raman specroscopy fithrul farmarind

Functional Group Analysis

To aid in fu nctional group an alysis, the following chart
illustrates se lect molec ul a r bonds and [heir co rresp ond ing
Raman a nd FTIR spectral characteristics. This in fo rmation
is designed to aid adva nced users with chemical identification,
and should be used in conjunction with responders ' expertise
and exper ience as well as additi ona l respon se rools.

Th e cha rt includes th e chemica l bond , the range at w hich
a bond would appea r on a spec[ral chan, th e functiona l
gro up and sample substances tha[ fa ll wi [hin [h at gro up.
For ec-;:a mpl e, FllR ana lysis oi alco ho ls a nd water would
show a characteristic broad peak between 2, 800 and
3,100 wavenumbcrs based o n the 0-H bond.

Wavenumbers
4000
Chemical names in the chan
a1e examples of t/1e tuncrional
g1 oup described

3500

0-H lbroad peak);
Alcohols and water
Ethanol
lsopropanol

S

N-H lnarrow peak(sll .
Amines, amides
- - - - - - Propylylamine
Benzamide
3000
C-H; Hydrocarbons
Hexane
1-octene
Toluene
S-H; Thiols

1-Hexanethiol
C=C; Alkynes
Hexyne
C=N: Cyanides
Potassium cyanide
Cyanobenzene
Q)

a:

c:

(ti

E
(ti
a:

2500

C=C: Alkynes
_ J Hexyne
C=N; Cyanides
- - - - - Potassium cyanide
Cyanobenzene

:1----

__

CS; Organic Sulpher
Compounds
Ethanethiol

N02; Organic
Nitrate Compounds
Nitropropane

j
1500

I
Ｍｾ＠

ｾ＠

C-C; Aromatics
Benzene

C-Br; Bromated
Organics
Bromoform
C-1; Iodated Organics
Ethyl iodide

Inorganic nitrate
lbroad)
Ammonium nitrate

ｾＭＡ＠

1000
0-0; Peroxides
Hydrogen peroxide

J

__J

N02; Organic Nitro
Compounds
Nitromethane

C-0; Alcohols.
- - Ethers, Esters
Ethyl ether

Inorganic Nitrate
Potassium nitrate
C-CI; Chorinated

Organics
Trichloroethylene

C=O: Ketones.
Aldehydes, Caboxylic
acids, Esters
Acetone
Benzaldehyde
Acetic acid
Ethyl acetate

2000

C=C; Alkenes
Hcptene
1-nonene

°'
c:
(ti

_ _ _ _ _ C-H; Hydrocarbons
Hexane
1-oclene
Toluene

ｾ＠

C-H: Unstaturated
- - - Hydrocarbons

il
r

［ｾＺｩ･ｮ＠
I
L________J

C-F; Fluorinated
Organics

Perfluoroctane
Inorganic
phosphate lbroadl
Sodium phosphate
C-CI; Chlorinated
Organics
Chloroform

500
Metal-0; Metal
Oxides

Raman Spectroscopy
• Spectral Range. 250 cm"' to 2875 cm·' • Doesn't require direct contact
• Spectral Resolution: 7cm"' to 10 crrr' • Scans through translucent contaioer'
• Laser Output 2500 mW to 350 mW
• Easily analyies aqueous solutions
• Tool of choice for HME identification

FTIR Spectroscopy

• Spectral Range. 650 cm"' to 4,000 cm"' • No sample color limitation
• Spectral Resolution: 4 cm
• Easily analyzes fl uorescent materials
• Collection Optics: ATR Diamond Crystal

compiled by Fithrul, farmarind.wordpress.com

PHARMACEUTICAL INSPECTION CONVENTION
PHARMACEUTICAL INSPECTION CO-OPERATION SCHEME

PE 009-11 (Annexes)
1 March 2014

GUIDE TO GOOD MANUFACTURING
PRACTICE FOR MEDICINAL PRODUCTS
ANNEXES

© PIC/S March 2014
Reproduction prohibited for commerc ial purposes .
Reproduction for internal use is authorised ,

provided that the source is acknowledged.
- -·-

Editor:

PIC/S Secretariat
14 rue du Roveray
CH-1207 Geneva

e-mail :
web site:

info@picscheme.org
http://vvww. picscherne.org

PE 009-11 (Annexes)

-·-

-

1 March 2014

compiled by Fithrul, farmarind.wordpress.com

Annex 8

Sampl ing of starting and packaging materia ls

ANNEX 8
SAMPLING OF STARTING AND PACKAGING
MATERIALS

PRINCIPLE
Sampling is an important operation in which onl y a small fraction of a batch is
taken. Valid conclusions on the whole cannot be based on tests which have
been carried out on non-representative samples . Correct sampling is thus an
essential part of a system of Quality Assurance.
Note: Sampling is dea lt wi th in Chapter 6 of the Guide to GMP , items 6.11 to
6.14. These supplementary guidelines give additional guidance on the
sampling of starting and packaging materials.

PERSONNEL
1.

