Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in...

7
Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden, 2 Elisabeth Morgan, 2 Amalendu Sarkar, 3 Andrew White 4 1 Thermo Fisher Scientific, San Jose, CA, USA; 2 Smithers Rapra, Shrewsbury, UK; 3 Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Transcript of Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in...

Page 1: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical ProductsKate Comstock,1 Andrew Feilden,2 Elisabeth Morgan,2 Amalendu Sarkar,3 Andrew White4 1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Page 2: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

2 Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical ProductsKate Comstock1; Andrew Feilden2; Elisabeth Morgan2; Amalendu Sarkar3; Andrew White4

1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

ConclusionThis study demonstrated a comprehensive extractable analysis workflow utilizing multiple techniques: HR-LCMS, GCMS, ICPMS, data processing software, anddatabase searching. This workflow followed recommended analytical methods byPQRI[2]. HS-GCMS was carried out but the data has not been reported.

The DI water and IPA extraction profiles of the four types of medical grade rubber stoppers were quickly established by using this workflow.

The UHPLC/HRAM full MS/HCD MS2 with rapid polarity switching in a single run data acquisition, coupled with novel database search, significantly increase the confidenceand throughput of routine extractable & leachable analysis, in particular for unknown components identification and structure characterization.

The GCMS and LCMS analysis are complementary to each other and necessary to give complete coverage of extractables.

References1. FDA CFR 21.94, CFR 66011(b) and 600.11(h), CFR 211.160 2. PQRI “L/E Recommendations to the FDA”

http://www.pqri.org/publications/index.asp

AcknowledgementsThe authors would like to thank Sukhy Toot for her contributions, Qure Medical for providing the rubber stoppers, Smithers Rapra for the extractions, and Buchi for the loanof the speed extractor.

Introduction Rubbers & plastics are widely used in medical & drug delivery devices and packaging materials. Extractables & leachables assessment of all materials, especially from elastomeric & oligomeric components, forms an integral part of the submission for approval of a new drug product or medical device [1].

Extractable = possible impact. Test the material

Leachable = actual impact. Test the product The mass spectrometer plays an important role in E&L identification and structure elucidation, as it is coupled with many techniques for definitive analysis, see figure 1. Here we present a comprehensive workflow for medical grade rubber stopper extractable analysis using multiple techniques including HR-LCMS, GCMS, and ICPMS, followed by data processes using novel software and database searching.

FIGURE 1. Potential analytical techniques with increasing chance of extractables and leachables being found as the molecular weight decreases.

LCMS Analyses Sample Preparation Four different types of medical grade rubber stoppers, sample-A, sample-B, sample-C, and sample-D, from Qure Medical, were extracted using DI water and IPA utilizing reflux extraction and a pressurized liquid extraction system. The extracts solutions were analyzed directly by LCMS.

Liquid Chromatography LC separations were carried out on the Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system consisting of: DGP-3000RS pump, WPS-3000RS sampler, TCC-3000RS column compartment, and DAD-3000RS UV detector Column: Thermo Scientific™ Hypersil C18, 2.1x150 mm 1.9 µm Column Temp: 50ºC LC Mobile phase: A: H2O B: MeOH C: 50 mM Ammonium Acetate

Mass Spectrometry MS analyses were carried out on the Thermo Scientific™ Q Exactive™ mass spectrometer using both electrospray ionization (ESI) and atmosphere pressure chemical ionization (APCI). High resolution full scan MS and top 3 MS/MS data were collected in a data-dependent fashion at a resolving power of 70,000 and 17,500 (FWHM m/z 200) with polarity switching. The scan range is m/z150-1500. Stepped NCE (Normalized Collision Energy) setting was: 30, 40, 50.

FIGURE 4. HR Full Scan and HCD MS2 for Component ID and Structure Elucidation

LCMS and GCMS results show DI water extracts using both Speedy and Reflux techniques were “clean”. “Triisopropanolamine” is the major extractable.

Complex profiles of IPA extractions were observed from both LCMS and GCMS analysis. Complete extractable list not shown.

IPA reflux shows higher extraction efficiency compared with IPA speed extraction. However, the speed extractor conditions were not optimized for this study

Results show that GCMS and LCMS analysis are complementary to each other andnecessary to give a fuller picture of the extractable profile.

FIGURE 5. mzCloud library Search Results for Irganox 1010

FIGURE 6. GCMS Chromatogram of IPA_REFLUX of Sample-D

FIGURE 7. ICPMS Results for the Four Rubber Stoppers (ppb)

GCMS AnalysesMethod and InstrumentationThe DI water samples were extracted with Hexane. The samples in 2.0 mL GC vials were introduced in split injection mode into the Thermo Scientific™ TRACE™ Ultra GasChromatograph using a Thermo Scientific™ TriPlus RSH™ Autosampler. TG-5ms (30 m x0.25mm x0.25µm) column was used. Compounds were detected and identified with the ISQ single Quad mass spectrometer.

mzCloud Spectral Database Searching

TABLE 1. Components Identified from IPA Reflux of Sample-A (Partial List)

FIGURE 3. MS Chromatogram of IPA Reflux of Sample-A (ESI+) ICPMS AnalysesThe ICPMS samples were prepared by placing the rubber stoppers in 25 ml DI water and25 ml 2% nitric acid and soaked at RT for 24 hours. The analyses were conducted onThermo Scientific™ iCAP™ Q ICP-MS with He KED (Kinetic Energy Discrimination) interference reduction mode setting. To determine if trace and potentially toxic metals were leached from the stoppers, theUSP<232> Class1 & 2 elements and additional elements which are commonly analyzed byICP-MS were determined.The analysis results for the four types of rubber stoppers showed that they are clean of all Class 1 & 2 elements, see Figure 7 for the ICPMS results. In addition, the system control software Qtegra provides a full 21CFR Part 11 tool set to operate under compliant environments.

Element Sample-1 DIwater

Sample-1 Nitric Acid

Sample-2 DI water

Sample-2 Nitric Acid

Sample-3 DIwater

Sample-3 Nitric Acid

Sample-4 DIwater

Sample-4 Nitric Acid LOD

75As (KED) ND ND ND ND ND ND ND ND <0.0233111Cd (KED) 0.009 0.006 0.003 0.007 0.010 0.009 0.276 0.070 <0.0023202Hg (KED) ND ND ND ND ND ND ND ND <0.0054208Pb (KED) 0.061 0.069 0.018 0.124 0.100 0.106 0.159 0.122 <0.0008

9Be (KED) ND ND ND ND ND ND ND ND <0.036211B (KED) ND ND 0.541 ND 0.853 0.596 ND ND <0.5229

23Na (KED) 14.326 29.535 8.197 13.575 25.630 20.648 30.074 18.071 <0.156824Mg (KED) 1.835 4.091 1.160 1.013 2.802 4.236 2.009 1.531 <0.023127Al (KED) 0.420 3.688 0.479 2.325 1.029 3.615 0.867 7.665 <0.3239K (KED) 8.246 11.185 5.921 5.295 13.580 11.235 16.088 6.645 <1.796448Ti (KED) 0.033 0.930 0.033 0.605 0.045 0.202 ND 0.107 <0.031451V (KED) 0.526 0.518 ND ND ND ND 0.051 0.044 <0.033952Cr (KED) ND 0.146 ND 0.103 ND 0.241 ND 0.108 <0.0072

55Mn (KED) 0.044 0.161 0.053 0.052 0.045 0.188 0.067 0.102 <0.006556Fe (KED) ND 9.962 ND 3.893 0.570 18.668 0.903 9.600 <0.017559Co (KED) 0.001 0.003 0.001 0.001 0.008 0.010 0.011 0.009 <0.000960Ni (KED) 0.097 0.123 0.047 0.075 0.254 0.420 0.715 0.281 <0.006365Cu (KED) 0.558 0.741 0.237 0.341 0.849 1.576 1.603 1.709 <0.00766Zn (KED) 13.137 11.712 31.153 43.436 4.049 13.275 81.914 51.445 <0.517378Se (KED) ND ND ND ND ND ND ND ND <0.078188Sr (KED) 0.045 0.070 0.037 0.032 0.082 0.087 0.076 0.078 <0.0036

95Mo (KED) ND ND ND ND ND ND ND ND <0.4285101Ru (KED) ND ND ND ND ND ND ND ND <0.0006103Rh (KED) ND ND ND ND ND ND ND ND <0.0648105Pd (KED) ND ND ND ND ND ND ND ND <0.0038107Ag (KED) ND ND ND ND ND ND ND ND <0.0226121Sb (KED) 0.007 0.017 0.007 0.020 0.017 0.124 0.040 0.052 <0.0012137Ba (KED) 0.259 0.231 0.165 0.270 1.646 2.420 1.250 1.125 <0.0124193Ir (KED) ND ND ND ND ND ND ND ND <0.0003195Pt (KED) ND ND ND ND ND ND ND ND <0.0004205Tl (KED) ND ND ND ND ND ND ND ND <0.0046

LCMS Result The all-in-one full scan/top3 ms/ms with polarity switching data acquisition using both ESI, APCI ionization ensures the detection of structurally diversified compounds. It provides comprehensive extractable profiles of the rubber stoppers, see Figures 2 and 3.

SR_Reflux_IPA_A_ESI_1 03/30/14 20:44:49Hypersil GOLD C18 150x2.1 1.9 um A: H2O B: MeOH C: 50 mMAmmonium AcetateRT: 0.00 - 45.01

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44Time (min)

0

10000

20000

30000

uAU

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

28.1526.72

30.2022.47

36.0729.0326.46 44.0625.32 31.430.90 42.3933.6721.851.85 17.78 36.89 40.8712.092.96 4.39 16.955.76 9.02 15.289.9730.20

22.47

36.0729.0326.6826.46

44.0631.43 42.3933.6721.851.02 20.5117.78 36.89 38.8812.09 39.9515.689.02 13.221.84 9.876.805.714.5328.15

26.72

29.410.90 22.43 24.941.85 21.58 32.45 33.66 36.03 38.04 40.87 43.5119.952.96 18.294.39 16.605.76 14.967.27 8.49 13.249.9732.42

36.0828.11 33.1230.2327.42 43.5636.99 42.4139.3525.1827.23

24.9023.70

20.5019.461.12 17.4316.29

NL: 6.29E9Base Peak MS SR_Reflux_IPA_A_ESI_1

NL: 2.65E9Base Peak F: FTMS + p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 6.29E9Base Peak F: FTMS - p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 3.45E4Total Scan PDA SR_Reflux_IPA_A_ESI_1

ESI (+/-) TIC

ESI (+)

ESI (-)

PDA

A search was also conducted against Thermo ScientificTM mzCloudTM Library, a newlydeveloped high resolution spectral database. mzCloud library provides several search criteria for small molecule structure identification using tandem mass spectra, including spectra, fragments, precursor ions, etc, all of which can be very useful for unknown structure elucidation. Figure 5 shows identification of Irganox 1010 using the ms/ms spectrum search feature. The accuracy of searching result is indicated by matching score between the query and library spectra. GCMS Instrument Conditions GCMS identified lower molecular weight and

volatile extractables which complement LCMS results. GCMS results showed that DI water extractions using both techniques were “clean” and more extractables were detectedfrom IPA extractions. Within the four rubber stoppers, sample C&D had more low molecular weight extractables detected by GCMS from both extraction techniques, see Figure 6 and 7.

