Class A comparison - Microlab Technologies
Contents
1. Installation operation cleaning and maintenance Install syringes according to the manual s instructions Operation parameters are dependant upon the type of fluids used Liquids with low vapor pressures will require slower fill aspiration speeds in order to avoid degassing the fluid while in the fluid path Liquids with high viscosity may require lower speeds in order to avoid overloading the syringe drives Bubbles in the fluid path may affect accuracy especially if they break loose and are dispensed Cleaning the fluid path may prevent bubble formation Syringe plunger tips can be easily damaged if not handled properly Pre wet the tip before installation into the barrel Avoid scratching or marring the plunger tip replace damaged plunger assemblies with new ones Leaks can result from a damaged tip thereby affecting accuracy and precision Rinsing the fluid path especially between applications with different fluids and after a work shift where salt solutions are used will prevent damage to the syringe plunger tips and the valves Halogenated solvents if left in the fluid path may reduce the life of the adhesive between the glass syringe barrel and the TLL fittings a thorough rinsing of these fluids after use is required In comparison with volumetric glassware the MICROLAB 500 provides extraordinary cost savings and sample preparation time reductions Its accuracy tolerances far exceed those of volu metric2. flasks graduated cylinders measuring pipets and burets Only transfer pipets can shower better accuracies The preci sion of the ML500 meets or exceeds that of Class A glassware especially when operator imprecision is considered that is the automated nature of the ML500 eliminates operator to operator inconsistencies Although the ML500 cannot claim better accuracy in all com parisons with Class A apparatus its benefits weigh heavily in favor of automating small volume sample preparation This is especially true since validations of individual preparation meth ods are relatively simple References 1 The United States Pharmacopeia USP 23 The National Formulary NF 18 United States Pharmacopeial Conven tion Inc Rockville MD 1995 2 1996 Annual Book of ASTM Standards Section 14 Gen eral Methods and Instrumentation Volume 14 02 American Society for Testing and Materials West Conshohocken PA 1996 3 Official Methods of Analysis of the Association of Official Analytical Chemists Fifteenth edition AOAC Arlington VA 1990 4 Hamilton MICROLAB 500B C Series User s Manual Revi sion C Instrument Calibration Diluters and dispensers such as the MICROLAB 500 can be periodically calibrated using the following procedure which is a gravi metric test based on a mix of Hamilton s QC method and the method outlined in ASTM E1154 Standard Specification for Piston or Plunger Operated Volumetric Appar
3. volumetric glassware USP methods specify the use of volumetric apparatus unless automated devices can demonstrate equivalent performance The following is a validation that the ML500 can be a preferred alternative to pipets burets and volumetric flasks Table 1 Accuracy specifications for Class A volumetric glassware The applicable ASTM standards are referenced in parentheses Tolerance mL Capacity Microvolumetric Burets Volumetric Transfer Graduated Measuring mL Vessels Flasks Pipets Cylinders Pipets E237 E287 E288 E969 E1272 E1293 0 5 0 006 1 0 010 0 006 0 01 2 0 015 0 006 0 01 3 0 015 0 01 4 0 020 0 01 5 0 020 0 02 0 01 0 05 0 02 6 0 01 7 0 01 8 0 02 9 0 02 10 0 020 0 02 0 02 0 02 0 10 0 03 15 0 03 20 0 03 25 0 030 0 03 0 03 0 03 0 17 0 05 30 0 03 40 0 05 50 0 05 0 05 0 05 0 25 100 0 10 0 08 0 08 0 50 200 0 10 250 0 12 1 00 500 0 20 2 00 1000 0 30 3 00 2000 0 50 6 00 Other Specifications amp Analytical Comparison of Performance Other Specifications Analytical Comparison of Performance Table 2 Precision data from Table 4 of ASTM E542 Standard If the specifications for Class A glassware and the MICROLAB Practice for Calibration of Laboratory Volumetric Apparatus 500 are compared at 1 mL transfer pipets are slightly better Vessel Size Reproducibili
