method development & laboratory participant perspective

Method Development & Laboratory Participant PerspectiveDenise McClenathanGroup Leader, Elemental Analysis Capability

A Tale of Two Labs in One

“Reference Lab”ICP-MS Method Development and Optimization

Total Digestion & Exhaustive ExtractionTablet and Raw Material Reference Values

“Participant Lab”Study Sample Analysis – Phase 2

ICP-MS and XRF

Elemental Impurities at P&G400+ DRUG PRODUCTS WITH WIDE RANGE OF MATERIALS

Excipients: Salts, Minerals, Botanicals, Organics, Polymers

Actives: Bismuth Subsalicylate, TiO2, ZnO, SnF2,NaF, SeS2, Al/Zr Based Actives

Co-Mingled: Trace EIs and Inorganic RMs

Analytical Challenges: Matrix Effects, Specificity, Digestion

ICP-MS Method Development

Method Developmentfor ICP-MS

Andrei Shauchuk

Kelly Smith

Uniform Procedures for Phase 2Approach to Development and Optimization

• Total Digestion Total Content AssessmentAchieve Mass Bal

• Exhaustive Extraction (w/o HF/HBF4)Evaluate Relative to Total Digestion

• Applicable to Wide Range of InstrumentationIndividually-Pressurized Vessels & Single Reaction Chamber

• Leverage ICP-MS/MS (QQQ) for Selectivity

Agilent 8800QQQ ICP-MS

MilestoneUltraWAVE with ECR

(HCl compatible)

Milestone UP

Titrate acid combination to most complex ingredient

Stabilize the analytes (ex. Hg with Au or HCl)

Prevent formation of insoluble fluorides (Mg, Al, Ca, etc.)– Complex excess fluoride with boric acid (2 step process)– Prepare ultra-trace HBF4 from HF & boric acid

Ultra-trace HBF4 not commercially available

Keys to Total Digestion ApproachesMORE THAN JUST “NUKING” THE MATRIX

Sample Preparation – Total DigestionRAW MATERIALS AND TABLETS

7

(1) Weigh sample Tablet (0.25 g)Raw material (0.01-0.15 g)

(2) Add reagents0.5 mL of HCl2.5 mL of HNO30.5 mL H3PO41.0 mL of HBF4

(3) Microwave digest

(4) Transfer to 50 mL tube and dilute to volume

(5) Prepare 50X dilution with internal standard

Single Reaction ChamberRamp and Hold @ 250 °C

Individually-Pressurized VesselRamp and Hold @ 180 °C

Magnesium Aluminum Silicate ResultsTOTAL DIGESTION AND ICP-MS/MS

0

2

4

6

8

10

12

1 2 3 4 5 6 7

Con

cent

ratio

n (µ

g/g)

BLOQ BLOQ

Analyte Detection Scheme

Concentration(µg/g)

Cobalt59 59 [He] 2.00

59 75 [O2] 1.83

Vanadium51 51 [NH3] 10.1

51 135 [NH3] 11.2

Confirmation of Specificity

Interferences at m/z 5943Ca16O+, 42Ca16O1H+, 24Mg35Cl+,

36Ar23Na+, 40Ar18O1H+, 40Ar19F+

Interferences at m/z 5134S16O1H+, 35Cl16O+, 38Ar13C+, 36Ar15N+, 36Ar14N1H+, 37Cl14N+,

36S15N+, 33S18O+, 34S17O+

Summation ApproachLEVEL 1 TABLET – TOTAL DIGESTION

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Raw MaterialsMagnesium Aluminum SilicateFerric Oxide RedSilicon Dioxide, As Co Hg Silicon Dioxide, Cd Pb NiMicrocrystalline CellulosePregelatinized StarchStearic AcidLactose

Finished ProductLevel 1 TabletC

once

ntra

tion

(µg/

g)

Mass Balance AssessmentLEVEL 1 TABLET – TOTAL DIGESTION

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14



once

ntra

tion

(µg/

g)


0

5

10

15

20

25

30

1 2 3 4 5 6 7 8 9 10 11 12 13 14



once

ntra

tion

(µg/

g)


0

10

20

30

40

50

1 2 3 4 5 6 7 8 9 10 11 12 13 14



once

ntra

tion

(µg/

g)

Mass Balance AssessmentALL TABLET LEVELS – TOTAL DIGESTION

Percent of the Predicted Value(Measured Tablet / RM Summation)

