Old Brown Glue

Woodworker's Journal – Issue: 112 – Posted Date: 11/30/2004

Written by: Michael Dresdner

For most woodworkers, the name W. Patrick Edwards evokes images of some of the finest, most complex marquetry on the planet, and so it should. Pat, as he is know to friends, is the expert who owns and runs The American School of French Marquetry, and executes eye-popping inlay to back up his well-deserved reputation. What you may not know is that he, his wife, Kristen Arrivee, and Matthew, one of their three sons, manufacture and sell Old Brown Glue.

What is Old Brown Glue? It's Pat's "boutique" version of hide glue, modified to give it better handling properties and a longer, yet variable, open time. It's used by Brian Boggs, Kelly Mehler, and a host of antique repair people and chair makers who love the controlled open time, which can vary from 20 minutes to an hour merely by changing the temperature in the shop.

As is often the case, the glue was created to fill a need by its developer, who then concluded that others would also appreciate having it. To understand how it came about, we need to get a bit of background both on hide glue, and on the type of work it benefits. In Pat's case, it's marquetry and antique restoration.

A native of San Diego, Pat has owned and operated Antique Refinishers, Inc. in the same location for the past 35 years. After studying in Paris at Ecole Boulle under Pierre Ramon, he focused his business on restoring pre-industrial furniture, (18th and 19th century or earlier) and specializing in veneer and marquetry. Four years ago, he and Kristen started the American School of French Marquetry at the same location. It is the only school outside of Paris that teaches the French method of marquetry, which allows you to make multiple copies quickly and accurately using only hand tools . Veneers are cut on a curious device called a "chevalet," or "marquetry donkey," that looks like a cobbler's bench on steroids. The school has been so successful that they are currently expanding the building.

"Originally, I started using a glue pot with hot hide glue. It did everything well. It was strong, transparent to stains, easy to clean up, reversible, economical, and the only adhesive which glues to itself both mechanically and chemically. That makes it ideal for repairing antique furniture, all of which, incidentally, was originally made with hide glue. If a joint breaks, you simply add more glue without having to clean off the old, and you get a perfectly strong joint."

"One of the trends I see today is the use of non-reversible glues. Furniture that will last will eventually need to be repaired. In fact, sometimes, you need to repair during construction. In my restoration work, reversibility is essential. Hide glue is reversible, even after many decades."

By manipulating heat and moisture, you can modify how hide glue behaves, affecting viscosity, open time, and cure time. This control is especially important with veneer work, chairs, and other complex assemblies. But hot hide glue requires almost constant attention, and sets too quickly for some veneer operations. "At times, you must overlap the veneer, because it can shrink and pull back from the joint while curing. Hence, you need a glue that takes a longer time to set, allowing the veneer to shrink before the seam is cut. Old Brown Glue does just that."

It started in 1996 during a European trip, where he joined a marquetry conservation group discussion on modifying hide glue to be liquid at room temperature. Back home, he started work on his own formula.

"While other liquid hide glues exist, I found they gave me uneven results, probably because the glue was not always fresh." Though dry glue granules have an indefinite shelf life, once it is mixed, it starts to deteriorate. "Other companies use preservatives, but that does not always work. Besides, some of the preservatives are hazardous, and since glue tends to get on your skin, I wanted to use only organic ingredients. I wanted to make glue as simple and safe as possible."

True to his mission, Old Brown Glue is made in small batches and sold fresh, like upscale coffee or handmade chocolate truffles. The shelf life is about one year after you buy it, but you can extend that time by keeping the glue in the refrigerator. "No one else sells fresh, liquid hide glue cooked in small batches," says Pat.

It's easy to see why. Old Brown Glue starts in the kitchen of their building, where Milligan and Higgins' 192 gram strength hide glue is cooked, along with the other ingredients, in a huge double boiler. It takes four days, during which time the pot must be stirred every hour. They bottle it themselves. "The shop dogs chew on any glue that drips onto the floor. That's not a problem," Pat says, "since the glue is nontoxic and almost pure protein." Each batch yields only 300 ounces.

"Unmodified hide glue gels at around 100 degrees. For Old Brown Glue, I reduced the gel point to about 80 degrees. When the glue first arrives, it appears to be a gelatinous solid, if the room temperature is below eighty degrees. Put the bottle in sunlight, or in warm water for a few minutes, and it is ready to use. It will stay liquid all day long once it is warmed."

Even more surprising than their manufacturing method is the way they do business. They are about to start taking credit cards, but up until now, they have simply sent out the glue, and asked people to send the money once they received it. It's an Old World honor system that dovetails nicely with this Old World product.

None of this lends itself to mass production. "Old Brown Glue remains a specialty product for a small group of customers who appreciate the features it has to offer. For 30 years I tended the hot hide glue pot, maintaining the glue," Pat told me. "Old Brown Glue is a convenient, easy-to-use improvement. It's made my life much easier, so I wanted to share it with other woodworkers."

Current Trends in Conservation of Marquetry Surfaces

W. Patrick Edwards, Furniture Conservator in Private Practice

Written by: W. Patrick Edwards

ABSTRACT: There have been several exciting and innovative new methods developed during the past decade, both in America as well as in Europe, for conservation of marquetry surface decorations. Since marquetry is composed of extremely diverse hardwood elements, often mixed with exotic materials such as ivory, bone, tortoiseshell, mother-of-pearl, metals, straw or horn, the rich decoration is itself relatively unstable over time, and further compromised by the natural decay of the animal glue and gomme lac finish which were designed to protect it. In the majority of early work the solid wood groundwork which supports the marquetry is itself the major cause of the damage. Traditional conservation techniques have included repair of damaged surfaces "in situ" as well as completely removing the surface.

