4 ms·

Why Roman concrete is stronger than it ever was, while modern concrete decays

44 points by jdkee 6y ago

gumby 6y ago
ISTR learning this in an HN post back in 2017. But I can't find it.
MichaelZuo 6y ago
For a civil engineer these ‘Roman concrete’ pop science articles must get a bit tiring after a while. It would be better to call it by a more accurate name ‘certain mixtures of high cost aggregate-cement concrete’. Then we can say modern concrete prepared with ‘certain mixtures of high cost aggregate-cement’ would also grow stronger over time too in similar conditions. And then the erroneous title becomes obvious.
- glandium 6y ago
  See also survivorship bias.
  - fuzzfactor 6y ago
    This was not ancient man, this was modern man in Ancient Roman times. So not really any more intelligent nor less so compared to our best materials engineers today. Nor any less talented at the Scientific Method or Continuous Improvement process over time. Nobody knows how many centuries or even millennia was spent before humans had concrete about like we have today, which only lasts a few centuries, since anything much lesser has already decayed without a trace. Further, nobody even knows how many more eons it took before Romans had the stuff that turned out to be the most long-lasting so far. No matter what, it might just plain take a few millennia into the future for us to get from where we are now to where they were then, you can't be sure but you can't let that stop you. >After the fall of the Roman empire, the recipe for making concrete was lost and a concrete of equal worth wasn’t re-invented until 1824 Not a very long term effort since then by comparison. Looks like a lot of catching up to do. Plus many people might not have actually been trying continuously to improve it that much in one single recent century compared to previous ancient centuries. Like Buckminster Fuller says most of the time for the masses once something novel and badly needed achieves popular adoption, further progress can halt right there. Barely good enough is the enemy of perfect. Now with a geologist like Jackson and her rare x-ray technique this is a 21st century workaround to try and compensate in only one lifetime for the loss of possibly millennia of previous actual human development. This could take a while and arrangements should be made to continue for more than one lifetime, or at least as long as possible beyond one university tenure. Even more so in this case since the materials testing phase can be expected to last more than a century itself after each cycle of continuous improvement, once cumulative progress has moved the overall longevity far enough beyond what we have now. This is survivorship bias in action, put to use. Many times employed in materials testing. The ancients definitely didn't have the same technolgy we have today, but what if many more of them were giving it all they had with greater determination compounded over very much longer periods of times? No doubt superior outcomes can be had from earlier, sometimes lost, technology. In finance more wealth can usually be extracted more quickly from generous reservoirs while they last compared to being built and protected from the ground up. For technology the biggest reservoir is the very long trailing edge which has more area under the curve than you are ever going to get in your little time no matter how late you come to the table and how long you stay. As I suspected, the longer you compound your concrete knowledge the better. With the wealth of more centuries of research behind you, it would be a lot easier to justify any increased cost of raw materials to fill a cement truck.
- souprock 6y ago
  Is that high cost mostly for an inherent reason, or mostly because such concrete isn't in mass production? What sort of cost difference would that be? If I actually wanted that type of concrete, how would I specify it and then ensure that I got it? (not ripped off by seller) Does it violate building codes?
  - MichaelZuo 6y ago
    It’s high cost for both reasons. If you wanted it in quantity you probably need to built your own cement factory. So at least a nine digit figure to get going. For smaller scales there’s a wide range of possibilities. You would likely need concrete, cement, and earth aggregate experts on staff to ensure you got it. The code issue would be dependent on your jurisdiction. How much would it cost hire the above + a code expert for say a hundred hours where your at? That’s a good starting figure.
mjbeswick 6y ago
One of the main reason modern concrete structures fail is because the are reinforced with steel, which expands when it corrodes causing cracks. Adding seawater to concrete used with steel reinforcement would cause it to fail much quicker. Many other kinds of reinforcement are available, but steel rebar is by far the most commonly used as it's cheap. Roman concrete structures have little tensile strength and depends on compression to maintain their integrity, so their design is much more limited.

