea what I just read meant but I’m glad there are people way smarter than me trying to make bolts as safe as possible!

Dude!  I’m a chemical engineer and a route developer and had resigned myself to being alone in my nerdy crossover of those two worlds.  Your blog is amazing, keep it up!

Kent Krauza wrote:

Dude!  I’m a chemical engineer and a route developer and had resigned myself to being alone in my nerdy crossover of those two worlds.  Your blog is amazing, keep it up!

Thanks, and yeah, I do what I do because I'm a nerd through and through, so it is great to find others that are nerdy about this stuff.

Can you break it down for someone who develops but is an idiot (me)?  I just wanna know if in relatively dry desert and mountain areas if 304 is cool to use.  I generally use 316 but if I find a good price soon 304 I jump at it.  

Kevin Mokracek wrote:

Can you break it down for someone who develops but is an idiot (me)?  I just wanna know if in relatively dry desert and mountain areas if 304 is cool to use.  I generally use 316 but if I find a good price soon 304 I jump at it.  

Basically, the author found a piece of 304 stainless chain that was exhibiting slower corrosion rates then the stainless around it.

He tested it in different ways and found that it had been “annealed” (or heated to a specific temp and cooled slowly) in a certain way and preposites that this process has made the chain more resistant to the corrosion.

…At least, that is what I took from it. I went to art school though, so mileage may vary, don’t quote me on it.

Super interesting read OP. Wish I could understand it even more but there was some complexity that I just wasn’t willing to google. Going to deep dive through your blog a bit more when I can!

Kevin Mokracek wrote:

Can you break it down for someone who develops but is an idiot (me)?  I just wanna know if in relatively dry desert and mountain areas if 304 is cool to use.  I generally use 316 but if I find a good price soon 304 I jump at it.  

Firstly, I should point out that all horror stories concerning aggressive corrosion of 304 have involved sea cliffs, and not just any sea cliff, but only those where high levels of environmental sulphate are available.  However, lack of evidence cannot be taken as evidence of lack, and there may be cases out there that will come to light.

So, what to do? I believe in reasons rather than rules, and thus consider the blanket ban on the use of 304 to be unsound. There are huge amounts of 304 giving excellent service right across the world, the trick is to understand its fatal weakness. So, can you use it in the regions you propose? I'd be gob-smacked if there was a problem.

However, by choosing 316 over 304 you can eliminate that known fatal weakness. These days the price differential is hardly enough not to choose 316.

Threads like this are one reason I love mountain project.  

David Reeve wrote:
However, by choosing 316 over 304 you can eliminate that known fatal weakness. These days the price differential is hardly enough not to choose 316.

That last paragraph sums it all up for me.  Exactly the reason why we use all 316 in the routes we develop now.  I feel that one of the responsibilities of a developer is to use the best materials out there and as David said the price differential is negligible.  Thank you for this and refreshing to see good and productive threads on mp such as this.

"Firstly, I should point out that all horror stories concerning aggressive corrosion of 304 have involved sea cliffs, and not just any sea cliff, but only those where high levels of environmental sulphate are available.  However, lack of evidence cannot be taken as evidence of lack, and there may be cases out there that will come to light."

I think it would be great to lead into any discussion about corrosion in 304 with a paragraph like this one. The ubiquitous five piece bolt is an example of an excellent piece of hardware in common usage in the US. The ones i placed 33 years ago still look like new here in the hot and wet Mid-Atlantic region.

1.  SS316 is hardly the solution to SCC.  It has huge SCC problems.  For just one, the OP article mentions that greater than 10% nickel helps somewhat.  But 316 can be anywhere from 10-14%, so it is not necessary high enough, and nickel has become expensive.  Another common corrosion benchmark is PREN.  SS304 has PREN 19, 316 is only slightly better at 24.  Metals that really resist SCC have PREN > 34.  One alloy in use for expansion bolts is EN 1.4462 duplex.   There are some slightly different versions, a newer one being 2205.  Other alloys such as 1.4529 are even better but cost twice as much.   secure.outokumpu.com/steelf…;Category=Forta
Not to be used at temperature less than -50°C (becomes brittle).    

example  http://bolt-products.com/SeaWaterSeries_000.htm   mentions an expansion bolt.  these are mostly glue-ins,  in which case Titanium is the standard.