Ace

toph

enon

e

Dim

ethy

lphe

nylm

etha

nol

L-Li

mon

ene

Dim

ethy

lben

zylm

ethy

l eth

er

Dip

heny

l eth

er

Triis

opro

pano

lam

ine

+Lon

gipi

nene

BH

T

Non

ylbe

nzen

e de

rivat

ive

Patc

houl

ene

1,5,

9-Tr

imet

hyl c

yclo

dode

catri

ene

C20

Uns

atur

ated

hyd

roca

rbon

Patc

houl

ene

isom

er

4/20/2014 3:47:32 AMIPA_REFLUX_DRT: 4.66 - 27.35

6 8 10 12 14 16 18 20 22 24 26Time (min)

01000000000

2000000000

3000000000

4000000000

5000000000

6000000000

7000000000

8000000000

9000000000

10000000000

11000000000

12000000000

13000000000

14000000000

Rel

ative

Abu

ndan

ce

6.6211.946.43

9.98 12.0110.6113.1110.95 14.4015.2615.396.68 18.1514.21 19.02 20.757.16 22.07 23.329.876.00 24.92 27.14

50 63 75 88 100 113 125 138 150 163 175 188 200 213 225 238 250 263 275 288 300 313 325 338 350 363 375 388 400 413 425 438 450

0

10

20

30

40

50

60

70

80

90

100155.070489.0598

163.1330

145.1225111.0442

101.0963

57.0702173.0810

217.1075

317.1964261.134067.0544 127.0753

255.1593

O

OO

OO

O

OO

OO

OOH

O

OO

O

OOO

OHO

OO

OOH

OO

O

O

O 2H

OOO

O

O 2HOO

O

O

O2HO OH

OO

OO

O OHO

OH

OH

180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580m/z

05

101520253035404550556065707580859095

100

Rel

ative

Abu

ndan

ce

452.3227C22H46O8N1.9203 ppm

435.2955C22H43O8

0.6606 ppm

(M+H)+

(M+NH4)+

OO

OO

OO

O

O

Bis[2-(2-butoxyethoxy0ethyl] adipateMolecular Formula : C22H42O8Formula Weight : 434.2880

FIGURE 2. LCMS Chromatogram of IPA Reflux of Sample A

Figure 7. GCMS Spectra of Compounds Identified in IPA_REFLUX of Sample-D (Partial)

Component Detection and Structure ElucidationThe High Resolution Accurate Mass (HRAM) data were processed using differential analysis software SIEVETM 2.1 for component extraction. ChemSpider database searching was carried out to obtain possible structures of extracted components. While many possible hits were obtained for each component, to determine the correct structures, “Thermo ScientificTM Mass FrontierTM Software”, a small molecule structure analysissoftware, was used. The “HighChem Fragmentation LibraryTM ” in Mass Frontier 7.0 has extensive published literature references. For each proposed structure, the “Fragments and Mechanisms” feature in Mass Frontier was used to generate predicted “fragments and mechanisms” through HighChem Library search, see figure 4-b. A high degree of correlation between predicted and experimental fragments (indicated in red, see figure 4-c) confirms the proposed structure. Mass Frontier then automatically annotates the matching fragments based on library search results, see figure 4-c.

m/z 452.3218

OO

OO

OO

O

O

NH4

OO

OO

OO

OH

Om/z 435.2952

+H

OO

O

OO

O

m/z 304.1880

Lib

OHO

O

O

O

OO

m/z 334.1986

Lib

OO

OHO

OO

O

Om/z 435.2952

+H

Lib

O 2HO

O

m/z 163.1329

rHB

rHB

Lib

rHB

m/z 452.3218

OO

OO

OO

O

O

NH4

OHO

Om/z 162.1250

Lib

OHO

Om/z 162.1250

Lib

OOH

m/z 89.0597

i

Lib

Summary of Reflux and Speed Extractions using IPA and DI Water

4-a. HRAM Full MS to Determine Elemental Composition

4-c. HRAM MS/MS Spectra For Structure Elucidation

4-b. HighChem Fragmentation Library Search to Predict Fragments and Mechanism

RT: 6.41

50 100 1500

50

100 105.0177.10

120.0451.12

77.9974.1391.16

134.07147

O

Acetophenone MW 120.04

RT:6.58

50 100 1500

50

100 43.01

121.08

77.11

45.1351.02

78.14105.05

91.1559.15

136.04141.04

Dimethylphenylmethanol M+ 136.09

OH

RT:10.93

50 100 1500

50

100 146.12

70.02

88.0842.03 98.1859.14

102.18

128.18158.09176.08

TriisopropanolamineM+ 191.15

OHN

OH

OH

RT:11.90

50 100 150 2000

50

100 57.01

205.14

81.15

145.16105.1467.12220.2391.04 177.1341.06

121.15131.16

149.17189.18

BHTM+ 220.18

OH

1,5,9 Trimethyl Cyclododecatriene MW. 204.35 Da

RT:12.46

50 100 150 2000

50

100 68.15

67.00

93.06

107.16

79.13 121.1853.12

41.15189.14

147.19161.16 204.19

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45Time (min)

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Rel

ative

Abu

ndan

ce

** * * * * *

*

P3

PEG P9

P13

P19 P20

P21

P18B B

B

Erucamide

Irganox1010

Phthalate

P10

P22Irganox 1076

DSDMA

* PPGB-BackgroundP- Extractable

P7P14

P15P16 P17B

P8(-)

P6(-)P5P4

P2

P1

P11

B

P12

Peak ID RT Mode Measured(M+H) +

Calculated(M+H) +

ElementalComposition

Error (ppm)

1 21.48 APCI (+) 235.1691 235.1693 C15H22O2 -0.672 22.43 ESI (-) 199.1707 (M-H) – 199.1693 (M-H) – C12H24O2 2.03 22.47 ESI (+) 435.2956 (M+NH4) + 435.2952 C22H42O8 0.864 24.94 ESI (-) 227.2017 (M-H) – 227.2006 (M-H) – C14H28O2 2.235 24.87 ESI (+) 219.1743 219.1743 C15H22O 1.86 26.72 ESI (-) 255.2330 (M-H) – 255.2391 (M-H) – C16H32O2 3.57 27.75 ESI (+) 282.2791 282.2791 C18H36ON -0.118 28.15 ESI (-) 283.2643 (M-H) – 283.2632 (M-H) – C18H36O2 2.89 28.19 ESI (+) 383.3396 (M+NH4) + 383.3396 (M+NH4) + (C24H47ON)+NH4 0.42

10 28.71 ESI (+) 284.2946 284.2948 C18H37O1N1 -0.511 29.14 ESI (+) 325.3097 325.3101 C21H40O2 1.312 29.36 ESI (+) 319.2992 319.2995 C22H38O1 -1.113 29.41 ESI (-) 311.2963 (M-H) – 311.2945 (M-H) – C20H40O2 2.414 31.43 ESI (+) 340.3568 340.3574 C22H46ON -1.815 31.77 ESI (+) 366.3729 366.3730 C24H48ON -0.516 32.52 ESI (+) 409.3100 409.3101 C28H40O2 -0.3417 33.70 ESI (+) 1227.00 1227.00 ?18 36.00 ESI (+) 1194.8170 (M+NH4) + 1194.8179 (M+NH4) + C73H108O12 -0.319 37.57 ESI (+) 663.4536 663.4537 C42H63O4P -0.1920 38.85 ESI (+) 522.5969 522.5972 C36H76N -0.2321 42.39 ESI (+) 548.5035 (M+NH4) + 548.5035 (M+NH4) + (C35H62O3)+NH4 -0.3322 44.08 ESI (+) 550.6285 550.6285 C38H80N1 -0.2

TABLE 2. Proposed Structures of Identified Compounds (Partial List)

O

O

1 O

O

OO

OO

OO

O

O2 3

6 OH

OOH

O

8NH

O

9

N 2H

O

10

13 OH

O

12OH

11 O

O

O

O O

O

O

O

OO

OH

OH

OH

OH

O

OH

O N

18

19

21 22

Palmitic Acid

Irganox 1010

Irganox 1076

Stearic Acid

Stearamide

OleamideN 2H

O

7

DSDMA Irgafos 168 oxidation product

GC

HPLC CE

SEC GFC

Polarity

Molecular Weight

Derivatization

Pyrolysis

HS-GC

OP

O

OO

Two fragments (highlighted) and corresponding mechanisms shown here to demonstrate the process.

HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.

Page 3: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

3Thermo Scientific Poster Note • PN-64219-IMSC 0814S

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock1; Andrew Feilden2; Elisabeth Morgan2; Amalendu Sarkar3; Andrew White4

1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Conclusion This study demonstrated a comprehensive extractable analysis workflow utilizing multiple techniques: HR-LCMS, GCMS, ICPMS, data processing software, and database searching. This workflow followed recommended analytical methods by PQRI[2]. HS-GCMS was carried out but the data has not been reported.

The DI water and IPA extraction profiles of the four types of medical grade rubber stoppers were quickly established by using this workflow.

The UHPLC/HRAM full MS/HCD MS2 with rapid polarity switching in a single run data acquisition, coupled with novel database search, significantly increase the confidence and throughput of routine extractable & leachable analysis, in particular for unknown components identification and structure characterization.

The GCMS and LCMS analysis are complementary to each other and necessary to give complete coverage of extractables.

References 1. FDA CFR 21.94, CFR 66011(b) and 600.11(h), CFR 211.160 2. PQRI “L/E Recommendations to the FDA”

http://www.pqri.org/publications/index.asp

Acknowledgements The authors would like to thank Sukhy Toot for her contributions, Qure Medical for providing the rubber stoppers, Smithers Rapra for the extractions, and Buchi for the loan of the speed extractor.

Introduction Rubbers & plastics are widely used in medical & drug delivery devices and packaging materials. Extractables & leachables assessment of all materials, especially from elastomeric & oligomeric components, forms an integral part of the submission for approval of a new drug product or medical device [1]. Extractable = possible impact. Test the material

Leachable = actual impact. Test the product The mass spectrometer plays an important role in E&L identification and structure elucidation, as it is coupled with many techniques for definitive analysis, see figure 1. Here we present a comprehensive workflow for medical grade rubber stopper extractable analysis using multiple techniques including HR-LCMS, GCMS, and ICPMS, followed by data processes using novel software and database searching. FIGURE 1. Potential analytical techniques with increasing chance of extractables and leachables being found as the molecular weight decreases.

LCMS Analyses Sample Preparation Four different types of medical grade rubber stoppers, sample-A, sample-B, sample-C, and sample-D, from Qure Medical, were extracted using DI water and IPA utilizing reflux extraction and the Buchi Speed Extractor. The extracts solutions were analyzed directly by LCMS.

Liquid Chromatography LC separations were carried out on the Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system consisting of: DGP-3000RS pump, WPS-3000RS sampler, TCC-3000RS column compartment, and DAD-3000RS UV detector Column: Thermo Scientific™ Hypersil C18, 2.1x150 mm 1.9 µm Column Temp: 50ºC LC Mobile phase: A: H2O B: MeOH C: 50 mM Ammonium Acetate

Mass Spectrometry MS analyses were carried out on the Thermo Scientific™ Q Exactive™ mass spectrometer using both electrospray ionization (ESI) and atmosphere pressure chemical ionization (APCI). High resolution full scan MS and top 3 MS/MS data were collected in a data-dependent fashion at a resolving power of 70,000 and 17,500 (FWHM m/z 200) with polarity switching. The scan range is m/z150-1500. Stepped NCE (Normalized Collision Energy) setting was: 30, 40, 50.

FIGURE 4. HR Full Scan and HCD MS2 for Component ID and Structure Elucidation

LCMS and GCMS results show DI water extracts using both Speedy and Reflux techniques were “clean”. “Triisopropanolamine” is the major extractable.

Complex profiles of IPA extractions were observed from both LCMS and GCMS analysis. Complete extractable list not shown.

IPA reflux shows higher extraction efficiency compared with IPA speed extraction. However, the speed extractor conditions were not optimized for this study

Results show that GCMS and LCMS analysis are complementary to each other and necessary to give a fuller picture of the extractable profile.

FIGURE 5. mzCloud library Search Results for Irganox 1010

FIGURE 6. GCMS Chromatogram of IPA_REFLUX of Sample-D

FIGURE 7. ICPMS Results for the Four Rubber Stoppers (ppb)

GCMS Analyses Method and Instrumentation The DI water samples were extracted with Hexane. The samples in 2.0 mL GC vials were introduced in split injection mode into the Thermo Scientific™ TRACE™ Ultra Gas Chromatograph using a Thermo Scientific™ TriPlus RSH™ Autosampler. TG-5ms (30 m x0.25mm x0.25µm) column was used. Compounds were detected and identified with the ISQ single Quad mass spectrometer.

mzCloud Spectral Database Searching

TABLE 1. Components Identified from IPA Reflux of Sample-A (Partial List)

FIGURE 3. MS Chromatogram of IPA Reflux of Sample-A (ESI+) ICPMS Analyses The ICPMS samples were prepared by placing the rubber stoppers in 25 ml DI water and 25 ml 2% nitric acid and soaked at RT for 24 hours. The analyses were conducted on Thermo Scientific™ iCAP™ Q ICP-MS with He KED (Kinetic Energy Discrimination) interference reduction mode setting. To determine if trace and potentially toxic metals were leached from the stoppers, the USP<232> Class1 & 2 elements and additional elements which are commonly analyzed by ICP-MS were determined. The analysis results for the four types of rubber stoppers showed that they are clean of all Class 1 & 2 elements, see Figure 7 for the ICPMS results. In addition, the system control software Qtegra provides a full 21CFR Part 11 tool set to operate under compliant environments.

