simultaneous determination of diethylene glycol and propylene glycol in pharmaceutical products by...

Tao Zhou1

Haiying Zhang2

Gengli Duan1

1Department of PharmaceuticalAnalysis, School of Pharmacy,Fudan University, Shanghai, PRChina

2Department of Biology, DezhouUniversity, Shandong, PR China

Original Paper

Simultaneous determination of diethylene glycoland propylene glycol in pharmaceutical productsby HPLC after precolumn derivatization withp-toluenesulfonyl isocyanate

A simple and reliable HPLC method was developed for the simultaneous quantita-tive analysis of diethylene glycol (DEG) and propylene glycol (PG) in pharmaceuticalproducts by precolumn derivatization. The derivatization reagent p-toluenesulfonylisocyanate (TSIC, 10 lL, 20% in ACN v/v) was added to 100 lL of the sample, andthen 10 lL of water was added. The resulting derivatives were separated using a C18

analytical column and a mobile phase composed of 0.01 M KH2PO4 buffer (adjustedto pH 2.5 with phosphoric acid) and ACN (47:53 v/v) at 1 mL/min and 258C. For detec-tion, UV light at 227 nm was used. The derivatization conditions including reactiontime, temperature, and concentration of TSIC were optimized. The calibrationcurves were linear from 0.062 to 18.6 lg/mL (r2 = 0.9999) and from 0.071 to 21.3 lg/mL (r2 = 0.9999) for DEG and PG, respectively. The RSD values of intra- and interdayassays were all below 4% for DEG and PG. The proposed method was then success-fully applied to analyze two Armillarisin A injection samples and two spiked syrupsamples.

Keywords: Derivatization / Diethylene glycol / HPLC / Propylene glycol / p-Toluenesulfonyl isocya-nate /

Received: March 7, 2007; revised: July 13, 2007; accepted: July 14, 2007

DOI 10.1002/jssc.200700097

1 Introduction

Diethylene glycol (DEG) is a clear, slightly viscous, highlyhygroscopic, and odorless liquid with many commercialuses such as antifreeze, solvent, humectant, hydraulicfluid and brake fluid, and so on. Accidental or suicidalingestion of DEG may cause neuro- and nephrotoxicity inhuman [1]. The structural formula of DEG is shown inFig. 1. Propylene glycol (PG) is a clear, hygroscopic, taste-less, odorless, and colorless clear oily liquid, commonlyused as moisturizer, lubricant, emulsification agent,additive, antifreeze, and solvent. The Food and DrugAdministration has determined it as “generally recog-nized as safe” for use in food, cosmetics, and medicines[2]. The structural formula of PG is shown in Fig. 1. PG ismore expensive than DEG, and therefore is profitable to

be substituted by DEG by pharmaceutical companieswithout strict manufacturing oversight. As a PG, glycerinor ethanol substitution, or as a contaminant in variouspharmaceutical products, DEG has been responsible formany mass poisonings (Table 1) [3–10]. In the Haiti out-break, the estimated lethal dose ingestion of DEG was1.63 mg/kg body weight, with a range of 0.35–5.40 mg/kg [9]. The developing countries, where most of the DEGmass poisonings have occurred, with fewer resources toimplement a much stricter pharmaceutical manufactur-ing oversight, will continue to be at a higher risk forsuch poisoning epidemics [11].

From 19th to 30th April 2006, nine patients died ofacute renal failure after an injection of Armillarisin Acontaining DEG in the 3rd Affiliated Hospital of Zhong-shan University (Guangzhou, China). It was supposed tobe caused by the insufficient quality control of raw mate-rials by the manufacturer [12]. Therefore, the need of areliable and inexpensive method for the simultaneousdetermination of DEG and PG is urgent.

