How to Make Roman Concrete, One of Human Civilization’s Longest-Lasting Building Materials

More than a mil­len­ni­um and a half after its fall, we still look back with won­der on the accom­plish­ments of the ancient Roman Empire. Few ele­ments of its lega­cy impress us as much as its built envi­ron­ment — or in any case, what’s left of its built envi­ron­ment. Still, the fact that any­thing remains at all of the struc­tures built by the Romans tells us that they were doing some­thing right: specif­i­cal­ly, they were doing con­crete right. Just how they made that aston­ish­ing­ly durable build­ing mate­r­i­al has been a sub­ject of research even in recent years, and we even fea­tured it here on Open Cul­ture back in 2017. But could we make Roman con­crete today?

Such is the task of Shawn Kel­ly, host of the Youtube chan­nel Cor­po­ral’s Cor­ner, in the video above. Using mate­ri­als like vol­canic ash, pumice and lime­stone, he makes a brick that looks more than sol­id enough to go up against any mod­ern con­crete.

As of this writ­ing, this sim­ple video has racked up more than three mil­lion views, a num­ber that reflects our endur­ing fas­ci­na­tion with the ques­tion of how the ancient Romans cre­at­ed their world — as well as the ques­tion addressed in the high­er-tech Prac­ti­cal Engi­neer­ing video below, “Was Roman Con­crete Bet­ter?”

The fact of the mat­ter is that, despite pos­sess­ing tech­nolo­gies the Romans could hard­ly have imag­ined, their con­crete lasts longer than ours. Why that should be the case comes down, in large part, to water: we put a great deal more of it into our con­crete than the Romans did, in order to pour it more cheap­ly and eas­i­ly. But this makes it more frag­ile and sub­ject to dete­ri­o­ra­tion over time (as seen in the ear­ly dilap­i­da­tion of cer­tain Bru­tal­ist build­ings), even despite our use of chem­i­cal addi­tives and steel rein­force­ment. Roman con­crete was also mixed with sea­wa­ter, which caused the for­ma­tion of crys­tals with­in the mate­r­i­al that actu­al­ly strength­ened it as it aged — thus cement­ing, as one wag in the com­ments puts it, the Romans’ place in his­to­ry.

If you would like to sign up for Open Culture’s free email newslet­ter, please find it here. Or fol­low our posts on Threads, Face­book, BlueSky or Mastodon.

If you would like to sup­port the mis­sion of Open Cul­ture, con­sid­er mak­ing a dona­tion to our site. It’s hard to rely 100% on ads, and your con­tri­bu­tions will help us con­tin­ue pro­vid­ing the best free cul­tur­al and edu­ca­tion­al mate­ri­als to learn­ers every­where. You can con­tribute through Pay­Pal, Patre­on, and Ven­mo (@openculture). Thanks!

Relat­ed con­tent:

How Did the Romans Make Con­crete That Lasts Longer Than Mod­ern Con­crete? The Mys­tery Final­ly Solved

The Roman Roads and Bridges You Can Still Trav­el Today

How Did Roman Aque­ducts Work?: The Most Impres­sive Achieve­ment of Ancient Rome’s Infra­struc­ture, Explained

The Beau­ty & Inge­nu­ity of the Pan­theon, Ancient Rome’s Best-Pre­served Mon­u­ment: An Intro­duc­tion

Roman Archi­tec­ture: A Free Course from Yale

Based in Seoul, Col­in Mar­shall writes and broad­casts on cities, lan­guage, and cul­ture. His projects include the Sub­stack newslet­ter Books on Cities, the book The State­less City: a Walk through 21st-Cen­tu­ry Los Ange­les and the video series The City in Cin­e­ma. Fol­low him on Twit­ter at @colinmarshall or on Face­book.


by | Permalink | Comments (74) |

Sup­port Open Cul­ture

We’re hop­ing to rely on our loy­al read­ers rather than errat­ic ads. To sup­port Open Cul­ture’s edu­ca­tion­al mis­sion, please con­sid­er mak­ing a dona­tion. We accept Pay­Pal, Ven­mo (@openculture), Patre­on and Cryp­to! Please find all options here. We thank you!


