Discovery and Formulation - Emerging Raman Applications and Techniques in Biomedical and Pharmaceutical Fields - page 227

the particle counting and shape evaluation tools is that it is generally dicult
if not impossible to confirm the identity of any individual particle. With a
number of inhalation products being combination products (i.e. contain more
than one active compound) discrimination between API particles becomes a
tougher challenge. However, the nominal collection volume and specificity of
Raman microscopy provides an elegant solution to this identification problem.
This was first demonstrated with the Raman mapping of deposits on impactor
plates from pressurised metred-dose inhalers (P-MDI) containing Ventide
[65]. Individual particles of each API present in the formulation (salbuta-
mol and beclometasone dipropionate) were easily identified from their Raman
spectra. From these data it was clearly demonstrated that neither API had
changed form or degraded in the formulation suspension. In the presence of
moisture, beclometasone dipropionate will convert to the monohydrate form,
so this result also suggested there was no significant moisture ingress into the
p-MDI canister. The Seretide /Advair formulation also contained two APIs:
salmeterol xinafoate and fluticasone propionate. This formulation was believed
to have better ecacy over the individual active compounds administered sep-
arately as the combination product allowed better co-deposition of the two
APIs. To evaluate this hypothesis impactor plates from combined and sepa-
rate API administrations were studied by Raman mapping [66]. The Raman
spectra obtained confirmed each particle's identity and form, and the chemical
images allowed identification of particles formed by API co-deposition. These
images were then analysed for differences in API distributions using a statis-
tical technique known as bootstrapping. The outcome was that significantly
more co-deposition occurs from APIs delivered simultaneously from the same
inhaler.
Although the use of Raman mapping has clear benefits to the analysis of
inhaled product, the approach has a number of limitations. These include

(i) that the method cannot be used quantitatively as the thickness of each
particle cannot be controlled or determined;
(ii) the method can be time consuming (a 128

× 128 μ m area mapped with a

m step size with a 10 s data acquisition takes 48 h to complete);
(iii) a significant part of the experiment time is wasted due to spectra being
acquired from blank areas of the impactor plate because the individual
or co-deposited particles are spatially resolved.

1 μ

This particular type of analysis is ideal for Raman imaging. As parti-
cles are spatially resolved there is no opportunity for spectral mixing from
materials below them, a problem inherent in Raman imaging of heteroge-
neous mixtures (e.g. tablet formulations). Similarly, as all points in the field
of view are analysed simultaneously there is no time wasted in the data ac-
quisition from blank regions in the sampling area. This has been exploited
by Doub et al. [67] in the analysis of Beconase AQ nasal spray. Imag-
ing areas at least a factor of 2 larger than the previously described map-
ping experiments, spectral acquisition took less than 3.5 h providing con-