Element Sample-1 DI water

Sample-1 Nitric Acid

Sample-2 DI water

Sample-2 Nitric Acid

Sample-3 DI water

Sample-3 Nitric Acid

Sample-4 DI water

Sample-4 Nitric Acid LOD

75As (KED) ND ND ND ND ND ND ND ND <0.0233 111Cd (KED) 0.009 0.006 0.003 0.007 0.010 0.009 0.276 0.070 <0.0023 202Hg (KED) ND ND ND ND ND ND ND ND <0.0054 208Pb (KED) 0.061 0.069 0.018 0.124 0.100 0.106 0.159 0.122 <0.0008

9Be (KED) ND ND ND ND ND ND ND ND <0.0362 11B (KED) ND ND 0.541 ND 0.853 0.596 ND ND <0.5229

23Na (KED) 14.326 29.535 8.197 13.575 25.630 20.648 30.074 18.071 <0.1568 24Mg (KED) 1.835 4.091 1.160 1.013 2.802 4.236 2.009 1.531 <0.0231 27Al (KED) 0.420 3.688 0.479 2.325 1.029 3.615 0.867 7.665 <0.32 39K (KED) 8.246 11.185 5.921 5.295 13.580 11.235 16.088 6.645 <1.7964 48Ti (KED) 0.033 0.930 0.033 0.605 0.045 0.202 ND 0.107 <0.0314 51V (KED) 0.526 0.518 ND ND ND ND 0.051 0.044 <0.0339 52Cr (KED) ND 0.146 ND 0.103 ND 0.241 ND 0.108 <0.0072

55Mn (KED) 0.044 0.161 0.053 0.052 0.045 0.188 0.067 0.102 <0.0065 56Fe (KED) ND 9.962 ND 3.893 0.570 18.668 0.903 9.600 <0.0175 59Co (KED) 0.001 0.003 0.001 0.001 0.008 0.010 0.011 0.009 <0.0009 60Ni (KED) 0.097 0.123 0.047 0.075 0.254 0.420 0.715 0.281 <0.0063 65Cu (KED) 0.558 0.741 0.237 0.341 0.849 1.576 1.603 1.709 <0.007 66Zn (KED) 13.137 11.712 31.153 43.436 4.049 13.275 81.914 51.445 <0.5173 78Se (KED) ND ND ND ND ND ND ND ND <0.0781 88Sr (KED) 0.045 0.070 0.037 0.032 0.082 0.087 0.076 0.078 <0.0036

95Mo (KED) ND ND ND ND ND ND ND ND <0.4285 101Ru (KED) ND ND ND ND ND ND ND ND <0.0006 103Rh (KED) ND ND ND ND ND ND ND ND <0.0648 105Pd (KED) ND ND ND ND ND ND ND ND <0.0038 107Ag (KED) ND ND ND ND ND ND ND ND <0.0226 121Sb (KED) 0.007 0.017 0.007 0.020 0.017 0.124 0.040 0.052 <0.0012 137Ba (KED) 0.259 0.231 0.165 0.270 1.646 2.420 1.250 1.125 <0.0124 193Ir (KED) ND ND ND ND ND ND ND ND <0.0003 195Pt (KED) ND ND ND ND ND ND ND ND <0.0004 205Tl (KED) ND ND ND ND ND ND ND ND <0.0046

LCMS Result The all-in-one full scan/top3 ms/ms with polarity switching data acquisition using both ESI, APCI ionization ensures the detection of structurally diversified compounds. It provides comprehensive extractable profiles of the rubber stoppers, see Figures 2 and 3.

SR_Reflux_IPA_A_ESI_1 03/30/14 20:44:49Hypersil GOLD C18 150x2.1 1.9 um A: H2O B: MeOH C: 50 mMAmmonium AcetateRT: 0.00 - 45.01

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44Time (min)

0

10000

20000

30000

uAU

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

28.1526.72

30.2022.47

36.0729.0326.46 44.0625.32 31.430.90 42.3933.6721.851.85 17.78 36.89 40.8712.092.96 4.39 16.955.76 9.02 15.289.9730.20

22.47

36.0729.0326.6826.46

44.0631.43 42.3933.6721.851.02 20.5117.78 36.89 38.8812.09 39.9515.689.02 13.221.84 9.876.805.714.5328.15

26.72

29.410.90 22.43 24.941.85 21.58 32.45 33.66 36.03 38.04 40.87 43.5119.952.96 18.294.39 16.605.76 14.967.27 8.49 13.249.9732.42

36.0828.11 33.1230.2327.42 43.5636.99 42.4139.3525.1827.23

24.9023.70

20.5019.461.12 17.4316.29

NL: 6.29E9Base Peak MS SR_Reflux_IPA_A_ESI_1

NL: 2.65E9Base Peak F: FTMS + p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 6.29E9Base Peak F: FTMS - p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 3.45E4Total Scan PDA SR_Reflux_IPA_A_ESI_1

ESI (+/-) TIC

ESI (+)

ESI (-)

PDA

A search was also conducted against Thermo ScientificTM mzCloudTM Library, a newly developed high resolution spectral database. mzCloud library provides several search criteria for small molecule structure identification using tandem mass spectra, including spectra, fragments, precursor ions, etc, all of which can be very useful for unknown structure elucidation. Figure 5 shows identification of Irganox 1010 using the ms/ms spectrum search feature. The accuracy of searching result is indicated by matching score between the query and library spectra. GCMS Instrument Conditions GCMS identified lower molecular weight and

volatile extractables which complement LCMS results. GCMS results showed that DI water extractions using both techniques were “clean” and more extractables were detected from IPA extractions. Within the four rubber stoppers, sample C&D had more low molecular weight extractables detected by GCMS from both extraction techniques, see Figure 6 and 7.

Ace

toph

enon

e

Dim

ethy

lphe

nylm

etha

nol

L-Li

mon

ene

Dim

ethy

lben

zyl m

ethy

l eth

er

Dip

heny

l eth

er

Triis

opro

pano

lam

ine

+Lon

gipi

nene

B

HT

Non

ylbe

nzen

e de

rivat

ive

Patc

houl

ene

1,5,

9-Tr

imet

hyl c

yclo

dode

catri

ene

C20

Uns

atur

ated

hyd

roca

rbon

Patc

houl

ene

isom

er

4/20/2014 3:47:32 AM IPA_REFLUX_D RT: 4.66 - 27.35

6 8 10 12 14 16 18 20 22 24 26 Time (min)

0 1000000000

2000000000

3000000000

4000000000

5000000000

6000000000

7000000000

8000000000

9000000000

10000000000

11000000000

12000000000

13000000000

14000000000

Rel

ative

Abu

ndan

ce

6.62 11.94 6.43

9.98 12.01 10.61 13.11 10.95 14.40 15.26 15.39 6.68 18.15 14.21 19.02 20.75 7.16 22.07 23.32 9.87 6.00 24.92 27.14

50 63 75 88 100 113 125 138 150 163 175 188 200 213 225 238 250 263 275 288 300 313 325 338 350 363 375 388 400 413 425 438 450

0

10

20

30

40

50

60

70

80

90

100155.070489.0598

163.1330

145.1225111.0442

101.0963

57.0702173.0810

217.1075

317.1964261.134067.0544 127.0753

255.1593

O

OO

OO

O

OO

OO

OOH

O

OO

O

OOO

OHO

OO

OOH

OO

O

O

O 2H

OOO

O

O 2HOO

O

O

O2HO OH

OO

OO

O OHO

OH

OH

180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 m/z

0 5

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95

100

Rel

ative

Abu

ndan

ce

452.3227 C 22 H 46 O 8 N 1.9203 ppm

435.2955 C 22 H 43 O 8

0.6606 ppm

(M+H)+

(M+NH4)+

OO

OO

OO

O

O

Bis[2-(2-butoxyethoxy0ethyl] adipate Molecular Formula : C22H42O8 Formula Weight : 434.2880

FIGURE 2. LCMS Chromatogram of IPA Reflux of Sample A

Figure 7. GCMS Spectra of Compounds Identified in IPA_REFLUX of Sample-D (Partial)

Component Detection and Structure Elucidation The High Resolution Accurate Mass (HRAM) data were processed using differential analysis software SIEVETM 2.1 for component extraction. ChemSpider database searching was carried out to obtain possible structures of extracted components. While many possible hits were obtained for each component, to determine the correct structures, “Thermo ScientificTM Mass FrontierTM Software”, a small molecule structure analysis software, was used. The “HighChem Fragmentation LibraryTM ” in Mass Frontier 7.0 has extensive published literature references. For each proposed structure, the “Fragments and Mechanisms” feature in Mass Frontier was used to generate predicted “fragments and mechanisms” through HighChem Library search, see figure 4-b. A high degree of correlation between predicted and experimental fragments (indicated in red, see figure 4-c) confirms the proposed structure. Mass Frontier then automatically annotates the matching fragments based on library search results, see figure 4-c.

m/z 452.3218

OO

OO

OO

O

O

NH4

OO

OO

OO

OH

Om/z 435.2952

+H

OO

O

OO

O

m/z 304.1880

Lib

OHO

O

O

O

OO

m/z 334.1986

Lib

OO

OHO

OO

O

Om/z 435.2952

+H

Lib

O 2HO

O

m/z 163.1329

rHB

rHB

Lib

rHB

m/z 452.3218

OO

OO

OO

O

O

NH4

OHO

Om/z 162.1250

Lib

OHO

Om/z 162.1250

Lib

OOH

m/z 89.0597

i

Lib

Summary of Reflux and Speed Extractions using IPA and DI Water

4-a. HRAM Full MS to Determine Elemental Composition

4-c. HRAM MS/MS Spectra For Structure Elucidation

4-b. HighChem Fragmentation Library Search to Predict Fragments and Mechanism

RT: 6.41

50 100 150 0

50

100 105.01 77.10

120.04 51.12

77.99 74.13 91.16

134.07 147

O

Acetophenone MW 120.04

RT: 6.58

50 100 150 0

50

100 43.01

121.08

77.11

45.13 51.02

78.14 105.05

91.15 59.15

136.04 141.04

Dimethylphenylmethanol M+ 136.09

OH

RT: 10.93

50 100 150 0

50

100 146.12

70.02

88.08 42.03 98.18 59.14

102.18

128.18 158.09 176.08

Triisopropanolamine M+ 191.15

OHN

OH

OH

RT: 11.90

50 100 150 200 0

50

100 57.01

205.14

81.15

145.16 105.14 67.12 220.23 91.04 177.13 41.06

121.15 131.16

149.17 189.18

BHT M+ 220.18

OH

1,5,9 Trimethyl Cyclododecatriene MW. 204.35 Da

RT: 12.46

50 100 150 200 0

50

100 68.15

67.00

93.06

107.16

79.13 121.18 53.12

41.15 189.14

147.19 161.16 204.19

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Time (min)

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Rel

ative

Abu

ndan

ce

* * * * * * *

*

P3

PEG P9

P13

P19 P20

P21

P18 B B

B

Erucamide

Irganox 1010

Phthalate

P10

P22 Irganox 1076

DSDMA

* PPG B-Background P- Extractable

P7 P14

P15 P16 P17 B

P8(-)

P6(-) P5 P4

P2

P1

P11

B

P12

Peak ID RT Mode Measured (M+H) +

Calculated (M+H) +

Elemental Composition

Error (ppm)

1 21.48 APCI (+) 235.1691 235.1693 C15H22O2 -0.67 2 22.43 ESI (-) 199.1707 (M-H) – 199.1693 (M-H) – C12H24O2 2.0 3 22.47 ESI (+) 435.2956 (M+NH4) + 435.2952 C22H42O8 0.86 4 24.94 ESI (-) 227.2017 (M-H) – 227.2006 (M-H) – C14H28O2 2.23 5 24.87 ESI (+) 219.1743 219.1743 C15H22O 1.8 6 26.72 ESI (-) 255.2330 (M-H) – 255.2391 (M-H) – C16H32O2 3.5 7 27.75 ESI (+) 282.2791 282.2791 C18H36ON -0.11 8 28.15 ESI (-) 283.2643 (M-H) – 283.2632 (M-H) – C18H36O2 2.8 9 28.19 ESI (+) 383.3396 (M+NH4) + 383.3396 (M+NH4) + (C24H47ON)+NH4 0.42

10 28.71 ESI (+) 284.2946 284.2948 C18H37O1N1 -0.5 11 29.14 ESI (+) 325.3097 325.3101 C21H40O2 1.3 12 29.36 ESI (+) 319.2992 319.2995 C22H38O1 -1.1 13 29.41 ESI (-) 311.2963 (M-H) – 311.2945 (M-H) – C20H40O2 2.4 14 31.43 ESI (+) 340.3568 340.3574 C22H46ON

-1.8 15 31.77 ESI (+) 366.3729 366.3730 C24H48ON -0.5 16 32.52 ESI (+) 409.3100 409.3101 C28H40O2 -0.34 17 33.70 ESI (+) 1227.00 1227.00 ?