4. 02 0 03 0 04 0 05 100000 mg mL ML500 Small Class A ML500 Large Class A Volume Glassware Volume Glassware Small Volume Large Volume Discussion amp Validation Methods Discussion Validating Sample Preparation Methods The MICROLAB 500 demonstrated superior performance in terms of cost savings by a factor of 884 and time reduction by a factor of 3 3 From Figure 1 it is apparent that the ML500 methods are not significantly different from the volumetric glassware methods The best fit line for the ML500 small volume method falls below the others because of the relative inaccuracies of the 50 uL and 1 mL syringes used From Table 5 the 50 uL sample side gener ally under dispensed negative error and the 1mL diluent side generally over dispensed Both syringes performed within specification however the result was dilutions that had lower than nominal concentrations The precision of the two general methods is comparable how ever the ML500 slightly out performed the glassware Judging by the best fit y intercepts the ML500 lines gave values that were closer to zero The experiment presented in this poster is just one example of validating an automated sample preparation method The primary validation protocol is to gravimetrically compare the MICROLAB 500 with the volumetric glassware that is or dinarily used in a particular procedure Determin
5. 22 0 9977730 29 0 9959486 23 0 99754 12 30 0 9956511 Hamilton Company Hamilton Bonaduz AG 4970 Energy Way P O Box 26 HAMILTON Sepu E Toll Free 800 648 5950 Bonaduz Switzerland THE MEASURE OF EXCELLENCE Fax 1 702 856 7259 Fax 41 81 641 2563 We 9001 Sb Tt Pde D Telephone 1 702 858 3000 Telephone 41 81 641 0101 Document No L10041 Rev A Hamilton Company 3 97 Printed in U S A
6. 500 with small volume syringes installed The calibration curves resulting from replicate injections of each sample concentration prepared with each method will be gener ated and compared In addition the quantity of methanol and the amount of time required to prepare the samples with each method will be monitored Equipment MICROLAB 530B Diluter Dispenser Hamilton Syringes 50 uL 500 uL 1 0 mL 10 0 mL Class A Pipets Pyrex 1 mL 4 mL 10 ml Class A Buret Pyrex 50 mL Class A Volumetric Flasks 25 mL 100 mL 200 mL 500 mL 1000 mL 2000 mL HPLC System Metering Pump LDC Milton Roy Constametric IInjector Rheodyne with 10 uL Sample Loop Absorbance Detector Kratos Analytical Spectroflow 757 Integrator Hewlett Packard 3396 Series II Column Hamilton PRP 1 5 um 150x4 1 mm Sartorius Balance Model R160P Sensitivity 0 01 mg Sartorius Balance Model MCS Sensitivity 0 001 mg Temperature Gage Solomat MPM with platinum Pt100 probe Weighing Vessels 50 mL plastic beaker with parafilm cover 300 uL microcup with lid Chemicals Methanol J T Baker Baker Analyzed HPLC Solvent Deionized water Milli Q Reagent Water System Acetaminophen Sigma Reference Standard Product number A 3035 Calibration Each pipet was gravimetrically evaluated to assure Class A ac curacy The MICROLAB 500 was evaluated at the experimental volumes both in the dispenser mode and in the diluter mode The results are shown in Table 5 Ex
7. HAMILTON THE MEASURE OF EXCELLENCE Comparing the Performance of Automated Diluters with the Accuracy and Precision Standards for Class A Volumetric Glassware Susan Mayes Hamilton Company 4970 Energy Way Reno Nevada 89502 USA Reprinted from the poster presentation at Pittcon 97 Atlanta March 18 1997 Table of Contents amp Objective Table of Contents ODJECUVE patisse in eere ae E tate desteastesstssieubessaetioes Page 2 DISCUSSION sierpecki aii aa E aE EAEE Page 8 Other Specifications es tus seis eee akin leniieed Page 3 Validating Sample Preparation Methods ee Page 8 Analytical Comparison of Performance cece Page 3 Factors Affecting Instrument Performance Page 9 Calibration and NIST Traceability