Sealing MEMS Devices with Glass

 

Krista Grayson

•

December 5, 2023

Micro-electromechanical systems (MEMS) have revolutionized various industries by integrating miniaturized mechanical and electrical components onto a single substrate, enabling the creation of highly sensitive and efficient devices. They are key elements in an array of medical, automotive, industrial, and defense applications.

a

However, the success of MEMS devices often hinges on maintaining a hermetic environment to protect their delicate internal components. This is where glass frit sealing technology comes into play, providing a superior solution for achieving reliable hermetic seals in precise applications like MEMS manufacturing and packaging.²

Understanding Hermetic Sealing and its Importance

Hermetic sealing involves creating an airtight barrier around a device to prevent the entry of contaminants, moisture, and other external elements. This sealing technique is crucial for MEMS devices as even minute environmental influences can alter their performance or lead to premature failure. In applications where stability, precision, and reliability are paramount, such as in the aerospace, medical, and telecommunications industries, achieving a hermetic seal is essential.²

Glass Frit Sealing: The Ideal Solution for MEMS

Among the various methods available to achieve hermetic seals, glass frit sealing stands out as a versatile and high-yield approach, particularly suited for MEMS applications. This technique leverages the unique properties of glass to create a reliable, robust, and precise encapsulation for MEMS devices while imposing minimal stress on the bonding surface. In a three-step process, a glass paste is screen-printed on a capping wafer, which is then bonded to the subject device through thermocompression for 10 minutes. During this process, 1000 mBar of force and 440 °C are applied to the material under a vacuum. Capable of bonding both hydrophobic and hydrophilic surfaces, this technique can be applied to almost all commonly used microsystem surface materials, such as aluminum, silicon, and glass.^3,4

Tailoring Precision Using the Coefficient of Thermal Expansion (CTE)

As the name implies, glass frit sealing makes use of glass particles, known as frit, which can be precisely formulated to match the coefficient of thermal expansion (CTE) of different materials.⁴ The CTE of a material refers to how its dimensions change with temperature fluctuations. By tailoring the glass frit’s composition, its CTE can be adjusted to closely match that of the MEMS device and the encapsulating material. This compatibility ensures that, when subjected to temperature variations, the seal remains intact without compromising the structural integrity of the device.²

Mo-Sci, a pioneering glass technology company, has been at the forefront of developing and perfecting glass frit sealing solutions for various high-tech applications, including MEMS devices. Its expertise lies in creating sealing glasses with customizable thermal expansion coefficients. With a diverse range of glass-metal and glass-ceramic seals that are meticulously matched in terms of CTE and are capable of enduring temperatures as high as 1600°C, Mo-Sci is an ideal partner for MEMS manufacturers seeking reliable hermetic sealing solutions.^2,5

The Versatility of Glass Frit Sealing

The applications of glass frit sealing extend beyond MEMS devices and encompass a range of cutting-edge technologies:

1. Solar Cells

Sealing glasses find utility in encapsulating perovskite photovoltaic elements. These elements are promising alternatives to traditional silicon solar cells due to their high efficiency and lower production costs. However, perovskite cells are highly sensitive to moisture, whereby even small amounts can completely prevent function. Laser-assisted bonding of glass frit sealing guarantees a durable hermetic barrier, shielding perovskite cells from moisture exposure and locking in lead-containing chemicals.²

2. Metal Ion and Thermal Batteries

In the evolving landscape of energy storage solutions, glass frit sealing plays a pivotal role in enhancing the reliability and longevity of metal ion batteries, including lithium-ion and sodium-ion batteries. These batteries require seals that can withstand high temperatures and resist chemical corrosion. Sealing glasses provide a resilient barrier that enables the efficient operation of these advanced battery technologies.

Sealing glass is also a viable solution for molten salt batteries. These batteries are highly dependent on sodium salts, including sodium-nickel and sodium-sulfur chloride, to achieve remarkable energy and power densities. For this reason, they are an appealing option for large-scale industrial and energy storage applications.

