Mixing BAC water yourself: the math and its pitfalls

Making bacteriostatic water (BAC water) yourself: The recipe requires exactly 9 milligrams of benzyl alcohol per milliliter of water, which for a 10-milliliter vial equals 90 milligrams. Since benzyl alcohol has a higher density than water, this amount corresponds to just under 0.09 milliliters, or roughly 9 units on an insulin syringe with a U-100 scale.

The practical hurdle lies in manual precision: You have to hit these 9 units exactly and prevent contamination in the vial, yet you cannot verify the success of either step at home. This guide explains the exact procedure and highlights the typical sources of error.

What do you need to make BAC water yourself?

To make BAC water yourself, you need exactly three components, where only purity and pharmaceutical quality matter.

Sterile water for injection. Sterile water for injection is a tested pharmaceutical product and is fundamentally different from simple distilled water. It is produced exclusively by distillation or reverse osmosis and meets significantly stricter limits for germ counts and impurities than commercially available demineralized water. This starting material is available, among others, in pack sizes of 5, 10, 400, and 1000 milliliters.

The source is the pharmaceutical trade: Since sterile water for injection is a regulated medicine, the original label always carries a batch number and an expiration date. This product does not exist in the range of drugstores or hardware stores; if batch number and expiration date are missing from the label, it is not approved water for injection.

Distilled water from the hardware store, battery water, boiled water, and tap water are not substitutes. They are not tested for germs or for fever-inducing substances.

Benzyl alcohol in pharmacopoeia quality. Benzyl alcohol is a pharmaceutical preservative. In pharmacopoeia quality, it must meet the official standards of the Ph. Eur. (European Pharmacopoeia) or USP (United States Pharmacopeia) with verified purity. Pharmaceutical benzyl alcohol has a content of 98 to 100.5 percent and a defined density of 1.043 to 1.049 grams per milliliter.

The same substance exists in three different commercial grades: as a technical product for industrial purposes, as a cosmetic grade for creams and perfumes, and in pharmaceutical quality for medicines. Chemically, it is the same liquid, but the purity criteria and testing standards differ fundamentally. For pharmaceutical quality, the pharmacopoeia strictly limits impurities: the peroxide value may be at most 5, and benzaldehyde and related substances are tested separately - for product intended for injection, using a stricter comparison solution than for any other use. These purity tests do not apply to technical and cosmetic grades. Such product often contains residual impurities that can trigger unwanted reactions in an injection solution. How far an individual container deviates from the standard cannot be determined, because no published limit values exist for these commercial grades.

You can recognize genuine pharmacopoeia quality by the pharmacopoeia designation on the label and by the batch-specific certificate of analysis (CoA), which documents the required content and density values through laboratory analysis. If these documents are missing, the actual purity remains unverified. This complete quality documentation explains the price difference between tested pharmacopoeia quality and technical-grade material from chemical suppliers.

Insulin syringes with U-100 scale. The U-100 scale is a standardized volume graduation for insulin syringes: 100 units (IU) correspond exactly to 1 milliliter, so 1 unit equals exactly 0.01 milliliter. This unit of measure applies uniformly to insulin syringes with a total capacity of 0.3, 0.5, or 1 milliliter. The nominal volume only limits the maximum fill amount; it does not change the scaling per unit.

The scale division determines the volume of each individual mark: syringes with 0.3 milliliter capacity usually have marks for half or whole units, 0.5-milliliter syringes for whole units, and 1-milliliter syringes for whole or double units. With a 1-milliliter syringe in 2-unit steps, one mark corresponds to 0.02 milliliter. Therefore, check the exact scale division of your syringe before each dose.

A finer scale division does not guarantee higher dosing accuracy: in a comparative study, half-unit markings improved neither the accuracy nor the repeatability of the dose. When measuring a single unit (0.01 milliliter), the tested syringes were clinically dangerously inaccurate.

What is the exact calculation for BAC water?

The pharmacopoeia specifies a guideline value of 9 milligrams of benzyl alcohol per milliliter for bacteriostatic water, which corresponds to a mass fraction of 0.9 percent. The permissible upper limit is 11 milligrams per milliliter (1.1 percent), so the usable target range is between 9 and 11 milligrams per milliliter.

The concentration specification of 0.9 percent necessarily refers to mass and not to liquid volume. Anyone who mistakenly uses a volume fraction of 0.9 percent will get a slightly higher content, which is usually still within the tolerance range but is mathematically imprecise. Taking density into account, 9 milligrams of benzyl alcohol corresponds mathematically to about 8.6 microliters.

Container	Benzyl alcohol	On the U-100 syringe
10 ml	90 mg	about 9 units
50 ml	450 mg	about 43 units
100 ml	900 mg	about 86 units

The solubility of benzyl alcohol in water is up to 4 grams per 100 milliliters, which is why the required amount of 9 milligrams per milliliter dissolves completely and easily in water.

Where do measurement errors occur when measuring by hand?

The permissible range of 9 to 11 milligrams per milliliter covers a span of about 20 percent, but in practice it is quickly exceeded by manual dosing errors when measuring small liquid volumes.

A study on the accuracy of measuring small volumes found actual deviations between 1.4 and 18.6 percent, despite a manufacturer tolerance of plus/minus 5 percent, depending on the individual user technique. The manual measurement error of an inexperienced person thus already uses up almost the entire allowed range. The authors therefore recommend measuring at least 20 percent of the syringe volume: 43 units on a 1-milliliter syringe easily meet this requirement, while 9 units on the same syringe fall below the limit.

