From Wiki

Back to Photographic Materials Chapter List
Back to PMG Photographic Processes

In progress: Seeking additional comments and images to develop this section

Photographic Materials Conservation Catalog
Photographic Processes
Date: Initiated April 2011
Contributors: Luisa Casella, Stephanie Watkins

The Photographic Materials Conservation Catalog is created and maintained by the Photographic Materials Group of the American Institute for Conservation for the convenience of the membership. The treatments, methods, or techniques described herein are provided for informational purposes. The reader assumes responsibility for any application results or interpretation of information.

Carbon Prints

Historical Facts

Eaxmples of Carbon Prints with different color pigments

Carbon printing is a dichromated colloid process invented by Louis Alphonse Poitevin in 1855. At this point the final image was not done by transfer (being more similar to a gum process). Halftones were hard to achieve: Only the surface area of the gelatin layer would harden in the halftones, thereby leaving them vulnerable to wash away during processing. The process was initially called “direct” carbon; It was latter suggested transferring the image to another support by Fargier or Joseph Wilson Swan (different sources disagree; although the French photographic patent website lists patent 46719 of 1860 by M. Fargier as being the first to refer to the transfer to other supports). The process is based in a colloid and a chromate. Many substances can be used as colloid: milk, glue, albumen, gelatin, etc., although gelatin is generally used for carbon prints. The colloid encapsulates the pigment that renders the image. Carbon prints can therefore appear in a number of colors. It was the use of this process in tri-color carbon prints that allowed for the first photographic images in color by Louis Ducos Du Hauron, which he patented in 1868.
The main period of use of this process was 1860 - 1940.

Identification Characteristics

Carbon print structure

Image layer: Composed of gelatin and pigment.
Color: Varies according with the pigment used.
Support: Commonly paper, type can vary wildly.
Material description: Carbon prints can consist of two or three layers (paper support, baryta, and image). Carbon prints (also referred to by commercial names such as Autotype or Fresson print) and woodburytypes both consist of an image formed of ink and gelatin. It is very difficult to distinguish them unless, for example, processing artifacts are visible (in carbon prints sometimes it is possible to see bubbles due to deficient contact in the transfer step). Another indicator of a woodburytype may be fluorescence of a shellac sub-coating (although the sub-coating could also be gelatin). Woodburytypes can also be mistakenly identified as being albumen when in very good condition. Image quality of carbon prints and woodburytypes is very high in terms of tonal range and detail rendering. They are commonly identified by a relief of the image area and little to no sign of fading present, depending on the stability of the used pigments. Carbon or lamp black was extensively used, but so were less stable pigments such as carmine (cochineal red). Image layers can be of single color or multiple colors. Although having high chemical stability, carbon prints have low mechanical resistance, being susceptible to physical damage (marring, abrasion, scratching, and in extreme cases, flaking or chipping) .

Process Chemistry
The process is based on the photosensitivity of chromium salts combined with a colloid (gelatin) which, once cross-linked and hardened by the photoreduced chromium salt, becomes insoluble in water in the exposed areas. The pigment that forms the image gets encapsulated in the colloid. Potassium dichromate (K2Cr2O7), also known as bichromate, is photo-reduced from a chromium (VI) compound to chromium (III). It is generally felt that the gelatin/albumen/gum Arabic serves as the electron source (just as the halide half of silver halide served as an electron source for the photoreduction of silver halide to form the latent image).
The chromium (III) cross-links the gelatin via a carboxylic acid group:
gelatin - - - - COO- - - - CrOH (2+) - - - - OOC - - - - - gelatin
The carboxylic acid group actually looks like an oxygen double-bonded to a carbon with an -OH group singly bonded to the same carbon. Since carbon can take 4 bonds maximum and double bond occupies 2 plus the single (total of 3) we have room for one more single bond to join the carboxylic acid group to the rest of the molecule (in this case, the rest of the gelatin). Note that not only does the chromium harden the gelatin making it swell less readily and dissolve less easily in hot water, but it also renders the colloid more lipophilic. Thus, one could wash off the more soluble, unhardened (unexposed) gelatin with hot water or one could apply an oil-rich ink that would preferentially stick to the hardened areas (the lipophilic or fat loving) areas (allowing to use the image like an ink stamp in which the light-hardened areas carry the ink so with black ink, the hardened areas would print black and everything else would be white (ink wouldn't stick to those areas). This gives us a basis from which to print directly (carbon) or indirectly using ink or even more remotely, woodburytype, where a plate is produced from which the actual prints are made.
SPSE notes that the carbon process is not the usual 14 H+, 6 electron reduction of Cr2O7(2-) (dichromate) to 2Cr(3+):
C2O7(2-) + 14 H(+) + 6e -> 2Cr(3+) + 7H2O
The carbon process is less pH sensitive since only 6H+s are involved and it's only a 3 electron process:
Cr2O7 (2-) + 6H(+) + 3e -> Cr(3+) + CrO4(2-) + 3H2O