Litchfield [13] has developed a GC method to simulta-neously analyze DEG and PG in blood. After this method,several other GC [14–17] methods have been published

Correspondence: Professor Gengli Duan, School of Pharmacy,Fudan University, 130 Yixueyuan Road, Shanghai 200032, PRChina.E-mail: [email protected]: +86-21-54237208

Abbreviation: TSIC, p-toluenesulfonyl isocyanate

i 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

2620 T. Zhou et al. J. Sep. Sci. 2007, 30, 2620 – 2627

J. Sep. Sci. 2007, 30, 2620 –2627 Liquid Chromatography 2621

for the determination of DEG in pharmaceutical prod-ucts, wines, manufactured food products, and humanplasma, most of which involved trimethylsilyl deriv-atives of DEG, glycol extraction from aqueous solutions,and evaporation, resulting in low recoveries [17]. GCcombined with MS (GC/MS) [18–19] was the best choicefor the confirmation of DEG, especially in humanplasma. TLC followed by sample extraction [20] and IRspectroscopy [21] can also be used as an alternative. Espe-cially, Kenyon et al. [22] developed a method to detectDEG contamination of glycerin and glycerin-based rawmaterials by TLC.

Meanwhile, several Fourier transforms near-infraredspectroscopic methods [23] and GC methods [24, 25] havebeen reported for analyzing PG in pharmaceutical prod-ucts, human blood, and urine samples.

In this study, we developed an HPLC method to analyzethe DEG and PG in the Armillarisin A injection samplesfrom Qiqihar No. 2 Pharmaceutical and Xi'an Lijun Phar-maceutical.

2 Experimental

2.1 Chemicals and reagents

p-Toluenesulfonyl isocyanate (TSIC) was purchased fromSigma–Aldrich (Dorset, UK). The derivatization reagentconsisted of 20% v/v TSIC in ACN and was freshly pre-pared. ACN (HPLC grade) was purchased from Merck(Gibbstown, NJ, USA). DEG, PG, monobasic potassiumphosphate (KH2PO4), and phosphoric acid were obtainedfrom Sinopharm Medicine Chemical Reagent (Shanghai,China). Armillarisin A injection (lot no. 06030501, Armil-larisin A injection A) was obtained from Qiqihar No. 2Pharmaceutical (Qiqihar, China). Armillarisin A injec-tion (lot no. 040515, Armillarisin A injection B) wasobtained from Xi'an Lijun Pharmaceutical (Xi'an, China).Paracetamol syrup (lot no. 060528166) containing PG asan excipient was purchased from Shanghai Johnson-Johnson Pharmaceuticals (Shanghai). Cough syrup (lotno. 20060303) without PG was purchased from JiangxiTengwangge Pharmaceutical (Nanchang, China). Waterwas purified using Milli-XQ equipment (Millipore, Bed-ford, MA, USA).

2.2 Apparatus and HPLC procedure

Chromatographic analyses were performed on an Agi-lent 1100 LC system (Agilent, Palo Alto, CA, USA) that wasequipped with a G1310A bin pump, a G1322A vacuumdegasser, a G1316A thermostated column compartment,a VWD variable wavelength UV/VIS detector, and an HP1100 series manual injector 20 lL fixed loop. The detec-tor was set at 227 nm and the peak areas were integratedautomatically using the Hewlett –Packard ChemStationsoftware program (Rev. A. 08. 03 [847]). The other appara-tus used included a Radiometer NEL pH 890 digital pHmeter that was equipped with a combined glass–calomelelectrode and an ultrasonic bath.

The analytical column was a DIAMONSIL C18 column(15064.6 mm2 id, 5 lm; Dikma, Beijing, China). Themobile phase composed of 0.01 M KH2PO4 buffer(adjusted to pH 2.5 with phosphoric acid) and ACN (47:53v/v). The flow rate was 1 mL/min. Chromatography was


Table 1. Some of the DEG mass poisonings that occurred in the 20th century

Year Nation Pharmaceuticalproducts

Number of deaths Substituted oradulterated materials

Reference

1937 United States Sulfanilamide elixir 107 Ethanol [3, 4]1988 India Glycerin ingestion 14 Glycerin [5]1990 Nigeria Paracetamol elixir 47 Unknown [6]1990 –1992 Bangladesh Paracetamol elixir 236 PG [7]1992 Argentina Propolis syrup 15 Ethanol [8]1996 Haiti Paracetamol syrup 85 Glycerin [9]1998 India Paracetamol syrup At least 33 Unknown [10]

Figure 1. Structural formulas of DEG, PG, and TSIC.


performed at 258C. The buffer was filtered through a0.45 lm pore size membrane filter, prior to mixing.

