Glazing: glass, acrylic, UV and reflections
Glazing is decided by weight, size and who is under it before it is decided by optics. This chapter reads the ultraviolet claim for what it covers, shows what filtering actually buys, and names the pieces that must not sit against glazing.
Start here Essentials
At minimum this chapter asks you to answer three questions in order: how big, how heavy, and who stands under it. The optics come last. Everything marked Go deeper is optional on a first pass.
- Work out the glazing weight for the size you need, in both materials. See what the difference is (Weight and hardware).
- Decide glass or acrylic from size, weight and who is under it. Do this before any coating question (Weight, size and who is under it).
- Choose an ultraviolet-filtering grade, or decide you are not buying one. Either way, know what it buys on your material (What the ultraviolet claim covers).
- Write the glazing line in the framing plan. Name the material, the thickness and the filtering (Framing plan).
Minimum time: about 20 minutes. Done when: the framing plan names a glazing material, a thickness and whether it filters ultraviolet, and you can say why in one sentence that is not about appearance.
1. Weight, size and who is under it Understand
Most people choose glazing by what they can see through it. The three questions that should come first are physical.
Weight. Glass is the mass in a framed picture, and it is the one well published figure in the whole subject. A trade association's bulletin, reissued as current national association data, gives flat glass by thickness: 1.2 pounds per square foot at 2.5 mm, 1.6 at 3 mm, 2.0 at 4 mm, 2.4 at 5 mm and 3.0 at 6 mm, with a density of 158 pounds per cubic foot for any thickness not on the table. Product-specific Picture framing glass is normally 2.5 mm, so a 24 by 36 inch sheet of it weighs about 7 pounds before anything else is in the frame.
Acrylic is lighter, and nobody publishes a table. Two makers publish a specific gravity of 1.19 for acrylic sheet, one under ASTM D792 and one under ISO 1183, which is as close to independent agreement as this subject gets; one of them states the consequence in its own words, that acrylic sheet is only half the weight of glass. The weight calculator computes acrylic from that figure and says on the page that it is computing rather than quoting.
Size. Sheet sizes are the nearest thing to a published answer to "how big before acrylic is the only option". One maker's conservation glass is listed at a maximum of 48 by 68 inches; its conservation acrylic runs to 80 by 120 and its anti-reflective acrylic to 72 by 120. Above about four feet in one dimension you are choosing between glass that has to be specially sourced and acrylic that is ordinary. A professional wiki gives a different threshold for strength: single strength glass may be used up to 30 by 40 inches, above that double strength, and very large works may need safety or laminated glass.
Who is under it. This is the question nobody publishes a figure for and the one that should decide the most cases. Glass breaks, and it breaks into pieces that cut. A maker lists the cases where it recommends acrylic instead: hanging and shipping heavy oversized artwork, children's rooms, earthquake zones, high traffic areas, and anywhere safety is a concern. The same maker's own glass product page states that its glass is unsuitable for friable materials, oversized artwork, frequently transported pieces, earthquake-prone locations, children's spaces and high-traffic public environments.
No published figure or size threshold for glazing over a bed, a crib or a stair wall was found in this research. The federal child-safety campaign on wall safety covers furniture and television tip-overs and does not mention pictures or mirrors at all. The nearest published anchor is the federal architectural glazing standard, which divides products at 9 square feet of glazing on one side and sets impact tests of 150 and 400 foot-pounds for the two categories; that regulation covers doors, sliding doors, glazed panels and bathtub enclosures, not picture frames, so a frame maker's reference to it is a borrowed benchmark rather than a legal requirement.
Use acrylic rather than glass over a bed, over a crib, in a child's room, on a stair wall, in a hallway at head height, and anywhere a fall would reach a person. No published figure or threshold supports that; it is our practice, derived from what glass does when it breaks and from the maker's own list of cases where it recommends acrylic. We would rather say so than dress it as a standard.
2. What acrylic costs you
Acrylic is not simply the better answer, and three of its costs are worth knowing before you choose it.
It scratches. One maker publishes the numbers: change in haze under a standard abrasion test of 11.2 percent after 10 cycles, 24 percent after 50 and 24.9 percent after 200. Hardness is Rockwell M-93 to M-95, against glass, which normal cleaning does not mark. Practically this means acrylic is cleaned rarely, with a soft cloth, and not with a paper towel.
It holds a static charge. That is the whole basis of the friable-media argument below, and it also means dust is attracted to it. Anti-static grades exist; ordinary acrylic is not one of them.