Carbon prints were made using what was referred to as carbon or pigment “tissue” – the carrier for the gelatin and pigment that will form the image. The term “tissue” can be confusing because it seems to refer to a thin paper when carbon tissue is actually thick (the gelatin layer will harden if kept in room environment. The tissue should be kept in a closed plastic bag or frozen). A watercolor pigment (originally carbon black, which gave the name of the process) is dispersed in a 10% aqueous solution of warm gelatin, and the resulting emulsion is coated on a paper. This can be done in two ways, in both of which the paper has to be perfectly leveled, on a slate. Casting is the more forgiving method, done by creating a “dam” on the paper, then pouring the gelatin, spreading it with a comb. Another method is applying the mixture on the edge of the paper and drawing it with a glass rod or foam brush.
The amount of gelatin and pigment affect the final results in the print:

  • the more gelatin in the tissue, the higher the image relief
  • the more pigment, the more depth in D-max and higher the contrast.

Modern, contemporary practitioners often bleach out silver from commercially made silver gelatin developing out printing paper for use with carbon printing, thereby avoiding this finicky step.
After drying, the pigment paper can be sensitized to light by soaking it for a few minutes in a solution of potassium or ammonium dichromate (the first is more common probably because ammonium dichromate is very unstable). The percentage and time in the sensitizing bath have an effect in the final result in the print:

  • the higher the amount of dichromate or the longer the sensitizing bath, the less contrast you get (flat image).
  • the less amount of dichromate or the shorter the time, the more contrasted the print.

In the historic literature, the percentage is usually 3% of dichromate.
After drying in the dark (only when dry the tissue is sensitive to light), the sensitized tissue can be contact printed. The negative should be masked around to get a safe white edge. Gelatin can peel easily at the edges of the support, so providing a white border helps protect the image. A strip test should be done to get the exposure time.
With light, the gelatin will harden from the top of the surface down. After exposure, the pigment tissue as well as the final support (which can be any high quality paper that has received a coating of hardened gelatin such as fixed out gelatin DOP) are soaked in cold water briefly (1 or 2 minutes). Because the receiving paper has hardened gelatin, the time of soaking has to be longer than for the tissue that has very soft gelatin. During this soaking procedure there must be great care in avoiding air bubbles that can be in contact with the tissue or final support surfaces, and will prevent the image layer to adhere leaving artifacts (this type of artifacts can, incidentally, be a distinction feature between carbon prints and woodburytypes – the latter will never present areas of no image due to bubbles). For this reason, this first water bath should be distilled water or tap water that was left to set overnight.
The papers are floated together, gelatin facing gelatin. They are taken out together and then squeegeed to remove the excess water. They are placed under weight for 10 minutes to cause transfer of the image layer. The papers should be in contact with a hard, smooth surface for optimum transfer. After the 10 minutes, the tissue/final paper combination is soaked in warm tap water, under agitation. The edges of the tissue will lift, as the unhardened gelatin dissolves and pigment will come out while gently massaging the edges with fingertip pressure.
The backing of the pigment tissue can eventually be removed which should be done very carefully in one even movement. Pulling off the backing in a discontinuous motion can leave streaks. The final support should continue to be washed in progressively cooler water until the image looks satisfactory. It takes some time for the image to be revealed and when this happens continue washing with cold water. The temperature of the water should be lowered slowly to prevent gelatine reticulation. A fine, smooth brush can be used to highlight certain areas (this should be done with extreme care for the entire image material can be removed accidentally).
The final washing has to be long enough so no soluble pigment remains in the print. A yellow dichromate stain may remain around the image where the edges of the tissue were. To remove the dichromate stain the print has to be immersed in a 5% solution of Potassium Alum until it clears.
The print is hung to dry to avoid any residual carbon creating disfiguring drying rings within the print surface if it is dried horizontally on a screen. A strong relief will be evident when wet, which will reduce as the image layer dries.

Historically, carbon prints were made on very thin papers, explaining why the edges were cut and the prints were usually mounted onto a support board.

  • Mechanical deterioration
    • The gelatin and paper do not react the same with humidity and temperature changes; when the atmosphere is dry, the gelatine retracts and the papers curl.
    • Cracks in the gelatin are probably linked to successive changes in the temperature and relative humidity. Cracking can also be linked to the fact that dichromate continues to harden gelatin. When the atmosphere dries, the more rigid gelatin layer cracks while the paper support contracts. Cracks often occur in high density areas and along edges. Humidity enters inside the prints by the edges. Gelatin is also thickest within the darkest pigment deposited areas.
    • Tears and folds are usually caused by handling manipulation given the thinness of the paper. Even if supported on a matboard, prints can buckle, also causing tears or folds if any pressure is exerted.
    • Carbon prints are sensitive to scratches, abrasions and skinning.
    • Mold spores (fungal, conidia) can develop because of the presence of gelatin.
  • Chemical deterioration
    • The image pigments in the carbon processes are commonly considered very stable although as different colorants were used, not all are equally stable (Nadeau, 1982, pp.177 – 183). It is believed that the colloid structure does help increase lightfastness of the pigment chosen.

