Glass

Back to the Objects Page

Objects Specialty Group Conservation Wiki
Contributors: Vanessa Muros, Molly Gleeson, Brittany Dolph, Geneva Griswold, Caitlin Mahony, Alexis North, Carinne Tzadik, Kara Fox, Rebecca Gridley, Madison Whitesell, Mackenzie Becker
Your name could be here! Please contribute.
Copyright: 2012. The Objects Group Wiki pages are a publication of the Objects Specialty Group of the American Institute for Conservation of Historic and Artistic Works. The Objects Group Wiki pages are published for the members of the Objects Specialty Group. Publication does not endorse or recommend any treatments, methods, or techniques described herein.

---

THIS ENTRY IS A DRAFT

History

It is widely accepted that glassmaking began in Mesopotamia in the Late Bronze Age, then spread to Syria and Egypt, then to the Mediterranean. There was widespread trade of glass goods in the Near East, but no archaeological evidence for its manufacture except in Egypt. However, glass vessels from Mycenae, which have been made in a very local style, were examined and shown to be made from glass from both Egypt and Mesopotamia. This points to a network of primary glassmaking locations, where ingots of solid glass were produced, and traded to be used in secondary production sites for finished objects (Smirniou 2011, 58-59).

Fritting or the manufacturing of frit is a process that was required in the earliest stages of glassmaking because of technological limitations with furnaces. Early societies that produced glass such as those in Egypt and Mesopotamia fritted because their furnaces could not reach and sustain temperatures over 1000 degrees Centigrade, which is needed to produce a homogenous glass without seeds (air bubbles) (Davison 2003:81). There is little written documentation about fritting, yet researchers have been able to study glass residues at recovered glass making sites to understand and reconstruct the process.

Materials and technology

Materials

Glass and frits

Glass is an amorphous (non-crystalline) solid material. Glass and frits are composed of three types of raw materials: network formers (silica or SiO₂), network modifiers (fluxes) in the form of alkali, and network stabilizers. Examples of fluxes, include soda (Na₂O) or potash (K₂O) and examples of stabilizers include lime (CaO) or magnesia (MgO). Other additives, such as colorants, de-colorants, or opacifiers, were perhaps accidental at first but were later added deliberately during manufacturing.

Once manufactured, soda-lime glass is more lustrous than potash glass and soda-lime glass is almost twice as durable as potash (Cronyn 1990:130). Lead may be added to replace some silica in a mixture to produce lead glass, which is softer and easily cut (for the production of lead crystal) and also used to make the fusible glass required for the manufacture of enamels (Cronyn 1990:129).

The melting point of pure silica glass is about 1700C but if other elements, modifiers are added to the glass, the MP falls to less than 1000C (Cronyn 1990:128).

Stained glass

“Stained glass” is glass colored by metallic oxide additives (ex. CuO) or by plating with a thin, cut section of colored glass adhered to one or both sides. The glass panels are then inserted into a network of lead came (Latin for “reed”) supports and soldered with a lead/tin (Pb/Sn) compound. The glass panels can then be painted, enameled, stained, etched, or engraved (Newton and Davison 1989:131). Silver-yellow was commonly used as a stain, while an opaque, dark brown or black iron oxide glass paint (grisaille) was used to outline forms and create shading# (Rousseau 2010:186).

Enamel

Similar to glass, enamels are made of a combination of silica, alkalis, alkaline earths, borax, lead compounds and metal oxides for coloring. Ideally, to match the physical characteristics of the metal, enamels contain a balance of soda and an alkali, such as borax (Maish and Scott 2001:10). Once the desired materials are selected, they must be combined and melted in a furnace to form a frit. After fritting, the enamel is poured out onto slabs and cooled. Once cooled, each enamel color is separately ground into a powder, and after washing and sifting to remove impurities, it can be placed on the surface of the metal in preparation for fusing.

Technology

Glass and frits

Producing a durable glass from the mixture of silica, alkali and stabilizers requires that the raw materials are chemically balanced. Early in the history of glassmaking it was found that a glass that was not chemically balanced would be easily soluble by water (Davison 2003:73). It was known that the addition of too much alkali to the glass batch had the effect of increasing the solubility of the resulting glass in water and that stabilizers could be added to the batch that would reduce the solubility. The content and balance of silica:flux:stabilizer in a glass is critical in determining its melting point (MP) and character (Cronyn 1990:129).

While frits are considered a material distinct from glass, it is thought that in antiquity, glass could only be made by the manufacture of frits as a first step. In fact, the development of glass technology in most ancient cultures probably grew out of frit manufacture, since it is unlikely that early furnaces could reach and sustain temperatures necessary to melt the glass components without the heavy addition of fluxes. As a result, the mixture retained much seed, or air bubbles, and the materials did not melt entirely.

Fritting is the preparatory step when the raw materials are mixed together and then heated to form a mass of sintered silicates (Nicholson and Henderson 2000:167). It is crucial that the heating process is properly conducted. The ancient recipes that do exist are very careful to emphasize this vital point in the process for it is at this point that the gas must be produced and escape into the atmosphere (Biek and Bayley 1979:1). If the temperatures are too high then the material becomes too viscous and the gas becomes trapped within the material. It is also important that during heating the mixture is consistently stirred, because it is undesirable to have and clumps of raw material (Davison 2003:82). The entire fritting process takes place in a fritting pan or crucible. There is not much information about these objects but examples have been found (Price 1976:114). Remnants of friable clay crucibles were discovered at the site Wadi el-Natrun in Egypt.