Comments (74)
You can skip to the end and leave a response. Pinging is currently not allowed.
  • Joan says:

    In case we should ever want to do things in the best way pos­si­ble. Polit­i­cal inter­fer­ence notwith­stand­ing.

  • John holland says:

    Fact is we dont know how to make con­crete
    I patent­ed a method and dis­played same at a local trade fair.
    I showed how all equip­ment was based on an erknekus assump­tion.
    That all aggre­gates srex­did­desrnr and con­crete tgusxre­qires an adjustable fre­quen­cy and inten­si­ty of vibra­tion.

    A Ger­man com­pa­ny tried to stale my patent hence I have reject­ed the west in my delib­er­a­tions.

  • Big o says:

    The fuck you say? Spell check that garbage.

  • Christ Jones says:

    Haha­ha… that’s right

  • tanetovic says:

    Anoth­er thing that hap­pens nowa­days with con­crete is car­bon­a­tion… And all of the steel rein­force­ment when in con­tact with air will rust to the point of total dete­ri­o­ra­tion and dam­age the struc­ture

  • robert j. walters says:

    Yes indeed , SEAWATER was the secret how­ev­er , the salin­i­ty of sea­wa­ter dur­ing Roman times is dif­fer­ent from the salin­i­ty in mod­ern sea­wa­ter …

  • Ignacio Granados says:

    The prob­lem with this is that Romans did­n’t use “iron bars“in as we do now with rein­forced con­crete. So from this point of view, we’re even, about all because it’s hard to think they just want make eter­nal build­ings, as we’re not look­ing for that either.

  • Ralph says:

    If the block soaks in water, what hap­pens if it freezes?

  • Lee S says:

    Lol… Not many sen­si­ble com­ments here .. but I find this fas­ci­nat­ing, I’m sure the use of rebar has not much to do with the dif­fer­ence between cur­rent and Roman con­crete, just theirs was bet­ter.
    Brains 2000 ago were just the same as ours.. just with less screen time… Big up the Romans…

  • Robert West says:

    It was salt­wa­ter so I doubt hav­ing it freeze will be a prob­lem.

  • Lee S says:

    Not that much dif­fer­ent mate, you would­n’t taste the dif­fer­ence… Salt water might have a dif­fer­ence in the process… I don’t know, but the dif­fer­ence in salin­i­ty of sea­wa­ter between now and then next to zero.. at least when it comes to mix­ing con­crete!

  • Lee S says:

    Not that much dif­fer­ent mate, you would­n’t taste the dif­fer­ence… Salt water might have a dif­fer­ence in the process of mak­ing con­crete. I don’t know, but the dif­fer­ence id salin­i­ty of sea­wa­ter between now and then next to zero.. at least when it comes to mix­ing concrete!im sure its just the romans had a bet­ter recipe than us.

  • Brooke Radding says:

    “Roman” con­crete is a type known as poz­zolan con­crete.

  • Brooke Radding says:

    That is not true. Theirs was not “bet­ter.” It’s well known what it is. It’s Poz­zolan con­crete. If it were bet­ter it would be in indus­tri­al use today. Mod­ern con­crete mix­es are engi­neered and spec­i­fied in con­struc­tion for dif­fer­ent appli­ca­tions. I did have one pro­fes­sor say, how­ev­er, that it is still not under­stood well. “We only know two things for cer­tain about con­crete,” he announced.
    “It’s gray and it cracks.”

  • Dave says:

    I read recent­ly that the Romans used vol­canic ash in the mix with lime . The ash was cre­at­ing a crys­tal grow that made it hard­er and more resilient as it aged. The exact method of mix­ing or ratios is yet to be found .

