223-66060

May 2017

LabSolutions

i-PeakFinder (New Peak Integration)

Instruction Manual
Guide to Adjusting Parameters

Read this manual thoroughly before you use the product.

Keep this manual for future reference.
This page is intentionally left blank.
 Introduction
Read this Instruction Manual thoroughly before using the product.

Thank you for purchasing this product.

This manual describes the operation for this product. Read this manual thoroughly before using
the product and operate the product in accordance with the instructions in this manual.

Keep this manual for future reference.

IMPORTANT
• If the user or usage location changes, ensure that this Instruction Manual is always kept
together with the product.
• If this manual or a product warning label is lost or damaged, immediately contact your
Shimadzu representative to request a replacement.

 Notice
• Information in this manual is subject to change without notice and does not represent a
commitment on the part of the vendor.
• Any errors or omissions which may have occurred in this manual despite the utmost
care taken in its production will be corrected as soon as possible, although not
necessarily immediately after detection.
• All rights are reserved, including those to reproduce this manual or parts thereof in any
form without permission in writing from Shimadzu Corporation.
• Microsoft and Windows are registered trademarks of Microsoft Corporation in the
United States and/or other countries. Other company names and product names
mentioned in this manual are trademarks or registered trademarks of their respective
companies. The TM and ® symbols are omitted in this manual.
• LabSolutions is registered trademark of Shimadzu Corporation in the United States and
Europe (Office for Harmonization in the Internal Market).
© 2017 Shimadzu Corporation. All rights reserved.
Original version is approved in English.

i-PeakFinder i
Introduction

This page is intentionally left blank.

ii i-PeakFinder
 Contents

Introduction ......................................................................................... i

Contents ............................................................................................ iii

1. i-PeakFinder ................................................................................. 1
1.1 When to Use i-PeakFinder....................................................................................................... 1
1.2 Using i-PeakFinder .................................................................................................................. 3

2. Adjusting the Baseline .................................................................. 5

2.1 Adjusting the Baseline ............................................................................................................. 5
Example: Chromatograms of Hydrocarbons Analyzed by GC with and Without Baseline
Estimation ..........................................................................................................................6
2.2 Fused Peaks and Baseline-Resolved Peaks ........................................................................... 7
2.3 Baseline Type Parameter ........................................................................................................ 7
Example: Processing of HPLC Gradient Curve in Under-estimated Baseline ..............................8
Example: Chromatograms of Hydrocarbons Analyzed by GC .....................................................10
Chapter Endnotes .......................................................................................................................... 14

3. Detecting and Unifying Peaks ..................................................... 15

3.1 Peak Detection ...................................................................................................................... 15
3.2 Peak Detection Sensitivity and Estimated Noise Range ....................................................... 15
[Detection Threshold] .........................................................................................................................15
[Noise Estimation Range] ...................................................................................................................15
Example: Chromatograms with High Noise Near Peak Tops and Excessive Peak Division ......15
3.3 Minimum Half Height Width and Maximum Half Height Width ............................................... 17
[Minimum Half Width] .........................................................................................................................17
[Maximum Half Width] ........................................................................................................................17
Example: LCMS Chromatograms with Peak Profiles Divided into Multiple Segments ..............18
3.4 Peak Baseline Height ............................................................................................................ 21
[Peak Baseline Height] .......................................................................................................................21
Example: Ensuring Area Repeatability by Raising the Peak-baseline Slightly ...........................22

i-PeakFinder iii
Contents

3.5 Unifying Peaks ....................................................................................................................... 23

Example: Chromatograms with Peak Distortion for LCMS Quantitative Peak Integration........ 23
3.6 Peak Filters ............................................................................................................................ 25
Example: Process for When a Downturn at the Peak End Point Due to Baseline Undulations
Causes a Long Narrow Region That Is Detected as a Peak ........................................ 25
Chapter Endnotes .......................................................................................................................... 28

iv i-PeakFinder
 1. i-PeakFinder
i-PeakFinder is a new peak integration algorithm for LabSolutions LC/GC/LCMS chromatograms.

