USP Purified Water: Standard, Specs & Generation

Chemically they are held to the same limits: the same conductivity requirement under USP general chapter 645 and the same total organic carbon limit of 500 parts per billion under 643. Water for Injection adds an endotoxin limit of 0.25 endotoxin units per milliliter and carries stricter expectations around production, storage and distribution. The practical consequence is that WFI is required where the product is parenteral or where a surface contacting a parenteral is given its final rinse, and Purified Water is appropriate for a great deal of oral, topical and non-sterile manufacture.

For oral dosage forms Purified Water is normally the correct grade, and specifying Water for Injection there is one of the more expensive habits in the industry. The monograph requires only that it be made from water meeting drinking water regulations by a suitable process, so the train follows the feed rather than a prescribed method: typically softening or antiscalant dosing, carbon or chemical reduction to remove chlorine and chloramine, reverse osmosis, then electrodeionization or mixed bed polishing. Feed water analysis across more than one season should drive that selection.

The monograph itself does not set a microbial limit for Purified Water, which surprises people. What exists is a recommended action level in USP general chapter 1231, generally 100 colony forming units per milliliter, and the expectation that you establish your own alert and action levels from your system history and your process risk. That is a meaningful distinction: exceeding 100 is not automatically an out-of-specification result, it is a trigger for investigation against limits you have justified. Water for Injection is handled the same way, at a considerably tighter 10 per 100 milliliters.

Control it at three points rather than one. In pretreatment, by sizing carbon for the actual organic and chloramine load and replacing it on performance rather than on a calendar. In generation, with reverse osmosis for bulk rejection and ultraviolet at 185 nanometres, which oxidizes organics and is a genuine total organic carbon reduction step, distinct from the 254 nanometre lamps used for disinfection. In distribution, by preventing biofilm, since a shedding biofilm is itself an organic source, and by choosing elastomers and filter materials whose extractables will not contribute.

It can be stored, and most systems do, but storage is where control is won or lost. A storage tank needs continuous circulation back through the loop rather than sitting static, a spray ball wetting the headspace so the upper surfaces are not left as a growth site, and a hydrophobic vent filter that is integrity tested and, where the tank is hot or steam sanitized, heated to prevent condensation blinding it. Sizing matters too: a tank far larger than demand turns over slowly and leaves water standing, which works against microbial control even when everything else is correct.

Through the three phase qualification program, the same structure used for Water for Injection. Phase one runs typically two to four weeks with daily sampling at every point of use, establishing that the system can consistently produce specification water and defining operating ranges. Phase two repeats over a similar span at reduced frequency to confirm consistency, and production is normally permitted after it. Phase three continues routine monitoring for a full year to capture seasonal feed water variation. Laboratory incubation sets the pace, so the schedule cannot be shortened with more people.

Conductivity is assessed in stages, and a Stage 1 in-line failure is not by itself out of specification. Stage 2 requires an off-line measurement at ambient temperature with the sample stirred to equilibrium against a limit of 2.1 microsiemens per centimeter, and Stage 3 applies pH-dependent limits after adding potassium chloride. The commonest cause of a Stage 1 failure that then passes Stage 2 is dissolved carbon dioxide, which depresses the reading without indicating ionic contamination. Genuine failures usually mean resin exhaustion, reverse osmosis membrane breakthrough or a failing electrodeionization stack.

No. Hot circulation at 80 degrees C or above is one way to control microbial growth, not a requirement of the monograph, and many Purified Water loops run at ambient temperature with ozone sanitization and ultraviolet destruction before each point of use. The trade is real: hot loops are forgiving of imperfect operational discipline but consume energy continuously and rouge considerably faster, so they need derouging sooner. Ambient ozonated loops use a fraction of the energy and rouge far less, at the cost of destruct units to monitor and maintain.

Yes, and that is the normal arrangement. The Water for Injection monograph requires the feed to meet drinking water regulations as a minimum, and in practice a Purified Water system is the usual feed to a still or a membrane WFI unit because presenting already-purified water reduces the load and improves reliability. Where this is done, remember the Purified Water loop becomes part of the WFI system boundary for qualification purposes: an excursion in the feed loop is a WFI system event, and the change control obligations extend across both.

The water quality comes from the USP monograph, with conductivity under general chapter 645 and total organic carbon under 643, plus the European Pharmacopoeia where product is supplied into Europe. USP general chapter 1231 provides the supporting guidance on system design, sanitization, monitoring and the establishment of alert and action levels. The hardware is a separate matter: ASME BPE governs surface finish, drainability, joining and documentation, ASME B31.3 the pressure piping design, and ASTM A967 the passivation. The facility quality system sits under 21 CFR 211.

A 10 gallon per minute generation package typically runs 300,000 to 800,000 dollars installed depending on feed water quality, pretreatment scope and whether hot sanitization is included. Distribution is priced separately at roughly 255 to 630 dollars per linear foot, and on most projects the loop exceeds the generation equipment, particularly where the point count is high. Point of use assemblies add 2,800 to 10,200 dollars each. Three phase qualification adds 58,000 to 195,000 dollars and scales with point count rather than with system size.

Yes, working nationwide from Kilmarnock, Virginia. We do not operate regional branches, so crews mobilize for planned work and dates are confirmed at quotation rather than promised as local availability. On a Purified Water project the front-end constraint is usually equipment lead time rather than crew scheduling, and most of the distribution weld volume can be prefabricated as shop spools before anyone travels. The qualification tail at the other end is governed by laboratory incubation, which is unaffected by where the contractor is based.

The monograph does not prescribe a method, only that it be produced from water meeting drinking water regulations by a suitable process and that it meet the specified limits. In practice the train is selected from the feed analysis. A typical sequence is multimedia or cartridge prefiltration, softening or antiscalant dosing to protect membranes from hardness, carbon or chemical reduction to remove chlorine and chloramine, reverse osmosis for bulk removal of ions, organics and microorganisms, then electrodeionization or mixed bed deionization to polish, with ultraviolet and final filtration as the design requires.

No. The Purified Water monograph sets conductivity and total organic carbon requirements but no endotoxin limit; that limit, 0.25 endotoxin units per milliliter, belongs to Water for Injection. Two caveats matter in practice. If your Purified Water feeds a WFI system, its microbial condition directly affects the endotoxin load the downstream unit has to handle, so it is often monitored anyway. And if a customer specification or your own risk assessment requires endotoxin control for a particular application, that becomes your commitment even though the monograph does not demand it.

Sanitization frequency should follow monitoring data rather than a fixed calendar, and the right interval depends on how the loop controls growth. A hot loop circulating at 80 degrees C or above suppresses growth continuously and needs periodic thermal cycling mainly as a precaution. An ambient loop sanitized with ozone or with periodic hot water typically runs a defined cycle daily or weekly, set during qualification and then confirmed by trending. The signal that an interval is wrong is escalation: if counts only stay in range by sanitizing more often or more aggressively, the cause is usually geometry rather than frequency.