Science & Technology in the Service of Conservation. IIC Congress, Washington DC. September 1982. p.61-64.
Invented in 1850, the albumen printing process was the most widely used technique for making photographic prints during the nineteenth century. Albumen prints are among the least stable silver photographic materials. The primary reasons for this instability are the structure of the silver and gold image and the chemical characteristics of albumen itself. The following investigations were aimed at understanding the mechanisms and rate-controlling factors of image deterioration in albumen prints, so that storage conditions for maximum permanence could be identified. The phrase 'image deterioration' refers to changes in the silver and gold photographic image such as fading, detail loss, shifts in image hue, and staining of non-image areas. Other serious types of deterioration such as cracking of the albumen layer and mold growth may be associated with image deterioration but are not considered in this paper. Probably 90 to 95 % of extant albumen prints display significant image deterioration.
Experimental albumen prints were prepared by double-coating egg white containing 1.25% ammonium chloride by weight onto an all rag 60g/kg uncoated photographic paper stock. Between coatings the sheets were bathed in a 70% solution of 2-propanol to coagulate the first layer of albumen Sensitization was accomplished by treating for three minutes in a solution of 10% silver nitrate also containing 1% citric acid. The samples were exposed with a 1000 watt mercury-vapor lamp (NuArc GW 114) equipped with an integrator. Processing involved washing in water for 10 minutes, toning for 10 minutes in sodium borate (decahydrate) 10g, gold chloride 0.5g. water 500 ml, fixing for five minutes in each of two baths composed of sodium thiosulfate (anhydrous) 1 50g, sodium carbonate 2g, water 1 liter, and a final wash of 30 minutes duration. Sample prints consisted of a 13-step gray scale 25 x 135mm.
The prints thus obtained were cool brown in color, and unexposed areas were white. Like all printing-out papers, albumen prints have a long tonal scale (log exposure range). Figure 1 shows characteristic curves (plot of visual reflection density vs. relative log exposure) for the albumen samples and a contemporary grade 1 enlarging paper. Sensitivity varies greatly between the materials; the curves have been superimposed to facilitate comparison. Of importance in understanding deterioration in albumen prints is the fact that highlight contrast in albumen prints is very low, i.e., the slope of the characteristic curve in the log exposure range from 0 to 1.0 is much lower than that of the developed-out paper. Even when the negative has good highlight tonal separation, an albumen print made from it will have rather small density differences in this log exposure range, and this is a circumstance which contributes to excessive loss of highlight detail upon deterioration. It may also be seen from comparing the overall shapes of the curves in Figure 1 why it is difficult to obtain satisfactory prints on modern papers from nineteenth-century negatives originally intended for albumen printing.
One of the most notable results of these investigations of albumen print deterioration was the finding that processing flaws such as the use of exhausted fixer and insufficient washing after fixation were not responsible for the generally poor condition of albumen prints surviving today. The results of the incubation studies detailed below suggested that an inherent instability in the presence of moisture and air, rather than poor processing, was the reason for the widespread deterioration now seen.
Fig. 1 Characteristic curves for a contemporary grade 1 enlarging paper and albumen paper. Curves have been superimposed for ease of comparison.
Washing is important in the processing of any photographic print because it removes the thiosulfate used in fixing. Residual thiosulfate may decompose (especially in the presence of moisture) and release sulfur to attack the silver image. To test the effects of poor washing, a series of 80 sample albumen prints was exposed and processed as above, except that the amount of washing time was varied. Four levels of washing were employed, the shortest being a mere dipping of the samples into standing water for five seconds, while the others consisted of one, five and 30 minutes in running water. There was no visible difference in the dry prints after processing between the extremes of washing. The samples were then incubated at 50°C and various relative humidities. Table 1 presents average density changes after 60 days of incubation. Two parameters of deterioration are measured in Table 1: the extent of yellow staining in highlight areas and the amount of image fading. Highlight yellowing is presented in terms of blue filter density gain in a non-image area, while image fading is presented in terms of red filter density loss in a high density area.