Parameter settings can be changed to modify the default processing methods used by i-PeakFinder. This
guide describes which parameters, or combinations thereof, are effective for making adjustments to
i-PeakFinder processes.

1.1 When to Use i-PeakFinder

 When the Baseline Is Undulating or When Shoulder Peaks Are Detected

[Algorithm] = [Chromatopac]
mAU
0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.4

-0.5

-0.6

-0.7

-0.8

-0.9

4.25 4.50 4.75 5.00 5.25 5.50 5.75 6.00 6.25 6.50 6.75 7.00 min

[Algorithm] = [i-PeakFinder], [Detection Threshold] = 2

mAU
0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

-0.1

-0.2

-0.3

-0.4

-0.5

-0.6

-0.7

-0.8

-0.9

4.25 4.50 4.75 5.00 5.25 5.50 5.75 6.00 6.25 6.50 6.75 7.00 min

Existing peak integration algorithms used to detect peaks based on parameters equivalent to slope
generally require using a time program for analysis, because baseline undulations (changes in baseline
slope as a function of time) also result in varying the degree of initial rise and fall of peaks as a function
of time.

Because i-PeakFinder can also detect shoulder peaks, as can be seen from peak integration results
near the center area, it is also useful for applications that require more accurate peak separation, such
for analyzing impurities.

i-PeakFinder 1
i-PeakFinder

 When Peaks Are Buried in Noise or When Peak Widths Vary

[Algorithm] = [Chromatopac], SLOPE = 100000
mAU

150

140

130

120

110

100

90

80

70

60

50

40

30

20

10

-10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 min

[Algorithm] = [i-PeakFinder], [Detection Threshold] = 5 (default)

mAU

150

140

130

120

110

100

90

80

70

60

50

40

30

20

10

-10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 min

Due to its superior ability to detect peaks with low S/N ratios, i-PeakFinder is able to correctly process
chromatograms even when peak widths vary within the chromatogram or when the noise level suddenly
increases.

 When Peak Distortion Is Large or When Peaks Are Divided

[Algorithm] = [i-PeakFinder], [Minimum Half Width] = 20
→ See Example: LCMS Chromatograms with Peak Profiles Divided into Multiple Segments.
3750

3500

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

-250
3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

In this example, both algorithms result in multiple peaks, given default parameter settings. However, to
unify peaks into a single peak, the Chromatopac algorithm requires manually integrating each peak,
whereas the i-PeakFinder algorithm only requires changing a single set of parameter settings for all
peaks.

2 i-PeakFinder
 i-PeakFinder

1.2 Using i-PeakFinder

i-PeakFinder can be specified on the [Integration] tab page either in the Method View or in the [Peak
Integration Parameters] dialog box.

Start by selecting [i-PeakFinder] in the [Algorithm] pull-down menu.

Infrequently used settings are specified in the [Integration Advanced Settings] window, which is
displayed by clicking the [Advanced] button on the right side of the window.

Respective settings are organized based on application and arranged in roughly the order processes
are performed.

Peak Detection Parameters

Parameters for peak detection include the following.
[Detection Threshold], [Minimum Half Width]*, [Integration Range], and all parameters on the [Peak
Detection] tab page in the [Integration Advanced Settings] window

* For LCMS. For LC or GC, the parameter is specified in the [Integration Advanced Settings] window.

i-PeakFinder 3
i-PeakFinder

Baseline Type and Peak Unification Parameters

Parameters for peak unification include the following.
[Baseline Type], and all parameters on the [Peak Unification] tab page in the [Integration Advanced
Settings] window

Peak Filter
Parameters for peak filtering include the following.
All parameters on the [Peak Filter] tab page, and all parameters on the [Peak Filter] tab page in the
[Integration Advanced Settings] window

LC/GC

LCMS

4 i-PeakFinder
 2. Adjusting the Baseline
The baseline depends significantly on the peak detection status. This chapter describes how to adjust the
baseline and what to do if the baseline cannot be adequately adjusted due to the peak detection status.

2.1 Adjusting the Baseline

i-PeakFinder adjusts for undulations and other baseline variations by automatically estimating the
chromatogram baseline and noise level.

