Calibration in the lab

 

See also chapter 14.D Qualifying laboratory instruments.

11 Definitions

In accordance with the EU GMP Guideline, Annex 18, calibration is defined as follows:

11.1 Persons

For each instrument, a system owner and a deputy must be named. The system owner is responsible for ensuring that the equipment is always in a faultless condition and for arranging the necessary maintenance and repair work. Finally, this person is also responsible for ensuring that the calibration is carried out at the planned times. He can also be consulted for training courses (equipment instruction).

The user is the person who works with the equipment temporarily. This person is responsible for proper operation of the equipment. In particular, he is obligated to check, before starting the analysis, that the equipment is calibrated and approved (see chapter 12 System suitability test (SST)).

11.2 Instruments

The essential functions are to be described in the instrument description. In addition, each piece of instrument is to be clearly labelled (instrument no., inventory no. etc.). If a piece of instrument consists of several components (instrument system), it must be ensured that it is possible at all times to track the individual components of which the instrument consists. This is most easily documented in the instrument log book. In practice, it has proven worthwhile to provide one component with an identification number to represent the entire system (e.g. auto-sampler in HPLC systems).

For each piece of instrument/each component, the critical parameters must be established. They form the basis for verifying proper operation (see chapter 12.1 Test intervals, test points, test instructions).

The test points are the critical functionalities of a piece of instrument, which are checked during calibration.

11.3 Working

Maintenance of analysis instrument is carried out regularly to ensure functional efficiency. Repair requires intervention in the equipment by the system owner, his deputy or a service technician, to restore functional efficiency.

Calibration of analysis instrument is the regular verification of the functional efficiency after start-up using a defined procedure. Usually, the result of this is the relationship of the measurement value to the true, known value of a standard used.

Adjusting of analysis equipment is the task of making the equipment ready for operation, with the elimination of systematic measurement deviations which could have a falsifying effect on the result. Adjusting requires intervention in the equipment.

Official verification is the operation in which the conformity of a measuring instrument with the calibration regulations is ascertained by the calibration authority. Stamping certifies that the measuring instrument complied with the verification regulations at the time of the test. This is an official operation.

12 Calibration instructions and record

Calibration of equipment is to be described in an SOP. The "umbrella SOP" basically specifies that all equipment used for GMP analyses must be calibrated.

In this SOP, it is advisable to define a generally valid calibration interval for all equipment for which there is initially no equipment-specific SOP.

Then, an individual SOP is to be compiled for each instrument or a set of instruments. Execution must be described accurately, in particular the calibrating equipment and reagents required must be listed therein. If special test techniques are used, these too must be specified.

The calibrations must be recorded and the documentation saved in the log book. In practice, the use of forms has proven useful for this purpose. It is advisable to include these in the SOP.

12.1 Test intervals, test points, test instructions

The equipment-specific SOP must contain specifications on the test interval, the test points and test instructions. For multi-component equipment, it is feasible that the test intervals for individual components will be different. The test points are derived from the description of the critical parameters (see chapter 14.D Qualifying laboratory instruments). The individual tests must be specified. Often they are already listed in the instrument manual or can be applied with slight changes. It is advantageous to describe the maintenance work in the same SOP.

In principle, calibration is always due if significant parts of a piece of equipment are replaced or if the equipment is moved to a new location. Re-qualification must also be taken into consideration (see chapter 14.D Qualifying laboratory instruments). Many calibrations can be covered by service agreements. This is a permissible procedure, for which the pros and cons must be weighed up against each other. External solution can be expensive and internal staff may lose in-depth equipment knowledge. In addition, it must be ensured that the external company implements the internal requirements. Above all, compliance with formal aspects (documentation) must be ensured.

In large companies, there are sometimes internal teams for calibration of equipment. Here, there is the risk that with this largely cost-effective solution, the individual employees will take their equipment responsibility slightly less seriously. The advantage of calibration by a system owner is justified by the better equipment knowledge and the higher level of responsibility. However, this is not always the most cost-effective solution.

13 Examples

13.1 Balance

The balances represent an essential element in the chain of analytical results. For each balance, extensive calibration must be carried out at least once a year. Parameters such as the following are checked:

• Measuring uncertainty, gives the permissible minimum initial weight via the repeated standard deviation (N = 10) (acc. to USP <41>)
• Analysis of the sample position
• Analysis of the linearity

This work can be processed well via service agreements.

Depending on the balance type, different procedures can be used for short-term (daily) and longer-term (annual) verification. A distinction is also made between tests that are carried out with built-in (automatic or manual) or external, certified control weights (manual). The balance calibration types are listed in figure 14.E-2 as an example:

It must be noted that calibration with external control weights is to be planned regularly - e.g. monthly - for all balance types (Note: the FDA has also already demanded daily verification here). It must be possible to justify the established procedure. For verification, it is also necessary to establish which control weights are used, preferably in the usual weighing range.

For some current balance types by Mettler-Toledo, the control weights are listed in the following overview (figure 14.E-3).

