ArtSci Entry: Roman Hydraulic (Marine) Concrete
Roman Hydraulic (Marine) Concrete
Introduction
The Romans first used hydraulic, or marine, concrete in
coastal underwater structures, probably in the harbors around Baiae (modern day
Baia) and Puteoli (modern day Pozzuoli) sometime around the end of the 2nd
century BC
Vitruvius, writing around 25 BC in his Ten Books on
Architecture, distinguished types of materials appropriate for the preparation
of lime mortars. For structural mortars, he recommended pozzolana (pulvis
puteolanus in Latin), the volcanic sand from the beds of Pozzuoli, which are
brownish-yellow-gray in colour in that area around Naples, and reddish-brown
near Rome. Vitruvius specifies a ratio of 1 part lime to 3 parts pozzolana for
mortar used in buildings and a 1:2 ratio for underwater work
In addition, evidence of opus reticulatum has been
found on concrete cores at a depth of 6m in some areas
Historical Materials and Methods
There are 4 ingredients for hydraulic concrete: rock
aggregate, lime, pozzolana (volcanic ash/pit sand), sea water.
Vitruvius explains that pit-sand (aka pozzolana) is the
preferred sand for concrete, however, river or beach sand is preferred in
stucco when it is thoroughly mixed with the mortar, while stone used as the
aggregate should be exposed to the elements for at least two years, and should
be appropriate for being exposed to water and the elements found within.
Travertine, which is a hard load calcium-based stone is good, but will crack
when exposed to fire. As such, it is a good option for being used in seawalls
and harbors underwater, where the chance of such heat is lessened
Slaked lime is the product of super-heated (1000F) limestone
or high-calcium-based seashells. “Lime made of close-grained stone of the
harder sort will be good in structural parts; lime of porous stone, in stucco”
Roman Concrete Construction Technique
The proportions for pit sand to lime is 3:1,
and for underwater work- river/beach-sand in a 2:1 ratio
Hot-mixing meant adding the lime directly to the dry mix,
then adding in the water instead of creating a slacked lime slurry (water +
lime) and then adding in the additional dry ingredients. Left over,
non-incorporated lime pieces would then slowly react to any water infiltration,
and in conjunction with the silica and alumina in the pozzolana, would continue
the crystallization process (called tobermorite) over time
Sample Roman concrete core
BAI.2006.03 (08/09)
Secca Fumosa pila
“The top of the core was 3.45 m below sea-level, the depth
of the core hole was 3.15 m, and the total length of core recovered was 2.9 m.
Very uniform concrete with many lime nodules. Generally, resembles the cores we
sampled from the experimental pila we constructed in the harbor of Brindisi.”
Harbor construction instructions
Five methods were used to place the concrete mix into the
water: 1) stakes driven into the ocean floor to create a box which was then
filled with the concrete (see instructions above), 2) boxes made on land
and filled, then sunk in the ocean, and 3) barges filled and then sunk. The
staked option is called a ‘cofferdam’ by Vitruvius, who adds two other methods,
the first being 4) a sloping platform being constructed into the water which
holds up large concrete blocks. The platform is then removed and the blocks
allowed to fall into the sea, creating a layer of large brick-like blocks
A/S Materials and
Methods
Concrete mix
1 part hydrated lime (1/2 powder and 1/2 clasts)
2 parts pozzolana
1 part roughly ground burnt brick
Seawater to bind into a thick paste.
Method- concrete
All dry ingredients were mixed together, then the seawater
was added until a thick, yet pourable, paste was formed. This was then poured
into molds. Small ‘blocks’ were made in a silicone tray that was heated in a
dehydrator at 95F for 10-12 hours. A larger ‘block’ was made in a resin mold
and placed in the same dehydrator. These became our sample ‘blocks.’
Method- harbor break-away
A plastic container was used for the miniature ‘world.’ Beach
sand was added to the container to create our replica land and slope, with a
base layer of rock and shells. Large ‘blocks’ were added to re-create method
4 before the sea-water was added, an example of such is found below.
