Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
This standard practice, designated D 4515 – 85 (Reapproved 2001), establishes a rigorous method for estimating the maximum period that a water sample containing organic constituents can be stored after proper collection and preservation without compromising analytical accuracy. Section 1 of the standard emphasizes that holding time is highly dependent on the sample matrix and the specific analyte of interest. Therefore, samples from each unique source must be individually validated.
The practice explicitly excludes concentrations less than one order of magnitude above the method detection limit (Section 1.4) and does not cover degradation occurring between collection and initial analysis (Section 1.3). A critical definition provided in Section 3.2.2 is the maximum holding time: the storage period beyond which the systematic error of the test exceeds the 99% confidence interval, which must be limited to a maximum error of 15% relative to the mean concentration established at the zero-time analysis.
Section 4.1 outlines the procedure: a large volume of the target water sample is carefully collected, preserved, and homogenized. A portion is analyzed immediately to define the baseline (“zero time”) mean concentration. The remaining sample is stored under standard conditions and analyzed in replicate at discrete, predetermined time intervals.
Before the study begins, each laboratory must generate its own precision data for the specific analytical method. For methods with high inherent variability, additional replicates are required at each time interval to ensure the 99% confidence interval is maintained within 15% of the zero-time mean. The measured concentrations are plotted against time; the maximum holding time is the interval from collection until the systematic error exceeds this calculated statistical limit.
The following table summarizes the core limits and requirements defined by the standard for conducting a holding time study:
| 🟦 Parameter | 📐 Specification / Threshold |
|---|---|
| Baseline Reference | Mean concentration found at zero time |
| Confidence Limit | 99% Confidence Interval |
| Maximum Systematic Error | ≤ 15% of the zero-time mean concentration |
| Minimum Analyte Level | At least 1 order of magnitude above detection limit |
| Analytical Replicates | Sufficient to maintain the 99% CI within 15% of mean |
| Sample Matrix Rule | Must be validated per specific water source |
The practice also references several companion ASTM standards essential for the overall procedure:
| 🎯 Standard | 📏 Title / Function |
|---|---|
| D 1129 | Terminology Relating to Water |
| D 1193 | Specification for Reagent Water |
| D 2777 | Practice for Determination of Precision and Bias of Applicable Methods |
| D 3694 | Practices for Preparation of Sample Containers and for Preservation of Organic Constituents |
🔍 What exactly qualifies as a “maximum holding time”?
According to Section 3.2.2, the maximum holding time is the storage interval following proper preservation during which the systematic error of the analysis remains within the 99% confidence interval, with this error not exceeding 15% of the mean concentration measured at zero time.
💡 Does this practice apply to trace-level contaminants?
No. Section 1.4 explicitly excludes analytes present at concentrations less than one order of magnitude above the method detection limit, as the statistical model becomes less reliable at such low levels.
📌 Can laboratories use holding times established for a different water type?
No. The standard stresses that holding time is highly matrix-dependent (Section 1.2). Laboratories must perform their own holding time studies for each specific sample source and target analyte to ensure data integrity is maintained.
⚡ Why is the zero-time analysis so critical?
The zero-time measurement establishes the baseline mean concentration. This value is the sole benchmark used to calculate the 99% confidence interval limits. All subsequent degradation is measured against this initial snapshot of the sample (Section 4.1).