Additional deviations are caused by the dead volume, i.e., the residual liquid that remains in the syringe cone and needle after fully pressing down. The standard ISO 7886-1 allows a dead space volume of up to 70 microliters for syringes under 5 milliliters; real measurements showed about 34 microliters for a 1-milliliter syringe without a needle and values between 31 and 74 microliters for needles alone.

With an injection, this dead volume is lost: With a target volume of 0.5 milliliters, a dead space of 50 microliters already corresponds to a loss of 10 percent; with 0.3 milliliters, it is 17 percent. When transferring, it is different: If you push the liquid completely through the needle into the target container, only the thin wall film remains in the syringe. If, on the other hand, you remove the needle before emptying, its entire liquid content is lost; accuracy therefore depends on the handling.

Why don't you notice errors when making BAC water?

An underdose of benzyl alcohol cannot be detected without a microbiological laboratory, since an insufficiently preserved solution remains completely inconspicuous in smell, appearance, and flow behavior. In the specialist literature, no minimum concentration is documented below which the protective effect reliably ceases; with too low a content, you leave the scientifically studied range.

An overdose of benzyl alcohol does not increase safety but destabilizes the protein structure. Premature unfolding, aggregation, or precipitation has been documented for growth hormone, IGF-1, interferon gamma, G-CSF, and the interleukin-1 receptor antagonist, among others; with a model protein, the unfolding temperature dropped by 5 degrees. A concentration that is too high thus endangers the stability and integrity of dissolved peptides.

Two scientific limitations apply to these findings: The available stability studies were mostly conducted with 0.9 percent benzyl alcohol, and direct comparison series of several concentrations are largely lacking. Furthermore, the data refer to larger proteins, while the effects on short-chain peptides are little researched.

Additionally, the pH value influences the preservative power: At 1 percent benzyl alcohol, germs in studies proved particularly sensitive between pH 4 and pH 5, but significantly more robust at pH 6. A targeted pH adjustment is not possible at home without laboratory equipment.

What does a sterile filter achieve with BAC water and what does it not?

A syringe filter with a 0.22 micrometer pore size separates bacteria and other microorganisms purely mechanically from the liquid pushed through, but does not chemically kill germs.

Fever-inducing metabolic products of dead germs (endotoxins) pass through the 0.22-micrometer filter unhindered. The FDA classifies the retention capacity of membrane filters against such pyrogens as narrowly limited, since these molecules are heat- and acid-stable and persist after cell death; filtration therefore cannot remediate impure starting materials.

Benzyl alcohol as a preservative also has no sterilizing effect: It only inhibits the subsequent multiplication of introduced germs but does not reliably kill pathogens that are already present.

How long is opened BAC water stable?

According to the specifications of the US Food and Drug Administration (FDA), the standard usage period for multi-dose injection containers is up to 28 days after first puncture, provided product-specific tests do not document a different period. This guideline presupposes the existence of standardized microbiological preservative effectiveness tests and limits the container size to a maximum of 30 milliliters.

The 28-day period serves as a safety limit with regular withdrawal and does not mark an automatic expiration date of the vial contents. Real contamination risks were studied on 384 opened multi-dose vials: Storage conditions proved to be the statistically significant risk factors, followed by the reuse of needles and the cleanliness of the withdrawal environment. The study showed no significant correlations for hand disinfection, printed expiration date, or stopper disinfection.

Germ contamination rates in studies vary between 0 and 18 percent of contaminated vials, with literature peaks up to 27 percent. In clinical practice, an evaluation of 1,223 weekly samples from 863 multi-dose vials with an average usage period of 9.5 days and a maximum period of 402 days found no germ detection at all.

For opened BAC water, the Injection Manager tracks two benchmarks: 28 days as a documented manufacturer specification that assumes frequent withdrawal under laboratory conditions, and about 90 days as community experience without evidence, with roughly weekly withdrawal and continuous refrigeration. The second number is neither a recommendation nor a measured value, but describes what users do.

These benchmarks apply to the opened water, not to your finished peptide solution. That has its own, shorter calculation. Never transfer the 90 days to a mixed vial.

From here on, it gets technical

On the choice of concentration: An explicit justification by the pharmacopoeias for exactly 0.9 or 1.1 percent is not found in the sources. In the specialist literature, 0.9 percent is considered the established standard for preserving multi-dose protein preparations.

On the mechanism: Benzyl alcohol shifts the equilibrium in favor of partially unfolded states and lowers thermodynamic stability. Sugars such as sucrose counteract this because they are excluded from the protein surface and favor a more compact form.

On tolerability: The EMA lists an acceptable daily intake of 0 to 5 milligrams per kilogram of body weight for benzyl alcohol. The so-called gasping syndrome in premature infants was observed at 100 to 200 milligrams per kilogram per day administered intravenously; the smallest toxic amount is not known. The FDA label carries the note that the product must not be used in newborns and excludes use in the spinal cord.

On sterile filtration: The retention of a filter is demonstrated by a test with a particularly small germ of about 0.3 to 0.4 micrometers, together with a test of the filter for integrity. Both are laboratory procedures.

On the in-use period: A generally applicable period for preserved multi-dose containers cannot be derived from the pharmacopoeias. The determination is product-specific and is based on stability data from the respective manufacturer.