The process of removing, restoring and re-gluing a complete surface has been used in the past as an accepted process for restoring and conserving marquetry. The success of this method relies on relative differences in the glues which are used both on the top and back of the marquetry. Previously, this difference was often so slight, as, for example, between fish and hide or bone glues, that only the most experienced restorer would choose this approach. As synthetic glues were developed and incorporated in this process, the risk of further damage to the marquetry was reduced to a more acceptable level, and it has gained in popularity. During the past decade, in fact, this process has been the standard practice in several workshops, both private and public. Material will be presented which illustratesthe most successful current practices in this field.

A more exciting approach to the problem of stabilizing a marquetry surface involves the re-hydration of the existing animal glues and allows the original glue to remain in place under the veneer. Work in this method has been developed by the author as well as Nicolas Boucher of the Museum of Decorative Arts in Paris, and is being considered in other workshops as the least invasive and most secure approach to the problems inherent in marquetry. Essentially, this involves the proper preparation of the surface through hydration, and the addition of a urea product to the animal glue which allows it to remain liquid much longer than normal. Finally, a controlled heat source is applied which re-heats the existing glue, causing it to adhere to the marquetry as it did originally. This process will be discussed and illustrated in detail, as, in certain cases, it represents the optimum solution available today.

Introduction

Since the Renaissance, European furniture in general and, to a lesser degree, American furniture has been decorated with a wide variety of techniques which require specific conservation and restoration procedures. One of the most interesting form of surface decoration is that of marquetry, which evolved from Italy into Germany and France in the 17th century, and remained the primary form of decoration in the larger cities of Europe throughout the 18th century.

Historically, there exist five distinct procedures used to create a marquetry surface: Tarsia Certosina, Tarsia Geometrica, Tarsia a Toppo, Tarsia a Incastro, and The Classic Method, more commonly know as "piece by piece". In general, marquetry is an assembled surface composed of various elements, often of different materials, which is glued onto a solid wood ground with animal collagen glues. Since marquetry designs can contain a wide variety of exotic hardwood veneers, animal horn, bone or ivory, mother-of-pearl, tortoiseshell, straw and different materials such as brass, copper, and pewter, the problems associated with the conservation are complex. Even with proper conservation and ideal environmental parameters marquetry surfaces are subject to predictable damage over time which requires some form of restoration. For example, the most common form of damage occurs when the solid wood ground shrinks and cracks through the marquetry surface. There is also the natural dehydration of the animal protein glues, inherent movement of the diverse materials used in the surface relative to heat and moisture, and the problems associated with the finishes used to protect the surface.

Traditional techniques used to restore damaged marquetry surfaces evolved directly from the original 18th century fabrication procedures, and remained essentially unchanged until the 1970's. One of the best references available today which describes these methods is "Restauration du Mobilier" by Daniel Alcouffe, published in 1977, in which the removal of a damaged marquetry surface is accomplished "piece by piece" using a warm iron and spatula. Other, more drastic, methods were also employed by workshops which included complete removal of the groundwood by hand planes and carefully sawing between the marquetry and the ground to separate the surface. Problems associated with each of these restoration methods limited their applications.

The purpose of this paper is to present a brief history of the completely new restoration and conservation developments since 1970 which have increasingly been accepted in private practice and museum workshops. These new methods involve the removal and repair of marquetry as a complete surface with much less risk of damage, as well as dramatically improved methods of rehydration of the existing animal collagen glues that conserve the original marquetry surface "in situ".

Damp Method for Boulle Marquetry

Between 1972 and 1975 an innovative approach to marquetry restoration was developed at the Institutet for Matérialkunsap in Stockholm. Two conservators, Rune Hakansson and Anna Østrop, under the direction of Björm Hallström, were faced with the problems of restoring a severely damaged Boulle marquetry surface on the floor of the Swedish royal coach, fabricated in Paris in 1699. The tortoiseshell and brass surface was in poor condition, with damages and missing elements, areas that had been worn thin by feet, prior restorations, and a solid wood ground that required extensive repair. To make restoration possible, a decision was made to completely remove the Boulle Marquetry from the ground. However, instead of removing the surface "piece by piece", as was the usual method at that time, a new system was created that would allow the removal of the entire surface in one piece. The advantages of this method were immediately obvious, and, after the successful completion of the project, were included in Pierre Ramond's book, "Marquetry", published in French in 1977. Dr Ramond referred to this new method as the "damp method" and included the essential details, which became widely available when his book was translated into English by Jacqueline Derenne, Claire Emili and Brian Considine in 1989.

Once this concept of complete removal of a marquetry surface was introduced to the Paris workshops in 1975, it quickly became the preferred method, undergoing further refinement and application at the Louvre and Versailles, as well as the Museum of Decorative Arts in Paris and Lyon, and remains today an accepted procedure. The removal of the entire surface in one piece preserved the original positions of the elements of the marquetry relative to each other, and made it easier to replace damaged or missing parts. The relative ease with which marquetry surfaces could be removed allowed for further research into the procedures used in the original fabrication, improving our understanding of the different methods. With the complete removal of the surface, visual and ultra-violet examination of the ground revealed the original design, as well as subsequent restorations. The original colors found on the reverse of the marquetry surfaces also illustrated how much the visible surface had changed over time.

The "damp method" for removal of a Boulle marquetry surface used by Hakansson and Østrop is as follows: The tortoiseshell and brass surface is first covered with a wet cloth under a layer of plastic and left for a period of 12 hours. During this phase, the animal glue absorbs the water and softens, the tortoiseshell becomes soft and flexible and many elements begin to lift from the ground. At this point a layer of Japanese paper is attached to the entire surface using Paraloid B 72 glue, which has the property of sticking to a wet surface, and which can be removed with tolulene. In addition, this glue makes the paper transparent, allowing visual examination of the surface during removal.

Once the marquetry surface is held together by the paper on the face, it can then be gradually removed as a single piece by inserting thin spatulas, using syringes to inject alcohol under areas which are still attached. Once lifted completely, the animal glue residue is removed from the back side of the marquetry and the front side of the groundwork with a sponge and warm water, being sure to quickly dry the moisture from both surfaces as soon as they are clean.

