Operational Qualification and

Performance Verification of
UV-visible Spectrophotometers

Technical Note

In principle, operational qualifica-

tion and performance verification
(OQ/PV) is not difficult to per-
form, because spectrophotome-
ters are relatively simple analyti-
cal instruments. However, in prac-
tice there are significant problems
because of the need to conform
closely to regulatory requirements
and because of the non-availability
of appropriate and easy-to-use
standards.

This technical note reviews the

regulatory requirements, the
advantages and disadvantages of
the standards currently available
and describes a new approach
which significantly reduces the
time required to perform OQ/PV.

Introduction mendations of pharmacopeias

Definition
In recent years quality require- have also become more influen-
tial. Verification of the continued The terms operational qualification and perfor-
ments, as outlined by ISO 9000, mance verification are used interchangeably in
GLP, GMP and NAMAS, have proper performance of UV-visible this publication. Both terms are used to mean a
assumed increasing importance spectrophotometers is an impor- series of tests designed to prove that an instru-
tant element of these quality ment is working within a set of required specifi-
and as a consequence, in the phar- cations.
maceutical industry, the recom- requirements.

Agilent Technologies
Innovating the HP Way
Regulatory requirements The statement intended and It is worthwhile noting that the
anticipated operating ranges is American Standard Testing
There are many regulatory critical when considering an Methods (ASTM) and the National
requirements that govern the use OQ/PV strategy. For example, if Institute of Standards and
of UV-visible spectrophotometers. the intended purpose is to mea- Technology (NIST) make similar
The following sections review the sure absorbance in the UV region statements. For example, the
most important of these. (as is the case for most pharma- ASTM states that the wavelengths
ceutical analyses), it is not appro- used for calibration should, if pos-
priate to verify photometric accu- sible, bracket the analytical wave-
Good laboratory practice and racy in the visible range—it must length and NIST comments on the
good manufacturing practice be done in the UV region and use of their various standards.
preferably at several wavelengths.
Good laboratory practice (GLP)
and good manufacturing practice
Similarly, it is not appropriate to
(GMP) requirements concerning
verify wavelength accuracy at the
the validation of instruments can
656.1 nm deuterium line because
be summarized as:
this is not a reliable indicator of
wavelength accuracy in the UV
“Documented verification that region. Further, as the vast majori-
the system or subsystem per- ty of samples to be measured in
forms as intended throughout the pharmaceutical laboratory are
representative or anticipated solutions contained in cuvettes, it
operating ranges.” 1 can be argued that the use of solid
or emission standards is inappro-
In reference to spectroscopy, the priate, because these standards
following is suggested: have different optical characteris-
tics from the samples that will be
“Where appropriate, periodic analyzed.
performance checks should be
carried out (for example, …
the resolution, alignment and
wavelength accuracy of spec-
trophotometers, etc.).” 2
United States Pharmacopeia3 The latter contains a cross-refer- The NIST4 provides a range of
ence: solid and liquid standards for
The regulatory requirements for determining wavelength accuracy,
UV-visible spectrophotometers are “For further details regarding photometric accuracy and stray
defined in the United States checks on both wavelength and light. The most important are sum-
Pharmacopeia (USP) XIII, Section photometric scales of a spec- marized in table 1.
851 on “Spectrophotometry and trophotometer, reference may
Light Scattering.” It states: be made to the following publi-
cations of the National
“Check the instrument for Institute of Science and
accuracy of calibration. … The Technology …”
wavelength scale may be cali-
brated also by means of suit-
able glass filters, which have
useful absorption bands
through the visible and ultravi-
olet regions. Standard glasses SRM# Type NIST Description
containing didymium (a mix- 930 Neutral density glass filters This SRM is for the verification and calibration
ture of praseodymium and of the transmittance and absorbance scales of
visible absorption spectrometers.
neodymium) have been widely
used. Glass containing holmi- 931 Cobalt and nickel solution in This SRM is for the verification and calibration
um is considered superior. nitric/perchloric acid mixture of the absorbance scales of ultraviolet and visible
absorption spectrometers having narrow band
passes.
For checking the photometric
935 Potassium dichromate solid for This SRM is for the verification and calibration
scale, a number of standard preparation of test solution of the transmittance and absorbance scales of
inorganic glass filters as well ultraviolet absorption spectrometers.
as standard solutions of
2031 Metal-on-quartz This SRM is for the verification and calibration
known transmittance such as of the transmittance and absorbance scales of
potassium chromate or potas- ultraviolet and visible absorption spectrometers.
sium dichromate are avail-
2034 Holmium oxide solution in This SRM is for use in the verification and
able.” perchloric acid calibration of the wavelength scale of ultraviolet
and visible absorption spectrometers having
nominal spectral bandwidths not exceeding 3 nm.

