Basavaiah K; Rajendraprasad N

Research Article

Austin J Anal Pharm Chem. 2017; 4(1): 1082.

High Performance Liquid Chromatographic Assay of Chlorpropamide, Stability Study and its Application to Pharmaceuticals and Urine Analysis

Basavaiah K¹* and Rajendraprasad N²

¹Department of Chemistry, University of Mysore, Mysuru, Karnataka, India

²PG Department of Chemistry, JSS College of Arts, Commerce & Science, B N Road, Mysuru, Karnataka, India

*Corresponding author: Basavaiah K, Department of Chemistry, University of Mysore, Manasagangothri, Mysuru, Karnataka, India

Received: February 27, 2017; Accepted: March 16, 2017; Published: March 22, 2017

Abstract

Chlorproamide (CLP) is a sulphonyl-urea derivative used in the treatment of type 2 diabetes mellitus. A rapid, sensitive and specific HPLC method with uv detection is described for the determination of CLP in bulk and tablet forms. The assay was performed on an Inertsil ODS 3V (150mm × 4.6mm; 5μm particle size) column using a mixture of phosphate buffer (pH 4.5), methanol and acetonitrile (30:63:7v/v/v) as mobile phase at a flow rate of 1mLmin^-1 and with uv detection at 254nm. The column temperature was 30?C and injection volume was 20μL. The retention behaviour of CLP as a function of mobile phase pH, composition and flow rate was carefully studied, and chromatographic conditions, yielding a symmetric peak with highest number of theoretical plates, were optimized. The calibration curve was linear (r = 0.9999) over the concentration range 0.5 – 300 μgmL^-1. The limits of detection (LOD) and quantification (LOQ) were found to be 0.1 and 0.3 μgmL^-1, respectively. Both intra-day and inter-day precisions determined at three concentration levels were below 1.0% and the respective accuracies expressed as %RE were =1.10%. Assays were performed under slightly altered chromatographic conditions, and the results were not significantly different from those obtained under optimum conditions, reflecting the robustness of the method. Inter-equipment and interanalysts deviations were insignificant testifying the ruggedness of the method. The method was validated for selectivity via placebo blank and synthetic mixture analyses. The method was applied to determine CLP in tablets and the results were comparable with those obtained by a reference method with respect to accuracy and precision. As part of stress testing, CLP was subjected to forced degradation under acid and base-induced hydrolysis, oxidation, heat and light, followed by assay using the developed method, and the results indicated that the drug was prone to oxidative degradation, but inert to other stress-conditions. The method was sensitive and selective enough to determine the drug content in spiked urine, and the results were satisfactory.

Keywords: Chlorpropamide; Assay; HPLC; Pharmaceuticals; Stabilityindicating

Introduction

Chlorpropamide (CLP), chemically called 4-Chloro-N- [(propylamino) carbonyl]benzenesulfonamide (Figure 1), is one of the sulphonyl-urea derivatives, which are widely used as hyperglycemic drugs used in the treatment of non-insulin dependent diabetes mellitus [1]. CLP is found to stimulate pancreatic β-cell insulin production, which causes the reduction of glucose levels in blood.

Figure 1: Structure of CLP.

    
    
    Figure 1: Structure of CLP.

Therapeutic use of the drug has resulted in the development of the many methods for its determination in body fluids and includes high performance liquid chromatography [2-7], gas-liquid chromatography [8-11], liquid chromatography-tandem mass spectrometry [12], micellar electrokinetic chromatography [13], electro spray mass spectrometry [14] and gas chromatography-mass spectrometry [15].

The drug has an official monograph in the United States Pharmacopeia [16] which has adopted an HPLC-uv detection method for the assay of CLP. British Pharmacopeia [17] describes a spectrophotometric method whereas European Pharmacopeia [18] offers a titrimetric method for the assay in tablets and bulk drug, respectively. In addition to the above, CLP in pharmaceuticals has been assayed by several other techniques such as titrimetry [19], spectrophotometry [20-23], electrophoresis [24], thinlayerchromatoghraphy [25, 26] and coulometry [27].