Personnel who ta ke samples should receive initial and on-going regular training
in the disciplines relevant to correct sampling. This training should include :
-,..

sampling plans ,

>
>
>
>

written sampling procedures ,
the techn iques and equipment for sampl ing ,
the risks of cross-contamination,
the precautions to be taken with regard to unstable and/or sterile
substances,

>

the importance of considering the visual appearance of materials ,
containers and labels ,

>

the importance of record ing any unexpected or unusual circumstances .

STARTING MATERIALS
2.

The identity of a complete batch of starting materials can normally only be
ensured if individual samples are taken from all the containers and an identity
test performed on each sample . It is permissible to sample only a proportion of
the containers where a va lidated procedure has been established to ensure that
no sing le container of starting material will be incorrectly identified on its label.

PE 009-11 (Annexes)

-81-

1 March 2014

compiled by Fithrul, farmarind.wordpress.com

2.2.48. Raman spectrometry

EUROPEAN PHARMACOPOElA 7.0

ｓ Ａ ｃ ｎ ａｌＭｔｏ
ｎ ｏｊ ｓ ｉ ｾ＠ RATI O
Th e detecti on limit and quantifi cation limit correspond to
signal-lo-noi se ratios of 3 and 10 respectively. The signal-to-noise
ratio (S/ N) is calcul ated usin g the expression :

S
JV

H

h

2H
h

height of th e peak corresponding to th e component
con cerned, in the electropherogram obtained with
th e prescribed reference so lutio n, measured from
the maximum of the peak to the extrapolated
baselin e of th e signal observed over a distance eq ual
to twenty tim es the width at half-height,
ran ge of the background in an electropherogram
obtained after injection of a blank, observed over
a distance equa l to twenty times the width at the
half-height of the peak in the e\ectropherogram
obtained with the prescribed reference soluti on and ,
if possible, situated equa lly aro und the place where
thi s peak wou ld be found.

A major limitati on of Ra man spectrometry is that impurities
may cause fluorescence that interferes wit h the detection of th e
mu ch weaker Raman signal. Fluorescence may be avoided by
choosing a laser source with a longer wavelength, for example
in the near infrared, as th e exciting lin e. The intensity of certain
Ram an lines may be enhanced in a number of ways, for in stance
in Resonance Raman (RR) and by Surface Enhanced Raman
Spectrometry (SERS) .

Du e to th e ;1arrow focus of th e irradiating laser beam, the
spectrum is typically obtained from on ly a few micro litres of
sample. Hence, sample inhomogenei ties must be considered,
unless the sample volume is increased, for examp le by rotation
of t he sample.
ID ENTTFICATION AND QUANTJTATTON USING REFERENCE
SUBSTANCES

Prepare the substance to be examined and th e reference
substance by the same procedure and record the spectra under
the same operational conditions. Th e maxi ma in the spectrum
obtai ned with th e substance to be exa min ed correspond in
position and relative intensity to those in th e spectrum obtained
01/ 2008:20248 with the reference substance (CRS).

2.2.48. RAMAN SPECTROMETRY
Raman spectrometry (inelasti c li gh t scattering) is a
li ght-scattering process in which the specimen under
examination is irradiated wi th intense monochromatic li ght
(us uall y laser li ght) and th e light scattered from the speci men is
ana lysed for frequency shifts.
Raman spectrometry is complementary to infrared spectrometry
in the sense that the two techniques both probe the molecular
vibrations in a material. However, Raman and infrared
spectrom etry have different relati ve sensitivities for different
functional groups. Raman spectrometry is particularly sensitive
to non-polar bonds (e.g. C-C sin gle or multiple bonds) and less
sensitive to polar bonds: Hence, wa ter, which has a strong
infrared absorption spectrum, is a weak Raman scatterer and is
thus well su ited as a solvent for Raman spectrometry.
Apparatus : Spectrometers for recording Raman spectra
typically consist of the following components:
a monochro matic li ghr source, typically a laser, wi th a
wavelengt h in th e ultrav iolet, visible or near-infrared region,
su itable optics (lens, mirrors or optical-fibre assembly) which
directs th e irrad iatin g light to and collects the scattered li ght
from the sample,
an optical device (monochromator or filter) that transmits the

frequ ency-5hifted Raman 5cattering and prevents the intense
incid ent frequency (Rayleigh scattering) from reachin g the
detector,
a dispersing device (grating or prism monochromator)
combined with wavelength-selecting slits and a detector
(usually a photomultipli er tube),
or:
a di spersin g device (grating or prism) combined with a
multichannel detector (usually a charge-coupled device
(CCD)),
or:
an interferometer with a detector that records the intensity
of the scattered light over time, and a data-handling device
that converts th e data to the frequ ency or wavenumber
domain by a Fourier-transform calculation.