After this initial heating, a sintered silica mass is produced that is then ground into a fine powder, referred to as “frit” (Davison 2003: 73-83). At this point the material can be refritted, used as a pigment, or heated into a glass (Nicholson and Henderson 2000:168). The end of the fritting process is a homogenous glass ingot that can be shaped, molded, or transported (Davison 2003:82).

By later periods, all glass producing societies had developed furnaces advanced enough to reach and sustain high temperatures, allowing the raw materials to be directly transformed into glass. This development made the fritting step unnecessary and therefore obsolete. In modern glass production, following the mixing and heating/melting, the glass is refined to remove bubbles and then can be shaped or formed. Forming methods include the float glass process, for making flat glass, as well as blowing, pressing, cutting and grinding.

Faience objects were made in three ways: self-glazing by efflorescence, cementing, or surface application. In efflorescence, the core material was mixed with water-soluble alkali salts like carbonates, chlorides, or sulphates of sodium. When drying, the salts emerge on the surface in a crust or bloom. Once the object was fired, the crust became a glaze and fused with the ceramic body. In cementing, the body was shaped and dried, then buried in a glazing powder. When heated, the glaze would stick to the surface in a uniform but thin coating. For a surface application, either a faience powder or slurry was applied to the body, then fired (Davison 2003, 84).

Enamel

For an object to be classified as a true enamel, the vitreous substance must actually be fused to the metal surface. In order to attain proper fusion to the metal, the enamel must have a coefficient of contraction equal to that of the metal. The purer the metal, the better it will receive and retain the enamel. Pure gold and copper are the most suitable and most commonly found throughout the ages (Tait 1971). The metal is formed into the desired shaped, and then must be cleaned and roughened to allow better adhesion of the enamel. The enamel powder is then fused to the metal by heating it in a furnace, to a temperature approximately between 750-850C.

Identification

A wide range of methods and techniques can be used to identify glass and glass-like materials, as well as analyze vitreous objects to obtain information on raw materials, method of manufacture, material sourcing, condition and provenance.

Visual Examination

Visual examination, with or without magnification, is generally used to identify glass. Examination of the surface or the interior of the glass can often provide clues as to the nature of the material and whether it is frit, faience or glass.

Because glass is an amorphous solid that has melted, it should have a relatively homogenous appearance. Small air bubbles are sometimes visible within the matrix. If glass is damaged, it often exhibits a conchoidal fracture pattern. If the glass is deteriorated, looking for changes to the surface such as iridescence, could help identify the material as glass.

Late Bronze Age (11-10th c. BC) faience bead from the site of Lofkend (Albania) The quartz grains in the core are visible and the interior has a "sugary" appearance. Frit can also have this appearance.(Image courtesy of the Lofkend Archaeological Project)

Frit on the other hand, will look different than glass. Because frit is not heated to a high enough temperature to melt all the components, it will not appear amorphous like glass. Silica grains will still be visible within the structure.

Faience is similar in appearance to frit, having a sintered quartz body which appears "sugary". The surface of faience will be glazed and therefore appear different than the core material.

Examination Methods/Techniques

Some of the techniques used to examine glass include:

• Optical Microscopy
• Scanning Electron Microscopy
• Ultraviolet light
• Radiography

Instrumental Analysis

Some of the techniques used to analyze glass to identify raw materials, source or provenance the glass, or understand the condition of objects made from vitreous materials include:

• Electron Microprobe Analysis (EMPA)
• Energy Dispersive Xray Spectrometry (EDS)
• Inductively Coupled Plasma Mass Spectrometry (ICPMS)
• Secondary Ion Mass Spectrometry (SIMS)
• Ultraviolet and Visible Spectrometry (UV-Vis)
• X-ray Diffraction (XRD)
• X-ray fluorescence (XRF)

Deterioration

The main factors that affect the extent and type of deterioration of glass and glass-like materials are its composition and environmental factors. In regards to composition, the component often cited as being most influential in the condition and preservation of glass are the materials added as the network modifiers or flux, and the amount of network stabilizer (Newton and Davison 1989). The environmental factors that affect glass are quite numerous, but the main deteriorants include moisture, pH, atmospheric pollutants, and microorganisms.

Effects of Composition on Deterioration

The types of fluxes used, the silica content of the glass and the ratio of lime content to flux can all affect the level and types of deterioration observed in glass. Glass made using sodium (Na) as the modifier are thought to be more stable than glasses made with mixed alkalis or with potash as the flux. Glasses with a higher silica content are also considered more stable and less prone to deterioration (Freestone 2001; Newton and Davison 1989). This is why Roman glass, made with a high silica content and a Na based modifier, is more stable that earlier glass made with mixes alkalis (containing Na and K) or Medieval glasses made with potash. The higher silica content and addition of Na in the glass network make the glass less prone to leaching of ions within the matrix (Watkinson, et al. 2005; Freestone 2001).

The metal oxides added to glass used as colorants and opacifiers also have an effect on deterioration (Freestone 2001). The presence of Al₂O₃, Fe₂O₃, and P₂O₅ in the glass can increase its resistance to corrosion or deterioration due to the low solubility of these oxides. It is thought that they immobilize the alkali ions preventing them from moving through the silicate network and being leached out of the glass matrix (Freestone 2001; Newton and Davison 1989).

