The following is based upon an excerpt from a letter by Michael Grutzeck in the journal of the American Chemical Society.
Concrete does not dry; it hydrates. The cement grains dissolve in water, and when the resulting solution becomes supersaturated with respect to a variety of hydrous calcium aluminate hydrate and calcium silicate hydrate phases, nucleation and precipitation occur. As more hydrates form, the previously plastic and flowable concrete mixture begins to stiffen (sets) and then continues to harden for days, weeks, and even years.
Concrete is a material developed over thousands of years largely by trial and error and until recently, poorly understood. Only a tiny fraction of today's users have any understanding of the chemistry. This investigation attempts to provide some understanding.
Concrete is usually made by mixing Portland cement, sand, and gravel with a minimum of water to create a very thick slurry, a kind of colloid called a sol. Colloids are incomplete combinations of two of the three phases of matter: gas, liquid, and solid. A sol is a colloid formed with solid particles suspended in a surrounding liquid. Like in the more familiar sol, skim milk, the solid typically blocks passage of light making even dilute sols opaque, appearing cloudy. (This is contrasted with solutions such as tea in which the solid dissociates into molecules small enough to allow the passage of light.) The sand and gravel are less expensive materials which often add strength to the concrete by providing diverse structures in the crystal structure.
As Grutzeck mentioned above, the cement itself forms a hydrate. Many chemical substances have uneven distributions of electrical charges which attract water. In low enough concentrations this allows such small molecules to be surrounded by water and thus form liquid solutions. At higher concentrations these molecules surrounded by water molecules form orderly crystal lattices, packing together into solids known as hydrates. Polar chemical bonds hold the water to the unevenly distributed electrical charged molecules. These are generally weaker than ionic and covalent bonds which attach the other atoms together in the solid. As a result, relatively gently heating can often break the polar bonds resulting in a separation of the water from the atoms more tightly bonded. Such a process is used to form Portland cement. Suitable rock is mined from natural deposits in the ground and heated in kilns driving off the water and leaving the dehydrated cement. This is shipped dry to near the location of use. When needed the dry cement is mixed in the desired proportions with dry sand and gravel before a measured amount of water is added. As the cement slowly dissolves in the water, the slurry is typically poured into molds of the desired shape. Tamping and otherwise vibrating the sol allows air bubbles to float out and causes the cement to flow around the rock and sand filling the mold to its surfaces. Over time as the cement continues to dissolve, the water becomes supersaturated (having a concentration in excess of the maximum soluble) and the excess hydrates begin to crystalize, setting into a solid structure which has the intended structural strength. Because of the complex crystal lattice, the process is slow but results in more strength and permanence than simpler crystal lattices. The resulting solid is known as concrete.
Chemical additives known as water reducers strengthen concrete because they allow one to produce a flowable concrete with significantly less water than a similar mixture made without such additives. As a class of chemicals, these additives tend to alter the surface charges present on the anhydrous (dry, without water) cement grains, which in turn causes them to repulse each other rather than agglomerate and thicken the mixture. As the amount of water is reduced, so are the water-filled spaces between the cement grains. With smaller distances to bridge and less space to fill, a given ad-mixture-enhanced concrete will harden and gain strength more rapidly than its more water-rich non-treated counterpart.
Professional scientists often have to formulate their own experiment. One of the hardest tasks is to find a question which one wishes to answer. This is often accompanied by library and internet searches to determine what is already known. Sometimes the question's answer will be totally unknown while in other cases one might choose to try to verify or extend what is believed to be known. After the question is chosen, the scientist must determine what procedure will possibly provide information needed to determine the sought answers. The procedure may be adopted or modified from what had been done previously or may be very original. Typically the first attempts may fail for a variety of reasons resulting in need for revisions and additional attempts to achieve the desired answers. Anticipate that this investigation will NOT go smoothly but will require time to ponder what went wrong and what revisions might avoid the difficulty and produce better information.
Communicating technical information such as observations and findings is a skill used by scientists but useful for most others. If you need course credit, use your observations in your journal to construct a formal report.