Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 102 (2013) 403–407

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journal homepage: www.elsevier .com/locate /saa

Estimation of ibuprofen in urine and tablet formulations by transmissionFourier Transform Infrared spectroscopy by partial least square

Abdul Rauf Khaskheli a,⇑, Sirajuddin a, S.T.H. Sherazi a, S.A. Mahesar a, Aftab A. Kandhro b,Nazar Hussain Kalwar a, Muhammad Ali Mallah a

a National Centre of Excellence in Analytical Chemistry, University of Sindh, Jamshoro 76080, Pakistanb Pakistan Council of Scientific and Industrial Research Laboratories, Karachi, Sindh, Pakistan

h i g h l i g h t s

" FT-IR procedure was developedusing partial least square (PLS)calibration model.

" PLS calibration was based on regionfrom 1807 to 1461 cm�1 ofibuprofen standards.

" Estimation of ibuprofen inpharmaceutical formulations andhuman urine samples.

1386-1425/$ - see front matter � 2012 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.saa.2012.10.021

⇑ Corresponding author. Tel.: +92 22 9213429; fax:E-mail address: arkhaskheli@gmail.com (A.R. Khas

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 10 February 2012Received in revised form 5 October 2012Accepted 13 October 2012Available online 23 October 2012

Keywords:IbuprofenTransmission FT-IR spectroscopyPartial least squareTabletsUrine samples

a b s t r a c t

A rapid, reliable and cost effective analytical procedure for the estimation of ibuprofen in pharmaceuticalformulations and human urine samples was developed using transmission Fourier Transform Infrared(FT-IR) spectroscopy. For the determination of ibuprofen, a KBr window with 500 lm spacer was usedto acquire the FT-IR spectra of standards, pharmaceuticals as well as urine samples. Partial least square(PLS) calibration model was developed based on region from 1807 to 1461 cm�1 using ibuprofen stan-dards ranging from 10 to 100 lg ml�1. The developed model was evaluated by cross-validation to deter-mine standard error of the models such as root mean square error of calibration (RMSEC), root meansquare error of cross validation (RMSECV) and root mean square error of prediction (RMSEP). The coeffi-cient of determination (R2) achieved was 0.998 with minimum errors in RMSEC, RMSECV and RMSEP withthe value of 1.89%, 1.63% and 4.07%, respectively. The method was successfully applied to urine and phar-maceutical samples and obtained good recovery (98–102%).

� 2012 Elsevier B.V. All rights reserved.

Introduction

Non-steroidal anti-inflammatory drugs (NSAIDs) are commonlyused to control the pain and inflammation occurred due to any dis-ease. Ibuprofen (2-[4-(2-methylpropyl) phenyl] propanoic acid) isan important anti-inflammatory, analgesic and antipyretic drugwith considerably less gastrointestinal adverse effect than otherNSAIDs [1–3]. In the literature it is clearly reported that ibuprofen

ll rights reserved.

+92 22 9213431.kheli).

is very effective for mild pain as well as severe pain attributable toinflammatory diseases like osteoarthritis, rheumatoid arthritis,ankylosing spondylitis, gout and Bartter’s syndrome [4,5].

For the quantification of ibuprofen in tablets, the BritishPharmacopoeia [6] describes a classical titrimetric method whilethe United States Pharmacopoeia [7] illustrates instrumentalhigh-performance liquid chromatographic (HPLC) method. Thetitrimetric method is not so accurate due to the problem of preciseend point detection, although HPLC method is more sensitive, butrequires a large amount of organic solvents. Pharmaceutical indus-tries require rapid and low cost methods for the accurate analysis

404 A.R. Khaskheli et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 102 (2013) 403–407

of their raw materials, finished products and process samples intheir quality control laboratories. To counter these problems,several analytical methods have been reported in the literaturefor the routine quantification of ibuprofen in bulk powder,pharmaceutical formulations and biological samples by spectro-photometry [8–10], spectrofluorimetry [11,12], polarography[13], conductometry [14], high-performance liquid chromatogra-phy [15–18], capillary electrophoresis [19,20], infrared spectrome-try [21], supercritical fluid chromatography [22] and protonmagnetic-resonance spectroscopy [23]. From last few decades,FT-IR spectroscopy has achieved much recognition as rapid analyt-ical technique in the laboratories of pharmaceutical industries forqualitative analysis of raw materials and finished products[24,25]. The advantages of FT-IR analysis are ease of operation,high sample turnover and no sample pretreatment, also predictionof one or more components through a single spectrum is possiblewith the application of chemometric techniques [26–29].Therefore, the aim of present study was to develop a simple, quick,inexpensive and accurate method for the routine determination ofibuprofen in tablet formulations and urine samples using FT-IRspectroscopy.