Taking calibration law and the recommendations of the balance manufacturer into account, many companies set tighter tolerance values than prescribed in the calibration law. For analytical balances, it is generally recommended to check the linearity with at least 2 weights. Here, only control weights for which there is a valid certificate may be used. For Mettler-Toledo control weights in accuracy class E2 and F1, the re-certification period is 5 years. For handling, the manufacturer's specifications must be followed, which prescribe, for example, special tweezers with gum mantles or suitable grease-free gloves. The control weights must be stored carefully and maintained in accordance with the manufacturer's specifications (degrease).

13.2 Volume measuring instruments

Calibrated glass pipettes and measuring flasks need not be calibrated. For automatic pipettes, a calibration interval of 6-12 months is usual. These control tests can also be assigned to third parties. Volume measuring instruments are usually checked gravimetrically. This is carried out on the Dilutor Hamilton Microlab 500, where the control test is executed once a year.

Critical parameters

• Accuracy
• Precision

Demineralised water, 0.01 N sulphuric acid and potassium dichromate p.a. are used (e.g. Fluka Item No. 1.04868). The dispenser function is checked gravimetrically, by weighing the aspirated and then ejected liquid volume on an analytical balance. The dilutor function is checked photometrically.

Dispenser function

The required volume is determined by weighing de-ionised water. The temperature of the water may only deviate from the ambient temperature by max. 2 °C. Calibration is carried out with a calibrated analytical balance. The temperature is measured with a certified thermometer.

5 ml of water are dosed in an Erlenmeyer flask and the atmosphere in the flask is saturated with water through repeated swivelling. The flask is placed on the analytical balance, tared and the desired volume is dosed. Here, with a variably adjustable volume, the volume of the flask/cylinder used must be checked as the full volume and as a partial volume of 50% and 10% of the full volume. Each fixed or variably adjustable volume amount must be ejected and weighed 7 times. The syringe is initially filled completely, whether or not the ejection is complete or partial. When determining partial volumes, the corresponding amounts are specified as aliquots. The displayed weight is printed out or noted. The syringe of the ejection pipe does not enter the liquid. The dispenser is operated at the standard speed (factory setting with specified parameters, e.g. syringe size).

During a measuring sequence, the Erlenmeyer flask must not be touched with bare hands. If the weighing range of the balance is exceeded during a measuring sequence, the flask is emptied, 5 ml water is dosed, it is retared and the measuring sequence is continued. For the 25 ml syringe, PE-tubes with a larger interior diameter must be used.

The following are calculated from the values measured per volume:

• the mean value
• the relative standard deviation (s_rel in%)
• and the recovery.

For the recovery and the mean value, the density of the water at the present measuring temperature must be taken into account.

Dilutor function

From stock solution A, a dilution series A is produced with the dilutor with a 250-mL sample syringe, 1 mL diluent syringe and standard sample tube.

From stock solution B, a dilution series B is produced with the dilutor with a 250-mL sample syringe, 1 mL diluent syringe and standard sample tube.

The solutions are measured against 0.01 N sulphuric acid with a UV/VIS spectral photometer at 350 nm in 0.5 cm half micro-cuvettes.

Requirements for the dilutor: For each dilution series, the fulfilment of the Lambert-Beer law is checked by determining the linearity (y = mx + b) and the regression (correlation coefficient R²), e.g. using an Excel sheet.

13.3 Photometer

The procedure is illustrated using the example of the UV/VIS Uvikon 922 spectral photometer by KONTRON. The calibration standard used is, for example, the secondary standard set for calibration of spectral photometers in the VIS range by HELLMA GmbH, Cat. No. 666-000. The methods are integrated in the equipment software of the spectrometer. The equipment is checked once a year. The equipment must be operated for at least 12 hours before calibration (conditioning of the bulbs).

Critical parameters

• Bulb intensity
• Abscissae values
• Ordinate values

Bulb intensity

• Intensity mode 200 nm (deuterium lamp adjustment)
• Intensity mode 580 nm (tungsten halogen lamp adjustment)

For the bulb intensity, the highest adjustable value is required. Setting the bulb intensity via the respective potentiometer up to the highest possible intensity (in accordance with user instructions). If the bulb intensity is too weak, it is set to the highest possible intensity via the respective potentiometer. If the intensity is still too weak, the bulb is replaced and then the bulb intensity is measured again with the new bulb.

Abscissae values

The standard filter set F1 of the secondary standard set for calibration of spectral photometers in the VIS range by HELLMA GmbH, cat. no. 666-000, is used.

The following settings are selected:

• Method: Wavelength Scan
• Ordinate scale: Absorption
• Slit width: 1 nm
• Scanning speed: 30 nm/min

Ordinate values

The standard filter set F2, F3 and F4 of the secondary standard set for calibration of spectral photometers in the VIS range by HELLMA GmbH, (cat. no. 666-000), is used, for example.