HPLC, which finds immense applications in the analysis of body fluids for CLP [2-7], has also been applied to the determination of drug in pharmaceuticals, chiefly due to its sensitivity, selectivity, speed and wide applicability. El-Gizawy [28] has reported an HPLC method for the determination of CLP in tablets. The author used a Zorbax SAX anion exchange column with phosphate buffer (pH 6.5) containing 1.7×10^-2M β-cyclodextrin/methanol of ratio 85:15 as mobile phase. In a method described by Kumasaka et al. [29], a gradient RP-HPLC system equipped with a PAD using ODS column and acetonitrile-ammonium acetate buffer as a mobile phase, was employed to determine the drug in tablets with recovery in the range of 90.7- 105.7%. After homogenization of the sample with methanol, the drug in tablets was assayed by Molins et al. [30].

A quantitative HPLC method for the determination of CLP and its hydrolysis products: p-chlorobenzene sulphonamide and p-toluene sulphonamide, in solid dosage forms was developed by Robertson et al. [31]. However, the method was not applied for the assay of drug in tablets. Dasgupta [15] developed an RP-HPLC method for CLP in tablets. There was no interference from the products of hydrolysis and the excipients present in tablets.

Isocratic separation and simultaneous estimation of four antidiabetics belonging to sulphonyl urea family viz. CLP, to lbutamide, glipizide and gliclazide was achieved by Nadkarni et al. [32] using a Metaphase Bioscience column (250×4.6mm; 5μm) as stationary phase. The mobile phase was 0.02M KH₂PO₄ buffer, pH 3.0- methanol (40:60) with UV-detection at 230nm. A linear relationship was found in the range 0.4-10μg mL^-1. The method was extended to the dosage forms of the four drugs.

The reported HPLC methods [28-32] have not been validated according to the current ICH guidelines and many of them suffer from low sensitivity and narrow linear ranges of applicability. None of the above method is stability-indicating, though stress testing is mandatory as per the ICH recommendations [33-35].

The objective of the present work was therefore, to develop a simple and sensitive HPLC method for the determination of CLP in pharmaceuticals, and to subject the drug to stress-testing via acid-, base-hydrolysis, oxidation, heating and photolysis to determine the stability-indicating ability of the developed method.

Experimental

Apparatus and software

Chromatographic analysis was performed with a Waters HPLC system (Waters Corporation, Milford, USA) equipped with Alliances 2695 series low pressure quaternary gradient pump, a programmable variable wavelength UV detector and auto sampler. Data were collected and processed using Waters Empower 2 software.

Materials and reagents

Pure CLP (99.9%) was kindly supplied by Deys Medical Stores (Mfg) Ltd, Kolkata, India, as a gift. Two brands of tablets containing CLP, Copamide-250 (Deys Medical Stores Ltd. Kolkata, India) and Diabinese-250 (Pfizer Pvt. Ltd. Mumbai, India) used in the investigation were purchased from local markets.

HPLC grade methanol and acetonitrile were purchased from Merck Ltd., India. Potassium dihydrogen orthophosphate, orthophosphoric acid, hydrochloric acid, sodium hydroxide and hydrogen peroxide were from Qualigens-India. Water purified by the Milli-Q system (Millipore, Milford, Massachusetts, USA) was used for mobile phase and sample diluent preparation. One litre of 10mM potassium dihydrogen orthophosphate solution was prepared by dissolving 1.4g of the salt in water and the pH was adjusted to 4.5 using triethylamine or dilute phosphoric acid. A 300mL portion of this buffer was mixed with 630mL of methanol and 70mL of acetonitrile, shaken well and filtered using 0.22μm Nylon membrane filter.

Hydrochloric acid (HCl, 0.1M) and hydrogen peroxide (H₂O₂, 5% v/v) were prepared by appropriate dilution of concentrated acid (Specific gravity 1.18) and commercial sample (30%) with water, respectively. Sodium hydroxide (NaOH, 0.1M) was prepared by dissolving the required quantity of chemical (S.D. Fine Chem Ltds, Bengaluru, India) in water.

Chromatographic conditions

Chromatographic assay was achieved on an Intersil ODS 3V (150mm ×4.6mm; 5μm particle size) column. A solution containing a mixture of phosphate buffer of pH 4.5, methanol and acetonitrile (30:63:7v/v/v) was used asa mobile phase. Mobile phase was used as a diluent. The flow rate was 1.0mLmin^-1, the detector wavelength was set at 254nm and the injection volume was 20μL. The column temperature was maintained at 30°C.