For many chromatograms, this feature can be used to obtain correct results automatically by simply
specifying the detection sensitivity based on the required number of peaks.

With baseline estimation (given a [Detection Threshold] setting of 1000)

　
uV
250000

225000

200000

175000

150000

125000

100000

75000

50000

25000

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 min

With baseline estimation disabled (given a [Detection Threshold] setting of 1000)

　
uV

275000

250000

225000

200000

175000

150000

125000

100000

75000

50000

25000

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 min

By simply adjusting the [Detection Threshold] setting, even the chromatograms above can be
automatically processed without any problem by using the parameter settings changed as described in
Example: in 2.3 Baseline Type Parameter.

i-PeakFinder 5
Adjusting the Baseline

With baseline estimation (given a [Detection Threshold] setting of 30)

mV

55

50

45

40

35

30

25

20

15

10

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

However, in cases where large baseline undulations prevent successful estimation or for types of
analytical methods that would produce better peak integration results with estimation disabled, good
results can be obtained by specifying peak integration parameter settings.

In this chapter, examples are used to describe how to obtain appropriate results.

Example: Chromatograms of Hydrocarbons Analyzed by GC with and Without Baseline

Estimation
For this analysis, better results were obtained by disabling baseline estimation.

With baseline estimation

uV

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

With baseline estimation disabled g

uV

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

6 i-PeakFinder
 Adjusting the Baseline

2.2 Fused Peaks and Baseline-Resolved Peaks

If a series of fused = unseparated peaks are detected, then the adjacent peaks are divided vertically
and, in general i, the peak-baseline is formed by a straight line between the starting point of the first
peak and the ending point of the last peak in that group of peaks.
(Fused peaks that satisfy certain specific conditions can also be processed with tailing process.)

If a group of peaks is separated, baseline segments are formed between the peaks and the overall
baseline is formed by connecting those baseline segments together.

Baseline-Resolved Peaks (Left) and Fused Peaks (Right)

For chromatograms separated into peaks and

baseline segments based on peak detection
mV

18
parameters (baseline resolution), such as
17
[Detection Threshold] or [Minimum Half Width], the
16 baseline segments are always used to form the
15 overall baseline, regardless of the Baseline Type
14
Parameter described below ii.
13
If the baseline is not tracked adequately by the
12
baseline estimation function or the level of tracking
11
needs to be adjusted, then consider first adjusting
10 peak detection parameters so that an appropriate
9
baseline area forms between peaks (i.e. refine
8
baseline estimation for detecting the desired
peaks).
→ See Example: .
7

18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min

Conversely, to ensure fused peaks (vertical division), such as in the case of the GC analysis of
hydrocarbons mentioned above, peak detection parameters must be adjusted so that no baseline
segments form between an uninterrupted series of detected peaks.
→ See Example: Chromatograms of Hydrocarbons Analyzed by GC.

2.3 Baseline Type Parameter

mV

18
Baseline estimation forms a straight peak-baseline
through groups of fused and vertically divided peaks,
17
except if the peak-baseline crosses the
16
chromatogram plot line, in which case it is always
15 bent downward to avoid the plot line.
14

13
In contrast, the peak-baseline can also be bent
12
upward iii by specifying peak height or baseline
length criteria, unless [Vertical Division] is specified
11
in the [Baseline Type] setting.
10

9 The level of peak-baseline bending can be adjusted

8
even if baseline segments did not form after the
7
adjustments above.
18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min

i-PeakFinder 7
Adjusting the Baseline

Example: Processing of HPLC Gradient Curve in Under-estimated Baseline

In the following chromatogram, baseline estimation was disabled iv for explanatory purposes.
mV

55

50

45

40

35

30

25

20

15

10

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

1 Specify parameters that does not result in misidentifying the baseline as a peak

For chromatograms with upward baselines (baselines that initially rise and then fall), the baseline
bulges can sometimes be misidentified as peaks. Therefore, if there are no necessary peaks after the
down slope, it is easier to use the [Integration Range] setting to exclude that section.