2032 Potassium iodide solid for This SRM is for use in the assessment of
preparation of test solution heterochromic stray radiant energy (stray light) in
ultraviolet absorption spectrophotometers.

Table 1
NIST standards
European Pharmacopoeia5
Control of wavelengths—Verify the wavelength scale using the
The European Pharmacopoeia absorption maxima of holmium perchlorate solution R, the line of
(EP) is based on the requirements a hydrogen or deuterium discharge lamp or the lines of a mercury
of national pharmacopoeias such vapor arc shown below. The permitted tolerance is ± 1 nm for the
as the British Pharmacopoeia (BP) UV range and ± 3 nm for the visible range.
and the Deutsche Arzneimittel-
buch (DAB) in Germany, and is 241.15 nm (Ho) 404.66 nm (Hg)
253.7 nm (Hg) 435.83 nm (Hg)
very specific in its requirements 287.15 nm (Ho) 486.0 nm (Db)
for UV-visible spectrophotome- 302.25 nm (Hg) 486.1 nm (Hb)
ters. The EP requirements are 313.16 nm (Hg) 536.3 nm (Ho)
334.15 nm (Hg) 546.07 nm (Hg)
described in the following sec- 361.5 nm (Ho) 576.96 nm (Hg)
tions. 365.48 nm (Hg) 579.07 nm (Hg)

Note that the British Control of absorbance—Check the absorbance using a solution of
Pharmacopoeia states that the potassium dichromate R at the wavelengths indicated in the follow-
ratio should “… be not less than ing table, which gives for each wavelength the exact values of the
1.5 unless otherwise specified in specific absorbance and the permitted limits. The tolerance for the
the monograph” and this is often absorbance is ± 0.001.
taken as a de facto minimum per-
formance specification for phar-
maceutical applications. Wavelength (nm) A (1 percent, 1 cm) Maximum Tolerance

235 124.5 122.9 to 125.2

257 144.0 142.4 to 145.7
313 48.6 47.0 to 50.3
350 106.6 104.9 to 108.2

Limit of stray light—Stray light may be detected at a given wave-

length with suitable filters or solutions. For example the
absorbance of a 1.2 percent m/V solution of potassium chloride R
in a 1-cm cell should be greater than two at 200 nm when com-
pared with water R as compensation liquid.