Sealing glasses are classed as a high-energy alternative to conventional polymeric or metal seals as they exhibit excellent resilience against demanding chemical environments but also against the rigorous operating temperatures inherent to molten salt batteries, which can range from 300 °C to 350 °C.²

3. High Temperature Sensors

Glass frit sealing also finds applications in high-temperature environments, such as automotive systems and chemical processing plants. The predictable thermal expansion and corrosion-resistant nature of sealing glass ensure the longevity and stability of sensors operating in extreme conditions.²

4. Solid Oxide Fuel Cells (SOFCs)

SOFCs hold tremendous promise for clean and efficient power generation, but their high operating temperatures present engineering challenges. To create high-temperature sealant materials for SOFCs, Mo-Sci currently utilizes two methods. One relies on a traditional glass-ceramic seal, wherein the glass undergoes crystallization to establish bonds with the sealing components.

The second approach involves the development of viscous-compliant glass seals. These seals remain vitreous throughout application and can self-heal, mitigating the risks associated with thermal stresses and ensuring the long-term stability of SOFCs.This groundbreaking technology is anticipated to play a pivotal role in facilitating the commercialization of SOFCs and driving their widespread adoption.^2,6

Embracing the Future with Glass Frit Sealing

Glass frit sealing technology has emerged as a transformative solution for achieving hermetic seals in MEMS devices and a wide array of other advanced applications. By precisely engineering the properties of sealing glasses, companies like Mo-Sci enable manufacturers to create highly reliable and robust encapsulation systems.

As industries continue to push the boundaries of technological innovation, the role of glass frit sealing in safeguarding sensitive components and ensuring optimal device performance becomes increasingly vital.

References and Further Reading

1. Forbes. Why Timing Must Be Tough Enough For Our Digital World. Available at: https://www.forbes.com/sites/forbestechcouncil/2021/09/02/why-timing-must-be-tough-enough-for-our-digital-world/ (Accessed on 10 August 2023).
2. Mo-Sci. Sealing Glass Applications. Available at: https://mo-sci.com/sealing-glass-applications/ (Accessed on 10 August 2023).
3. Chang H-D, et al. (2010). High hermetic performance of glass frit for MEMS package. 2010 5th International Microsystems Packaging Assembly and Circuits Technology Conference. https://doi.org/10.1109/IMPACT.2010.5699539
4. Knechtel R. (2015). Chapter 31 – Glass Frit Bonding. Handbook of Silicon Based MEMS Materials and Technologies (Second Edition). https://doi.org/10.1016/B978-0-323-29965-7.00031-2
5. Mo-Sci. Matching Coefficient of Thermal Expansion in Glass Seals. Available at: https://mo-sci.com/matching-cte-in-glass-seals/ (Accessed 10 August 2023).
6. Mo-Sci. Sealing Glass. Available at: https://mo-sci.com/products/sealing-glass/ (Accessed on 10 August 2023).

source:https://mo-sci.com/sealing-mems-devices-with-glass/?utm_source=Mo-Sci+Newsletter&utm_campaign=b5090c88ed-EMAIL_CAMPAIGN_2023_09_28_06_45_COPY_01&utm_medium=email&utm_term=0_-cf8dcfb60f-%5BLIST_EMAIL_ID%5D&mc_cid=b5090c88ed&mc_eid=0ab94327fb

Spikes on Frit Castings

Credit: The Crucible.com

It is frequent to have castings from frit with spikes, needles, or prickles around the edges. 

Causes

These spikes result from the glass touching the edge of the mould or separator during the hottest part of the firing. The glass particles first begin to compact as the glass rises toward the fusing temperatures. As the temperature increases toward the casting temperature it begins to expand both horizontally and vertically from that compact mass. As it cools, the glass sinks down and retreats from the edge. This movement leaves some small bits of glass stuck to the sides. The glass contracts as it cools, leaving the spikes as it contracts from its hottest state. 

Avoidance

The usual recommendation is to mound frit in the middle and let it flow to the outside. Still, the glass flows to the outside of the mould at casting temperature and it touches the sides. Leaving the risk of creating spikes. Accurate measuring of the amount of glass to charge the mould with is important. With the right amount of glass, the mould will not be overfilled and so, reduce the spiking. 

Measuring the weight of glass for the mould is not difficult. In many cases, the manufacturer of the mould has done the work for you. If you need to calculate the weight of glass required for the mould, it is not difficult. A method is given here. In short, you use a dry fill of the mould. Measure the volume (using the metric system) and multiply by the specific gravity to get the weight in grams. 

Larger chunks of glass tend to produce fewer spikes than smaller frit. Usually longer soaks at top temperature are required to fully form the glass with smaller frit. It is also possible to drip glass into the mould from a pot suspended above the mould. Accurate measurement of the weight will still be important. But add 100gms/4oz. to the amount to allow for the glass that will stick to the pot.