If that is not possible, the misidentification can probably be eliminated by using the [Maximum Half
Width] setting to exclude false peaks* greater than the required peak half width. (See 17 on page
[Maximum Half Width])
*: Bulges in the baseline misidentified as a peak

In this example, an [Integration Range] setting of 0 to 25 min is specified.

mV

55

50

45

40

35

30

25

20

15

10

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

2 Specify peak detection parameters that result in baseline segments (standard method)

Normally, the [Detection threshold] setting is adjusted, which involves dividing the chromatogram plot
line into peaks and baselines based on an S/N ratio as the threshold value. However, if possible,
adjustments to [Detection threshold] can be minimized or eliminated by setting [Noise estimation
range] to the non-peak area which the user feels to have the highest noise level, so that the user can
set the noise value in the S/N ratio as desired.

Because the noise estimation function estimates the noise based on the chromatogram signal
frequency, if an area of the baseline with a gentle bend is specified, a low noise component can be
estimated based on that bend.

8 i-PeakFinder
 Adjusting the Baseline

In this example, the baseline area range between 23 and 24 minutes is specified, near the
shouldered baseline area where the chromatogram has finished rising.
mV

55

50

45

40

35

30

25

20

15

10

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

v
If necessary, make minor adjustments to the [Detection Threshold] setting. (In the following
example, change the setting from 5 to 3.)
mV mV

19 19

18 18

17 17

16 16

15 15

14 14

13 13

12 12

11 11

19.5 20.0 20.5 21.0 21.5 min 19.5 20.0 20.5 21.0 21.5 min

2 Bend the fused peak-baseline upward during baseline processing (alternative method)

This method involves moving the peak-baseline for the fused peaks so that it more closely matches
the chromatogram plot line shown after 19 minutes in the chromatogram in step 1 above (straight
peak-baseline corresponding to vertically divided peaks).

The level of baseline bend can be specified in the edit box that is enabled by selecting the [Baseline
Type] checkbox and clicking [Baseline Length] and [Valley/Height Ratio] in the pull-down menu.

In this example, [Baseline Length] was set to 0.5 minutes, so that any straight areas of fused
peak-baseline longer than that setting are bent upward. サンプル名:Test
mV

55

50

45

40

35

30

25

20

15

10

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

i-PeakFinder 9
Adjusting the Baseline

Increase Likelihood of Peak Separation by Increasing Peak Baseline Height (for Reference)
Height = 5 (default) Height = 25
mV mV
19

18
18

17
17

16
16

15
15

14 14

13 13

12 12

11 11

10 10

9 9

8 8

7
7

18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min 18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min

The [Peak Baseline Height] setting (on the [Peak Detection] tab page in the [Integration Advanced
Settings] window) determines the height of peak starting and ending points based on the specified
multiple of noise standard deviations. Increasing the setting raises the peak-baseline position, which
shortens the length of leading and tailing ends.

That results in peaks with narrower bases and increases the likelihood of peak separation which
means peak-baseline separation between peaks. This technique can be especially effective when
used in conjunction with the method described above.

Because the result is determined based on the standard deviation of noise, a relatively modest initial
setting should probably be specified for chromatograms with a high noise level, whereas a
significantly large initial value should be specified for chromatograms with low noise.

Example: Chromatograms of Hydrocarbons Analyzed by GC vi

With default settings, the baseline estimation function is enabled, but the peak-baseline automatically
tracks the chromatogram plot line.

Consequently, basically the completely opposite settings to the settings described in Example: are
specified to induce fused peaks (straight peak-baseline corresponding to vertical division).
uV

100000000

90000000

80000000

70000000

60000000

50000000

40000000

30000000

20000000

10000000

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 min

1. Disabling Baseline Estimation (and Simultaneously Decreasing the Peak Baseline Height)

If [Vertical Division] is selected for [Baseline Type], the baseline estimation function is disabled and
[Peak Baseline Height] is set to 0.01 automatically.
(The [Tailing Off] checkbox is also enabled.)

10 i-PeakFinder
 Adjusting the Baseline

In this chromatogram, peaks are detected so continuously that the default [Detection Threshold]
setting of 5 results in almost all peaks being unseparated, so that vertical division results in a straight
peak-baseline.
uV

100000000

90000000

80000000

70000000

60000000

50000000

40000000

30000000

20000000

10000000

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 min

2. Obtaining the Desired Peak Profile

Given the excessive peak division in this example, adjust the division level.