The UIAA says
- 316(L) or 304(L) is NOT appropriate for any area where SCC has been documented or is suspected.
- Use of 316L grade or better alloy for corrosion resistance is recommended for all outdoor anchor components in locations where SCC has never been documented and there is no reason to suspect its presence.

- The UIAA in 2020 added a category rating of UIAA 123 SCC which means a bolt can be used in an SCC environment, but it's not clear if anyone has qualified to this.

2.  SCC can happen away from sea cliffs, especially at limestone/dolomite and especially at karst formations.  UIAA map   google.com/maps/d/viewer?mi…;hl=en&ll=-7.63333123551244e-14%2C62.30778065000004&z=2

tom donnelly wrote:

1.  SS316 is hardly the solution to SCC.  It has huge SCC problems.  For just one, the OP article mentions that greater than 10% nickel helps somewhat.  But 316 can be anywhere from 10-14%, so it is not necessary high enough, and nickel has become expensive.  Another common corrosion benchmark is PREN.  SS304 has PREN 19, 316 is only slightly better at 24.  Metals that really resist SCC have PREN > 34.  One alloy in use for expansion bolts is EN 1.4462 duplex.   There are some slightly different versions, a newer one being 2205.  Other alloys such as 1.4529 are even better but cost twice as much.   secure.outokumpu.com/steelf…;Category=Forta
Not to be used at temperature less than -50°C (becomes brittle).    

example  http://bolt-products.com/SeaWaterSeries_000.htm   mentions an expansion bolt.  these are mostly glue-ins,  in which case Titanium is the standard.

The UIAA says
- 316(L) or 304(L) is NOT appropriate for any area where SCC has been documented or is suspected.
- Use of 316L grade or better alloy for corrosion resistance is recommended for all outdoor anchor components in locations where SCC has never been documented and there is no reason to suspect its presence.

- The UIAA in 2020 added a category rating of UIAA 123 SCC which means a bolt can be used in an SCC environment, but it's not clear if anyone has qualified to this.

2.  SCC can happen away from sea cliffs, especially at limestone/dolomite and especially at karst formations.  UIAA map   google.com/maps/d/viewer?mi…;hl=en&ll=-7.63333123551244e-14%2C62.30778065000004&z=2

The predominant failure mechanism discussed is SSC (sulphide stress cracking) not SCC however to date, the UIAA Anchor Working Group and/or David has yet to receive a failed anchor confirmed to be have been manufactured from 316. 

Everything analysed by Tomas Prosek and others at the UIAA 'adopted' lab in Prague was fabricated from 304, regardless of whether the failure was attributed to SSC or SCC.

Stating that a metal grade lacks sufficient corrosion resistance is meaningless without characterising what type and severity of environmental corrosion it is expected to resist.

316, because of the molybdenum and elevated nickel content, provides improved corrosion protection compared to 304, for the majority of environments where climbers are installing fixed protection.

Titanium is great but climbers have to balance their budget with what they think will work, for them, in their environment and in most cases where titanium anchors are being used, it is because developers are fully aware of the corrosion issues involved.

  

tom donnelly wrote:

1.  SS316 is hardly the solution to SCC.  It has huge SCC problems.  For just one, the OP article mentions that greater than 10% nickel helps somewhat.  But 316 can be anywhere from 10-14%, so it is not necessary high enough, and nickel has become expensive.  Another common corrosion benchmark is PREN.  SS304 has PREN 19, 316 is only slightly better at 24.  Metals that really resist SCC have PREN > 34.  One alloy in use for expansion bolts is EN 1.4462 duplex.   There are some slightly different versions, a newer one being 2205.  Other alloys such as 1.4529 are even better but cost twice as much.   secure.outokumpu.com/steelf…;Category=Forta
Not to be used at temperature less than -50°C (becomes brittle).    

example  http://bolt-products.com/SeaWaterSeries_000.htm   mentions an expansion bolt.  these are mostly glue-ins,  in which case Titanium is the standard.