When the spectra recorded in the solid state show differences in
the positions of the maxima, treat th e substance to be examined
and the reference substance in the same manner so that they
crystallise or are produced in the same form, or proceed as
described in the monogra ph, then record the spectra.
While Beer-Lambert's law is not valid for Ram an spectrometry,
Raman intensity is directly proportional to th e concentration of
the sca tteri ng species. As for other spectroscopic techniques,
quantitation can be performed using known amounts or
concentrati ons of reference substances. Owin g to the small
spatia l resolution of th e technique, care must be taken to ensure
representative samples of standards and unknowns, for examp le
by making sure that they are in th e same physical state or by
using an internal standard for liquid samples.
IDENTIF'TCATION AN D QUANTJTATION USING SPECTRAL
LIBRARIES AND STATISTICAL METHODS FOR
CLASSIFICATION AND CALIBRATION
Con trol of instrument performance. Use the appara tus
according to the manu facturer's instructions and carry out
the prescribed calibrations and system performance tests
at regular intervals, depending on the use of the apparatus
and the substances to be examined. When using Raman
spectro metry for quantitative determinations, or when setting
up spectral reference libraries for (chemometric) classification
or calibration, particular care should be taken to ensu re
that corrections are made or measures are taken to control
the variability in wavenumber and response-intensity of the
instrumentation .
Verification of the wavenumber scale. Verify the wavenumber
sca le of the Raman shift (normally expressed in reciprocal
centimetres) using a suitable standard which has characteristic
maxima at the wavenumbers under investigation, for example,
an organic substance, an Ne lamp or Ar· plasma lines from an
argon-ion laser.

The calibration measurement should be matched to the sample
type, i. e. a solid calibration sample should be used for solid
sa mples and a liquid calibration sample for liquid samples.
Choose a suitable substance (e.g. indene, cyclohexane or
naphthalene) for which accurate wavenumber shifts have
PREPARATION OF THE SAMPLE
been establ is hed (see Table 2.2.48.-1). The indene sample can
favourably be placed in an N.MR tube, evacuated and sealed
Raman spectra can be obtained from solids, liquids and gases
eith er directly, or in glass containers or tubes, generally without under inert gas, and stored cool in the dark to avoid degradation
of the sample.
prior sample preparation or diluti on.

82

See the information section on general monographs (cover pages)

compiled by Fithrul, farmarind.wordpress.com

EUROPEAN

ｐｈ ａｒｾ

ｉａｃｏｐｅ＠

7.0

2.2.54. Isoelectric focusing

Table 2.2 .48.-1. - Wave number shifts (and acceptable

tolerance.1) of cyclohexmie, indene and naphthalene.
cyclohexane

A

indene

6

naphthalene '1

Comparison of the spectra or transfor ms of the spectra or
quanti tative prediction of properties or amounts in th e material
in qu estion may invo lve the use of a suitable chemometric or
statistical classification or calibration techniq ue.

3056.4 (± 15)

01/ 2008:20249

2938.3 (± l 5)
2923.8 (± 15)

2.2.49. FALLING BALL VISCOMETER
METHOD

2852.9 (± 15)
l 609. 7 (± l.0)

J57G.6 (± 1.0 )

1444 .4 (± JO)

l 552.6 (± l.0)

1464.5 (± .1.0)

1266.4 (± l.O)

J 205.2 (± l.0)

1382.2 (± l .0)

1028.3 (± 1.0)

1018.6 (± l.O)

1021.6 (± 1.0)

801.3 (± J.O)

730.5 (± l 0)

763.8 (± l.0)

533.9 (± 1.0)

5J3.8 (± J.0)

l.1 4 7.2 (± LO)

ll57.G (± l.0)

1
'

S tandard guide for Raman shift standards for spectrom eter
calibration (Ame rican Society for Testing a nd Mate ria ls ASTM E 184 0).
" D. A. Corter, W. R. Thompso n, C. E. Tay lor a nd J. E. Pe mbert on,

Applied Spectroscopy, 1995, 49 (11), 1561-1576.

Verification of the response-intensity scale. The absolute and
relative intensities of the Raman bands are affected by several
factors including:
th e state of polarisation of the irradiating li ght,
th e state of polarisation of the collection optics,
the intensity of the irradiating li ght,
differences in instrument response,
differences in focus and geometry at sample,
differences in packing density for solid sa mples.
Appropriate acceptance criteria will vary with the application
but a day-to-day variation of± 10 per cent in re lative band
intensiti es is achievable in most cases.

The determi na tion of dynamic viscosity of Newtoni an liquids
usin g a suitable fa lling ball viscom eter is performed at
20 ± 0.1 °C, unless otherwise prescribed in the monograph. Th e
time requ ired for a test ball to fall in the liquid to be examined
from one rin g mark to the other is determined. If no stricter
limit is defined for th e equipment used the result is valid only if
2 consecutive measures do not differ by more than 1.5 per cent.
Apparatus. Th e falling ball viscometer consists of: a glass
tube enclosed in a mantle, which allow precise control of
tem perature ; six balls made of glass, nickel-iron or stee l with
different densiti es and diameters. Th e tube is fi xed in such a way
that the axis is inclin ed by 10 ± 1 ° with re ga rd to th e vertical.
Th e tube has 2 ring marks which define the di stance the ball has
to ro ll. Commerci ally available apparatus is supp lied with tables
givin g the constants, th e density of the balls and th e suitability
of the different balls for the expected range of viscosity.
Method. Fill th e clean, dry tube of the viscometer, previously
brought to 20 ± 0.1 °C, with the liquid to be examined, avoiding
bubbles. Add the ball suitable for the range of viscosity of the
li quid so as to obtain a falling time not less than 30 s. Close
th e tube and maintain the solution at 20 ± 0.1 °C for at least
15 min . Let th e ball run through th e liquid between the 2 ring
mar ks once without measurement. Let it run aga in and measure
with a stop-watch, to th e nearest one-fifth of a second, the time
required for the ball to roll from th e upp er to the lower ring
mark. Repeat the test run at least 3 times.
Calculate the dynamic viscosity fl in mi Iii pascal seconds using
the formula:
I)