  • Euwe says:

    Took the words right out of my mouth

  • Kraig Richard says:

    If the­o­ry valid, resist­ing acids of weath­er might like­ly be a com­bi­na­tion of dif­fer­ent rea­sons. Right bal­ance of var­i­ous par­ti­cle sizes, some kind of impu­ri­ty cre­at­ing a reac­tion. Maybe they added aged camel piss. It’s hard to believe whst we can’t say about their cement… if we can send a Tes­la (with­out Musk in it) into infin­i­ty and watch it for­ev­er with our star­ry eyed space scope.

  • Tony says:

    No shit. LITERALLY LMAO!!!

  • John boy says:

    U’s say toma­to we say to-mar‑o

  • Hugo Gallipoli says:

    exac­ta­mente, es así, la ten­den­cia esper­a­da es hacia la mete­orización o la desin­te­gración y no solo por los cam­bios de tem­per­atu­ra en el día, el mes, el año, sino que el aumen­to de la entropía es siem­pre algo espon­tá­neo, pero este con­cre­to cristal­iza cada vez más, ¿los alu­mi­ni-sil­i­catos se entre­cruzan?, segu­ra­mente pero ¿quien sabe?, hay que con­tin­uar estu­dian­do su estruc­tura inte­ri­or.

  • Brian says:

    Fun fact, the longevi­ty of the con­crete used as the base of the stat­ue of Lib­er­ty and the Brook­lyn bridge is because they were made of Rosendale cement, which is stronger than Port­land cement but takes longer to set up. You can find the old kilns used to process lime­stone in the Hud­son Val­ley, near and in towns like Rosendale for which the cement is named.

  • Joe Sickler says:

    I would like to learn more of this dif­fer­ent types of cement n how its all mixed n how each bond

  • Roger says:

    In Chica­go, they build infe­ri­or roads pur­pose­ly so they can rebuild again next year to keep those bud­gets in use, or they don’t get the mon­ey next year.

  • Sean Laughlin says:

    I live in Chica­go and con­firm your state­ment. Real­ly is ridicu­lous how poor­ly our roads are made here. 294 N at the Bensenville bridge being resur­faced 2 yrs after con­struc­tion?

  • Ibuyminefromtheman says:

    It real­ly comes down to water if you want high­strenth con­crete don’t put it in a mix­er a truck or a pump. Just bare­ly wet it. So it looks like wet sand at the beach. Then scoop it and pack it into the foot­ing. It will fin­ish the same as the poured con­crete but instead of being 5 to 9000psi its 20,000psi. Roman’s hand mixed used very lit­tle water and packed it in. You will get the same qual­i­ty out of a sack of quikcrete if you use the bare min­i­mum of water. Had mix and pack don’t pour the con­crete. More water equals much low­er strength. Whicm means cracked dri­ve­ways.

  • Anonymous says:

    Seri­ous­ly? Do u have anger issues?
    Or just a bet­ter speller than you syn­drome?

  • Dennis The Menace says:

    Very interesting…some of U guys (gals?) are real­ly fun­ny!! And talkin’ bout syn­dromes, some of U may have been pre­vi­ous­ly diag­nosed with “my pee pee is big­ger than yours syn­drome” lol 😆
    That said, I do believe that using water spar­ing­ly is crit­i­cal in get­ting the most outa ur con­crete!

  • James Bone says:

    Bad con­crete is why that con­do­mini­um col­lapsed in Flori­da last year or when­ev­er it was. I per­son­al­ly live in a 3‑story build­ing that was lit­er­al­ly “poured” and the out­er walls & floor are made of con­crete rein­forced with steel. The build­ing was poured in the 1970s so should still be good for a few more decades I would guess. But I don’t real­ly know…

  • OC says:

    Out of curios­i­ty, does any­one know what site is send­ing folks to our site/post today?

    Thanks for any insights,
    Open Cul­ture

  • James Bone says:

    Job secu­ri­ty for road builders…

  • James Bone says:

    I found your arti­cle in my Google News feed…

  • Heywood Jabloeme says:

    I’m new here.
    I opened Google on my phone and it was list­ed as an “inter­est­ing sto­ry I might like”. I have zero idea how arti­cles like this are cho­sen. Most often it’s for garbage I do not care about. This was cool though.

