Casting Glass




About: An engineer, seamstress, cook, coder, and overall maker. Spent a summer at Instructables; got a degree in E: Neural Engineering at Olin College; made a microcontroller (; now thinking about climate...

This was a project I did in a group for my Materials Science class. Since I had access to a laboratory full of materials testing equipment, I'll give you some extra scientific data. But if you happen to have access to casting equipment at home, you should be able to produce similar effects!

Necessary materials:
-something to smash glass
-materials to make an investment mold (wax, investment plaster, flasks, a vacuum sealer)
-something big enough to hold glass over investment at around 850 degrees C
(Please read through the steps first- this materials list is probably not comprehensive.)

I'm not an expert- none of us had tried working with glass before this project. If you have advice, please comment!

Step 1: Make Wax Pieces

Make positives of whatever it is you plan to cast. Sculpting, carving, or braiding thin wax pieces is easiest.

Another option is to copy an existing object using silicone molding and a wax injector. I used mold-making rubber, but here is an instructable for a different way to make silicone molds. Make sure there is an opening from the side of your mold to your object so that you can inject wax with the wax injector. I made several wax copies of a phoenix figurine using the silicone mold pictured.

The objects which cast best in glass for us were the solid figurines, such as the animals pictured. However, some of the more delicate objects (some rings, a phoenix) also came out well.

Attach your objects to a wax sprue.

Step 2: Make an Investment

Place your sprue into a stainless steel flask and seal the holes. You will pour your investment into the flask to completely cover the wax pieces.

Follow the instructions for your investment mix to properly mix and set your investment.

Put the investment in the furnace to melt out the wax, leaving holes for your glass to fill.

Step 3: Gather Base Glass

Find some glass. You can purchase glass, or you can repurpose old glass. In this project, we used four different types of glass. I have details on them below.

While it's helpful to do analysis on the glass- particularly thermomechanical analysis- I assume you don't have access to a full lab.   The important thing you need to know is the softening temperature of your glass, so that when you cast, you can make sure to be above it. Internet searching your glass type will probably yield decent results.

Wine bottle glass
This glass is soda-lime container glass. This is the most common type of glass currently in use. It's a nice green color, and perhaps you can see the ideological appeal behind turning empty wine bottles into art and jewelry.
Composition data was taken using energy-dispersive X-Ray spectroscopy (EDS) on a scanning electron microscope (SEM). Our wine bottle sample was 50% Silicon, 22% Oxygen, 15% Calcium (lime) 9% Sodium (soda), 3% Carbon, 1% Aluminum, and 1% Potassium. (Compositions listed by elemental weight %. My percentages won't add up perfectly because they are rounded.)
When subjected to thermomechanical analysis (TMA), the sample showed a softening point of 641C.

Pyrex made its name in glass for laboratories by using borosilicate glass, which has a very low coefficient of thermal expansion (so it won't explode). We used Pyrex cookware, and found it to be more of a hybrid borosilicate-soda-lime glass.
Composition: 47% Silicon, 22% Oxygen, 10% Calcium, 10% Sodium, 5% Boron, 4% Carbon, 1% Mg, 1%Al.
TMA showed a softening point of 660C.

Red  and Purple Gaffer's Glass
The red and the purple glasses are glassblowers' glass. The red seems to be potassium silicate glass, and the purple a lead glass. Lead is added to glass to increase the refractive index- to make it sparkle more. Examples of leaded glass include some cut crystal stemware and Swarovski crystals. I'm not sure what the potassium content does to the red glass- brief internet research suggests that it may be a strengthening agent. The absence of lime in the sample may also help it to avoid devitrification (where heated glass develops crystal structure and becomes opaque- see Step 8: Clean and polish).
Red glass composition: 37% Silicon, 16% Oxygen, 11% Potassium, 8% Sodium, 4% Carbon, 3% Boron
Red glass TMA: softening point at 631C.
Purple glass composition: 47% Lead, 25% Silicon, 11% Potassium, 8% Oxygen, 6% Barium, 2% Carbon, 1% Sodium.
Purple glass TMA: 627C

Step 4: Smash Glass Into Pieces

Take your pieces of glass somewhere safe, wrap them up in fabric, then hit them multiple times with a hammer. We used a blacksmith's hammer and worked on a paved surface.

