Project: A Study of Niello

sirelesar

Oct 14, 20236 min read

Updated: Dec 10, 2023

Figure 1:

The Cheddar Brooch: 9 centimeters in diameter and dates to between 800 and 900 C.E. © South West Heritage Trust

Introduction:

Niello is a material of a sulphurated metal or alloy of metals primarily used to create a black inlay for jewelry or other metal artwork. It is usually applied in the form of a ground, glass-like powder and with the aid of a flux. It is melted into a recess in the artwork, thus providing a contrasting background. The metal or metals used in the sulfide blend are usually silver, copper and lead. Prior to the first century A.D., a singular metal was used in the sulfide blend, usually matching the base material of the art1. This approach is often termed Roman niello. So, with a silver piece of art, the niello would consist of silver-sulfide (Ag2S) and for copper art, the niello would be copper-sulfide (Cu2S). Around the fifth century A.D., there is evidence1 to show that an alloy of metals began to be used, thus lowering the ultimate melting temperature of the niello, making the application easier. The alloy was usually of silver and copper, thus creating a niello of silver-sulfide and copper sulfide (Ag2S + Cu2S). Later in the eleventh century A.D., an alloy of Silver, Copper and Lead were used, thus lowering the melting temperature even further1. However, the introduction of lead introduced limitations onto which metals the niello could be applied.

All these alloy ratios of two or three metals have various melting temperatures and limitations of application.

Theophilus Presbyter (1070-1125)5 is a Latin pseudonymous author of the 12th Century who compiled documents describing artform processes including niello. Benvenuto Cellini (1500-1571)6 is an Italian goldsmith, sculptor and author who documented the processes for many artforms including niello. This project is a study of which of the medieval niello blends as stated Theophilus and Cellini has the lowest melting temperature with an aspect of identifying any application limitations of that blend. These melting temperatures will be compared to the eutectic blend, or blend that has the proven lowest melting temperature2. An accommodating goal of this study is to demonstrate the methodology of production and application from each of the sources to determine which ones produce a desirable finish.

Approach:

Several sources that reference the three-metal niello have varying ratios of the silver, copper and lead. Theophilus2 promotes a ratio of 4-2-1 (56%Ag-30%Cu-14%Pb). Cellini3 cites the ratio of 1-2-3 (17%Ag-33%Cu-50%Pb). The eutectic ratio has been shown to be around a 2.5-3.5-4 ratio (25%Ag-35%Cu-40%Pb). To demonstrate the application temperatures of each, all niello recipes stated are created using the process described by the author. They are then cast in a thin rod form by pouring the molten mixture in a steel channel. Once cooled to room temperature (72F), the rods are tested by snapping them to confirm that the silver did not come out of solution and granulate during the cooling process. Once the niello blend has been confirmed to be of uniform consistency, one ounce is placed in a crucible without flux and heated with a consistent heat to time how long it takes to liquify the sample, and at what ultimate temperature the blend melts. A series of these observations are averaged, and the average recorded in summary.

In these tests, the borax flux normally utilized when melting and setting the niello will not be applied to the test bases so that the ultimate melting temperature of each can be determined without a modifier. After the melting temperatures are determined, the various niello recipes are then applied to silver, copper, brass and bronze pieces with a flux to test the adherence and effect on the piece.

Materials:

For the most part, the consumed elements used are the same as identified in the texts. A few of the implements are slightly different in materials based on availability. The method of heating is different for convenience.

Element Use	My Materials	Period Materials
Metal Components	Pure Silver (99.9%) Pure Copper (99.9%) Lead	Pure Silver Pure Copper Lead
Sulfer	Pure Sulfer (99.5%)	Pure Sulfer
Flux	Borax	Borax
Stirring Implement	Graphite Rod	Charcoal
Crucible	Quartz	Clay
Heat Source	MAP Gas Torch	Charcoal Forge

Process for Utilizing these Materials:

For producing the Theophilus2 niello, the silver and copper are placed into a small crucible. The lead and “some” Sulfur are placed together separately. There is no indication in his publication of how much sulfur is used in this mix. I am using the equivalent weight of the metal constituents as a basis. The silver and copper are then melted and stirred with a stick of charcoal. I am using a rod of graphite instead of the charcoal. The lead/sulfur mix is then poured into the hot crucible and melted along with the silver and copper. A second crucible is prepared with the remaining sulfur. The first hot crucible contents are then poured into the second cold crucible with the sulfur. The second crucible is then heated until all the contents are melted. Afterwards the contents are poured into an iron ingot mold. I use a steel channel, which serves the same purpose. Theophilus states to hammer it prior to completely cooling, with continuous heat-hammer cycles until it is in a thin stick form. During the heat-hammer cycles, the niello cannot get too cool or else it will shatter, nor too hot to where it melts. It should be hammered while orange hot. He then has it crushed into a fine powder and placed into a quill for use, which is not being done for this experiment, as the stick of niello is the form that will be tested for melting temperature. Theophilus’ niello is normally applied to the work piece by applying a water that has been mixed with borax until turbid to the work piece, then applying the ground niello and heating the work piece until the niello flows into the work surface.

Cellini’s3 approach is slightly different. He directs to melt the silver and copper together in a blast furnace until well mixed, add the lead, quickly take it out of the heat, and stir it with a piece of charcoal. Then the lead will produce a bit of scum on the surface, which is to be removed with the charcoal. This will produce a clean and blended metal mix. Simultaneously, a fist-sized earthenware flask with a finger-wide opening is to be filled half-full of fine ground sulfur. The amount of sulfur can be approximated from this direction and appears to be about 6 ounces of sulfur for a 6-ounce amount of total metal. Then the melted metal is poured into the flask and the flask is corked with clay. Then the flask is shaken until the metal cools to produce a granulated niello. Take the granulated niello and put it back into the crucible with a grain of borax and re-cast it 2-3 times, breaking it up between castings. For this experiment the last casting will not be broken up so that the pencil of niello can be used in the test.

Results:

Theophilus Niello Notes:

This niello was difficult to make. It took significant heat to alloy the elements, and when adding the sulfur, it took additional heat to bring it back to a melted state. Once cast, it had a fine dark-grey color. It was clear when the niello had accepted sufficient sulfur, as the cast pencil was void of silver grains. The snap test of the pieces with the correct hue was easy and clean, revealing a uniform cross-sectional property of a dark, tight granular structure. The usage temperature and time to melt were significantly higher than the Cellini blend, which is understandable as the Cellini blend is much closer to the eutectic ratio.

Cellini Niello Notes:

This niello was relatively easy to make, even with the three cycles of heat-cool in refining the blend. Once refined it had similar color to the Theophilus blend, with similar snap characteristics and grain size. This blend and methodology of creation produce a product that is more forgiving to apply to artwork than the Theophilus blend as the base material is not required to be heated as hot, thus protecting fine detail that may overheat during the application. This niello does not adhere as well to bronze as the Theophilus blend does. It is very forgiving on gold and silver substrates.

Melting temperatures of the base materials4:

Silver: 961 ℃ Copper: 1084 ℃ Lead: 328 ℃

Experimental Results:

Niello Blend	Blend Ratio (Ag-Cu-Pb)	Ratio Percentages (Ag-Cu-Pb)	Melt Time	Melt Temp. deg C
Theophilus	4-2-1	56%-30%-14%	2m-5s	675
Cellini	1-2-3	17%-33%-40%	26s	560
Eutectic	2.5-3.5-4	25%-35%-40%	N/A	440-560

Conclusion:

The 15th Century Cellini niello is easier to use and construct than the 11th Century niello as described by Theophilus and is much closer to a eutectic blend. This makes sense as additional time had prepared the evolution of the craft. This experiment taught me how to make and apply niello. It also demonstrated to me that the Cellini approach is easier to make and easier to apply. The only increased limitation is it is not as applicable to use on bronze substrates. I will use this knowledge to further my use of niello in my artwork.

Bibliography:

1. La Niece, Susan – Niello: An Historical and Technical Survey Reprinted from The Antiquaries Journal 1983 Volume LXIII Part II

2. Hawthore, John G Theophilus – On Divers Arts The Foremost Medieval Treatise on Painting, Glassmaking and Metalwork Dover Publications, London 1963 ISBN 048623784

3. Ashbee, C.R. – The Treatises of Benvenuto Cellini on Gold Smithing and Sculpture Dover 1967 ISBN 1428604308

4. Avalone, Eugene A & Baumeister, Theodore – Mark’s Standard Handbook for Mechanical Engineers, Mc-Graw Hill 1987 ISBN: 007004127X

5. Wikipedia – Theophilus Presbeter - https://en.wikipedia.org/wiki/Theophilus_Presbyter

6. Wikipedia – Benvenuto Cellini - https://en.wikipedia.org/wiki/Benvenuto_Cellini