method development & laboratory participant perspective
TRANSCRIPT
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
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)
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
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 2 TABLET – TOTAL DIGESTION
0
5
10
15
20
25
30
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 2 TabletC
once
ntra
tion
(µg/
g)
Mass Balance AssessmentLEVEL 3 TABLET – TOTAL DIGESTION
0
10
20
30
40
50
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 3 TabletC
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.
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
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)
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
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.
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
XRF Results for Tablet 2COMPARISON TO REFERENCE VALUE
Con
cent
ratio
n (µ
g/g)
Reference (ICP-MS)XRF
0
5
10
15
20
25
30
1 2 3 4 5 6 7*
*Not measured
XRF Results for Tablet 3COMPARISON TO REFERENCE VALUE
Con
cent
ratio
n (µ
g/g)
Reference (ICP-MS)XRF
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
Standard Concentration(µg/g)
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
Standard Concentration(µg/g)
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