The most common deterioration noticed are cracking of the binder in the high density areas, as well as planar distortion due to differential dimensional changes of the gelatin and support. The hygroscopicity, uneven thickness and potential brittleness of the binder layer influences treatment. There is little success in treating carbon prints with binder problems.
The image pigments in these processes are commonly very stable, although different colorants were used by different practitioners and not all equally stable (Nadeau, 1982, pp.177 – 183).

Concerns and limitations in treatment:
Carbon prints commonly require treatments involving flattening, repairing tears or image layer consolidation. Prints are frequently mounted onto a secondary support that reduces planar distortion.

  • Issues with flattening:
    • Heat – high enough temperature to relax the gelatin can cause it to “melt” and darken, as well as break-down its structure causing it to be brittle
    • Humidity – gelatin will swell during humidification and can crack during drying
    • Pressure – excessive pressure may alter the surface characteristics of the gelatin
  • Issues with flattening secondary support:
    • Humidifying and drying – The response of the secondary mount to moisture, pressure as well as its drying rate is different from that of the print, thereby imposing many restrictions to treatment options. Be careful of the amount and method of introducing and removing moisture to and from the paper support and gelatin layers.
  • Issues with consolidation:
    • Consolidant – it is necessary to swell and relax the gelatin so it can go back to its original place, and maintain this after drying. Successful methods include: localized humidification alone or followed by application of gelatin or methylcellulose, or mixture of gelatin, starch and glycerin; application of layer of gelatin; application of synthetic adhesive (Beva D8) with heat; cellulose acetate (Edmondson, Mellon Workshop, 2004). Local application of gelatin can be effective (see Dana Hemingway’s treatment) but it can bulk on top of the area to consolidate or stay on top of the image layer (failing to consolidate the area).
    • Application Method – the areas to consolidate are commonly very fine cracks. It is difficult for consolidants to penetrate these and reattach the image layer to the base. As expected, larger molecular or long chain adhesives tend to sit on the surface and not penetrate through the cracks to the support. Methods to introduce adhesives include brushing, spraying, or vaporizing through use of a nebulizer or ultrasonic machine.

Housing and Storage Considerations
Carbon prints should be stored horizontally. Un-mounted carbon prints should be matted or placed in rigid folders to prevent physical damage from handling. French inlay mounting prevents curling of the carbon print and allows the print to be easily handled. Carbon prints pasted onto a matboard should be either mounted in passé-partout or protected by a paper envelope. A polyester envelope can be also used if the picture needs to be manipulated very often and not displayed or stored in an environment with large relative humidity fluctuations or high RH.

Maintaining relative humidity between 30 to 50% and a temperature around 67°F (18°C) is recommended (B. Lavedrine, Les Collections Photographiques, Guide de Conservation Preventive, Arsag, Paris, 2000 p.111). It is also necessary to avoid large relative humidity and temperature fluctuations. Environmental fluctuations leads to the gelatin cracking.

Exhibition and lighting:
According to Wagner et. al. (2009), carbon prints may be in the category of Very Light Sensitive, and should have a "total exposure per year (unless otherwise noted) - 5,000 ft-c hours (50,000 lux hours); e.g. 3 ft-c for 5 months at 10 hours per day or 5 ft-c for 3 months at 10 hours per day. Rest for 3 years minimum between display cycles." This is the case of carbon prints "with non-earth, colored pigments (not carbon black or earth pigments) or on poor-quality papers." Or, they can be in the category of Less Light Sensitive, in which case they should receive a "total exposure per year - 30,000+ ft-c hours (300,000 lux hours); e.g. 10 ft-c for 9 months at 10 hours per day. Rest for 1 year minimum between display cycles." This is the case of carbon prints with the following characteristics: "Support paper in good condition and light stable. Colorant known to be carbon or other earth pigment."

Emergency Recovery
Emergency due to water damage may affect unhardened gelatin to soften. Avoid touching the surface of the image. If damage was done by soiled water, rinse in clean water and air dry. Paper supports may be weakened by water immersion and can tear easily. Mold (fungi, conidia) can grow on damp paper and gelatin.


  • Wagner, Sarah S., Constance McCabe, and Barbara Lemmen.1990 (updated 2009). "Guidelines for Exhibition Light Levels for Photographic Materials". Original version compiled in 1990 as in-house guidelines for US Library of Congress by Sarah S. Wagner. Published as Guidelines for Exhibition Light Levels for Photographic Materials Topics in Photographic Preservation, Vol 9, 2001 and by the National Park Service website:

Process and Historic Material

Contemporary Practice

Back to Photographic Materials Chapter List
Back to PMG Photographic Processes
Back to previous page