Attachment C-GIPSA WI Vita3r01.pdf

Determination of Vitamin A as Retinyl Palmitate in Processed-

Grain Commodities

WORKING INSTRUCTIONS &

METHOD VALIDATION

United States Department of Agriculture Grain Inspection, Packers and Stockyards Administration

Technical Services Division Analytical, Reference and Testing Services Branch

Don Kendall, Chief of ARTS Branch

WI: Vita3r01 (Filename: Vita3r01.doc)

Title: Determination of Vitamin A as Retinyl Palmitate in Processed-grain Commodities

Lab: USDA, GIPSA, FGIS, TSD, ARTS

Program: Commodities Testing Program

Revision: 1

Replaces: 0

Effective: 8/28/2000

Method Flowchart

Extraction

5.0 g sample

7 mL IPA/3 mL H2O then 25 mL hexane or chloroform/acetone

Shake 2 Hrs

Evaporation Allow extract to settle or Centrifuge

Determination

HPLC/UV

Silica column, Isocratic hexane: ethyl ether:ethyl palmitate

UV Detection at 326 nm

Inject 50 µL

Transfer 1 mL to vial and evaporate

Add 1 mL sat. hexane and mix

Summary Procedure for the Validation of Vitamin A Samples

1. Scan each chromatogram to verify that the integration has been applied properly and consistently and that the peaks have been assigned to the right compound.

2. Run the “VitaQC” macro on each calibration curve and check that each calibration parameter is within the specification (CV values for the 13-cis and the all-trans compounds, the 13-cis/all-trans ratio, and the target areas).

3. Run the “VitaResults” macro to generate the summary file. In this file look at each “check calibration”

(CC) sample and confirm that the value is within 5% of the target value of 24,200 IU/Kg.

4. Compare each check sample result with the results found in the “Trace_Lab” database. Any result that is over 3 standard deviations from the average result in the database shows that the method was not in statistical control when the batch was analyzed and the samples must be rerun. If a result is between 2 and 3 standard deviations from the mean, the batch QC results should be carefully inspected. Only one out of twenty results (5%) should be in this range.

5. Create a copy of the results batch file at this point. The name should be “Copy of batch_name.xls.”

Remove lines from the actual results file that are not real sample results, but leave QC samples in the file.

Run the “VitaCert” macro to generate the certificate. Double-check to see that the internal ID numbers have been properly assigned and that the quality control naming conventions have been adhered to. QC samples will have the following designations in the internal ID field and will have a null customer ID.

QC naming specifications:

RB for reagent blank CC for calibration check CS for check sample

Full name for check sample includes a data label designating the check sample and the commodity type for that check sample. An example would be CS012000APF. The date is in month/day/year format with no spaces between.

6. Finally run the “Vita_Access” macro to generate an input file to the Trace_lab database. Then go into the database and run the “samplehistory” and “results” macros to pull the data in. Check the tables to be sure the data was pulled in properly.

If any of the specifications outlined in 1-4 are out of the target range, the lab supervisor must be notified.

Most of these situations will require rerunning that batch of samples.

Method Revision History

Changes in Revision 1

The method was changed so that all processed-grain commodities except all-purpose and bread flours are analyzed using the acetone:chloroform solvent system. Comparison analyses showed that acetone:chloroform often gave higher results than hexane on duplicate samples run in the same batch. In sets of four or more bulgur samples, the results were also less variable with acetone:chloroform than with hexane.

The column was changed from the Zorbax Sil column to the Zorbax RX-Sil column. Due to an apparent change in manufacturing, it is now necessary to use the RX-Sil column. This column contains a silica which is purer than that in the Sil column. The standard Sil column caused irreversible binding of the retinyl palmitate and this was observed as a slowly increasing peak size when replicate injections were made of the retinyl palmitate standard. This occurred even after careful conditioning of the column with isopropyl alcohol and water.

The peak width in the integrator events table was changed from 0.10 to 0.01 minutes to allow for a more accurate integration of the 13-cis retinyl palmitate peak.

All references to “solid commodities” were changed to “processed-grain commodities.”

Table of Contents

1. Purpose and Scope of Application

2. Analyst Qualifications and Responsibilities

3. References

4. Safety and Hazardous Waste

5. Equipment

6. Materials

7. Quality Control Procedures

7.1 Definitions

7.2 Preparation of Standards

7.2.a Preparation of the Concentrated Standard 7.2.b Determination of the Concentrated Standard 7.2.c Preparation of the Calibration Standard

7.3 Preparation of Quality Control Samples

7.3.a Reagent Blank 7.3.b Check Sample

8. Instrument Parameters

8.1 High-Performance Liquid Chromatograph (HPLC)

8.1.a HPLC Conditions 8.1.b HPLC Performance Evaluation/Troubleshooting

9. Sample Analysis

9.1 Solvent Preparation

9.1.a Acetone:choloroform Solvent 9.1.b HPLC Mobile Phase

9.2 Extraction

9.3 Evaporation

9.4 Determination

10. Data Validation

10.1 Peak Shape and Retention Time

10.2 Integration of Peaks

10.3 Signal-to-Noise Ratio

10.4 Interferences

10.5 Calibration Table

10.6 Sample Dilutions

10.7 Check Calibration

10.8 Control Charting

11. Data Reporting

11.1 Vitamin A Results Macro

11.2 Reporting Procedure

12. Isomer Calculation Equations

13. Excel Macros

13.1 Vitamin A Results

13.2 Vitamin A QC

13.3 Vitamin A Certificate

13.4 Vitamin A Access

14. Method Validation

14.1 Overview

14.2 Results and Discussion

14.3 Summary and Tables

1. Purpose and Scope of Application

The purposes of these Working Instructions (WI) are to establish the operational parameters, methodology, and requirements for the quality assurance and acceptability of data in the determination of vitamin A as retinyl palmitate in processed-grain commodities. These commodities include all-purpose and bread flour (APF and BF), bulgur (BUL), soy-fortified bulgur (SFB), soy-fortified sorghum grits (SFSG), corn meal (CM), soy-fortified corn meal (SFCM), corn-soy blend (CSB), and wheat-soy blend (WSB). A mixture of 30% acetone and 70% chloroform is used as the extraction solvent for all commodities except APF and BF.

For these samples hexane is used as the extraction solvent.

2. Analyst Qualifications and Responsibilities

The analyst(s) will receive proper training in the conduct of these WI and will follow the WI as written.