Establishm ent of a spectral reference library. Record the
spectra of a suitable number of materia ls which have been fully
tested (e.g. as prescribed in a monograph) and which exhibit
the variation (manufacturer, batch, crystal modification, particle
size, etc.) typical of the material to be analysed. The set of
spectra represents the information that defin es th e similarity
border or quantitative limits, wh ich may be used, e.g. to identify
the substance or control the amount formed in a manufacturing
process. Th e number of substances in the database depends
on the specific application. The collection of spectra in the
database may be represented in different ways defined by the
mathematical technique used for classification or quantitation.
The selectivity of the database which makes it possible to
identify positively a given material and distinguish it adequately
from other materials in the database is to be established during
the validation procedure. This selectivity must be challenged on
a regular basis to ensure ongoing validity of the database ; this
is especiall y necessary after any major change in a substance
(e.g. change in supplier or in the manufacturing process of
the material) or in the set-up of th e Raman instrument (e.g.
verification of the wavenumber and response repeatability of
the spectrometer).

k

= k (p1 - p2)

X

t

constant, expressed in millimeter squared per
second squared,
density of th e ball used, expressed in grams per
cubic centimetre,
density of the liquid to be examined, expressed
in grams per cubic centimetre, obtained by
multiplying its relative density ､ ｾｧ＠ by 0.9982,
falling time of the ball, in seconds.
01/2010:20254

2.2.54. ISOELECTRIC FOCVSlNGt6)

GENERAL PRINCIPLES
Tsoelectric focusing (IEF') is a method of electrophoresis
that separates proteins accordin g to their isoelectric point.
Separation is carried out in a slab of polyacrylamide or
agarose gel that contains a mixture of amphoteric electro lytes
(ampholytes). When subjected to an electric field, the
ampholytes migrate in the gel to create a pH gradient. In some
This database is then valid for use only with the originating
cases gels containing an immobilised pH gradient, prepared by
instrument, or wit h a similar instrument, provided the
incorporating weak acids and bases to specific regions of the gel
transferred database has been demonstrated to remain valid.
network during the preparation of the gel, are used. When the
Method. Prepare and examine the sample in the same manner as applied proteins reach the gel fraction that has a pH that is th e
for the establishment of the database. A suitable mathematical same as their isoelectric point (pl), their charge is neutralised
and migration ceases. Gradients can be made over various
transformation of the Raman spectrum may be calculated to
ranges of pH , according to the mixture of ampholytes chosen.
facilitate spectrum comparison or quantitative prediction.
(6)

Thi s cliaplcr bas undergone pharmacopoe ia! harmonisation . Sec chapter 5.8. Pharmacopoeia / harmo11isatio11.

General Notices (1) apply to all monographs and other texts

compiled by Fithrul, farmarind.wordpress.com

83

USP 37

General In formation

STANDARD ERROR OF THE LABORATORY (SEL) is a ca lcu lati on
based on repeated readings of one or more samp les to estimate the precision and/or accuracy of the refe rence laboratory method, depending on how the data were collected.
STAN DARD ER ROR OF PRED ICTION (S EP) is a m easure of model
accuracy of an analytical method based on applying a given
ca libration model to the spectral data from a set of samples
different from but simil ar to those used to calculate the ca libration model. SEP is the sta ndard deviation of the residua ls
obtained from comparing the values from the refe rence laborato ry to those from th e method under test for th e specified samples . SEP provides a measure of the mod e l accuracy
expected when one measures future samples.
SURFACE REFLECTANCE, also known as specular reflection, is
that portion of the radiation not interacting with the sa mple
but simp ly reflecting back from the sample surface layer
(sample- air interface).
TRANSFLECTION is a transmittance measurement technique
in which the radiation traverses the sample twice. The second time occurs after the radiation is reflected from a surface behind the sample.
TRANSM ITIAN CE is represented by the equation:

T= 1/10 or T=

lQA

in which I is the intensity of the radiation transmitted
through the sample; 10 is the intensity of the radiant energy
in cident on the sample and includes losses due to solvent
absorption, refraction, and scattering; and A is the absorbance.