Resolution power—When prescribed in a monograph, measure

the resolution of the apparatus as follows: record the spectrum of a
0.02 % V/V solution of toluene R in hexane R. The minimum ratio
of the absorbance at the maximum at 269 nm to that at the mini-
mum at 266 nm is stated in the monograph.
Recommended OQ/PV test
Type of Test USP EP Comment
parameters and standards
Wavelength accuracy
• Deuterium discharge lamp 4 Only two peaks
When the above regulatory • Mercury discharge lamp 4
requirements are considered • Holmium oxide glass 4 Position of peaks may vary from batch to batch
and must be calibrated.
together there are many inconsis- • Holmium perchlorate solution 4 4 Peaks throughout UV and visible ranges.
tencies and, in some cases, dis-
Photometric accuracy
agreements. However, it is possi- • Neutral density glass filters 4 Blocks in UV range so only useable for visible
ble to define a strategy of tests range
and standards that essentially sat- • Metal on quartz filters 4 Often problems with interreflection errors and
very temperature sensitive
isfies all requirements. This is • Potassium dichromate solution 4 4 Has a tendency to decompose
illustrated in table 2 where the
implied USP methods (based on Stray light
• Sodium nitrite solution Not essential but gives a good data point in the
NIST standards) and the EP meth- mid-UV range at 340 nm
ods are summarized and then the • Potassium iodide solution 4 Use for 220 nm
GLP filter of suitability for pur- • Potassium chloride solution 4 Use for 200 nm
pose (liquid standards calibrated Resolution
in the UV range) is laid over the • Toluene in Hexane 4
matrix.
Table 2
Regulatory test requirements
In addition, although not specified
by the USP or EP, it is generally
accepted that to characterize
properly the performance of a The major problem in doing per- the performance characteristics
spectrophotometer the parameters formance verification is the fact have changed in a way which
noise, baseline flatness and stabili- that all of the above are wave- could affect the quality of the ana-
ty should also be tested. length dependent and, in the case lytical results.
of stray light, standard dependent
Linearity is also often considered as well. It is not practicable to per- It should be noted that compli-
to be an important factor for per- form all tests at all wavelengths so ance with the above criteria deter-
formance verification. However, in practice a few wavelengths, rep- mined with an appropriate set of
the primary instrumental reason resentative of the intended pur- reference standards does not nec-
for non-linearity is tested for. pose, should be used. The results essarily guarantee that a particular
Other reasons for non-linearity are of these test should be compared analysis can be performed with a
strongly sample or sampling to the instrument performance required accuracy and linearity.
dependent and for this reason it is specifications and/or to the instru- Since many parameters are sam-
the authors opinion that this para- mental performance requirements ple dependent this can only be
meter is best measured during a for the methods in use. demonstrated with an appropriate
system suitability test as Performance verification then system suitability test that uses
described below. effectively demonstrates whether the sample itself.
Practical aspects
Solid Standards Liquid Standards Comment
Handling Straightforward, but user Normally prepared by user Straightforward, but user
Whilst liquid standards have the must take care to avoid using pure chemicals, must take care to avoid
advantage that an appropriate set dust or scratching. calibrated balance and dust or scratching.
can fulfill the regulatory require- volumetric flasks.
User requires some skill to
ments as outlined above they do eliminate cross contamina-
have disadvantages when com- tion. Using liquid standards
pared to solid standards. Some is to some extent a test of
proper sample handling.
attempts have been made to
reduce these disadvantages by Calibration User must calibrate each If prepared accurately, Because the cuvette con-
sealing liquid standards into standard individually. these are absolute taining the sample is also a
standards that do not factor, each standard must
cuvettes but the results are not require calibration. be calibrated individually.
entirely satisfactory. The relative
advantages and disadvantages of Stability Good, but require recalib- If not properly stored, Generally good, but some
ration at 1–2 years. some standards show (potassium dichromate)
the different types of standards significant instability. are less stable than solid
are compared in table 5. Standards are best pre- standards. User may need
pared fresh for each test. to recalibrate at 6-monthly
intervals.

Cost High, because each stan- Can be low, because High, because each standard
dard must be individually chemical costs are must be individually calibra-
calibrated and then recali- minimal and no calibra- ted and then recalibrated at
brated at intervals. tion is required. Can also intervals.
be prepared in bulk.