Upon incubation, increased highlight yellowing and fading were quickly apparent in the poorly washed samples. As expected, harmful results from the residual thiosulfate were more pronounced with increasing relative humidity during incubation. However, the data indicated that the samples were effectively washed within a relatively short time, and further, that moisture induced severe deterioration quite independent of washing time. Repetition of the experiment yielded essentially similar results. The fact that albumen prints were washed fairly rapidly was published in the 1890s by Haddon and Grundy [1], and these results agree with that finding. The very thin paper support of albumen prints and the absence of a baryta layer are important factors. Our investigation also showed that prolonged washing of albumen prints leads to increased fading and yellowing upon incubation.
The consequences of the use of exhausted fixing baths in the processing of albumen prints are interesting and somewhat complex. Three separate experiments were performed, all similar in design, involving approximately 70 samples each. Albumen print samples were exposed and processed as above, except that they were fixed for five minutes in 250 ml of a fixing solution (compounded as in 2.1 above) which became progressively exhausted during the course of the experiment. Exhaustion of the fixing solution was accomplished by immersing in it sheets of sensitized but unexposed albumen paper. The level of fixer exhaustion is expressed as the number of 8 x 10 inch albumen prints fixed in the bath prior to the fixation of the experimental samples for incubation.
As the fixing solution became progressively more exhausted the image color of the samples was progressively altered. These differences in image color were immediately apparent after processing, and consisted of a gradual blackening (shift toward neutrality) of the image hue, together with a gain in density. Prints fixed in exhausted fixer also showed high levels of highlight yellowing and staining, visible immediately after processing and which became very much worse upon incubation.
The combination of extremely high stain levels in the highlights and a blackening of the image hue can be regarded as evidence of the use of exhausted fixer, and these phenomena are occasionally seen in historical prints. Inadequate washing has quite different manifestations in a deteriorated print, including more fading and highlight detail loss. The blackening associated with the use of exhausted fixer may be the result of a partial conversion of the image to silver sulfide or the result of an enlargement of the image grains from a deposition of silver on the image from silver-thiosulfate complexes.
* No. of 8 x 10s previously fixed
Table 2 presents the results of a fixer exhaustion experiment after 60 days incubation at 50°C. The data are averages of three duplicates. The effects of fixer exhaustion on the amount of image fading (red filter density loss in a high density area) are rather small. The amount of moisture present during incubation is much more of a determinant of density loss than is the degree of fixer exhaustion. The effect of fixer exhaustion on stain formation in highlight areas, however, is dramatic. Loss of highlight detail during high humidity incubation was worse for the fresh fixer samples than for the moderately exhausted fixer samples; at very high levels of exhaustion highlight detail loss during 95% RH incubation exceeded that of the fresh fixer samples.
The above experiments with processing variables in albumen prints demonstrated that (1) adequate washing of albumen prints was accomplished fairly rapidly and (2) the use of exhausted fixing solutions is associated with easily recognizable symptoms notably high levels of stain and image blackening. The majority of deteriorated historical albumen prints resemble those experimental samples which were properly processed, but subjected to high humidity incubation. They do not resemble those which received flawed processing.
The most important factor in the preservation of albumen prints is the control of relative humidity. From all incubation testing done in this laboratory, it appears that relative humidity should be maintained in the range of 30-40%. The brittleness of the albumen layer and paper support at very low relative humidity as well as the increased image deterioration associated with high humidity dictate that a moderately low RH be maintained. The experimental basis for this recommendation is given below, together with an account of the types of image deterioration caused in well-processed prints by moist incubation.
Extensive incubation studies on albumen prints have provided an overview of the way in which albumen prints deteriorate when subjected to air, heat and moisture. There are four principal aspects to this deterioration:
The primary rate-controlling factor for all of these forms of deterioration is the moisture content of the albumen layer and paper support. Temperature is also an important determinant of the rate of deterioration.
