A staining solution that worked last month gives a weak or patchy result today. In most cases the powder has not changed; the solution has. This article covers six stains and, for each, the fault that most often causes the problem, what causes the fault, and what to do about it.

Hematoxylin: not ripened, or over-oxidised

Hematoxylin is not itself a dye. It is almost colourless, and it stains only after oxidation to hematein, which then binds a metal mordant, usually aluminium or iron.^[1][2] The oxidation removes two hydrogen atoms and converts one of the two catechol rings into a quinone (Figure 1). Every hematoxylin recipe depends on this step.

The colour of the solution shows its state. A fresh solution of hematoxylin in water is pale yellow-brown; a ripened solution is deep brown. With an aluminium salt, the unripened solution is a pale, transparent violet and the ripened solution a deep, opaque purple.^[1]

Two faults follow from this. The first is a solution that has not been ripened. Ripening in air takes weeks to months, so most recipes now add sodium iodate, and the solution can be used as soon as it has cooled.^[1][2] The second is a solution that has been oxidised too far. Oxidation does not stop at hematein; it continues for as long as the solution exists, and the solution is exhausted when too little hematein remains.^[1] Over-oxidised solutions form precipitates and stain poorly.^[3] This is also the reason hematoxylin is sold rather than hematein. A solution made from hematoxylin holds a reserve that continues to supply hematein, whereas a solution made from hematein loses strength from the first day.^[1] For the same reason, many recipes add only enough oxidant to convert half of the hematoxylin, and leave air to oxidise the rest slowly.^[1]

Two other faults resemble over-oxidation. Hematein and its aluminium complex are only sparingly soluble in water, so they settle as a sediment or form a metallic film on the surface. Recipes add glycerol or ethylene glycol to keep them dissolved. These glycols were once thought to prevent over-oxidation, but their function is solubility.^[2] In routine use, the most common reason an alum hematoxylin stops working is alkaline tap water carried over from the slides. Alum hematoxylins are made acidic: Gill's formulation contains acetic acid and Mayer's citric acid, because the acid keeps the stain selective for nuclei.^[2] Tap water neutralises the acid and the lake precipitates. A small addition of acetic acid often restores the solution.^[1]

Iron hematoxylins differ. The ferric salt is both the mordant and the oxidant, so no separate oxidant is added.^[1][3]

A hematoxylin solution is usable from ripening until the hematein is consumed. In daily use, most solutions that stop working have been neutralised by tap water, not exhausted by age.

Product: Hematoxylin, CAS 517-28-2, ≥99.0%

Alcian blue: precipitation, and a powder that differs from the original

Alcian blue is a copper phthalocyanine with four cationic side-chains (Figure 3). It is used at pH 2.5 to stain glycosaminoglycans and acid glycoproteins. The four positive charges bind the sulfate and carboxylate groups of these targets, and the same charges make the dye sensitive to salt. Two faults are common.

The first is precipitation in salt solutions. The critical electrolyte concentration methods stain from solutions that contain magnesium chloride, and not every batch of alcian blue remains dissolved under those conditions. The Biological Stain Commission therefore tests each batch in 0.5 M magnesium chloride at pH 5.7, and certifies for those methods only the batches that do not precipitate.^[4][5] A batch that precipitates in a salt solution may be unsuitable for that method rather than defective. In 3% acetic acid at pH 2.5 the solution is more stable; for a 1% solution that forms a precipitate on storage, the supplier's advice is to pass it through a 0.22 or 0.45 µm filter.^[7]

The second fault is that the powder differs from the one the recipe was written for. Production of the original alcian blue 8G ended more than fifty years ago. The dyes now sold under that name are copper phthalocyanines with at least four different solubilising side-chains, and the Commission classifies most of them as "alcian blue variants" rather than the original dye.^[4][6] Current batches also often contain more than twice as much dye as the batches used when the published methods were written.^[4] A weight taken from an old recipe can therefore give twice the intended amount of dye and heavier background staining. The correction is described in our article on dye content: divide the weight of dye the method requires by the dye content of the batch.^[15]

With alcian blue, the cause is as often the batch as the solution.

Product: Alcian blue 8GX, CAS 33864-99-2, dye content ≥80%

Trypan blue: precipitate counted as cells

Trypan blue is used as a 0.4% solution in saline for the dye-exclusion viability test. The known fault is a precipitate that increases with storage time and with freezing, and orange-red fibrous aggregates that form after refrigeration.^[8][9] The solution still stains. The problem arises in the counting chamber, where an automated counter can count the particles as cells and so alter both the total count and the viability figure.^[10]

Suppliers give three remedies: filter through a 0.22 µm membrane; allow the stock to settle and do not shake or vortex it; or warm the vial to 37 °C for ten minutes.^[8][9] We found no primary paper on the precipitate; the information above is supplier practice.