My view is that with dams, it is better to use a straight sided shape with fibre cushioning around the outside. When annealed and cool, clean it well. Then fire polish with a slow ramp to 540°C/1000°F followed by a quick ramp to the fire polish temperature. This will polish the sides of the piece that were in contact with fibre paper.

Homemade Glass Powder

Summary of a question: Frit made in an electric coffee grinder and cleaned with magnet still produced black specks in the finished piece. What’s happening?

 

 Most coffee grinders use stainless steel blades. Most culinary stainless is not magnetic. So, you are left with flecks of steel in your powder that the magnet cannot remove.

 My experience with homemade powder has not been positive. This has led me to buy powders. But I still use home-made frit. You can wash and sieve out the powder and steel particles at the same time into a bucket or basin. This will leave you with clean frit from fine to as coarse as you want.

 Do not put the residue down the drains. It will block your drains after a time. Instead, you can let the glass settle to the bottom of the bucket and pour off the clear water. Let the remaining water evaporate, and wrap up the sediment for waste collection. Or if you have a garden, you can empty the water and sediment onto the ground. It makes for good drainage over time.

 

The contamination in home-made glass powders make it best to buy powders and make your own frit.

  

Frit Additions to Shaped Pieces

It is possible to stick frit to slumped pieces. But soaking for a long time – several hours - at 650°C/1200°F is required to stick the frit.  The added pieces will remain relatively sharp. You need to observe frequently from 600°C/1111°F to make sure that the form of the glass is not distorting. 

Credit: Pyramid Gallery, Smyth and Zebrak

Although it is possible, adding pieces to already shaped objects is not best practice, nor will it frequently give satisfactory results.  If the slump is shallow, it is more possible to do this successfully than steep or highly shaped forms.  But the most suitable practice is to flatten the piece, then tack fuse the pieces onto it. Follow this fusing with the new slump or drape.  This flattening process will not be possible with all shapes. 

The best results will be achieved by accepting what you have and make a new piece with the planned additions from the start.

This process will not be suitable for draped glass as the glass will drape further during this low temperature soak.

 

I've a book that gives more detail. Low Temperature Kilnforming, an evidence based approach to scheduling  or at Bullseye

Home Made Frit Maker

Recently, when looking for a small frit maker, no shop had one in stock.  Having heard of making one from plumbing pipes, I went in search of material.  I came across stainless steel pipe and caps.

 

The practical size seemed to be 50mm.  Short sections of threaded pipe were available with matching caps.  That forms the containment cylinder.

 

A threaded 25mm pipe and cap can be fitted loosely into the larger one, and so forms the plunger or piston.

 

There needs to be a handle.  It could be a turned piece of wood to fit the inside of the pipe.  In this case, I obtained a reducing connector to fit a 12mm pipe to the plunger and topped it with another cap. 

The whole was put together in less than a minute, once all the parts were assembled.

The completed frit maker

Galvanised pipe would be cheaper but carries the possibility of introducing zinc into the frit.  Stainless steel risks introducing non-magnetic particles into the frit.  As I sieve out powder from my own frit making before washing, I am not too concerned about steel contamination.  If you want powder, use uncoated mild steel so the contamination can be drawn out with a strong magnet.

Playing in the Sandbox

This process provides flowing, abstract images that can be used as autonomous pieces or formed into other objects, such as free drops, bowls, cut for jewellery or into pattern bars.  The appearance provided is unique to this combination of using frit and pressing.

In principle, this process is the same as creating sand pictures.  The process is in three stages: making the box, adding frit, and pressing.

The Sandbox

Determine the size of the box.  It should not be more than two-thirds the size of your kiln shelf depending on thickness.  Thicker glass pressed to 6mm will spread more than thinner.  As a guide, 12mm should have an allowance to spread to about 1.3 times the original size; 19mm should have an allowance to spread about 1.5 times the original dimensions.

Cut two sheets of the same size from clear fusing glass. One will be the front. The other will be the back.

Determine whether the image you are creating will be portrait, landscape, or square.  Orient the sheets in the appropriate way to have the top away from you.  Choose the top piece of the pair and cut two 6mm strips from the designated top.  This gives you a lip to be able to pour the frit into the box easily.