At this point, an assembly board the same size as the marquetry is prepared. Several assembly boards are used during the process, allowing the transfer of the marquetry surface from the face side to the glue side as required. Brown Kraft paper is moistened slightly and attached to the edges of a flat panel of plywood slightly larger then the size of the marquetry. Once dry, the Kraft paper stretches tightly, and the marquetry surface is glued face side up on the paper with animal glue, and placed in a press.

After removal from the press, the Japanese paper and Paraloid B 72 glue are completely removed from the face with tolulene, leaving a clean surface. At this time all missing elements are replaced and damaged elements repaired. Once the marquetry is repaired, it is cut away from the first assembly board and glued face down onto a second assembly board, using animal glue. The allows the removal and cleaning of the Kraft paper and animal glue on the back side of the marquetry, which is done with cold water and gentle scraping. The water is quickly removed and the marquetry surface is ready to glue onto the original groundwork with animal glue. To remove the Kraft paper and glue from the face, the same method of cold water and scraping is used, and the surface is ready for finishing.

Dry Method for Wood Marquetry

The "damp method" was first developed for Boulle tortoiseshell and brass and when it was applied to wood marquetry surfaces, it produced the undesirable effect of expanding the wood elements in different directions, making it dangerous for complicated designs. Thus, several restorers (notably Pierre Costerg, Michel Tigréa, and Michel Jamet, among others) modified this system, creating what is known as the "dry method" for removing marquetry.

The dry method for removal of wood marquetry surfaces is described in detail in Pierre Ramond's book as follows: The finish is removed (fig.1), and a layer of cotton gauze or thin fabric is stretched across the surface and held at the edges with tape. Over this gauze is brushed a layer of neoprene glue which is allowed to dry about 20 minutes. This layer of gauze and glue acts to keep the marquetry surface intact during removal from the ground.

Using a hot iron or heat gun (fig.2) with temperatures between 50 and 60 degrees centigrade, the original animal glue is softened and a spatula is introduced between the ground and the marquetry surface. This action continues until the entire marquetry surface is removed in one piece, at which time the old glue is cleaned as usual from the back of the marquetry and the front of the ground. One reason neoprene glue is used is that it has a resistance to heat as high as 200 degrees centigrade.

As in the wet method described previously, the marquetry is now glued face side up onto an assembly board with animal glue. The neoprene glue and gauze is removed from the face, using trichlorethylene or pure alcohol. Repairs are made (fig.3) at this stage of the process. Once all missing and damaged elements are restored, the marquetry is glued face side down to a second assembly board. Finally, the paper and animal glue is removed from the counter face, using cold water and gentle scraping. The marquetry is ready to glue back to its original position (fig.4).

Dry Method for Boulle Marquetry

The success of the dry method of wood marquetry removal encouraged several restorers and conservators (such as Nicolas Boucher, Michel Jamet, and Yannick Chastang, among others) to develop a dry system for restoration of Boulle tortoiseshell surfaces. One distinct advantage of this proceedure is the preservation of the original finish and mastic. This is a recent development which promises to become the next standard in the treatment of Boulle surfaces. For this process it is not necessary to remove any of the existing finish. Standard transparent adhesive cellulose tape is firmly applied to the entire surface which is to be lifted. There is a wide variety of clear tapes on the market, and less expensive brands seem to have a better resistance to heat, so one is selected which resists deformation below 60 degrees centigrade. Virtually all clear tapes are damaged at higher temperatures, and this is an advantage since the tape will be damaged before the marquetry is affected by excess heat. The tape is left on the marquetry for 24, which provides better adhesion.

As in the dry method for wood marquetry, the surface is now lifted using a heat gun and spatula, working from the edges slowly and carefully. The lowest heat setting is used, as excessive heat makes the tortoiseshell brittle and causes the metal elements to expand. Once removal is accomplished, the old animal glue is cleaned and the marquetry is attached face side up to an assembly board with animal glue. This allows the removal of the tape from the surface, using white spirits as necessary. It is often the case with Boulle marquetry that the individual thickness of the materials used varies greatly due to the wide range of densities and hardness of tortoiseshell, ivory, horn, mother-of-pearl, brass, copper and other materials. One distinct advantage of the dry process is that the various elements of the existing marquetry can be restored to a flat face position, conserving any finish and engraving which exists, and eliminating the need for further sanding or scraping of the surface.

This procedure is called "repoussage" or embossing, and involves the pressing of the marquetry surface in a press, in such a way that the individual elements of the design are pushed into a flat face position.

As the marquetry surface is transferred face down onto the assembly board in the usual way, a softer material (such as card stock, many layers paper, thin carpet, or wall board) is placed on the glue side so that the pressure of the press acts to push all elements forward into position. A thin (4 mil) layer of plastic film is placed between the glue surface and the softer material to prevent any adhesion.

To allow for the uneven thickness of the restored marquetry panel when it is glued back into its original position, either a very thin (1 mil) layer of cork or a thick mastic/glue mixture is used under the marquetry. Another approach, suggested by Michel Jamet, is to conserve all existing animal glue residue on the counter face of the marquetry as well as the ground, and carefully replace the restored marquetry in its original position, adding a fresh layer of animal glue to insure proper adhesion.

Rehydration of Old Animal Glues

As a result of these recent developments in the restoration of damaged marquetry surfaces, a method has been developed which allowed the conservation "in situ" of less damaged marquetry surfaces.

In 1987 Nicolas Boucher, working with Roch Pyet, restored the surface of a marquetry commode belonging to the Musée Vouland in Avignon, France, using as experimental system that allowed the rehydration of the existing animal glues without removal of the marquetry. The results were encouraging and in 1991 the same method was used on a 17th century commode at the Museum of Decorative Arts in Paris, also with dramatic results.

After several more years of study and application, this procedure, called the "rehydration of old animal glues" was published in the November 1995 issue of the French magazine "L'Objet D'Art". This important article, by Nicolas Boucher, was called "decisive progress in the restoration of marquetry furniture".