ee eeeeeeeeees Page 4 CONCIUSIONS wis csscsindvscsstisiesestens aonana Page 9 Experimental sisian norse ee Aae esis EREA Page 4 AEE e Te E E ETTE AE Page 9 RESUS enecens a codes gitsers Page 6 Calibration of Hamilton Diluters Dispensers Page 10 Calculating Return on Investment 0 0 eee eeeeeeeeees Page 6 Objective Automated sample preparation devices such as the Hamilton MICROLAB 500 ML500 reduce preparation time reagent volume requirements and waste disposal costs In addition to these benefits laboratory managers technicians and auditors require that the accuracy of these instruments meets the criteria established for Class A
8. atus The procedure is rather generic allowing the instrument user to set his her own speci fications for accuracy and precision Published specifications for new instruments are found in the User s Manual and original test results are shown on the Performance Test Report s shipped with the instrument The user can specify desired test volumes drive speed and other conditions according to the particular applications and requirements I Summary The general procedure is based on determining the weighing results of water samples delivered by the instrument Volume dispensed is calculated based on the density of water at specific temperatures II Limitations This method is not recommended for volumes below 1 uL and certain procedural modifications are required for volumes of 25uL and less There is no upper volume limit HI Equipment Materials Environment A Laboratory balances required for the test method should meet or exceed the following performance specifications be calibrated regularly with the appropriate traceable weights and be regular ly maintained Test volume uL Balance sensitivity mg 1 10 0 001 10 100 0 01 100 0 1 B Use a balance table or suitable equivalent to minimize vibration Cover its working surface directly in front of the balance with a dark smooth nonglare material Keep the balance area reasonably free of draft currents and the ambient area free of excessive dust C Use a calibrat
9. e the accuracy and precision of both using dispense volumes that match those that would actually be used in preparing samples If the performance of the ML500 meets or exceeds that of the glassware then the ML500 is deemed a suitable equivalent If the ML500 does not at first meet the defined specifications assure that the factors affecting performance next section have been addressed In addition the accuracy of the individual syringes used on the ML500 contribute significantly to the ac curacy of the instrument Different syringes may provide better performance A supplementary method of validating an automated sample preparation method is to actually prepare the samples with both the glassware and the instrument and compare analytical results as was done in the experimental section of this presentation Each application must be evaluated for suitability on a case by case basis Factors Affecting Performance Conclusions amp References Factors Affecting Instrument Performance Conclusions e Choose the appropriate parts Syringes must be chosen based on the sample sizes required For best performance dispensed volumes should be between 10 and 80 of total syringe volumes Tubing gauge must be of the correct size For small volumes use the smaller gauge 18 For relatively highly viscous fluids use the large gauge 12 Assure that the outlet tubing is tapered Hand probes are available for various applications