 

Box formed with bottom and sides glued to back and front.  The filling lip shows on the right.

From another piece of clear glass cut two 6mm strips for the sides.  If you cut them the same length as the side of the glass, they will stick above the back about 3mm. You can cut this off, but it really is not a worry for the construction of the box.  These strips form the spacers to allow the frit to be poured into the box.  Their thickness will determine the amount of frit needed to fill the box.

Get out the back sheet and clean and prepare it for attaching the strips. My preferred method is to glue the bottom 6mm strip on its edge with super glue.  It is advisable to wear plastic gloves when gluing the strips, to avoid sticking your fingers to the glass.  Super glue cures quickly and does not delay the construction of the box.  It burns out cleanly without any health and safety concerns.  Place a thin film of super glue on one edge of the strip.  Attach it to the bottom by placing it carefully at the edge of the sheet.  Do the same for the sides.

  

When the strips are stuck down to the back, place  a thin line of super glue on the top edge of the strips in preparation for attaching the top sheet.  Using a strip of wood placed at the bottom of the backing glass will help in placing the sheet accurately. Lower the sheet from contact with the bottom to the strips forming the sides of the box.

When the glue is cured, inspect the sides of the box for gaps. If there are gaps, use clear Sellotape to seal the gaps in the sides. It will burn off cleanly in the kiln.

Adding the Frit.

Place the box on an easel or other support so it is slightly tipped backwards.  This helps ensure the box does not fall toward you while working on it.  It also allows the frit to slide toward the bottom rather than bouncing off the other frit.

 

The early stages of filling with the box on a stand

The size of frit you choose to use will affect the final appearance.

·        Generally, powder will appear greyer and more opaque than frit. This is due to the multiplicity of tiny bubbles between the grains of powder.

·        Fine and medium frit provide more clarity than powder.

·        Coarse frit provides the most clarity, but with fizzy bubbles between pieces of frit.

When preparing to place the frit in the box, it is a good idea to take small amounts out of jars and place it into small cups to avoid contamination of the main source of the frit.

Pouring the frit into the sandbox

You can use a jeweller’s scoop or a teaspoon to move the frit from the cup to the box.  Tip the frit into the box above where you want the colour to be placed.  

Moving the frit with a skewer

If the frit does not land just where you want it, you can move it with stiff wire that is long enough to reach the bottom of the box.  Gently sweep the frit with the end of the wire toward the place you want the coloured frit to be.

Using a jewellers scoop to add the frit.

Continue adding colours to create the profile and shapes you wish.

You can make additional alterations to the way the frit is placed.  You can poke the frit from one layer into lower layers with a stiff wire by pushing the wire directly downward.  You cannot do this more than 2 or 3 centimetres deep, as the frits and powders become compacted.

A thick copper wire being used to poke down from an upper layer to the lower ones.

When filled to the top or to your desired level, use the fourth strip to close the box.  If full, glue the strip to the top.  If not full, cut strip to the length needed to drop into the opening of the box.  Place a couple of drops of super glue on the top of the already placed strip to keep it in place while moving to the kiln.

The Pressing

Prepare the shelves

You will need two shelves for each pressing. One is the base to hold the glass and the spacers.  The other is to provide the weight to press the glass thinner.

Clean off old kiln wash from the shelves. Experience shows that adding new kiln wash over old for this process promotes the sticking of the kiln wash to the glass.  Add new kiln wash that performs well at extended times at upper temperatures.  I find Bullseye shelf primer works very well.

Once partially dried, with the pink beginning to pale, you can smooth the surface brush marks.  Some use balled up material such as tights to rub over the surface.  I find very good results from rubbing lightly over the kiln washed surface with a sheet of paper between the palm of my hand and the shelf.  The advantage of doing this smoothing while slightly damp is that no dust is created that needs to be cleaned away.  The disadvantage is that too much pressure will pull bits of kiln wash from the shelf.

Do not use fibre papers as the separator.  The glass will be moving within the space between the shelves.  It will pick up and incorporate parts of the fibre paper, if used.

If you have shelves of different thicknesses, reserve the thickest shelf for the upper, pressing one.  If all your shelves are the same size, put a second on top for adequate weight, or add heavy bricks or a steel weight to the top shelf.  (Note: if you use bricks for weights, they need to be dried first.  A two-hour to three-hour soak at 95C should be sufficient.)