  • Ron says:

    To say our con­crete is infe­ri­or is false. The big secret, i believe, is car­bon… eas­i­ly acquired from a fire pit. The right amount of car­bon makes con­crete incred­i­bly strong. For ref­er­ence watch the doc­u­men­tary on the con­struc­tion of the Petronas (Twin) tow­ers in Kuala Lumpur.

  • Daniel Miller says:

    Sci­ence X digest

  • Doug says:

    I like to know what engi­neer­ing back­ground goes into test­ing the hard­ness of con­crete by scratch­ing it with a fin­ger­nail and tap­ping it with your knuck­le.
    I’ve had stale bread that would pass those tests.

  • David Unvert says:

    I did a sci­ence fair project in 7th grade using tap water, deion­ized water and sea­wa­ter to make con­crete and the sea water was the strongest which was oppo­site of my hypoth­e­sis

  • Clelie says:

    Where did you get the Vol­canic Ash and Pumice (Ingre­di­ents). I’d like to make some.

  • Mike Sanchez says:

    Either that or the fact that 50GB of hacked info on rul­ing elites was stored as pro­tec­tion by soft­ware founder.

  • BillyBaroo says:

    “More than a mil­len­ni­um and a half after its fall, we still look back with won­der on the accom­plish­ments of the ancient Roman Empire.”

    I real­ize that the appeal to ancient­ness is the hook for this arti­cle and that 1500 years is allot of time; how­ev­er the Roman Empire did­n’t come to an end until the fall of Con­stan­tino­ple in the 15th cen­tu­ry.

    Those who lived in the city called them­selves Romans and so did the Mus­lim invaders who final­ly con­quered it.

  • Withnail says:

    mod­ern cement does the same thing, fly ash from pow­er sta­tions is used.

  • Lord Piddlepuck says:

    The Romans made an almost eter­nal king­dom work but in the end, they had lunatics like Nero and Caligu­la rule their demise, the once infal­li­ble Empire now crum­bled to the ground, though hoist­ed upon their own petard, the wealthy seem to always cream to the top like sour milk, then the rich refuse to help any­one but them­selves, the same rack­et has been going on since Cain.

  • Damiel says:

    My Google feed rec­om­mend­ed this arti­cle.

  • Leo says:

    The use of fly­ash /carbon has also neg­a­tive impact on plac­ing con­crete and dura­bil­i­ty. Car­bon or fly­ash floats on water it weak­en­ing to top lay­er of con­crete when poured. Water is need­ed For work­a­bil­i­ty and mak­ing the con­crete lev­el.
    In my expe­ri­ence fly­ash or cabon will come to the sur­face and will cause scal­ing and Spald­ing in freez­ing con­di­tions.
    As far as I know or read in roman times in con­crete there was use of blood for use of airen­train­ment.
    Also intrest­ing is the state­ment that salt water is used,new con­crete has tobe free of sodi­um or cal­ci­um for mak­ing the new con­crete.
    Great dis­cus­sion thank you.

  • David says:

    Repub­li­cans make it, so it’s designed to fail. The Romans nev­er con­sid­ered mak­ing such a thing.

  • Kaden says:

    Repub­li­cans made it, because Democ­rats nev­er work and they hate progress.

    Now, what the HELL did you put a polit­i­cal state­ment on a non-polit­i­cal arti­cle? Was it about pol­i­tics? No, it was about cement.

    This teenag­er thought the arti­cle was inter­est­ing.

  • Thomas Edward Shultz says:

    What if the steel were com­plete­ly cleaned of rust, (or imme­di­ate­ly at the end of its pro­duc­tion process, before any rust could form), then coat­ed with some anti-rust com­pound?
    I worked in a ware­house, sell­ing steel repair for use in con­struc­tion. I know, and can cer­ti­fy, almost all steel rein­force­ment rebar is rusty as hell, when bought for con­struc­tion use. They don’t give a f___.

