New studies of ancient concrete could teach us to do as the Romans did | Hacker News Reader

New studies of ancient concrete could teach us to do as the Romans did

phys.org

213 pointsdnetesn72 comments

According to the history on wikipedia's page on Tobermorite [1], this has been known at least since 2013.

[1] https://en.wikipedia.org/w/index.php?title=Tobermorite&actio...

I've been reading lately about the properties of lime mortarand concrete (check "building with lime"). I have an early 50's garaje that has rammed earth walls. Because recent covers with portland and plastic paint, the humidity is destroying the wall from the inside.

Amazingly the solution seems to be more breathable mortar cover using lime in the traditional way. This goes contrary against all the recomendations from modern buiding codes.

Lime mortars and concretes are very interesting for not technical (bridges and skyscrappers) applications. Lime cures absorbing CO2 and "petrifying" for months. It has much lower compression strength, but that helps to preserve the stone used to build. When there is thermal or ground movement, the part that breaks is the lime binder and not the rock, making it easier to repare (concrete is so strong that it breaks the whole wall). Also for small fisures, lime is self healing, it disolves with humity and fills the micro cracks. This way you can see this old european buildings that are actually leaning slightly in some direction but holding without mayor failures.

Another advantage is that it's breathable when using it as a wall cover. You don't develope "humity" stains in your wall nor it leaches white salts like the porland cement.

It can also be used to stabilize a road or ground around a house without pouring concrete, specially if you have problems with mud. Just mix a small part of lime for square meter of land, wet and ram. The ground will hold much better heavy traffic and absorb water when it rains. They are using this technique in some southamerican jungle roads as it's cheap and holds well. I've also seen videos of this method being used to improve the ground of factories.

Painting or sculping with cured lime (lime that has been rehydrated and wet for months, to allow it to cristalize). Gives you a sanitizing surface because it's extremely alcaline.

It's not a super material, in fact the structural properties are mediocre compared to modern ones, but the overall advantages are quite interesting for small buildings, decoration, and given that you are roman you could buid quite durable structures.

Very different concepts of how building materials should behave compared with modern materials (more isolation, more strength and stiffness).

About the rebar, I've read some articles about using basalt rebar (basalt glass fibers binded with epoxy). They say it's structurally stronger than steel and not degrading nor corroding. I don't know if it really works.

Edit 1 & 2: typos and formating. Writting from the phone..

Interesting, any reading recommendations books/links for this?

Most sources are from restorers of old buildings and lime associations.

Building with Lime: A Practical Introduction by Stafford Holmes et al. Link: http://a.co/3BdZ2tp

There is a book that it's virtually impossible to find: Artes de la cal by Ignacio Garate Rojas, who was restorer of the Alhambra de Granada and several other monuments in Spain.

http://anfacal.org/media/Biblioteca_Digital/Construccion/Mez...

http://cornishlime.co.uk/information/lime-in-building/

Soil stabilization: http://www.graymont.com/sites/default/files/pdf/tech_paper/l...

Hope it helps

Thanks, much appreciated!

Having apprenticed as a stone mason and then gone on to work with lime, I can definitely thumps up this post.

For anyone interested, here's the supplier I found some years ago and have been very happy with: http://www.limes.us/

Also lots of interesting info about working with lime mortars.

This is such an interesting point. When I was young my dad and I remodeled a 300 year old timbered home. Many people who had done the same had used foamed concrete between the beams and used modern paint. The outcome was similar to what happened to your garaje. Humidity lead to the beams starting to rod. So my dad and I used clay bricks and used lime based plaster and colors. Much more similar to how timbered houses used to be. Not only did the work come out well, but working with those materials especially the clay was a great experience. Traditional mortar will mess up your hands if you don't wear gloves whereas working with the clay was like a spa treatment.

One thing I recall reading is that rebar-reinforcing concrete is a major factor in concrete deterioration. So while concrete structures today don't last nearly as long as the Roman Pantheon (poured in the second century AD), they're also really cheap to build, can go really high, and last for 40 or more years (which given how cheap it is, the architects probably thought was long enough).

Rebar is a big problem. But concrete without it has terrible tensile strength. If you want to build in earthquake country, or have long concrete beams, there has to be something with tensile strength in there.