Placing

Place the sandbox centrally on the shelf.  If you are doing more than one, ensure there is plenty of space between the pieces and from the edge, so they don’t contact each other, or drip over the edge of the shelf.  The allowances given for the size of the sandbox are a guide.

Two sandboxes placed on separate shelves

Place spacers of the desired thickness around the four corners of the shelf to restrict the extent of thinning.  This also regulates the evenness of the glass across the whole surface.  Usually, 6mm is a desirable height for the pressing.  Other thicknesses can be chosen for different purposes.  The spacers can be steel washers, although they will spall in the cooling stages of the firing.  If you have pieces of ceramic of the desired height, they can be used.  Fibre paper stacked up to the appropriate height are surprisingly robust spacers.  They also provide a cleaner set of spacers than steel.

A corner of the shelf with the 6mm fibre spacer

Place the upper shelf gently down onto the glass piece. The glass at this stage is taking the whole of the weight of the pressing shelf.  The shelf must be placed both gently and evenly down onto the glass to avoid breakage.

Check that everything is in place. This may require additional, directional light such as from your mobile phone or a torch.  It is now ready to fire.

The Firing

This assembly of materials has a lot of mass.  It is 2 to 3 times the normal mass for a standard firing.  

Pressing shelf placed on top of the glass sandbox

This promotes variations in practice:

• ·        Even with this additional mass, you can fire quickly.  This is because the glass is in small pieces and that the mass of the shelves gains heat slowly. 
• ·        The greater mass does require longer soaks than a normal fuse firing. 
• ·        The upper temperature for a full fuse is required to get the glass to a sufficiently low viscosity to allow the glass to move.
• ·        The long soak at the top temperature does not promote devitrification as in normal fusing.  My speculation is that the glass is not exposed to the air, so the devitrification cannot form. 
• ·        A further difference in a pressing firing is that the annealing can be at the rate for the final thickness of the glass.  The mass of the shelf and weights above the glass means the glass is cooling evenly from both sides, unlike normal fusing.  The glass may be cooling more slowly than programmed, but the programmed rates limit any possibility of too rapid a cooling.

A schedule for a 12mm thick Bullseye piece with a 19mm upper shelf might look like this:

300°C/hr to    670C       for   180 minutes

300°C/hr to    816C       for   180 minutes

AFAP      to    482C       for  120 minutes

55°C/hr   to    427C       for   0 minutes

99°C/hr   to    370C       for   0 minutes

200°C/hr to    50C         for   0 minutes

Off

A piece of 19mm should be slower:

150°C/hr to    670  for   240 minutes

150°C/hr to    816  for   240 minutes

AFAP      to    482  for   120 minutes

45°C/hr   to    427  for   0 minutes

90°C/hr   to    370  for   0 minutes

180°C/hr to    50    for   0 minutes

Off  

Both these schedules assume the final thickness of the glass will be 6mm.  The schedule for glasses other than Bullseye only needs to have the top and annealing temperatures altered to the ones appropriate to the glass.

Results

The pressed glass will have the texture of the shelves on both sides.  Normally, no kiln wash will be stuck to the glass.  If there is kiln wash to be removed, you can do this by abrasive means – sandblasting, diamond pads, wet and dry sandpapers or Dremel style tools.  It is important to keep the glass damp during this process.

Untreated result of pressing

If the surface of the glass is without sticking kiln wash or other marks, you can use it with the matte surface without further kiln work.  You can also fire polish the piece to get a glassy surface, once you have thoroughly cleaned it.

Alternatives

Tape box together

After super gluing the bottom and side strips, you can bind the box together with clear Sellotape.  Pull off at least three strips of tape and set them where you can reach them easily.  Place the upper sheet on the prepared base. Move the box to the edge of the work surface so a little of the box hangs over.  The first stage is to place a strip of tape at right angles to the side to bind the top to the bottom.  Do this for each of the three sides.  When the top is securely attached to the base and sides tape along the length of each of the three sides. 

This shows on the lower left a loosened piece of sellotape on the edge of the sandbox.

This process avoids any difficulty in attaching the top.   Attempting to use only Sellotape to bind the box together is very difficult and requires at least three hands.

Spacers for the frit

Spacers do not always need to be strips on edge.  The spacers can be one or two wider strips placed on their sides to provide the needed height.  They can be coloured, forming a border; but remember the border will become curved. The strips will need to be glued to the back.  The top can be attached with super glue, or taped to the sides and back.

