653.78 653.80 653.82 653.84 653.86 653.88 653.90

m/z

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

Re

lative

Ab

un

da

nce

0

20

40

60

80

100

0

20

40

60

80

100

653.8029R=70741

653.7707R=68199

653.8031R=136928

653.8247R=153121

653.8034R=243449

653.8039R=591659

653.8018R=589119

653.7977R=539419

653.8040R=593824

653.8016R=599008

653.7977R=596861

653.8070R=594065

NL:3.81E5

met_120K_5e4#1652-1668 RT: 8.91-9.00 AV: 17 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:2.28E5

met_240k_5e4#920-932 RT: 8.88-9.00 AV: 13 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:1.78E5

met_500k_5e4_ic_20ul#493-496 RT: 8.95-9.01 AV: 4 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:1.06E5

met_1m_5e4_ic_20ul#247-251 RT: 8.91-9.06 AV: 5 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:2.42E2

C 15 H12 I3 NO4 +H: C 15 H13 I3 N1 O4

p (gss, s /p:40) Chrg 1R: 600000 Res .Pwr . @FWHM

500K, Full Scan, AGC 5e4

1M, Full Scan, AGC 5e4

Simulation, 600K

779.690 779.695 779.700 779.705 779.710 779.715

m/z

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100

Re

lative

Ab

un

da

nce

0

20

40

60

80

100

0

20

40

60

80

100

779.7000R=230550

779.7004R=522954

779.6979R=510670

779.6940R=689868

779.7005R=549176

779.6981R=515043

779.6944R=582233

779.7162R=524131

779.7104R=570932

779.7003R=518070

779.6983R=565962

779.6945R=723343

779.7007R=540434

779.6982R=539804

779.6944R=547080

779.7037R=435971

779.7067R=302769

NL:1.61E5

met_500k_5e4_ic_20ul#535-543 RT: 9.72-9.87 AV: 9 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:6.35E4

met_1m_tsim#50-54 RT: 9.75-9.89 AV: 5 T: FTMS + p ESI SIM ms [197.0777-781.2000]

NL:3.64E4

met_1m_tsim_1e3#49-54 RT: 9.69-9.86 AV: 6 T: FTMS + p ESI SIM ms [197.0777-781.2000]

NL:1.32E5

met_1M_5e4_IC_20ul#270-272 RT: 9.75-9.82 AV: 3 T: FTMS + p ESI Full ms [150.0000-1000.0000]

NL:2.43E2

C 15 H11 I4 NO4 +H: C 15 H12 I4 N1 O4

p (gss, s /p:40) Chrg 1R: 550000 Res .Pwr . @FWHM

RESULTS

Performance considerations and validation

In FTMS measurements, ion cloud coherence determines the length of time over which useful

measurements can be made. Cloud coherence can be compromised especially for less dense ion

clouds, and this can lead to intensity-dependent decay. This in turn can lead to inaccurate isotope

abundances, as shown below.

ABSTRACT

Purpose: This work surveys key performance problems and their solutions in Orbitrap mass

spectrometry at high resolving powers and presents example applications benefiting from high-

resolution Orbitrap analysis.

Methods: A Thermo Scientific™ Orbitrap Fusion™ Lumos™ Tribrid™ mass spectrometer was modified

by (1) ultra-precisely machined Orbitrap electrodes; and (2) carefully controlled injection conditions.

Results: Improvements in performance enabling routine analysis at resolving power 1,000,000 (at m/z

200), are presented, along with results from applications in metabolomics and lipid fluxomics.

INTRODUCTION Since its commercial introduction in 2005, the Orbitrap mass analyzer has been widely adopted by the

mass spectrometry community. Historically, applications in which Orbitrap analysis has been

particularly useful are those which require high resolving power. For example, in many proteomics

studies, resolution of 60k or 120k (at m/z 200) is more than sufficient to accurately determine the

monoisotopic mass of multiply charged peptides across the mass range of interest. However, some

applications require higher resolutions and until now they remained served only by Fourier Transform

Ion Cyclotron Resonance (FT-ICR) instruments. Here we describe an Orbitrap-based mass

spectrometer capable of dependable and routine analysis at a resolving power of 1M. Performance

criteria are discussed and applications examples are provided.