Immediately after the publication of this article, a small group of restorers and conservators formed with the goal of perfecting this method. Leading this group was Nicolas Boucher, conservator at the Museum of Decorative Arts in Paris, and Marie-Christine Triboulot, professor of wood technology at ENSTIB (Ècole Nationale Supérieure des Technologies et Industries du Bois) located in Nancy, France. The name of the study group was chosen as ADEN and the purpose was to meet annually to suggest specific areas of research into the problems associated with marquetry restoration and conservation. International interest in this group has grown quickly, and at the most recent meeting in 1997 there were 95 members from 12 countries.

The results of this collaboration between the Museum of Decorative Arts in Paris and ENSTIB in Nancy, under the direction of the group ADEN has been the publication of new research into this field each of the past 3 years. To date the areas of interest include: methods of accelerating the aging of marquetry surfaces and animal glues; various additives used to modify the properties of animal collagen glues manufactured and used, and how their working properties can be restored after accelerated aging; and methods of restoring curved and shrunken ground wood to original dimensions using polyethelene glycol (PEG). These articles are listed in the references at the end of this article. The group ADEN can be accessed through the Internet, under the site "Musée des Arts Décoratifs" (http://www.ucad.fr).

One of the most interesting ideas has been the modification of the traditional animal collagen glues, specifically with small amounts of thiourea or urea, so that the glue remains liquid at room temperature. Using this modified glue has made the process of rehydration of old animal glues reliable and applicable to a wide range of projects.

As in the dry method used for Boulle surfaces, the rehydration process preserves the original finish. In fact, the presence of the existing finish is necessary to slow down the penetration of the water, allowing the moisture to penetrate deeply into all damaged areas. The surface is covered with wet cloth or tissue (fig.5) and allowed to sit briefly (from 20 minutes to a few hours) until lifting of the elements begins to occur. Constant observation is required as different marquetry surfaces respond at different rates.

Once deformation of the surface begins to occur (fig.6), the water is removed and animal collagen glue modified by the urea is brushed onto the surface, using syringes and spatulas to work the glue fully under any damaged areas of the veneer. Since the glue is modified and remains liquid at room temperature, there is no real pressure to work quickly, and care can be exercised to insure full application of glue, which may take up to 30 minutes.

Onto this marquetry surface coated with the modified glue is placed a thin (1.2 mm) sheet of lexan plastic (fig.7). On top of the plastic is placed two sheets of card stock with a temperature sensor in between.

On top of this layer is placed a special heating screen cut to shape with aluminum bars attached at each end. Once in the press this heating screen is connected to an electrical transformer which provides 400 amps at 4 volts for 20-45 minutes, until an operating temperature of 65 degrees centigrade is reached (fig.8). After 5 minutes at this temperature, the power is turned off and the marquetry is left in the press for an additional 72 hours. The marquetry must be kept under pressure, covered by a piece of cellophane, until completely dried, about a week.

This same procedure is used for Boulle marquetry with minor changes. Before the rehydration process is started, all the loose brass work is removed and reglued in place using fish glue. After this is done, the Boulle surface is rehydrated in the usual manner, placed in the press with the plastic caul and heating elements, and left for 72 hours.

Conclusion

During the past two decades there have been significant developments in the field of marquetry conservation and restoration. An important study group, ADEN, has been formed in France, and the collaborative efforts of a diverse group of professionals promises to make further advances in this field in the near future.

The purpose of this new research has been to create new methods which will allows less invasive and more controlled restoration of damaged marquetry surfaces as well as suggest new procedures for conservation of existing marquetry surfaces. As a direct result of this work, several different techniques are currently being used to aid in the restoration of marquetry surfaces and groundwork, as well as conservation methods which preserve the surface in its original position, including the glue, mastic, engraving and finish.

Acknowledgements

I have been fortunate to have had a great deal of assistance in the preparation of this paper, and I personally wish to thank those in my profession who have patiently answered many my questions during the years. These include Pierre Ramond, professor of marquetry at ecole Boulle, Paris, Michael Jamet, ebeniste in Paris, Nicolas Boucher and Pierre Costerg, conservators at Museum of Decorative Arts, Paris, Marie-Christine Triboulot, professor of wood technology at ENSTIB, Nancy, Yannick Chanstang, Conservator of furniture at the Wallace Collection, London, and Brian Considine, conservator at the J.Paul Getty Museum, Los Angeles.

About the Author

W. PATRICK EDWARDS, a native Californian, has been in private practice restoring and conserving pre-industrial furniture at his workshop in San Diego since 1969. A graduate of University of California, San Diego with a degree in physics and history, he has completed graduate studies in the fields of decorative arts and pre-industrial technology at Winterthur, Delaware, University of Massachusetts and ecole Boulle, Paris.

References

Alcouffe, Daniel. 1977. "Restoration du Mobilier." Office du Livre S.A. Fribourg, Suisse.

Boucher, Nicolas. 1995. "La réhydration des colles anciennes". L'Estampille L'Objet D'Art, Number 296.

Dos Santos, Judith and Mourey, Nicolas. 1997. "Amélioration des operations de recollage des placages et de redressements des panneaux-suports lors de la restoration de panneaux décoratifs." ENSTIB, 27 rue du Merle Blanc BP 1041, 99051 Epinal, France.

Garcet, Aurélie. 1996. "Etude des colles d'origine animale utilisées pour la restoration de marqueteries anciennes." ENSTIB, 27 rue du Merle Blanc BP 1041, 88051 Epinal, France.

Lavigne, Elodie and Monteau, Laurent. 1995. "Réhydration des colles animales lors de marqueteries anciennes." ENSTIB, 27 rue du Merle Blanc BP 1041, 88051 Epinal, France.

Ramond, Pierre. 1989. "Marquetry." Les Editions H. Vial, 8 Rue des Moines, 91410 Dourdan, France. Ramond, Pierre. 1996. "Chefs D'Oeuvre des Marqueteurs, Tome II." Les Editions H. Vial, 8 Rue des Moines, 91410 Dourdan, France.