Material and methods

Reagents and samples

A stock solution (1000 lg ml�1) of ibuprofen (purity 99%-Merck,Darmstadt, Germany) was prepared by means of dissolving 0.1 g ofstandard ibuprofen in 100 mL of chloroform 99.98% purity (Fisherscientific UK limited). Working standards were prepared by takingthe required volume from the stock and diluted with chloroform.All solutions were stored in the dark at laboratory temperatureand in glass vessels with tight cover. The tablet samples containingibuprofen were purchased from drug store Hyderabad, Pakistan.

Sample preparation

Sample preparation procedure involves only grinding of tabletsamples followed by dissolution in chloroform for FT-IR measure-ment. After weighing, the tablet samples were ground to finepowder in mortar to minimize particle size. Quantitative analysesof solutions of ibuprofen in chloroform were performed in a cellwith KBr windows and variable optic pathway (Wilks). The KBrwindows were scanned from 4000 to 400 cm�1 on Thermo Nicolet5700-FT-IR spectrometer.

Collection and preparation of urine samples

Urine samples were collected from healthy volunteers after 2 hthe doze of an ibuprofen tablet. 2 ml of urine sample was passedthrough DSC-18 column, which was pre-washed with 2 ml ofmethanol. Then column was washed with 5 ml of 5% methanol fol-lowed by 1 ml chloroform. After washing, the urine sample wasspiked with known concentration of ibuprofen to check thepercent recovery of ibuprofen.

FT-IR spectral measurements

Thermo Nicolet 5700 FT-IR spectrometer fitted with a demount-able KBr liquid cell and DTGS detector was used for the recordingof infrared spectra of standard, tablet and urine samples. OMNICsoftware version 7.3 was used to control the instrument. Themid-IR spectral range from 4000 to 400 cm�1 was selected with to-tal 16 scans and resolution of 4 cm�1. A fresh background spectrum

of chloroform was taken before recording the spectra of each sam-ple and standard.

FT-IR calibrations

Ten standards of ibuprofen ranging from 10 to 100 lg ml�1 inchloroform were prepared. PLS calibration model was developedto assess unknown quantity of ibuprofen present in pharmaceuti-cal tablets and urine samples. For quantitative determination,Turbo Quant (TQ) analyst software was applied. The FT-IR spectraof ibuprofen standards were opened in TQ analyst program whichis better option to save the time and labor to measure peak height,peak width and peak area as compare to manual calculations.Through cross validation, an excellent calibration curve was ob-tained between actual and predicted values of ibuprofen standards.

Results and discussion

FI-IR has now recognized as rapid analytical technique for bothqualitative and quantitative analysis due to outstandingperformance of recently developed softwares and chemometrics.The main objective of current study was to develop a rapid methodusing FT-IR spectroscopy for the assessment of ibuprofen in phar-maceutical tablets and urine samples. The existing methods usinghigh cost instruments which although provide good sensitivity butat the same time they suffer from draw backs of time consumptionas well as intensive labor involved. PLS multivariate calibrationmodel was applied in order to correlate the data collected betweendifferent variables on same observations in multi-componentmixtures. PLS works on principle of computing large amount ofinformation which is common between different variables withmaximal covariance.

Analysis of pharmaceutical samples

For quantification of ibuprofen in pharmaceutical samples, sim-ple Beer’s law and PLS calibrations were developed using standardsof ibuprofen ranging from 10 to 100 lg ml�1 on FT-IR under opti-mized parameters. Good linearity (R2 = 0.99820) with excellent re-sponse of FT-IR with increasing concentration of standards wereobserved in PLS calibration model in comparison to the Beer’slaw as shown in Fig. 1. The outcome of calibration clearly revealedsignificant sensitivity and accuracy of the FT-IR method. Table 1shows the comparison of the two calibration models Beer’s lawand PLS.

The results for the ibuprofen concentration found in the tabletsamples and mean values of five replicate measurements withlow standard deviation revealed that method is well reproducible.

There was very good conformity between obtained results andvalues labeled on formulations and results achieved were accord-ing to limits of pharmacopoeia. The evaluation of proposed methodwas also checked by standard addition method with knownamounts of ibuprofen to the tablet sample as shown in Fig. 2a.

The statistics of ibuprofen recovery tests (Table 2) on three setof samples revealed high recovery performance (98.53%, 99.66%,and 102.02%) with reasonable precision (CV = 1.6%, 1.3% and1.5%) of the proposed method. Satisfactory recovery is 90–108%for concentrations around 0.1% or 100 mg/100 g [30,31].