2.3 Standard solutions

DEG and PG were dissolved in ACN as stock solutions(18.6 lg/mL DEG and 21.3 lg/mL PG). These solutionswere all stored at 48C. Working standard solutions werefreshly prepared by serial dilution of the stock solutionwith ACN.

2.4 Sample preparation

The derivatization reagent TSIC (10 mL, 20% in ACN v/v)was added to 100 lL of the sample in a plastic tube. Thetube was vortex-mixed for 30 s and allowed to stand atroom temperature for 10 min. To eliminate the excessderivatization reagent, 10 lL of water was added and thetube was vortex-mixed for 30 s again. The resulting deriv-ative of 20 lL was injected into the HPLC system.

2.5 Analysis of Armillarisin A injection and spikedcommercial syrup

For Armillarisin A injection, 12.19 mg of Armillarisin Ainjection A or 10.53 mg of B was transferred to a 100 mLvolumetric flask and diluted with ACN, sonicated for5 min, and then filled to volume with the same solventto make the solution I, respectively. To a 10 mL volumet-ric flask, 1 mL of the solution I was transferred, anddiluted with ACN to make the solution II. Solution II wasderivatized according to Section 2.4 and injected for theHPLC analysis.

For the paracetamol syrup, 11.2 mg of sample wastransferred to a 200 mL volumetric flask and dilutedwith ACN, sonicated for 5 min and then filled to volumewith the same solvent. Then the solution was derivatizedaccording to Section 2.4 and injected for the HPLC anal-ysis.

For the cough syrup sample, 12.8 mg of sample wastransferred to a 100 mL volumetric flask and dilutedwith ACN, sonicated for 5 min, and then filled to volumewith the same solvent to make the solution I. To a 10 mLvolumetric flask, 1 mL of the solution I was transferredand diluted with ACN to make the solution II. Solution IIwas derivatized according to Section 2.4 and injected forthe HPLC analysis.

The amount of DEG and/or PG in the injections and syr-ups was calculated from the related linear regressionequation.

3 Results and discussion

3.1 Derivatization

3.1.1 Derivatization reagent

Before the HPLC analysis, derivatization was required forDEG and PG to enable UV detection. Sulfonyl isocyanatessuch as TSIC are a series of compounds with very strongnucleophilic activity. The rate of reaction of sulfonyl iso-cyanates with hydroxyl-containing compounds isextremely fast [26, 27]. The reaction generates p-toluene-sulfonylcarbamic ester, which is very suitable for theHPLC-UV analysis. Since sulfonyl isocyanates could reactwith water rapidly to generate sulfonamides, the excessderivatization reagent could be eliminated by the hydrol-ysis reaction. Therefore, TSIC was selected for the deriva-tization of DEG and PG. The structural formula of TSIC isshown in Fig. 1.

3.1.2 Optimization of derivatization conditions

To the best of our knowledge, the reaction conditions,namely, temperature, time, and concentration of TSICare the prime parameters affecting the derivatization ofTSIC with hydroxyl-containing compounds. In this work,optimization of the derivatization parameters was per-formed. The derivatization reagent TSIC (20% in ACN v/v)was added to the standard solution with 18.6 lg/mL DEGor 21.3 lg/mL PG in ACN (the highest concentration usedfor the calibration curve) in a plastic tube, respectively.Then the tube was vortex-mixed and allowed to stand atdifferent temperatures for 10 min. After adding water, itwas injected for the HPLC analysis as described above.The result showed that the rate of derivatizationincreased with temperature from 10 to 208C, and was sta-ble for temperatures from 20 to 608C (Fig. 2A).