The Supervisory Chemist (or Project Leader) is responsible for ensuring that the WI are followed and modified as necessary or appropriate. All revisions to these WI must be approved by the Chief of the Analytical, Reference and Testing Services Branch prior to implementation.

3. References

• Thompson, J. N.; Hatina, G.; Maxwell, W. B. “Α High Performance Liquid Chromatographic Determination of Vitamin A in Margarine, Milk,” Partially Skimmed Milk, and Skimmed Milk, J.

AOAC Int. 1980, 63, 894-898. (HPLC method is based on this reference).

• Woollard, D. C.; Indyk, H. “The HPLC Analysis of Vitamin A Isomers in Dairy Products and their significance in Biopotency Estimations,” J. Micronutrient Analysis 1986, 2, 125-146. (Basis for the inclusion of the 13-cis isomer in the Vitamin A concentration).

• Ross, A. C. “Separation of Long-Chain Fatty Acid Esters of Retinol by High-Performance Liquid

Chromatography,” Anal. Biochem. 1981, 115, 324. (Extinction coefficients for retinol and retinyl esters are identical).

• Hubbard, R. “Geometric Isomerization of Vitamin A, Retinene and Retinene Oxime,” J. Am. Chem.

Soc. 1956, 78, 4662-4667. (Reference for the extinction coefficients for all-trans and 13-cis retinyl palmitate in hexane).

• Baldingh, J.; Cama, H. R.; Collins, F. D.; Morton, R. A.; Gridgeman, N. T.; Isler, O.; Kofler, M.;

Taylor, R. J.; Welland, A. S.; Bradbury, T. “Pure All-trans Vitamin A Acetate and the Assessment of Vitamin A Potency by Spectrophotometry,” Nature (London) 1951, 168, 598. (Reference for the extinction coefficients for all-trans and 13-cis retinyl palmitate in hexane).

• Qian, H.; Sheng, M. “Simultaneous determination of fat-soluble vitamins A, D and E and pro-vitamin

D2 in animal feeds by one-step extraction and high-performance liquid chromatography analysis,” J.

Chrom. 1998, 825, 127-133. (Basis for the use of acetone:chloroform as an extraction solvent).

• Analytical Methods Committee “Determination of Vitamin A in Animal Feedingstuffs by High-performance Liquid Chromatography,” Analyst (London) 1985, 110, 1019-1026. (Traditional saponification method used in the method validation studies).

4. Safety and Hazardous Waste

Hexane, chloroform, and ethyl ether are extremely flammable and are health hazards. They can be absorbed through the skin or the lungs. Ethyl ether is a peroxide former and should not be stored for more than one year. Please date each can of ether when it is received. Use latex/nitrile gloves, eye protection, and perform operations with these solvents in the hood.

For additional information refer to the Technical Services Division Chemical Hygiene Plan.

5. Equipment

This section is meant to specify the equipment being used in the laboratory during the method validation process. Other equipment can be substituted for these WI as long as the supervising chemist/technician approves. In some cases, side-by-side testing will be necessary to verify equivalency.

a) High-performance liquid chromatograph, Model 1100 with auto sampler, quaternary pumping system, vacuum degasser, variable-wavelength UV detector, and Chemstation software version 7. (Agilent Technologies, Wilmington, DE).

b) Analytical balance, capable of weighing to 0.0001 g.

c) Ultraviolet spectrophotometer, HP 8452 (Agilent Technologies, Wilmington, DE).

d) Centrifuge, Model GS-6 (Beckman Instruments, Palo Alto, CA).

e) Orbital shaker, Model 3520 (Lab-line Instruments Inc., Melrose Park, IL).

f) Nitrogen evaporator, Turbovap LV (Zymark Corporation, Hopkinton, MA).

6. Materials

This section is meant to specify the materials being used in the laboratory during the method validation process. Other materials can be substituted for these WI as long as the supervising chemist/technician approves. In some cases, side-by-side testing will be necessary to verify equivalency.

a) Hexane, Burdick and Jackson, UV grade (#BJ216-4, VWR Scientific, Chicago, IL).

b) Ethyl Ether, anhydrous, BHT stabilized, ACS Grade (#E138-4, Fisher Scientific, Pittsburgh, PA).

c) Ethyl Palmitate, 99% (#28,691-5, Aldrich Chemical, Milwaukee, WI).

d) Water, high purity, through Milli-Q RO Plus/Milli-Q UV water purification system (Millipore Corp., Marlborough, MA).

e) Acetone, Burdick and Jackson, HPLC grade (#BJ015-4, VWR Scientific, Chicago, IL).

f) Chloroform, HPLC grade (#C607SK-4, Fisher Scientific, Pittsburgh, PA).

g) 2-Propanol, HPLC grade (#A451-4, Fisher Scientific, Pittsburgh, PA).

h) Retinyl Palmitate, all trans, type IV, synthetic (#R3375, Sigma, St. Louis, MO).

i) Cottonseed Oil (#17,991-4, Aldrich Chemical, Milwaukee, WI).

j) Syringe Filters, 25 mm, PTFE, 0.45 µ, Whatman (09-927-30D, Fisher Scientific, Pittsburgh, PA).

k) Hamilton Gastight Syringes, 5 mL; #1005, 1.0 mL; #1001, 500 µL; #1750, 250 µL; #1725, 100 µL;

#1710, 50 µL; #1705, 25 µL; #1702, 10 µL; #701 (Hamilton Company, Reno, NV).

l) Centrifuge Tubes, Falcon, 50 mL, polypropylene (#14-959-49A, Fisher Scientific, Pittsburgh, PA).

m) Autosampler vials and 11 mm snap caps, Agilent #5182-0545 and #5182-0550, respectively (Agilent Technologies, Wilmington, DE).

n) Adjustable 1-5 mL FinnPipet with disposable tips (#21-377-196 and #21-377-51, Fisher Scientific, Pittsburgh, PA).

o) Pipettor, 100 to 1000 µL, Eppendorf (#53511-582, #53508-819 (tips), VWR Scientific, Chicago, IL)

p) Analytical column, Zorbax RX-SIL (normal phase silica), 250 x 4.6 mm, 5µ (#880975-901, Agilent

Technologies, Wilmington, DE).

q) Guard Column, Zorbax SIL, 4.6 x 12.5 mm, 5µ (#820950-901, Agilent Technologies, Wilmington, DE).

r) Guard Column Holder (#820777-901, Agilent Technologies, Wilmington, DE).

s) Precolumn filter with 0.5 µ frit (#A-101x, Upchurch Scientific, Oak Harbor, WA).

t) Vials, 20 mL, pre-cleaned (Quorpak Company, Bridgeville, PA).

u) Bottle-top Dispensers, Dispensette III, 1-10 mL (#4701 341, BrandTech Scientific, Inc).

v) Bottle-top Dispensers, Dispensette III, 2.5-25 mL (#4731 351, BrandTech Scientific, Inc).

w) Culture Tubes, 16 x 100 mm (#60825-618, VWR Scientific, Chicago, IL).

x) Volumetric Flasks, class A, 10 mL (VWR Scientific, Chicago, IL).