  • Thomas Shultz says:

    I wrote REBAR, not “repair”! Sh!t.

  • Bryan says:

    They do, or did, mar­ket an epoxy coat­ed rebar that solved the rust issues, but the price was pro­hib­i­tive and I don’t think it Caught on.

  • Mike says:

    This showed up in my stan­dard Google “news/stories” feed.
    Was under your open­cul­ture link.

  • Brian k says:

    @Thomas Edward Shultz

    My under­stand­ing is that sur­face rust on the steel at the time of pour­ing is of no con­se­quence because the acids absorb the sur­face oxi­da­tion which actu­al­ly helps the steel to bond to the cement dur­ing the cur­ing process. Once the con­crete has cured how­ev­er, if the steel begins to rust — that is a prob­lem because it results in the de-bond­ing of the steel from the con­crete and even­tu­al spalling. How does the steel start to rust once it’s embed­ded in con­crete? Mois­ture migrates into the con­crete due to its poros­i­ty. If this mois­ture is exposed to freeze/thaw cycles it can cause crack­ing. Then the spi­ral of dis­in­te­gra­tion begins: the cracks formed by freeze-thaw cycles intro­duce more mois­ture, more freez­ing induced expan­sion and crack­ing, and even­tu­al dis­in­te­gra­tion of the struc­tur­al integri­ty of the con­crete. Throw road salts into this process and you have a recipe for the US infra­struc­ture spend­ing bill.

  • Maria Hicks says:

    Me too, Google news!

  • Will says:

    From Google News aggre­ga­tor on Sam­sung fold home screen.

  • Kurt says:

    Google news feed for me

  • Zafiel says:

    Umm, I think you are all for­get­ting that these build­ings we cur­rent­ly see are their crit­i­cal infra­struc­tures. We have even few­er remains of nor­mal infra­struc­ture from their time. So to say that Roman con­crete in gen­er­al is bet­ter than our mod­ern con­crete is to say that high grade con­crete of the past is being com­pared to our nor­mal con­crete.

    The best com­par­i­son for our con­crete that would be used is prob­a­bly the con­cretes used in high­ly sen­si­tive mil­i­tary infra­struc­ture and gov­ern­men­tal build­ings.

  • pimalu says:

    Busi­ness are business…you know..

  • pimalu says:

    Bet­ter use stain­less steel bars

  • Some guy says:

    I found this page from my news­feed

  • John deforest says:

    Did u notice most spelling Nazis usu­al­ly have lit­tle or no sub­stance to con­tribute !

  • Aldarindi says:

    Thanks for an inter­est­ing and thought pro­vok­ing seg­ment.
    One of my engi­neer­ing teach­ers told me that the con­struc­tion of many of the ancient world’s con­crete struc­tures employed the delib­er­ate use of voids, emp­ty ampho­ra and such, to give the struc­tur­al advan­tages of mas­sive thick­ness, with min­i­mum expen­di­ture of mate­ri­als, and a struc­ture of least weight, for its dimen­sions.
    Thanks again this was fun.

  • J Moore says:

    Nice sum­ma­ry of the basic con­cepts of roman con­crete and the issues which sur­round it. These are dis­cussed in more detail on my father’s web site, romanconcrete.com and in his book “The Roman Pan­theon, The Tri­umph of Con­crete”. Its great that Mr. Kel­ly actu­al­ly made test sam­ples and showed the impor­tance of water ratio on con­crete strength and how mod­ern addi­tives can allow less water to be used while main­tain­ing work­a­bil­i­ty.
    My father, David Moore P. E., actu­al­ly attempt­ed to recre­ate roman con­crete itself using nat­ur­al mate­ri­als such as poz­zolan from Italy and lime and to sim­u­late the process of com­pact­ing to min­i­mize water usage. But he passed on in 2012 before he could fin­ish the research. He loved to study this mar­velous mate­r­i­al and the cre­ative Romans who made it!