Be very careful- note in the picture of red glass that the fabric has cuts in it. Our fabric was extra-tough canvas, and this just shredded it.

Use gloves, maybe a mask if you're worried about dust. Wear shoes. Pour your smashed glass into safe containers to transport it.

IMPORTANT: Don't smash your glass too small! We overdid it. You want inch-sized cubes at the smallest. If you make chunks that are too small, the opacity of your glass will decrease.

Step 5: Melt Glass Pieces

Put your glass in a clean graphite* crucible. Heat in a furnace until glass pieces all meld together. This should be around 800 degrees C for a long time, or closer to 1000C for a short time (but no  promises that your glass will be pretty). 

Let glass cool in crucible.

It is also possible to skip this step if you can find a different way to position your glass pieces directly over the opening to your investment.

*We used graphite, but you probably want to find something else. Ceramic, perhaps? Maybe stainless steel? Our graphite crucibles ended up disintegrating, which made a huge mess.

Step 6: Kiln Cast Glass

Now you need a furnace big enough to hold your investment in the flask and your glass piece over it. We simply upended the crucibles full of glass over the mouth of the investment.

Close furnace and turn the heat up to 800C.
Leave it in there for six days.

We also put some little ceramic boats with extra glass chunks in the furnace, just so that we'd have some visual feedback.

(We did 900C for 12 hours, but our glass didn't turn out very pretty. The six-day figure is from this source:

The last picture shows our graphite crucibles disintegrating all over the furnace. That's why you should use a different material, I guess...

When you're casting metal, you can just quench your investment in cold water and it will pop right off, giving you access to your pieces. However, glass cracks when cooled too quickly, so you'll have to let it slow cool. Turn your furnace off and let it sit until it returns to room temperature.

Step 7: Recover Glass Pieces

When your investments are cool, remove them from the furnace. Put them somewhere that can get messy.
Wear a dust mask so that you don't breathe too much investment powder in.

Scrape at the investment around the edges of the flask. It should be pretty easy to scrape and make a big powdery mess. If you can, try to scrape around all of the investment until it dislodges from the flask. If not, just dig down into the investment to treasure hunt for your pieces. Be careful- depending on how delicate your pieces are, you could accidentally break them.

Once you're out of the flask, put your pieces in a big sink and run water over them while continuing to scrape and clean. You're trying to remove all of the investment from your glass pieces.

Once you're down to just glass pieces and glass sprue, you'll want to take your pieces off of the sprue. I used a tabletop diamond bandsaw. (Note to my friends: if you ever want to get me an expensive present, this is it.)
If you don't have one of those, try scoring and very carefully breaking your pieces off of the sprue, then sanding down the sharp edges.

Step 8: Clean and Polish

Wash your pieces with soap and water. Use a brush to get into the crevices.

If you're lucky like me, you have access to an ultrasonic cleaner. If that's the case, put soap and water in that, throw your piece in, and let it vibrate for five minutes. Rinse and repeat for all pieces. Then replace your soap and water with a coarse polish, vibrate and repeat. And one more time with a fine polisher (I used an oil-based diamond slurry). That should get your glass looking lovely!

For flat sides, you can also try grinding with wet fine-grit sandpaper.

Our pieces, as you'll note from the pictures, have a distinct non-glass quality to them. Specifically, they're opaque.
We're not entirely sure what happened to cause this, but since our processing was rather messy, there are a number of possible points where this could have occurred:
-Melting could have introduced debris into the material. Our crucibles weren't perfectly clean, so little bits of dust and graphite got mixed in.
-Kiln casting occurred at a temperature higher than strictly necessary for slumping. We were trying to save time, but we might have introduced defects into our product.
-Devitrification could have occurred in our glass- crystals could have formed. Clear materials, such as glass and clear plastic, have amorphous microscopic structures, which means that they have no crystal structure. Sometimes, when glass is cast, it forms a milky, opaque surface, where the glass has crystallized around foreign particles.