Similarly, to evaluate the effect of reaction time on thederivatization, the derivatization reagent (20% in ACNv/v) was added to the standard solution with DEG or PGat the highest concentration used for the calibrationcurve in a plastic tube. Then the tube was vortex-mixedand allowed to stand at room temperature (208C) fordifferent time periods. The result shows that, at the reac-tion temperature of 208C, 10 min was enough to com-plete the derivatization reaction for DEG and PG (Fig. 2B).

To study the effect of concentration of TSIC on the deri-vatization, the derivatization reagent with a concentra-tion of 2, 5, 10, 15, 20, and 30% of TSIC (in ACN v/v) wasadded to the standard solution with DEG or PG at thehighest concentration used for calibration curve in aplastic tube. Then the tube was vortex-mixed and allowedto stand at room temperature (208C) for 10 min. Theresult shows that, at the reaction temperature of 208C,15% of TSIC was sufficient to complete the derivatizationreaction for DEG and PG (Fig. 2C), and 20% of TSIC waschosen.



3.1.3 Effect of water in aqueous samples

Owing to the potential neutralization of the derivatiza-tion reagent by water in the aqueous samples, the aque-ous injection and syrup will behave differently compared

to the ACN. To study the water influence, the DEG and PGstandard solutions diluted with pure ACN and withwater–ACN mixture (1:100, 1:200, 1:600, 1:1000, 1:2000v/v) were assayed and compared (Table 2). Statistical anal-


Figure 2. Effects of (A) temperature, (B) time, and (C) con-centration of TSIC on the reaction of TSIC with DEG andPG.

Table 2. Water influences in the assay of the DEG and PG in the aqueous samples

Spiked concentra-tion (lg/mL)

Obtained concen-tration of analytesin pure ACN

Obtained concentration of analytes in water –ACN mixture

1:2000a) 1:1000 1:600 1:200 1:100

0.062 0.063 0.062 0.061 0.055 – b) –DEG 0.62 0.62 0.61 0.61 0.56 0.41 0.30

18.6 18.4 18.3 18.2 17.2 13.4 9.60.071 0.072 0.071 0.070 0.062 – –

PG 0.71 0.70 0.69 0.68 0.62 0.47 0.3121.3 21.6 21.1 20.9 19.6 13.9 10.2

a) The ratio of water/ACN in water –ACN mixture v/v.b) Unable to be quantified.


ysis of the results did not show any significant differencebetween the obtained concentration of DEG and PGstandard solutions diluted with pure ACN and withwater –ACN mixture (1:1000 and 1:2000 v/v). Therefore, a1000-fold dilution of aqueous samples with ACN is suffi-cient to avoid the water influence.

3.2 HPLC separation optimization

HPLC separation optimization was carried out in orderto validate an efficient method for analysis of the DEGand PG. Parameters such as detection wavelength,mobile phase, and optimum pH have been studied. Sev-eral binary mobile phases using different percentages ofACN or methanol and several aqueous buffers (phos-phoric acid, formic acid, and acetic acid at pH rangedfrom 2.0 to 5.0) were investigated. Among the differentmobile phases tested, the ACN and the KH2PO4 buffer(pH 2.5) were chosen, because shorter retention timeswere obtained with well resolved and narrow peaks. Thepercentages of the ACN and the buffer in the mobilephase were also optimized to achieve the separation ofanalytes as soon as possible with an optimum resolution.The detection at 227 nm in UV was chosen on the basis ofthe maximal UV absorption of the derivatives of DEG andPG.

3.3 Validation

3.3.1 Method linearity

We prepared the calibrators by diluting DEG and PG inACN. Standard solutions with concentrations of 0.062,

0.186, 0.62, 1.86, 6.2, 18.6 lg/mL DEG and 0.071, 0.213,0.71, 2.13, 7.1, 21.3 lg/mL PG were assayed. The calibra-tion curves were obtained by plotting the peak areas ofDEG and PG against their concentrations. The linearregressions were also calculated. A least-squares linearregression evaluation of the area ratio (Y) versus concen-tration (X) relationship gave Y = 203.8X + 2.63(r2 = 0.9999) for DEG, and Y = 286X – 1.36 (r2 = 0.9999) forPG, respectively. The contents of DEG as impurity in thepharmaceutical products may be much lower than thecontents of it as replacement of PG. Because the methodwas developed to assay the DEG in both kind of cases, thewide concentration range of calibration curves is useful.