Its cleaner is product-specific. A professional wiki says solvent-containing cleaners cause microcracks and other surface disturbances on acrylic. Confusingly, the maker of one coated acrylic says the opposite about its own product, recommending common household glass cleaners and specifically not acrylic cleaners or polishing agents. The resolution is that coatings differ: follow the instructions for the product you actually bought, and where you do not know, use a dry soft cloth. Product-specific
Two more facts worth carrying. A conservation institute notes that large pieces of plastic must be 3 mm thick to avoid bowing, a thickness that often will not fit an average frame. And a professional wiki warns that tape must not be applied to acrylic, because the adhesive is then impossible to remove, which matters when a framed piece travels.
Against that, two real advantages beyond weight: acrylic does not shatter, and acrylic is slightly permeable to moisture where glass is a true vapor barrier, which a preservation center notes as a difference rather than as a verdict.
3. What the ultraviolet claim covers
This is the part of the subject most likely to be oversold, and the figures that correct it are published.
What the makers claim. One maker states that its conservation glass, conservation acrylic, museum glass, anti-reflective acrylic and laminated glass each block up to 99 percent of ultraviolet in the 300 to 380 nanometre range. The same page records the trade definition behind it: per a framers' association guideline, glazing may be labelled as providing ultraviolet protection only if it blocks at least 97 percent of ultraviolet in the 300 to 380 nanometre range. Product-specific
What testing found. A conservation institute tested seventeen glass and plastic filters in 2012 and published the results. Ordinary glass reduced ultraviolet by about 26 percent, plus or minus 11. Glass filters targeted at museum applications averaged 83 percent, plus or minus 14. Rigid plastic filters promoted in the conservation market cut 99 percent, plus or minus 1. Against an older and stricter specification only 35 percent of the seventeen filters qualified; against the 97 percent standard, 88 percent did. The institute's own recommendation is to look for filters reducing more than 90 percent. Test it
So: the plastic figures and the claims agree, and the glass figures do not. An average of 83 percent with a spread of 14 points is a different product from one blocking 99, and a reader paying a premium for coated glass is entitled to know that the only independent testing in this research returned a lower number than the label.
What removing ultraviolet actually buys. This is the best answer anywhere to "is museum glass worth it", and it is a table. The same conservation institute publishes the light dose that causes a just-noticeable fade of each ISO Blue Wool standard, in millions of lux hours, with ultraviolet present and with ultraviolet removed:
| ISO Blue Wool | 8 (most stable) | 7 | 6 | 5 | 4 | 3 | 2 | 1 (most fugitive) |
|---|---|---|---|---|---|---|---|---|
| Dose with ultraviolet present | 120 | 50 | 20 | 8 | 3.5 | 1.5 | 0.6 | 0.22 |
| Dose with ultraviolet removed | 1,000 | 300 | 100 | 30 | 10 | 3 | 1 | 0.3 |
Read the two ends. The most stable material goes from 120 to 1,000, a factor of more than eight. The most fugitive goes from 0.22 to 0.3, a factor of about 1.4. Physical principle The same institute states the general case directly: for colors that are sensitive to light, ultraviolet usually contributes less than half of the fading and often only about one tenth.
That is the sentence this chapter exists for. Removing ultraviolet buys a great deal on stable material and almost nothing on fugitive material, because visible light fades things too. The same institute adds the other half of the case for buying it anyway: for many artifacts, such as paintings with permanent pigments or monochrome prints and drawings, the yellowing and disintegration of the media and support by ultraviolet is the major form of deterioration during uncontrolled lighting.
Put your own wall through the exposure box, which prints both rows against your measured level. A filter is worth buying. It is not a solution, and nothing in this chapter makes a sunny wall safe.
A percentage on a glazing label is a product claim about one band of the spectrum, measured by the maker, and the only independent testing found in this research returned a lower average for glass with a wide spread. No glazing stops fading. Glazing buys time, and the exposure calculator is where you see how much. If you want a number for your own piece, the honest answer is that nobody can tell you which Blue Wool number your colorants are without testing them.
4. Friable media: the live disagreement
Pastel, charcoal, chalk and soft graphite sit loose on the paper, and the question is whether a static charge on plastic glazing can lift them. Five publishers answer, and they do not agree.