7. Quality Control Procedures

7.1 Definitions

Reagent Blank (RB). Sample where the method is run with only reagents. Its purpose is to check for impurities that may interfere with the analysis.

Check sample (CS). A portion of each commercially-fortified commodity containing approximately the target level (24,200 IU/Kg) of retinyl palmitate. The check sample is analyzed with each batch of samples as a check on the method performance. It indicates the performance of the method on a “real” sample.

Neat Standard. Pure all-trans retinyl palmitate standard purchased from a commercial vendor.

Concentrated Standard. Standard solution prepared from the neat standard and UV-grade hexane (ca 1000 µg/mL).

Calibration Standard (STD1, STD2, or STD3). A solution containing a known quantity of retinyl palmitate prepared from the concentrated standard and UV-grade hexane.

Calibration Check (CC). A calibration standard that is run during sample analysis to check for any drift in the instrumental response.

Internal ID (VIAYCONT). Sample ID that is assigned when the sample enters the Trace Analysis Laboratory. “VIA” stands for vitamin A analysis, “Y” is a single digit refering to the fiscal year, and “CONT” is a counter that starts with 0001 and continues for each sample assigned until the end of that fiscal year.

Batch (VIAYRCNT). A batch of samples consists of one reagent blank, one check sample for each commodity being analyzed, and up to ninety samples. A batch of samples is processed, as much as possible, as a single unit. This means that all the samples are processed on the same day, in the same lab environment, and in exactly the same way. When the batch is analyzed on the instrument it includes a calibration standard analyzed at 3 levels, one reagent blank, at least one check sample, one to ninety samples, and a check calibration every 10 injections. In naming the batch “VIA” refers to vitamin A analysis, “YR” is the fiscal year the batch was analyzed in , and “CNT” is a batch counter starting at 1 that begins day one of that fiscal year.

7.2 Preparation of Standards

7.2.a Preparation of the Concentrated Standard

1. Bring the neat all-trans retinyl palmitate standard to 30 °C.

2. Weigh 10-12 mg of retinyl palmitate into a 20 mL, amber, precleaned vial using a 500 µL Hamilton syringe (about 1 drop) or a disposable Pasteur pipet.

3. Add 10 mL of hexane using the adjustable pipet and disposable 5 mL tip. Cap the vial.

4. Mix by sonicating for 15 minutes.

5. Label the vial with the chemical name, concentration (1 mg/mL), date, solvent, lot number, and your initials.

7.2.b Determination of the Concentrated Standard

This determination must be performed each time a batch of samples is analyzed, due to the rapid evaporation of hexane from the standard.

1. Make sure that the standard has come fully to room temperature if it has been refrigerated.

2. Add 50.0 µL of the 1 mg/mL concentrated standard to a 10 mL volumetric flask and dilute to the mark with hexane.

3. Mix the solution by vortexing or inverting. The concentration will be about

5 µg/mL.

4. Turn on the HP 8452 diode-array UV spectrophotometer and the computer. The lamp must be on for at least one hour before performing this test. Set the system for “General Scanning” from 200 nm to 400 nm.

5. Fill a clean quartz UV cell with hexane, put it into the instrument, and run “Scan

Blank.” Remove the cell, discard the hexane, and dry the cell with a stream of nitrogen. Transfer about 3 mL of the ca 5 µg/mL solution prepared above into the UV cell. Put the lid on the UV cell. Place the cell back into the spectrophotometer as soon as possible and activate “Scan.”

6. Place the cursor at the highest point on the curve near 325 nm and record the displayed absorbance value. The maximum must occur at 326 nm +/- 2 nm. If the maximum is not in the range of 324-328 nm, inform the supervising chemist/technician.

7.2.c Preparation of the Calibration Standard

1. Make sure that the concentrated standard has come fully to room temperature if it has been refrigerated.

2. Calculate the volume of concentrated standard to add to generate a standard containing 121.0 IU (24,200 IU/Kg sample) of total retinyl palmitate:

Volume (µL) = 33.25 / ABS Where ABS is the absorbance for the

5 µg/mL solution measured in step 7.2.b.

IMPORTANT: This equation is correct only for cis/trans peak area ratios of

0.035 to 0.055. If analysis of the standard is out of this range, a new standard should be prepared or refer to section 12, equation (11) for the calculation that includes the actual area ratio.

3. Label a centrifuge tube “STD” and add 7 mL of 2-propanol, 3 mL of Millipore water and 25 mL of a 30% acetone:70% chloroform mixture using the bottle-top dispensers. To prepare a standard for APF and BF use 25 mL of hexane instead of the acetone:chloroform solution. Then add 100 µL of cottonseed oil to the tube with an Eppendorf pipet.

4. Add the calculated volume of the concentrated standard to the centrifuge tube using a 50 µL or 100 µL (closest match) Hamilton syringe. Do not push the plunger abruptly to 0, but push it slowly and steadily to deliver the standard. Do not immerse the needle into the hexane. When the entire standard is expelled from the syringe, touch the needle to the inside of the centrifuge tube to remove the last drop. Put the cap onto the tube snugly and shake.

5. Let the solution settle for a few minutes.

6. For hexane standards, transfer portions of the top layer to several 2 mL autosampler vials and label the vials, STD. Prepare enough vials to run a check calibration every 10 injections and at the end of the sequence. The concentration is 24,200 IU/Kg total retinyl palmitate. For acetone:chloroform standards, the portions should be taken from the lower layer. Be sure to push air out while inserting the pipet through the top aqueous layer to avoid drawing in any of this layer.

