Thanks for your suggestion Bruce, it's just that I found everything is too focussed on pharma style validation and mostly for proteins and antibodies. The validation for those is a bit too hardcore for us as a developer/manufacturer of polymers for medical devices but I will continue to hunt for answers.
Last night I made a start on some ideas for a draft protocol which I pasted here since we can't attach files. Hey rick1112 or anyone else if it gives even one person just one idea that steers you in way that saves any time at all then great.
Anyone else if you can be bothered suggesting what I have missed or what I really don't need, your suggestions will be much appreciated. Remembering that it's for medical device polymers so there is some regulation to it, but nothing so fancy like antibodiesfor example.
So here it is, still a big mess but it's a start.
PURPOSE
To verify that the WatersAlliance 2695 GPC system performs as intended and provides reproducible results. XXXX What about something to state this is sort-of combined PQ and method validation?XXXX
BACKGROUND
The GPC system is used to estimate the molecular weight (MW) and to determine the polydispersity of polymers, prepolymers and other materials that are soluble in either DMF or THF. All instrument control, data capture and processing is performed by Empower 2 software.
Mobile phase carries the dissolved sample through a bank of heated columns consisting of one pre-column and three separating columns. The flow is maintained at a consistent nominal rate of 1.0 ml/min. Analytes are detected using a differential refractive index detector.
The MW is calculated from the retention time which relies on an accurate and consistent flow rate in order to generate meaningful results.
SCOPE
There are two GPC systems, one dedicated to running DMF with 0.05 M LiBr as the mobile phase and the other dedicated to running THF. Samples and standards are dissolved in mobile phase used on whichever system they are run on.
Flow rate accuracy and consistency which are by far the most crucial elements in GPC. Characteristics are listed below, along with a description of the tests that will be performed to validate them, or with a note explaining why they are not relevant.
Specificity
The bank of columns in use was chosen to be specific to molecules within the range of 100 to 600,000. A calibration curve is generated using polystyrene standards with an Mp (peak MW) within this range, from 390 to 419000 making this the range for accurate results. If a sample is outside this range it will be indicated as having an Mp below 390 or above 419000 and the result will be interpreted as being a best-estimate, since the calibration curve will have been extrapolated outside that range.
*Note that in special cases for the lower MW standards, if there is enough resolution of the monomer peaks they can be labeled and their Mp (peak MW) calculated so as to extend the lower MW limit, depending on the degree of resolution. The Senior Research Chemist (X we can give Timmy a cooler job title? XXX) will be consulted in these cases.
Accuracy
COMPANY requires only that GPC results describe the MW and polydispersity characteristics. Samples are not prepared in a quantitative manner as this is not necessary.
The flow rate will however be checked for accuracy, as will the column oven temperature accuracy.
Precision
Repeatability – Multiple injections will be made from multiple standards and the results compared for consistency. This will also serve as an injection volume repeatability check.
Intermediate Precision – A cut-down version of the repeatability check will be performed by multiple analysts on multiple days.
Reproducibility – XXXX This is the multiple lab comparison, I don’t know how to write something to exclude it seeing as COMPANY doesn’t have another site, at least not yet XXXX.
Detection limit
Not applicable to MW or polydispersity determination.
Quantitation limit
Not applicable to MW or polydispersity determination.
Linearity (line of best fit)
Linearity for GPC relates more to the ability of the bank of columns to retain molecules for a length of time that is proportional to their size. The proportionality within the specified range is approximately linear, but note this is not linearity as typically used for HPLC applications. Instead the line of best fit is used.
Polystyrene standards (11 in total) will be injected and a standard curve generated by plotting their size vs retention time.
Additionally, multiple standards will be dissolved in the same solution in order to verify that a valid calibration curve can be obtained using fewer injections.
Range
There are documented guidelines as detailed by the column supplier for the recommended concentration of samples. Where the MW is within the range that will be tested at COMPANY, the concentrations are listed below.
MW Concentration
Up to 25,000 <0.25 %
25,000 to 200,000 <0.1%
200,000 to 2,000,000 <0.05%
Often the expected molecular weight of a sample cannot be accurately known before testing. Within a certain range, the sample concentration has little effect on the MW or polydispersity result. Standards and samples will be prepared at a range of concentrations to determine the range of concentrations needed for accurate results.