Why Not Period Glue?

Journal of American Period Furniture– Volume: 2 – January 2002

Written by: W. Patrick Edwards

Period furniture makers today have a variety of traditional and modern adhesives available, and it is sometimes difficult to choose the perfect glue for the job. However, a careful examination of the working characteristics of these glues makes it evident that the best choice is traditional animal protein hydrolyzed collagen glue, known simply as "hot glue." This paper will present many of the facts and some of the fiction surrounding the most commonly used furniture glue in history.

HISTORY OF COLLAGEN GLUE

The use of animal collagen glue by man has been traced back over 8000 years, with the discovery of artifacts in caves near the Dead Sea which contain this material as an adhesive. Egyptians used collagen glues 4000 years ago. Historical records indicate large specialized glue factories were established in Europe at the end of the 17th century and America at the beginning of the 19th century. Hot glue was used by furniture makers exclusively until the start of the 20th century, when the development of synthetic glues began to change the market for this product. American furniture makers were some of the first to abandon the use of animal glues in favor of these new synthetic products, while in England and Europe the use of traditional glues continued until well after the second World War. Today, traditional animal glues are generally used in America by antique restorers, museum conservators and musical instrument makers and restorers.

ANIMAL AND VEGETABLE GLUE

There are two basic types of natural adhesives which are commonly used: animal and vegetable. A wide variety of vegetable glues are derived from starches, gums, cellulose, bitumen and natural rubber, and have specialized applications. Animal glues are derived from casein (a milk protein used in paint), blood albumen (used in plywood), and collagen (used in woodworking). All of these adhesive products are organic in nature and non toxic to humans.

CHEMICAL MAKEUP OF ANIMAL GLUE AND FISH GLUE

Animal glues are adhesives which are essentially high polymer proteins derived from hydrolyzed collagen. Footnote 1 These organic colloids are comprised of complex proteins found in animal hides, connective tissues and bones. This protein has two elements that define its characteristics: chondrin, which gives it adhesive strength, and gluten, which gives it gel strength (gelatin). Footnote 2

These glues are made using a rather simple process, which hasn't changed much over the ages. The raw material is first conditioned in a water solution with lime (calcium hydroxide). Then the pH value is adjusted by adding a dilute mineral acid and rinsed in water. Footnote 3 Then the process of cooking begins, and while the material is cooked the water/protein solution is extracted and filtered. The protein which is collected by the filters is dried and ground up as a final product. The resulting glue is then tested as to viscosity (fluidity) and gel strength (stiffness of gel formation), and graded on a scale from 50 to 512. Footnote 4 Lower grades dry slower and are more flexible and higher grades dry faster and harder. Glue chip glass is made using hide glue with a 135 gram strength, which allows the glue to actually tear off the surface of the glass as it sets. Woodworkers can choose between 164, 192, and 251 gram strengths, which have slightly different working characteristics. 192 gram strength is the most popular, and allows hammer veneering, "rubbed" joints, and adequate working times when the wood is preheated to 95 degrees.

Animal hide and bone glues set in a two part process which first begins by cooling from 145 degrees to room temperature, and then completely drying by evaporation during the next 12 to 24 hours. This allows the traditional woodworker to use this glue to his advantage, since hammer veneering and "rubbed" joints both require a glue with a rapid initial grab as it cools. In addition, the strong initial hold of these glues allows clamps to be removed and reused on another job while the first project dries overnight.

Protein glues form a chemical (molecular) bond as well as a mechanical bond. This means that fresh animal glue will reactivate and chemically bond to previous animal glue surfaces, as well as forming a strong mechanical bond with wood surfaces and other natural fibers. Hide glue sticks to surfaces by an electrochemical attraction, or "specific adhesion." It is one of the few truly reversible glues, which can be changed from liquid to solid and back again with the addition or subtraction of heat and moisture.

MODIFICATION OF ANIMAL GLUES

Hydrolyzed collagen glues are easily modified with a wide variety of additives. It can be formulated with all water soluble materials, such as sorbitol, glycols, sugars, syrups, metal salts and sulfonated oils. Efforts to make it more water proof have involved using 1% aluminum sulfate, alum (aluminum potassium sulfate), tannins and formaldehyde fumes. KCl (salt) and potash prevent brittleness and crazing over time. A 5% glycerin additive makes the glue flexible enough for canvas backing on tambours. Adding 5-10% or more by weight of urea extends the gel time, and also increases flexibility, producing a liquid hide glue at room temperature. All of these additives reduce the actual strength of the glue somewhat, but the final result is still an adhesive which is stronger than the wood surface. All protein glues contain some preservatives and foam control agents, which do not affect their working characteristics.

TRADITIONAL USE/TECHNIQUES

All pre industrial woodworking processes which required adhesives were designed to maximize the working characteristics of animal collagen glues. In the 18th century, shop stoves and open fires were used to preheat the wood surfaces, and during the 19th century furniture factories used specially heated rooms to allow more assembly time. The rapid setting action of the hot glue as it cooled was used to great advantage when hammer veneering, as it allowed the worker to lay veneers directly onto the final surface, cutting the joints as the work progressed. The simple technique of "rubbing" in glue blocks was used universally from cabinetmakers to clock makers, and allowed rapid and solid construction without nails, screws or clamps. Glue was modified with additives to allow the veneering of turned columns, which became a popular form of furniture decoration, and requires a reversible glue to accomplish. Shaped, heated cauls were constructed which allowed veneering crotch mahogany on simple OG moldings, and, by the mid 19th century, shaped cross grain veneer rosewood moldings appeared. When large areas of veneer had to be laid by hand, all hands in the shop would work together to hammer veneer it in place before it cooled. By the end of the century, large factories used heated, hydraulic presses to veneer large areas more easily, and with less labor.