10. ed thermometer D Use a weighing vessel that has a total volume about 10 to 50 times the test volume If possible also use fits over the outside of the cover to a cover that vessel top don t allow the come into contact with the test liquid The vessel should be plastic glass metal or some other nonporous material The cross sectional area of the opening should be as small as possible for evaporation control E Handle the vessel with forceps or tweezers F Use deionized water 10 IV Procedure A Introduction Deliver a total of n samples into a weigh ing vessel and weigh each sample after delivery Repli cate all motions and time intervals in each sampling cycle as precisely as possible Keep the distance between the balance and the diluter dispenser to a minimum B Preparation Select the analytical equipment and materi als Prepare the instrument to be evaluated by installing the desired syringe s tubing hand probe valve or valve assembly Program the instrument in order set the de sired dispense volumes and syringe drive speeds Ensure that the room equipment and materials including the prepared water are thermally equilibrated Ensure that electronic balances have had sufficient warm up time to stabilize C Place a small amount of water in the weighing vessel between 2 and 30 sample amounts D Place the instrument s inlet tubing into a water reservoir Prime the instrument Perform one asp
11. ges and volumetric glassware used to prepare dilutions Pr MLS500 Small ML500 Large Class A Small Class A Large 0 005 mg mL 5 uL of 50 pL syringe 50 uL of 500 pL syringe 1 mL pipet 10 mL pipet 9950 uL of 1 mL syringe 9 95 mL of 10 mL syringe 200 mL flask 2000 mL flask 0 01 mg mL 10 uL of 50 yL syringe 100 uL of 500 uL syringe 1 mL pipet 10 mL pipet 990 uL of 1 mL syringe 9 90 mL of 10 mL syringe 100 mL flask 1000 mL flask 0 02 mg mL 20 uL of 50 uL syringe 200 uL of 500 uL syringe 4 mL pipet 10 mL pipet 980 uL of 1 mL syringe 9 80 mL of 10 mL syringe 200 mL flask 500 mL flask 0 04 mg mL 40 uL of 50 uL syringe 400 uL of 500 uL syringe 1 mL pipet 4 mL pipet 960 uL of 1 mL syringe 9 60 mL of 10 mL syringe 25 mL flask 100 mL flask 0 05 mg mL 50 uL of 50 uL syringe 500 uL of 500 uL syringe 10 mL pipet 25 mL of a 50 mL buret 950 uL of 1 mL syringe 9 50 mL of 10 mL syringe 200 mL flask 500 mL flask Experimental Results amp Calculating Return on Investment Chromatographic Conditions Six injections of each of the 20 samples were chromatographed in a manner similar to the assay described in the USP monograph Operating conditions Flow rate 2 mL min temperature ambient injection volume 10 uL mobile phase 3 1 deionized water metha nol detection 243 nm Results The results are summarized in Figure 1 and Table 7 Table 7 Comparing cost time and statistical regression results The cost of me
12. irate dispense cycle and discard the effluent When testing the sample side of a diluter use the probe to aspirate and dispense the water Change the drive speeds if undue splashing of the dispense occurs E Open door of balance chamber place weighing vessel on balance pan and close door of balance chamber F Tare the balance Aspirate one sample Retrieve weigh ing vessel from the balance chamber deliver complete sample and return the vessel to the balance pan closing the door to the chamber Observe and record balance readout In some instances it may be possible and more appropriate to dispense into the vessel without removing it from the balance G Repeat step F until 10 samples have been weighed Note Perform the weighing cycles as quickly as possi ble but without compromising the integrity of the liquid delivery or the precision of the technique of the operator H Measure and record the water temperature V Procedure Modifications Calibration continued For volumes of 25 uL and less follow these guidelines 1 Use a very small vessel such as a microwell cup having a total volume of about 300 uL Avoid han dling the vessel by hand as finger oils will provide a source of error Assure a cap for the vessel is used as well 2 Dispense the aliquot onto the inside wall of the ves sel and not directly into the mass of water 3 Determine and use an evaporation coefficient With out dispensing any samp