18 36.00 ESI (+) 1194.8170 (M+NH4) + 1194.8179 (M+NH4) + C73H108O12 -0.3

19 37.57 ESI (+) 663.4536 663.4537 C42H63O4P -0.19 20 38.85 ESI (+) 522.5969 522.5972 C36H76N -0.23 21 42.39 ESI (+) 548.5035 (M+NH4) + 548.5035 (M+NH4) + (C35H62O3)+NH4 -0.33 22 44.08 ESI (+) 550.6285 550.6285 C38H80N1 -0.2

TABLE 2. Proposed Structures of Identified Compounds (Partial List)

O

O

1 O

O

OO

OO

OO

O

O2 3

6 OH

OOH

O

8 NH

O

9

N 2H

O

10

13 OH

O

12 OH

11 O

O

O

O O

O

O

O

OO

OH

OH

OH

OH

O

OH

O N

18

19

21 22

Palmitic Acid

Irganox 1010

Irganox 1076

Stearic Acid

Stearamide

Oleamide N 2H

O

7

DSDMA Irgafos 168 oxidation product

GC

HPLC CE

SEC GFC

Polarity

Molecular Weight

Derivatization

Pyrolysis

HS-GC

OP

O

OO

Two fragments (highlighted) and corresponding mechanisms shown here to demonstrate the process.

HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.

Page 4: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

4 Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock1; Andrew Feilden2; Elisabeth Morgan2; Amalendu Sarkar3; Andrew White4

1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Conclusion This study demonstrated a comprehensive extractable analysis workflow utilizing multiple techniques: HR-LCMS, GCMS, ICPMS, data processing software, and database searching. This workflow followed recommended analytical methods by PQRI[2]. HS-GCMS was carried out but the data has not been reported.

The DI water and IPA extraction profiles of the four types of medical grade rubber stoppers were quickly established by using this workflow.

The UHPLC/HRAM full MS/HCD MS2 with rapid polarity switching in a single run data acquisition, coupled with novel database search, significantly increase the confidence and throughput of routine extractable & leachable analysis, in particular for unknown components identification and structure characterization.

The GCMS and LCMS analysis are complementary to each other and necessary to give complete coverage of extractables.

References 1. FDA CFR 21.94, CFR 66011(b) and 600.11(h), CFR 211.160 2. PQRI “L/E Recommendations to the FDA”

http://www.pqri.org/publications/index.asp

Acknowledgements The authors would like to thank Sukhy Toot for her contributions, Qure Medical for providing the rubber stoppers, Smithers Rapra for the extractions, and Buchi for the loan of the speed extractor.

Introduction Rubbers & plastics are widely used in medical & drug delivery devices and packaging materials. Extractables & leachables assessment of all materials, especially from elastomeric & oligomeric components, forms an integral part of the submission for approval of a new drug product or medical device [1]. Extractable = possible impact. Test the material

Leachable = actual impact. Test the product The mass spectrometer plays an important role in E&L identification and structure elucidation, as it is coupled with many techniques for definitive analysis, see figure 1. Here we present a comprehensive workflow for medical grade rubber stopper extractable analysis using multiple techniques including HR-LCMS, GCMS, and ICPMS, followed by data processes using novel software and database searching. FIGURE 1. Potential analytical techniques with increasing chance of extractables and leachables being found as the molecular weight decreases.

LCMS Analyses Sample Preparation Four different types of medical grade rubber stoppers, sample-A, sample-B, sample-C, and sample-D, from Qure Medical, were extracted using DI water and IPA utilizing reflux extraction and the Buchi Speed Extractor. The extracts solutions were analyzed directly by LCMS.

Liquid Chromatography LC separations were carried out on the Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system consisting of: DGP-3000RS pump, WPS-3000RS sampler, TCC-3000RS column compartment, and DAD-3000RS UV detector Column: Thermo Scientific™ Hypersil C18, 2.1x150 mm 1.9 µm Column Temp: 50ºC LC Mobile phase: A: H2O B: MeOH C: 50 mM Ammonium Acetate

Mass Spectrometry MS analyses were carried out on the Thermo Scientific™ Q Exactive™ mass spectrometer using both electrospray ionization (ESI) and atmosphere pressure chemical ionization (APCI). High resolution full scan MS and top 3 MS/MS data were collected in a data-dependent fashion at a resolving power of 70,000 and 17,500 (FWHM m/z 200) with polarity switching. The scan range is m/z150-1500. Stepped NCE (Normalized Collision Energy) setting was: 30, 40, 50.

FIGURE 4. HR Full Scan and HCD MS2 for Component ID and Structure Elucidation

LCMS and GCMS results show DI water extracts using both Speedy and Reflux techniques were “clean”. “Triisopropanolamine” is the major extractable.

Complex profiles of IPA extractions were observed from both LCMS and GCMS analysis. Complete extractable list not shown.

IPA reflux shows higher extraction efficiency compared with IPA speed extraction. However, the speed extractor conditions were not optimized for this study

Results show that GCMS and LCMS analysis are complementary to each other and necessary to give a fuller picture of the extractable profile.

FIGURE 5. mzCloud library Search Results for Irganox 1010

FIGURE 6. GCMS Chromatogram of IPA_REFLUX of Sample-D

FIGURE 7. ICPMS Results for the Four Rubber Stoppers (ppb)

GCMS Analyses Method and Instrumentation The DI water samples were extracted with Hexane. The samples in 2.0 mL GC vials were introduced in split injection mode into the Thermo Scientific™ TRACE™ Ultra Gas Chromatograph using a Thermo Scientific™ TriPlus RSH™ Autosampler. TG-5ms (30 m x0.25mm x0.25µm) column was used. Compounds were detected and identified with the ISQ single Quad mass spectrometer.

mzCloud Spectral Database Searching

TABLE 1. Components Identified from IPA Reflux of Sample-A (Partial List)

FIGURE 3. MS Chromatogram of IPA Reflux of Sample-A (ESI+) ICPMS Analyses The ICPMS samples were prepared by placing the rubber stoppers in 25 ml DI water and 25 ml 2% nitric acid and soaked at RT for 24 hours. The analyses were conducted on Thermo Scientific™ iCAP™ Q ICP-MS with He KED (Kinetic Energy Discrimination) interference reduction mode setting. To determine if trace and potentially toxic metals were leached from the stoppers, the USP<232> Class1 & 2 elements and additional elements which are commonly analyzed by ICP-MS were determined. The analysis results for the four types of rubber stoppers showed that they are clean of all Class 1 & 2 elements, see Figure 7 for the ICPMS results. In addition, the system control software Qtegra provides a full 21CFR Part 11 tool set to operate under compliant environments.

Element Sample-1 DI water

Sample-1 Nitric Acid

Sample-2 DI water

Sample-2 Nitric Acid

Sample-3 DI water

Sample-3 Nitric Acid

Sample-4 DI water

Sample-4 Nitric Acid LOD

75As (KED) ND ND ND ND ND ND ND ND <0.0233 111Cd (KED) 0.009 0.006 0.003 0.007 0.010 0.009 0.276 0.070 <0.0023 202Hg (KED) ND ND ND ND ND ND ND ND <0.0054 208Pb (KED) 0.061 0.069 0.018 0.124 0.100 0.106 0.159 0.122 <0.0008

9Be (KED) ND ND ND ND ND ND ND ND <0.0362 11B (KED) ND ND 0.541 ND 0.853 0.596 ND ND <0.5229

23Na (KED) 14.326 29.535 8.197 13.575 25.630 20.648 30.074 18.071 <0.1568 24Mg (KED) 1.835 4.091 1.160 1.013 2.802 4.236 2.009 1.531 <0.0231 27Al (KED) 0.420 3.688 0.479 2.325 1.029 3.615 0.867 7.665 <0.32 39K (KED) 8.246 11.185 5.921 5.295 13.580 11.235 16.088 6.645 <1.7964 48Ti (KED) 0.033 0.930 0.033 0.605 0.045 0.202 ND 0.107 <0.0314 51V (KED) 0.526 0.518 ND ND ND ND 0.051 0.044 <0.0339 52Cr (KED) ND 0.146 ND 0.103 ND 0.241 ND 0.108 <0.0072

55Mn (KED) 0.044 0.161 0.053 0.052 0.045 0.188 0.067 0.102 <0.0065 56Fe (KED) ND 9.962 ND 3.893 0.570 18.668 0.903 9.600 <0.0175 59Co (KED) 0.001 0.003 0.001 0.001 0.008 0.010 0.011 0.009 <0.0009 60Ni (KED) 0.097 0.123 0.047 0.075 0.254 0.420 0.715 0.281 <0.0063 65Cu (KED) 0.558 0.741 0.237 0.341 0.849 1.576 1.603 1.709 <0.007 66Zn (KED) 13.137 11.712 31.153 43.436 4.049 13.275 81.914 51.445 <0.5173 78Se (KED) ND ND ND ND ND ND ND ND <0.0781 88Sr (KED) 0.045 0.070 0.037 0.032 0.082 0.087 0.076 0.078 <0.0036

95Mo (KED) ND ND ND ND ND ND ND ND <0.4285 101Ru (KED) ND ND ND ND ND ND ND ND <0.0006 103Rh (KED) ND ND ND ND ND ND ND ND <0.0648 105Pd (KED) ND ND ND ND ND ND ND ND <0.0038 107Ag (KED) ND ND ND ND ND ND ND ND <0.0226 121Sb (KED) 0.007 0.017 0.007 0.020 0.017 0.124 0.040 0.052 <0.0012 137Ba (KED) 0.259 0.231 0.165 0.270 1.646 2.420 1.250 1.125 <0.0124 193Ir (KED) ND ND ND ND ND ND ND ND <0.0003 195Pt (KED) ND ND ND ND ND ND ND ND <0.0004 205Tl (KED) ND ND ND ND ND ND ND ND <0.0046

LCMS Result The all-in-one full scan/top3 ms/ms with polarity switching data acquisition using both ESI, APCI ionization ensures the detection of structurally diversified compounds. It provides comprehensive extractable profiles of the rubber stoppers, see Figures 2 and 3.

SR_Reflux_IPA_A_ESI_1 03/30/14 20:44:49Hypersil GOLD C18 150x2.1 1.9 um A: H2O B: MeOH C: 50 mMAmmonium AcetateRT: 0.00 - 45.01

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44Time (min)

0

10000

20000

30000

uAU

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

28.1526.72

30.2022.47

36.0729.0326.46 44.0625.32 31.430.90 42.3933.6721.851.85 17.78 36.89 40.8712.092.96 4.39 16.955.76 9.02 15.289.9730.20

22.47

36.0729.0326.6826.46

44.0631.43 42.3933.6721.851.02 20.5117.78 36.89 38.8812.09 39.9515.689.02 13.221.84 9.876.805.714.5328.15

26.72

29.410.90 22.43 24.941.85 21.58 32.45 33.66 36.03 38.04 40.87 43.5119.952.96 18.294.39 16.605.76 14.967.27 8.49 13.249.9732.42

36.0828.11 33.1230.2327.42 43.5636.99 42.4139.3525.1827.23

24.9023.70

20.5019.461.12 17.4316.29

NL: 6.29E9Base Peak MS SR_Reflux_IPA_A_ESI_1

NL: 2.65E9Base Peak F: FTMS + p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 6.29E9Base Peak F: FTMS - p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 3.45E4Total Scan PDA SR_Reflux_IPA_A_ESI_1

ESI (+/-) TIC

ESI (+)

ESI (-)

PDA

A search was also conducted against Thermo ScientificTM mzCloudTM Library, a newly developed high resolution spectral database. mzCloud library provides several search criteria for small molecule structure identification using tandem mass spectra, including spectra, fragments, precursor ions, etc, all of which can be very useful for unknown structure elucidation. Figure 5 shows identification of Irganox 1010 using the ms/ms spectrum search feature. The accuracy of searching result is indicated by matching score between the query and library spectra. GCMS Instrument Conditions GCMS identified lower molecular weight and

volatile extractables which complement LCMS results. GCMS results showed that DI water extractions using both techniques were “clean” and more extractables were detected from IPA extractions. Within the four rubber stoppers, sample C&D had more low molecular weight extractables detected by GCMS from both extraction techniques, see Figure 6 and 7.