Trypan blue precipitate does not stop the stain from working, but it distorts the count.

Product: Trypan blue, CAS 72-57-1, dye content ≥70%

Methylene blue: conversion to azures

Methylene blue is the parent of a family of dyes. Under oxidising or alkaline conditions it loses methyl groups one at a time to give azure B, azure A and azure C (Figure 4). This process is called polychroming, and it is how azure B, the dye that stains chromatin purple in blood films, is produced commercially: by partial purification of polychromed methylene blue.^[4] In Wright and Giemsa stains the conversion is intended, because the combination of azure and eosin gives the Romanowsky colours.

The fault is that the conversion continues in the bottle. In a mixed blood stain, methylene blue continues to demethylate to azures as the solution ages, and azure–eosin salts precipitate, so the colour balance of a stored working solution changes and precipitates appear.^[11] Manufacturers follow the balance by the ratio of absorbance at 650 nm (the blue dyes) to absorbance at 525 nm (eosin); for a Giemsa-type stain the ratio is about 1.7–1.8.^[11] Polychromed methylene blue is itself a mixture of many dyes, which is why these stains have always been difficult to standardise.^[12] The usual practice follows from this: keep the stock in methanol, dilute it into buffer immediately before use, and discard the diluted solution afterwards.

For an aqueous methylene blue stock without eosin or alkali, the sources we read give neither a rate of conversion nor a shelf life, so we do not state one.

In a blood stain, methylene blue is meant to convert to azures. The fault is allowing the conversion to continue in the working solution.

Product: Methylene blue trihydrate, CAS 7220-79-3, ≥99.5%

Crystal violet: precipitate on the smear

The Gram stain uses crystal violet as a solution of about 2.3% dye with ammonium oxalate and 20% ethanol, the Hucker formulation (Figure 5).^[13] The common fault is dye that crystallises in the bottle and is deposited on the smear, where the crystals can be mistaken for organisms. Laboratory procedures therefore examine a control slide for precipitate with each new batch, and filter the crystal violet into a clean bottle before use; the solution is not centrifuged.^[14] If a deposit has formed, the manufacturer's instruction is to warm the solution to about 35 °C and stir; the filtered solution can be used again.^[13]

Two other faults resemble a crystal violet problem. The first is the iodine. Gram's iodine loses strength on storage,^[13] and a weak mordant gives weak retention of crystal violet. The second is contamination of the stain. A case report describes Gram-negative bacilli seen in smears that were in fact growing in the crystal violet bottle;^[16] filtration through a bacteriological filter before use removes them.

A Gram stain usually fails because of precipitate or weak iodine, not because of the dye.

Product: Crystal violet, CAS 548-62-9, dye content ≥88%

Resazurin: reduced before the assay

Resazurin is the one stain in this article that is meant to change. It is a redox indicator. As supplied it is dark blue and only weakly fluorescent; living cells reduce it to resorufin, which is pink and strongly fluorescent, and that fluorescence is the viability signal.^[17] On reduction the absorption maximum moves from about 605 nm to about 573 nm, and the fluorescence of resorufin is measured at about 580–590 nm.^[17][18] Resorufin can be reduced a second time, to hydroresorufin (dihydroresorufin), which is colourless and does not fluoresce (Figure 6).^[17]

The fault is that both reductions also occur without cells. The first is caused by light: prolonged exposure of the reagent to light raises the background fluorescence, and the background in wells without cells also rises over hours of incubation.^[17] A stock that has become pink already contains resorufin, and resorufin carried into the assay is read as living cells. For this reason one supplier states that a trace of resorufin in the powder makes it unsuitable for bioassays.^[19] The second reduction is caused by an excess of reducing capacity: with high cell numbers and long incubations, resorufin is reduced further to the colourless form, the signal falls, and healthy wells appear dead.^[17]

The storage advice from assay suppliers follows from this. Keep the reagent in the dark. A frozen stock tolerates at least ten freeze–thaw cycles. A stock in daily use can be kept tightly closed at room temperature for some weeks. A stock solution prepared from the powder is best used the same day and discarded within a month.^[17][20] Serum in the medium adds background of its own, so wells without cells, containing the same medium, are the control for all of these effects.^[17]

Resazurin is meant to become pink in the well, not in the bottle. A pink stock has already been partly reduced.

Product: Resazurin sodium salt, CAS 62758-13-8, ≥90.0%

One rule that covers all six

Each of the faults above is caused by a precipitate or by an oxidation. Filtering before use removes the first; storage in the dark, in a full and closed bottle, slows the second. For the weight of dye to use, see our article on dye content.^[15]

© 2026 NorrChemica. All rights reserved.
Content may not be reproduced without written permission.