TOC is the number that tells you a water system is working before anything else does. It's non-specific, it's fast, and it's sensitive enough that a drift in the trend is a conversation — not a deviation. Here's how the test actually works, what the acceptance value is, and what the analyzer has to hold to keep the numbers honest.
Why TOC is the watch parameter
Total organic carbon is a non-specific indicator. That's the right way to start, because it sounds like a weakness and it isn't. TOC doesn't tell you what is in the water. It tells you that something organic is there — from natural compounds, from what the process adds, from biological growth — without you having to identify a single molecule.
For a water system, that's the right kind of information. You're not trying to catch one contaminant. You're trying to catch the water system changing. And the earliest, cheapest, most general signal of change you have is the total amount of carbon moving through it. A microbiology result takes days. A TOC number takes minutes. The trend is the instrument, and the number is the reading.
The number you're defending: 500 ppb
The pharmacopoeial acceptance value for total organic carbon in water for injection and purified water is 500 ppb. That's the line the batch record is written against, and it's the number your analyzer has to be able to report with confidence — well below it, on a routine day, and right around it, on the day the system is having problems.
And the value matters less than the trend. A single reading at 480 ppb is a number. A water system that was at 120 ppb for a year and is now at 480 ppb is a water system that is changing, and the TOC program is how you find out in the current batch instead of in the investigation. The analyzer's job is to make the trend honest: same method, same conditions, same instrument, every time.
How a TOC number actually gets made
Here's the mechanism, and the trap inside it.
TOC is measured by destroying the organic carbon in a sample and detecting the product. The mainstream approach is oxidation — persulfate combined with UV, for example, which is the method behind Standard Methods 5310C and EPA Method 415.3, and is what ISO 8245 covers for TOC determination in water. The organic carbon oxidizes to CO2, and the instrument detects the change.
The trap is that the sample also carries inorganic carbon — carbonates and bicarbonates. That inorganic fraction (TIC) is not organic, but it ends up as CO2 too, and it will read as TOC unless the instrument deals with it. So a TOC analyzer has to remove or quantify the inorganic fraction before the organic fraction gets measured. The quality of that separation is a big part of the quality of the number.
That's why the test isn't "run the sample." It's a sequence: sample in, purge, oxidize, measure — and the sequence is what the method describes. Change the sequence and you've changed the test.
The three-stage determination
For purified water, the pharmacopoeial approach runs three stages on the same sample, and each stage answers a different question.
Pre-treatment TOC — the total carbon in the sample as it arrives. Organic plus inorganic. This is the raw number.
Post-purge TOC — after the inorganic carbon has been stripped off, the carbon that remains is the organic fraction. The gap between pre-treatment and post-purge is the inorganic load, and a big gap tells you the sample carried a lot of carbonate.
Post-treatment TOC — after oxidation, what's left. The result of the test is this stage, and it's the number that goes against the acceptance value.
The three stages together are what make the number defensible. A result with all three is a result you can explain. A result with only the final number is a result someone else has to take on faith.
What an analyzer has to hold
The acceptance value is 500 ppb. That tells you the spec floor for the instrument, and it's lower than people expect.
Detection limit with headroom. You don't want the instrument's detection limit sitting at your acceptance value — you want an order of magnitude under it, so that a normal result is a real measurement and not noise near the floor. The MedIntegrity TA-100 specifies a detection limit of 5 µg/L, which is 5 ppb. A routine 100–200 ppb result is measured well inside the instrument's honest range. The TA-200's specified limit is 50 µg/L — fine for the system-monitoring range, a different instrument for a different job.
Range. The TA-100 runs 0 to 1000 mg/L, extensible; the TA-200 runs 0 to 3000 mg/L, extensible. The water-monitoring work lives in the low hundreds of ppb. The range exists so the same instrument can also handle process water, wastewater, or a spiked reference without a change of method.
Reproducibility. A trend is only a trend if the same sample gives the same answer twice. The TA-100 and TA-200 are both specified at ±3% reproducibility. On a 200 ppb result that's about 6 ppb of repeat-to-repeat spread — small enough that the trend line is the signal, not the noise.
Response time. Water systems get tested a lot. The TA-200 is specified under 4 minutes per determination, and the TA-100 within 4 minutes. When a system has points at several locations and a shift has a window, four minutes per sample is the difference between finishing the rounds and losing the window.
Sample conditions. Real water isn't always dilute. The TA-100 is specified for salinity up to 85 g/L and runs on nitrogen at 99.995% purity. The nitrogen matters: the purge line is part of the method, and a gas with organics in it adds carbon to the sample. The instrument that can't hold its own blank can't report your trend.
Records, and the audit trail behind the number
A TOC result is a regulated number. It lives in a water system program that has to show the data integrity chain: who ran it, on what instrument, with what method, and what the result was — for years, because the trend is the point.
So the instrument side has to carry the record. The TA family is specified with full 21 CFR Part 11 compliance, conditions developed in accordance with GAMP guidelines, and the IQ/OQ/PQ qualification documentation package included with the instrument. The TA-100 stores at least five years of history; the TA-200 carries an electronic testing record. The records leave the instrument by USB or to a PC, so the result can get into the LIMS in the shape your quality system reads.
The qualification package is the part that keeps the number auditable. The instrument that arrives is the instrument the validation file describes — IQ, OQ, PQ, done, and in the file. That's the sentence the auditor reads when the water system trend is being examined.
When a number moves
A TOC excursion is a procedure, not a panic.
First, confirm the method. Run a reference — a known value, a system blank — through the same instrument the same day. If the reference comes back right, the instrument is honest and the water is the story. If the reference is off, the instrument is the story, and the water gets retested, not released.
Then look at the obvious causes: the purge line, the nitrogen, a change in the treatment upstream, a point in the distribution loop that's been sitting. Retest at the affected point. A single high value is a retest. A rising trend across points is a system event, and the investigation starts now, in the current batch.
The rule that keeps this sane: the trend decides, the single value doesn't. One high number, method confirmed, system normal — retest and move on. Three numbers drifting up across a month, system normal on paper — that's the water system telling you something, and the TOC program is how you catch it before it's a water for injection problem.
Lab and semi-online
The TOC work sits in two places. The lab, where the daily and periodic monitoring runs and the numbers are reduced into the trend. And the process, where semi-online or at-line measurement takes the sampling out of the loop — the water runs, a sample is drawn automatically, and the number is there before the batch record is signed.
The TA family covers both, with automated sampling options for the at-line work. The instrument choice follows the monitoring program: the points you test, the frequency, and whether the sample comes to the instrument or the instrument comes to the sample.
Standards and sources
This guide draws on the primary documents below. Read them with your water system monitoring program open — the gap between the two is usually where the program lives.
USP <643> — Water for Injection · ISO 8245 — Water Quality — Determination of Total Organic Carbon · EPA Method 415.3 — Organic Carbon, Total · FDA Guidance for Industry — Water System Monitoring
Specifying a TOC analyzer for a water program?
Tell us your monitoring points and your method — we'll come back with configuration, qualification documentation and records integration.
Talk to Our Team