Water

Water is one of the main causes of deterioration of buried glasses. Water deteriorates glass via two mechanisms: ion exchange and network dissolution (Schalm, et al. 2011; Freestone 2001; Newton and Davison 1989). In ion exchange, mobile cations are leached from the glass structure and are substituted with a proton from the environment (Schalm, et al. 2011). In the case of water, a hydrogen ion from the water substitutes an alkali ion (such as Na or K) that has been removed from the glass network. The result is a hydrated, silica rich layer sometimes referred to as a “gel”. Because hydrogen is much smaller than the alkali ion it replaced, and forms a strong bond with oxygen, it creates a glass network that is weaker and contracts, making it more prone to further deterioration (Freestone 2001). Analysis of weathering layers in glass has found not only that it is silica rich, but can often be enriched in other ions such as iron, aluminum, manganese and other cations due to weathering.

As ion exchange occurs, the pH of the water on the surface of the glass begins to increase if it is not replenished and can cause the second deterioration mechanism, network dissolution (Schalm, et al. 2011; Freestone 2001. At a pH above 9, hydroxyl ions in the solution attack the silicate network causing the dissolution of silica (Newton and Davison 1989). It has been found that acidic conditions are more deteriorative to glasses when they had a high lime content, greater than 10% (Newton and Davison 1989). This means that the pH of the solution on the surface greatly affects the types of deterioration occurring and what ions are leached from the glass matrix, as well as the rate at which it occurs.

Ion exchange is thought to be the primary form of deterioration of archaeological glass due to the relatively neutral pH of groundwater, but both mechanisms can occur simultaneously. The rate at which one occurs over the other affects the appearance and composition of the deteriorated surface. If loss of the alkali occurs at a faster rate that network dissolution, then the surface is comprised mainly of hydrated silica. If network dissolution is the primary cause of glass corrosion, no silica rich layer is created but instead the surface of the glass is etched away (Freestone 2001).

In addition to the deterioration caused by liquid water, moisture , or water vapor, can also cause deterioration. Water from humid environments can penetrate the surface of the glass and can result in similar deterioration mechanisms as those described above. (Newton and Davison 1989).

Atmospheric Pollutants

Atmospheric pollutants, and in particular sulfur dioxide, are often cited as a cause of deterioration, especially in the case of stained glass (Newton and Davison 1989). However there has been no direct evidence that sulphur dioxide attacks glass. Instead, the reaction of sulfur dioxide with glass occurs as part of several stages that result in the deterioration of glass by the atmosphere.

In the initial stages of deterioration in the atmosphere, water attacks the surfaces as described above and forms hydroxides. Carbon dioxide in the air then converts these hydroxides to carbonates. In the final stages, sulfur dioxide interacts with the carbonates converting them to sulfates (Newton and Davison 1989).

Microorganisms

Microorganisms suchs as lichens, mosses, algae, fungi and bacteria can be found on glass and help promote deterioration (Newton and Davison 1989). Lichens, mosses and algae do not require nutrients to grow but they cannot grow on clean glass. Therefore the surface of the glass must be altered either through pitting, dirt or grease in order for them to attach to the surface. These organic growths do not alter the glass directly but promote deterioration by trapping moisture on the surface resulting in the types of deterioration observed by liquid water or moisture. In the case of lichens, water has been found to be trapped between the lichen and surface of the glass due to capillary action. CO₂ released from the lichen into the water also helps to promote chemical attack of the glass.

Fungi and bacteria have been found to attack glass, but need a source of nutrients to survive. Fungi seem to be able to use CO₂ from the atmosphere to obtain their food (Newton and Davison 1989). Certain species of bacteria are able to metabolize certain metallic ions in the glass causing further surface deterioration (Perez y Jorba, et al 1980). For example ferrobacteria can metabolize manganese in glass. Analysis of microorganism activity on stained glass has shown the presence of sulphur-reducing bacteria causing damage. The formation of calcium oxalate on the surface of stained glass has been observed as the result of microorganisms such as bacteria and lichens secreting oxalic acid onto layers of calcite formed on the surface of the glass.

Deterioration of Stained Glass

Stained glass can exhibit types of deterioration not seen in buried glasses. For example, weathering crusts are primarily composed of gypsum or syngenite. These corrosion deposits occur on glass that has too much lime or too little silica because the calcium leaches out and forms salts (Griffiths 1989: 129). The glass can exhibit fracturing or stress due to thermal expansion.

The paint layers on stained glass can serve to protect glass or can be damaged due to different factors. The paint is often more stable than the glass substrate. Yellow stain and grisaille, fired onto the glass at about 1250˚ F, can serve either as a protective layer or cause deterioration of the glass surface (Davison 1989: 192). The effects of enamel paint may be related to the paint’s alkali component, promoting corrosion if the content is too high while protecting it if it is too low. As well, the protective effect of silver stain may be due to the sodium ions in the surface having been replaced with silver ions, as “leaching of the silver would not make the solution alkaline in the way that sodium ions would” (Davison 1989: 192). Neither hypothesis has yet been validated. Paint loss is associated with desiccated surfaces that flakes the paint. This is sometimes caused by excess calcium sulfates on the surface (Griffiths 1989: 129).

Deterioration of Enamels

In addition to the deteriorants described above, enamels have been found to also suffer damage due to the raw materials used as colorants. Though in glass making the proportions of silica-alkali-flux and other components could be adjusted to increase the stability of the glass (Davison 2003: 73)], in enamels the balance of the components was altered in order to achieve desired colors. It appears that color took precedence to the stability of the enamel for it has been observed that some enamels began deteriorating soon after their manufacture. Currently throughout collections of the world, enamels are showing differing level of deterioration based upon their color (Drayman-Weisser 2003;Hughes 1987; Smith, et al. 1987). Conservation efforts to understand more clearly the deterioration of enamels has led to investigations of the ions involved in the active deterioration of enamels.

