C41 is the colour-negative film process and chemistry kits are available to do this at home. It is strongly recommended that you become familiar with black and white processing before attempting colour processing.
If your time is free, it is quite possible to develop colour film for approx $1 per roll with off-the-shelf kits, an appealing option for those of us where professional labs are charging close to $10/roll.
The C41 process is carried out at 37.8C and requires tight temperature control in order to prevent colour shifts and crossover (shadows and highlights having a different hue). The process steps are:
Each step is described below...
The film is pre-washed using process-temperature water, which serves the following purposes:
A total of at least 5:00 washing time should be used, with two changes of water, e.g. a 2:00 wash followed by a 3:00 wash. All wash-water should be at process temperature (37.8C).
The colour developer is CD-4. It develops the silver images in each layer of the film and the development by-products are designed to interact locally with the dye-couplers in the film to produce dyes where development occurred.
The more exposed an area of film is, there more development activity occurs and therefore the more dye that is formed, hence it being a negative process. The image on the film is darkest where the film was exposed to the most light.
The standard development time is 3:15. If you are using a non-replenishing chemistry then the development time will be extended when the developer is re-used in order to account for the partial exhaustion; you will need to consult the table provided with your kit for these times. Temperature control is critical during development.
Pushing and pulling is achieved by adding or subtracting (respectively) 0:30 to the development time for each stop of push/pull desired, e.g. a 1-stop push with fresh developer will have a development time of 3:45.
Agitation should be continuous during development. Development duration is fairly important; it is the time between when the developer first hits the film and the next bath (stop or bleach) hits the film. Development is still occurring in the film after the bulk of the developer has been poured out and there is a small quantity remaining on the film.
A quick wash step here (one change of water, approx 30s) will reduce forward-contamination of the bleach and prolong its life. Water is not a very effective stop agent however, so this bath should be long enough just to swill excess developer from the tank and spirals.
The bleach is responsible for converting metallic silver (produced during development) back into silver halide so that it can be removed by the fixer. The process runs to completion, so it is safe (even recommended) to run the bleach process for a couple of minutes longer than specified, e.g. 8:00 instead of 6:30, especially if the bleach is nearing its rated capacity. Temperature is not critical.
Bleach requires oxygenation to be active. Immediately before use, take a half-full bottle of bleach and shake it vigorously for a few seconds to promote oxygenation.
Because the silver is to be all removed, a rehalogenating bleach is not required and typical C41 bleaches will often remove some of the metallic silver entirely.
This wash step removes bleach from the emulsion and is necessary because the bleach can destroy the fixer. Incomplete washing at this stage can lead to short fixer life and therefore poor fixing.
Using the rotary processor, I change the water once per minute for about 7 minutes, using 1.5-2x as much as is necessary to cover the film. Always have at least one change of water be the max rated capacity (generally 1L for Jobo) of the tank.
The fixer (usually an ammonium thiosulfate based rapid fixer, similar to B&W fixers) is responsible for removing silver halide from the emulsion. Because the bleach has converted all the metallic silver back to halides, all of the silver will be removed from the image by the fixing step, leaving only the dyes produced by the developer.
Fixing proceeds to completion, therefore it is permissible or even advisable to fix for slightly longer than specified, particularly when the fixer is reaching its rated capacity. Temperature is not critical.
This wash stage is responsible for removing all fixer from the film in order for it to be archivally stable.
Using the rotary processor, I change the water about 8 times in 10 minutes (longer washes near the end, shorter-washes near the beginning), using 1.5-2x as much as is necessary to cover the film. Always have at least one change of water be the max rated capacity (generally 1L for Jobo) of the tank.
The stabiliser bath contains biocidal agents (anti-fungals, mostly) that will help ensure nothing grows on the film in storage. DO NOT WASH the film after it has been in the stabiliser bath; the stabiliser must dry onto the film.
Stabiliser (and the surfactants it contains, similar to Photoflo) have a reputation for gumming up tanks, spirals and funnels (Jobo Lift). I never put stabiliser through the Jobo Lift from a perhaps-irrational fear that it will be difficult to wash out and likewise don't bother putting it in the tanks either. Instead, I transfer each spiral (after the wash) to a 2L plastic container full of stabiliser and let it sit there for the required time before draining and drying the film.