The diamond slurry polishing should have removed any surface devitrification, but since the glass was mixed with foreign debris throughout in the melting step, this isn't enough evidence to cross off devitrification as a possible source of opacity. So I took samples of each type of cast glass to the SEM to look for crystal structure.

Two of the pictures above are through the SEM. They show the red glass at 60X and 500X magnifications. No clear crystal structure emerges, because the surface is too textured. The glass looks porous- there are hollows for many tiny bubbles within the glass. 


Step 9: Show Off Your Homemade Glass Pieces!

If you made anything, be proud! This is a really difficult and time-intensive process.

I'd love to see pictures of other people's homemade cast glass!

I hope I didn't bore you with the science :)



    • Growing Beyond Earth Maker Contest

      Growing Beyond Earth Maker Contest
    • Classroom Science Contest

      Classroom Science Contest
    • Stone Concrete and Cement Contest

      Stone Concrete and Cement Contest

    23 Discussions


    2 years ago

    I use 'quenching' (thermal shock) to make smaller chunks of glass out
    of bigger ones. One interesting feature of glass is that it has high
    surface tension so if you have a chunk of glass with a lot of fractures
    in it, you can remove the fractures by heating the chunk up enough
    without the chunk melting. Unfortunately, I use a microwave kiln so I do
    not have temperatures to talk about - only # of minutes in kiln. The
    3rd skull from the left, top row was made from a bunch of thermal
    shocked glass pieces that were heated up enough to remove the fractures
    from the individual chunks w/o actually melting them - in the process,
    'welding' the individual chunks into an interesting cubist-like form.

    Chris Logan

    6 years ago on Step 6

    Cooling metal too quickly can generate cracks. But almost always leaves a less-desirable molecular structure.

    Don't cool it faster than you have to.

    5 replies
    Chris LoganSelkeyMoonbeam

    Reply 6 years ago on Step 6

    Not unless you have very specific requirements for material properties.

    Cooling iron, zinc, steel in cold water will leave microfractures throughout the surface if it doesn't shatter the part outright.

    Cooling aluminum, magnesium results in a stupidly soft and unmachinable alloy, regardless of your starting material. 6061 after a water quench ends up with properties like 1001.

    Just let your parts air cool. You avoid putting unnecessary molecular and physical stress on them.

    Chris LoganChris Logan

    Reply 6 years ago on Step 6

    Air cooling avoids that stress. And results in a part that is easier to machine or handwork finished.

    Quenching will cause immediate fracturing in the glass, regardless of type. Controlled regulation of the temperature from the annealing point, through the strain point, and then down to room temperature is required. Rapidly cooling the glass though the strain point will cause a tempering of the glass but this is not done with the casting method as the imperfections in the surface would negate the process. Each glass will have its own temperature and time range requirements and continually increases with thickness of the casting.

    Great Hacker

    3 years ago on Introduction

    As a glassblower, I know a thing or two about glass and I would definitely have to say it was devitrification that made the glass crystalline. Because it was in such small molds with negligible convection, the glass would quickly become crystalline. I would recommend let the glass flow into the mold for about 3-4 hrs and then ramp down at 100f/hr.


    5 years ago on Step 9

    Bore us? I thought it was fascinating. I've just started working with glass. I'm using low melt 90 & 96 coefficient frit (very finely ground glass).

    I take it that the word "investment" means whatever mold your actually sacrificing in the kiln because one needs to destroy it to get to the final molded glass object.

    Anyhow, the frit I'm using melts at about 2000 degrees and I leave it in the kiln for 2 hrs. I've had my glass come out opaque like yours and also nice and transparent. I thought mine came out opaque because I fired it either too long, or too hot, or both.

    I enjoyed your Instructable and learned something, too.

    2 replies

    Thanks for the great comment! The frit casting sounds significantly more effective than my version; I wish I'd asked around online more before the project, as that would have been a good process for us.
    I also wonder if the smashed glass we used became contaminated during the smashing step (grit from the pavement), which could also have contributed to its oddness as well.

