Reconstitution Water Where Type I Is Not Available: Sourcing and Assessing a Substitute

Where a Type I polishing unit is out of reach, a laboratory has four realistic sources of water: sealed pharmacopoeial sterile water bought through a hospital or pharmacy wholesaler, bottled reagent-grade water imported through a laboratory distributor, water dispensed by a neighbouring institution that does run an ultrapure system, and whatever a point-of-use unit of its own can produce. Which one is right is decided by three buyer-side questions — what can actually be obtained on a reliable cycle, what the label can be shown to mean, and how the water is kept after the seal is broken. The chemistry of the grades themselves belongs to the reference literature, and a reader who needs what each laboratory water grade actually specifies should take it from there; this article stays on sourcing and assessment.

The problem is specific to markets without a laboratory-supply chain. In much of the continent, the word 'distilled' on a shop shelf describes battery top-up water, and a bottle labelled 'Type I' may have no standard, no lot number and no date behind it. Bacteriostatic water, which is a registered medicine rather than a laboratory reagent, has its own supply problem and its own article in this cluster.

What the grades mean, and which properties matter to a buyer

Two specification systems are in circulation and their numbering does not correspond. ISO 3696 defines three grades for the analysis of inorganic chemicals and states that it does not apply to water for organic trace analysis or for biological or medical analysis ^[1]. ASTM D1193 defines four reagent-water types with separate microbiological grades, and its current revision allows each type to be produced by any technology provided the constituent limits are met ^[2]. For a buyer, that has one practical consequence: the process named on a label tells you very little, and the grade number tells you nothing unless the standard is named alongside it.

Three properties are worth asking a supplier about, because they fail independently: ionic content, reported as conductivity or resistivity; organic content, reported as total organic carbon; and biological content, reported as a microbial count and, where the downstream work is biological, an endotoxin limit. A supplier who can quote only the first has told you about one of three.

What is realistically obtainable, and how it is labelled locally

The first row is the one people overlook. Pharmacopoeial water is made under good manufacturing practice, where the quality claim attaches to a qualified, monitored production system rather than to an individual bottle ^[3]. That is the reason a sealed pack from a medicines wholesaler can be a better answer than an unlabelled carboy from a laboratory reseller: the documentation chain is regulated, the pack has a lot number and an expiry, and the regulator's register can be checked against it. What it does not give you is volume at a low price, or any assurance once the seal is broken.

Assessing an unfamiliar supply

Treat a new source the way you would treat any new supplier: document first, measure second, and accept nothing on appearance. A few checks cost little and catch most of the problems.

1. Read the label for a named standard, a grade or type, a lot number and a date. A grade with no standard, or a product with no lot, is a claim with nothing behind it.
2. Ask for the certificate for that lot. A supplier claiming conformity to ISO 3696 should be able to show results for the parameters the standard specifies, which include pH, conductivity, oxidisable matter, absorbance, evaporation residue and silica ^[1].
3. Measure conductivity on opening with a meter you have checked against a reference solution, and write the reading into the receipt record with the lot number.
4. Inspect the container: material, seal, whether it has been refilled, and whether the bottle has sat in sunlight on a shop floor.
5. Record where it was bought and in what condition, so a later problem can be traced to a lot and a source rather than to 'the water'.

Be clear about what the meter can and cannot do. A high conductivity reading is a reason to reject a supply. A low one is not a reason to accept it, because most organic contaminants are electrically neutral and bacteria do not register at all; a deionisation system can deliver excellent resistivity while its resin bed hosts a growing microbial population ^[4]. For biological work the concern that matters most is endotoxin, which outlives the bacteria that shed it, and no bench meter will show it. That is a documentation question, answered by the supplier's certificate or by a sealed pharmacopoeial product, not by a reading.

Point-of-use purification: what it can and cannot fix

A small cartridge deioniser or reverse-osmosis unit fed from mains water is the usual next step, and it does solve the ionic problem well. It does not, on its own, solve the organic or the biological one: ion-exchange resin removes no organic carbon, and a resin bed left standing is a surface for bacterial growth ^[4]. The good-manufacturing-practice guidance on pharmaceutical water treats stagnation, dead legs and irregular sanitisation as the recurring causes of microbial failure in any water system, however well it was specified on the day it was installed ^[3].

The less obvious cost is supply. Every cartridge, filter and ultraviolet lamp in the unit is an imported consumable with its own lead time, and a purifier with an exhausted cartridge is worse than no purifier, because it still dispenses water that looks and measures like the real thing until the conductivity alarm is noticed. Treat the replacement parts as critical stock and plan them the way you plan anything else you cannot buy next week.

Power is the other constraint. A unit that has to run to produce water produces nothing during an outage, and neither does its conductivity monitor. In South Africa the position has improved: Eskom last implemented load-shedding on 15 May 2025 and completed a full year without it in May 2026 ^[5], and its April 2026 winter outlook projected continued stability while flagging a critical supply-adequacy risk between 2029 and 2030 if new generating capacity is delayed ^[6]. Elsewhere, outages remain a routine operating condition for a large share of firms ^[7]. Either way, plan production for the hours when supply is dependable, and do not respond to an unreliable grid by producing ultrapure water in bulk and storing it, which trades one problem for a worse one.

Storage and recontamination in a warm climate

Water of any grade starts to deteriorate from the moment it leaves its production system or its sealed pack. It takes up carbon dioxide from the air, leaches from its container, and — the part that matters most in a 30 °C room — gives any microorganism introduced on opening a warm, nutrient-poor but perfectly adequate place to grow. The pharmaceutical guidance is built around the same principle: microbial quality has to be controlled through storage and distribution, not only at the point of production ^[3].

• Buy sealed sterile water in the smallest pack that suits a working session, rather than a large container that stays open for weeks.
• Write the date and time of opening on the container, and discard on a rule you have written down rather than on appearance.
• Keep stock out of direct sun and off the floor of an uncooled store room; the label's storage condition is part of the product's specification.
• Never top up or decant into a container that previously held something else, and never return unused water to its source bottle.
• Record the source, lot and opening date against every preparation made with it.

When the water is the limiting factor on the work

For much routine work the water is not the weak link, and spending on it is money taken from something that is. Rinsing, water baths and autoclave feed need clean water, not ultrapure water. The cases where the water becomes the limiting factor are narrower and predictable: trace analysis and mass spectrometry, where dissolved organics and ions appear in the result; cell-based work, where endotoxin does; and any preparation that has to remain sterile, where only a sealed sterile product will do.

Where the work falls into one of those categories and no verifiable water is available, there are three honest options. Bring in a small quantity of certified water with the next consignment and reserve it for the steps that need it. Arrange access to a neighbouring institution's system and collect close to the time of use. Or proceed, record the water as an uncontrolled variable with its source and lot, and interpret the results with that limitation stated. What is not defensible is using an unverified local product and recording it as though it met a grade it was never shown to meet.