Standard CLP solution

A stock standard solution containing 1000μg mL-1 CLP was prepared by dissolving accurately weighed 100mg of pure CLP in the mobile phase in a100mL calibrated flask.

General procedures

Procedure for bulk drug

Procedure for preparation of calibration curve: Working standard solutions containing 0.5-300μg mL^-1 CLP were prepared by appropriate dilution of stock solution. Aliquots of 20μL were injected (triplicate) and eluted with the mobile phase under the stated chromatographic conditions. The average peak area versus concentration was plotted. Alternatively, the regression equation was derived using mean peak area-concentration data and concentration of unknown was computed from the regression equation.

Procedure for tablets: Tablet powder equivalent to 20mg CLP was transferred into a 100mLvolumetric flask containing 60mL of the mobile phase. The mixture was sonicated for 20min to achieve complete dissolution of CLP, and the content diluted to volume with the same solvent to yield a concentration of 200μg mL^-1 CLP, and filtered through a 0.45μm nylon membrane filter. The tablet extract was injected on to the HPLC column in five replicates.

Procedure for placebo blank and synthetic mixture: A placebo blank of the composition of talc (15mg), starch (20mg), lactose (15mg), sodium alginate (20mg), calcium gluconate (15mg) and magnesium stearate (20mg) was made and its solution prepared as described under ‘assay procedure for tablets’ by taking about 20mg. To 10mg placebo, 10mg of pure CLP was added and homogenized, transferred to a 50mL calibrated flask and its extract prepared as described under “procedure for tablets”. Placebo blank and synthetic mixture solutions were injected on to the column in five replicates each.

Procedure for stress study: Ten mg of pure CLP was transferred separately into three different 50mLvolumetric flasks and dissolved with 10mL of mobile phase. Added 5mL of 0.1M HCl, 0.1M NaOH or 5% H₂O₂ separately, and the flasks were heated for 2h in a water bath maintained at 80°C. Then, the solutions were cooled, neutralized by adding base or acid and the volume in each flask was brought to the mark with mobile phase. An appropriate volume (20μL) was injected and chromatographed. Solid state thermal degradation was carried out by exposing pure drug to dry heat at 105°C for 2h. For photolytic degradation study, pure drug in solid state was exposed to 1.2 million lux hours in a photo stability chamber. The sample after exposure to heat and light was used to prepare 200μgmL^-1 solutions in mobile phase separately, and the chromatographic procedure was followed.

Procedure for analysis of spiked human urine: Twenty mg of pure CLP was taken in a 50mL volumetric flask containing 10mL of drug free-urine, 5mL of mobile phase and 20mL of methanol were added and the contents were mixed well and the volume was brought to the mark with diluent. The solution was filtered through 0.22μm nylon membrane filter. Aliquots of 20μL were injected (n = 5) and eluted with the mobile phase under the stated chromatographic conditions after diluting to 100, 200 and 300μg mL^-1 levels. The concentration of CLP was calculated using the regression equation, and percent recovery of CLP added determined.

Results and Discussion

With a view to achieve excellent linearity, sensitivity and selectivity, the method was optimized and validated in accordance with the current ICH guidelines [35]. The typical chromatograms obtained for blank and pure CLP under optimized HPL Cconditions are depicted in Figure 2.

Figure 2: Chromatograms for: a) Blank (mobile phase) b) Pure CLP solution (200 μg mL^-1).

    
    
    Figure 2: Chromatograms for: a) Blank (mobile phase) b) Pure CLP solution
(200 μg mL^-1).

Method development

The conditions affecting the retention behaviour of CLP were carefully studied and optimized so as to obtain satisfactory system suitability parameters such as retention time, number of theoretical plates and tailing factor. Mobile phase pH, composition and flow rate were the parameters of vital importance studied. The results of optimization of these parameters are compiled in Table 1.

Effect of pH and ionic strength of buffer

Mobile phases of pH ranging from 3.0-7.0, adjusted with sodium hydroxide or phosphoric acid were tried. The study revealed that the effect of pH of phosphate buffer was not critical over the pH range 4.5 to 6.5. Table 1 show that pH 4.5 was most appropriate and giving well defined peak and the highest number of theoretical plates. At lower and higher pH asymmetrical peak and smaller number of theoretical plates were observed.