There's stainless steel rebar.[1] That holds up well against corrosion, but costs 8x as much as the cheap stuff. There was a fad for epoxy-coated rebar, but it's not holding up well in practice; the coating is too fragile.

The original Panama Canal locks are holding up well after a century. But the lock walls don't have rebar. The lock gates are all steel. The new locks have severe concrete flaws.[2]

[1] https://www.youtube.com/watch?v=FrFcGx_UPow [2] http://gcaptain.com/a-concrete-sample-was-pulled-from-the-ne...

This is a pet peeve of mine, and makes me want to scream every time I see some click bait about how much better roman concrete was. If their concrete was so great, where are the ruins of the toman skyscrapers?!? :)

Why would you need to build skyhigh, if there is still plenty of land available for low-rise building?

Why would you live in a city when there's plenty of suburban sprawl?

There are lots of reasons of efficiency to build up even if you can build out.

Romans had neither elevators or air conditioning. They did build on a fairly large scale though. I believe 150 foot tall structures were not uncommon, and those are the ones still standing.

That's simply a false dichotomy. Many of the world's great cities still mostly enforce height limits that the Romans could have built to. Because actually living like an ant among 100 story skyscrapers is not what most humans want. A medium density and mixed use city of max ~150 foot high buildings has very few disadvantages and analogizing it to USA suburban sprawl is just laughable nonsense. Have you ever been to e.g. Rome, Paris, Amsterdam, Florence ... most of Western Europe?

Also, less than 150 years ago, there was one seventh of the population on this planet than there is now. In fact, during the time of the Roman Empire there were only 300 million people on this rock, 45 million or so were within the Empire[1].

Skyscrapers are only a good solution if you have over-population. No-one in their right mind would ever want to live or work in one if they didn't have to.

[1] http://channel.nationalgeographic.com/killing-jesus/articles... (first Google hit)

Ancient Rome grew beyond 1 million inhabitants in the 2d century. Chang'an (Xi'an), China in the 8th century, Bagdad around the 10th century. There were big cities before.

https://en.wikipedia.org/wiki/List_of_largest_cities_through...

Engineers are expensive! Residential housing would have been wooden and (relatively) temporary.

I think a big reason ancient towers didn't really exist is less because they couldn't be built, and more that nobody wanted to climb up 5 or 10 floors if they could avoid it. No sane person would go up and down 20 floors if they could avoid it.

The colosseum is 150 feet tall, which isn't exactly short. Beyond that point, I think a big problem is getting the resources up high enough to keep building, as well as convincing people to climb that far up.

The only multistory residential buildings of antiquity I know of is the old city of Sana'a: https://en.wikipedia.org/wiki/Sana%27a#Old_City Five to nine stories tall, made of rammed earth, not even concrete! (Not enough water for concrete in Yemen) Continuously inhabited for ~2500 years.

@idlewords went there back in 2014, before the Saudis invaded: http://idlewords.com/2014/07/sana_a.htm

Rome had multistorey (3 to 4 of them) residential buildings, and opposite to today, the lower floors were the prestige floors - less climbing, more robust materials. The top floors were generally poorly constructed, out of lightweight materials.

In Manhattan, buildings over 100yo have taller ceilings on the first couple floors (where the retail shops would be and the owners would live), and then low ceilings on the highest floors for the scriveners and production workers. Of course, it's an issue today when lots of software companies want light (i.e., high floor) and high ceilings.

People don't realize what a paradigm shifting technology elevators were.

The Secret Life of Machines - The Lift (Elevator):

https://www.youtube.com/watch?v=dBLa105TQeg

These were called insulae[1]. Part of the reason the lower floors were higher prestige was due to the constant risk of fire. In the event of a fire those on the upper floors were in a far worse position in terms of survival.

1. https://en.wikipedia.org/wiki/Insula_(building)

Appart from the the ground floor (for business), it wasn't until the invention of elevators that the top floors were made prestige floors. The bel étage or piano nobile [1] was the first floor above ground level (for relative privacy from the street level). Top floors were not penthouse flats they were attics or lower class housing [2]

The first proto skyscrapers were built in the late 19th century. The Equitable Life Assurance Building (1870) and the Home Insurance Building (1885) were office buildings. The idea of a Penthouse appartment [3] dates from the 1920s. And as buildings are only slowly built and renovated, the first floors were still prestige floors well into the 20th century.

