Reference library

Bacteriostatic Water vs Deionized Water

Bacteriostatic water is sterile, preserved with 0.9% benzyl alcohol; deionized water is neither. Compare sterility, endotoxin specs, cost, and which to use.

Bacteriostatic water is sterile water containing 0.9% w/v benzyl alcohol as an antimicrobial preservative, packaged in sealed multi-dose vials with sterility and endotoxin documentation attached to the lot. Deionized water is water that has had dissolved ionic species stripped out by ion-exchange resin — it is graded by resistivity and total organic carbon, it is not sterile, it contains no preservative, and it carries no endotoxin specification unless it has been further processed. They are not interchangeable materials, they are not tested against the same standards, and they cost roughly two orders of magnitude apart per milliliter. The choice between them is a documentation and analytical-interference question, not a purity question — deionized water is frequently the purer of the two.

The specifications each material is actually held to

Bacteriostatic water is a compendial product. The relevant specification is Sterile Water for Injection carrying benzyl alcohol at 0.9% w/v — that is 9 mg/mL, or about 83 mM benzyl alcohol (MW 108.14 g/mol). It is tested for sterility under USP <71>, for antimicrobial effectiveness under USP <51>, and for bacterial endotoxins under USP <85>, where the limit for water for injection is 0.25 EU/mL. Vials are typically 10 mL or 30 mL, closed with an elastomeric stopper rated for repeated needle entry, and labeled with a 28-day in-use period after first stopper puncture regardless of how much volume remains.

Deionized water is graded under ASTM D1193 or ISO 3696 and the grade is the whole story:

  • Type I / ultrapure: resistivity 18.2 MΩ·cm at 25 °C, TOC typically <10 ppb, often <5 ppb on a well-maintained polisher.
  • Type II: resistivity >1 MΩ·cm, TOC under 50 ppb — adequate for buffer preparation and glassware rinsing.
  • Type III: resistivity >0.05 MΩ·cm (>4 MΩ·cm on some scales), suitable for rinsing but not for trace analysis.
  • USP Purified Water: conductivity limit 1.3 µS/cm at 25 °C (Stage 1), plus a TOC limit of 500 ppb.
  • HPLC-grade bottled water: filtered to 0.2 µm, UV-treated, sold in 1 L and 4 L bottles with a lot-specific certificate.

Notice what is missing from every deionized water grade: nothing in that list constrains microbial count. Resistivity measures ions. Bacteria and neutral organic molecules are poor conductors, so a recirculating loop can read a flawless 18.2 MΩ·cm while carrying a substantial biofilm-shed colony count. A resistivity meter is not a sterility test and has never been one. That single fact accounts for most of the confusion around this comparison.

Why the preservative matters analytically, not just microbiologically

This is the part most write-ups on the topic skip entirely. Benzyl alcohol is an aromatic compound with strong UV absorbance — a λmax near 257 nm with meaningful absorbance running down through 210–220 nm, exactly the wavelength range where peptide and small-molecule reference standards are quantified by their amide backbone.

At 9 mg/mL, benzyl alcohol is present at 1,000 to 10,000 times the concentration of a typical analytical standard prepared at 0.1–1 mg/mL. Consequences at the instrument:

  • A reversed-phase gradient starting at 5% organic will hold benzyl alcohol on-column and elute it as a large, tailing peak that can obscure early-eluting impurities.
  • Benzyl alcohol oxidizes slowly to benzaldehyde, which absorbs near 250 nm and shows up as a growing secondary peak in aged vials — an impurity that belongs to the diluent, not the sample.
  • In ESI-MS, an 83 mM co-solute contributes background across the low-mass region and competes for charge, degrading signal for the analyte you actually injected.
  • Benzyl alcohol has documented effects on protein and peptide conformational stability in the literature, including promotion of aggregation in some formulations — a materials-compatibility property worth checking against the specific reference material before it ever enters a sample vial.

A resistivity meter reads 18.2 MΩ·cm on water that may carry a substantial microbial load. Ion purity and microbiological control are separate specifications, and no single number covers both.