7.3 Preparation of Quality Control Samples

7.3.a Reagent Blank

1. Label a centrifuge tube RB.

2. Add 7 mL of 2-propanol, 3 mL of Millipore water and either 25 mL of 30% acetone:70% chloroform solution or 25 mL of hexane to a 50 mL centrifuge tube using the bottle-top dispensers.

7.3.b Check Sample

1. Label a collection vial “CSDATECOM” where DATE is the date label on the check sample in “MODYYR” format and “COM” refers to the abbreviation for the commodity (ie, APF, BF, BUL, SFB, SFSG, CM, SFCM, WSB, CSB).

2. Weigh 4.95 to 5.04 g of the ground check sample into the centrifuge tube.

3. Add 7 mL of 2-propanol, replace the cap, and tap firmly on the edge of the benchtop to fully wet the commodity. Add 3 mL of water to the tube and again tap on the benchtop to mix. Wait at least 10 minutes and then add 25 mL of hexane or acetone:chloroform solution.

8. Instrument Parameters

8.1 High-Performance Liquid Chromatograph (HPLC)

8.1.a HPLC Conditions

Pump (1100)

Flow

2.0 mL/min

Mobile Phase (isocratic)

96% water sat. hexane:

4% ethyl ether:

0.1% ethyl palmitate

Stop Time

10 min

Column

Type

Zorbax RX-SIL

Particle Size

5 micron

Dimensions

4.6 mm x 250 mm

Temperature

40 °C

Injection

Volume

50.0 µL

Draw Speed

200 µL/min

UV Detector Stop Time as Pump: 10 min

Wavelength

326, 4 nm

Reference Wavelength

450, 80 nm

Slit Width

4 nm

Flow Cell Path Length

10 mm

Peak width

0.05 min

Sampling Interval

0.320 sec

Calibration

Replace

All Levels

Integration Events:

Area Reject

0.50

(“Event_VWD1A”)

Slope Sensitivity

1.75

Peak Width

0.01

Height Reject

0.50

Shoulders

OFF

Baseline at Valleys On

2.20

Integration On

2.20

Integration Off

5.0

Report Parameters

Destination

Printer, File

File Types

.DIF, CSV, TXT, XLS

Report Style

Short

Quant. Results Sorted By

Signal

Chrom. Orientation

Portrait

Chrom. Time Size

0.00 to 10.00 min

Chrom. Response Range

Full

Calibration Table Parameters

Type

External Standard

Basis

Peak Areas

Rel. Reference Window

0.5 min

Curve Type

Linear

Origin included

Calibration Table

Cis/trans=0.045 (0.035-0.055)

13-cis retinyl palmitate

Level 1

433.5

Level 2

Level 3

1,734

All-trans retinyl palmitate

Level 1

11,666.5

Level 2

23,333

Level 3

46,666

8.1.b HPLC Performance Evaluation/Troubleshooting

1. Check the instrument pressure reading on the Chemstation after the system has been on for one hour with the correct mobile phase. The pressure should not be more than 10 bar higher than the normal reading for the HPLC using the same mobile phase and flow rate. The normal pressure reading will be different for each HPLC/column combination. To determine the correct reading, check the logbook for the last pressure reading entry for a vitamin A batch. The reading will probably be between 70-80 bar.

2. If the reading is too high, install a new precolumn filter. If this does not bring down the pressure, install a new guard column. If the pressure is still high, report the problem to the supervising chemist/technician.

3. If the reading is too low, look for leaks. The most common place for a leak is at the precolumn filter, guard cartridge holder, or at the column. Tighten fittings carefully, only applying just enough pressure to stop the leak. “Cranking” on fittings will ruin the fitting and cause a larger leak.

4. Adjust the UV signal attenuation on-screen until the “noise” fills up 10-20% of the chromatogram. Pump for at least 60 minutes to fully equilibrate the system.

Check to see that the baseline (imaginary line through the center of the noise) has been flat for at least 5 minutes.

5. Inject the standard (using the Vita_com method) and wait 10 minutes until the report emerges at the printer.

6. The retention time for retinyl palmitate must be between 2.5 and 3.5 minutes. If it is not within this range, carefully prepare a new batch of mobile phase according to these WI and equilibrate the HPLC for at least 30 minutes before reinjecting the standard. If the peak is still not within this range, report this to the supervising chemist/technician.

7. The retinyl palmitate peak must be gaussian-shaped with no “leading edge” or

“trailing edge.” The main thing to look for is a symmetrical peak. That is, if you draw an imaginary line down through the center of the peak, the resulting halves are mirror images. If you see a fronting peak (leading edge) or a tailing peak (trailing edge) report this to the supervising chemist/technician. If you must troubleshoot this problem, first try mixing up a new batch of mobile phase.

If that doesn’t work, try a new guard and/or analytical column.

8. Measure the noise on the lowest-level standard report chromatogram, just after the retinyl palmitate peak. Measure the peak height from the center of the noise (imaginary line) to the top of the peak. Calculate the signal-to-noise by dividing the peak-height measurement by the noise-height measurement. This result must be at least 10.0 to proceed. If the signal-to-noise ratio is <10.0, report this to the supervising chemist/technician.

9. The last performance parameter to check is the target area. The middle-level standard should give an area of 299 +/-5% for the hexane standard and 273 +/- 5% for the acetone:chloroform standard. If the areas are not within the specified range, report this to the supervising chemist/technician. (NOTE: These numbers are instrument dependent. Determine these numbers before running samples).

10. If all of these performance parameters are within range, the sequence for the determination of this batch of samples can be started. Refer to section 9.4 to start the HPLC determination. Also see section 10 for additional performance checks to be made after the sequence has begun.

9. Sample Analysis

9.1 Solvent Preparation

9.1.a Acetone:choloroform Solvent: 30% acetone:70% chloroform

1. Rinse a clean 4L jug and a 2000 mL graduated cylinder with HPLC grade acetone.

2. Measure 1200 mL of HPLC grade acetone using the 2000 mL graduated cylinder and add it to the 4L jug.

3. Measure 1400 mL of chloroform using the 2000 mL graduated cylinder and add it to the 4L jug. Repeat this process for a total of 2800 mL chloroform.

4. Invert the jug several times to mix (careful to release pressure, if heating occurs) and allow the solvent to come to room temperature. It will help to place the jug in a sink of room temperature water to speed up the process.

9.1.b HPLC Mobile Phase: 96% wet hexane/4% ethyl ether/0.1% ethyl palmitate

1. To prepare the “wet hexane” solution, add at least 200 mL of Millipore water to a 4L jug of hexane and shake well. Let this jug stand overnight.