(1120) RAMAN SPECTROSCOPY
INTRODUCTION
Raman spectroscopy shares many of the principles that
apply to other spectroscopic measurements discussed in
Spectrophotometry and Light-Scattering (851 ). Raman is a vibrational spectroscopic technique and is therefore related to
infrared (IR) and near-infrared (NIR) spectroscopy. The Raman effect itself arises as a result of a change in the polarizability of molecular bonds during a given vibrational mode
and is measured as inelastically scattered radiation.
A Raman spectrum is generated by exciting the sample of
interest to a virtual state with a monochromatic source, typically a laser. Light elastically scattered (no change in wavelength) is known as Rayleigh scatter and is not of interest in
Raman spectro metry, except for marking the laser wavelength. However, if the sample relaxes to a vibrational energy level that differs from the initial state, the scattered radiation is shifted in energy. This shift is commensurate with the
energy difference between the initial and final vibrational
states. This "inelastically scattered" light is referred to as
Raman scatter. Only about one in 1 0 6- l 0 8 photons incident
on the sample undergoes Raman scattering. Thus lasers are
employed in Raman spectrometers. If the Raman-scattered
photon is of lower energy, it is referred to as Stokes scattering. If it is of higher energy, it is referred to as anti-Stokes

I

(1120 ) Raman Spectroscopy 959

scattering. In practice, nearly all analytically useful Raman
measurements make use of Stokes-shifted Raman scatter.
Th e appearance of a Raman spectrum is mu ch like an infrared spectrum plotted lin early in absorbance. The intensities, or the number of Raman photons counted, are plotted
against the sh ifted e nergi es . Th e x-axis is generally labe led
"Raman Shift/cm- 1 " or "Wavenumber/cm-1 " . Th e Raman
sh ift is usually exp ressed in wavenumber and represents the
difference in the absolute wavenumber of the peak and the
lase r wavenumber. The spectrum is interpreted in the same
manner as th e corresponding mid-infrared spectrum. The
positions of the (Raman shifted) wavenumbers for a given
vibrational mode are identical to the wavenumbers of the
corresponding bands in an IR absorption spectrum. However, the stronger peaks in a Raman spectrum are often weak
in an IR spectrum, and vice versa. Thus the two spectroscopic techniques are often said to be complementary.
Raman spectroscopy is advantageous because quick and
accurate measurements can often be made without destroying the sample (solid, semisolid, liquid or, less frequently,
gas) and with minimal or no sample preparation. The Raman spectrum contains information on fundamental vibrational mod es of the sample that can yield both sample and
process understanding. The signal is typically in the visible
or NIR rang e, allowing efficient coupling to fiber optics. This
also means that a signal can be obtained from any medium
transparent to the laser light; examples are glass, plastics, or
samples in aqueous media. In addition, because Raman
spectra are ordinarily excited with visible or NIR radiation,
standard glass/quartz optics may be used. From an instrum e ntal point of view, modern systems are easy to use, provide fast analysis times (seconds to several minutes), and are
re liable. However, the danger of using high-powered lasers
must be recognized, especially when their wavelengths are
in the NIR and, therefore, not visible to the eye. Fiber-optic
probes should be used with caution and with refe rence to
appropriate government regulations regarding lasers and laser classes .
In addition to "normal" Raman spectroscopy, there are
several more specialized Raman techniques . These include
resonance Raman (RR), surface-enhanced Raman spectroscopy (SERS), Raman optical activity (ROA), coherent antiStokes Raman spectroscopy (CARS), Raman gain or loss
spectroscopy, and hyper-Raman spectroscopy. These techniques are not widely employed in pharmaceutical laboratories, and are not addressed in this general information chapter.

QUALITATIVE AND QUANTITATIVE RAMAN
MEASUREMENTS
There are two general classes of measurements that are
commonly performed by Raman spectrometry: qualitative
and quantitative.

Qualitative Raman Measurements
Qualitative Raman measurements yield spectral information about the functional groups that are present in a sample. Because the Raman spectrum is specific for a given

Official from May 1, 2014
Copyright (c) 2013 The United States Pharmacopeial Convention. All rights reserved.

compiled by Fithrul, farmarind.wordpress.com

960 (1 120 ) Raman Spectroscopy I General In formation

compound, qualitative Raman measurements can be used
as a compendia! ID test, as well as for stru ctural elu cidation.

Quantitative Raman Measurements
For instrumen ts equipped with a detector that measu res
optica l power (such as Fourier transform [FT]-Raman spectromete rs), quantitative Raman meas urements utilize th e
fo ll owin g relatio nship betwee n signal, \, at a given wavenumber, v, and the concentration of an analyte, C:
S,. = Kcr,.(v, - v1,) 4 P0C
in which K is a constant that depends on laser beam diam eter, coll ection optics, sample volume, and temperature; er,. is
the Raman cross section of the particul ar vibrational mod e;
v, is the lase r wavenumber; v fJ is th e wavenumber of th e vibration al mod e; and P0 is th e laser power. Th e Raman cross
secti o n, crw is chara cteristic of th e nature of the particular vibrational mode. Th e sample volume is defined by size of the
focus of the lase r bea m at th e sampl e, th e optic being used
for focusing, and th e optical properti es of the sample itself.
Spot sizes at the sample can rang e from less than 1 µm for a
microprobe to 6 mm for a large area sample system. For
Raman spectrometers that measure the number of photons
per second (such as change-coupled device [CCD]-Raman
spectrometers) the corresponding equation is:

5,. = Kcr,.vL(vL- v1,)3P 0 C
From the above equations, it is apparent that peak signal is
directly proportional to concentration. It is this relationship
that is th e basis for the majority of quantitative Raman applications.