| Publisher and date | Position |
|---|---|
| Conservation institute, note dated 1996 | Glass picks up very little electrostatic charge and is therefore recommended for all friable media. Plastic glazing readily picks up and holds a charge, which can lift particles from charcoal, pastel, chalk and conté crayon, so plastic is not recommended for works with friable images or cracked or flaking paint. If plastic must be used, increase the distance: eight-ply board rather than four, or a double mat with two superimposed windows |
| State archives, undated | Use only glass for artworks in powdery media like chalk, charcoal or pastels |
| Preservation center, 2019 | Conventional acrylic carries a strong static charge and must never be used for pastels, charcoal or light tissue-like papers; use ultraviolet-filtering glass or anti-static acrylic instead |
| Professional wiki, current | Acrylic is now the norm for glazing framed artworks and glass is now considered a historic glazing material. When framing artwork with friable media, ensure at least an eight-ply mat's worth of space between the glazing and the artwork, and use acrylic with anti-static properties |
| Conservation center, July 2022 | Recommends acrylic glazing instead of glass, to avoid shattering and to lighten the package; names one anti-reflective, anti-static, ultraviolet-filtering acrylic as usable with pastels and other friable media |
| Maker, current | Its anti-static protection exceeds glass and immediately dissipates static; publishes a surface resistivity static decay of 0.01 seconds under a federal test method, and recommends the product for static sensitive materials including textiles, charcoals and pastels |
What changed is the coating, not the physics. The 1996 note and the state archives are describing ordinary acrylic, which does hold a charge. The current sources are describing anti-static acrylic, which is a different product, and the preservation center's wording shows the hinge: it rules out conventional acrylic and allows anti-static acrylic in the same sentence.
This manual's default: anti-static acrylic with a spacer, or glass with a spacer. Either works. What does not work is ordinary acrylic against a pastel, and what is not enough on its own is a single four-ply mat. The trade specification asks that pastels and chalk drawings be spaced at least 5 to 6 mm from the glazing and that multiple mounts or deep spacers be used for migrant or delicate pigments; it also says a framer should not apply fixative to such work. The professional wiki's eight-ply figure is the same instruction in different units.
One further warning from the 1996 note that nothing has superseded: a cited study found that taping glass for transport may damage unfixed pastel, because static built up during taping and tape removal can transfer pastel from the work to the glass. Carry pastels flat and face up. Moving has the rest.
5. Coatings and reflections
Reflection is a viewing problem, and it has two solutions: a coating on the glazing, and the room.
The maker's published figures for its own range give the scale of what a coating buys. Conservation glass and conservation acrylic reflect about 8 percent from both faces together and transmit more than 89 and 92 percent respectively. Anti-reflective museum glass reflects less than 1 percent and transmits more than 97. Anti-reflective acrylic reflects less than 1.6 percent and transmits more than 98. Laminated anti-reflective glass reflects less than 1 percent. Product-specific
That difference is real and visible. Whether it is worth the premium depends on how much light the room throws at the wall and where people stand, which is a room question rather than a glazing question. Learn Interior Lighting's chapter on layers and glare covers the room side: where a lamp has to be so its reflection does not land in your eye, and why moving a fixture often does more than changing the glass.
One coating to avoid for work on paper. A conservation institute explains why etched non-glare glass is not recommended: its haze only clears when the glass is in direct contact with the artwork, which conservation framing forbids. The professional wiki says the same thing. Coated reflection-control glass behaves differently and can be used at a distance of up to about three matboards.
And one product fact worth knowing before someone offers you a bargain: ordinary picture framing glass blocks very little ultraviolet. The same maker's comparison table lists its own non-conservation clear glass at 45 percent against 99 for the conservation grade, and the conservation institute's testing put ordinary glass at about 26 percent.
Go deeper 6. The published figures, side by side
Everything a maker publishes about its own glazing, with the independent testing beside it.
| Product | Thickness | Maximum sheet | Reflection | Transmission | Ultraviolet blocked (claimed) |
|---|---|---|---|---|---|
| Conservation glass | 2.5 mm | 48 by 68 in | 8 percent | over 89 percent | up to 99 percent, 300 to 380 nm |
| Museum glass (anti-reflective) | 2.5 mm | 48 by 68 in | under 1 percent | over 97 percent | up to 99 percent |
| Conservation acrylic | 3.0, 4.5, 6.0 mm | 80 by 120 in | 8 percent | 92 percent | up to 99 percent |
| Anti-reflective anti-static acrylic | 3.0, 4.5, 6.0 mm | 72 by 120 in | under 1.6 percent | over 98 percent | up to 99 percent |
| Laminated glass | 4.4, 8.8 mm | 72 by 120 in | under 1 percent | over 98 percent | up to 99 percent |
| Ordinary framing glass (same maker's premium clear) | 2.5 mm | not stated | 8 percent | over 90 percent | 45 percent |
| Independent testing, 2012 | ordinary glass about 26 percent ± 11; museum-targeted glass filters average 83 percent ± 14; rigid plastic conservation filters 99 percent ± 1 | tested, not claimed | |||
Laminated glass is the third option and the one least often mentioned. The maker's own classifications for its laminated product are the only published safety classifications for framing glazing found in this research: it meets a swinging punch bag test to one standard's Class B and the federal architectural glazing Category I rating, and a pendulum tire drop test classification. It is recommended by the maker for parchment, albumen, panel paintings and other media sensitive to humidity fluctuations, which is a different reason again: glass is a true vapor barrier.
