Michael P. Dosch
Rev. Nov 2000
It is well-known that photographic chemicals change their potency over time. The decline in activity is accelerated when these chemicals are mixed in water solution (1, 2). An experiment was designed to describe the decrease in activity of phenidone powder, phenidone in aqueous solution, and a phenidone-ascorbic acid film developer, when these are kept for various lengths of time. A second part tested various scratch-mixed metol-hydroquinone (also called MQ or D76-type) developers. Exhaustion was arbitrarily defined as a 0.2 log density unit decrease in density of the most-dense section of a step tablet, as compared to the density of the same step when developed at the initial time interval. The main outcome was measured by densitometer.
Phenidone developers: Mixing one phenidone developer ("PC") from dry phenidone powder produced inconsistent results. A stock solution of phenidone was exhausted within 4 weeks. Storage life of more than 6 months was obtained through a different method of mixing PC, from PX, and from XtolTM (Eastman Kodak, Rochester NY). The last three named were also more consistent than the first two methods.
D76-type developers: The longest lasting was a metol-Kodalk variant without hydroquinone (called DK76b here) at 22 weeks. The rest were all exhausted in 6-12 weeks, except D76TM (Eastman Kodak, Rochester NY) which lasted more than 6 months and was the most consistent performer. However, with the commercial D76, attempts made to exclude air in an accordian-type plastic bottle reduced both the storage life and the consistency considerably.
It was concluded that the commercial products D76 and Xtol are consistent performers whose storage life is underestimated by their product literature (at least, when mixed with distilled water and stored under the conditions of this experiment). For scratch mixers, DK76b and two varieties of a phenidone-ascorbic acid developer warrant consideration since their performance is consistent and they last nearly 6 months before exhaustion.
Phenidone (1-phenyl-3-pyrazolidone), phenidone A, film development, phenidone-ascorbic acid film developer, XtolTM, Mytol, photographic solutions-storage, black and white photography, D76, MQ, metol, hydroquinone.
Phenidone (1-phenyl-3-pyrazolidone, CAS No. 92-43-3), also called phenidone A, is a developing agent useful for black and white film or paper development. Its formula weight is 162.2 (3). It is said to lose its potency after about six months (2), when kept as the dry chemical. This claim is difficult to evaluate, since no studies are cited in the sources. Perhaps even more importantly, storage conditions will vary from darkroom to darkroom. Unless these are specified, it is difficult or impossible to apply safe-storage figures to one's own photographic processes. DimezoneTM (Phenidone B; 4-hydroxymethyl-4-methyl-1-phenyl-3-pyrazolidone) is thought to resist oxidation in solution better than phenidone A (4), but it is not as readily available. Phenidone A can be substituted for phenidone B on a weight-for-weight basis (4).
Like most developing agents, phenidone should lose potency at a faster rate due to oxidation once it is mixed in water solution, particularly if stored in partially-filled bottles. One suggestion for keeping phenidone, which also aids in measuring small quantities, is to mix a stock solution (Table 1). This stock solution was chosen for testing. It has been modified from the formula given in the source by the substitution of sodium metabisulfite for sodium bisulfite, and a change in the total amount of phenidone from 2 to 1.5 gm/L, so that each 100 mL contains 0.15 gm of phenidone.. No claims about the keeping properties of this solution were stated in the original source (5). "The preservative will have no appreciable effect on any developing formula" (4).
|Table 1 Phenidone stock solution|
|Distilled water (65.5 ºC /150 ºF)||750 mL|
|Sodium metabisulfite||6 gm|
|Water to make||1.0 L|
Phenidone has been combined with ascorbic acid derivatives into a new developer, XtolTM. It is thought to produce better image quality (less granularity, increased sharpness) and faster film speed, store longer once mixed, be more environmentally friendly, and present less risk of allergic reactions than developers based on metol or hydroquinone (2, 6). It also mixes at room temperature. Because of these advantages, it was chosen for test, as well as a a modified version, suitable for compounding by anyone with access to the raw materials ("PC" developer, see Table 2). It was modified from the Paul Lewis Mytol formula (see 2, 6) by the substitution of ascorbic acid (formula weight 176.1) for sodium ascorbate (formula weight 198.1). Scales are generally only accurate to the nearest 0.1 gm. To prepare phenidone 0.15 gm, dissolve 0.3 gm Phenidone in 200 mL distilled water (65.5 ºC /150 ºF), then use 100 mL.