FACTORS AFFECTING QUANTIFICATION

Sample-Based Factors
Th e most important sample-based factors that del eteriously affect quantitative Raman spectrometry are fluoresce nce, sample heating, absorption by the matrix or th e sample itself, and the effect of polarization. If the sample matrix
in cludes fluor escent compounds, the measured signal will
usuall y contain a contribution from fluorescence. Fluorescence will be.observed only if the laser excitation wavelength ove rlaps with an absorption band of a fluorescent
compound. Fluorescence is typically observed as a broad
sloping background underlying the Raman spectrum. Fluoresce nce can cause both a baseline offset and reduced signal-to-noise ratio. The wavelength range and intensity of
the fluoresc ence is dependent on the chemical composition
of the fluores ce nt material. Because fluorescence is generally
a mu ch more efficient process than Raman scattering, even
very minor amounts of fluorescent impurities can lead to
significant degradation of the Raman signal. Fluoresce nce
can be reduced by using longer wavelength excitation sources su ch as 785 nm or 1064 nm. However, it should be remembered that the strength of the Raman signal is proportional to (vL - v;J 4, so the advantage of using a long-wavelength excitation lase r to minimize fluorescence is at least
partially offset by the reduced strength of the Raman signal.

USP 37

Th e greatest signal-to-noise ratio wil l be obtained by bala ncing fluoresce nce rejection, signal strength, and detector response.
FluoresC"e nce in solids can sometimes be mitigated by exposing the samp le to the lase r radiation for a period of time
before measu rement. This process is ca lled photobl eaching,
and ope rates by degrading the highly absorbing species .
Ph otobleac hin g is less effective in liquid s, where th e sample
is m obi le, o r if th e amount of fluorescent material is more
than a tra ce.
Sample heating by th e lase r source can cause a variety of
effects, such as physical form change (melting), polymorph
co nversion, or sample burning. Th e chance for sample heatin g is grea tes t when th e spot size at the sa mple is the smallest, i.e., when a microprobe is being us ed . This is usua lly an
issue for colored, highly absorbing species, or very small
particles that ha ve low hea t transfer. Th e effe cts of sample
heating are usually obse rvabl e either as changes in the Raman spectrum over time or by visual inspection of the sampl e.
Besi des decreasing the laser flux, a variety of methods can
be employed to diminish lase r-induced heating, such as
m ovi ng the sample or lase r during the measurement or improvin g the heat transfer from the sample with thermal co nta ct or liquid imm ersi on.
Absorpti on of th e Raman signal by the matrix or the sampl e itself can also occur. This problem is more prevalent with
long-waveleng th FT-Raman systems where the Raman signal
can overlap with an NIR overtone absorption. This effect will
be depend ent on the op tics of the system as well as on the
sample prese ntation. Associated with this effect is variability
from scattering in solids as a result of packing and particlesize differences. Th e magnitude of all of these effects, however, is typi cally less severe than in NIR because of the limited depth of penetration and th e relatively narrower waveleng th reg ion sampled in Raman spectrosco py.
Finally, it should be recognized that laser radiation is polarized and th e Raman spectra of crystalline materials and
other oriented samples can differ significantly depending on
the way that the sample is mounted . If the Raman spectrom ete r is capable of producing linearly polarized radiation
at th e sample then a polarization scramble r is recommended for routin e sample analysis.

Sampling Factors
Raman spectroscopy is a zero-background technique, in
that the signal at the detector is expected to be zero in the
absence of a sample. This situation can be contrasted with
absorption spectrometry, where the signal at the detector is
at a ma ximum in the absence of a sample. Zero-background
tec hniques are inherently sensitive because small changes in
sampl e concentration lead to proportionate changes in the
signal level. The instrument will also be sensitive to other
sources of light that can cause sample-to-sample variations
in the measured signal level. In addition, a large background
signal caused by fluores ce nce will lead to an increased noise
level (photon shot noise). Thus it may be very difficult to use
th e absolute Raman signal for direct determination of an analyte. Other potential sources of variation are changes in the
sample opacity and heterogeneity, changes in the laser
power at the sample, and changes in optical collection ge-

.
.
.
Official from May 1, 2014
Copyright (c) 2013 The United States Pharmacope1al Convention. All rights reserved.

compiled by Fithrul, farmarind.wordpress.com

Genera l Information

USP 37

ometry or sample position . These effects can be minimized
by sampling in a reproducib le, representative manner. Carefu l design of the instrumentation can reduce these effects
but they cannot be eliminated enti re ly.
Use of an internal reference standard is the most common
and robust method of el iminating variations caused by ab solute intensity fluctuations. The re are several choices for
this approach. An interna l standard can be deliberately added, and iso lated peaks from this standa rd can be employed;
o r a band due to a mo iety such as an aromatic rin g, the
Rama n cross-section of w hi ch does not change w ith the way
the sa m ple is prepared, can a lso be used. For solutio n spectra, an iso lated so lvent band can be e m ployed beca use t he
so lvent w ill remain relatively u nc hanged from sa m p le to
sample. Al so, in a fo rmu lation, an excip ient peak can be
used if it is in su bstantial excess co m pa red to t he ana lyte.
The entire spectrum can also be used as a reference, w it h
the ass umption that laser and sa m ple-orientation changes
w ill affect the e nt ire spect rum eq ually.
A second important samplin g -based facto r to conside r is
spectral co n ta mi nation. Raman scatte ri ng is a weak effect
th at can be m asked by a num be r of exte rn al sou rces. Co m m on contamination so u rces include sample-ho lder a rtifacts
(conta ine r or su bstrate) a n d amb ien t light. Typica lly, t hese
issues ca n be identified and reso lved by careful expe rim entation.