    I'm actually molding a cement lizard for my front yard. I wanted the eyes to be shaped like quarter moons and tried to cut tile and shape them that way, way too hard. so I stumbled onto molding glass while I was on Amazon and thought I'd give it a try. I'm flying by the seat of my pants as I'm using a kiln I built myself out of bricks and a propane tank, Raku style. Everything's trial and error with me, but I noticed that when I backed off with the heat and shortened the time my glass started to clear up.

    I found a lot of info reading the comment sections under products on Amazon and, also, watching You Tube on the subject.


    6 years ago on Step 6

    So there are a couple of issues that will go into how you should anneal your glass, glasses behavior through temperature ranges is very different than for metal. To give a brief explanation, all glasses contain specific COE, or coefficient of expansion, which determines the rate of contraction though its strain point on the way down to room temperature from the annealing point. For most modern soda-lime and lead based casting glasses this is between 104 and 85 COE, while boric based pyrex glasses have a common COE of 33, higher he number, the. In general for casting the higher the COE the lower the viscosity of the comparative glass at the same temperature, lead based glasses (104 COE) will begin to melt as low as 400*C while pyrex (33COE) would need to be brought to above 1500*C for any effect. Most Casting glasses have a investment casting temperature of 850*C- 870*C, higher temperatures will further reduce the viscosity, but most molding materials will begin to rapidly deteriorate higher than this.


    6 years ago on Introduction

    Thanks for the info. I am quite familiar with metal casting investment, but I am not sure about chemistry compatibility with glass. (I was not able to post under your reply; seems to be a bug with the captchas)

    1 reply

    Reply 6 years ago on Introduction

    Oh, I see.
    The same investment should work; the teacher whose materials I was using had only worked with metal before, and I just used the metal materials for glass.


    6 years ago on Introduction

    Thanks for the info. I am quite familiar with metal casting investment, but I am not sure about chemistry compatibility with glass.


    6 years ago on Introduction

    What type of investment did you use? Can you share who you purchased it from too? Thanks!

    1 reply

    Reply 6 years ago on Introduction

    I'm sorry, but I don't actually know... this was done in a classroom setting, and the powder we made our investment from was purchased by the teacher.
    However, here is another excellent casting instructable: and the author of that one recommends 'satin cast' by kerr for detailed jewelry casting in the comments.

    Chris Logan

    6 years ago on Step 6

    Additionally, your graphite, that's dissolving away, is just oxidizing into oblivion... CO2 generation.

    I'm sure some of it was absorbed by the glass.

    First of all, this is really, really cool.

    Secondly, what the hell is up with the purple gaffer's glass? Are they nuts?!? Who puts that much lead in something? Obviously, this is not directed at you, selkeymoonbeam, - just whatever criminally insane freak that decided to make that available to the public without a giant - WARNING! THIS GLASS IS INSANELY TOXIC! - label all over it.  Lead in glass will leach out in 5% (vinegar-level) solutions of acetic acid - I don't want to think about how much would leach out in stomach acid if someone (i.e. someone under the age of six) swallowed a bead made of that glass.  <rant ends>

    Thirdly, I wonder if it's the potassium that makes the wine bottles green.  European forest glass frequently had a greenish cast to it, and it seems likely that it was because glass makers were using ash/potash for their various alkali carbonates, which would contain (in that region) I high percentage of potassium.  That is just very interesting ....

    Fourthly, that is just a very cool instrucable.

    1 reply

    My understanding is that the old scandinavian / german glass is green because the soil used in the batch contained iron oxide. Today those impurities are removed, and Iron Oxide and Chromium are added to make deep green glass, like wine bottles.

    When glass is re-melted over and over, it tends to develop a greenish tint. This might happen because a piece of metal from the blow pipe was re-melted, or from something in the furnace like the heating elements degrading (my guess).

    I wouldn't hesitate to put a piece of yellow glass in my mouth (even though it contains cadmium), but if I'm melting it in the fire, I'll definitely be working under a ventilation system. I don't think using lead crystal glassware on a day to day basis poses any serious health risks, but storing alcohol in it for any long period of time should be avoided.

    This instructable is great! I enjoyed the laboratory approach very much.