3.3.2 LOQs and LODs

The LOQs were found to be 0.062 lg/mL for DEG and0.071 lg/mL for PG (both are the highest concentrationsused for the calibration curve), derived from 106S/N.The LODs were found to be 0.019 lg/mL for DEG and0.022 lg/mL for PG, derived from 36S/N. Because theLOQ and LOD are also related to the necessary dilution ofaqueous samples, they are at most 1000 times higher foraqueous samples than for nonaqueous samples. Becausethe concentrations of DEG in pharmaceutical productsranged from 14.4 to 72% in known mass poisoning [3–10], the LOQ for DEG in aqueous samples was still lowenough for analyzing them.

3.3.3 Precision and accuracy

The intra- and interday precision of the DEG and PG wereevaluated (Table 3). The standard solutions diluted withACN at concentrations of 0.062, 0.62, and 18.6 lg/mL forDEG, and 0.071, 0.71, and 21.3 lg/mL for PG wereassayed. The interday precision was evaluated for fivecontinuous days in a week. SD and RSD values were calcu-lated using standard methods. All RSDs of intra- andinterday assays were below 4%.

The accuracy of the method was evaluated by a recov-ery study of DEG and PG spiked at the levels of 10 and20% w/w in the Armillarisin A injection A, the Armillari-sin A injection B, the paracetamol syrup, and the coughsyrup. The average recoveries were calculated by the fol-lowing equation: recovery (%) = [(amount obtained –original amount)/amount spiked]6100%. The results arelisted in Table 4.


Table 3. Precision (n = 5) of spiked DEG and PG in ACN

Spikedconcentration(lg/mL)

Obtainedconcentration(lg/mL)(mean l SD)

RSD ofintradayassay (%)

RSD ofinterdayassay (%)

0.062 0.063 l 0.002 2.4 2.5DEG 0.62 0.62 l 0.01 2.0 2.4

18.6 18.3 l 0.3 1.9 2.00.071 0.073 l 0.002 2.3 3.0

PG 0.71 0.70 l 0.01 2.1 2.421.3 21.2 l 0.4 1.8 1.9

Table 4. Accuracy (n = 5) of DEG and PG

Amount spiked(% w/w)

Armillarisin Ainjection A recovery(%) (mean l SD)

Armillarisin Ainjection B recovery(%) (mean l SD)

Paracetamol syruprecovery (%)(mean l SD)

Cough syruprecovery (%)(mean l SD)

DEG 10 98.5 l 1.15 98.6 l 2.12 97.9 l 1.79 99.1 l 2.3120 98.3 l 3.87 99.7 l 3.54 98.3 l 2.89 98.7 l 2.59

PG 10 99.9 l 1.20 98.1 l 1.43 98.3 l 2.34 98.3 l 2.3420 99.1 l 2.23 98.5 l 2.51 98.2 l 2.76 98.5 l 2.64


3.3.4 Selectivity

Representative chromatograms of blank (derivatizationwithout any added DEG or PG) and ACN with DEG and PGboth assayed as described above (9.3 lg/mL DEG and10.7 lg/mL PG) are shown in Fig. 3A and B. DEG and PGwere both well separated with good symmetry withretention times of 5.6 and 6.6 min, respectively. Asshown in Fig. 3A, no significant direct interference in theblank was observed at the retention time of the analytes.The analysis was completed within 12.8 min. Figure 3C–F show the chromatograms of content assays of DEG and/or PG in Armillarisin A injection A, Armillarisin A injec-tion B, paracetamol syrup, and cough syrup.


