Making the Recycled Glass Test Bricks

We have initially chosen Redart Fireclay as the standard clay for this project.  We received dry clay samples from two brick manufacturers in California, and tested them for plasticity, grog content, water requirement, and firing temperature.  The brick manufacturers’ clays are proprietary, but Redart, made by Resco Products of Pennsylvania, was the best commericially available surrogate we could find.

For the brick mixes we tested, the total sand and grog was about 50 percent by dry weight.  Therefore, our initial mix is
  50 percent dry Redart Fireclay
  50 percent 12 mesh glass

16 percent water by dry weight of clay plus glass seemed to create a pressable mix.  One-inch thick solid bricks are desired.  The formula for one brick is:
  800 grams 12 mesh glass
  800 grams dry Redart Fireclay
  256 grams room temperature water

Here’s the procedure:

Weigh 800 grams of 12 mesh glass into a mixing bowl

Add 800 grams dry redart Fireclay

Blend the dry materials together, then add 256 grams of water and mix by hand

Form the wet glass into a loaf, and place it into a wooden mold

 Press the clay mass into the mold by hand

Put a lid on the mold

Press the brick in a press made from a bottle jack

Remove the brick from the mold and place it on a warming plate with a 200 degree F surface

When the brick is dry, put it into the kiln for firing

Initial Absorption Tests

The first task is to determine a firing profile that will result in a brick meeting structural standards.  5 percent overnight absorption was chosen as the initial target.  5 percent absorption will exceed ASTM standards for freeze/thaw for bricks.  Absorption tests were carried out by weighing the brick, then putting it under water overnight, then wiping off the surface and weighing it again.  Absorption is the difference between the two numbers divided by the dry weight.

The dried bricks were subjected to the following standard profile:

2 minutes to 625 F
30 minutes at 625 F
4 minutes to 1832 F (this max temp varies)
30 minutes at 1832 F (this soak time varies)
The maximum temperature and the final soak time were varied while making pairs of bricks.

Here are the initial results:

The left scale is Absorption, the bottom scale is soak time in minutes

 At 1868 F, the granules of glass were beginning to pop out of the brick.  This temperature was considered to be too high.  The initial firing profile to focus on is holding for 60 minutes at 1850 F.


Let’s take a step back and review why we’re doing this.

The traditional use for recycled glass, melting it back into new containers, requires that the glass be collected, sorted by color, and processed to the standards of the container manufacturers.  Environmentally, it’s a plus because re-melting old containers to make new ones saves about ten percent of the energy it takes to make a new container.

In addition to the energy savings, using recycled glass reduces emissions.  When virgin batch materials are melted, they lose about 15 percent of their initial weight up the stack.  Most of the emissions are in the form of nitrous oxides and carbon dioxide. Not so long ago we thought carbon dioxide emissions were benign.  Now we know better.  At least most of us do.  

Re-melting recycled glass emits almost zero emissions. The only measurable emissions come from burning off the organic contaminants (labels), which are less than one percent of the processed glass.

Over the last fifteen years, thanks to minimum content laws in California, the fiberglass industry has also become a major user of recycled glass.  The advantage of using recycled glass in fiberglass is that it doesn’t need to be color sorted.  One disadvantage is that it needs to be crushed finer.  And it still needs to be quite clean, or it can ruin very expensive equipment in the melting and spinning of fiberglass.

A relatively recent development in the glass processing industry has been the installation of automated optical-pneumatic sorting equipment.  The sorting machines eliminate the need for hand-sorting glass, and do an excellent job of getting rid of inorganic contaminants like ceramics.  The automated equipment, however, creates a new large waste stream.  To sort the glass, first the system breaks the bottles into pieces smaller than one inch.  Then the optical equipment is able to pick out contaminants and colors down to as small as 3/8-inch.  But when you break a bottle, some of the pieces are inevitably smaller than 3/8-inch.  So now the large glass processors are generating thousands of tons per year of 3/8-inch and smaller glass mixed with labels and small pieces of aluminum. This project is partially about developing uses for that material. The ceramic processes described below can tolerate a high level of contaminants.

Using recycled glass in either container manufacturing or fiberglass manufacturing presupposes the existence of a container or fiberglass manufacturing plant within a reasonable distance.  Years ago Argonne Labs conducted a research project that determined the breakeven point for shipping recycled containers by truck at about a one hundred mile radius, in terms of the energy used in transport vs. that saved in manufacturing.  And transport fuel prices being what they are, that relationship may be worse today.

In the Pacific Northwest, where this is being written (WA, OR, ID, AK), in over 500,000 square miles of territory there are only two container manufacturing plants and no fiberglass manufacturing.  Recycled glass is a loser for recycling programs throughout most of this region.  So to sustain collection programs, alternative uses are required.  Over the years many uses like using glass as a construction aggregate, filtration medium, blasting abrasive, etc., have been proven to work technically.  Reports on some of those applications can be seen at

Problems with the “aggregate” type applications include the processing that’s still needed to turn the glass into a specification material, and the low comparative value of natural aggregates.  Collecting glass, processing it into sand, then selling it for less than $100 per ton sounds like a tough proposition.