After each spiral has been unloaded, it (the spiral, NOT the film) is washed with hot (50-60C) water to remove all traces of stabiliser and surfactants and prevent contamination of later processes using the same spiral.
Dry the film in preferably-filtered air. If dust reaches the film while wet, it will sink into the emulsion and never be removed, whereas dust landing on dried film can be blown off.
The film may look slightly milky and purple on one side when wet but it will dry to the usual burnt-orange colour.
As of 2011, C41 chemistry is a little difficult to find (and ship!), but the following kit options at least are available:
Kodak seems to have discontinued their C41 kits.
If you're processing lots of film, it is possible to buy all of the components separately from Fuji or Kodak in large quantities (e.g. to make 100L). Bulk chemistry is by far the cheapest option per-roll if you are processing large quantities but you're buying about $500 of chemistry up-front. If you're not processing a lot of film then the developer may go off before you exhaust it.
Professional C41 chemistry has three active baths: developer, bleach and fixer. If you can get a kit like this, it will be the most-reliable option and give the highest-possible quality results.
Some home kits are available that combine the latter two baths in a single "blix" stage, making the process slightly simpler to carry out, but this has the drawback that bleach and fixer are mixed into a single solution, at which point they begin attacking each other and suffer short shelf-life.
Fresh blix-based kits are reported to produce results indistinguishable by eye from bleach+fix kits, however the shorter shelf-life of the blix can mean that later processing results will be (non-obviously) sub-par and the archival longevity of the film can be negatively affected. Don't go there if you can avoid it, particularly considering that the use of blix can save only a few minutes of processing time.
You need an accurate thermometer; a "colour thermometer" designed for this purpose is the best option. Second-best is a digital fever thermometer designed for measuring humans (it covers the correct range and should be accurate) but they often read quite slowly. You can also use a digital aquarium thermometer but it's best to buy a few (they're about $2 each) and find the one that best matches an accurate reference colour-thermometer.
The primary change in equipment requirements compared to black and white is driven by the need for accurate temperature control. The easiest solution is to buy a rotary processor (e.g. Jobo) that provides constant agitation and automated temperature control, but these are not cheap.
It is also possible to build a home-brew C41 processor using simple inversion tanks (e.g. Paterson), a water bath and heating element: commonly known as a slow cooker and available from kitchen stores. If you're brave you can control the temperature by hand using a dimmer but a better option is to buy an off-the-shelf PID (proportional/integral/differential) controller to handle the temperature control.
The cheapest option is to make up a large (at least 20L) quantity of water at about 42C in a plastic container; it will cool slowly due to its large thermal mass. Watch the temperature carefully, if you have enough water the drop will be less than 1C during a 4-minute period and this is accurate enough to process the film. Start processing when the water is slightly too-hot (e.g. 38.2C) and complete it 3:15 later when the water has reached about 37.4C. Add boiling water to bring it back to 40C for the next batch of film and repeat.
When using a large water bath, perform agitation with the tank on the surface of the water, or dunk it in the water at each inversion.
The temperature control (assuming Jobo sensor is calibrated) should be set to 38.0C which, with evaporation off the tank in a 24C room, will result in about 37.8C in the tank according to Fuji instructions.
Jobo 2xxx tanks in a CPA-2 or CPP-2 should be operated at "P" speed. 3xxx tanks should be operated at "4" speed, perhaps slightly slower. Agitation should be set to reverse after two revolutions in each direction, though some report that results with non-reversing agitation are fine.
A Jobo Lift is highly recommended, especially for the frequent water changes performed while washing.
The Fuji kit has a rated developer life of a couple of weeks only once mixed and partially used but if stored under butane in a refrigerator (under 4C but never frozen), it can keep for at least four months without any degradation in performance. This has been verified with a pair of identically-exposed test rolls of Portra.
I have used only the Fuji 5L kit and can't speak from experience on any of the other chemistry options. I use a Jobo CPP-2 with lift; I've never bothered with any of the manual/hacky temperature-control schemes.