Ace

toph

enon

e

Dim

ethy

lphe

nylm

etha

nol

L-Li

mon

ene

Dim

ethy

lben

zyl m

ethy

l eth

er

Dip

heny

l eth

er

Triis

opro

pano

lam

ine

+Lon

gipi

nene

B

HT

Non

ylbe

nzen

e de

rivat

ive

Patc

houl

ene

1,5,

9-Tr

imet

hyl c

yclo

dode

catri

ene

C20

Uns

atur

ated

hyd

roca

rbon

Patc

houl

ene

isom

er

4/20/2014 3:47:32 AM IPA_REFLUX_D RT: 4.66 - 27.35

6 8 10 12 14 16 18 20 22 24 26 Time (min)

0 1000000000

2000000000

3000000000

4000000000

5000000000

6000000000

7000000000

8000000000

9000000000

10000000000

11000000000

12000000000

13000000000

14000000000

Rel

ative

Abu

ndan

ce

6.62 11.94 6.43

9.98 12.01 10.61 13.11 10.95 14.40 15.26 15.39 6.68 18.15 14.21 19.02 20.75 7.16 22.07 23.32 9.87 6.00 24.92 27.14

50 63 75 88 100 113 125 138 150 163 175 188 200 213 225 238 250 263 275 288 300 313 325 338 350 363 375 388 400 413 425 438 450

0

10

20

30

40

50

60

70

80

90

100155.070489.0598

163.1330

145.1225111.0442

101.0963

57.0702173.0810

217.1075

317.1964261.134067.0544 127.0753

255.1593

O

OO

OO

O

OO

OO

OOH

O

OO

O

OOO

OHO

OO

OOH

OO

O

O

O 2H

OOO

O

O 2HOO

O

O

O2HO OH

OO

OO

O OHO

OH

OH

180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 m/z

0 5

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95

100

Rel

ative

Abu

ndan

ce

452.3227 C 22 H 46 O 8 N 1.9203 ppm

435.2955 C 22 H 43 O 8

0.6606 ppm

(M+H)+

(M+NH4)+

OO

OO

OO

O

O

Bis[2-(2-butoxyethoxy0ethyl] adipate Molecular Formula : C22H42O8 Formula Weight : 434.2880

FIGURE 2. LCMS Chromatogram of IPA Reflux of Sample A

Figure 7. GCMS Spectra of Compounds Identified in IPA_REFLUX of Sample-D (Partial)

Component Detection and Structure Elucidation The High Resolution Accurate Mass (HRAM) data were processed using differential analysis software SIEVETM 2.1 for component extraction. ChemSpider database searching was carried out to obtain possible structures of extracted components. While many possible hits were obtained for each component, to determine the correct structures, “Thermo ScientificTM Mass FrontierTM Software”, a small molecule structure analysis software, was used. The “HighChem Fragmentation LibraryTM ” in Mass Frontier 7.0 has extensive published literature references. For each proposed structure, the “Fragments and Mechanisms” feature in Mass Frontier was used to generate predicted “fragments and mechanisms” through HighChem Library search, see figure 4-b. A high degree of correlation between predicted and experimental fragments (indicated in red, see figure 4-c) confirms the proposed structure. Mass Frontier then automatically annotates the matching fragments based on library search results, see figure 4-c.

m/z 452.3218

OO

OO

OO

O

O

NH4

OO

OO

OO

OH

Om/z 435.2952

+H

OO

O

OO

O

m/z 304.1880

Lib

OHO

O

O

O

OO

m/z 334.1986

Lib

OO

OHO

OO

O

Om/z 435.2952

+H

Lib

O 2HO

O

m/z 163.1329

rHB

rHB

Lib

rHB

m/z 452.3218

OO

OO

OO

O

O

NH4

OHO

Om/z 162.1250

Lib

OHO

Om/z 162.1250

Lib

OOH

m/z 89.0597

i

Lib

Summary of Reflux and Speed Extractions using IPA and DI Water

4-a. HRAM Full MS to Determine Elemental Composition

4-c. HRAM MS/MS Spectra For Structure Elucidation

4-b. HighChem Fragmentation Library Search to Predict Fragments and Mechanism

RT: 6.41

50 100 150 0

50

100 105.01 77.10

120.04 51.12

77.99 74.13 91.16

134.07 147

O

Acetophenone MW 120.04

RT: 6.58

50 100 150 0

50

100 43.01

121.08

77.11

45.13 51.02

78.14 105.05

91.15 59.15

136.04 141.04

Dimethylphenylmethanol M+ 136.09

OH

RT: 10.93

50 100 150 0

50

100 146.12

70.02

88.08 42.03 98.18 59.14

102.18

128.18 158.09 176.08

Triisopropanolamine M+ 191.15

OHN

OH

OH

RT: 11.90

50 100 150 200 0

50

100 57.01

205.14

81.15

145.16 105.14 67.12 220.23 91.04 177.13 41.06

121.15 131.16

149.17 189.18

BHT M+ 220.18

OH

1,5,9 Trimethyl Cyclododecatriene MW. 204.35 Da

RT: 12.46

50 100 150 200 0

50

100 68.15

67.00

93.06

107.16

79.13 121.18 53.12

41.15 189.14

147.19 161.16 204.19

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Time (min)

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Rel

ative

Abu

ndan

ce

* * * * * * *

*

P3

PEG P9

P13

P19 P20

P21

P18 B B

B

Erucamide

Irganox 1010

Phthalate

P10

P22 Irganox 1076

DSDMA

* PPG B-Background P- Extractable

P7 P14

P15 P16 P17 B

P8(-)

P6(-) P5 P4

P2

P1

P11

B

P12

Peak ID RT Mode Measured (M+H) +

Calculated (M+H) +

Elemental Composition

Error (ppm)

1 21.48 APCI (+) 235.1691 235.1693 C15H22O2 -0.67 2 22.43 ESI (-) 199.1707 (M-H) – 199.1693 (M-H) – C12H24O2 2.0 3 22.47 ESI (+) 435.2956 (M+NH4) + 435.2952 C22H42O8 0.86 4 24.94 ESI (-) 227.2017 (M-H) – 227.2006 (M-H) – C14H28O2 2.23 5 24.87 ESI (+) 219.1743 219.1743 C15H22O 1.8 6 26.72 ESI (-) 255.2330 (M-H) – 255.2391 (M-H) – C16H32O2 3.5 7 27.75 ESI (+) 282.2791 282.2791 C18H36ON -0.11 8 28.15 ESI (-) 283.2643 (M-H) – 283.2632 (M-H) – C18H36O2 2.8 9 28.19 ESI (+) 383.3396 (M+NH4) + 383.3396 (M+NH4) + (C24H47ON)+NH4 0.42

10 28.71 ESI (+) 284.2946 284.2948 C18H37O1N1 -0.5 11 29.14 ESI (+) 325.3097 325.3101 C21H40O2 1.3 12 29.36 ESI (+) 319.2992 319.2995 C22H38O1 -1.1 13 29.41 ESI (-) 311.2963 (M-H) – 311.2945 (M-H) – C20H40O2 2.4 14 31.43 ESI (+) 340.3568 340.3574 C22H46ON

-1.8 15 31.77 ESI (+) 366.3729 366.3730 C24H48ON -0.5 16 32.52 ESI (+) 409.3100 409.3101 C28H40O2 -0.34 17 33.70 ESI (+) 1227.00 1227.00 ?

18 36.00 ESI (+) 1194.8170 (M+NH4) + 1194.8179 (M+NH4) + C73H108O12 -0.3

19 37.57 ESI (+) 663.4536 663.4537 C42H63O4P -0.19 20 38.85 ESI (+) 522.5969 522.5972 C36H76N -0.23 21 42.39 ESI (+) 548.5035 (M+NH4) + 548.5035 (M+NH4) + (C35H62O3)+NH4 -0.33 22 44.08 ESI (+) 550.6285 550.6285 C38H80N1 -0.2

TABLE 2. Proposed Structures of Identified Compounds (Partial List)

O

O

1 O

O

OO

OO

OO

O

O2 3

6 OH

OOH

O

8 NH

O

9

N 2H

O

10

13 OH

O

12 OH

11 O

O

O

O O

O

O

O

OO

OH

OH

OH

OH

O

OH

O N

18

19

21 22

Palmitic Acid

Irganox 1010

Irganox 1076

Stearic Acid

Stearamide

Oleamide N 2H

O

7

DSDMA Irgafos 168 oxidation product

GC

HPLC CE

SEC GFC

Polarity

Molecular Weight

Derivatization

Pyrolysis

HS-GC

OP

O

OO

Two fragments (highlighted) and corresponding mechanisms shown here to demonstrate the process.

HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.

Page 5: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

5Thermo Scientific Poster Note • PN-64219-IMSC 0814S

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock1; Andrew Feilden2; Elisabeth Morgan2; Amalendu Sarkar3; Andrew White4

1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Conclusion This study demonstrated a comprehensive extractable analysis workflow utilizing multiple techniques: HR-LCMS, GCMS, ICPMS, data processing software, and database searching. This workflow followed recommended analytical methods by PQRI[2]. HS-GCMS was carried out but the data has not been reported.

The DI water and IPA extraction profiles of the four types of medical grade rubber stoppers were quickly established by using this workflow.

The UHPLC/HRAM full MS/HCD MS2 with rapid polarity switching in a single run data acquisition, coupled with novel database search, significantly increase the confidence and throughput of routine extractable & leachable analysis, in particular for unknown components identification and structure characterization.

The GCMS and LCMS analysis are complementary to each other and necessary to give complete coverage of extractables.

References 1. FDA CFR 21.94, CFR 66011(b) and 600.11(h), CFR 211.160 2. PQRI “L/E Recommendations to the FDA”

http://www.pqri.org/publications/index.asp

Acknowledgements The authors would like to thank Sukhy Toot for her contributions, Qure Medical for providing the rubber stoppers, Smithers Rapra for the extractions, and Buchi for the loan of the speed extractor.

Introduction Rubbers & plastics are widely used in medical & drug delivery devices and packaging materials. Extractables & leachables assessment of all materials, especially from elastomeric & oligomeric components, forms an integral part of the submission for approval of a new drug product or medical device [1]. Extractable = possible impact. Test the material

Leachable = actual impact. Test the product The mass spectrometer plays an important role in E&L identification and structure elucidation, as it is coupled with many techniques for definitive analysis, see figure 1. Here we present a comprehensive workflow for medical grade rubber stopper extractable analysis using multiple techniques including HR-LCMS, GCMS, and ICPMS, followed by data processes using novel software and database searching. FIGURE 1. Potential analytical techniques with increasing chance of extractables and leachables being found as the molecular weight decreases.

LCMS Analyses Sample Preparation Four different types of medical grade rubber stoppers, sample-A, sample-B, sample-C, and sample-D, from Qure Medical, were extracted using DI water and IPA utilizing reflux extraction and the Buchi Speed Extractor. The extracts solutions were analyzed directly by LCMS.

Liquid Chromatography LC separations were carried out on the Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system consisting of: DGP-3000RS pump, WPS-3000RS sampler, TCC-3000RS column compartment, and DAD-3000RS UV detector Column: Thermo Scientific™ Hypersil C18, 2.1x150 mm 1.9 µm Column Temp: 50ºC LC Mobile phase: A: H2O B: MeOH C: 50 mM Ammonium Acetate

Mass Spectrometry MS analyses were carried out on the Thermo Scientific™ Q Exactive™ mass spectrometer using both electrospray ionization (ESI) and atmosphere pressure chemical ionization (APCI). High resolution full scan MS and top 3 MS/MS data were collected in a data-dependent fashion at a resolving power of 70,000 and 17,500 (FWHM m/z 200) with polarity switching. The scan range is m/z150-1500. Stepped NCE (Normalized Collision Energy) setting was: 30, 40, 50.

FIGURE 4. HR Full Scan and HCD MS2 for Component ID and Structure Elucidation

LCMS and GCMS results show DI water extracts using both Speedy and Reflux techniques were “clean”. “Triisopropanolamine” is the major extractable.

Complex profiles of IPA extractions were observed from both LCMS and GCMS analysis. Complete extractable list not shown.