|Table 2 Phenidone-Ascorbic acid developer ("PC")|
|Distilled water||750 mL|
|(27 ºC / 80 ºF)|
|Sodium sulfite anhyd.||60 gm|
|Sodium metaborate||4 gm|
|Ascorbic acid||10.7 gm|
|Sodium metabisulfite||3 gm|
|Distilled water to make||1.0 L|
Another phenidone-ascorbic acid developer ("PX") was compounded for testing. This was a formula modified by the author from the patent documents for XtolTM (6) by omitting the chelating agent EDTA, the use of phenidone A instead of B, and the use of ascorbic acid instead of sodium isoascorbate (Table 3).
|Table 3 Phenidone-ascorbic acid developer ("PX")|
|Distilled water (65.5 ºC /150 ºF)||200 mL|
|Sodium sulfite anhyd.||10 gm|
|Sodium metaborate||4 gm|
|Distilled water (27 ºC / 80 ºF) to make||750 mL|
|Sodium sulfite anhyd.||75 gm|
|Sodium metabisulfite||3.5 gm|
|Ascorbic acid||10.7 gm|
|Distilled water to make||1.0 L|
Once the methods were established, it was decided to study several MQ developers. These are listed in Table 4. D76, as produced commercially by Eastman Kodak, was chosen as a control. After all, the reason for compounding chemical solutions is to ensure optimum freshness and consistent activity. If I can't do better than purchased D76, it is time to re-examine my assumptions. The published formula for D76 is given (5). The commercial product D76 probably contains proprietary additives in addition to the ingredients shown. D76H is advocated by Anchell and Troop (2) as consistent. As compared to D76, the formula omits hydroquinone, and increases metol to 2.5 gm. My customary developer is DK76a, which is D76 with the substitution of sodium metaborate (Kodalk) for borax. I also created and tested DK76b, which is similar to D76H, but with a similar substitution of sodium metaborate for borax. Finally, I studied what I call MC here. This formula is a metol-ascorbic acid developer (7) which is thought to promote surface development, which is associated with higher definition and less graininess. Also, I was interested in it as an example of an ascorbic acid developer.
|Table 4 Metol-hydroquinone|
|(D76- or MQ-type) developers|
|Water mL 125 °F/52 °C||750||750||750||750||750|
|Sodium sulfite anhyd. gm||100||100||100||100|
|Ascorbic acid gm||10|
|Borax decadhydrate gm||2||2|
|Sodium metaborate gm||2||2||35|
|Potassium bromide gm||1|
|Water to make||1.0 L||1.0 L||1.0 L||1.0 L||1.0 L|
The purpose of the first part of the study was to describe the loss of activity over time of:
A secondary purpose of the study was to use the various developers, become familiar with them, and create a family of curves that would provide useful information to anyone desiring to compound them. The importance of this study is not only economic ("How can I get solutions to last for the maximum amount of time and save money?") but the removal of another barrier to consistency in black and white processing ("How can I be sure I am using chemicals and solutions as long as they are effective, and no longer?").
Hypotheses It was expected that dry phenidone powder would lose activity after six months, and phenidone stock solution, PC developer, PX developer, or XtolTM after one to two months. It was expected that most of the MQ deveopers would lose activity after 1-2 months, and that D76H would last longer.
Assumptions It was assumed that loss of activity could be measured as the difference between the maximum density recorded from a step tablet contact printed on film initially, compared to the density of the same step on a similar piece of film, exposed and developed under the same conditions, but at a later time, with "aged" chemicals or solutions. It was assumed that if these test procedures, time intervals, and storage conditions were specified, other photographers could benefit from the information.