2. Rinse a 1L HPLC solvent reservoir and a 1000 mL graduated cylinder with methanol and then with hexane.

3. Add 960 mL of wet hexane to the 1L reservoir using a 1000 mL graduated cylinder.

4. Add 40 mL of ethyl ether to the 1L reservoir using a 50 mL graduated cylinder or a syringe.

5. Add 1 mL of ethyl palmitate to the reservoir using the FinnPipet. It may be necessary to heat the ethyl palmitate in the GC oven at 30 °C to liquify.

6. Swirl the reservoir for 15 seconds to mix. After installing the reservoir on the HPLC, open the prime/purge valve and pump at 10.0 mL/min for a few minutes to purge the air bubbles from the system. Reset the flow and close the valve.

9.2 Extraction

1. Weigh 4.95 to 5.04 g of the commodity sample into a 50 mL centrifuge tube.

2. Continue to weigh out all samples including the check samples (one for each commodity being analyzed).

3. Add 7 mL of 2-propanol to the first sample tube, replace the cap and tap firmly on the edge of the benchtop to fully wet the commodity. Repeat this procedure for all the samples.

4. Add 3 mL of water to each tube, replace each cap and again tap on the benchtop to mix.

Wait at least 10 minutes and then add 25 mL of acetone:chloroform solution to each tube or hexane if running APF or BF samples.

5. Shake the tubes for 2 hours at 250 rpm on the orbital shaker.

9.3 Evaporation

1. Place the tube racks upright and let sample extracts settle out. For acetone:chloroform extracts, centrifuge at 2500 rpm for 5 minutes. Label one set of autosampler vials with the appropriate sample IDs.

2. For acetone:chloroform extracts, filter about 3 mL of the lower layer through the syringe filter (0.45 micron, PTFE) into a culture tube. Be aware that there may be only a single layer. In this case just take the aliquot from the middle of the tube. Filtering can be performed with the FinnPipet by cutting off about 5 mm from the end of a pipet tip so that it will fit snugly into the syringe filter. Measure 1 mL of the filtered extract into an autosampler vial. For hexane extracts, transfer 1 mL of the top hexane layer from each sample tube to an autosampler vial using the FinnPipet.

3. Place the vials in the custom vial rack and then into the Turbovap LV nitrogen evaporator. Evaporate the samples at 45 °C with a very low flow of nitrogen. The nitrogen flow should be adjusted by starting with the pressure at 0 and increasing carefully until the liquid surface in each vial is moving, but without any splashing.

Evaporate just to dryness (about 15 minutes for hexane and about 30 minutes for acetone:chloroform). Remove the vials promptly.

4. Add 1 mL of water-saturated hexane to the first autosampler vial and cap immediately.

Repeat until all vials are capped. Shake each vial briefly to mix.

9.4 Determination

1. Load the method, “Vita_com” and turn on the pump and lamp if they are not already on.

2. The instrument should equilibrate at 2.0 mL/min and 40 °C for at least 60 minutes before sample analysis begins. The HPLC pump should be started before the extraction and evaporation steps to allow enough time for equilibration and analysis of the calibration standards.

3. Enter the analysis sequence so that the calibration standards are analyzed first, from low to high, the QC samples, and finally the samples to be determined. The calibration levels consist of 12,100, 24,200 and 48,400 IU/Kg of total retinyl palmitate in the solid sample.

The levels are obtained using a single calibration standard with injection sizes of 25 µL, 50 µL, and 100 µL. A check calibration using a separate calibration vial must be run after every 10 samples to check for instrument drift. The continuing calibration is run as “sample,” not as “calibration” in the sequence table. Save the sequence and print out a copy. Start the sequence by choosing “Run Sequence.”

4. Load the standard, QC samples, and samples into the autoinjector and double-check the sequence printout to be sure the samples are in the correct order. Start the sequence.

5. Check the first report to be sure the retinyl palmitate peak is found and integrated. If the retention times are slightly out of range, abort the sequence, enter the new retention time of the retinyl palmitate peaks into the calibration table, save the method, and restart the sequence. If the peaks are not integrated properly, check the integrator parameters in the method against those listed in section 8.1.a. After adjusting the parameters, the method can be saved and the sequence restarted.

6. When the three calibration levels have been analyzed, run the Excel macro, VitQC, to determine if the CV, %CIS, and Target area values are within specification. If the curve is not within the specifications listed on the report, prepare a new calibration curve and restart the sequence.

10. Data Validation

10.1 Peak Shape and Retention Time

Refer to section 8.1.b for the acceptable retention time range and peak shape. If the peak shape has deteriorated or the retention time has drifted out of range, the whole sequence must be reinjected. If more than 24 hours have passed since the initial sequence was started, the samples must be reanalyzed from the beginning.

10.2 Integration of Peaks

Correct integration of the retinyl palmitate peaks can be determined by observing the “integration line” drawn at the base of the peak in the “chromatogram” section of the report. Ideally, the line should begin at the point where the response just increases above the baseline and stop at the point where the response just meets the baseline. Unfortunately, it is not always easy to determine where the baseline really is. For these WI we will define two cases.

In the first case, the baseline has “shifted” while the peak eluted and either the front edge of the peak is higher or lower than the back edge. When this occurs, the peak integration line should stay along the baseline and another vertical line should be drawn up to the start or end of the peak.

This shifting may have occurred because of a “drifting” of the instrument response, or because of an interfering peak. Under no circumstances should the peak be integrated using this method if the baseline has shifted more than 25% of the total peak height. If this is the case, the sample must be reinjected or reanalyzed to eliminate the interference.

In the second case, a broad interference peak exists under the peak and both the front and the back edges of the peak are elevated off of the baseline. Be careful that you know where the baseline really is. You can determine this by looking at the whole chromatogram and drawing a line from the response at 0 time to that at 10 minutes. This is usually the true baseline. If it is determined that the retinyl palmitate peaks occur on a broad interference, then the peak must be “skimmed” off of the interference peak. That is, the integration line should be drawn from the point where the response just increases above the “interference baseline” and stop at the point where the response just meets the “interference baseline.” The “interference baseline” is an imaginary line drawn that outlines the expected shape of the broad interference peak.