Limits of detection and quantification

The analysis at lowest concentration which produced substan-tial signal was repeated eleven times and calculated by followingformula [30,31]:

LOD ¼ 3� SD� C=M

Fig. 1. Calibration plot of ibuprofen standards in the range of 10 to 100 lg ml�1.

Table 1Comparing the results obtained in analyzing standard solution of IBP.

Beer’s law PLS

Peak height Peak area

Spectral region (cm�1) 1710 cm�1 1807–1461 cm�1 1807–1461 cm�1

Number of factors – – 7RMSEC – – 1.89RMSEP – – 4.07R2 0.496 0.644 0.998

A.R. Khaskheli et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 102 (2013) 403–407 405

where SD is standard deviation; C is the concentration of analyteand M is the mean peak area.

LOQ was determined by the same way with following equation:

LOQ ¼ 10� SD� C=M

For the determination of ibuprofen the lowest concentration to bedetected through this method was found to be 0.77 lg ml�1 andquantification limit was found to be 2.54 lg ml�1.

Fig. 2a. Pharmaceutical tablet sample and sp

Analysis of urine samples

For analysis of ibuprofen in urine samples, same calibration thatwas used for tablet samples was applied as it also covered lowerrange of the analyte. Therefore, precise and reproducible resultswere obtained.

One urine sample was spiked with 10, 20 and 30 lg ml�1 ofibuprofen to check response and accuracy of method. Fig. 2b repre-sents the extended FT-IR spectra from 1800 to 1650 cm�1 withthree spiked level of ibuprofen.

Recovery results of ibuprofen in urine samples after spikingknown concentrations of standard are given in Table 3. Appropriatepercent recoveries (98.6% and 100.15%) of the standard added tosample with very small variation clearly verified the reliability ofmethod for urine samples also.

All measurements were taken in five replicates to ensure repro-ducibility as very low values of standard deviation were observedwith respect to mean values. From FT-IR spectra of urine samplesand standards of ibuprofen, it could be easily deduced that involve-ment of interference from the sample matrix is very negligible. Theresidual mean standard error of calibration (RMSEC) value of 1.89

ikes of 30, 50 and 70 lg ml�1 ibuprofen.

Table 2Recovery result of ibuprofen from tablet samples after spiking known concentrations of standard.

Conc.(A) lg ml�1 Conc.(B) lg ml�1 By FT-IR Acceptable recovery (%) AOAC [30]

Conc.(C) Recoverya (%) CVb (%)

1 30 20 49.56 ± 1.25 98.53 1.62 50 20 69.83 ± 0.8 99.66 1.33 70 20 91.42 ± 1.4 102.02 1.5 90–108

(A) Exogenous addition.(B) Before addition.(C) After addition.

a Recovery (%) = (C � B)/A � 100.b Coefficient of variation was obtained from the mean of five replicate tests.

Fig. 2b. Extended FT-IR spectra of blank, urine sample with two spikes of 20 and 30 lg ml�1 ibuprofen standard.

Table 3Recovery results of ibuprofen in urine samples after spiking known concentrations ofstandard.

Conc.(A)

lg ml�1Conc.(B)

lg ml�1By FT-IR

Conc.(C) Recoverya

(%)CVb

(%)

1 10 11 20.86 ± 1.3 98.6 2.12 20 11 31.03 ± 0.8 100.15 2.1

(A) Exogenous addition.(B) Before addition.(C) After addition.

a Recovery (%) = (C�B)/A � 100.b Coefficient of variation was obtained from the mean of five replicate tests.

406 A.R. Khaskheli et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 102 (2013) 403–407

was achieved after comparison of actual and computed concentra-tion for all standards which provide evidence about the accuracyand precision of method.

Conclusion

Transmission FT-IR spectroscopic method was developed usingKBr windows for the determination of ibuprofen. A PLS model wassuccessfully developed with the help of TQ analyst software. Themain goals achieved through this method include analytical sim-plicity, enhanced rapidity, better accuracy and improved sensitiv-ity for quantification of ibuprofen. The developed method was

easy to handle, cheap and very effective for quantification of ibu-profen in tablet formulation and urine samples.

Acknowledgment

The authors would like to thank Ex-Director (Prof. M. I. Bhang-er), National Centre of Excellence in Analytical Chemistry (NCEAC),University of Sindh, Jamshoro, Pakistan for providing the opportu-nity to carry out this research work.

References

[1] D. Velasco, Ch.B. Danoux, J.A. Redondo, C. Elvira, J.S. Román, P.S. Wray, S.G.Kazarian, J. Control. Release 149 (2011) 140–145.