So what else can be done with the glass?  How about using it as a raw material in other ceramics manufacturing?  The container plant has already invested large amounts of energy to turn the mostly crystalline virgin raw materials into an amorphous structure.  That embodied energy can then be used in other kiln processes.  The container manufacturing plant operates furnaces at up to 2500 degrees F to melt the virgin materials.  Once it’s glass, at temperatures as low as 1250 degrees it can be softened and made into other things.

That’s what this project is about.  Estimates for the energy it takes to make a brick vary widely.  Building for Environmental and Economic Sustainability  (BEES 3.0 – NISTIR 6916), October, 2002, page 57, estimates the manufacturing energy to make a brick at 1238 btu/lb.  Energy and Ceramics (Elsevier Press, 1980), the article “Economy in Fuel Consumption of Bricks Manufacturing Plants,” by E. Facincani, put the energy consumption at 1760 btu/pound.  A NICE3 study estimated a high-efficiency natural gas kiln’s energy usage at 2180 Btu/pound of product  (“Low-Thermal-Mass Kiln Installation At Pacific Clay Products, Inc.,” California Energy Commission Contract Number: 400-96-019).

    BEES        1238 btu/pound
    Facincani    1760 btu/pound
    NICE3    2180 btu/pound

Most interesting, the article by Facincani estimated the endothermic energy at 400 btu/pound.  The endothermic energy is the unrecoverable energy, because it is consumed in the chemical transformation of the raw material.

As early as 1972, during the first energy crisis associated with the politics of the Mid-East, the U.S. Bureau of Mines sponsored a study entitled “Waste Glass as a Flux for Brick Clays” (Report of Investigations 7701), by M. E. Tyrrell and Alan H. Goode.  They reported that energy could be saved by using recycled glass in brick manufacturing.

 More recently, a project sponsored by WRAP in the UK came to the same conclusions.  The WRAP report can be downloaded at
In a test, that report says that making bricks without glass used 2825 btu/pound, while with 5 percent glass, 2249 btu/pound was needed.  Energy consumption per pound is a function of scale as well as temperature, so the WRAP report may not be a reasonable comparison.

All of the studies so far do to things:
  they focus on very fine (200 mesh and finer) glass, and
  they treat glass as a flux.
A flux is defined in McGraw Hill Dictionary of Scientific and Technical Terms as “A substance used to promote the fusing of minerals or metals.”  The definition implies that the minerals are doing the fusing, and the flux is the promoter.  The promotion can be by creating better thermal conductivity between grains or acting as a catalyst of some sort.

There are several problems associated with using only 200 mesh glass.  Among them are
  there are no large-scale producers of 200 mesh glass
  compared with mined clay, 200 mesh is expensive to make 
  at higher percentages, 200 mesh glass decreases the workability of the clay mixture.

Instead of treating the glass as a fine flux, what if we think of it as an aggregate and a glue?  Brick manufacturers already add up to 50 percent of a combination of sand and coarse temperature-stable pieces called “grog” to reduce shrinkage, reduce water requirement, and accelerate drying.  

So add glass in a gradation similar to grog and sand, at 50 percent of dry weight.  The mixture forms and dries like a normal brick blend.  In fact it needs somewhat less water because the glass particles absorb no water, compared with grog particles, which tend to be porous.

At room temperature, the brick clay forms and dries normally.  When heated, however, at temperatures as low as 1250 F, the glass particles begin to fuse together, giving the brick body strength.  To get good strength, the brick needs to be heated considerably higher, because the viscosity of the glass must be reduced enough to surround the clay particles.

But the clay particles never reach their reaction temperature.  The function of the clay has been to give the clay workability and green strength at room temperatures, then to hold the brick in shape as the glass particles fuse to final strength.  That’s the principle behind this project.  I call it the “principle of reactive aggregate,” because the glass acts as an aggregate at low temperatures, then becomes the reactive material at higher temperatures, reversing the normal state of affairs, where the aggregate is the inert material, as it is in concrete or normal brick manufacturing.

Early in this project we sent a cover letter, sample tiles, and a summary of this principle to ceramic manufacturers in California.  With two exceptions, we got no traction.  In fact, on trying to visit three manufacturers to speak with them in person, we basically got the “bum’s rush” out the door.  This is understandable, I suppose.  If your entire career was predicated on certain chemical reactions happening at certain temperatures, and someone came along with an idea that threatened the importance of those reactions, you might respond pretty defensively as well.

One brick manufacturer gave a serious trial, but decided to focus on the finer glass only and insisted on trying to use old technology furnaces.  One thing about using glass is that old-fashioned brick firing on rail cars stacked four feet high with multiple day firing schedules won’t work.  Glass is more sensitive to temperature and time variation than clay, so furnaces made with lightweight insulation and good instrumentation are necessary.  If the upside is saving half the energy and more than half the time, then welcome to the 21st century.

One tile manufacturer in California gave us a fair hearing.  As a result of that contact, we are currently developing a monolithic glazed countertop that will be five-feet long, made from a single slab of recycled glass and ceramic, and fired in a few hours start to finish.

There will be a lot more about energy issues as the project proceeds.  Now on with the testing.

next page