1 https://en.wikipedia.org/wiki/Piano_nobile

2 https://en.wikipedia.org/wiki/Chambre_de_bonne

3 https://en.wikipedia.org/wiki/Penthouse_apartment

Yep, and the scheme remained more or less the same in Italy where traditional homes (for the wealthy) have been till the 1900's organized as follows:

At ground floor (usually with higher celings than the rest of the house) dining room, sitting room, library, study, most of the "luxury" (marble, columns, paneling, paintings, etc.) was there, as they were the only parts of the house that visitors would see.

In the basement the kitchen and cellar (where cook and other servants worked).

On first floor (what probably our US friends will call second) the bedrooms for the owners.

On second (last) floor, right under the roof, usually with a reduced height AND extremey cold in winter and extremely hot in summer, the servant quarters.

Thanks for the idlewords link. His writing is very capturing, I've been reading the blog for hours.

Maybe it's lightning?

The Yonging Pagoda was 138 metres high. It was destroyed in 534 when it was struck by lightning and caught fire.

https://en.wikipedia.org/wiki/Yongning_Pagoda

Another example of tall Roman buildings still standing today is Roman aqueducts.

Examples: http://0.tqn.com/d/ancienthistory/1/S/U/I/Rome_5.jpg

http://3.bp.blogspot.com/-z_W3YC_mG0Y/Tb55AUhJ0dI/AAAAAAAAAE...

And, of course, the Pantheon, at 43 meters. It was the largest dome in the world for over 1300 years (https://en.wikipedia.org/wiki/List_of_largest_domes#Worldwid...)

The Nerja Aqueduct was built in the 19th century. But yes roman aqueducts were impressive https://en.wikipedia.org/wiki/Pont_du_Gard was 48.8m high (160feet).

The skyscraper was primarily made practical by elevators.

plumbing I think is also a major reason as well as why not just spread out. They werent as geographically contained as we are.

They were banned by building codes - insulae were limited to something like five stories due to fires. A problem that is tragically still relevant.

who wants skyscrapers in absence of elevators?

Interesting the doc built of nickel steel rebar. My grandfather owned a fishing boat built during WWII. Because of shortages copper nails were unavailable, so they used nickel. Still in good shape after 75 years.

That said my experience with material science issues has mostly been sadness and fail. As in.

Them: We will solve this problem with Super Duper X material.

Me: I think Super Duper X is going to give you problems.

Them: But it's Super Duper!!!

Me: Yes and it has Super Duper failures too.

Them: If you're so smart what would you use?

Me: I have no idea, why don't you consult a professional?

Them: See so you don't know what you are talking about.

Them: Besides professionals are expensive.

later... Oh noes Super Duper X isn't working!!!

Rust: The Longest War https://amzn.com/1451691602

"US vs Memory Safety"

flyash 1/4, cement 3/4 to neutralize chloride attacks

There's this new thing - probably not a fad - to just add plastic strips to the concrete while its in the mixer. This is popular with people building with polystyrene block forms e.g. houses and pools as it saves oodles of effort, and with experiments to '3d print' concrete buildings.

I kinda expect it to take off.

It poses a different kind of problem during recycling. With iron it's simple and easy to seperate. If you add plastic strips I imagine that it's quite a different story.

Cement production is already the very major polluter on a global scale:

https://en.wikipedia.org/wiki/Cement#Environmental_and_socia... especially https://en.wikipedia.org/wiki/Cement#/media/File:Global_Carb...

Surely, making the recycling process more complicated does not help.

First, plastics are on a spectrum of bio-degradableness and I have no idea what plastics are used to strengthen concrete.

Second, to recycling concrete you crush it. Crush it to powder and you can use it as aggregate in another batch of concrete. Leave it as big chunks and you can use it as fill. In this kind of recycling, the plastic can go with it just like any other.