Sample Arsenic Cadmium Mercury Lead Cobalt Nickel Vanadium

Tablet 1 100% 108% 67% 105% 100% 103% 98%

Tablet 2 99% 101% 84% 98% 101% 103% 101%

Tablet 3 98% 98% 97% 97% 97% 97% 100%

Sample Preparation – Exhaustive ExtractionRAW MATERIALS AND TABLETS

15

(1) Weigh sample Tablet (0.25 g)Raw material (0.01-0.15 g)

(2) Add reagents10 mL of HNO350 µL of 1000 ppm Au

(3) Microwave digest

(4) Transfer to 50 mL tube and dilute to volume

(5) Prepare 50X dilution with internal standard

Single Reaction ChamberRamp and Hold @ 175 °C

Individually-Pressurized VesselRamp and Hold @ 175 °C

Exhaustive Extraction – Tablets and Raw MaterialsCOMPARISON TO TOTAL DIGESTION

Arsenic Cadmium Mercury Lead Cobalt Nickel Vanadium

Tablet 1 98% 98% 93% 94% 99% 101% 99%

Tablet 2 106% 103% 95% 102% 97% 100% 99%

Tablet 3 111% 103% 96% 99% 98% 93% 100%

Magnesium Aluminum Silicate 92% N/A N/A 98% 95% 100% 77%

Ferric Oxide 93% N/A N/A N/A N/A N/A 105%

SiO2 (As, Hg, Co) 110% N/A 118% N/A 113% N/A N/A

SiO2 (Cd, Pb, Ni) N/A 106% N/A 105% N/A 107% N/A

N/A – Results BLOQ

Percent of Total Digestion(Exhaustive Extraction vs RM Summation)

Reference Lab Learnings

Exhaustive extraction can be equivalent to total digestion when the procedure is appropriately optimized for the matrix.

ICP-MS Participant Results

Total Digestion Exhaustive ExtractionICP-MS & ICP-MS/MS

Agilent 8800 QQQ ICP-MS

MilestoneUltraWAVE

Agilent 7900 ICP-MS

Instrumentation / ApproachRAW MATERIALS AND TABLETS

Usa Rattanaudompol

Total Digestion Results for Level 1 TabletCOMPARISON TO REFERENCE VALUE

Triple Quad Single Quad

Perc

ent o

f Ref

eren

ce V

alue

0%

20%

40%

60%

80%

100%

120%

140%

1 2 3 4 5 6 7

*Represents LOQ Value

*

Summation Approach & Mass Balance AssessmentLEVEL 1 TABLET – TOTAL DIGESTION



once

ntra

tion

(µg/

g)

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 140

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14

ICP-MS (SQ) ICP-MS/MS (QQQ)

Summary of ICP-MS Results for Tablets

Sample Preparation ICP-MS Arsenic Cadmium Mercury Lead Cobalt Nickel Vanadium

Tablet 1Total

QQQ 96% 96% 26% 97% 94% 90% 93%SQ 96% 97% 26% 96% 95% BLOQ 97%

ExtractQQQ 90% 96% 26% 94% 98% 100% 100%SQ 92% 97% 26% 94% 98% BLOQ 99%

Tablet 2Total

QQQ 97% 104% 22% 103% 95% 97% 97%SQ 98% 105% 21% 102% 96% 93% 102%

ExtractQQQ 95% 103% 21% 101% 97% 100% 103%SQ 97% 104% 22% 103% 97% 97% 101%

Tablet 3Total

QQQ 101% 107% 58% 105% 96% 101% 90%SQ 100% 109% 56% 106% 98% 97% BLOQ

ExtractQQQ 100% 107% 62% 100% 101% 104% 84%SQ 99% 107% 62% 103% 100% 99% BLOQ

COMPARISON TO REFERENCE VALUES

Percent Recovery vs Reference

Summary of ICP-MS Results for RM Summation

Sample Preparation ICP-MS Arsenic Cadmium Mercury Lead Cobalt Nickel Vanadium

Tablet 1Total

QQQ 102% 96% 153% 94% 98% 99% 100%SQ 103% 98% 147% 97% 100% 23%* 103%

ExtractQQQ 103% 99% 158% 94% 99% 96% 98%SQ 101% 97% 159% 96% 99% 23%* 93%

Tablet 2Total

QQQ 107% 105% 106% 101% 102% 100% 86%SQ 104% 105% 118% 103% 99% 41%* 97%

ExtractQQQ 104% 105% 127% 100% 99% 94% 95%SQ 102% 104% 128% 102% 99% 42%* 88%

Tablet 3Total

QQQ 107% 105% 106% 101% 102% 100% 86%SQ 107% 107% 102% 104% 103% 96% 0%*

ExtractQQQ 107% 108% 110% 102% 103% 99% 67%*SQ 105% 106% 110% 104% 104% 98% 0%*

COMPARISON TO REFERENCE VALUES

Percent Recovery vs Reference

*BLOQ values set to zero in calculation

Difference in Mercury Results

0

10

20

30

40

50

3/2017 9/2017 4/2018 10/2018 5/2019 12/2019 6/2020 1/2021

800

900

1000

1100

Hg

Con

cent

ratio

n (µ

g/g)