Figure 3. HPLC chromatograms resulting from the analysis of: (A) blank (DEG and PG free); (B) standard sample (equivalent to9.3 lg/mL DEG and 10.7 lg/mL PG); (C) determination of DEG and/or PG in Armillarisin A injection A; (D) determination of DEGand/or PG in Armillarisin A injection B; (E) determination of DEG and/or PG in the spiked commercial paracetamol syrup sample;and (F) determination of DEG and/or PG in the spiked commercial cough syrup sample: peaks I and II refer to the derivatives ofDEG and PG, respectively.

Table 5. Quantitatively analytical results of DEG and PG inArmillarisin A injection

Armillarisin Ainjection A

Armillarisin Ainjection B

Obtainedconcentra-tion (% w/w)

RSD(%, n = 5)

Obtainedconcentra-tion (% w/w)

RSD(%, n = 5)

DEG 31.1 3.8 nda) –PG nd – 49.4 3.5

a) nd: Not detected.


3.3.5 Stability studies

Stability of derivatives was determined by checking theirconcentrations within three successive days of storage at+48C and the data were compared with freshly preparedsamples. Statistical analysis of the results did not showany significant difference. This indicates that the deriv-atives of DEG and PG are stable for at least 3 days.

3.4 Application to the pharmaceutical products

3.4.1 Analysis of Armillarisin A injection producedby two different companies

The developed method was applied to analyze authenticsamples, Armillarisin A injection A and Armillarisin Ainjection B. The Armillarisin A injection A caused the poi-soning of 64 patients out of which 9 died in Guangzhou,China, and the Armillarisin A injection B was producedby another company.

An unpublished report by China State Food and DrugAdministration (SFDA) said that the content of DEG inArmillarisin A injection A was 30% w/w. In this study, theobtained contents of DEG were 31.1% w/w in ArmillarisinA injection A and not detected in B. The obtained con-tents of PG were 49.4% w/w in Armillarisin A injection Band not detected in A (Table 5). Considering the investiga-tions by China SFDA, Drug Administration of GuangzhouMunicipality and the 3rd Affiliated Hospital of Zhong-shan University, and the result in this work, the DEG con-tamination in Armillarisin A injection as a substitutionfor PG may be the cause of the mass poisoning and deathdescribed in Section 1.

3.4.2 Analysis of spiked commercial paracetamolsyrup and cough syrup sample

In order to make sure that the proposed method couldalso be used to analyze the syrup samples, it was appliedto determine the DEG and/or PG in them. We madespiked commercial paracetamol syrup and cough syrupsamples by adding known amounts of DEG to simulatesyrups contaminated with DEG, and we did not spike PG

in any sample. The quantitatively analytical results arelisted in Table 6.

4 Concluding remarks

For the first time, an HPLC method following derivatiza-tion with TSIC was developed for the determination ofDEG and PG. After thorough optimization and valida-tion, the procedure was successfully applied to the deter-mination of DEG and PG in two Armillarisin A injectionsamples from different producers; one of which causedmass poisoning and death in Guangzhou, in April 2006,and in the spiked paracetamol syrup and cough syrupsamples. It has been shown that the proposed methodwas a reliable method that could be used for quality con-trol of the pharmaceutical products.

5 References

[1] Morris, J. A., Nelson, A. A., Calvery, H. O., J. Pharmacol. Exp. Ther.1942, 74, 266 – 273.

[2] U. S. Food and Drug Administration, Fed. Regist. 1982, 47, 27819.

[3] Geiling, E. M. K., Cannon, P. R., J. Am. Med. Assoc. 1938, 111, 919 –926.

[4] Wax, P. M., Ann. Intern. Med. 1995, 112, 456 – 461.

[5] Pandya, S. K., BMJ 1988, 297, 117 – 119.

[6] Okuonghae, H. O., Ighogboja, I. S., Lawson, J. O., Nwana, E. J.,Ann. Trop. Paediatr. 1992, 12, 235 – 238.

[7] Hanif, M., Mobarak, M. R., Ronan, A., Rahman, D., Donovan, J. J.,Jr, Bennish, M. L., BMJ 1995, 311, 88 – 91.

[8] Ferrari, L. A., Giannuzzi, L., Forensic. Sci. Int. 2005, 153, 45 – 51.