Personally I prefer to build with wood.

http://nautil.us/issue/24/error/why-we-should-let-the-panthe...

There was some news years ago that there was a bacteria that could hold the key to ensuring modern rebar reinforced concrete could repair itself over time: http://www.nydailynews.com/news/national/concrete-fix-thanks...

Pozzolanic concrete with basalt rebar supposedly addresses the spalling and other issues with long-duration reinforced concrete use cases. There are various pozzolanic concrete recipes being tried to reproduce the list Roman recipe. No one I can find has combined the two and stuck it in seawater for 50+ years, though.

The sea water theory in the article does not adequately explain the aqueducts still standing in Europe in Italy/France/Spain/etc after 2000 years with no contact with salt water after construction, although I am not certain if the concrete in those things is used in a structural way similar to what would be expected of buildings.

My limited understanding trying to find out more about pozzolanic concrete (planning for my house extension in about a handful of years) is that those who study this stuff believe it stood up by itself like when used in the aqueducts with plenty of strength and durability, without constant immersion in salt water. Don't know if someone has drilled a core sample into those aqueducts to try to determine if salt water was used to mix that Roman concrete, though; would be interesting to find out.

What drew attention to it was that modern concrete immersed in seawater, even without rebar, doesn't appear to have the same durability as ancient Roman concrete. Don't know how much of this is attributable to survivorship bias, though; would be unfortunate if all this attention turns out to be over a tiny fraction of a percent of ruins that happened to survive.

I recall some experiments being done on the engineering side of my university using carbon rod for rebar. I don't recall how well the thermal expansion coefficients of the two materials matched up, but the professor claimed a concrete + carbon bridge would last 200 years, due to lack of rust.

This was in the late 90's, so we still have a few more years to go before the test is complete. ;)

Regarding rebar - in my country communists build a lot of "wielka płyta" (big pre-stressed concrete panels) commie blocks. Around 20% of Poles still live in these. These buildings were built since sixties, and designed to stay for 50 years, but so far they seem to hold rather well, none of them were destroyed so far in a building catastrophe, not even when there were gas explosions in some blocks.

They were a little modified from the standard commie thing (for example stainless steel was used for joining the panels), and they had lots of usability problems (noise and temperature insulation), but from durability POV they seem to work better than expected. I wonder if in next decades 20% of the country will have to buy a new house, cause that might be a huge economic change.

Again? These Scientists must have the memory of goldfish - It seems like this is cracked every other month in the last 5 years or so

Seriously. Pozzolans and hydraulic cements have been well understood for a very long time.

https://en.wikipedia.org/wiki/Pozzolan

This article seems to imply that only Roman marine concrete is more lasting because of the ongoing reaction with the sea water, but I was under the impression that all Roman concrete was more lasting... Is it just the presence of rebar that makes today's concrete so short lived or is there something else as well?

I'm no expert, but I've read that their use of volcanic rock is a major factor - since "pumice" has a swiss-cheese like surface, and that tends to add strength to concrete. Also, they used incredibly little water in their mixtures - there are descriptions of them pounding in the concrete (as opposed to the way we "flow" it.) I understand that studies have shown that concrete mixed with very little water is much stronger. I also read that fast-setting concrete has been found to be very weak - and was never actually tested in any significant way before it was approved. So very long curing times result in much stronger concrete.

Obviously, all these discoveries are being fought by the building industry - so none of these things are reflected in current code. Instead, they just argue about rebar while ignoring the other discoveries.

Just because it's stronger doesn't mean it's more durable, which the article points out. Portland concrete has higher compressive strength but doesn't last as long.

Well, old concrete, even pre-"quick set" has been found to be much more "durable" (colloquially speaking - not really an engineering term.) And in most applications it isn't concrete's compressive strength which causes failure - for either old concrete or new concrete. So I would hesitate to defend current building practices based on the lone metric of compressive strength. But, again, I'm not a structural engineer (though I do know one who works for the state, and he isn't allowed to express his real opinions about the danger of our current infrastructure - many building failed in earthquakes which they should have survived easily.)

From my experience with epoxy, thicker viscosity and slower-curing is well known to result in a stronger end material. I'm not saying this is also true with concrete, but there is a parallel precedent.