10.3 Signal-to-Noise Ratio

Refer to section 8.1.b for the acceptable signal-to-noise ratio for the 12,100 IU/Kg standard. If the signal-to-noise ratio is <10.0, the quantitative result is not reliable and must not be reported. This signal-to-noise ratio does not include separated background interferences. It refers only to the random noise created by the instrumentation. If this noise cannot be determined due to matrix interferences, and standard peaks have acceptable signal-to-noise ratios, it can be concluded that the results for the sample are within the signal-to-noise ratio specification.

10.4 Interferences

When another peak is merged with the retinyl palmitate peak such that either the start or end of the peak is elevated more that 25% of the total peak height and the interfering peak’s width at half height is within 3 times that of the retinyl palmitate peak, the sample must be reinjected. After reinjection, if the interference peak still appears, the sample must be reanalyzed. When the interfering peak is >3 times the width of the analyte peak, it is classified as a broad interference and can be integrated as described in section 10.2.

10.5 Calibration Table

When the coefficients of variation of the response factors in the calibration table are out of range (>3% for the 13-cis and >1% for the all-trans) and the sequence was analyzed anyway, results that are “not detected” can still be reported. Other sample results will have increased variability and should not be reported. In this case new standards must be prepared and all samples reinjected.

10.6 Sample Dilutions

Samples with results that are >75,000 IU/Kg must be diluted and reinjected so that they are within the calibrated range. A standard dilution factor of 5 should be used. For example, if a sample result was 80,000 IU/Kg, the sample must be diluted by adding 100 µL of the original sample and 400 µL of water-saturated hexane to a new autosampler vial. The sample result must be multiplied by 5 to get the final result.

10.7 Check Calibration

Check calibration injections must fall within the range of 0.95 to 1.05 times that of the 24,200 IU/Kg responses. If they do not, the ten samples after this check calibration standard must be reinjected along with a new calibration curve. Reprocessing is not an option here since the out-of-range result could have been caused by an injection problem.

10.8 Control Charting

One check sample is analyzed for each commodity in each batch of samples. The result should be plotted on a control chart with previous results for that commodity. For most commodities, a difference of >15% from the mean should be a warning sign. If the result is greater than 2 standard deviations from the mean, the supervising chemist/technician should be notified. If the result does not lie within 3 standard deviations of the mean the whole batch of samples for that commodity must be reanalyzed.

11. Data Reporting

11.1 Vitamin A Results Macro

Running this macro will produce a summary of all sample results for this batch. It is important to check each injection for proper integration before running the macro. The macro adds results for the 13-cis and all-trans isomer to give a total retinyl palmitate result. It also rounds the numbers so that 3 significant figures are reported in IU/Kg. The Hamilton syringes used to make the calibration standard are good only to 3 significant figures. For this reason, the final result must be rounded to 3 significant figures. A result of 24,350 IU/Kg will be rounded and reported as 24,400 IU/Kg.

A control chart should be generated showing the historical results of each check sample analysis.

The chart should show the mean, two, and three standard deviations. The charts allow for a quick check that the analyses were in statistical control (within 3 standard deviations of the mean).

11.2 Reporting Procedure

The Vitamin A Certificate macro should be run to generate the lab certificate. The certificate will list the customer ID (or LAB ID), the internal ID, the sample type, and the result in both IU/Kg and IU/lb. The supervising chemist/technician must sign the certificate before the results can be released.

The Vitamin A Access macro should be run to generate an input file into the Trace_Lab database.

12. Isomer Calculation Equations

It has not proven possible to obtain a 100% pure all-trans retinyl palmitate standard. The Sigma standard used in the GIPSA lab usually contains about 4.5% of the 13-cis isomer. For this reason the 13-cis compound must be taken into account when the standard is prepared. In addition, the 13-cis isomer also has biological activity at about 75% the level of the all-trans isomer. The 13-cis isomer should be quantified in samples containing this isomer and the amount added to the all-trans result. The following derivation was used to generate the equation used in section 7 and to generate the equations used to calculate the concentrations of the 13-cis and all-trans isomers in the HPLC calibration table given in section 8.

The extinction coefficients for all-trans and 13-cis retinyl palmitate were taken from the literature (see section 3). The λmax values for the all-trans and 13-cis isomers in hexane are 326 and 328 nm, respectively.

Since our measurements are being taken 326 nm instead of the λmax of the 13-cis isomer, the extinction coefficient for this isomer has been adjusted downward by 2% from the literature value.

εcis = 47,300 cm-1 l mol-1 εtrans = 52,100 cm-1 l mol-1 Molecular weight = 524.9 g/mol (all-trans or 13-cis retinyl palmitate) Molecular weight = 286.5 g/mol (retinol) 1 IU = 0.3 µg of retinol

In this method, the UV spectrophotometer is used to standardize the concentration of the all-trans retinyl palmitate. Since it is actually a mix of the two isomers, the following equation follows:

(1) transcisnm AAA +=326 where A326nm is the total UV absorbance at 326 nm Acis is the UV absorbance contributed by 13-cis retinyl palmitate Atrans is the UV absorbance contributed by all-trans retinyl palmitate

From Beer’s Law we know that

(2) cisciscis bcA ε=

(3) transtranstrans bcA ε= where ε is the extinction coefficient b is the path length c is the concentration

Analysis of the standard by HPLC allows the determination of the area ratios of the two isomers by UV detection at 326 nm and allows us to write equation (4).

(4) σ ε ε transtrans ciscis trans cis cb cb

AreaPeak AreaPeak where Peak_Area is the area of the HPLC peaks for the two isomers detected at 326 nm.

Rearranging equation (4) gives (4a):

(4a) cis transtrans cis cc ε σε

Combining equations (1)-(3) and plugging (4a) back in gives equations (5) and (6).

(5) )11(

)900,524(326 σ ε + cis cis Ac

And

(6) )1(

)900,524(326 σε + trans trans

Ac where c in both cases is in units of µg/mL

These equations allow the calculation of the concentration of either isomer from the total absorbance of the solution and the isomer ratio σ of that same solution determined by HPLC.

The target concentration of total retinyl palmitate for processed-grain commodities is 24,200 IU/Kg. This leads to equation (7):

(7) 200,24=+ transcis cc

Adding a factor of 0.75 to equation (4a) to account for the lower biological activity of 13-cis retinyl palmitate and then plugging (4a) into equation (7) gives equation (8):

(8) σεε σε transcis trans cisc

)75.0( )200,24)(75.0(

Equation (8) allows the calculation of the concentration of the cis isomer from the total retinyl palmitate in IU/Kg. For a σ of 0.045 and a total retinyl palmitate concentration of 24,200, the concentration of the 13-cis isomer is 867 IU/Kg. This makes the all-trans isomer 23,333 IU/Kg.