A Gossen LunaPro Digital light meter and Beseler Audible/Repeating Enlarger Timer (Catalog No. 8177) were used to ensure consistency of film exposure. The same step tablet was used throughout. An XRite 810 densitometer was calibrated before each use, and used to read all films. An Eastman Kodak Process Thermometer Type III was used to establish a temperature of 20 degrees C for all developing solutions, and for the tempering water bath used to keep solution temperatures constant during the course of development, stop bath, and fixer. A GraLab 300 timer was used to time processing steps. An AccuLab V-400 electronic scale, readable to 0.1 gm, was used to weigh chemicals. It was allowed to warm up for 30 minutes, and calibrated against a supplied 200 gm weight, before each use. The materials used are specified in Table 5 ("Formulary" is Photographers Formulary, Condon Montana, 800-922-5255, http://www.photoformulary.com).
|Table 5 Stock numbers|
|Sodium sulfite anhyd.||Formulary||10-1340|
|Acetic acid 28%||Kodak||146 2829|
|Universal Rapid Fixer||Ilford||741 850|
|PhotoFlo 200||Kodak||146 4502|
Only distilled water was used to prepare developers and the phenidone stock solution. All developers were diluted to working strength with distilled water. Whether diluted or not, all developers were used one-shot and then discarded. Tap water was used in all other solutions unless specified below.
Tmax 400TM 4 x 5 inch sheet film from two 100-sheet boxes (expiration March 2001) was used for all tests. Over the course of the experiment it was stored in the freezer, in its original container or a 25 sheet sized box, wrapped in a sealed plastic bag to exclude moisture. The freezer temperature was not monitored, but no power failures were encountered during the experiment. The film package was removed from the freezer at least three hours (small box) or six hours (large box) prior to exposure. Development began within an hour after exposure (usually immediately). For each test in Part 1 (phenidone-ascorbic acid developers), one sheet of film was exposed in contact with the step tablet, and developed for 11 minutes. In Part 2 (D76 types), standard development time was 8 minutes. The density of the most-dense area of the step tablet was determined initially, and then at each time interval. Testing of each solution ceased when exhaustion occurred, defined as a 0.2 log D unit density loss, as compared to the maximum initial density.
Testing procedures were based on methods specified by Phil Davis (8). Consistent film exposure was ensured by using the same light source (Beseler 45M with condenser head), the same time and f/stop setting (0.4 sec and approximately f/11), and the same step tablet throughout. The enlarger head height or f stop of the enlarger lens (but not the time) was adjusted for each exposure, which produced incident light meter readings of EV +0.5 on the easel. Temperature of the developer, and the tempering water bath, were measured and brought to 20 degrees C (68 degrees F) prior to beginning development. Film was processed in BTZS tubes (Darkroom Innovations, Fountain Hills, AZ), with constant manual rotary agitation.
For selected developers, a family of curves was generated in Plotter (9) in the first week of the experiment to describe their characteristics. Testing intervals planned for phenidone developers are shown in Table 6. Because it was anticipated that MQ-type developers would exhaust more quickly, it was planned to test them every one to two weeks initially, then every month after 8 weeks had elapsed.
|Storage life of...||Solution||Dilution||Testing interval||Preparation|
|1 Powder||PC from powder||1 + 1||Initial, then monthly||Mix PC fresh each month from dry phenidone|
|2 Stock solution||PC from stock||1 + 0||Initial; each week x 4; every two weeks x 4; then monthly||Mix PC at each interval from phenidone 0.15% stock solution (prepared at initial time interval)|
|3 PC||PC from stock||1 + 0||Initial; each week x 8; every two weeks x 2; then monthly||Use PC (mixed from stock at initial interval)|
|4 PX||PX||1 + 1||Initial; each week x 8; every two weeks x 2; then monthly||Use PX (mixed at initial interval)|
|5 XtolTM||XtolTM||1 + 1||Initial; each week x 8; every two weeks x 2; then monthly||Use XtolTM (mixed at initial interval)|
At the end of development, film was placed in SB1 stopbath (48 mL acetic acid 28%, in 1L tap water) for one minute, then fixed in a tray of Ilford Universal Rapid Fixer (1+3 in tap water) with constant rocking agitation. Fixing time was set at 5 minutes (normal clearing time in this lab with this combination is approximately 1.5 to 2 minutes). Film was rinsed in tap water for 3 to 5 minutes, then rocked in a tray of sodium sulfite 2% for 5 minutes as a wash aid. Washing in an archival 4 x 5 washer (Red Village Products, Geneva NY) proceeded for 25 minutes. Film was rinsed in PhotoFlo 200 (1 mL in 200 mL distilled water) for one minute, then hung to dry at room temperature.