APPARATUS

I

(1120) Raman Spectroscopy 96 1

4 . Wave length p rocess ing unit
5. Detector and electronics
EXCITATI ON SOURCE ( LASER)

Table 1 identifies several common lasers used for pha rmaceutica l applications or Rama n spectrometry. UV lasers have
also been used for speciali zed app lications but have va rio us
d rawbacks that limi t their uti li ty for gene ra l analytica l measu rements. As m o re app lications fo r UV lasers are described,
it is like ly that they may beco m e mo re com m on for Ra m an
spectrometry.
SAMP LI NG DEVICE
Seve ra l samp li ng a rrangements are possib le, includ in g direct optical inte rfaces, mi croscopes, fibe r opt ic-based probes
(eithe r noncontact or imm e rsio n optics), and sa m ple chambers (includ ing spec ia lty sa m p le h o ld e rs a n d a u tomated
sample changers). The samp li ng optics can a lso be designed
to obtain the po larization-depe nden t Ra m a n spectru m,
wh ich often contains add itio nal info rm atio n . Se lectio n of
th e sampling device wi ll ofte n be dictated by t he ana lyte
and sample. However, cons ide rations suc h as sa m p lin g volum e, speed of the m eas ure m e nt, lase r safety, a nd reprod ucib ili ty of sa m p le p resentation sho ul d be eva luated to op tim ize the samp li ng device fo r any give n app li catio n .
FILTERING DEVICE

Components
All m ode rn Ra m an m easu re m e nts invo lve ir radiatin g a
sa m p le w ith a lase r, co ll ectin g the scatte re d radi atio n, rejecti ng the Rayle igh -scattered lig h t, d iffe re n tiating t h e Ram an photons by wave length, and detecti ng the res ul t in g
Ra m a n spectru m. All co mm e rcia l Ra m a n in strum e n ts the refore share t he fo ll owing com m on features to perform th ese
fu n ctions:
1 . Excitation so urce (lase r)
2 . Sa m p ling device
3. Device to filte r/ rejec t ligh t scattered at the laser wavele ngth

Th e in tensity of scattered li gh t a t the laser wave length
(Rayle ig h) is m any orde rs of m ag ni t ude g reate r than the
Rama n signal a n d mu st be rejecte d p rio r to th e d etecto r.
Notch fi lters are almost uni versa lly used fo r t his p u rpose and
provide exce ll ent rejecti on and st abi li ty co m bi n ed with
sma ll size. The tra d itio na l us e of mul t istag e m o noc h rom ators for th is purpose, alth oug h sti ll viab le, is now ra re. In additio n, va ri ous fil te rs o r p hys ica l ba rri e rs to shie ld the sa m ple
fro m external rad iation so u rces (e.g ., roo m li ghts, lase r p lasm a li nes) m ay be req ui red depe nd ing on the co ll ectio n geo m et ry of the instru m e n t.

Table 1 Lasers Used in Pharmaceutical Applications
Laser ),, nm (n e are st
whole number)

Type

Typical Power
at Laser

Wavelength Range, nm
(Stokes Region, 100 cm- 1
to 3000 cm - 1 shift)

Comments

NI R Laser s
Commonly used in Fou ri er transform instrum en ts

1064

Solid state
(N d:YAG)

Up to 3 W

1 075- 1563

830

Diode

Up to 300 mW

827- 980

Typically li mited to 2000 cm 1 ; Rama n sh ift
because of CCD spectral response; less
common than the other lase rs

785

Diode

Up to 500 mW

791 - 1027

Mosl widely used dispersive
Rama n laser

632.8

He- Ne

Up to 500 m W

637- 781

532

Doubled
(Nd:YAG)

Up to 1 W

535-632.8

High flu orescence ri sk

514.5

Ar+

Up to 1 W

51 7-608

Hiqh fluo rescence ri sk

488-632.8

Ar+

Up to 1 W

490-572

High fluo rescence risk

Visible Lasers
Relatively sma ll fl uorescence risk

Official from May 1, 2014
.
.
Copyright (c) 2013 The United States Pharmacopetal Convention . All rights reserved.

compiled by Fithrul, farmarind.wordpress.com

962 (1120) Raman Spectroscopy

I

General In fo rm ation

USP 37

WAVELENGTH PROCESSING UNIT

LASER WAVE LENGTH

The wavelength scale may be encoded by ei ther a scanning monochromato:, a gratin g polychromator (in CCDRaman spectrometers) or a two-beam in terferometer (in FTRaman spectrometers). A discussion of the specifi c benefits
and drawbacks of each of the dispersive designs compared
to the FT ins trument is beyond the scope of this cha pter.
Any properly qualified instruments shou ld be suitabl e for
qualitative measu rements. However, care must be taken
when selectin g an in stru ment for quantitative measurements, as dispersion and response lin ea rity might not be
uniform across th e full spectral range.