Dividing the all-trans target concentration by 1000 and multiplying by 5 gives 116.7 IU, which is the amount in a single sample tube required to give a concentration of 23,333 IU/Kg in the sample. This amount in International Units of Vitamin A can be converted to µg of all-trans retinyl palmitate using the ratio of the molecular weights of retinol and retinyl palmitate along with the definition that says 1 IU = 0.3 µg of retinol.

(9) ( ) atrpatrp ug

IU

ug mg mgIU 12.64

3.0

/5.286 /9.52467.116 = where atrp = all-trans retinyl palmitate

Equation (9) shows that a 5g sample with a concentration of 24,200 IU/Kg contains 64.12 µg of all-trans retinyl palmitate.

The equation for the volume of standard solution required in µL to prepare a 24,200 IU/Kg standard (or to add 64.12 µg of all-trans retinyl palmitate) is given in equation (10) where ctrans is in µg/µL.

(10) gcVol transL µµ 12.64))(( =

We can substitute equation (6) in for ctrans in equation (10) along with a factor of 200 to account for the dilution of the concentrated standard (1000 µg/mL) to that used to measure the UV (5 µg/mL) to give equation (11).

(11)

)1(82.31 A

Vol L σ µ

For σ = 0.045 equation (12) results.

(12)

25.33 A

Vol L =µ

13. Excel Macros

13.1 Vitamin A Results

This macro produces a spreadsheet summarizing the results for the batch result information. No template is stored in Sheet1 or in Sheet2.

' Vitamin A Res Macro ' Macro written 3/16/99 by Tim Norden Sub VitaminA_Res() 'Prompt the user for the subdirectory and datafile name.

'The MainDir and TemplateName statements can be changed to reflect 'where the datafiles are located.

Instrument = InputBox("Please enter the instrument number:", "Instrument") If Instrument = "1" Then MainDir = "Q:\1\DATA\" Else: MainDir = "Q:\2\DATA\" End If BatchName = InputBox("Enter the Batch Name:", "Batchname") StoreFile = MainDir & BatchName & "\" & BatchName Row = 4 'Test each datafile to determine if the directory holds the valid calibration data.

'This is done by looking in Report00.dif see if it is a "Level 4" calibration sample.

NextFile = Dir(MainDir & BatchName & "\*.d", vbDirectory) Workbooks.Add Cells(1, 1) = BatchName Range("A1").Select With Selection.Font .Name = "Arial" .FontStyle = "Regular" .Size = 12 .Strikethrough = False .Superscript = False .Subscript = False .OutlineFont = False .Shadow = False .Underline = xlNone .ColorIndex = xlAutomatic End With Selection.Font.Bold = True ActiveWorkbook.SaveAs Filename:="New.xls" 10 Do While NextFile <> "" FirstFile = MainDir & BatchName & "\" & NextFile & "\" & "Report00.dif" Workbooks.Open Filename:=FirstFile Test = Cells(1, "A") ActiveWorkbook.Close If Test = "Report Title" Then InfoFile = MainDir & BatchName & "\" & NextFile & "\" & "Report02.dif" DataFile = MainDir & BatchName & "\" & NextFile & "\" & "Report03.dif" Else InfoFile = MainDir & BatchName & "\" & NextFile & "\" & "Report00.dif" DataFile = MainDir & BatchName & "\" & NextFile & "\" & "Report01.dif" End If

'First, paste in the BatchName and get the Sample Name from the datafile.

Row = Row + 1 If Row <> 5 Then GoTo 20 Workbooks.Open Filename:=DataFile Range("H1:H4").Select Selection.Copy Windows("New.xls").Activate Range(Cells(Row - 2, 2), Cells(Row - 2, 2)).Select Selection.PasteSpecial Paste:=xlAll, Operation:=xlNone, SkipBlanks _ :=False, Transpose:=True Windows(2).Activate ActiveWorkbook.Close

20 Workbooks.Open Filename:=DataFile Range("E1:E4").Select Application.CutCopyMode = False Selection.Copy Windows("New.xls").Activate Range(Cells(Row, 2), Cells(Row, 2)).Select Selection.PasteSpecial Paste:=xlAll, Operation:=xlNone, SkipBlanks _ :=False, Transpose:=True Windows(2).Activate ActiveWorkbook.Close

Workbooks.Open Filename:=InfoFile Range("B1").Select Application.CutCopyMode = False Selection.Copy Windows("New.xls").Activate Range(Cells(Row, 1), Cells(Row, 1)).Select ActiveSheet.Paste Windows(2).Activate ActiveWorkbook.Close NextFile = Dir() Loop 'Check for dashes and replace them with zeros Windows("New.xls").Activate For i = 5 To Row check = ActiveSheet.Cells(i, 2) If check = "-" Then Cells(i, 2) = "0" check = ActiveSheet.Cells(i, 3) If check = "-" Then Cells(i, 3) = "0"

Next i

'We just need to saveas, print and do some file cleanup.

'The intent is not to change any files except the newly created one.

ActiveSheet.Cells(3, 4) = "total retinyl palmitate" ActiveSheet.Cells(3, 5) = "total retinyl palmitate" ActiveSheet.Cells(4, 4) = "(IU/Kg)" ActiveSheet.Cells(4, 5) = "(IU/LB)" Range("D5").Select ActiveCell.FormulaR1C1 = "=IF((RC[-2]+RC[-1])-10000>0, ROUND(RC[-2]+RC[-1], -2), ROUND(RC[-2]+RC[-1], -1))"

Range("D5").Select Selection.AutoFill Destination:=Range(Cells(5, 4), Cells(Row, 4)), Type:=xlFillDefault Range("E5").Select ActiveCell.FormulaR1C1 = "=IF(((RC[-3]+RC[-2])*0.454)-10000>0, ROUND(((RC[-3]+RC[-

2])*0.454), -2), ROUND(((RC[-3]+RC[-2])*0.454), -1))"