Chemicals were kept in tightly closed, amber glass bottles. After mixing, all developers and the phenidone stock solution were stored in narrow mouth 500 mL size amber glass bottles, partially filled (unless noted otherwise). These were placed in a basement darkroom, in a closed cabinet, and a weekly log of temperatures was kept. The temperature of this room ranged from 18.2 to 23.8 degrees C (64.7 to 74.8 degrees F) during the course of data collection (April 1999 to January 2000).
Densitometer readings were recorded for each time interval in Microsoft Excel. Graphs were prepared in Excel or Phil Davis' Plotter software. Missing values were interpolated for the graphs so they would be clearer, but tables of original data with missing values were also prepared. Curve shapes were plotted for the initial family of curves for each developer solution at varying times of development, and for the initial and final film test for selected developers.
Log density readings for the phenidone developers are found in Table 7. The results of one film test are shown in each cell of the table (i.e. these are not averages of two or three film readings). Coefficient of variation is the sample standard deviation, divided by the sample mean. It, as well as standard deviation and range, are ways to show how much variation there was in density, over the time course of the study.
The initial density for Xtol is much higher than the other developers. The three developers with the least variation are PC from stock, PX, and Xtol. Useful life, defined as the time before a change of 0.2 units from the initial density, was longest in these three developers as well.
|Week||PC from powder||Stock solution||PC from orig stock||PX||Xtol|
|Coeff. of variation||0.30||0.44||0.07||0.08||0.04|
Figure 1 shows the same results for the phenidone-ascorbic acid developers graphically. The initial densities are shown as zero, and changes over time are shown as deviations above or below the zero horizontal line. Values were interpolated for intervals which were not tested to produce a more readable graph.
Table 8 shows similar data and summary statistics for the MQ-type developers. The initial density for MC is much higher than the other developers. The developer with the least variation is clearly Kodak D76, although the variability is also low in D76 (no air), D76H, and DK76b. Useful life (time to change of 0.2 units from the initial density) was longest in D76, followed closely by DK76b. All the rest lasted less than 14 weeks.
|Week||D76||D76 (no air)||D76H||DK76a||DK76b||MC|
|Coeff. of variation||0.04||0.07||0.10||0.08||0.07||0.07|
|Exhaustion (weeks)||> 24||6||12||14||22||8|
Figure 2 shows the same results for the MQ-type developers graphically. The initial densities are shown as zero, and changes over time are shown as deviations above or below the zero horizontal line.
Curve shapes at exhaustion did not differ greatly from that found initially. They gave the appearence of a curve with shorter development time. Figure 3 shows PC developer curves initially and at 28 weeks.
Figure 3 PC curves at 0 & 28 weeks.
Figure 4 shows the curves for D76H developer initially and at 14 weeks.
Figure 4 D76H at 0 & 14 weeks.
The curve families for the phenidone-ascorbic acid developers were similar in appearence. Thus PC or PX could be expected to produce generally similar results to Xtol in gradation, while they differed in development time needed and effective film speed.
Figure 5 Xtol
Figure 6 PC developer
Figure 7 PX developer
The curve families for the MQ-type developers were also similar in appearence. DK76b produced identical results to Kodak D76 in terms of effective film speed, gradation, and development time.
Figure 8 D76 (Kodak)
Figure 9 DK76a
Figure 10 DK76b
Xtol was the most-active of these developers, producing a maximum density of 1.8, as compared to 1.0-1.2 for all other phenidone developers tested, with a standard developing time of 11 minutes.
Phenidone was hard to measure accurately. This may have been the cause of the wide swings in activity I saw when testing PC developer made from the dry phenidone at each time interval. This is due in some part to the fact that any scale made to measure to the nearest 0.1 gm will read "0.2 gm" when masses of 0.151 to 0.249 gm (0.098 gm range) are on the balance. This range is an significant fraction- nearly half- of the mass you are weighing. Compare this to the same errror with a larger mass: when you are weighing 10.0 grams, the scale displays "10.0 gm" with weights from 9.951 to 10.049 gm (still a 0.098 gm range, but this time only around 1% of the weight on the pan).
Because of the inherent difficulty of weighing small quantities of phenidone, I had high hopes for the stock solution mixed at the initial interval, and used to measure phenidone in making PC developer at each testing interval. But the activity of this stock solution was significantly poorer than the keeping qualities of PC, PX, or Xtol developer themselves. I wouldn't recommend using it.