The concentration of phosphate buffer

The effect of concentration of phosphate buffer on the chromatographic performance was investigated. Different concentrations in the range of 5 to 15mM were studied. The study revealed that the optimum chromatographic performance could be achieved with 10mM phosphate buffer. Although 5 and 8 mM gave the highest numbers of theoretical plates they were not selected due to the very poor reproducibility of retention times as well as the relatively high values of the tailing factors.

Type of organic modifier

Methanol or acetonitrile as organic modifier failed to yield a symmetric peak. A combination of methanol and acetonitrile (63:7 v/v) was the organic modifier of choice giving a symmetric and highly sensitive peak.

Ratio of organic modifier: The effect of ratio of organic modifier on the performance characteristics was investigated using mobile phases containing 30-63% methanol and 7-35% acetonitrile in different ratios. Table 1 shows that 63% methanol and 7% acetonitrile was the best, giving well defined peak and the highest number of theoretical plates. Different ratios of buffer and organic modifiers were tried and the optimum ratio was phosphate buffer, methanol and acetonitrile in the ration of 30:63:7v/v/v.

The effect of flow rate: The effect of flow rate on the symmetry, sensitivity and retention time of the peak was studied in the range of 0.5-1.5 mLmin^-1 and a flow rate of 1.0 mLmin^-1 was optimal for better symmetry and reasonable retention time (Table 1).

Table 1: Effect of ratio of organic modifier, pH and ionic strength of buffer on the number of theoretical plates.




  
    Ratio
      (A/B/C)^a
    Number
      of
      theoretical
      plates (N)
    pH of
      the
      medium
    Number
      of
      theoretical
      plates (N)
    Flow
      rate,
      mL
      min^-1
    Number
      of
      theoretical
      plates (N)
  
  
    15/50/35
      20/55/25
      25/60/15
      30/63/7
      35/45/20
      40/35/25
      40/30/30
    4212
      5571
      5685
      6154
      5718
      5572
      5015
    3.0
      3.5
      4.5
      5.5
      6.5
      6.8
      7.0
    3785
      3893
      6681
      5954
      5818
      4236
      4029
    0.50
      0.75
      0.85
      0.95
      1.00
      1.25
      1.50
    4172
      4643
      4795
      5763
      6199
      5328
      4927
  
  
    ^aA: Phosphate buffer; B: Methanol and    C-acetonitrile.



Table 1:  Effect of ratio of organic modifier, pH and ionic strength of buffer on the
number of theoretical plates.

Choice of column: Different columns including, Inertsil ODS 3V (150mm×4.6mm, 5μm particle size) column; Hypersil BDSC8 (250mm × 4.0mm, 5.0μm particle size) thermo column; Luna C18 (250mm×4.0mm, 5.0μm particle size); Chromatopack (250mm × 4.6mm, 5μm particle size) column and Zorbax XDB (250mm×4.0mm, 5.0μm particle size) column were tried for better performance characteristics. The first column (Inertsil ODS 3V) was found to be highly suitable with respect to suitability parameters.

Choice of detection wavelength: The UV detector response of CLP in methanol showed a maximum at 254nm with maximum absorptivity.

Method validation

Linearity: Linearity was studied by preparing standard solutions of different concentrations from 0.5 to 300 μg mL^-1, plotting a graph of mean peak area against concentration and determining the linearity by least-square regression quation. The calibration plot was linear over the concentration range 0.5-300 μg mL^-1 (n= 3) (Figure 3) and is described by the equation:

Figure 3: Calibration curve.

    
    
    Figure 3: Calibration curve.

y = m x + b

where y is the mean peak area, x the concentration of CLP in μg mL^-1, m slope and b intercept. The slope (m), y-intercept (b) and their standard deviations were calculated and are presented in Table 2. The calculated correlation coefficient (r = 0.9999) confirms the linear relationship between the peak areas and concentration of CLP.

Table 2: Linearity and regression parameters.




  
    Parameter
    Value
  
  
    Linear    range, μg mL^-1
      Limits of    detection, (LOD), μg mL^-1
      Limits of    quantification, (LOQ), μg mL^-1
      Regression    equation, y*
      Slope (m)
      Intercept    (b)
      Standard    deviation of intercept (S_b)
      Standard    deviation of slope (S_m)
      Correlation    coefficient (r)
    0.5 - 300
      0.10
      0.30
       
      44395
      211943
      169.2
      827.9
      0.9999
  
  
    *y=mx+b,    where y is the mean peak area, x concentration in μg mL^-1, b    intercept, m slope.