Range("E5").Select Selection.AutoFill Destination:=Range(Cells(5, 5), Cells(Row, 5)), Type:=xlFillDefault Columns("A:E").AutoFit Columns("A:A").Select Selection.NumberFormat = "@" Columns("B:E").Select Selection.NumberFormat = "#,##0" With Selection .HorizontalAlignment = xlCenter .VerticalAlignment = xlBottom .WrapText = False .Orientation = xlHorizontal End With Columns("B:E").Select Selection.Columns.AutoFit With ActiveSheet.PageSetup .PrintTitleRows = "" .PrintTitleColumns = "" End With ActiveSheet.PageSetup.PrintArea = "" With ActiveSheet.PageSetup .LeftHeader = "" .CenterHeader = "" .RightHeader = "" .LeftFooter = "" .CenterFooter = "" .RightFooter = "" .LeftMargin = Application.InchesToPoints(0.75)

.RightMargin = Application.InchesToPoints(0.75) .TopMargin = Application.InchesToPoints(1) .BottomMargin = Application.InchesToPoints(1) .HeaderMargin = Application.InchesToPoints(0.5) .FooterMargin = Application.InchesToPoints(0.5) .PrintHeadings = False .PrintGridlines = True .PrintComments = xlPrintNoComments .PrintQuality = 600 .CenterHorizontally = False .CenterVertically = False .Orientation = xlPortrait .Draft = False .PaperSize = xlPaperLetter .FirstPageNumber = xlAutomatic .Order = xlDownThenOver .BlackAndWhite = False .Zoom = 100 End With Sheets("Sheet1").Name = BatchName Workbooks("New.xls").SaveAs Filename:=StoreFile Kill "New.xls" Windows("VitaRes.XLS").Activate ActiveWorkbook.Close saveChanges:=False End Sub

13.2 Vitamin A QC

This macro generates means and coefficients of variation (CV) for each set of response factors. To be acceptable, the CV must be <1.0% (or 3.0% for the cis isomer). The correlation coefficient must be at least 1.0000 to continue analyzing the batch. Percentage cis isomer must be 0.035 to 5.5%.

The spreadsheet contains the following template stored in Sheet1 and saved with the macro:

' Vitamin A QC Macro ' Macro written 3/19/98 by Tim Norden Sub VitA_QC() 'Prompt the user for the subdirectory and datafile name.

'The MainDir and TemplateName statements can be changed to reflect 'where the datafiles are located.

Instrument = InputBox("Please enter the instrument number:", "Instrument") If Instrument = "1" Then MainDir = "Q:\1\DATA\" Else: MainDir = "Q:\2\DATA\" End If BatchName = InputBox("Enter the Batch Name:", "Batchname") filename = InputBox("Enter the File Name containing the Calibration Data without the Extension or Press Enter to Automatically Find the Files:", _

Template

0 0 0 0 13-cis retinyl palmitate 0 0 0 0 all trans retinyl palmitate 1 406.5 1.083593 375.1408 13-cis retinyl palmitate 1 8869.5 21.60995 410.436 all-trans retinyl palmitate 2 813 2.0748 391.8451 13-cis retinyl palmitate 2 17739 43.60413 406.8192 all-trans retinyl palmitate 3 1626 4.176197 389.3495 13-cis retinyl palmitate 3 35478 87.06573 407.4852 all-trans retinyl palmitate mean 385.4451 mean 408.2468 std. dev. 9.0106 std. dev. 1.9249

CV 2.34% <3.00 CV 0.47% <1.00

Corr. Coeff. 0.9999 =1.0000 Corr. Coeff. 1.0000 =1.0000

CIS 24.14783

%CIS 27.7% 3.5-5.5% Target Area 43.60 299h (273ac)

0.5

1.5

2.5

3.5

4.5

0 500 1000 1500 2000

Series1

0 10000 20000 30000 40000

Series1

"Calibration Data File Directory") If filename <> "" Then CalFile = MainDir & BatchName & "\" & filename & ".d\" & "Report01.dif" InfoFile = MainDir & BatchName & "\" & filename & ".d\" & "Report02.dif" ExcelFile = MainDir & BatchName & "\" & filename & ".d\" & "Report01.xls" StoreFile = MainDir & BatchName & "\" & filename & ".d\" & filename GoTo 20 End If 'Test each datafile to determine if the directory holds the valid calibration data.

'This is done by looking in Report00.dif see if it is a "Level 3" calibration sample.

NextFile = Dir(MainDir & BatchName & "\*.d", vbDirectory) 10 Do While NextFile <> "" Pos = InStr(NextFile, ".D") filename = Left(NextFile, Pos - 1) LevelFile = MainDir & BatchName & "\" & filename & ".d\" & "Report00.dif" CalFile = MainDir & BatchName & "\" & filename & ".d\" & "Report01.dif" InfoFile = MainDir & BatchName & "\" & filename & ".d\" & "Report02.dif" ExcelFile = MainDir & BatchName & "\" & filename & ".d\" & "Report01.xls" StoreFile = MainDir & BatchName & "\" & filename & ".d\" & filename Workbooks.Open filename:=LevelFile level = Cells(17, "B") ActiveWorkbook.Close NextFile = Dir() If level <> "3" Then GoTo 10 'First, get the calibration data from the datafile, Report01.dif.

20 Workbooks.Open filename:=CalFile Columns("G:I").Select Selection.Delete Shift:=xlToLeft Range("C1:G3").Select Application.CutCopyMode = False Selection.Copy Windows("vitaQC.XLS").Activate Sheets("Template").Select Range("A4").Select ActiveSheet.Paste

Windows("report01.dif").Activate Range("C4:G6").Select Application.CutCopyMode = False Selection.Copy Windows("vitaQC.XLS").Activate Sheets("Template").Select

Range("G4").Select ActiveSheet.Paste 'Next, get the areas of the cis and trans isomers from the datafile, Report01.xls.

Workbooks.Open filename:=ExcelFile Sheets("Peak").Select check = "0" Row = 2 Do While check <> "" check = Cells(Row, 18) If check = 1 Then GoTo 30 Row = Row + 1 Loop 30 ActiveSheet.Cells(Row, 14).Select Selection.Copy Windows("vitaQC.XLS").Activate Sheets("Template").Select Range("C15").Select ActiveSheet.Paste 'Now get the pathname and sampleinfo data from the datafile, Report02.dif.

Workbooks.Open filename:=InfoFile Range("B1").Select Selection.Copy Windows("vitaQC.XLS").Activate Sheets("Template").Select Range("A1").Select ActiveSheet.Paste Application.CutCopyMode = False With Selection.Font .Name = "Arial" .FontStyle = "Regular" .Size = 14…