When veneers were first sawn by specialized workers before 1800, the thickness was uneven and the surface rough. A toothing plane was created, which used a toothed iron, set at a scraping angle, to further reduce the veneers in thickness and make the surface more even to glue. At the same time these toothing planes were used to prepare the surface of the groundwork for the veneer, and, if the surface was flat with no defects, the evidence of even toothing marks was said to prove the "truth" of the work. This meant it was ready for the application of veneer, since the hot glue would eventually pull the veneer down into any depressions that might exist on the groundwork and show up in the finish later. As veneers were mechanically sawn after 1800, the use of the toothing plane was no longer required to prepare the veneers, but it continued in use for preparing the groundwork. As well as showing the "truth" of the work, it was used to increase the surface area of the glue joint, as well as provide a stronger mechanical bond with dense woods. After the introduction of sliced veneers, and the final transformation of the industrial revolution changed the way furniture was produced, the toothing plane and its use became obsolete.

There are a variety of traditional animal glue applications that continue to be used by modern craftsmen. Rabbit skin glue is necessary for laying gold leaf properly. Instrument makers and restorers have a wide variety of applications that depend on animal glues. For example, the fact that these glues can be colored and mixed with many components allows the addition of plaster of paris to glue for laying ivory keys. Marquetry workers add different colors to the glue to restore Boulle tortoise shell and make mastic. Fish glue has properties which make it perfect for exotic materials, such as tortoise shell, horn, leather, shark skin, cloth and metals. Footnote 5 Fish glue is a liquid glue with strong cold tack grip, and its use to glue brass, pewter and copper in Boulle marquetry is further strengthened when the metal is first rubbed with a fresh clove of garlic. Animal bone and hide glues are used individually and mixed together for all types of woodworking. Diluted glues are used for veneer sizing and flattening, as well as for sizing end grain and porous woods before sanding.

PROBLEMS WITH SYNTHETIC GLUES

There are several problems associated with synthetic glues which make them unattractive to furniture makers. These problems are often overlooked in favor of the generally perceived "easy to use" features that make a ready to use product handy around the shop. One of the most overlooked problems is the most obvious: lack of reversibility. Most furniture makers today do not consider the future problems synthetic glues create when it comes time to repair their creations. However, all furniture is subject to use and damage, and all furniture needs to be repairable if it is to survive the generations. Synthetic glues cure by a catalytic conversion from one chemical to another, and are irreversible. This means to take apart furniture made with synthetic glues requires destructive intervention and physical removal of all glue prior to repair.

Modern glues have a mechanical bond only, and require tight fitting joints and even clamping. They do not bond to themselves, and set up unevenly, remaining wet in one area of the joint while setting dry in another. These glues seal the wood surface and prevent stains and finishes from penetrating evenly. They are difficult to sand and remove from the surface when set. One of the greatest problems is the lack of resistance to sheer forces, which allow the wood to "creep" along the glue joint. This "creep" allows veneer joints to open up, and solid wood joints to move over time as wood movement occurs relative to humidity and temperature fluctuations, as well as wood shrinkage due to aging.

A more serious consideration when using synthetic glues is the toxic nature of the solvents which are included in their formulation. This represents, in some cases, immediate health concerns for the wood worker who might be constantly exposed to these solvents on a daily basis, and requires informed decision making as to what kinds of protection are required for their safe use.

WORKING WITH PROTEIN GLUE

Working with traditional animal glues is a simple process. The glue in dry form has a unlimited shelf life, stored in a dry container and kept away from heat. To prepare the glue for use, just add cold water and let sit overnight. It is not really important how much water you add, as long as it completely covers the glue. If you mix by weight, use 1.8 parts of water to 1 part of glue. If you mix by volume, just cover the dry glue with more water than glue. Once all the water has been absorbed, put the gelled glue into a double boiler and cook it on a low heat. A variety of materials can be used for the double boiler, such as copper, iron enamel, glass, stainless steel or aluminum. Use a stainless steel meat thermometer to monitor the glue temperature. Keep it constantly at 145 degrees F (60 degrees C), and add more water as needed to replace that lost to evaporation. A foil cover can be kept loosely over the top of the glue pot while cooking to reduce evaporation. A good quality round bristle brush is best as an applicator.

Traditional woodworkers used subjective tests to monitor the hot glue as it cooked. The odor should be pleasant if the glue is good, and smell bad if the glue is overheated or has been damaged by mold. The viscosity is measured by lifting the glue brush about a foot over the pot and letting the glue drip back down. It should be thin and liquid, with no lumps. You can test the strength by putting a small amount of hot glue between your finger and thumb, and rubbing together until it cools. The strength is then measured by pulling the finger and thumb apart several inches and looking at the protein strands which appear like spider webs. The longer the strands the stronger the glue. The color of protein glue when freshly cooked is light amber and it continues to darken as it is cooked. As long as it is not overheated it remains quite strong. If the glue temperature reaches 212 degrees it is ruined. If there is mold in the glue it can be very hard to see, but the glue will remain lumpy at operating temperature, and should be discarded. The glue pot and brushes must then be cleaned by boiling in water before a fresh batch is made.
Footnote 6

THE IMPORTANCE OF REVERSIBILITY

Animal protein glues are the only easily reversible glues available to woodworkers. All modern synthetic glues convert from one chemical form to another by using a catalyst. Once converted these synthetic glues are difficult or impossible to undo. Since protein glues react to heat and moisture, they can be easily converted from liquid to solid and back again, even after a century or more of time. Footnote 7 This is a primary reason why these glues have continuously been used in the restoration field. The existing original glue can be softened or cleaned with warm water, and the new application of hot hide glue will completely bond with the previous glue. For example, if you want to repair a rush seat chair, or any joint that can't be fully taken apart, you can drill a small hole into the joint, inject some warm water and then some hot glue using a syringe. Injecting alcohol or vinegar into an old glue joint will dry out the glue and make it more brittle, which can make it easier to take apart. Steam can also be applied to veneers which make it possible to remove and repair these surfaces.