The UIAA says
- 316(L) or 304(L) is NOT appropriate for any area where SCC has been documented or is suspected.
- Use of 316L grade or better alloy for corrosion resistance is recommended for all outdoor anchor components in locations where SCC has never been documented and there is no reason to suspect its presence.

- The UIAA in 2020 added a category rating of UIAA 123 SCC which means a bolt can be used in an SCC environment, but it's not clear if anyone has qualified to this.

2.  SCC can happen away from sea cliffs, especially at limestone/dolomite and especially at karst formations.  UIAA map   google.com/maps/d/viewer?mi…;hl=en&ll=-7.63333123551244e-14%2C62.30778065000004&z=2

I think Francis, above, has covered the fact that I am specifically talking about SSC not SCC. Have a read of my earlier post for my reasoning for assigning aggressive 304 anchor corrosion to SSC rather than SCC. Link here.

I can answer some of the other points you raise.

Yes, I agree that 316 is vulnerable to SCC. It differs from 304 in that it contains 2% of molybdenum which elevates the pitting corrosion potential over that of 304. This bit of electrochemistry serves to reduce the number of pits that form and thus potentially initiate SCC. As you say, it doesn't raise the PREN high enough to totally suppress SCC. However, it doesn't bear on SSC resistance.

The new rock anchor corrosion categories introduced by the UIAA are to be welcomed for sure. However, as I pointed out to Safe Com, the prescribed tests are all specific to just one type of corrosion, and that is chloride-driven stress cracking. Other mechanisms do exist for which the prescribed categories may serve little purpose. To this end, they added a caveat that specifically mentions environments of high sulphate.

There are a number of locations where literally thousands of 304 bolts have been installed, only to be dangerously corroded in well under 10 years. My position is that every one of these locations has high environmental sulphate levels that sponsors aggressive SSC due to the presence of sulphate reducing bacteria. I am yet to see any examples of SSC failure numbering in the tens, let alone the thousands.

I know this sounds like heresy, and I know very competent labs have assigned some failures to SCC, but in my opinion, all have lacked adequate investigation. The evidence presented could be interpreted as either SSC or SCC. Once again in my opinion, I fail to see how the laboratory conditions required to induce SCC can be replicated at the crag. A final point is that we see brittle fracture associated with this corrosion phenomenon. Hydrogen embrittlement is part and parcel of SSC and not SCC.

Science is a slow self-correcting process. All of this will be amended and tweaked in the correct direction as time marches on.

David Reeve wrote:

There are a number of locations where literally thousands of 304 bolts have been installed, only to be dangerously corroded in well under 10 years. My position is that every one of these locations has high environmental sulphate levels that sponsors aggressive SSC due to the presence of sulphate reducing bacteria. I am yet to see any examples of SSC failure numbering in the tens, let alone the thousands.

Typo?

Is the debate about seaside corrosion a climbing specific thing, or is it generally not figured out? There's millions of times more steel used more in buildings/bridges/etc on the coast than bolts, so I would have assumed they had the financial incentive to figure this out and we could just apply that knowledge to climbing. Or are engineers building bridges also arguing about SSC vs SSC and grades of stainless?

David Reeve wrote: ...Yes, I agree that 316 is vulnerable to SCC. It differs from 304 in that it contains 2% of molybdenum which elevates the pitting corrosion potential over that of 304. ...

David, did you state the inverse of what you intended? Doesn't the addition of molybdenum decrease the pitting corrosion potential? Or, that could be stated as: elevates the pitting corrosion *protection*.

Austin Donisan wrote:

Typo?

Is the debate about seaside corrosion a climbing specific thing, or is it generally not figured out? There's millions of times more steel used more in buildings/bridges/etc on the coast than bolts, so I would have assumed they had the financial incentive to figure this out and we could just apply that knowledge to climbing. Or are engineers building bridges also arguing about SSC vs SSC and grades of stainless?