Although all the above-mentioned forms of deterioration generally occur simultaneously, highlight yellowing can be produced independent of the rest through the mechanism of the Maillard, or 'protein-sugar', reaction [2]. This reaction is very thermally sensitive [3] as well as moisture sensitive, and is favored under warm, moderately moist conditions. The extent of the yellowing generated by the Maillard reaction will vary according to the amount of glucose contained in the albumen coating.
Fig. 2 Highlight yellowing produced by various levels of relative humidity during incubation at 50°C.
The yellow staining of low density areas is a much more severe problem with albumen prints than in almost any other type of photographic print. There can be (and usually are) many contributing causes for the yellowing, and a full discussion of them is out of place here. A list of factors leading to yellowing would include the Maillard reaction, formation of colloidal silver, photo-oxidation of the albumen layer, and migration of products from secondary supports. Incubation of well-processed albumen prints at 50°C and at various humidity levels provided the relationship between ambient moisture and highlight yellowing shown in Figure 2. Highlight yellowing is expressed as blue filter density gain in a non-image area.
Fig. 3 Loss of highlight detail in an albumen print after 60 days of incubation at 50°C, 95% RH. Top print is an unincubated control.
Fig. 4 Contemporary enlarging paper after 60 days of incubation at 50°C, 95% RH. Top print is an unincubated control. Compare with detail loss in albumen print shown in Figure 3.
Highlight detail loss, overall image fading, and image color change in albumen prints are all related to chemical and microstructural changes in the silver and gold image. They are characteristic of gold-toned printing-out papers and differ significantly from the usual forms of deterioration encountered in black and white prints. The most significant change in terms of visual information is the loss of highlight detail. During moist incubation the low densities (highlights) of an albumen print are attacked most severely. The initial stages of deterioration in low density areas are often marked by a slight gain in density, but this soon gives way to density loss. Ultimately the low density areas fade to such an extent that they disappear completely and blend invisibly with the background stain. This may be seen in Figure 3, which shows an albumen print after 60 days incubation at 50°C, 95% RH, together with an unincubated control print. The detail loss in the albumen print sample can be compared with Figure 4, which shows a print on contemporary bromide paper (fiber-base) incubated under the same conditions, together with an unincubated control.
Fig. 5 Relationship between ambient relative humidity and image fading (density loss) in albumen prints after 60 days incubation at 50°C.
In addition to the difference in the amount of highlight detail loss, albumen prints typically lose much more density over the entire tonal scale during moist incubation than contemporary develop-out prints do. The relationship between the amount of density loss in high-density areas of albumen prints and the ambient moisture during incubation is shown in Figure 5. Fading is expressed as the amount of red filter density loss in an area of original density c. 1.45. Of particular importance is the dramatic increase in the amount of fading when the ambient RH exceeds 60%.
4.5 Image Color Changes
The changes in image hue which accompany the deterioration of albumen prints during moist incubation can be generally characterized as a shift from cool, purplish-brown hues toward warm, yellowish-brown colors. These changes depend somewhat on the original density of the area in question, the gold toning treatment used, the thickness of the albumen coating, and many other factors.
The incubation experiments described above demonstrate that albumen prints are not equipped to withstand 'ordinary' conditions in temperate climates, where summer heat and humidity are quite high. While well-processed and gold-treated filamentary (developed-out) silver images are resistant enough to air and moisture to be considered a stable medium, clearly albumen prints are not. The poor condition of the great majority of albumen prints surviving today testifies to their inherent instability.
Those albumen prints which have survived in good condition usually have been somehow shielded from air and moisture. There can be many mechanisms and casual arrangements which serve to establish a seal against air and moisture during the storage of photographs. Although room humidity may be high, moisture will penetrate closed containers and stacks of photographs, etc., only very slowly. The extent of deterioration will depend on the temperature, which actually reaches the print, and how long the adverse conditions are prolonged. Many other factors may also play a role in deterioration.