Results and Discussion
The first two rounds did not have the brick aggregate. All
tests used the hot-mix method, with the dry ingredients being mixed before the
sea water was added.
Test 1) 1:2; lime clasts: pozzolana
In the first test, lime clasts were included. The resulting
blocks crumbled with slight pressure, though lime clasts were able to be seen
in the broken bricks similar to the historical sample.
Test 2) 1:2 lime : pozzolana
In the second test, the lime was ground fine before being
mixed in. The resulting blocks were more uniform in nature, harder, and showed
no lime clasts when broken apart. The texture was smoother, and had a more
clay-like slip texture. The top of the large block had calcium crackling. The
cross section showed the type of consistency we would expect from a smooth
stucco mortar, not concrete.
Test 3) 0.5:0.5:1:2 lime clasts : ground lime powder : brick aggregate :
pozzolana
The third test incorporated 1-part lime (half lime
clasts/half finely ground lime powder), 1-part crushed brick aggregate (both
fine and medium-fine size), and 2 parts pozzolana. These bricks were the
strongest of the three tests, with a smooth exterior and good distribution of
lime clasts and brick aggregate throughout the sample. Two batches were made,
one with small bricks and one with larger blocks. Both were dried in a food
dehydrator at 95F for 10-12 hours before being ‘cured’ in salt water for two
weeks.
The small bricks held up just fine. The larger block broke
apart when added to the seawater as the lime reacted. The mix was remade with
more water into the hot mix to allow for the lime to fully slake before being
dried and reused. This method allowed for the final product shown. The smaller
bricks were used to show the edge of a ‘road,’ leading to the ‘seawall’.
Final model: Dry, top and side view
Final Model: Wet, top and side view
While the large blocks showed calcium hydroxide (lime) skins, they did maintain their cohesion until they were physically snapped in half to create half blocks to fit the tank. Even though these blocks broke under force while being placed within the tank, they blocked water infiltration to the sand in between the 'road' and the 'seawall'.
References
Primary References
Oleson, J. P., Brandon, C., Cramer, S. M., Cucitore, R.,
Gotti, E., & Hohlfelder, R. L. (2004). The ROMACONS Project: A Contribution
to the Historical and Engineering Analysis of the Hydraulic Concrete in Roman
Maritime Structures. International Journal of Nautical Archaeology, 33(2),
199-229.
Vitruvius. (1914). Ten Books on Architecture, Book 1 and Book
5 (Vol. 1). (M. Morgan, Trans.) London, UK: Harvard University Press.
Secondary References
Blackmann, D. (1982). Ancient Harbours in the Mediterranean,
Part 1. International Journal of Nautical Archaeology, 11, 79-104.
Brandon, C., Hohlfelder, R., & Oleson, J. (2008, April
20). The Concrete Construction of the Roman Harbours of Baiae and Portus
Iulius: The ROMACONS 2006 field season. International Journal of Nautical
Archaeology. Retrieved from https://web.uvic.ca/~jpoleson/New%20Material/Baia%202006/Baia%202006.htm
Department of Civic and Environmental Engineering,
University of Wisconsin-Madison. (2023). Roman Concrete. Retrieved from Ancient
Engineering Technologies: https://ancientengrtech.wisc.edu/roman-concrete/
Lechtman, H., & Hobbs, L. (1986). Roman Concrete and the
Roman Architectural Revolution Ceramics and Civilization. In W. Kingery (Ed.),
High Technology Ceramics: Past, Present, Future. American Ceramics Society
(Vol. 3).
Seymour, L., Maragh, J., Sabatini, P., di Tommaso, M.,
Weaver, J., & Masic, A. (2023, January 6). Hot mixing: Mechanistic insights
into the durability of ancient Roman concrete. Science Advances, 9(1).
doi:10.1126/sciadv.add1602
Tertiary References
Campbell, P. (2017, December 14). New underwater discoveries
in Greece reveal ancient Roman engineering. Retrieved from The Guardian:
https://www.theguardian.com/science/2017/dec/14/new-underwater-discoveries-in-greece-reveal-ancient-roman-engineering
Comments
Post a Comment