Glue reversibility is essential when working with veneers, inlay and marquetry surfaces. It is necessary to be able to glue and unglue veneers while building a pattern on the surface. Protein glues allow easy repair and replacement of damaged veneers, by using heat and moisture. The glues are easy to clean off the surface, either with water or sanding, and are not affected by stains, solvents and finishes. Thinned animal glue mixed with sawdust makes a good mastic for marquetry.

A technique used in conservation of marquetry surfaces for several years demonstrates the advantage of traditional animal protein glue. Since antique veneer and marquetry surfaces in poor condition often have losses in finish, missing elements, loose areas, and cracks, the original animal glue dries out and looses its grip. This original layer of protein glue, often centuries old, can be rejuvenated by a simple process. The surface is first made wet with distilled water and paper towels. Plastic is placed over the towels to keep the water in. Constant observation is necessary to determine when the veneer, mastic and marquetry elements begin to lift. The surface is then immediately dried off and covered with a traditional animal glue modified by urea to make it stay liquid longer. This modified glue is worked into all the cracks and under any loose veneer. Once the glue is rubbed into the surface, it is covered with a thin sheet of Lexan plastic. A heated aluminum plate is then tightly pressed over the Lexan plastic and left in place 24 hours. Once the marquetry is removed from the press and the plastic is removed, the surface glue can be cleaned up with cold water. The new glue "re activates" the old glue, through the veneer and mastic, and restores the grip of the original glue without any damage to the surface. Footnote 8

CONCLUSION

In our efforts to research and understand the furniture of the past, we are obliged to recreate the process used by pre industrial cabinet makers. This implies a knowledge of the types of tools and methods used, as well as the materials selected and the reasoning behind their selection. One of the most important elements of all aspects of furniture making is the selection of adhesives. If we strive to recreate the period furniture as faithfully as possible in the world today, we must adopt the hot glue pot as a permanent fixture in the wood shop. It is every bit as important to the process as the selection of tools and wood.

REFERENCES AND SOURCES

Jay Utzig, Milligan and Higgins, www.Milligan1868.com
Craig Brougher 816-254-1693, www.player-care.com/hide_q-a.html
Craig Deller, www.deller.com
Frank Ford, www.frets.com
Hubbard harpsichord, www.hubharp.com
Eugene Thorndahl, Bjorn Industries, North Carolina, 704-364-1186, www.bjorn.net
W. Patrick Edwards, www.wpatrickedwards.com

FOOTNOTES:

1. Product Information pamphlet, Milligan & Higgins, (p.2)
"Animal glue is a protein derived from collagen...described as hydrolyzed collagen with the following formula:

C102 H149 O38 N31(COLLAGEN) + H2O------------->C102 H151 039 N31 (ANIMAL GLUE PROTEIN)

The approximate chemical composition of glue protein is as follows:
carbon, 51.3%; hydrogen, 6.4%; oxygen, 24.1%; nitrogen, 18.2%."

2. Bookbinding and the Conservation of Books, Stanford Internet site
"An adhesive consisting of organic colloids of a complex protein structure obtained from animal materials such as bones and hides in meat packing and tanning industries. Glue contains two groups of proteins: chondrin, which accounts for its adhesive strength, and gluten, which contributes jelling strength. Animal glue is a protein derived from the simple hydrolysis of collagen, which is the principle protein constituent of animal hide, connective tissue and bones."

3. Ibid
"Hide and bone glues make up the two major types of animal glue. Hide glue, which is by far the superior of the two, yields a fairly neutral pH in solution, usually in the range of 6.5 to 7.4, although wider variations are possible. Bone glue is generally acidic, having pH values of 5.8 to 6.3. A glue having a high acidity absorbs less water and tends to set more slowly than a glue having low acidity."

4. Product Information pamphlet, Milligan & Higgins, (p.6)
M&G "The gel (or jelly) strength of a glue is determined with the Bloom gelometer. It is the measure of the rigidity of a gel formed by a 12.5% glue solution at 10 degrees C ...The Bloom unit is a measure of the force (weight in grams) required to depress a 0.5 inch diameter plunger 4 mm into the surface of the gelled sample."

5. The Restoration of Old Wood Furniture Marquetry: Protein Glues, Their Analysis, Upgrading and Rehydration"
M.C. Triboulot, N. Boucher, et al, "Holzforschung und Holzverwertung" Nr. 3/1996, (p.63)
"Fish glue, contrary to the collagen based gelatins, appears to be particularly rich in the amino acid phenylalanine...proteins secreted by marine organisms to adhere to submerged rocks are particularly rich in phenolic groups to increase the resistance of the protein to the action of the water."

6. Test for quality of glue
Test: 1 oz glue (2 tablespoons) in 1 lb water (about 1 pint) let sit 12 hours. pour off surface water and weigh glue gel. 5 times the original weight or more is excellent. Solidity and coherency of the mass indicates the strength.

7. Product Information pamphlet, Milligan & Higgins, (p.2)
"Animal glues are soluble only in water. They are insoluble in oils, greases, alcohols, and other organic solvents. When placed in cold water, the glue particles absorb water and swell to form a spongy gel. When heated the particles dissolve to form a solution. When the solution is cooled the glue forms an elastic gel. This property is thermally reversible, and upon application of heat the gel liquefies. The gelling or melting point of an animal glue solution will vary from below room temperatures to over 120 degrees F, depending upon glue grade, concentration, and the presence of modifiers."