The following settings are selected:

• Method: Fixed Wavelength
• Ordinate scale: Absorption
• Slit width: 1 nm
• Adjustment time: 2 s

13.4 HPLC system

The calibration of complete systems, shown here using the example of an HP 1100 HPLC system, is more time-consuming. In such cases, the manufacturers often offer suitable software-based calibration packages. It is then up to the users to decide if the entire package should be used or only parts of it. If you are dealing with a mixed environment, i.e. equipment by different manufacturers, it may be worth determining the best procedure from all packages that covers all equipment regardless of the manufacturer.

The HPLC system consists of the following elements: pump system, degasser, autosampler with injector unit, column oven, detector unit and data analysis system.

The wavelength accuracy (Diode Array Detector = DAD or Variable Wavelength Detector = VWD), bulb intensity, temperature accuracy, temperature stability, noise, flow rates, injector precision and linearity, cross over, gradient mixer function (A/B and C/D), detector linearity and degasser function (vacuum) are checked as critical parameters.

Calibration should be performed annually and after significant changes in the system.

Before calibration, thorough maintenance of the entire system must be executed. The detector should be switched on approximately 24 hours in advance. The specifications defined for the individual tests are prescribed by the HP ChemStation software. The evaluation is carried out automatically by the software. All tests are described in detail in the "Operational Qualification/Performance Verification for Complete HP 1100 Series HPLC Systems" manual.

The tests are executed in the order shown in figure 14.E-7.

For calibration, the material listed in figure 14.E-8 is required.

Preparation

• Installation of the columns
• Eluent A: Water, Eluent B: Acetonitrile, condition columns with A:B 85:15 v/v (approx. 30 min)
• Switch to the "Verification" menu, select "Standard Tests" under Options and select all available tests with the "Edit" option. For each test, call up the "set up" and enter the limits and waiting time (see figure 14.E-9).
• The VWD wavelength test is called up under the "Enhanced Test" option and executed separately (easier to execute than in the "Standard Test" option).
• The gradient mixer function test is executed separately (see gradient composition check).
• Time required: Around 6-7 hours per piece of equipment must be scheduled.

  Figure 14.E-9 Waiting times and limits for HP 1100  

  Test	Limit	Waiting time
  Intensity	at max: 10000 at min: 5000 lower wavelength: 220 nm upper wavelength: 350 nm	0 min
  Holmium oxide test	Accuracy: 2 nm	0 min
  Temperature accuracy	left / right: 2 °C	0 min
  Wavelength accuracy VWD	at min (244.3 nm) at min (205 nm) at max (272.2 nm): 2 nm	No waiting time, test is executed independently
  Wavelength accuracy DAD	at min (244.3 nm) at max (272.2 nm): 2 nm	10 min, break before start
  Injector precision	Retention time: 1.5% RSD Peak area: 1% RSD Peak height: 2% RSD	0 min after DAD test, otherwise 10 min, break before start
  Injector linearity	Correlation: >0.999 RF Linearity: 5% RSD	0 min
  Detector linearity/cross over	Correlation < 0.999 RF linearity 5% RSD Cross over Area: 0.2% Height: 0.4% Quantity: 0.2%	5 min
  Background noise, flow rate, temperature stability	ASTM: 0.5 mAU Wander: 0.2 mAU Drift (DAD): 2.0 mAU/h Drift (VWD): 2.0 mAU/h Temperature stability: 0.5 °C Flow rate accuracy: 5%	10 min
  Gradient mixer function (only for HP1100 with degasser)	Waves: 0.5% Step accuracy: 0.7% Precision: 0.5%	30 min system conditioning after switching to isopropanol
  Degasser function		Degasser shuts down electronically in the case of a loss of power (warning light switches from green to red)

The following control tests figure 14.E-9 must be executed:

Time intervals for maintenance

14 Decision

14.1 Requirements, tolerances, specifications

For each test point, the corresponding acceptance criteria (requirements, tolerances, specifications) must be established. In practice, it has proven worthwhile discussing these with the manufacturer. They should be established "sensibly" in order not to unnecessarily restrict the handling leeway, and on the other hand in order to be able to set selective criteria. The manufacturer's documentation often contains limits that seem implausibly generous, on the basis of which precise working under GMP would not be possible.

14.2 Equipment release

If all test points comply with the acceptance criteria, the equipment is released by the system owner or by a person authorised by him until the next test time. To this end, an easily visible label is to be applied to the equipment, which specifies the test date, the test person and the date of the next test (month/year). In practice, a green approval sticker (so-called "TÜV sticker") has proven its worth. This can easily be procured and has the advantage that its meaning is easily understood.

14.3 Out of calibration

In the event of being out of calibration, the equipment must be rejected immediately. The equipment is to be clearly labelled accordingly, e.g. with a red "Rejected" sticker, labelled with the date/signature and this is to be noted in the equipment log book. Further measures for the equipment must be stipulated accurately. This can include repeating the calibration (re-calibration), alignment or other measures such as maintenance, repair or even disposal of the equipment. The further steps to be taken must be defined specific to the equipment. It is also necessary to check which analysis results obtained with the equipment are to be re-evaluated retrospectively. The measures must be documented.