Ultrapure deionized water has essentially no UV chromophore and no MS background of its own, which is why it, and not preserved water, is the standard blank and sample diluent for chromatographic work. If the purpose is generating an HPLC or LC-MS record against a certificate of analysis, the preserved product is the wrong material for the injection.

What each one costs, honestly

Per-milliliter economics diverge sharply:

  • A 30 mL bacteriostatic water vial commonly retails $8–$25, which is $0.27–$0.83 per mL.
  • 4 L of HPLC-grade water runs roughly $30–$60, or about $0.008–$0.015 per mL — a 50× to 100× difference.
  • An in-house Type I polishing system costs $6,000–$15,000 installed, with consumable cartridges and UV lamps at $400–$900 per year. At 10 L/day that amortizes below $0.01 per liter inside three years.

Vials also carry storage constraints that bulk water does not: controlled room temperature at 20–25 °C with excursions permitted 15–30 °C, unopened shelf life commonly 24–36 months, and that hard 28-day post-puncture window. Ultrapure water has the opposite problem — it degrades fast in an open container. An 18.2 MΩ·cm aliquot left in an open beaker absorbs atmospheric CO₂ and can fall to 1–2 MΩ·cm within an hour. It is produced at the point of use for a reason.

When neither purchase is worth making

The inconvenient section, stated plainly.

Do not buy bacteriostatic water bundled at a reference-materials checkout if your work is analytical. At $0.83/mL it is among the highest-margin line items a supplier can attach to an order, and for any workflow ending at a UV detector or a mass spectrometer, the preservative is a contaminant you are paying a premium to introduce. A general laboratory distributor sells the identical compendial product, and Type I water from a bench polisher is free at the margin.

Do not buy ultrapure deionized water in bulk if you lack the storage discipline. A 10 L carboy of 18.2 MΩ·cm water is a depreciating asset. Ion exchange leaves no residual biocide, so a partially used container held at room temperature for weeks is microbiologically uncontrolled and its resistivity has already drifted. Buying a case to save 20% is usually a false economy.

Do not buy either without documentation you can actually read. A preserved-water vial with no lot number, no expiry, and no sterility statement is an unverifiable claim. If a supplier cannot produce the batch record, the preservative content is an assertion printed on a label. The same test applies to reference materials generally — see quality for what a usable certificate of analysis contains, and the library for how supplier documentation practices differ in the field.

Do not assume "sterile" and "pure" are the same axis. Bacteriostatic water is microbiologically controlled and chemically loaded. Ultrapure water is chemically clean and microbiologically uncontrolled. Neither dominates the other. A lab that needs both properties simultaneously needs a filtered, endotoxin-tested water product tested to both, not a substitution.

Do not treat distilled, deionized, and USP Purified as synonyms. Distillation removes non-volatiles and most microorganisms but passes volatile organics. Deionization removes ions and passes neutral organics and microbes. They fail differently, and a certificate that says only "distilled" tells you nothing about TOC.

Known failure modes worth naming: resistivity meters that read fine on contaminated loops; stoppers punctured past the 28-day window and used anyway; preserved water used as an HPLC blank, producing a "ghost peak" chased for days; and bulk DI water stored in HDPE for months, leaching plasticizer that shows up as an unidentified LC-MS ion series.

What this site does and does not provide

We supply documented reference materials for laboratory research use, with lot-traceable analytical records. We do not publish preparation, dilution, or handling protocols of any kind, and nothing here is intended for use in humans or animals — the terms are set out in research use only and the disclaimer. Questions about a specific lot's analytical record, batch traceability, or documentation should go through contact.

The practical takeaway: pick the water by the specification your protocol has to satisfy and by the detector at the end of the workflow. If the answer is a chromatogram, the preservative is a liability. If the answer is a microbiologically controlled multi-dose container with compendial documentation, resistivity is beside the point. Very few workflows genuinely need both, and the ones that do should be buying a product tested to both — not substituting one specification for the other and hoping the difference does not surface in the data.


Supplied strictly for in-vitro laboratory research by qualified researchers and institutions. Not a drug, food, cosmetic or dietary supplement. Not for human or veterinary consumption. Not evaluated by the FDA.

Every batch ships with its own certificate.

Purity, identity and lot number, documented for the exact vial you receive.

How we test