[9] O'Brien, K. L., Selanikio, J. D., Hecdivert, C., Placide, M. F., Louis,M., Barr, D. B., Barr, J. R., Hospedales, C. J., Lewis, M. J., Schwartz,B., Philen, R. M., St Victor, S., Espindola, J., Needham, L. L., Dener-ville, K., JAMA 1998, 279, 1175 – 1180.

[10] Singh, J., Dutta, A. K., Khare, S., Dubey, N. K., Harit, A. K., Jain, N.K., Wadhwa, T. C., Gupta, S. R., Dhariwal, A. C., Jain, D. C., Bhatia,R., Sokhey, J., Bull. World Health Organ. 2001, 79, 88 – 95.

[11] Wax, P. M., J. Toxicol. Clin. Toxicol. 1996, 34, 517 – 520.

[12] Zhang, E., Cao, J., Huang, L., China Pharm. 2006, 17, 850 – 851.

[13] Litchfield, M. H., Analyst 1968, 93, 653 – 659.

[14] Lawrence, J., Chadha, R. K. Y., Lau, P., Weber, D. F., J. Chromatogr.1986, 367, 213 – 216.


Table 6. Quantitatively analytical results of DEG and PG in the spiked commercial paracetamol syrup and cough syrup

Spiked commercial paracetamol syrup Spiked commercial cough syrup

Spikedconcentration(% w/w)

Obtainedconcentration(% w/w)

Accuracy(%)

RSD(%, n = 5)

Spikedconcentration(% w/w)

Obtainedconcentration(% w/w)

Accuracy(%)

RSD(%, n = 5)

DEGa) 13.3 13.1 –1.5 2.7 17.6 17.3 –1.7 3.1PGb) – 0.953 – 3.5 – ndc) – –

a) We spiked known amounts of DEG in the spiked commercial paracetamol syrup and cough syrup.b) We did not spike PG in the spiked commercial paracetamol syrup and cough syrup.c) nd: Not detected.


[15] Castle, L., Cloke, H. R., Startin, J. R., Gilbert, J., J. Assoc. Off. Anal.Chem. 1988, 71, 499 – 502.

[16] Brooks, J. B., Basta, M. T., Alley, C. C., Holley, J. S., El Kholy, A. M.,J. Chromatogr. 1984, 309, 269 – 272.

[17] Baffi, P., Elneser, S., Baffi, M., De Melin, M., Baffi, P., Elneser, S.,Baffi, M., De Melin, M., J. AOAC Int. 2000, 83, 793 – 801.

[18] Maurer, J. E., Kessler, C., Arch. Toxicol. 1988, 62, 66 – 69.

[19] Maurer, H. H., Peters, F. T., Paul, L. D., Kraemer, T., J. Chromatogr.B Biomed. Sci. Appl. 2001, 754, 401 – 409.

[20] Stahl, E., Thin-Layer Chromatography, Springer-Verlag, New York1969.

[21] Hummel, D., Identification and Analysis of Surface-Active Agents byInfrared and Chemical Methods, Interscience Publishers, New York1962.

[22] Kenyon, A. S., Xiaoye, S., Yan, W., Wai Har, N. G., Prestidge, R.,Sharp, K., J. AOAC Int. 1998, 81, 44 – 49.

[23] Broad, N. W., Jee, R. D., Moffat, A. C., Eaves, M. J., Mann, W. C.,Dziki, W., Analyst 2000, 125, 2054 – 2058.

[24] Houze, P., Chaussard, J., Harry, P., Pays, M., J. Chromatogr. 1993,619, 251 – 257.

[25] Yu, D. K., Sawchuk, R. J., Clin. Chem. 1983, 29, 2088 – 2090.

[26] Zuo, M., Gao, M. J., Liu, Z., Cai, L., Duan, G. L., J. Chromatogr. B Ana-lyt. Technol. Biomed. Life Sci. 2005, 814, 331 – 337.

[27] Ulrich, H., Chem. Rev. 1965, 65, 369 – 376.


simultaneous determination of diethylene glycol and propylene glycol in pharmaceutical products by...

Documents