8. The Restoration of Old Wood Furniture marquetry:
Protein glues, their analysis, upgrading and rehydration"
M.C. Triboulot, N. Boucher, et al, "Holzforschung und Holzverwertung" Nr. 4/1996, (p.61)
"Protein glues solidify by dehydration, that is the loss of water to the environment (wood, air). However, a certain amount of water remains bound to the solid adhesive. Over many years or under drastic temperatures or lack of moisture in the air even this water is to a high extent lost and the glue becomes very brittle and weak and loses both adhesion and cohesion. Thus rehydration, the reestablishment of this water, becomes necessary to reverse the state of the glue to one of both good adhesion and good cohesion."
Note: For a more complete discussion of this procedure, see "Postprints of the Wooden Artifacts Group", American Institute for Conservation, San Diego conference, June 1997, paper presented by W. Patrick Edwards, "Current Trends in Conservation of Marquetry Surfaces", (p. 27)

Master Class:
Low-tech method for veneer columns

Written by: W. Patrick Edwards

During the early 19th century, American and European furniture makers commonly used veneered columns as both structural and decorative elements. When I began to restore antiques more than 30 years ago, I wondered how the columns were made. My research originally suggested that the turned columns were coated with hot animal glue, and then wrapped with veneer followed by wet jute webbing. When the column was rotated in front of a fire, the webbing would heat and shrink, effectively clamping the veneer in place. Unfortunately, my early efforts to duplicate this method met with such dramatic failures that I abandoned the idea for 20 years. It remained one of the mysteries of the trade, and I continued to search for answers to the problem.

I joined an international conservation group in France about 10 years ago, which was researching how protein glues could be modified with urea to make them liquid at room temperature. A lightbulb went off in my head. Finally, I was able to see how the early woodworkers could easily wrap veneers around columns using animal glue. They modified the glue!

I began my own research and eventually found a glue recipe and veneering method that work perfectly. Using this method I have veneered a wide variety of columns, ranging from a diameter of about 1 in. to as large as 12 in. The process is simple and effective, and it adds a dimension to my work that makes it dramatic and unique.

Use elastic bands and liquid hide glue
After making my glue discovery, my next attempt at gluing veneer to columns employed rubber inner tubes cut into strips, but they left marks and produced uneven pressure. Then I discovered Rep Bands at a medical-supply outlet. These 4-in.-wide elastic bands are sold in 50-yd. (part #A518015) and 6-yd. (part #A508805) rolls by Sammons Preston Rolyan (800-558-8633). I use the plum color, which is the strongest, and I cut lengths according to the job.

For liquid hide glue I use Old Brown Glue, which is a modified liquid animal glue that I make and sell (619-298-0864; www.wpatrickedwards.com). You also may experiment with Milligan and Higgins 192-gram glue (518-762-4638; www.milligan1868.com) by adding common urea (sold at chemical-supply outlets as a fertilizer) until it remains liquid at room temperature.

To thin it to a runny viscosity for this job, Old Brown Glue should be warmed above 80°F. A bath of hot tap water works fine. By the way, the bottled glue can be rewarmed many times without losing its effectiveness.

You also will need a single piece of veneer large enough to cover the column and a solid wood turning for the core. I prefer to use either poplar or Honduras mahogany for the core wood, depending on the quality of the project. Mahogany turns and takes glue slightly better than poplar.

To work properly, the veneer shouldn't have any cracks or flaws, and the grain should run lengthwise along the column. It is possi-ble to use burl veneer or veneer with seams, but that method is too complicated to cover in this short article.

Turn a column and secure it to the bench
Turn the solid-wood column using normal lathe tools, producing either a straight column or a straight-tapered column. It is extremely important that the column be smooth, because any bumps or dips will telegraph through the veneer.

I mount most of my columns in metal capitals, which means that I turn slightly smaller tenons at each end of the column.

To hold columns for veneering, I made long, oak benchdogs and drove a long wood screw into the tip of each one, with the point protruding on the face (left). I adjusted the clamping pressure to allow the work to be rotated with some resistance. Mount the column so that it is about 6 in. above the bench. Place newspaper under the column to protect the bench from drips.

Be sure to tape off the tenon areas to protect them from the mess.

Apply the veneer
First, wrap a paper template around the column to help you size the piece of veneer. Soak the veneer in a tray of hot water for about 15 minutes.

Remove the veneer from the hot-water bath and use paper towels to soak up the excess water on the surface. Next, apply a generous coating of glue (using a brush or a roller) to the glue face of the veneer. I also like to heat up the wood column, using a hot-air gun, before the glue.
Now wrap the veneer around the column, holding it in place with your hands.

Start wrapping a length of the elastic band evenly from one end of the column to the other, allowing the excess glue to squeeze out onto the newspaper.

The start is critical. It helps to have a raw wood area-the tenon areas work well for this-in which to anchor the wide elastic band before it contacts the veneer. Also, don't allow the veneer to twist as you wrap the rubber banding; it will be impossible to cut the seam on a spiral.

>Overlap each previous layer by almost half, so you get almost a double layer of banding along the column. If the wrapping doesn't go as planned, unwrap it and start again. You have about 10 minutes of open time before the glue starts to become gummy. The elastic band will not stick to the glue, and excess glue can be cleaned up with cold water.

Wait a day, and then fix the seam
Modified animal-protein glue is ideal for this application because it takes longer than synthetic glue to set up and cure. During the first 24 hours, the animal glue remains flexible enough to allow the wood to shrink as it loses moisture.

After 24 hours the veneer will have shrunk enough to allow you to cut an accurate seam. First, remount the column in the benchdogs and remove the elastic band. Dampen an abrasive pad (3M makes these) with cold water to remove as much excess glue as possible at this point, to avoid later sanding. Wipe the surface dry immediately.

Place a straightedge along the center of the seam overlap and hold it in place with either spring clamps or tape. Using a veneer saw, cut along the straightedge through both layers of veneer. If you use a knife, watch its tendency to follow the grain.

Now use a warm iron or another heat source to heat the area along this cut. To test the temperature of your iron, hold it near your face. If it feels dangerous-warmer than hot tea- lower its temperature. When the glue loosens along the seam, reach under the veneer to remove the small scrap of veneer that remains, then brush a bit more glue under the seam, reheat the joint, and use a veneer hammer to press the joint closed. Replace the elastic band and place the column aside for another 24 hours. When the column is completely dry, remove the elastic band, and wipe off any leftover glue residue with a damp, cool abrasive pad. Finally, sand the dry surface.