IPA reflux shows higher extraction efficiency compared with IPA speed extraction. However, the speed extractor conditions were not optimized for this study

Results show that GCMS and LCMS analysis are complementary to each other and necessary to give a fuller picture of the extractable profile.

FIGURE 5. mzCloud library Search Results for Irganox 1010

FIGURE 6. GCMS Chromatogram of IPA_REFLUX of Sample-D

FIGURE 7. ICPMS Results for the Four Rubber Stoppers (ppb)

GCMS Analyses Method and Instrumentation The DI water samples were extracted with Hexane. The samples in 2.0 mL GC vials were introduced in split injection mode into the Thermo Scientific™ TRACE™ Ultra Gas Chromatograph using a Thermo Scientific™ TriPlus RSH™ Autosampler. TG-5ms (30 m x0.25mm x0.25µm) column was used. Compounds were detected and identified with the ISQ single Quad mass spectrometer.

mzCloud Spectral Database Searching

TABLE 1. Components Identified from IPA Reflux of Sample-A (Partial List)

FIGURE 3. MS Chromatogram of IPA Reflux of Sample-A (ESI+) ICPMS Analyses The ICPMS samples were prepared by placing the rubber stoppers in 25 ml DI water and 25 ml 2% nitric acid and soaked at RT for 24 hours. The analyses were conducted on Thermo Scientific™ iCAP™ Q ICP-MS with He KED (Kinetic Energy Discrimination) interference reduction mode setting. To determine if trace and potentially toxic metals were leached from the stoppers, the USP<232> Class1 & 2 elements and additional elements which are commonly analyzed by ICP-MS were determined. The analysis results for the four types of rubber stoppers showed that they are clean of all Class 1 & 2 elements, see Figure 7 for the ICPMS results. In addition, the system control software Qtegra provides a full 21CFR Part 11 tool set to operate under compliant environments.

Element Sample-1 DI water

Sample-1 Nitric Acid

Sample-2 DI water

Sample-2 Nitric Acid

Sample-3 DI water

Sample-3 Nitric Acid

Sample-4 DI water

Sample-4 Nitric Acid LOD

75As (KED) ND ND ND ND ND ND ND ND <0.0233 111Cd (KED) 0.009 0.006 0.003 0.007 0.010 0.009 0.276 0.070 <0.0023 202Hg (KED) ND ND ND ND ND ND ND ND <0.0054 208Pb (KED) 0.061 0.069 0.018 0.124 0.100 0.106 0.159 0.122 <0.0008

9Be (KED) ND ND ND ND ND ND ND ND <0.0362 11B (KED) ND ND 0.541 ND 0.853 0.596 ND ND <0.5229

23Na (KED) 14.326 29.535 8.197 13.575 25.630 20.648 30.074 18.071 <0.1568 24Mg (KED) 1.835 4.091 1.160 1.013 2.802 4.236 2.009 1.531 <0.0231 27Al (KED) 0.420 3.688 0.479 2.325 1.029 3.615 0.867 7.665 <0.32 39K (KED) 8.246 11.185 5.921 5.295 13.580 11.235 16.088 6.645 <1.7964 48Ti (KED) 0.033 0.930 0.033 0.605 0.045 0.202 ND 0.107 <0.0314 51V (KED) 0.526 0.518 ND ND ND ND 0.051 0.044 <0.0339 52Cr (KED) ND 0.146 ND 0.103 ND 0.241 ND 0.108 <0.0072

55Mn (KED) 0.044 0.161 0.053 0.052 0.045 0.188 0.067 0.102 <0.0065 56Fe (KED) ND 9.962 ND 3.893 0.570 18.668 0.903 9.600 <0.0175 59Co (KED) 0.001 0.003 0.001 0.001 0.008 0.010 0.011 0.009 <0.0009 60Ni (KED) 0.097 0.123 0.047 0.075 0.254 0.420 0.715 0.281 <0.0063 65Cu (KED) 0.558 0.741 0.237 0.341 0.849 1.576 1.603 1.709 <0.007 66Zn (KED) 13.137 11.712 31.153 43.436 4.049 13.275 81.914 51.445 <0.5173 78Se (KED) ND ND ND ND ND ND ND ND <0.0781 88Sr (KED) 0.045 0.070 0.037 0.032 0.082 0.087 0.076 0.078 <0.0036

95Mo (KED) ND ND ND ND ND ND ND ND <0.4285 101Ru (KED) ND ND ND ND ND ND ND ND <0.0006 103Rh (KED) ND ND ND ND ND ND ND ND <0.0648 105Pd (KED) ND ND ND ND ND ND ND ND <0.0038 107Ag (KED) ND ND ND ND ND ND ND ND <0.0226 121Sb (KED) 0.007 0.017 0.007 0.020 0.017 0.124 0.040 0.052 <0.0012 137Ba (KED) 0.259 0.231 0.165 0.270 1.646 2.420 1.250 1.125 <0.0124 193Ir (KED) ND ND ND ND ND ND ND ND <0.0003 195Pt (KED) ND ND ND ND ND ND ND ND <0.0004 205Tl (KED) ND ND ND ND ND ND ND ND <0.0046

LCMS Result The all-in-one full scan/top3 ms/ms with polarity switching data acquisition using both ESI, APCI ionization ensures the detection of structurally diversified compounds. It provides comprehensive extractable profiles of the rubber stoppers, see Figures 2 and 3.

SR_Reflux_IPA_A_ESI_1 03/30/14 20:44:49Hypersil GOLD C18 150x2.1 1.9 um A: H2O B: MeOH C: 50 mMAmmonium AcetateRT: 0.00 - 45.01

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44Time (min)

0

10000

20000

30000

uAU

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

28.1526.72

30.2022.47

36.0729.0326.46 44.0625.32 31.430.90 42.3933.6721.851.85 17.78 36.89 40.8712.092.96 4.39 16.955.76 9.02 15.289.9730.20

22.47

36.0729.0326.6826.46

44.0631.43 42.3933.6721.851.02 20.5117.78 36.89 38.8812.09 39.9515.689.02 13.221.84 9.876.805.714.5328.15

26.72

29.410.90 22.43 24.941.85 21.58 32.45 33.66 36.03 38.04 40.87 43.5119.952.96 18.294.39 16.605.76 14.967.27 8.49 13.249.9732.42

36.0828.11 33.1230.2327.42 43.5636.99 42.4139.3525.1827.23

24.9023.70

20.5019.461.12 17.4316.29

NL: 6.29E9Base Peak MS SR_Reflux_IPA_A_ESI_1

NL: 2.65E9Base Peak F: FTMS + p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 6.29E9Base Peak F: FTMS - p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 3.45E4Total Scan PDA SR_Reflux_IPA_A_ESI_1

ESI (+/-) TIC

ESI (+)

ESI (-)

PDA

A search was also conducted against Thermo ScientificTM mzCloudTM Library, a newly developed high resolution spectral database. mzCloud library provides several search criteria for small molecule structure identification using tandem mass spectra, including spectra, fragments, precursor ions, etc, all of which can be very useful for unknown structure elucidation. Figure 5 shows identification of Irganox 1010 using the ms/ms spectrum search feature. The accuracy of searching result is indicated by matching score between the query and library spectra. GCMS Instrument Conditions GCMS identified lower molecular weight and

volatile extractables which complement LCMS results. GCMS results showed that DI water extractions using both techniques were “clean” and more extractables were detected from IPA extractions. Within the four rubber stoppers, sample C&D had more low molecular weight extractables detected by GCMS from both extraction techniques, see Figure 6 and 7.

Ace

toph

enon

e

Dim

ethy

lphe

nylm

etha

nol

L-Li

mon

ene

Dim

ethy

lben

zyl m

ethy

l eth

er

Dip

heny

l eth

er

Triis

opro

pano

lam

ine

+Lon

gipi

nene

B

HT

Non

ylbe

nzen

e de

rivat

ive

Patc

houl

ene

1,5,

9-Tr

imet

hyl c

yclo

dode

catri

ene

C20

Uns

atur

ated

hyd

roca

rbon

Patc

houl

ene

isom

er

4/20/2014 3:47:32 AM IPA_REFLUX_D RT: 4.66 - 27.35

6 8 10 12 14 16 18 20 22 24 26 Time (min)

0 1000000000

2000000000

3000000000

4000000000

5000000000

6000000000

7000000000

8000000000

9000000000

10000000000

11000000000

12000000000

13000000000

14000000000

Rel

ative

Abu

ndan

ce

6.62 11.94 6.43

9.98 12.01 10.61 13.11 10.95 14.40 15.26 15.39 6.68 18.15 14.21 19.02 20.75 7.16 22.07 23.32 9.87 6.00 24.92 27.14

50 63 75 88 100 113 125 138 150 163 175 188 200 213 225 238 250 263 275 288 300 313 325 338 350 363 375 388 400 413 425 438 450

0

10

20

30

40

50

60

70

80

90

100155.070489.0598

163.1330

145.1225111.0442

101.0963

57.0702173.0810

217.1075

317.1964261.134067.0544 127.0753

255.1593

O

OO

OO

O

OO

OO

OOH

O

OO

O

OOO

OHO

OO

OOH

OO

O

O

O 2H

OOO

O

O 2HOO

O

O

O2HO OH

OO

OO

O OHO

OH

OH

180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 m/z

0 5

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95

100

Rel

ative

Abu

ndan

ce

452.3227 C 22 H 46 O 8 N 1.9203 ppm

435.2955 C 22 H 43 O 8

0.6606 ppm

(M+H)+

(M+NH4)+

OO

OO

OO

O

O

Bis[2-(2-butoxyethoxy0ethyl] adipate Molecular Formula : C22H42O8 Formula Weight : 434.2880

FIGURE 2. LCMS Chromatogram of IPA Reflux of Sample A

Figure 7. GCMS Spectra of Compounds Identified in IPA_REFLUX of Sample-D (Partial)

Component Detection and Structure Elucidation The High Resolution Accurate Mass (HRAM) data were processed using differential analysis software SIEVETM 2.1 for component extraction. ChemSpider database searching was carried out to obtain possible structures of extracted components. While many possible hits were obtained for each component, to determine the correct structures, “Thermo ScientificTM Mass FrontierTM Software”, a small molecule structure analysis software, was used. The “HighChem Fragmentation LibraryTM ” in Mass Frontier 7.0 has extensive published literature references. For each proposed structure, the “Fragments and Mechanisms” feature in Mass Frontier was used to generate predicted “fragments and mechanisms” through HighChem Library search, see figure 4-b. A high degree of correlation between predicted and experimental fragments (indicated in red, see figure 4-c) confirms the proposed structure. Mass Frontier then automatically annotates the matching fragments based on library search results, see figure 4-c.

m/z 452.3218

OO

OO

OO

O

O

NH4

OO

OO

OO

OH

Om/z 435.2952

+H

OO

O

OO

O

m/z 304.1880

Lib

OHO

O

O

O

OO

m/z 334.1986

Lib

OO

OHO

OO

O

Om/z 435.2952

+H

Lib

O 2HO

O

m/z 163.1329

rHB

rHB

Lib

rHB

m/z 452.3218

OO

OO

OO

O

O

NH4

OHO

Om/z 162.1250

Lib

OHO

Om/z 162.1250

Lib

OOH

m/z 89.0597

i

Lib

Summary of Reflux and Speed Extractions using IPA and DI Water

4-a. HRAM Full MS to Determine Elemental Composition

4-c. HRAM MS/MS Spectra For Structure Elucidation

4-b. HighChem Fragmentation Library Search to Predict Fragments and Mechanism

RT: 6.41

50 100 150 0

50

100 105.01 77.10

120.04 51.12

77.99 74.13 91.16

134.07 147

O

Acetophenone MW 120.04

RT: 6.58

50 100 150 0

50

100 43.01

121.08

77.11

45.13 51.02

78.14 105.05

91.15 59.15

136.04 141.04

Dimethylphenylmethanol M+ 136.09

OH

RT: 10.93

50 100 150 0

50

100 146.12

70.02

88.08 42.03 98.18 59.14

102.18

128.18 158.09 176.08

Triisopropanolamine M+ 191.15

OHN

OH

OH

RT: 11.90

50 100 150 200 0

50

100 57.01

205.14

81.15

145.16 105.14 67.12 220.23 91.04 177.13 41.06

121.15 131.16

149.17 189.18

BHT M+ 220.18

OH

1,5,9 Trimethyl Cyclododecatriene MW. 204.35 Da

RT: 12.46

50 100 150 200 0

50

100 68.15

67.00

93.06

107.16

79.13 121.18 53.12

41.15 189.14

147.19 161.16 204.19

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Time (min)