Acrylic physical properties, from the two makers' published sheets: specific gravity 1.19; light transmission 92 percent total; Rockwell hardness M-93 to M-95; maximum recommended continuous service temperature around 71 to 88 degrees Celsius; coefficient of linear thermal expansion about 0.000072 metres per metre per degree Celsius, which is why a large sheet needs clearance in the rabbet. One maker warrants that its 92 percent light transmittance will not lose more than 2 to 15 percent of its light-transmitting ability over 30 years.
7. See it See it
The Okonkwos decide, twice. Nia and Marcus have fourteen photographs going above a sofa, with two children in the room, and three of the prints are irreplaceable. Known: each framed piece is 16 by 20 inches, which is 2.2 square feet, so 2.5 mm glass is about 2.7 pounds a sheet; the room gets afternoon sun through a west window and the wall reads 450 lux at its brightest. Assumed: that the three unique prints are at the more fugitive end, which nobody can confirm without testing them. Open: nothing about the material. Children climbing on a sofa under fourteen sheets of glass settles it, and the setup question answers itself: acrylic, and anti-static is no more expensive at this size than plain. For the three unique prints the exposure box says the wall spends about twenty times the published annual budget, so ultraviolet filtering buys them a factor of about 1.4 at the fugitive end and the real answer is to scan them and hang the copies. They do both.
- Ultraviolet filters, a conservation institute: the only independent testing of filter performance in this research, with the percentages and the spreads.
- Light, ultraviolet and infrared, the same institute: the Blue Wool dose table above, and the statement that ultraviolet usually causes less than half of the fading and often only a tenth.
- Conservation clear glass, a maker: the full comparison table across one range, and the trade definition of a "UV protection" label at 97 percent.
- Approximate weight of architectural flat glass: the weight table, in both unit systems, with the density for anything not on it.
8. Practice Practice
Minimum: one piece, with a glazing line written in the framing plan and a reason that is not about appearance. Time: about 30 minutes. Materials: your sizes, the weight calculator, and a frame shop's sample chips if you can get them.
- Work the weight both ways. Put your glazing size into the weight box with 2.5 mm glass, then with 3 mm acrylic, and write both numbers down. For a small piece the difference is trivial; at 24 by 36 it is not.
- Answer the three questions in order and write the answers. How big, how heavy, who is under it. If any of the three points at acrylic, that is your answer and the optics do not override it.
- Look at sample chips against the actual piece. In the room it will hang in, at the hour you will see it. Anti-reflective glazing is the one thing in this chapter you can judge by eye, and it is worth judging in place rather than in a shop.
- Write the glazing line in the framing plan. Material, thickness, filtering, and the sentence that explains it.
- Run the exposure box (Go deeper; optional). With a measured level from the wall you are considering, the exposure box shows what ultraviolet filtering buys on material at each end of the sensitivity range.
9. Your pictures Your pictures
10. Check your work Check your work
Criteria
- The glazing line names a material and a thickness, not a brand adjective.
- The reason for the choice is physical: size, weight, breakage risk, or a stated optical preference, in that order.
- If the piece is friable, the glazing choice is paired with at least 5 to 6 mm of space, and the acrylic is anti-static.
- If you bought ultraviolet filtering, you can say what it buys on your material and what it does not.
- Nothing in the package touches the glazing.
Worked example
Yusuf Ferrante (fictional) has a 30 by 40 inch painting on stretched canvas by a friend. He checks the chapter and finds he has no glazing decision to make: a stretched canvas is usually framed without glazing at all, and a conservation institute says that if glazing is used it must never touch the paint surface and the framing method must include a spacer. His 22 by 28 inch embroidered textile does need glazing, and the same institute says spacers stop the glazing crushing the three-dimensional quality of embroidery and reduce the risk of moisture condensing on the glazing. He chooses acrylic for the textile, on weight and on the apartment's humidity, with a spacer, and writes it down. Two pieces, two different answers, neither of them about reflections.
Common mistakes
- Choosing by optics first. Anti-reflective glass over a stair wall is still glass over a stair wall.
- Reading 99 percent as the measured figure. It is the claim. The only independent testing in this research averaged 83 percent for glass filters, with a 14 point spread.
- Treating ultraviolet filtering as fade protection. It is partial fade protection, and how partial depends entirely on the material.
- Putting ordinary acrylic over a pastel. This is the one case where the older agency guidance and the newer practice both agree something is wrong, even though they disagree about the fix.
- Cleaning acrylic with a glass cleaner, or with a paper towel. Follow the instructions for the product you bought; where you do not know, a dry soft cloth.
What is still unresolved
- How much space the stack actually gives in front of the art: the depth box.
- What the finished piece weighs, and what will hold it: Hanging it.
- Whether this wall is one a sensitive piece should be on at all: Where it can hang.
- Which level of framing the glazing choice sits in: The levels.