I didn't make a formal test of the gradation of PC or PX compared to Xtol. But the shapes of the curve families are reasonably similar, so they would be worth exploring further. They did not produce the same effective film speeds as Xtol. They required longer developing times to produce the same densities on film. But either lasted six months or so before exhaustion. If you are interested in scratch-mixing your own Xtol, either formula would be worth testing.
Finally, I was surprised by how hard phenidone is to dissolve. I tried dissolving it in a small quantity of isopropyl alcohol, which was not particularly effective. Heat was less effective than I would have guessed before beginning. I had to stir phenidone for 7-10 minutes before it would dissolve, even with distilled water at 150-170 degrees F.
There's certainly nothing wrong with Kodak D76. Its useful life exceeded 6 months in this experiment. Its variability (standard deviation) week to week was half that of any other developer tested.
Strangely, attempts to prolong its useful life by storing it in an accordian-type plastic bottle and excluding air resulted in less time before exhaustion occurred! I can only guess that the plastic container was not impervious to oxygen in the room air. Otherwise I cannot account for this- excluding air from the bottle "should" have worked, according to everything I have read.
D76H has been advocated by Anchell and Troop (2) as a developer which is much more stable than D76. I found that it was reasonably stable (exhaustion 12 weeks).
Two of the other developers in this experiment are modifications of D76H. DK76b is a developer with increased metol and no hydroquinone (like D76H), but with sodium metaborate substituted for borax. MC is a developer with increased metol, but with ascorbic acid instead of hydroquinone, and a large amount of sodium metaborate instead of sodium sulfite.
DK76b turned out to be the developer of choice. Its gradation, film speed, and developing times mirror those of D76 or my former standard developer, DK76a. And it was quite stable week to week, with a very long shelf-life (22 weeks) besides. Relying on metol alone rather than the MQ superadditive combination may mean that this developer might be more compensating (less tendency to block highlights), like D23 is said to perform.
MC still interests me, in spite of its modest shelf-life of 8 weeks, because of the large quantity of ascorbic acid I now have on hand, and because it has the potential for compensating development (no hydroquinone), room temperature compounding (no hydroquinone), and sharpness (no sodium sulfite). I will be examining the curve family and pictorial qualities of this developer in the future.
Why are my results so much more liberal than published recommendations? Kodak suggests that the shelf life of D76 is 1-2 months. I found excellent activity remained at 24 weeks (6 months). I would speculate that glass bottles were important and would recommend their use instead of plastic- they are impervious to oxygen. The difference in shelf-life between a single batch of D76, split between a glass bottle and a Falcon plastic accordion-type bottle was notable. Secondly, consistent pH and lack of any contaminants makes distilled water a clear choice if you want maximum storage life and consistent results. Again, this is speculation, because my data only relate to storage life, I did not test what factors promote it in any given developer.
When a manufacturer makes their recommendations, it is on the basis of multiple tests, with a variety of storage temperatures, water samples, and conditions of storage such as varying temperatures. My data show a way, certainly not the only way, in which to produce prolonged shelf-life in your own darkroom.
This method doesn't formally evaluate curve shape, or the gradation of photographs taken with progressively aging developer. Nor does it evaluate any corrective actions (could one simply develop longer and get good results with an "exhausted" 9 week old batch?). The results can be applied to other photographer's process controls only to the degree their storage conditions match mine. Finally, it is unknown how long the phenidone I received in April 1999 had been on the shelf prior to shipping.
The formal language I use masks one technical flaw. Because of budget and time constraints, I tested each developer with one film sheet at each time. Had I tested two or three sheets of film, I would have had a better idea of the test to test variability, and been able to apply statistical techniques such as life table analysis or analysis of variance. Without these techniques, I cannot state with confidence what part of the week to week variability is due simply to random variations in the methods (such as the small variations between one sheet of film and another), and what part is due to change in developer activity over time.
It was concluded that the commercial products D76 and Xtol are consistent performers whose storage life is underestimated by their product literature (at least, when mixed with distilled water and stored under the conditions of this experiment). For scratch mixers, DK76b and two varieties of a phenidone-ascorbic acid developer (called PC and PX here) warrant consideration since their performance is consistent and they last nearly 6 months before exhaustion.
Return to the top of this page.
The author has no financial or paid consulting relationship with the manufacturers of any device or material mentioned here. The recommendations made here are the author's personal opinions based on his own darkroom practice, consultation with published materials and colleagues, and his own research.