Table 5
Comparison of practical aspects of different standard types

A practical solution:
Standard Blank Test
liquid standards in ampules
40 g/l holmium oxide in 10 % perchloric acid Wavelength accuracy
10 % perchloric acid
The problems associated with the
preparation and use of liquid stan- 60 mg/l potassium dichromate in 0.01 N sulfuric acid Photometric accuracy
0.01 N sulfuric acid
dards have been virtually eliminat-
ed by making them available as 50 g/l sodium nitrite in water Water Stray light at 340 nm
prepared solutions in sealed
10 g/l sodium iodide in water Water Stray light at 220 nm
ampules. To use them it is only
necessary to break open the 12 g/l potassium chloride in water Water Stray light at 200 nm
ampule, pour in the curette and
0.02 % v/v toluene in hexane Hexane Resolution
measure. The contents of the stan-
dards kit and the purpose of each Table 6
standard are summarized in Contents of liquid standards kit from Agilent Technologies
table 6.
The only additional requirement is Additional test recommenda- In practice users have developed
water as blank for the non-critical tions their own strategies for perform-
stray-light tests, all other refer- ing system suitability. For exam-
ence solvents are included. The To ensure continued proper opera- ple:
advantages of these standards are tion between OQ/PV testing, and
that they: to ensure suitability of the system • Measure and calibrate using
for the specific analyses being per- one standard with a concentra-
• save time because no prepara- formed, the following additional tion equal to 100 percent of the
tion is required tests are recommended. expected component concen-
tration. Then measure and
• eliminate false OQ/PV failures Instrument selftest quantify the standard and the
due to errors in preparation The Agilent ChemStation software standard diluted by a factor
for UV-visible spectroscopy of 2. The results of both sam-
• are inexpensive compared to includes a set of self-test routines ples should be within a speci-
other calibration standards that check proper electronic and fied percentage of the known
optical operation of the spec- concentration. Remeasuring
trophotometer as well as wave- the standard demonstrates the
• include a certificate of analysis
length accuracy checks with the quality of the initial measure-
for traceability, and
two lines at 486.1 and 656.0 nm ment.

• have been shown to have from the deuterium lamp.

• Measure the standard and then
excellent stability.
System suitability measure a series of dilutions of
System suitability should not be the standard and calculate the
Errors due to poor sample han- extinction coefficient
dling can still occur but, if desired, confused with method validation.
System suitability is designed to (absorbance/concentration)
these can be significantly reduced for each concentration. The
by using a flow cell instead of a evaluate the components of the
analytical system to show that the values of the extinction coeffi-
standard cuvette. The test solu- cients should not vary by more
tions are drawn into the cell using performance of the system meets
the standards required by the than a specified percentage.
a syringe and the potential for
cross contamination is greatly method. While method validation
reduced. A further advantage of is performed once at the end of
this procedure is that it saves con- method development, system suit-
siderable time compared to using ability tests are performed on a
a standard cuvette which must be given system periodically to deter-
rinsed carefully several times mine its adequacy or effective-
between standards. Test have ness. System suitability require-
shown that a full OQ/PV including ments for chromatography sys-
noise and baseline flatness tests tems have been well defined but
can be performed in about 30 min- no similar definition exists yet for
utes on the Agilent 8453 UV-visible UV-visible spectroscopy systems.
spectrophotometer and its perfor-
mance verification software.

Because diode-array spectropho-

tometers such as the Agilent 8453
have excellent stability and relia-
bility it is only necessary to repeat
the OQ/PV tests at 6–12-month
intervals.
Summary References 4
National Institute of Standards
Performing OQ/PV that exactly 1 and Technology, Standard
meets international regulatory Alford, J.S., F.L. Cline, F.L., Reference Materials Catalog,
requirements and GLP has, until “PMA’s Computer System Section 204, Optical Properties,
now, been very time consuming Validation Committee, Computer pages 104–106, 1995,
and prone to errors. Using liquid System Validation—Staying Gaithersburg, MD, USA
standards in ampules virtually Current,: Installation
eliminates these problems. Qualification,” Pharmaceutical 5
Technology, 1990, September, European Pharmacopoeia, 3rd
88–104 Edition, Section 2.2.25,
“Absorption Spectrophotometry,
2 Ultraviolet and Visible,” pages
“EURACHEM Guidance Document 28–29, 1996, Strasbourg, France
No. 1/WELAC Guidance
Document No. WGD 2:
Accreditation for chemical labora-
tories: Guidance on the interpreta-
tion of the EN 45000 series of
standards and ISO/IEC Guide 25,”
1993

3
United States Pharmacopeia
XXIII/National Formulary XVIII,
“Spectrophotometry and light-
scattering,” General Chapter 851,
page 1830, 1996, The United
States Pharmacopoeia
Convention, Inc., Rockville, MD,
USA

For the latest information and services visit

our world wide web site:
http://www.agilent.com/chem

Copyright © 1997 Agilent Technologies

All Rights Reserved. Reproduction, adaptation
or translation without prior written permission
is prohibited, except as allowed under the
copyright laws.

Publication Number 5965-7438E

Agilent Technologies
Innovating the HP Way