44 points by jdkee 6y ago

ISTR learning this in an HN post back in 2017. But I can't find it.

For a civil engineer these ‘Roman concrete’ pop science articles must get a bit tiring after a while. It would be better to call it by a more accurate name ‘certain mixtures of high cost aggregate-cement concrete’. Then we can say modern concrete prepared with ‘certain mixtures of high cost aggregate-cement’ would also grow stronger over time too in similar conditions. And then the erroneous title becomes obvious.

glandium 6y ago
See also survivorship bias.
- fuzzfactor 6y ago
  This was not ancient man, this was modern man in Ancient Roman times. So not really any more intelligent nor less so compared to our best materials engineers today. Nor any less talented at the Scientific Method or Continuous Improvement process over time. Nobody knows how many centuries or even millennia was spent before humans had concrete about like we have today, which only lasts a few centuries, since anything much lesser has already decayed without a trace. Further, nobody even knows how many more eons it took before Romans had the stuff that turned out to be the most long-lasting so far. No matter what, it might just plain take a few millennia into the future for us to get from where we are now to where they were then, you can't be sure but you can't let that stop you. >After the fall of the Roman empire, the recipe for making concrete was lost and a concrete of equal worth wasn’t re-invented until 1824 Not a very long term effort since then by comparison. Looks like a lot of catching up to do. Plus many people might not have actually been trying continuously to improve it that much in one single recent century compared to previous ancient centuries. Like Buckminster Fuller says most of the time for the masses once something novel and badly needed achieves popular adoption, further progress can halt right there. Barely good enough is the enemy of perfect. Now with a geologist like Jackson and her rare x-ray technique this is a 21st century workaround to try and compensate in only one lifetime for the loss of possibly millennia of previous actual human development. This could take a while and arrangements should be made to continue for more than one lifetime, or at least as long as possible beyond one university tenure. Even more so in this case since the materials testing phase can be expected to last more than a century itself after each cycle of continuous improvement, once cumulative progress has moved the overall longevity far enough beyond what we have now. This is survivorship bias in action, put to use. Many times employed in materials testing. The ancients definitely didn't have the same technolgy we have today, but what if many more of them were giving it all they had with greater determination compounded over very much longer periods of times? No doubt superior outcomes can be had from earlier, sometimes lost, technology. In finance more wealth can usually be extracted more quickly from generous reservoirs while they last compared to being built and protected from the ground up. For technology the biggest reservoir is the very long trailing edge which has more area under the curve than you are ever going to get in your little time no matter how late you come to the table and how long you stay. As I suspected, the longer you compound your concrete knowledge the better. With the wealth of more centuries of research behind you, it would be a lot easier to justify any increased cost of raw materials to fill a cement truck.
souprock 6y ago
Is that high cost mostly for an inherent reason, or mostly because such concrete isn't in mass production? What sort of cost difference would that be? If I actually wanted that type of concrete, how would I specify it and then ensure that I got it? (not ripped off by seller) Does it violate building codes?
- MichaelZuo 6y ago
  It’s high cost for both reasons. If you wanted it in quantity you probably need to built your own cement factory. So at least a nine digit figure to get going. For smaller scales there’s a wide range of possibilities. You would likely need concrete, cement, and earth aggregate experts on staff to ensure you got it. The code issue would be dependent on your jurisdiction. How much would it cost hire the above + a code expert for say a hundred hours where your at? That’s a good starting figure.

mjbeswick 6y ago

One of the main reason modern concrete structures fail is because the are reinforced with steel, which expands when it corrodes causing cracks. Adding seawater to concrete used with steel reinforcement would cause it to fail much quicker. Many other kinds of reinforcement are available, but steel rebar is by far the most commonly used as it's cheap. Roman concrete structures have little tensile strength and depends on compression to maintain their integrity, so their design is much more limited.