MATERIALS AND METHODS Samples. For calibration and characterization, a standard calibration solution was used, which

consisted of n-butylamine, caffeine, MRFA, and Ultramark 1621 dissolved in 50:49:1 acetonitrile,

methanol, and formic acid. For the metabolite standards experiment, 24 common metabolites were

mixed at a concentration of 100 pg/µL. 20 µL were then injected on to a Thermo Scientific™ Hypersil

GOLD™ column (100x2.1 mm). Analytes were separated using a 20-min gradient starting at 5% buffer

B (99.9:0.1 acetonitrile:formic acid) and increasing to 35% B (buffer A was 99.9:0.1 water:formic acid),

at a flow rate of 300 µL/min. For the flavonoid and lipid fluxomics work, sample information can be

found in the referenced presentations.

Mass spectrometry. All mass spectrometry was performed on a Orbitrap Fusion Lumos Tribrid mass

spectrometer equipped with the 1M Option (Fig. 1). The 1M Option consists of ultra-precisely machined

high-field Orbitrap electrodes used in combination with a C-trap and injection optics with improved

manufacturing tolerances. The UHV system was baked at a temperature of approximately 120oC until

the base pressure was ≤ 5 × 10-11 torr. Applied voltages were calibrated according to an automated

calibration routine as described below.

B

A

FIGURE 1. Orbitrap Fusion Lumos mass spectrometer equipped with the 1M Option:

(A) high-precision Orbitrap electrodes; (B) C-trap and injection optics with improved

manufacturing tolerances.

C-trap

Lenses

Deflector

The problem described above is worse for longer transients, i.e., high resolving powers. To

ensure quantitative accuracy and adherence to all performance objectives, we employ an

automated calibration routine. This builds upon the routine previously described [1], which

employs a genetic algorithm and tailored scoring in order to navigate the multidimensional

optimization surface that arises in C-trap/Orbitrap calibration.

FIGURE 2. Intensity-dependent signal decay can lead to inaccurate isotopic

abundances for smaller ion clouds (green) relative to larger (red).

time

amp

litu

de

time

amp

litu

de

m/z

m/z

Ideal isotope ratio

Low isotope ratio

Caffeine, m/z 195

A+1

A+2

Avg Rp = 1100197

MRFA, m/z 524

FIGURE 4. With the improvements shown described in methods and with proper

calibration, good results can be obtained as shown above. Each species is quad-

isolated (20 m/z isolation width), AGC target 5e4 ions.

Avg Rp = 654752

A+2

A+3

FIGURE 5. The above results can

be improved slightly by

exercising more control over the

ion population. At right is the

quad-isolated A+2 peak of MRFA,

(isolation width 1 Da), AGC target

2000 ions. With internal

calibration turned on, mass

accuracy is << 1 ppm.

Avg Rp = 650031

Avg S/N = 1647

FIGURE 6. We characterize

the high-resolution

behavior of the C-

trap/Orbitrap combination

by finding the signal-to-

noise ratio at which the 18O

and 13C2 peaks start to lose

baseline separation. This

starts when the mass

difference (Δm) between

the two reaches about 1.75

mDa (about 2.5 mDa

theoretical), as the number

of ions in the peak cluster

increases.

Example applications

The figures below show fine isotopic structure revealed in a metabolites standard, on an LC

timescale.

Liothyronine (m/z 651)

Thyroxine

(m/z 777)

Simulation 550K

1M, SIM IW 7, AGC 1e3

1M, Full Scan, AGC 5e4

Liothyronine, A+2

Orange Juice Matrix

Base Peak Chromatogram

Extracted Ion

Chromatogram Putative

Flavonoid Conjugates

FIGURE 7. Two examples

of metabolite fine structure

are shown at left, acquired

on an LC timescale

(chromatography trace

above). Data were

acquired with different

strategies, showing that

isolation and control of the

ion population yields the

best results.

Thyroxine, A+2

Liothyronine is a synthetic

form of triiodothyronine, a

hormone used in treatment of

thyroid conditions.