I seem to recall reading something about concrete density. Roman concrete when wet was about as malleable as Silly-Putty. It took a lot of work to shape it into its forms and eliminate air pockets. Our concrete is looser and pours when wet, so it's a lot easier to work with, but it's also weaker.

Easier to work with -> faster and cheaper.

The Romans optimised for long-term durability; modern builders optimise for cost.

It's like any manufacturing industry: building expensive stuff that lasts forever will naturally diminish your business in the long run. It's much better to make "cheap but good enough" stuff that you throw away and re-make every X years. Such is the way of the private profit.

Also, union construction guys are more expensive than slaves. Remember that manhours were cheap in the Roman era.

There is a big difference in the "base" cement, even modern Pozzolanic cement is more lasting than Portland.

Pozzolan has been considered for years an "inferior" cement (IMHO wrongly) because it is much slower in setting/curing.

The typical compresssion test for concrete (it depends on countries/standards) is carried on samples of 28 days of age. A concrete made with Portland cement normally is very near, (let's say conventionally 85-90%) to the maximum strength it will ever reach (the maximum is reached usually within months).

A concrete made with Pozzolanic cement, tested in the same manner is more likely to be in the 50-60% range, Pozzolan will normally continue to harden (for years), often overtaking the Portland.

Additionally there is a "mass effect" particularly if the environment has a high humidity, that makes the samples being "less representative" of the actual concrete mass in the case of Pozzolanic (a "huge" thickness of Pozzolanic cement has generally speaking more compression resistance than what the test sample may prove).

Now if you have to build multi-storey buildings or - say -high raise pillars/columns one of the key factors is time, the soon you can remove formworks and scaffolding and go to the next storey or level is essential and a much faster setting Portland cement has been preferred for all the 20th century, before durability was even thought of.

Nowadays there is a trend towards "blended" cements, see:

https://en.wikipedia.org/wiki/Pozzolan

https://en.wikipedia.org/wiki/Pozzolan#Use

I live in a limecrete building from 1857, in the Scottish Borders.

It is similar to a tower block in minature, a girder frame with cast walls, only 3 stories high - but the rest of the village is all 2 story stone cottages. Innovative technology for 1857 !

But being limecrete it has set into limestone over the last 150 years - a very long lasting technology.

The only downside is it suffers from the 1960s portland-cement based harling (pebble dash) which keeps the moisture close to the stone - hopefully we can change this to a lime render and limewash as it should be.

Maybe someone in this thread will know but I've always feared Roman concrete suffers from survivor bias.

Do we know Roman concrete was always great or are we just looking at structures that happen to last?

Aren't the structures that lasted 2000 years also the ones that are worth taking a closer look at?

I suspect "curing time" is probably part of the issue.

As a mental exercise, how many projects would get funding if their concrete took 10 years to cure? So, how far down do you have to dial that before people are willing to pay the money and how much impact does that have on "durability"?

Romans built large scale concrete structures in months, not years. Given that these would require a great many layers of cement, not just one, the upper bound for curing time would have to be several days, a week at the very edge of possibility.

Keep in mind that concrete usually isn't allowed to fully cure before construction moves on. Most concrete houses only fully set after 2-3 years and truly large structures, famously Hoover dam, can take centuries.

This is why Hoover Dam was poured in planned, interlocking segments, with cold water piped through it to accelerate curing.

A contiguous pour would have taken an estimated 125 years to cure, even assuming no fatal shrinkage occurred.

Surely cooling retards curing. And it is commonly done to avoid inducing thermal stress and causing cracks.

Not necessarily - the reaction that produces heat is purely chemical, by taking this heat away you're not going to stop it, but you are going to give more headroom for the reaction to continue.

They probably overbuilt so that the half cured concrete would support the structure. Hence it's still around 2000 years later.

Look at the old railroad bridges and viaducts built in the 1890s that are still going strong. They continue to exist because they are grossly overbuilt.

Another interesting ancient concrete technology is using rice to form a composite: https://www.acs.org/content/acs/en/pressroom/newsreleases/20...

Some hempcrete/hemp lime binders are also made with pozzolites.

Reply on news.ycombinator.com