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Rel

ative

Abu

ndan

ce

* * * * * * *

*

P3

PEG P9

P13

P19 P20

P21

P18 B B

B

Erucamide

Irganox 1010

Phthalate

P10

P22 Irganox 1076

DSDMA

* PPG B-Background P- Extractable

P7 P14

P15 P16 P17 B

P8(-)

P6(-) P5 P4

P2

P1

P11

B

P12

Peak ID RT Mode Measured (M+H) +

Calculated (M+H) +

Elemental Composition

Error (ppm)

1 21.48 APCI (+) 235.1691 235.1693 C15H22O2 -0.67 2 22.43 ESI (-) 199.1707 (M-H) – 199.1693 (M-H) – C12H24O2 2.0 3 22.47 ESI (+) 435.2956 (M+NH4) + 435.2952 C22H42O8 0.86 4 24.94 ESI (-) 227.2017 (M-H) – 227.2006 (M-H) – C14H28O2 2.23 5 24.87 ESI (+) 219.1743 219.1743 C15H22O 1.8 6 26.72 ESI (-) 255.2330 (M-H) – 255.2391 (M-H) – C16H32O2 3.5 7 27.75 ESI (+) 282.2791 282.2791 C18H36ON -0.11 8 28.15 ESI (-) 283.2643 (M-H) – 283.2632 (M-H) – C18H36O2 2.8 9 28.19 ESI (+) 383.3396 (M+NH4) + 383.3396 (M+NH4) + (C24H47ON)+NH4 0.42

10 28.71 ESI (+) 284.2946 284.2948 C18H37O1N1 -0.5 11 29.14 ESI (+) 325.3097 325.3101 C21H40O2 1.3 12 29.36 ESI (+) 319.2992 319.2995 C22H38O1 -1.1 13 29.41 ESI (-) 311.2963 (M-H) – 311.2945 (M-H) – C20H40O2 2.4 14 31.43 ESI (+) 340.3568 340.3574 C22H46ON

-1.8 15 31.77 ESI (+) 366.3729 366.3730 C24H48ON -0.5 16 32.52 ESI (+) 409.3100 409.3101 C28H40O2 -0.34 17 33.70 ESI (+) 1227.00 1227.00 ?

18 36.00 ESI (+) 1194.8170 (M+NH4) + 1194.8179 (M+NH4) + C73H108O12 -0.3

19 37.57 ESI (+) 663.4536 663.4537 C42H63O4P -0.19 20 38.85 ESI (+) 522.5969 522.5972 C36H76N -0.23 21 42.39 ESI (+) 548.5035 (M+NH4) + 548.5035 (M+NH4) + (C35H62O3)+NH4 -0.33 22 44.08 ESI (+) 550.6285 550.6285 C38H80N1 -0.2

TABLE 2. Proposed Structures of Identified Compounds (Partial List)

O

O

1 O

O

OO

OO

OO

O

O2 3

6 OH

OOH

O

8 NH

O

9

N 2H

O

10

13 OH

O

12 OH

11 O

O

O

O O

O

O

O

OO

OH

OH

OH

OH

O

OH

O N

18

19

21 22

Palmitic Acid

Irganox 1010

Irganox 1076

Stearic Acid

Stearamide

Oleamide N 2H

O

7

DSDMA Irgafos 168 oxidation product

GC

HPLC CE

SEC GFC

Polarity

Molecular Weight

Derivatization

Pyrolysis

HS-GC

OP

O

OO

Two fragments (highlighted) and corresponding mechanisms shown here to demonstrate the process.

HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.

Page 6: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

6 Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products

Comprehensive Analysis of Extractables from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock1; Andrew Feilden2; Elisabeth Morgan2; Amalendu Sarkar3; Andrew White4

1Thermo Fisher Scientific, San Jose, CA, USA; 2Smithers Rapra, Shrewsbury, UK; 3Qure Medical, Rock Hill, SC, USA; 4 Buchi UK Ltd, Oldham, UK

Conclusion This study demonstrated a comprehensive extractable analysis workflow utilizing multiple techniques: HR-LCMS, GCMS, ICPMS, data processing software, and database searching. This workflow followed recommended analytical methods by PQRI[2]. HS-GCMS was carried out but the data has not been reported.

The DI water and IPA extraction profiles of the four types of medical grade rubber stoppers were quickly established by using this workflow.

The UHPLC/HRAM full MS/HCD MS2 with rapid polarity switching in a single run data acquisition, coupled with novel database search, significantly increase the confidence and throughput of routine extractable & leachable analysis, in particular for unknown components identification and structure characterization.

The GCMS and LCMS analysis are complementary to each other and necessary to give complete coverage of extractables.

References 1. FDA CFR 21.94, CFR 66011(b) and 600.11(h), CFR 211.160 2. PQRI “L/E Recommendations to the FDA”

http://www.pqri.org/publications/index.asp

Acknowledgements The authors would like to thank Sukhy Toot for her contributions, Qure Medical for providing the rubber stoppers, Smithers Rapra for the extractions, and Buchi for the loan of the speed extractor.

Introduction Rubbers & plastics are widely used in medical & drug delivery devices and packaging materials. Extractables & leachables assessment of all materials, especially from elastomeric & oligomeric components, forms an integral part of the submission for approval of a new drug product or medical device [1]. Extractable = possible impact. Test the material

Leachable = actual impact. Test the product The mass spectrometer plays an important role in E&L identification and structure elucidation, as it is coupled with many techniques for definitive analysis, see figure 1. Here we present a comprehensive workflow for medical grade rubber stopper extractable analysis using multiple techniques including HR-LCMS, GCMS, and ICPMS, followed by data processes using novel software and database searching. FIGURE 1. Potential analytical techniques with increasing chance of extractables and leachables being found as the molecular weight decreases.

LCMS Analyses Sample Preparation Four different types of medical grade rubber stoppers, sample-A, sample-B, sample-C, and sample-D, from Qure Medical, were extracted using DI water and IPA utilizing reflux extraction and the Buchi Speed Extractor. The extracts solutions were analyzed directly by LCMS.

Liquid Chromatography LC separations were carried out on the Thermo Scientific™ Dionex™ UltiMate™ 3000 RSLC system consisting of: DGP-3000RS pump, WPS-3000RS sampler, TCC-3000RS column compartment, and DAD-3000RS UV detector Column: Thermo Scientific™ Hypersil C18, 2.1x150 mm 1.9 µm Column Temp: 50ºC LC Mobile phase: A: H2O B: MeOH C: 50 mM Ammonium Acetate

Mass Spectrometry MS analyses were carried out on the Thermo Scientific™ Q Exactive™ mass spectrometer using both electrospray ionization (ESI) and atmosphere pressure chemical ionization (APCI). High resolution full scan MS and top 3 MS/MS data were collected in a data-dependent fashion at a resolving power of 70,000 and 17,500 (FWHM m/z 200) with polarity switching. The scan range is m/z150-1500. Stepped NCE (Normalized Collision Energy) setting was: 30, 40, 50.

FIGURE 4. HR Full Scan and HCD MS2 for Component ID and Structure Elucidation

LCMS and GCMS results show DI water extracts using both Speedy and Reflux techniques were “clean”. “Triisopropanolamine” is the major extractable.

Complex profiles of IPA extractions were observed from both LCMS and GCMS analysis. Complete extractable list not shown.

IPA reflux shows higher extraction efficiency compared with IPA speed extraction. However, the speed extractor conditions were not optimized for this study

Results show that GCMS and LCMS analysis are complementary to each other and necessary to give a fuller picture of the extractable profile.

FIGURE 5. mzCloud library Search Results for Irganox 1010

FIGURE 6. GCMS Chromatogram of IPA_REFLUX of Sample-D

FIGURE 7. ICPMS Results for the Four Rubber Stoppers (ppb)

GCMS Analyses Method and Instrumentation The DI water samples were extracted with Hexane. The samples in 2.0 mL GC vials were introduced in split injection mode into the Thermo Scientific™ TRACE™ Ultra Gas Chromatograph using a Thermo Scientific™ TriPlus RSH™ Autosampler. TG-5ms (30 m x0.25mm x0.25µm) column was used. Compounds were detected and identified with the ISQ single Quad mass spectrometer.

mzCloud Spectral Database Searching

TABLE 1. Components Identified from IPA Reflux of Sample-A (Partial List)

FIGURE 3. MS Chromatogram of IPA Reflux of Sample-A (ESI+) ICPMS Analyses The ICPMS samples were prepared by placing the rubber stoppers in 25 ml DI water and 25 ml 2% nitric acid and soaked at RT for 24 hours. The analyses were conducted on Thermo Scientific™ iCAP™ Q ICP-MS with He KED (Kinetic Energy Discrimination) interference reduction mode setting. To determine if trace and potentially toxic metals were leached from the stoppers, the USP<232> Class1 & 2 elements and additional elements which are commonly analyzed by ICP-MS were determined. The analysis results for the four types of rubber stoppers showed that they are clean of all Class 1 & 2 elements, see Figure 7 for the ICPMS results. In addition, the system control software Qtegra provides a full 21CFR Part 11 tool set to operate under compliant environments.

Element Sample-1 DI water

Sample-1 Nitric Acid

Sample-2 DI water

Sample-2 Nitric Acid

Sample-3 DI water

Sample-3 Nitric Acid

Sample-4 DI water

Sample-4 Nitric Acid LOD

75As (KED) ND ND ND ND ND ND ND ND <0.0233 111Cd (KED) 0.009 0.006 0.003 0.007 0.010 0.009 0.276 0.070 <0.0023 202Hg (KED) ND ND ND ND ND ND ND ND <0.0054 208Pb (KED) 0.061 0.069 0.018 0.124 0.100 0.106 0.159 0.122 <0.0008

9Be (KED) ND ND ND ND ND ND ND ND <0.0362 11B (KED) ND ND 0.541 ND 0.853 0.596 ND ND <0.5229

23Na (KED) 14.326 29.535 8.197 13.575 25.630 20.648 30.074 18.071 <0.1568 24Mg (KED) 1.835 4.091 1.160 1.013 2.802 4.236 2.009 1.531 <0.0231 27Al (KED) 0.420 3.688 0.479 2.325 1.029 3.615 0.867 7.665 <0.32 39K (KED) 8.246 11.185 5.921 5.295 13.580 11.235 16.088 6.645 <1.7964 48Ti (KED) 0.033 0.930 0.033 0.605 0.045 0.202 ND 0.107 <0.0314 51V (KED) 0.526 0.518 ND ND ND ND 0.051 0.044 <0.0339 52Cr (KED) ND 0.146 ND 0.103 ND 0.241 ND 0.108 <0.0072

55Mn (KED) 0.044 0.161 0.053 0.052 0.045 0.188 0.067 0.102 <0.0065 56Fe (KED) ND 9.962 ND 3.893 0.570 18.668 0.903 9.600 <0.0175 59Co (KED) 0.001 0.003 0.001 0.001 0.008 0.010 0.011 0.009 <0.0009 60Ni (KED) 0.097 0.123 0.047 0.075 0.254 0.420 0.715 0.281 <0.0063 65Cu (KED) 0.558 0.741 0.237 0.341 0.849 1.576 1.603 1.709 <0.007 66Zn (KED) 13.137 11.712 31.153 43.436 4.049 13.275 81.914 51.445 <0.5173 78Se (KED) ND ND ND ND ND ND ND ND <0.0781 88Sr (KED) 0.045 0.070 0.037 0.032 0.082 0.087 0.076 0.078 <0.0036

95Mo (KED) ND ND ND ND ND ND ND ND <0.4285 101Ru (KED) ND ND ND ND ND ND ND ND <0.0006 103Rh (KED) ND ND ND ND ND ND ND ND <0.0648 105Pd (KED) ND ND ND ND ND ND ND ND <0.0038 107Ag (KED) ND ND ND ND ND ND ND ND <0.0226 121Sb (KED) 0.007 0.017 0.007 0.020 0.017 0.124 0.040 0.052 <0.0012 137Ba (KED) 0.259 0.231 0.165 0.270 1.646 2.420 1.250 1.125 <0.0124 193Ir (KED) ND ND ND ND ND ND ND ND <0.0003 195Pt (KED) ND ND ND ND ND ND ND ND <0.0004 205Tl (KED) ND ND ND ND ND ND ND ND <0.0046

LCMS Result The all-in-one full scan/top3 ms/ms with polarity switching data acquisition using both ESI, APCI ionization ensures the detection of structurally diversified compounds. It provides comprehensive extractable profiles of the rubber stoppers, see Figures 2 and 3.