Robustness
In order to generate accurate results, standards and samples tested by GPC will be run using the same fixed parameters, regardless of the sample. Parameters are listed below.
DMF Mobile Phase THF Mobile Phase
Mobile phase DMF with 0.05 M LiBr THF
Diluent DMF with 0.05 M LiBr THF
Columns Sytragel HT2, HT3 & HT4 Sytragel HT2, HT3 & HT4
Column temperature 70 °C 30 °C
Detector temperature 50 °C 30 °C
Flow rate 1.0 ml/min 1.0 ml/min
Injection volume 100 µl 100 µl
Samples must be run using the same parameters as the standards. For this reason, no changes to the parameters will be made unless for a special investigation. If this occurs then it will be fully documented, but for routine use all diluent and instrument parameters are fixed as described above.
MATERIALS AND EQUIPMENT
10 ml volumetric flask
Digital stopwatch
Digital thermometer with temperature probe
Polystyrene standards within the range 100 to 600,000 molecular weight
DMF with 0.05 M LiBr, filtered, for use as diluent and mobile phase
THF for use as diluent and mobile phase
5 ml disposable syringe and PTFE syringe filters
GPC vials and PTFE lined silicone septa caps
Polystyrene standards from Polymer Standards Service
Standard Mw Mn Mp Polydispersity
1 489 405 370 1.21
2 582 498 474 1.11
3 891 807 890 1.10
4 2780 2620 2770 1.06
5 6480 6240 6520 1.04
6 9000 9590 9890 1.07
7 18200 17900 18200 1.02
8 42300 41400 44200 1.02
9 115000 109000 120000 1.05
10 177000 174000 177000 1.02
11 434000 420000 419000 1.03
METHOD AND ACCEPTANCE CRITERIA
Specificity
Each of the 11 standards used to generate the standard curve must have an Mp within the range 100 to 600000 as specified by the column manufacturer.
*All samples must have a MW within the range covered by the standards.
Accuracy
Run the GPC system at 1.0 ml/min in XXXX mode (not recycle or purge mode I don’t know what mode that would be called XXXX) until the entire length of the waste line has been primed and is free of all bubbles. Switch the detector to purge mode then place the waste line into a 10 ml volumetric flask. When ready to begin, switch the detector to XXXX mode again and begin timing using a stopwatch. Record the time taken for the waste to reach the 10 ml mark of the volumetric flask.
*The waste solvent must reach the 10 ml mark within 600 ± 6 sec (10 mins ± 1 %).
Place a calibrated temperature probe in the column oven and record the temperature every 5 min for the duration of an injection.
*The temperature must be within 30 ± 1 °C or 70 ± 2 °C for all readings.
Precision
After generating a standard curve in the Linearity section, prepare additional solutions of 3 different polystyrene standards covering the MW range at the recommended concentration as previously described.
Repeatability – Inject each solution 5 times then calculate and record the Mp (Peak MW) results and the peak areas for each injection.
*The Mp for each injection of each standard must be within ± 5% of the value stated on the certificate and the RSD of the Mp for each standard must be not greater than 1.0 % *The RSD of the peak areas for each standard must be not greater than 1.0 %.
X1% is made up, yeah I said it but I can’t remember what you told me to say insteadXX
Intermediate Precision – Two other analysts will prepare their own set of the same 3 standards as used for repeatability and will inject them 3 times each on separate days.
*Acceptance criteria are the same as for repeatability.
Reproducibility – Don’t know what to write, see this section in the SCOPE.
Detection limit
Not applicable.
Quantitation limit
Not applicable.
Linearity (line of best fit)
Prepare a solution of each standard within the range XXX to XXX at the recommended concentration as previously described. Inject the standards in duplicate and generate a standard curve using whichever nth order fit is necessary.
*The regression of the line of best fit must have an R value of not less than 0.9999 and all of the residuals must be within ± 5.0 %.
Prepare solutions combining 2 standards that have as different MW as possible so that 6 vials are used instead of 11. Do the same again but combine 3 standards instead, so that 4 vials are used.
*Chromatograms of the multiple standards per vial must have baseline resolution of all standard peaks. The standard curve line of best fit must have an R value of not less than 0.9999 and all of the residuals must be within ± 5.0 %.