13. ished for one specified temperature realize that significant error may result if the test is done at a temperature different from that which the density is reported Gravimetric testing of dilutions of two different and inter dependent test liquids is beyond the scope of this proce dure 11 Calibration Calculations VI Calculations C Single dispense in accuracies can be calculated from the volume dispensed V and the expected volume A Ifan evaporation coefficient Corp was V determined correct each mass reading m Accuracy 100 x V V V Noor m T Cevap D Calculate the average dispensed volume from the indi B Calculate the volume of each dispense V by divid vidual dispensed volumes V where i is 1 to n in this ing each corrected mass value by the density of case 10 water at the measured temperature Refer to the table Vae Vit Vitet Vio 10 below for density values E Calculate the instrument accuracy Density of Water at Various Temperatures Accuracy 100 x V re V V Taken from CRC Handbook of Chemistry amp Physics 77th edition 1996 97 page 6 10 F Calculate the standard deviation SDEV of the calcu lated volumes C g cc C g cc SDEV V V no n 1 p 17 0 9987769 24 0 9972994 18 0 9985976 25 0 9970480 G Determine the coefficient of variation precision 19 0 9984073 26 0 9967870 CV 100 x SDEV V 20 0 9982063 27 0 9965 166 21 0 9979948 28 0 9962371
14. le replicate the weighing routine Repeat to obtain 10 values each represent ing the amount evaporated from the vessel during each cycle Add the average of these readings to each sample weighing See the next section Calcula tions To further optimize the procedure in addition to the above small volume guidelines 1 Use degassed water 2 Use the density of water from the CRC Handbook table based on the temperature read to the nearest 0 1 C The table in this procedure only lists the densi ties based on temperatures read to the nearest 1 C 3 Assure that the relative humidity of the testing envi ronment is 45 75 4 Assure that the temperature of the testing environ ment and equipment remains constant to 0 5 C dur ing the course of the test and that no direct sunlight enters the testing area C Here are some guidelines for various sample sizes 1 For validation of a new dispense dilution method use a sample size of 30 instead of 10 2 For quick performance checks such as at a monthly preventative maintenance interval or when tubing or valves are replaced use a sample size of 4 3 When anew syringe is installed onto the instrument for the first time use the proscribed sample size of 10 For a test liquid other than water use that liquid s density in the calculations Most liquids are not as well speci fied at various temperatures as water If the density of the non water liquid is only publ
15. mple tests are run at 10 pL 50 uL and 300 uL dispense volumes using deionized water The ML500 used in this study was calibrated at these volumes and many others Please refer to the Experimental portion of this presentation A calibration procedure describing the details of testing these instruments gravimetrically is found on page 10 of this poster reprint The procedure is based on the method found in ASTM E1154 Standard Specification for Piston or Plunger Operated Volumetric Apparatus The ML500 is calibrated via an unbroken chain of calibrations traceable to the National Institute of Standards and Technology NIST The links in the chain of traceability and the associated uncertainties are illustrated in Table 4 Table 4 NIST traceability of the ML500 Parameter Step Description Uncertainty Temperature 1 NIST calibration 0 00006 K 2 Vendor standard 0 005 K 3 Vendor probe 0 05 K 4 Hamilton probe 0 05 K 5 Fluid temperature Mass 1 NIST calibration 0 00000281 g 2 Vendor standard 0 000005 g 3 Hamilton standard 0 000007 g 4 Hamilton balance 0 000005 g 5 Fluid mass Summary HPLC of acetaminophen will be the vehicle for comparing the MICROLAB 500 with Class A pipets burets and volumetric flasks Five concentrations of acetaminophen will be prepared using four methods Large volume volumetric ware small vol ume volumetric ware the ML500 with large volume syringes installed and the ML