Steel bridges must be kept fully coated to protect from weather or they will corrode.  Most steel reinforced concrete near coasts in the past was made with generic steel rebar or tensioning.  Much of it eventually corrodes and collapses.  Today some stainless is being used, as well as some other rebar options:  galvanized coating (imperfect, but most common), epoxy coating (often gets holes in the epoxy) , boron (good for straight runs, but can not be bent).

A trillion dollar corrosion problem is steel pipelines for natural gas and petroleum.  No great material solution has been found that is cheap enough.

Austin Donisan wrote:

Typo?

Is the debate about seaside corrosion a climbing specific thing, or is it generally not figured out? There's millions of times more steel used more in buildings/bridges/etc on the coast than bolts, so I would have assumed they had the financial incentive to figure this out and we could just apply that knowledge to climbing. Or are engineers building bridges also arguing about SSC vs SSC and grades of stainless?

A climbing specific thing.

The issue of Stress Corrosion Cracking (SCC) in austenitic stainless steels and sulphate based corrosion (SSC) has been known for decades. Hydrogen embrittlement for a lot longer. Climbing is a recreational sport where climbers have been installing anchors largely outside of defined standards requirements for decades. Ours is a very different context to that of a construction environment where building specifications and working standards etc are core to everything and non negotiable.

The cause of the aggressive corrosion associated with Southern Thailand, Taiwan, Cayman Islands to name but a few well-known examples, was originally thought to be caused by Stress Corrosion Cracking (SCC) because of Angele Sjong's study in 2008. Southern Thailand became ground zero I think, simply because so many climbers were visiting and breaking bolts. The Cayman Islands were also having issues which is what led to John Byrnes later developing titanium glue-in anchors (Tortuga bolts). The assumption at the time was that magnesium in the limestone was combining with chloride in saltwater to form magnesium chloride, which is corrosive, so when a Petzl hanger was subjected to an ASTM G36 boiling magnesium chloride test under lab conditions, it unsurprisingly cracked. This logically fitted the environment in Thailand where similar conditions of humidity, material type, stress were known to have caused catastrophic SCC failures in swimming pool roofs that were made from stainless steel. I helped with re-bolting in Southern Thailand during 2000-2001 and it was assumed then that the problem was because of anchor type (expansion bolts) so re-bolting switched to glue-ins but they failed too so then the adhesive type was suspected as being too porous and not sufficiently insulating the anchor material (the switch to Hilti RE 500 in the belief a pure epoxy would save the day) but the failures continued by which time some routes had been re-bolted several times over. Angele's work seemed to identify the corrosion mechanism but did not actually characterise the chemical environment. Myths such as; seaside limestone sport crags will always have SCC and it's saltwater mixing with the magnesium in the rock started to spread as fact, but ultimately, could not explain why similar aggressive corrosion was occurring at Long Dong, in Taiwan, where the geology is sandstone and entirely different.

Many people have been involved in the Safe Com Anchor Working Group work but David spearheaded the analytic chemistry in characterising the chemical environment which identified that sulphur was present in the crag environment, at higher Molar concentrations than the natural sulphur cycle could be producing. Volunteers conducted wall wash sampling at climbing areas in known aggressive locations but also and importantly, benign areas too, sending the samples to David for his laboratory analysis, which identified elevated chlorides BUT also elevated sulphate compounds. Specifically Calcium Sulphate in Southern Thailand and Aluminium Sulphate in Taiwan, at very nasty concentrations. Undersea fumaroles exist in these locations and are the logical source of the external sulphur. The data collected from crags globally has consistently supported aggressive corrosion occurring where sulphates are present and not just chlorides. Data from crags in coastal areas where stainless bolts were installed but continue to be reliable have been sampled and found to be absent of sulphate conditions.

