One of the most important considerations when you build anything is the material. It impacts everything from the look and feel to the more fundamental characteristics like weight and strength. But maybe the most significant aspect of a construction material is its cost. Infrastructure is not made to be glamorous, and it’s often paid for by you and me through taxes, so we like to keep the costs down. And there’s one construction material that’s cheaper than just about anything else out there: dirt.

You’ve probably don’t think much about the strength of the soil beneath your feet, but some of us have dirtier minds than others. Just about every structure out there sits on the ground. And anyone who’s ever built a sand castle knows that soil is not all that strong. So if we want our buildings and bridges and pipelines and roads and anything else that sits on the ground to keep standing, and especially if we want to use earth itself as a construction material, we’re going to need some geotechnical engineering.

Soils are frictional materials. Rather than being held together by molecular bonding like steel or by a binder like the cement in concrete, their strength almost completely depends on internal friction between the soil particles themselves. Remember back to your physics class that the frictional force is proportional to the normal force. Push harder perpendicular to the sliding plane, and you increase the amount of force required to slide the two materials. If we want to avoid sliding, the frictional force can be considered the shear strength. The more friction, the more strength against shearing. Just like the simple block on the plane, the shear strength of soil depends on the internal forces too. But unlike that example, soils have an infinite number of potential sliding planes all at once.

Let’s look at a sample of soil and apply a vertical force. If you analyze a horizontal failure plane, our force is completely perpendicular, or normal, so it is increasing the shear strength of the soil. But if we look at an angled plane, things change. On this plane, the force is partly acting normal, increasing the strength, but it’s also partly acting in parallel to the plane increasing the shear stress. The steeper the angle of the failure plane, the more the vertical force contributes to shear stress and the less it adds to the shear strength. If the shear stress exceeds the strength, sliding occurs and we say that the material has failed. This is why granular materials generally can’t stand vertically. The weight of the material itself is enough to cause a shear failure along an angled plane.

Pour out some sand on a table, and you’ll notice that the pile forms a slope. The angle of this slope is called the angle of repose which is the steepest angle at which a soil can naturally rest. In other words, this is the slope at which the shear stresses within the soil due to its own weight are exactly equal to the shear strength caused by internal friction. Any steeper and the soil will slide. Let’s look back at our sample of soil. If we put the sample back in the ground, now it’s surrounded by additional soil that can apply horizontal pressure. This is called the confining pressure, and it helps to balance out vertical forces like the weight of the soil itself. This confining pressure is the reason that a granular material can be stable at a slope, but usually won’t be stable vertically.

This can be a problem if you’re trying to build an earthen structure for two reasons. First, it takes about twice as much material than if you’re using something that can stand vertically. And second, is space. In crowded cities, space is at a premium. If you’re building an earthen structure, every foot that you go up in height, you have to go out that far as well, or even further. So what’s a geotechnical engineer to do? What if there was a way to add confining pressure to the soil, without having to build on a slope. Behold, reinforced earth.

Just like rebar in concrete, you can create an incredibly strong composite material with soil just by adding reinforcing elements. A wall created in this way is called mechanically stabilized earth, or MSE. And if you look closely, MSE walls are everywhere. I did a quick demonstration of how this works. I cut up some circles of paper towel and layer them into the sand in a cup. Without any reinforcement, the wet sand can stand up vertically, but as soon as you apply a load, you get failure. Even just a few discs of paper towel to reinforce the soil allow the sand to hold up a 15 pound weight. So what’s happening here? The tension in the reinforcement is generating confining pressure in the soil. This pressure acts perpendicularly to the failure planes, increasing the shear strength of the sand.

Building an MSE wall in real life works exactly the same way, and they are primarily used in highway projects, especially on the approaches to elevated roadways. Compacted soil is added in layers with reinforcing elements in between each layer. Most MSE walls have a facing of interlocked concrete panels usually with some kind of decorative pattern, and these facing systems are what make them so recognizable. Any time you see a vertical wall of tessellated concrete panels, you can almost be sure that there’s reinforced earth behind it. By the way, the only purpose of these panels is to look nice, and keep the soil on the edges of the wall from raveling. The wall would be completely stable without the concrete facing, just not as pretty.

So how strong is an MSE wall. The simple answer is stronger than you would think. I tried one more demo. I built an 8” cube out of plywood. Just like the cup, I layered in sand and squares of reinforcement. For this first test, I used scraps cut from an old t-shirt. This part is not to demonstrate the strength of MSE, but rather to show that the soil really is transferring load into the reinforcement. Engineers often use exaggerated response in drawings or demonstrations to help get a feel for how forces are acting in the system. The t-shirt material is stretchy, so when the vertical load is transferred into the reinforcement, it spreads out and deforms. I wanted to demonstrate this because it may not be obvious in the next example.

For the next test, I used pieces of fiberglass window screen as reinforcement. This is a much stiffer material that does not deform under load. Under about 70 pounds it didn’t budge. Under my weight, even bouncing up and down, it didn’t budge. Let’s try something heavier. For the sake of science, we should probably have a control test with no reinforcement. But this is an engineering channel, not a science channel, and we all know what would happen to a block of dry sand under a car wheel. The wheel load on my car is probably on the order of 600 pounds and you can barely even see movement as the weight of the car is transferred to the cube. Unfortunately, I don’t have a hydraulic press, so my Mazda grocery hauler is about the heaviest thing I could think of to test the homemade MSE cube, so I moved on to dynamic loading I dropped this 20 pound barbell from about 6 feet up to simulate what would happen if you drop a 20 lb weight on the cube from 6 feet up. Almost no damage. In fact if I had a facing system to keep the edges intact, you probably wouldn’t have known the difference.

Dirt was probably your first construction material. We are born geotechnical engineers, trying to build taller and stronger earthen structures from probably even before we could talk. With a little bit of reinforcement, we’ve transformed this dirty propensity into a simple, inexpensive, construction material that you probably drive over every day. Thanks for watching, and let me know what you think.