Laser wavelength variation can impact both th e wavelength precision and the photometri c (signal) precision of a
given instrument. Even the most stable current lasers can
vary sli ghtly in their measured wavelength output. The laser
wavelength must therefore be confirmed to ensu re that the
Raman shift positions are accu rate fo r both FT-Raman or dispersive Raman instruments. A reference Raman sh ift sta ndard material such as those outli ned in ASTM El 840-96
(2002) 1 or other suitab ly ve rifi ed materials can be utili zed
for this purpose. [NOTE- Reliable Rama n shift standa rd va lues for frequently used liquid and solid reagents, required
for wave number cali bration of Raman spectrometers, are
provided in the ASTM Standard Guide cited. These va lu es
can be used in addition to the highly accurate and precise
low-p ress ure arc lamp emission lines that are also ava ilable
for use in Ram an instrument calibration.] Spectrometric
grade mate ri al can be purchased from appropriate suppli ers
for this use. Certain instrum ents may use an intern al Ram an
stan dard separate from the primary op ti cal path. External
calibration devi ces exactly reprod uce th e opti cal path taken
by the scattered radiation. [N OTE-When chemical sta nd ards are used, care must be taken to avo id contamination
and to confirm standa rd stability.]
Unl ess the ins trum ent is of a continuous calibration type,
the primary wavelength ax is ca li bration sho uld be performed, as per vendor procedures, ju st prior to meas uring
the laser wavelength. For externa l calibration, th e Raman
shift stand ard shou ld be placed at the sample loca ti or;i and
meas ured using appropriate acquisition parameters. Th e
peak center of a strong, well-resolved band in th e spectral
reg ion of interest should be evaluated. The position can be
assessed man ually or with a suitable, valid peak-picking algorithm. Th e software provided by the vendor might measure th e laser wavelength and adjust the laser wavelength
appropriately so that this peak is at th e proper position. If
the vendor does not provide this functionality, the lase r
wavelength sho uld be adjusted manually. Depending on the
type of laser, the laser wavelength can vary with temperature, current, and voltage. Wavelength tolerances can vary
dep ending on the specifi c application.

DETECTOR
Th e si li con-based CCD array is the most common detector for dispersive in strum ents. The cooled array detector allows measu rements over th e spectra l range from 4500 to
100 cm- 1 Ram an sh ift with low noise when most visible lase rs, such as fr equency-doubled neodym ium-doped yttrium-aluminum-g arnet (Nd:YAG) (532 nm) or helium-neon
(632.8 nm) lasers, are used. When a 785-nm diod e laser is
used, the wavelength range is red uced to about 31 00 to
100 cm- 1. The most commonly used CCD has its peak
wavelength respons ivity when match ed to th e commonly
us ed 632.8-nm He- Ne gas laser or 785-nm diode laser. FT
instrum en ts typically use single-channel germanium or indium- gallium- arsenid e (lnGaAs) detectors responsive in the
NIR to ma tc h the 1064-nm excitatio n of a Nd :YAG lase r.

Calibration
Ram an instrument ca libration involves three components:
primary wavelength (x-axis), laser wavelength, and intensity
(y-axis).
PRIMARY WAVELENGTH (X-AXIS)
In the case of FT-Raman instruments, primary wavelength-axis calibrati on is maintained, at least to a first approxim ation , with an internal He-Ne laser. Most disp ersive
instrum en ts utilize atomic emission lamps for primary wavelength-axis calibration . In all instruments suitable for analytical Raman measu rements, the vendor will offer a procedure
of x-a xis ca libration that can be perform ed by the user. For
disp ersive Ram an instrum ents, a calibration based on multiple atomic emission lines is preferred. Th e validity of this calibrati on approach can be verified subsequent to laser wavelength calibration by using a suitable Raman shift standard.
For sca nning dispersive instruments, calibration might need
to be perform ed more frequently, and precision in both a
scanning and static operation mode may need to be verifi ed.1

SIGNAL LEVEL (Y-AXIS)
Calibrati on of the photometric axis can be critical for successful quantifi cation by using certain analytical methods
(chemometri cs) and method transfer between instruments.
Both FT-Raman and dispersive Raman spectrometers should
undergo similar calibration procedures. The toleran ce of
photometri c precision acceptable for a given measurement
should be assessed during the method development stage .
To calibrate th e photometric response of a Raman instrument, a broad-band emission source should be used . Th ere
are two accepted methods . Method A utilizes a tungsten
white light source.2 Th e output power of such sources is
' NIST-tracea ble tungsten w hi te light so urce statement: While the ca libration

1

ASTM El 840-96 (2002) Standard Guide for Raman Shift Standards for Spectrometer Calibration, ASTM Intern atio nal, 1 00 Barr Harbor Drive, PO Box

of the Raman frequen cy (o r Raman shift, cm- 1) ax is using pure materials and
an existing ASTM stand ard is well accepted, tech ni ques fo r ca libration of the
Raman intensity axis are not. In tensity ca librations of Raman spectra can be

GOO, West Consho hocken, PA, USA 19428-2959.

acco m pli shed with certified white ligh t sou rces .

Official from May 1, 2014
Copyright (c) 2013 The United States Pharmacopeial Convention. All rights reserved .

compiled by Fithrul, farmarind.wordpress.com