Once sulphate based corrosion was identified, analysis focused on sulphur bacteria and hydrogen embrittlement in tandem with the material analysis of failed anchors (Tomas Prosek's UIAA funded work in Prague). Crucially the formal material analysis of failed anchors highlighted that too many assumptions were being made by climbers about the composition of fixed anchors. Often samples were assumed to be 316 only to be something else such as 304 or 303.

David Reeve wrote:

I think Francis, above, has covered the fact that I am specifically talking about SSC not SCC. Have a read of my earlier post for my reasoning for assigning aggressive 304 anchor corrosion to SSC rather than SCC. Link here.

Good point, and I don't know what are the real differences as far as material corrosion results.  I was assuming they aren't that different.

There are a number of locations where literally thousands of 304 bolts have been installed, only to be dangerously corroded in well under 10 years. My position is that every one of these locations has high environmental sulphate levels that sponsors aggressive SSC due to the presence of sulphate reducing bacteria. I am yet to see any examples of SSC failure numbering in the tens, let alone the thousands.

I know this sounds like heresy, and I know very competent labs have assigned some failures to SCC, but in my opinion, all have lacked adequate investigation. The evidence presented could be interpreted as either SSC or SCC. Once again in my opinion, I fail to see how the laboratory conditions required to induce SCC can be replicated at the crag. 

I think you need more evidence to prove that at locations other than your primary test cliff.

A final point is that we see brittle fracture associated with this corrosion phenomenon. Hydrogen embrittlement is part and parcel of SSC and not SCC.

I thought hydrogen embrittlement also applied to SCC.

Austin Donisan wrote:

Typo?

Is the debate about seaside corrosion a climbing specific thing, or is it generally not figured out? There's millions of times more steel used more in buildings/bridges/etc on the coast than bolts, so I would have assumed they had the financial incentive to figure this out and we could just apply that knowledge to climbing. Or are engineers building bridges also arguing about SSC vs SSC and grades of stainless?

I think others have covered this point, but I'll add a couple of comments.

Yes, there are huge amounts of architectural stainless steel exposed to the marine environment, but is that relevant? I see three factors that make the aggressive corrosion phenomenon we observe on sea cliffs quite different to say that which might occur on the balustrades of apartment buildings. I am working on a post that attempts to formally analyse quite what it is that is special. For now, I'll just need to wave my hands a bit. :-)

Firstly, only those sea cliffs facing an ocean that provides a source of elemental sulphur are relevant. I believe this comes down to location with respect to underlying plate tectonics and associated volcanism.

Secondly, rock anchors are placed into what is the living biological entity of the natural crag environment. This environment supports the sulphur oxidising bacteria that convert the incoming sulphur into sulphuric acid and thus sulphate. With a source of sulphate we are primed for the third factor.

Thirdly, because rock anchors, by design, are embedded into the rock, we end up presenting metallic iron in a low oxygen environment. Given sufficient sulphate, it is inevitable that sulphate reducing bacteria (SRB) will gate-crash our party.

If the steel is of a type resistant to the penetration of hydrogen, the consequences of attack by SRB are superficial. However, for a material like cold-worked 304, hydrogen entrainment allows the bacteria to penetrate deep into the metal leading to stress cracking.

If we consider the Costa Blanca, we see the aggressive corrosion of 304 on the sea cliffs, but I've read no reports of problems with manmade infrastructure. Why? My guess is that the missing piece of the puzzle is the second item above. The thin soil of the clifftop and its ledges is necessary to support the sulphur oxidising bacteria.

ClimbBaja wrote:

David, did you state the inverse of what you intended? Doesn't the addition of molybdenum decrease the pitting corrosion potential? Or, that could be stated as: elevates the pitting corrosion *protection*.

Yes, the effect of molybdenum is to increase protection against pitting corrosion.

The pitting corrosion potential, Vpit, is elevated, and must be exceeded for a pit to be initiated. In practice this means that the chloride ion activity, or temperature, must be increased if pitting is to be initiated. Put the other way around, this translates as 316 being more resistant to pitting initiation than 304.

It is worth noting that just the initiation of pitting can be predicted in this way. The actual growth of the pit, and further, the propagation to a stress crack is something else again.