  • Tamerin says:

    Lol what it seems like we could make con­crete on that lev­el it’s just cheap­er and more con­ve­nient not to.

    Say­ing there’s is bet­ter because com­mon­ly we use one that’s pret­ty cheap even though we have the abil­i­ty to make many dif­fer­ent qual­i­ties and kinds of con­crete is so sil­ly.

  • carlar says:

    found it on a flip­board email, some­one flipped it into Arche­types

  • Eric griffin says:

    Bout time some­one final­ly said it .there are some oth­er minor addi­tions you can add or sub­tract .but less water and pack­ing it .cal­ci­um And oth­er things add to more crack­ing .nev­er set fence posts in wet mixed concrete.just pour in dry bags.they will hard­en when they do.tgere won’t be that mois­ture col­lect­ing void that holds water between hard ened con­crete and post.from wind or fence move­ment it also depends on it’s use a less brit­tle con­crete mix that is much soft­er than a hard­er stronger on3 may last longer life it it was used in some­thing need­ing to Alex more

  • Eric griffinits not from anywhere with much knowledge.especially Craftsman's first hand type.m says:

    Bout time some­one final­ly said it .there are some oth­er minor addi­tions you can add or sub­tract .but less water and pack­ing it .cal­ci­um And oth­er things add to more crack­ing .nev­er set fence posts in wet mixed concrete.just pour in dry bags.they will hard­en when they do.tgere won’t be that mois­ture col­lect­ing void that holds water between hard ened con­crete and post.from wind or fence move­ment it also depends on it’s use a less brit­tle con­crete mix that is much soft­er than a hard­er stronger on3 may last longer life it it was used in some­thing need­ing to Alex more

  • Leni says:

    Learn from the best. In the case of Romans…they learned pret­ty much every­thing from Greeks and to a less­er extent the Etr­uscans.

  • Me says:

    Noth­ing worse than the gram­mar Nazi

  • Phat Khat says:

    Google feed for me, too.

  • Tim Kluxdal says:

    Google feed. Love arti­cles like this.

  • Jpm says:

    I hate to be the guy try­ing to chip a hole or work with such a hard concrete..but for struc­ture pur­pos­es I’ve worked on so many dif­fer­ent jobs..some con­crete amazes me how strong it. Is cure­ing the answer.maybe

  • Alex says:

    Work­ing on bridges the last cou­ple years I noticed all the rebar is epoxy coat­ed a light green…
    I’d guess this is to elim­i­nate the oxi­da­tion fac­tor and degra­da­tion of the con­crete…
    Alex

  • Steven says:

    A retired IRS exec­u­tive told me in the 90’s that over near­ly 20% of the fed­er­al tax dol­lar is graft from road con­struc­tion.
    It’sTexas as well. High­way 45 going thru Hous­ton, except for 1day, has been under con­stant recon­struc­tion.
    Sir, you are cor­rect. Crim­i­nal behav­ior is too often rerward­ed in these Unit­ed States

  • Mike says:

    Roman con­crete uses Poz­zolan­ic cement — not as we use, Port­land cement. Hence this vol­canic deriv­a­tive pro­duces for more coher­ent struc­tures.
    PS I haven’t a clue what you said in your mes­sage — too much gob­blede­gook — maybe you need to learn to use a spell check­er?

  • Kevin says:

    I wish some­one had updat­ed this thread. In Jan­u­ary of 2023 MIT researchers released their study show­ing the miss­ing ele­ment to Roman con­crete. That miss­ing ele­ment was the sub­sti­tu­tion of quick lime instead of slaked lime. To put it suc­cinct­ly, by uti­liz­ing the quick lime, The exother­mic reac­tion of quick lime and water cre­ates a high heat that forms lime clasts. These lime clasts react to water from cracks which may form over time due to dif­fer­ent rea­son­ings and when water goes through these cracks it inter­acts with the lime class which then bub­ble up and seal those cracks. This should be added to this thread as it is an impor­tant ele­ment and appar­ent­ly the final miss­ing piece.

Leave a Reply

Quantcast