Active deterioration on enamels includes the phenomena of crizzling, weeping, spalling, and salt efflorescence, which result in the loss of translucency, cracking and pitting of surface, and loss of material (Drayman-Weisser 2003: 280). While enamels clearly suffer damage due to handling and corrosion of the underlying metallic substrate, evidence also shows that there is inherent instability enamel that leads to its own deterioration (Drayman-Weisser 2003). The severity of deterioration has been associated in many collections to color. For example, both Davidson and Drayman-Weisser note the severe deterioration associated with blue, mauve, and violet enamels (Davison 2003: 321; Drayman-Weisser 2003: 281). On the other hand, Smith, et al. (1987: 109) notes the level of stability of flesh toned enamels. Both Drayman-Weisser and Smith provide the elemental analysis of these colored enamels, which were achieved through XRF, XRD or electron beam microprobe to connect the color to the ion. In the case of the related blue and violet colors, these enamels were found to have higher concentrations of potassium ion. The flesh tones were found to consist of higher amounts of lead (Drayman-Weisser 2003: 289). Both the instability of the blues and stabilities of flesh colors are caused by these ions’ interaction with water. High amount of potassium ions will lead to a hydration of the glass surface. When the glass layer becomes hydrated the silica network is attacked and dissolves. If this layer dehydrates, shrinkage occurs causing cracks or spalling. Lead stabilizes the glass surface because it is water resistant.

Types of Deterioration Observed on Glass and Glass-like materials

• Dulling causes glass to become less transparent (Newton and Davison 1989).
• Iridescence gives the surface of glass a rainbow-like appearance with the colors visible changing as light hits the glass at different angles. It initially can appear in patches or a thin layer or the surface, but as the glass continues to deterioration, the iridescence can flake off and eventually cause powdering of the glass. Iridescence is the result of the continued breakdown of a silica rich layer on the surface of the glass (Newton and Davison 1989).
• Strain cracking is used to describe a series of small cracks observable through the glass that run in all directions. As the vessel continues to deteriorate, the glass takes on a sugary appearance and eventually disintegrates. This cracking may likely be due the hydrated silica gel layer on the surface of the glass drying out. (Newton and Davison 1989).
• Pitting
• Corrosion crusts on glasses can vary in thickness , color and morphology. The appearance seems to be influenced by the composition of the glass, burial environment or atmosphere and the length of time buried or exposed to atmospheric conditions. The structure is often described as lamellar and has been described as pale in color or can be dark if it incorporates ions such as iron and manganese (Newton and Davison 1989). The corrosion crust is rich in silica and has a decreased amount of network modifiers and stabilizers (Freestone 2001). The crust is often hydrated.
• Black or brown staining is often found on glasses from buried contexts (Newton and Davison 1989). The staining can appear to be dendritic in morphology. Analysis of these stains has identified manganese and iron in the stains (Weber, et al. 2007). The Mn and Fe could have been diffused into the glass matrix from the environment or are due to enrichment of those elements in weathering layers or corrosion crusts or their movement to the surface from the glass matrix (Watkinson et al. 2005).
• Salt efflorescence
• Glass Disease (also known as “sick glass”) has been observed on many glasses made between the 17th-19th centuries due to deterioration. When glass is “weeping” droplets of moisture are visible on the surface as a result of alkali stabilizers–Ba, Al, Ca, Mg- leaching from the glass (Davison 1989: 191). When the surface liquid evaporates as the air dries, small cracks known as crizzling remain (Pilosi and Wypyski 2001: 68). The presence of moisture also causes opaque or dull corrosion layers, while fluctuations in the relative humidity can cause flakes of glass to detach from the surface. Objects prone to crizzling exhibit imbalanced compositions- typically, too much alkali and too little stabilizer. A suitable lime (CaO) content (more than 5%) is lacking that would help to prevent hydrolysis, causing instead a surplus of alkali, such as KO or NaO (Newton and Davison 1989; Brill 1972: 46). The presence of potash flux has been found to exacerbate this type of deterioration. The degradation process commences when the alkali hydrates, caused by humidity fluctuation, which can begin both gradually or rapidly.

• Devitrification One type of deterioration that affects glass is devitrification. Archaeologists and conservators use this term in connection to the loss of vitreous nature of the glass (Davison 1989:5). It should be noted that the term devitrification is also considered the process of crystallized products appearing on the surface of the glass during the firing process. Glass is a solid material composed of four main elements: silica (sand), alkalines/fluxes (sodium carbonate or potassium carbonate), stabilizers (calcium oxide or magnesia), and colorants. The composition and ratio of these elements, particularly the proportion of silica to alkali and stabilizers determines the stability of the glass (CCI 2011). Physical properties such as transparency, brittleness, and hardness are characteristic of each glass composition, the most important being chemical durability which is the key to deterioration (Romich 2006:164).