Table 2:  Linearity and regression parameters.

Precision and accuracy: Pure drug solution at three concentration levels was injected in seven replicates during the same day and over a period of five days (five injections per day) as part of intra-day and inter-day precision and accuracy study. Error and standard deviations were calculated for each data set. Pooled standard deviation was computed to determine the inter-day precision.

The percent relative error which is an indicator of accuracy is =1.1% and is indicative of high accuracy. The peak area-based and retention time-based %RSD values were <1% and reflective of very high repeatability and reproducibility of the developed method (Table 3).

Table 3: Results of accuracy and precision study.




  
    CLP
      injected,
      μg mL^-1 
    Intra-day (n = 7) 
    Inter-day (n = 5) 
  
  
    CLP
      found,
      μg mL^-1
    %
      RE^a
    %
      RSD^b
    %
      RSD^c
    CLP
      found,
      μg mL^-1
    %
      RE^a
    %
      RSD^b
    %
      RSD^c
  
  
    100
      200
      300
    100.5
      201.6
      301.9
    0.50
      0.80
      0.63
    0.54
      0.42
      0.39
    0.27
      0.36
      0.63
    101.1
      198.7
      302.1
    1.10
      0.65
      0.70
    0.45
      0.29
      0.36
    0.30
      0.45
      0.36
  
  
    ^aRelative    error; ^bRelative standard deviation based on peak area;^cRelative    standard deviation based on retention time.



Table 3:  Results of accuracy and precision study.

Method robustness: Robustness of the method was evaluated by performing analysis under slightly altered chromatographic conditions like, detection wavelength, flow rate, mobile phase composition and column temperature, and the results were compared with those of the optimized chromatographic conditions. The result expressed as intermediate precision (%RSD) is summarized in Table 4 and speaks of the robustness of the method.

Table 4: Results of method robustness reported as intermediate precision.




  
    Condition
      altered
    Modification
    Mean peak
      area
      ± SD*
    %
      RSD
    Mean R_t
      ± SD*
    %
      RSD
    Mean
      theoretical
      plates
      ± SD*
    %
      RSD
    Mean
      tailing
      factor
      ±SD*
    %
      RSD
  
  
    Actual
    -
    9069466
      ± 43327
    0.48
    4.902±
      0.004
    0.082
    6218±
      6.439
    0.10
    1.109±
      0.003
    0.27
  
  
    Column
      temperature
    30 and
      30 ± 2 ?C
    9067345
      ± 46618
    0.54
    4.921±
      0.002
    0.061
    6155±
      6.392
    0.11
    1.112±
      0.004
    0.36
  
  
    Mobile
      phase
      composition
    (Buffer:
      methanol)
      30:63:7
      28:65:5
      32:60:8
    9067945
      ± 45414
    0.51
    4.909±
      0.003
    0.061
    6236±
      5.732
    0.09
    1.115±
      0.003
    0.27
  
  
    Flow rate
    1.0 and
      1.0± 0.1 mL min^-1
    9068129
      ± 48196
    0.53
    4.918±
      0.002
    0.041
    6189±
      5.572
    0.09
    1.105±
      0.004
    0.36
  
  
    Wavelength
    254 and
      254 nm ±
      1 nm 
    9069992
      ± 52312
    0.58
    4.927±
      0.004
    0.081
    6234±
      6.963
    0.11
    1.101±
      0.003
    0.27
  
  
    *Mean    value of three determinations at CLP concentration of 200 μg mL^-1.



Table 4:  Results of method robustness reported as intermediate precision.

Method ruggedness: The ruggedness of the method was assessed by comparing results of assay performed by three analysts using the same column and also by a single analyst with three columns of the same type but belonging to different batches. The inter-analysts and inter-columns deviations were not significant. These results are presented in Table 5.

Table 5: Results of method ruggedness (n=3).




  
    Variable
    Mean
      Peak
      area
      ± SD*
    %RSD
    Mean
      Rt ±
      SD*
    %RSD
    Mean
      theoretical
      plates
      ±SD
    %RSD
    Mean
      tailing
      factor±
      SD*
    %RSD
  
  
    Analysts
      (n=3)
    9067945
      ± 45414
    0.51
    4.918±
      0.002
    0.041
    6234±
      6.963
    0.11
    1.115±
      0.003
    0.27
  
  
    Columns
      (n = 3)
    9278856
      ± 53628
    0.58
    5.216±
      0.003
    0.058
    6196±
      7.565
    0.12
    1.042±
      0.004
    0.38
  
  
    *Mean    value of three determinations at CLP concentration of 200 μg mL^-1.