SR_Reflux_IPA_A_ESI_1 03/30/14 20:44:49Hypersil GOLD C18 150x2.1 1.9 um A: H2O B: MeOH C: 50 mMAmmonium AcetateRT: 0.00 - 45.01

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44Time (min)

0

10000

20000

30000

uAU

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

20

40

60

80

100

Relat

ive A

bund

ance

28.1526.72

30.2022.47

36.0729.0326.46 44.0625.32 31.430.90 42.3933.6721.851.85 17.78 36.89 40.8712.092.96 4.39 16.955.76 9.02 15.289.9730.20

22.47

36.0729.0326.6826.46

44.0631.43 42.3933.6721.851.02 20.5117.78 36.89 38.8812.09 39.9515.689.02 13.221.84 9.876.805.714.5328.15

26.72

29.410.90 22.43 24.941.85 21.58 32.45 33.66 36.03 38.04 40.87 43.5119.952.96 18.294.39 16.605.76 14.967.27 8.49 13.249.9732.42

36.0828.11 33.1230.2327.42 43.5636.99 42.4139.3525.1827.23

24.9023.70

20.5019.461.12 17.4316.29

NL: 6.29E9Base Peak MS SR_Reflux_IPA_A_ESI_1

NL: 2.65E9Base Peak F: FTMS + p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 6.29E9Base Peak F: FTMS - p ESI Full ms [150.00-1500.00] MS SR_Reflux_IPA_A_ESI_1

NL: 3.45E4Total Scan PDA SR_Reflux_IPA_A_ESI_1

ESI (+/-) TIC

ESI (+)

ESI (-)

PDA

A search was also conducted against Thermo ScientificTM mzCloudTM Library, a newly developed high resolution spectral database. mzCloud library provides several search criteria for small molecule structure identification using tandem mass spectra, including spectra, fragments, precursor ions, etc, all of which can be very useful for unknown structure elucidation. Figure 5 shows identification of Irganox 1010 using the ms/ms spectrum search feature. The accuracy of searching result is indicated by matching score between the query and library spectra. GCMS Instrument Conditions GCMS identified lower molecular weight and

volatile extractables which complement LCMS results. GCMS results showed that DI water extractions using both techniques were “clean” and more extractables were detected from IPA extractions. Within the four rubber stoppers, sample C&D had more low molecular weight extractables detected by GCMS from both extraction techniques, see Figure 6 and 7.

Ace

toph

enon

e

Dim

ethy

lphe

nylm

etha

nol

L-Li

mon

ene

Dim

ethy

lben

zyl m

ethy

l eth

er

Dip

heny

l eth

er

Triis

opro

pano

lam

ine

+Lon

gipi

nene

B

HT

Non

ylbe

nzen

e de

rivat

ive

Patc

houl

ene

1,5,

9-Tr

imet

hyl c

yclo

dode

catri

ene

C20

Uns

atur

ated

hyd

roca

rbon

Patc

houl

ene

isom

er

4/20/2014 3:47:32 AM IPA_REFLUX_D RT: 4.66 - 27.35

6 8 10 12 14 16 18 20 22 24 26 Time (min)

0 1000000000

2000000000

3000000000

4000000000

5000000000

6000000000

7000000000

8000000000

9000000000

10000000000

11000000000

12000000000

13000000000

14000000000

Rel

ative

Abu

ndan

ce

6.62 11.94 6.43

9.98 12.01 10.61 13.11 10.95 14.40 15.26 15.39 6.68 18.15 14.21 19.02 20.75 7.16 22.07 23.32 9.87 6.00 24.92 27.14

50 63 75 88 100 113 125 138 150 163 175 188 200 213 225 238 250 263 275 288 300 313 325 338 350 363 375 388 400 413 425 438 450

0

10

20

30

40

50

60

70

80

90

100155.070489.0598

163.1330

145.1225111.0442

101.0963

57.0702173.0810

217.1075

317.1964261.134067.0544 127.0753

255.1593

O

OO

OO

O

OO

OO

OOH

O

OO

O

OOO

OHO

OO

OOH

OO

O

O

O 2H

OOO

O

O 2HOO

O

O

O2HO OH

OO

OO

O OHO

OH

OH

180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 m/z

0 5

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95

100

Rel

ative

Abu

ndan

ce

452.3227 C 22 H 46 O 8 N 1.9203 ppm

435.2955 C 22 H 43 O 8

0.6606 ppm

(M+H)+

(M+NH4)+

OO

OO

OO

O

O

Bis[2-(2-butoxyethoxy0ethyl] adipate Molecular Formula : C22H42O8 Formula Weight : 434.2880

FIGURE 2. LCMS Chromatogram of IPA Reflux of Sample A

Figure 7. GCMS Spectra of Compounds Identified in IPA_REFLUX of Sample-D (Partial)

Component Detection and Structure Elucidation The High Resolution Accurate Mass (HRAM) data were processed using differential analysis software SIEVETM 2.1 for component extraction. ChemSpider database searching was carried out to obtain possible structures of extracted components. While many possible hits were obtained for each component, to determine the correct structures, “Thermo ScientificTM Mass FrontierTM Software”, a small molecule structure analysis software, was used. The “HighChem Fragmentation LibraryTM ” in Mass Frontier 7.0 has extensive published literature references. For each proposed structure, the “Fragments and Mechanisms” feature in Mass Frontier was used to generate predicted “fragments and mechanisms” through HighChem Library search, see figure 4-b. A high degree of correlation between predicted and experimental fragments (indicated in red, see figure 4-c) confirms the proposed structure. Mass Frontier then automatically annotates the matching fragments based on library search results, see figure 4-c.

m/z 452.3218

OO

OO

OO

O

O

NH4

OO

OO

OO

OH

Om/z 435.2952

+H

OO

O

OO

O

m/z 304.1880

Lib

OHO

O

O

O

OO

m/z 334.1986

Lib

OO

OHO

OO

O

Om/z 435.2952

+H

Lib

O 2HO

O

m/z 163.1329

rHB

rHB

Lib

rHB

m/z 452.3218

OO

OO

OO

O

O

NH4

OHO

Om/z 162.1250

Lib

OHO

Om/z 162.1250

Lib

OOH

m/z 89.0597

i

Lib

Summary of Reflux and Speed Extractions using IPA and DI Water

4-a. HRAM Full MS to Determine Elemental Composition

4-c. HRAM MS/MS Spectra For Structure Elucidation

4-b. HighChem Fragmentation Library Search to Predict Fragments and Mechanism

RT: 6.41

50 100 150 0

50

100 105.01 77.10

120.04 51.12

77.99 74.13 91.16

134.07 147

O

Acetophenone MW 120.04

RT: 6.58

50 100 150 0

50

100 43.01

121.08

77.11

45.13 51.02

78.14 105.05

91.15 59.15

136.04 141.04

Dimethylphenylmethanol M+ 136.09

OH

RT: 10.93

50 100 150 0

50

100 146.12

70.02

88.08 42.03 98.18 59.14

102.18

128.18 158.09 176.08

Triisopropanolamine M+ 191.15

OHN

OH

OH

RT: 11.90

50 100 150 200 0

50

100 57.01

205.14

81.15

145.16 105.14 67.12 220.23 91.04 177.13 41.06

121.15 131.16

149.17 189.18

BHT M+ 220.18

OH

1,5,9 Trimethyl Cyclododecatriene MW. 204.35 Da

RT: 12.46

50 100 150 200 0

50

100 68.15

67.00

93.06

107.16

79.13 121.18 53.12

41.15 189.14

147.19 161.16 204.19

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Time (min)

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

Rel

ative

Abu

ndan

ce

* * * * * * *

*

P3

PEG P9

P13

P19 P20

P21

P18 B B

B

Erucamide

Irganox 1010

Phthalate

P10

P22 Irganox 1076

DSDMA

* PPG B-Background P- Extractable

P7 P14

P15 P16 P17 B

P8(-)

P6(-) P5 P4

P2

P1

P11

B

P12

Peak ID RT Mode Measured (M+H) +

Calculated (M+H) +

Elemental Composition

Error (ppm)

1 21.48 APCI (+) 235.1691 235.1693 C15H22O2 -0.67 2 22.43 ESI (-) 199.1707 (M-H) – 199.1693 (M-H) – C12H24O2 2.0 3 22.47 ESI (+) 435.2956 (M+NH4) + 435.2952 C22H42O8 0.86 4 24.94 ESI (-) 227.2017 (M-H) – 227.2006 (M-H) – C14H28O2 2.23 5 24.87 ESI (+) 219.1743 219.1743 C15H22O 1.8 6 26.72 ESI (-) 255.2330 (M-H) – 255.2391 (M-H) – C16H32O2 3.5 7 27.75 ESI (+) 282.2791 282.2791 C18H36ON -0.11 8 28.15 ESI (-) 283.2643 (M-H) – 283.2632 (M-H) – C18H36O2 2.8 9 28.19 ESI (+) 383.3396 (M+NH4) + 383.3396 (M+NH4) + (C24H47ON)+NH4 0.42

10 28.71 ESI (+) 284.2946 284.2948 C18H37O1N1 -0.5 11 29.14 ESI (+) 325.3097 325.3101 C21H40O2 1.3 12 29.36 ESI (+) 319.2992 319.2995 C22H38O1 -1.1 13 29.41 ESI (-) 311.2963 (M-H) – 311.2945 (M-H) – C20H40O2 2.4 14 31.43 ESI (+) 340.3568 340.3574 C22H46ON

-1.8 15 31.77 ESI (+) 366.3729 366.3730 C24H48ON -0.5 16 32.52 ESI (+) 409.3100 409.3101 C28H40O2 -0.34 17 33.70 ESI (+) 1227.00 1227.00 ?

18 36.00 ESI (+) 1194.8170 (M+NH4) + 1194.8179 (M+NH4) + C73H108O12 -0.3

19 37.57 ESI (+) 663.4536 663.4537 C42H63O4P -0.19 20 38.85 ESI (+) 522.5969 522.5972 C36H76N -0.23 21 42.39 ESI (+) 548.5035 (M+NH4) + 548.5035 (M+NH4) + (C35H62O3)+NH4 -0.33 22 44.08 ESI (+) 550.6285 550.6285 C38H80N1 -0.2

TABLE 2. Proposed Structures of Identified Compounds (Partial List)

O

O

1 O

O

OO

OO

OO

O

O2 3

6 OH

OOH

O

8 NH

O

9

N 2H

O

10

13 OH

O

12 OH

11 O

O

O

O O

O

O

O

OO

OH

OH

OH

OH

O

OH

O N

18

19

21 22

Palmitic Acid

Irganox 1010

Irganox 1076

Stearic Acid

Stearamide

Oleamide N 2H

O

7

DSDMA Irgafos 168 oxidation product

GC

HPLC CE

SEC GFC

Polarity

Molecular Weight

Derivatization

Pyrolysis

HS-GC

OP

O

OO

Two fragments (highlighted) and corresponding mechanisms shown here to demonstrate the process.

HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.

Page 7: Comprehensive Analysis of Extractables from Rubber Stopper ... · from Rubber Stopper used in Medical Devices and Pharmaceutical Products Kate Comstock, 1 Andrew Feilden,2 Elisabeth

www.thermofisher.com©2016 Thermo Fisher Scientific Inc. All rights reserved. HighChem Fragmentation library is a trademark of HighChem, Ltd. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific prod-ucts. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative for details.

PN-64219-EN 0716S

Africa +43 1 333 50 34 0Australia +61 3 9757 4300Austria +43 810 282 206Belgium +32 53 73 42 41Canada +1 800 530 8447China 800 810 5118 (free call domestic)

400 650 5118

Denmark +45 70 23 62 60Europe-Other +43 1 333 50 34 0Finland +358 9 3291 0200France +33 1 60 92 48 00Germany +49 6103 408 1014India +91 22 6742 9494Italy +39 02 950 591

Japan +81 45 453 9100Latin America +1 561 688 8700Middle East +43 1 333 50 34 0Netherlands +31 76 579 55 55New Zealand +64 9 980 6700Norway +46 8 556 468 00Russia/CIS +43 1 333 50 34 0

Singapore +65 6289 1190Spain +34 914 845 965Sweden +46 8 556 468 00Switzerland +41 61 716 77 00UK +44 1442 233555USA +1 800 532 4752