Devitrification in the archaeological sense, is a natural process that chemically degrades siliceous materials in the archaeological record (Romich 2006:164). Devitrification occurs when the surface of the glass becomes crystalline through the absorption of moisture from the environment. The sodium and potassium oxides existing in glass materials are hygroscopic (meaning retaining moisture), thus the surface of the glass inherently absorbs the moistures of its environment. The absorbed moisture combined with the exposure to carbon dioxide causes the oxides to convert to sodium or potassium carbonate. In moist environments, drops of moisture will appear on the surface of the glass, causing the hygroscopic carbonates to leach out of the glass leaving behind a fragile, hydrated silica material. This fragile silica deposit becomes partially crystalline, exposing an identifying white, cloudy haze (Hamilton 2011). This process of chemical weathering is known as devitrification. The white, cloudy haze covering the glass surface will indicate devitrification. Once the devitrified surface has been identified and documented, the object must be very carefully cleaned in a tap water or deionized water bath. The cleaning of the object must be done with caution, as to avoid removing important layers of the glass surface. Once the object has been thoroughly cleaned, it should be allowed to air dry. Apply a solution of 5-10% Paraloid B-72 in an acetone solvent with a soft brush to the surface of the glass. This solution will fill in the porous surface making the glass appear clearer and more transparent. Lastly, it is important to store the glass in a dry environment no higher than 40-45% relative humidity. Keeping the object out of a moist environment will prevent the glass from returning back to a state of deterioration (Hamilton 2011).

Conservation and care

This information is intended to be used by conservators, museum professionals, and members of the public for educational purposes only. It is not designed to substitute for the consultation of a trained conservator.

• To find a conservator, please visit AIC's Find a Conservator page.
• To learn how you can care for your personal heritage, please visit AIC's Resource Center.
  

Preventive Conservation

Objects made of glass, faience and frit should be stored and displayed in a stable environment that will deter further deterioration. Glass is hygroscopic and therefore a stable relative humidity (between 40-45%) and temperature (approximately 22˚ Celsius)is needed (Koob 2010: 131). Fluctuations causing condensation (RH above 70%) or dehydration are avoided. This is especially true for objects that are deteriorated or unstable. Particle and gaseous pollutants should also be controlled.

For unstable glass, creating a stable storage or display environment can often be best accomplished through the creation of microclimates using conditioned silica gel and pollution scavengers. For example, the Victoria and Albert Museum in London uses conditioned Artsorb silica gel that is equilibrated to 38% to store their unstable glass (Davison 2003, 316).

Interventive treatments

Cleaning

Whether or not a glass object can safely be surface cleaned should be carefully evaluated. Glass that shows any surface instability such as weathering, deterioration, lifting gilding, or cold painting may not be safely cleaned with dry or aqueous methods. Glass with structural damage such as cracks or losses will be vulnerable during cleaning.

Structurally stable glass can be washed in a plastic tub with mildly warm water and diluted detergent and rinsed with deionized water (Koob 2006, 39). A nonionic detergent such as Synperonic A-7 or Triton XL-80N should be diluted between 1:10-1:20 (Koob 2006, 40; Davidson 2003: 200, 273). While in the bath, mechanical action can be applied with a soft cotton paper towel, plastic bottle brushes (such as those for lab glassware) or a paint brush. Gloves should not be worn to ensure the safest possibly handling of the slippery glass surface. Excessive residual water should be gently toweled off and then the glass can be allowed to air dry. For structurally compromised glass or localized cleaning, the water and detergent can be applied by swab and rinsed with deionized water.

Some surface grime or stains may remain after surface cleaning and require solvents to fully remove. Greasy surface dirt may be removable with an alcohol (such as acetone). Koob (2003) recommends starting with a dilute concentration (1:1) in deionized water, followed by pure alcohol if necessary. Especially stubborn surface grease or wax can be removed with naphtha (Koob 2003; 43).

Stain Reduction

Glass may have surface staining from use that cannot be removed with general surface cleaning. Reduction of such stains may be possible but should be carefully considered in regards to each specific object as many reducing agents may also affect the glass surface under certain conditions.

Organic stains can be reduced with 10-15% sodium hydroxide in water (Koob 2006; 43). The solution should be applied for 10-30 seconds using a synthetic brush and then thoroughly rinsed with deionized water. Proper PPE and caution should be used when handling the sodium hydroxide solution. Davidson (2003; 202) also mentions hydrogen peroxide “diluted to 10 vol.” in water as a method of organic stain reduction.

A 10-15% sodium hydroxide solution can also be used to dissolved silicate deposits (which are rarely seen on glass) without damaging the glass beneath (Koob 2006; 43). The same application method as organic staining above can be used.

Iron stains can be removed with a cotton poultice of 5-10% solution of oxalic acid in water left on the surface for up to 15 minutes (Koob 2006, 43). Gluconic acid at a low pH (such as 7) may reduce iron staining safety but all chelators run a risk of attacking the glass structure (especially at a higher pH) and may make the glass more vulnerable to future deterioration (Davidson 2003; 202).

Carbonate deposits, often the result of tap water evaporation, can be dissolved in 3-5% nitric acid (Koob 2006; 43). The acid can be applied by cotton swab and then thoroughly rinsed with soap and water, followed by deionized water. It is worth noting that mineral acids, including nitric acid, hydrochloric acid, and hydrofluoric acid can all severely damage unstable glass and even stable glass (Davidson 2003; 203).

Stabilization

Consolidation

Glass consolidation in objects conservation presents unique challenges due to the material’s inherent properties, diverse composition, and historical significance. These challenges are compounded by the need for ethical considerations and the reversibility of treatments.

Glass compositions differ significantly across historical periods and manufacturing techniques, influencing how glass deteriorates and how it reacts to treatments. For example, soda-lime glass is prone to crizzling and surface delamination, whereas potash-rich glass is susceptible to weeping and environmental degradation (Brill 1972). Such variability complicates the selection of suitable consolidants and treatment protocols.