Table 5:  Results of method ruggedness (n=3).

Selectivity: Selectivity of the method was evaluated by injecting the mobile phase, placebo blank, pure drug solution and tablet extract. No peaks were observed for mobile phase and placebo blank and no extra peaks were observed for tablet extracts (Figure 4). Synthetic mixture, when analysed at 200μgmL-1concentration level, yielded percent recoveries of 98.7% with standard deviation 0.81% indicating the absence of interference from the tablet excipients in the assay.

Figure 4: Chromatograms obtained for: a) placebo blank and b) tablet extract (200μgmL^-1 CLP).

    
    
    Figure 4: Chromatograms obtained for: a) placebo blank and b) tablet extract
(200μgmL^-1 CLP).

Solution stability: The drug solution stored at lab temperature (25±2°C) was injected (n = 3) at different time intervals of 0, 12 and 24 h, and chromatograms were recorded. At the specified time interval, RSD (%) for the peak areas was calculated, and pooled standard deviation was computed for three sets of data generated at the end of 24h period. Low values of %RSD (=1%) are reflective of stability of solution over a period of 24h, and no significant change in the elution time of the peak and system suitability parameters: retention time, theoretical plates and tailing factor. The results of this study are compiled in Table 6.

Table 6: Results of solution stability, reported as intermediate precision.




  
    Time, hour
    Mean
      Peak
      area
      ± SD*
    Pooled
      %RSD
    Mean
      Rt ±
      SD*
    Pooled
      %RSD
    Mean
      theoretical
      plates
      ±SD
    Pooled
      %RSD
    Mean
      tailing
      factor±
      SD*
    Pooled
      %RSD
  
  
    0
    2784999
      ± 8319
    0.62
    5.575±
      0.003
    0.078
    9019±
      5.775
    0.13
    1.118±
      0.003
    0.41
  
  
    12
    2792397
      ± 9118
    5.572±
      0.004
    9025±
      6.192
    1.122±
      0.005
  
  
    24
    2812795
      ± 8217
    5.583±
      0.004
    8954±
      5.861
    1.121±
      0.004
  
  
    *Mean    value of three determinations for CLP concentration of 200 μg mL^-1 at each time interval.



Table 6:  Results of solution stability, reported as intermediate precision.

Application to tablets: The developed method was applied to the determination of CLP in two brands of tablets. The same tablet powder was assayed by European Pharmacopoeial method [18] for comparison. The comparison method is based on the titration of drug in alcohol against 0.1M sodium hydroxide to a phenolphthalein end point. The results were compared statistically by applying the Student’s test for accuracy and F-test for precision. As shown by the results compiled in Table 7, proposed method and the reference method do not differ significantly with respect to accuracy and precision, at the 95% confidence and for 4 degrees of freedom.

Formulation brand name	Nominal amount, mg	CLP found (%) ± SD (n = 5)
Formulation brand name	Nominal amount, mg	Reference method	Proposed method	t-value	F- value
Copamide	250	99.81±1.25	99.58±1.46	0.27	1.36
Diabenese	250	99.79±0.97	100.3±1.27	0.72	1.71
Tabulated t-value at 95% confidence level is 2.77; Tabulated F-value at 95% confidence level is 6.39.

Tablet studied	CLP in tablet, �g mL^-1	Pure CLP added, �g mL^-1	Total found, �g mL^-1	Pure CLP recovered* (%NTG ±SD)
Copamide	99.58	50	150.9	102.6±1.04
	99.58	100	202.9	103.3±0.81
	99.58	150	247.6	98.68±1.09
Diabenese	100.3	50	152.6	104.6±1.05
	100.3	100	201.3	101.0±0.63
	100.3	150	248.9	99.07±0.72
*Mean value of three determinations.

Degradation condition	% Degradation
Degradation condition	% Degradation
Acid hydrolysis	No degradation
Base hydrolysis	No degradation
Oxidation	19.2
Heat (105�C, 3 hours)	No degradation
Light (1.2 million lux hours)	No degradation