[2] D.R. Mehlisch, M.D.S. Aspley, S.E. Daniels, D.P. Bandy, Clin. Ther. 32 (2010)882–895.

[3] B.J. Whittle, Fundam. Clin. Pharmacol. 17 (2003) 301–313.[4] C.D. Palma, R. Di Paola, C. Perrotta, E. Mazzon, D. Cattaneo, E. Trabucchi, S.

Cuzzocrea, E. Clementi, Pharmacol. Res. 60 (2009) 221–228.[5] A.O. Santini, J.E. de Oliveira, H.R. Pezza, L. Pezza, Microchem. J. 84 (2006) 44–

49.[6] British Pharmacopoeia, vol. 1, Pharmaceutical Press, London, 1998.[7] United States Pharmacopeia National Formulary, USP 26, NF 21, Rockville,

2003, p. 947.[8] V. Tantishaiyakul, N. Phadoongsombut, S. Kamaung, S. Wongeisansri, P.

Mathurod, Pharmazie 54 (1999) 111–114.[9] D. Balsi, K.K. Mahalanabis, B. Roy, J. Pharm. Biomed. Anal. 16 (1998) 809–812.

[10] A.A. Wahbi, E. Hassan, D. Hamdy, E. Khamis, M. Barary, Pak. J. Pharm. Sci. 18(2005) 1–6.

[11] L.A. Hergert, G.M. Escandar, Talanta 60 (2003) 235–246.[12] P.C. Damiani, M. Bearzotti, M.A. Cabezón, J. Pharm. Biomed. Anal. 25 (2001)

679–683.

A.R. Khaskheli et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 102 (2013) 403–407 407

[13] C. Kanout, P. Boucly, E. Guernet-Nivaud, M. Guerent, Ann. Pharm. Fr. 43 (1985)265–269.

[14] F.A. Aly, F. Belal, Pharmazie 49 (1994) 454–455.[15] R. Canaparo, E. Mumtoni, G.P. Zara, C. Dellapena, E. Berno, M. Costa, M. Endi,

Biomed. Chromatogr. 14 (2000) 219–226.[16] B. Vermeulen, J.P. Remon, J. Chromatogr. B Biomed. Sci. Appl. 749 (2000) 243–251.[17] C. Herkenne, A. Naik, Y.N. Kalia, J. Hadgraft, R.H. Guy, J. Invest. Dermatol. 127

(2007) 135–142.[18] H. Farrar, L. Letzig, M. Gill, J. Chromatogr. B 780 (2002) 341–348.[19] K. Makino, Y. Itoh, D. Teshima, R. Oishi, Electrophoresis 25 (2004) 1488–1495.[20] R. Hamoudova, M. Pospısilova, J. Pharm. Biomed. Anal. 41 (2006) 1463–1467.[21] E. Dreassi, G. Ceramelli, P. Corti, M. Massacesi, P.L. Perruccio, Analyst 120

(1995) 2361–2365.[22] N.K. Jagota, J.T. Stewart, J. Chromatogr. 604 (1992) 255–260.

[23] S. Husain, M. Kifayatullah, R. Sekhar, J. AOAC Int. 77 (1994) 1443–1446.[24] M. Blanco, D. Valdes, M.S. Bayod, F. Fernandez-Mari, I. Llorente, Anal. Chim.

Acta 502 (2004) 221–227.[25] O.Y. Rodionova, L.P. Houmøller, A.L. Pomerantsev, P. Geladi, J. Burger, V.L.

Dorofeyev, A.P. Arzamastsev, Anal. Chim. Acta 549 (2005) 151–158.[26] S.T.H. Sherazi, M. Ali, S.A. Mahesar, Vib. Spectrosc. 55 (2011) 115–118.[27] X.M. Chong, C.Q. Hu, Y.C. Feng, H.H. Pang, Vib. Spectrosc. 49 (2009) 196–203.[28] M.C. Sarraguca, J.A. Lopes, Vib. Spectrosc. 49 (2009) 204–210.[29] X. Wanga, Q. Fua, J. Sheng, X. Yang, J. Jia, W. Du, Vib. Spectrosc. 53 (2010) 214–

217.[30] AOAC International, Guidelines for Single Laboratory Validation of Chemical

Methods for Dietary Supplements and Botanicals, Arlington, VA, 2002.[31] N.N. Memon, S.T.H. Sherazi, F.N. Talpur, M.I. Bhanger, J. Assoc. Off. Anal. Chem.

Int. 94 (2010) 1249–1254.