Pressing without a box

It is possible to use the pressing technique without a box or frit.  You can arrange clear and coloured cullet on the shelf.  The arrangement needs to be such that there are no gaps between the pieces.  This means that the glass will probably be 3 to 4 layers thick.  Be careful to avoid creating thick layers of dark colour by interfiling with clear. Place the spacers at the corners of the shelf for the thickness desired and fire.  The slower rate of firing (as for 19mm) should be used.

This sandbox process is a combination of arranging frits and pressing.

Further information is available in the ebook Low Temperature Kiln Forming.

Colour Dilution of Powders

Sometimes you do not have a tone or shade of a colour you need for your project.  Other times you want to have a gradation of shade across a piece.  There is the obvious solution of mixing a colour with clear to produce lighter shades.  But there is a difficulty when mixing clear with powders to fuse. The result is often a pointillist effect with points of light coming through the colour. There are several approaches to this difficulty.

One way is to use a powder made from a tint of the colour.  But sometimes there is not a tint made. Sometimes you do not have that tint in stock. So, you must look to other solutions.

Credit: www.warm-glass.co.uk

An alternative is to use clear powder to mix with the intense colour you want to dilute.  You will need to test varying proportions of clear to colour to get the tone you need.  You may be surprised at the amount of clear needed.  And there still is the slight possibility of points of light coming through the clear.

Another possibility is to use one of the less dense white powders to mix with the colour.  White powders such as the Bullseye 000243, translucent white, or the 000113, dense white are possible.  The very dense or lacy whites are not as suitable. One is too opaque, the other is uneven in colour. Again, testing will be required, and you may be surprised at how little is required to alter the tone.

One other way I have used is to mix fine frit with the powder.  This has less control than the other methods but can provide significant dilution of the intense colours.  If you want to see if this is suitable, you can follow this process. 

Add a few drops of water to the clear frit in a small container. Close it and shake to get all the frit coated with a film of water. If after shaking the frit is not “clumping” you can add a little more. Too much water will create a slurry which is not suitable.  So, add only a small amount of water at a time until the frit is like damp sand on the beach. Any excess water must be poured off. 

Add powder to the damp frit, and shake well again to coat the frit with powder. If the frit does not seem to be fully coated, add a little more powder.  The film of water on the frit allows the powder to adhere temporarily to the frit.  

This mixture can then be applied to the surface and smoothed with a pallet knife. This will not guarantee there are no clear pinpoints, but it will reduce them to a minimum. You will not have the subtle differences in tone that sifting powder can give you, but it is a cost-effective way of diluting intense powder colours that can have advantages over mixing powders.

Of course, the various methods of diluting colour described here can be used to combine powders to produce new colours.

Glue boiling

People often find that their glued pieces shift in the firing.  Sometimes dramatically. This can be from placement or boiling of the glue.

Almost all commonly available glues evaporate well below the “sticky” range of glass. 

The boiling points of some of the popular glues:

PVA (also known as: wood glue, white glue, carpenter's glue, school glue, Elmer's glue in the US, or PVA glue)  

Boiling point: 112°C (234°F)

Super glue (and other cyanoacrylate glues) 

boiling point 54-56°C (129-133°F)

Lacquer (hair spray)

Boiling point: 185 to 189°C (365 to 372°F)

CMC  (carboxymethyl cellulose includes wallpaper paste, vitragel and most fusing glues)

boiling point: 260–270°C (500–518°F)  

melts at 274°C 

Aloe vera gel  

ignition point: ca. 232°C therefore, its boiling point is lower.

This shows the popular glues used in kilnforming are not effective above 300°C and many with much lower boiling points.  This means that glues only hold glass in place when cold. Glues may be useful in getting items with many pieces into the kiln, but they will not hold them in place until the glass becomes sticky.  The glass, in general, needs to be at or above the slumping temperature to begin to stick together.

If glue doesn’t work, what can I do?

Other means than glues are required to support the moveable items until the glass become sticky at around 620°C (about 700°C for float glass).

If placing frit supports is not possible without showing, such as in a tack fuse, you can use mechanical means. Two of these are grinding flat spots on rolling pieces; and placing supports under the balanced items. Other support and damming methods will depend on the nature of the project.  In general, if the pieces will not stay in place without glue before placing in the kiln, the pieces will move in the kiln after the glue evaporates.