Glass is non-porous but is an inherently brittle material, often presenting with micro-cracks that are not always visible to the naked eye. These cracks increase the risk of damage during consolidation and may cause uneven penetration of consolidants. According to Koob (2006), stress concentration at fracture edges can exacerbate damage if treatments are not carefully controlled.

Maintaining the optical clarity and appearance of glass is critical in conservation, especially for decorative or functional objects. Disparities in refractive indices between the original glass and the consolidant can result in visible repairs, disrupting the object’s aesthetic integrity. Furthermore, some consolidants may yellow over time, further compromising the visual fidelity of the conserved object.

One of the fundamental principles of conservation is reversibility. However, achieving a balance between stabilization and reversibility in glass treatments is difficult. Consolidants like epoxies or acrylic resins may offer short-term stability but often become brittle or discolored with age, making future removal challenging (Down 2015).

Many types of historical glass are hygroscopic, meaning they absorb moisture from the environment, which exacerbates degradation. For example, fluctuating humidity levels can lead to further crizzling or stress fractures, complicating consolidation efforts (Newton and Davison 1989).

Innovative approaches, such as the use of silica-based sol-gel consolidants and laser-assisted bonding, are being explored to address these challenges. These materials can deeply penetrate glass without altering its appearance and are seen as more durable and stable compared to traditional consolidants. Nanosilica, for example, can be applied as a colloidal solution that fills micro-cracks and reinforces the structure, providing a less invasive alternative to conventional methods. These methods aim to provide greater material compatibility, reversibility, and long-term stability. While Paraloid B72 remains standard (Koob 2006), innovations such as nanomaterials are promising (Yanming Zhang, Liping Huang, and Yunfeng Shi 2019). However, continued research is essential to evaluate their long-term performance in practical conservation environments.

Structural treatments

Joining Fills

Glass can be joined with a variety of materials including Paraloid B72 or the epoxies like Hxtal NYL-1. Selecting the correct material depends on a variety of factors including opacity, refractive index, and working time.

Opaque glass

Transparent glass

Blind cracks

Aesthetic reintegration

Loss compensation, fills, casting, molding, re-touching, finishing, etc.

Transparent fills

Colored Fills

Opaque Fills

Using molds

Surface treatments

Polishing, coatings, etc.

References

History & Technology

Biek, L. and J. Bayley. 1979. Glass and other Vitreous Materials. World Archaeology: Early Chemical Technology 11(1): 1-25.

Hatton, G.D., Shortland, A. J., and M. S. Tite. 2008. The production technology of Egyptian blue and green frits from second millennium BC Egypt and Mesopotamia. Journal of Archaeological Science 35: 1591-1604.

Moorey, P. R. S. 1999. Ancient Mesopotamian materials and industries. Winona Lake, Indiana: Eisenbrauns.

Nicholson, P. T. and J. Henderson. 2000. Glass. In Ancient Egyptian Materials and Technology'5, eds. Nicholson, P. T. and I. Shaw. Cambridge: Cambridge University Press. 195-226.

Maish, J.P. and Scott, D.A. 2001. Glass enamel on a 4th century BC Madedonian gold wreath.In Annales du 15e Congres de l’Association Internationale pour l’Histoire du Verre, New York. 8-12.

Ogden, J. 2000. Metals. In Ancient Egyptian Materials and Technology, eds. Nicholson, P. T. and I. Shaw. Cambridge: Cambridge University Press. 148-176.

Price, J. 1976. Glass. In Roman Crafts, eds.D. Strong and D. donBrown. London: Duckworth.

Shortland, A. J. and M. S. Tite. 2008. Production Technology of Faience and Related Early Vitreous Materials. Oxford, England: Oxford University School of Archaeology.

Smirniou, M. 2011. Direct evidence of primary glass production in Late Bronze Age Amarna, Egypt. Archaeometry 53(1): 58-80.

Conservation Issues, Treatment & Preventive Care

Augerson, C., and J.M. Messinger II. 1993. Controlling the Refractive Index of Epoxy Adhesives with Acceptable Yellowing after Aging. Journal of the American Institute for Conservation 32(3): 311-314.

Barack, S. 2016. 3D Printing and Fills on Glass Vessels: A Case Study from the Fraunces Tavern Museum. In Recent Advances in Glass and Ceramics Conservation 2016. Paris: International Council of Museums - Committee for Conservation (ICOM-CC). 199-204.

Bradley, S.M. and S. E. Wilthew. 1984. The Evaluation of Some Polyester and Epoxy Resins Used in the Conservation of Glass. ICOM 7th Triennial Meeting, Copenhagen, 1984 Preprints Vol. 2: 84.20.5-.9.

Brill, R. H. 1972. "Crizzling: A Problem in Glass Conservation." Journal of Glass Studies, 14, 34–46.

Bristow, H., and J.D. Cutajar. 2017. Archaeological Glass Conservation: Comparative Approaches and Practicalities of Using Acrylic Resin Films as Gap Fills. American Institute for Conservation Objects Specialty Group Postprints 24: 188-206.

Canadian Conservation Institute. 2007. CCI Notes 5/1 Care of Ceramics and Glass. Accessed April 15, 2013

Cronyn, J. M. 1990. The elements of archaeological conservation. London: Routledge.