LC/MS of metabolite standards

Flavonoids in orange juice

18O 213C

Filter on fine

structure

pattern

FIGURE 8. Using fine isotope

structure as a pattern filter

allowed detection of 128

putative flavanoid conjugates

in a mixture containing >4000

compounds.

For more, see ASMS ‘17 Proceedings:

WP457, T. Stratton et al.

Lipid Fluxomics

In fluxomics, the objective is the quantitation of functional molecular turnover over many different

molecular species of interest.

The figure below describes measurements of lipid biosynthesis. Normally such measurements are

made using tritium labeling; however, very few laboratories are able to practice this method due to

strict controls on radioactive materials and handling. Here deuterium labeling is used, and the

labeled and unlabeled forms can be baseline separated using ultra-high resolution Orbitrap MS.

Precursor: TAG 52:3 Fragment: TAG 52:3 HCD

FIGURE 9. Intact deuterium-label triacylglyceride (TAG 52:3) is fragmented to yield a

diacylglerceride (DAG 34:1), which retains a double-bond. The fragment DAG is at lower

m/z where higher resolving power allows the 13C and D peaks to be easily separated.

CONCLUSIONS

Using ultra-precisely machined Orbitrap electrodes and employing exact control over ion injection

conditions, transients of 2 seconds yield good performance at a mass resolving power of 1,000,000 (at

m/z 200).

Optimized calibration and tuning provide accurate isotope fidelity and robust space charge

performance

Data obtained with the 1M Option provide important information about a variety of research topics.

REFERENCES

1. J. Canterbury et al., Poster M309, ASMS 2013, “Automatic Isotope Ratio Optimization on High-field

Orbitrap Mass Analyzers.”

ACKNOWLEDGEMENTS

We are grateful to Matthew Mitsche of the University of Texas Southwestern Medical Center for examples

from his lipid fluxomics work.

We gratefully acknowledge Dae-Eun Lee for valuable assistance during the course of this work. We

thank Hartmut Kuipers and Oliver Lange for help with data processing. Finally we thank Raman Mathur

and Chad Weisbrod for helpful discussions.

TRADEMARKS/LICENSING

© 2017 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher

Scientific and its subsidiaries. This information is not intended to encourage use of these products in any

manner that might infringe the intellectual property rights of others.

For more, see ASMS ‘17 Proceedings

MOB 10:10 am,

Matthew Mitsche et al.

FIGURE 3. Top left, complex surface

representing the response of an

isotope score (a composite of

isotope fidelity measurements for

caffeine, MRFA, and UM1522) to

changes in two important optics

elements, the deflector and an

adjacent lens. While complex, this 2-

D surface captures only a portion of

the complexity of the entire multi-

dimensional surface, which includes

responses from 6 other optics

elements.

Bottom left, this complexity can be

significantly reduced by using a more

selective score (see ref [1]), which

makes the true optimum more clear.

Thyroxine is a thyroid hormone

responsible for regulation of

metabolism.

18O 13C2

18O 13C2

13C

15N

13C2

18O

15N13C 2H13C

15N34S

13C34S 13C2

15N

13C233S

18O13C

13C3

2H13C2

34S

15N13C

13C33S

18O

13C2

2H13C

RT: 0.00 - 25.02

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

21.59

1.01

1.37

8.99

9.82

4.990.90

21.30

3.98

19.3620.04

11.56

20.224.740.22

19.14

18.82

20.5818.67

18.4217.661.84

11.71 17.2617.01

2.28 16.103.00 22.68 24.1611.20 15.3012.87 13.925.72 6.52 7.79 8.44

NL:2.97E7

Base Peak MS met_1M_5e4_IC_20ul

34S

13C33S

18O

13C2

2H13C

15N13C

Jesse Canterbury1, Michael Senko1, Seema Sharma1, Tim Stratton1, Vlad Zabrouskov1, Eduard Denisov2, Alexander Makarov2 Thermo Fisher Scientific, 1San Jose, CA, USA and 2Bremen, Germany

Performance metrics and example applications: Orbitrap mass spectrometry at resolving power 1M

PO65013--EN 0617S