This evaporation can be so fast as to be called boiling.  Glue boils off in the kiln whether dry before firing or not. Sometimes there is enough glue or rapid enough heating to cause displacement of the glass by the force of the boiling. Just as in boiling water, the evaporation can be explosive.  The force of the built-up pressure of glue trapped under glass can move small pieces relatively large distances.

How do I avoid boiling the glue?

• Use as little glue as possible.
• Use it at the edges of the pieces.
• Do not place it in the middle of large pieces.
• When you do use glue, advance slowly to at least 300°C allow the glue to evaporate, rather than boil.  I’d suggest a rate of 50°C per hour would be slow enough to avoid the boiling of sparingly applied glue.

Best of all, use no glue.

Quickly fired glue - wet or dry - boils.  Sometimes with enough force to move the glass significant distances.  Avoid gluing as much as possible and use sparingly when needed.

Making Thin Sheets

The question of how to make thin sheets arises from time to time.  Unless you are a glass manufacturer, it is unlikely you can make large, thin glass sheets.  But you can approximate making thin sheets by two methods that I know.

Sintering

One of these is sintering.  This is firing the glass to a low temperature and soaking for a long time.  The common form of this is powder wafers. 

By using a screen to deposit an even layer of glass powder you can make very thin, but delicate sheets of glass.  The procedure I would use is a screen of about 45 – 60 threads per inch.  This is coarse enough to allow the powder through, but not so fine as to “reject” large amounts of the coarser particles. 

You can screen the powder directly onto a kiln washed shelf, or onto Thinfire or Papyros.  You will not be able to move the unfired powder on a sheet of paper or fibre paper without changing the thickness and shape of the screened powder.  It must be laid down onto the separator directly on the shelf.  You can of course, move the shelf to the kiln if you can get in without tipping it.

Method

Support the screen about 3mm above the surface to allow the powder to fall through.

Make a ridge of powder at one end of the screen.  Using a smooth straight edge wide enough to cover the whole of the screen, lightly spread the powder from the starting end to the other. Then repeat drawing the powder to the starting end.  Make about five repeats of this – that is 10 passes, to get enough powder laid down to form about 0.5 to 1mm sheet.  You will need to experiment with the number of passes to get what you want.

Do not try to press the powder through the screen.  That will only wear the screen out quickly and may tear it.  Each pass should be a light spreading of the powder.  It is heavy enough to fall through the screen without additional force.

You could, of course, just sift the powder over the area you want to cover and judge by eye how even the layer is.  It is possible that your observation is good enough, but it is more likely that you will have thick and thin areas.  Often even at sintering temperatures, the thin is pulled toward the thicker, leaving small or large holes.   By screening the powder, you know you will have an even layer

Firing

The kind of schedule to use to sinter the glass particles together without changing their structure is the following:

220°C to 482°C , soak for 60 mins
55°C to 593°C, 10 minutes
28°C to 665°C for 5 mins
as fast as possible to 482°C for 30 mins
28°C to 427°C, no soak
55°C to 370°C, no soak
110°C to 50°C, no soak

This will work for most fusing glasses.

This slow firing allows enough heat to penetrate the glass grains that they will stick together without changing shape or developing holes.  I admit the anneal cool is very cautious.  You can experiment with quicker cools if you want to speed the process.

  

Pressing

This is a technique of thinning already existing sheets of glass.  Typically, you will have a 6mm or thicker piece of glass that you want to be 3mm or less.  Paul Tarlow has described this kiln pressed glass very well in his books and on the fusedglass.org site.

In essence, you use a pair of kiln shelves.  Kiln wash both shelves.  Place the glass to be thinned on one shelf.  At the outer edges of the shelf put down spacers of the thickness you want the glass to be after pressing.  This will keep the upper shelf from settling down too much and more importantly unevenly.  Place the other shelf, kiln washed side down, on top of the glass.  Be sure the spacers are in places where they can support the upper shelf.

If you are thinning from 6mm to 3mm, normally you do not need any additional weight on top of the upper shelf.  But the thinner you want the glass to be, the greater the weight needs to be.  It could be another shelf, fire bricks or steel weights.

When scheduling the annealing remember you must take account of the mass of the weight on top of the glass.  You will need a much longer temperature equalisation soak and a much slower annealing cool.  

Further information is available in the ebook Low Temperature Kiln Forming.