'Preservation' of albumen prints in institutional collections almost certainly leads to the disruption of the microclimate, favorable or otherwise, in which they had previously existed. Once part of a collection, there may be more opportunities for prints to equilibrate with the relative humidity of room air than had existed for most of the prints' previous existence. Few collections strictly maintain moderately low relative humidity and many have no humidity control whatever. There is a danger that many of the near-pristine albumen prints left at present will lose that distinction within a relatively short time, because the protective microclimate in which they have existed has been disrupted.
The rate of deterioration of albumen prints at a given set of conditions decreases over time, so that the maximum rate of fading is always observed in the initial stages of the process. Therefore prints which have remained in excellent condition are much more sensitive to their environment than those which have already deteriorated. From Figure 5 it may be seen that excessively high relative humidity is not required in order for serious deterioration to occur. Unless temperature and relative humidity are strictly controlled, it is unreasonable to assume that the albumen prints now in good condition will remain so for very long. The importance of humidity control in the preservation of albumen and similar gold-toned printing-out papers cannot be overstated.
For storage of albumen prints the following conditions are recommended:
Albumen is subject to more photochemical degradation than gelatin because the proteins of egg white are richer in the aromatic amino acids tryptophan, tyrosine and phenylalanine [4]. These are UV absorbers and tryptophan in particular has a highly colored oxidation product [5]. To minimize photochemical degradation of the albumen, prints on display should be illuminated with tungsten lamps. Little research has been done, and until more is known about the effects of light on albumen prints, it is best to apply the usual standard for vulnerable materials and aim for a level of illumination of 50 lux (5 footcandles) at the print surface. Daylight and fluorescent illumination should be avoided; if they are unavoidable, UV-absorbing shields on windows and lamp tubes and UV-filtering framing materials are recommended.
The question of appropriate storage enclosure materials for albumen prints has not been satisfactorily resolved. Because the Maillard or 'protein-sugar' reaction which leads to the yellowing of the albumen layer is accelerated by alkaline conditions [6], the use of alkaline-buffered paper storage enclosures is not recommended. Incubation of experimental albumen prints in contact with alkaline buffered papers results in increased yellowing under humid conditions [7]. The use of acidic enclosures is also to be avoided. An interim recommendation for paper enclosures for albumen prints is the use of a neutral, high alpha-cellulose paper without carbonate buffering. Uncoated polyester and cellulose triacetate also appear safe.
Proper care of albumen prints must be based on the awareness that they are extraordinarily sensitive to moisture. There is a striking contrast between albumen prints and modern black and white materials in that the latter are often poisoned by inadequate processing, yet are reasonably stable if well processed in the first place. On the other hand, albumen prints are usually adequately processed but are inherently far less stable. One important lesson of this investigation into albumen print deterioration is that photographic materials have very different characteristic behaviors. Major historical materials must be investigated individually if their preservation needs are to be properly understood.
This research was supported by grants from the National Museum Act (administered by the Smithsonian Institution) and the National Historical Publications and Records Commission.
1, Haddon, A., and Grundy, F.B., 'On the Amounts of Silver and Hypo Left in Albumenised Paper at Different Stages of Washing', British Journal of Photography, 40 (1893), 511-512.
2. Hedge, J., 'Dehydrated Foods—Chemistry of Browning Reaction in Model Systems', Agricultural and Food Chemistry, 1 (1953), 928-943.
3. Spanyar, P., 'The Importance of the Maillard Reaction in Food Chemistry', Period. Polytech. Chem. Eng., 20 (1976), 37 1-378.
4. Fevold, H.L., Advances in Protein Chemistry, 6(1951), 187-252.
5. Schnell, P.G., Vadehra, D.V., and Baker, R.C., 'The Cause of Discoloration of Hard Cooked Egg Rolls', Journal of Food Science, 34 (1969), 423-426.
6. Bergquist, D.H., 'Egg Dehydration' in Egg Science and Technology, AVI Publishing, Westport, 1977.
7. Reilly, J., 'The Role of the Maillard, or Protein-sugar Reaction in Highlight Yellowing of Albumen Photographic Prints' in Preprints of the 10th Annual Meeting of the American Institute for Conservation, Milwaukee, 1982.