Davison, S., and N.S. Brommelle. 1984. A Review of Adhesives and Consolidants Used on Glass Antiquities. In Adhesives and Consolidants: Preprints of the Contributions to the Paris Congress. London: International Institute for Conservation of Historic and Artistic Works. 191-194.

Davison, S. 1998. Reversible Fills for Transparent and Translucent Materials. Journal of the American Institute for Conservation 37(1): 35-47.

Davison, S. 2009. A History of Joining Glass Fragments. In Holding it All Together: Ancient and Modern Approaches to Joining, Repair and Consolidation. London: Archetype Publications in association with the British Museum. 107-112.

Davison, S. 2003. Conservation and Restoration of Glass, Second Edition. San Diego: Butterworth Heinemann.

Down, J.L. 1984. The Yellowing of Epoxy Resin Adhesives: Report on Natural Dark Aging. Studies in Conservation29(2): 63-76.

Down, J.L. 1986. The Yellowing of Epoxy Resin Adhesives: Report on High-intensity Light Aging. Studies in Conservation 31(4): 159-170.

Down, J.L. 2015. The Stability of Adhesives Used in Conservation. Canadian Conservation Institute.

Drayman-Weisser, T. 2003. The Early Painted Enamels of Limoges in the Walters Art Museum: Historical Context and Observations on past Treatments. Journal of the American Institute for Conservation 42(2): 279-312.

Freestone, I. C. 2001. Post-depositional changes in archaeological ceramics and glasses. In Handbook of archaeological sciences, eds. D.R. Brothwell and A.M. Pollard. London: John Wiley & Sons, Ltd. 615-625.

Gridley, R. and K. Stamm. 2019. Seeing Clearly: Casting Epoxy Fills for Glass Objects Using Transparent Molds. American Institute for Conservation Objects Conservation Specialty Group Postprints 26.

Gridley, R. and K. Stamm. 2019. Using Double-walled PVC Foil Molds to Separately Cast Large Epoxy Resin Fills for Glass Objects. In Recent Advances in Glass, Stained-Glass, and Ceramics Conservation 2019. 101-110.

Griffiths, D.R. and A.M. Feuerbach. 2001. The Conservation of Wed Medieval Window Glass: A Test Using an Ethanol and Acetone Mixed Solvent System. Journal of the American Institute for Conservation 40(2): 125-136.

Hamilton, Donny. 2011. Conservation of Glass. In: Conservation Research Laboratory of the Nautical Archaeology Program. Texas A&M University. Accessed April 15th, 2013.[2].

Hughes, M. J. 1987. Enamels: Materials, Deterioration and Analysis. In From Pinheads to Hanging Bowls: The Identification, Deterioration and Conservation of Applied Enamel and Glass Decoration on Archaeological Artifacts, Occasional Papers no. 7, eds. L. Bacon and B. Knight.

Koob, S.P. 2006. Conservation and Care of Glass Objects. London: Archetype Publications in association with The Corning Museum of Glass.

Koob, S.P., A. Roy, and P. Smith. 2000. New Techniques for the Repair and Restoration of Ancient Glass. In Tradition and Innovation: Advances in Conservation: Contributions to the Melbourne Congress. London: International Institute for Conservation of Historic and Artistic Works. 92-95

Koob, S.P. 2002. Four Beakers or Two Beakers? Only Your Conservator Knows for Sure... American Institute for Conservation Objects Specialty Group Postprints 9: 114-130.

Koob, S.P. 2003. Tips and Tricks with Epoxy and Other Casting and Molding Materials. American Institute for Conservation Objects Specialty Group Postprints 10: 158-172.

Koob, S.P. 2004. Cleaning Glass: A Many-faceted Issue. American Institute for Conservation Objects Specialty Group Postprints 11: 60-70.

Koob, S.P. 2010. An Experimental Treatment for Severely Crizzled Glasses. In Glass and Ceramics Conservation 2010: Interim Meeting of the ICOM-CC Working Group. Corning: Corning Museum of Glass.128-132.

Koob, S.P., S. Benrubi, N.A.R. Van Giffen, and N. Hanna. 2011. An Old Material, a New Technique: Casting Paraloid B-72 for Filling Losses in Glass. In Symposium 2011: Adhesives and Consolidants for Conservation: Research and Applications: Proceedings. Ottawa: Canadian Conservation Institute. 1-14.

Koob, S.P., Van Giffen, N.A.R., Kunicki-Goldfinger, J.J. and Brill, R.H., 2018. Caring for glass collections: the importance of maintaining environmental controls. Studies in Conservation, 63(sup1):146-150.

Kunicki-Goldfinger, J.J., 2008. Unstable historic glass: symptoms, causes, mechanisms and conservation. Studies in Conservation, 53(sup2): 47-60.

Lemajič, G. 2006. Transparent PVC Mould: Replacing Missing Pieces on Hollow Glass Objects. Icon News 3: 46-48.

Mertik, N., and G. Lemajič. 2007. Conservation of a Glass Cup Using a Transparent Mould Technique. In Glass and Ceramics Conservation 2007. 83–84.

Messinger, J.M., II, and P.T. Lansbury. 1989. Controlling the Refractive Index of Epoxy Adhesives. Journal of the American Institute for Conservation 28 (2): 127-136.

Mills, Allan A. 1987. The Conservation and Restoration of Mediaeval Stained Glass Windows by ‘Gel-Plating’. Studies in Conservation 32(3): 122-136.

Newton, R. and Davison, S. 1989. Conservation and Restoration of Glass. Oxford: Butterworth-Heinemann.