16. perimental Table 5 Calibration results for the ML500 Each calibration at each volume involved 10 samples Accuracy is reported as percent error inaccuracy precision is reported as the coefficient of variation CV in percent For comparison the specifications for volu metric pipets per E969 are listed where applicable in terms of percent error calculated from the published tolerance Syringe Volume Dispensed Volume Dilution Ratio Error Precision Pipet Error uL 50 pL 5 n a 0 45 0 51 10 n a 0 849 0 250 20 n a 0 643 0 170 40 n a 0 003 0 137 50 n a 0 372 0 093 1 0 mL 10 n a 0 078 1 034 50 n a 0 151 0 304 300 n a 0 167 0 143 950 n a 0 069 0 011 1000 n a 0 081 0 018 0 6 50 uL and 1 0 mL 1000 1 199 0 059 0 041 1000 1 99 0 049 0 008 1000 1 49 0 058 0 012 1000 1 24 0 056 0 021 1000 1 19 0 048 0 058 500 pL 50 n a 0 020 0 431 100 n a 0 014 0 287 200 n a 0 174 0 059 400 n a 0 071 0 065 500 n a 0 044 0 047 1 2 10 0 mL 9500 n a 0 247 0 019 10000 n a 0 258 0 006 0 20 500 uL and 10 mL 10000 1 199 0 284 0 013 10000 1 99 0 289 0 013 10000 1 49 0 301 0 009 10000 1 24 0 308 0 011 10000 1 19 0 318 0 009 Sample Preparation First an acetaminophen concentrate of mg mL in 3 1 water methanol was prepared From that five dilutions were prepared also with 3 1 water methanol as the diluent using each sample preparation method Table 6 is a summary of the equipment used Table 6 Summary of syrin
17. thanol was based on 28 per 4 L bottle and the cost of waste disposal was based on 520 per 55 gallon drum Method ML500 Small ML500 Large Class A Small Class A Large Volume Methanol Used 1 25mL 12 5mL 186 mL 1040 mL Cost of Methanol Used 0 01 0 09 1 30 7 28 Volume Waste Generated 5 mL 50 mL 742 mL 4259 mL Cost of Waste Generated 0 01 0 13 1 85 10 40 Sample Preparation Time 25 min 25 min 75 min 75 min Clean Up Time 5 min 5 min 15 min 25 min Total Cost 0 02 0 22 3 15 17 68 Total Time 30 min 30 min 90 min 100 min Best Fit Line Data y intercept 456 1644 3285 2894 slope 11 76x10 12 27x10 12 44x10 12 46x10 R 0 996568 0 999459 0 999115 0 997672 Calculating Return on Investment The price of the ML530B is 3 500 Calculating the differences large and small as obtained in Table 7 and assuming a technician s hourly wage of 10 the return on investment ROD is between 1 5 and 3 4 weeks Class A Large vs ML500 Small Class A Small vs ML500 Large Solvent Costs 17 68 0 02 17 66 3 15 0 22 2 93 Labor Costs 10 x 100 30 60 11 67 10 x 90 30 60 10 00 Cost per Set 17 66 11 67 29 33 2 93 10 00 12 93 Sets 3500 29 33 119 3500 12 93 271 ROI 119 x 0 5 hr 60 hours 271 x 0 5 hr 136 hours Figure 1 Best Fit of Data For Each Sample Preparation Method 700000 600000 500000 400000 300000 Peak Area 200000 100000 S 0 01 0
18. ty Reproducibility Product Tolerance mL mL Transfer Pipet 6uL Transfer 1 0 002 0 2 MICROLAB 500 10 pL Pipets 2 0 002 0 1 Microvolumetric Flask 10 pL 5 0 002 0 04 Measuring Pipet 10 pL 10 0 003 0 03 Buret 20 pL 15 0 005 0 03 Graduated Cylinder 50 pL 25 0 005 0 02 50 0 007 0 014 If the tolerance specifications are compared at the lowest volume 100 0 010 0 01 specified the ML500 is much better than Class A glassware Flasks 10 0 005 0 05 25 0 005 0 02 Product Tolerance 50 0 007 0 014 MICROLAB 500 10 uL 100 0 011 0 011 1 mL syringe 0 3 pL Burets 10 0 003 0 03 Transfer Pipet 0 5 mL 6pL 25 0 005 0 02 Microvolumetric Flask 1 mL 10 pL 50 0 007 0 014 Measuring Pipet 0 1 mL 100 0 012 0 012 1 mL total vol 10 nL Volumetric Flask 5 mL 20 pL Table 3 Accuracy and Precision data for the MICROLAB 500 Buret 50 pL The performance of the ML500 is specified by percent error at 10 mL total volume 20 pL various percents of stroke using a 1 mL syringe Precision is Cylinder 0 1 mL represented as the coefficient of variation 5 mL total volume 50 pL Percent of Stroke Accuracy Precision within 1 5 3 0 1 5 5 30 1 2 0 5 30 100 1 0 0 2 Experimental Calibration amp NIST Traceability Experimental Each MICROLAB 500 is tested before leaving the Hamilton facilities This evaluation involves a gravimetric calibration of each syringe drive at three volumes One milliliter syringes are installed and 10 sa
Download Pdf Manuals
Related Search