SiO2

INVESTIGATING UNEXPECTED DATA

Analysis Date

Lab error?

Stability? Level 1 Tablet Level 2 Tablet Level 3 Tablet

Difference in Mercury Results

0

10

20

30

40

50

3/2017 9/2017 4/2018 10/2018 5/2019 12/2019 6/2020 1/2021

800

900

1000

1100 SiO2

Level 1 Tablet Level 2 Tablet Level 3 Tablet

INVESTIGATING UNEXPECTED DATA

Analysis Date

Lab error?

Stability?

Use RM result to extrapolate back to

formulation date

Hg

Con

cent

ratio

n (µ

g/g)

ICP-MS Learnings

Good agreement across digestion approaches and ICP-MS systems, with a few exceptions.

Mercury was unstable in the tablets AGAIN.

XRF Participant Results

Wavelength Dispersive

XRF

Instrumentation / ApproachTABLET ANALYSIS

Bruker Tiger S8WD-XRF

Standard PreparationBlend Raw MaterialsAdd Liquid Standard

Dry in FurnaceGrind & Press

Sample PreparationGrind & Press

Christina Haven

XRF Results for Tablet 1COMPARISON TO REFERENCE VALUE

0

5

10

15

20

25

1 2 3 4 5 6 7

Con

cent

ratio

n (µ

g/g)

Reference (ICP-MS)XRF

*

*Not measured


Con

cent

ratio

n (µ

g/g)


0

5

10

15

20

25

30

1 2 3 4 5 6 7*

*Not measured


Con

cent

ratio

n (µ

g/g)


0

10

20

30

40

50

1 2 3 4 5 6 7*

*Not measured

Arsenic Cadmium Mercury Lead Cobalt Nickel Vanadium

Tablet 1 92% 157% N/A 63% 172% 200% 102%

Tablet 2 81% 94% N/A 58% 87% 107% 82%

Tablet 3 85% 87% N/A 64% 73% 100% 211%

N/A – Not Measured

Percent of Reference Value

XRF Results – All TabletsCOMPARISON TO REFERENCE VALUE

Calibration and Drift – Arsenic DIGGING DEEPER INTO THE XRF RESULTS

Standard Concentration(µg/g)

Residual Error

QCRecovery

1 4.5 -60% 100%2 30 -12% N/A3 15 -9% N/A4 9.0 22% N/A5 9.0 83% N/A6 45 6% N/A7 4.5 9% 86%8 23 -5% N/A

Calibration and Drift – Lead DIGGING DEEPER INTO THE XRF RESULTS


Residual Error

QCRecovery

1 2.2 -5% 68%2 3.8 -5% N/A3 6.3 -63% N/A4 11 -87% N/A5 16 -92% N/A6 8.8 -1% N/A7 2.4 -29% 79%8 3.5 20% N/A

Calibration and Drift – Vanadium DIGGING DEEPER INTO THE XRF RESULTS


Residual Error

QCRecovery

1 52 -11% 83%2 78 -13% N/A3 122 -20% N/A4 222 -16% N/A5 65 -1% N/A6 300 3% N/A7 31 3% 89%8 160 -5% N/A

XRF Learnings

XRF performed better than expectations.

Not practical for EI screening on ever-changing number of products and materials.

MIGHT work for control method for a formulation

Broader Learnings

• Investing in alternate approaches / instrumentation for flexibility, robustness, business continuity

• Digging into “WHY” for method training/transfers

• Balancing familiarity of the method and matrices with embracing the “fresh perspectives”

WHAT WE ARE DOING DIFFERENT

Acknowledgements

Procter & GambleKelly SmithAndrei ShauchukUsa RattanaudompolChristina HavenRoy Dobson

Trace Analytical Challenges Sub-TeamDonna Seibert, PerrigoJames Harrington, RTI International

Improving everyday life.

method development & laboratory participant perspective

Documents