Perez y Jorba, M., Dallas, J.P., Bauer, C., Bahezre, C., and J.C. Martin. 1980. Deterioration of stained glass by atmospheric corrosion and micro-organisms. Journal of Material Science 15: 1640-1647.

Pilosi, L. and M.T. Wypyski. 2001. Technical Examination and Conservation of Glass. The Metropolitan Museum of Art Bulletin, Ars Vitraria: Glass in the Metropolitan Museum of Art 59(1): 66-68.

Roemich, H. 2006. Chapter 7: Glass and Ceramics. In: Conservation Science: Heritage Materials, Cambridge. 160-184.

Rousseau, D.R. Keeping Out the Wind: Repair and Restoration History in Stained Glass Windows. Glass and Ceramics Conservation 2010, ed. H. Roemich. ICOM Committee for Conservation in association with The Corning Museum of Glass. International Council of Museums. 185-193.

Schalm, O., K. Proost, K. de Vis, S. Cagno, K. Janssens, F. Mees. P. Jacobs, and J. Caen. 2011. Mangenese staining of archaeological glass: the characterization of Mn-rich inclusion in leached layers and a hypothesis of its formation. Archaeometry 53(1): 103-122.

Schuck, T., A. Fischer, and B. Schwahn. 2019. Acrylic Barrier Surface Coatings for Epoxy Resin Adhesive Bonds in Glass Conservation: Evaluation of Bond Strength and Reversibility. In Recent Advances in Glass, Stained-Glass, and Ceramics Conservation 2019. 131-140.

Smith, R., Carlson, J.H., and R.M. Newman. 1987. An Investigation into the Deterioration of Painted Limoges Enamel Plaques c. 1470-1530. Studies in Conservation 32(3): 102-113.

Stamm, K., G. Lemajič, and L. Pilosi. Vacuum-formed PVC Moulds for Casting Epoxy Resin Fills in Glass Objects. In Recent Advances in Glass, Stained-Glass, and Ceramics Conservation 2013. 69-75.

Tait, H. 1971. A thousand years of enamel. Exhibition catalog for the Warski to mark their Diamond Jubilee in London, 1911-1971. Printed in Enland by Graphics Press Limited, London.

Tennent, N. H., J. H. Townsend, and D.D. Froment. 1984. Factors Affecting the Refractive Index of Epoxy Resins. In ICOM Committee for Conservation 7th Triennial Meeting: Copenhagen. Paris: International Council of Museums. 84.20.27-84.20.28.

Tennent, N. H., J. H. Townsend, and N.S. Brommelle. 1984. The Significance of the Refractive Index of Adhesives for Glass Repair. In Adhesives and Consolidants: Preprints of the Contributions to the Paris Congress. London: International Institute for Conservation of Historic and Artistic Works. 205-212.

Tennent, N. H. 1999. The Conservation of Glass and Ceramics: Research, Practice and Training. London: James & James.

Tennent, N. H., and S. P. Koob. 2010. An Assessment of Polymers Used in Conservation Treatments at the Corning Museum of Glass. In Glass and Ceramics Conservation 2010: Interim Meeting of the ICOM-CC Group.Corning: Corning Museum of Glass. 100-109.

Thickett, David, and Denise Ling. 2021. Investigation of weeping glass deterioration under controlled relative humidity conditions. Studies in Conservation: 1-7.

Van der Wielen, R., and S. de Groot. 2019. Thermocasting of PARALOID B-72: Solvent-Free Production of Acrylic Flat-Glass Restoration Casts. In Recent Advances in Glass, Stained-Glass, and Ceramics Conservation 2019. 111-120.

Van Giffen, N. A. R., S. P. Koob,  and R. O'Hern. 2013. New Developments for Casting Paraloid® B-72 for Filling Losses in Glass. In Recent Advances in Glass, Stained-glass, and Ceramics Conservation 2013. Zwolle: SPA Uitgevers. 53-60.

Van Giffen, N. A. R., and S. P. Koob. 2016. Cleaning and Repairing Blaschka Glass Models: Challenges in the Preservation of Exquisite Reproductions of Nature. In Recent Advances in Glass and Ceramics Conservation 2016. Paris: International Council of Museums - Committee for Conservation (ICOM-CC). 205-214.

Verhaar, Guus. 2018. Glass Sickness: Detection and Prevention: Investigating unstable glass in museum collections. [Thesis, fully internal, Universiteit van Amsterdam]. https://dare.uva.nl/personal/pure/en/publications/glass-sickness-detection-and-prevention(669f19fd-9169-40f7-aec3-8bd568d9ab8c).html (accessed Oct. 5, 2023.).

Watkinson, D., Weber, L., and K. Anheuser. 2005. Staining of archaeological glass from manganese-rich environments. Archaeometry 47(1): 69-82.

Weber, L., Eggert, G., and D. Watkinson. 2007. A closer look at brown staining on archaeological glass. In Glass and Ceramics Conservation 2007, Intermediate meeting of the ICOM-CC working group, August 27-30, 2007, Nova Gorica, Slovenia, ed. L. Pilosi. 35-45.

Wight, K. 2009. The Aging of Fills in Glass and High-Fired Ceramics. Masters Thesis for Winterthur/University of Delaware Program in Art Conservation.

Yanming Zhang, Liping Huang, and Yunfeng Shi. 2019. Silica Glass Toughened by Consolidation of Glassy Nanoparticles. Nano Letters 19 (8), 5222-5228. DOI: 10.1021/acs.nanolett.9b01634.

Back to the Objects Page