To adjust the separation level for the group of peaks without destroying the unseparated status, use
the settings on the [Peak Unification] tab page ([Integration Advanced Settings] window) to resolve
the state of excessive division based on the characteristics between adjacent peaks.

For this example, [Unify Peaks by Separation Width] was set to 1.2, [Unify Peaks by Depth Ratio] to 1,
and [Unify Peaks by Shoulder Ratio] to 2.
uV

100000000

90000000

80000000

70000000

60000000

50000000

40000000

30000000

20000000

10000000

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 min

Caution: Perform a Trial to Confirm Results Are Acceptable Before Using the Peak Detection
Function to Unify Fused Peaks
Attempting to determine the division level using the peak detection function (reducing the peak
detection sensitivity by either increasing the [Detection Threshold] setting value or setting a
[Minimum Half Width] value) will result in baseline segments between peaks or at the end of the
group of peaks, so that the overall peak-baseline is bent or elevated.

It is important to understand that, unlike the peak unification function, which simply merges adjacent
peaks, the peak detection function separates the peak and baseline area from the plot line of the
chromatogram. (That results in unifying the peaks.)
(For an explanation of the differences between peak detection parameters and peak unification
parameters, see the next chapter.)

i-PeakFinder 11
Adjusting the Baseline

The chromatogram used for this explanation was adequate for unifying the peaks to achieve the level
of division shown above, but it can sometimes produce results like those shown in the example
below.
(For this example, the [Minimum Half Width] setting is 10.)
uV

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

Using Peak Integration to More Reliably Ensure the Peak-baseline Does Not Bend (for
Reference)
If the peak-baseline is bent in step 1, then the peak detection sensitivity value must be increased (by
decreasing the [Detection Threshold] setting), so that there is no separation between peaks.

Setting [Detection Threshold] to 0 results in all noise spikes, regardless of their width, being detected
as peaks and maximizes the likelihood of peaks being unseparated. However, to obtain more robust
results, specify the [Noise Estimation Range] range (on the [Peak Detection] tab page in the
[Integration Advanced Settings] window) to zero (from 0 to 0 minutes, for example).

Doing so disables the noise estimation function, with noise assumed to be at the minimum 1 µV level.
That minimizes the N value for [Detection Threshold], which is the S/N ratio threshold value, and
maximizes the sensitivity accordingly.

Of course, that tends to cause excessive peak division, but peaks can be unified again later,
according to the procedure described in step 2, above, or, for applications such as distillation GC
analysis, for example, temporary excessive division will not affect final results, because all the peaks
are eventually unified into one peak.

In this case, a [Minimum Half Width] setting of 50 was specified for the chromatogram in step 2.
(Instructions for using the [Minimum Half Width] setting is provided below.)
uV

100000000

90000000

80000000

70000000

60000000

50000000

40000000

30000000

20000000

10000000

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 min

12 i-PeakFinder
 Adjusting the Baseline

This page is intentionally left blank.

i-PeakFinder 13
Adjusting the Baseline

Chapter Endnotes
i
Unseparated peaks that satisfy certain specific conditions can also be processed with tailing process.

ii
If [Baseline Type] is set to [Vertical Division], then peak detection parameters are preset internally to
reduce the likelihood of forming baseline areas.

iii
This is referred to as valley-to-valley or base-to-base peak-baseline resolution. The software uses a
time program command name "Base to Base On/Off" to switch the base-to-base mode ON or OFF.

iv
Selecting the [Peak Baseline Height] checkbox on the [Integration Advanced Settings] - [Peak
Detection] tab page disables baseline estimation and enables baseline adjustment based on
specifying a peak-baseline height. In this example, the explanation assumes a setting of 5.

v
The [Detection Threshold] setting is also applied to bulges within peaks (noise and undulations), but if
the peak detection sensitivity is increased ([Detection Threshold] setting value is decreased) to detect
those increases as peaks, it could affect how the peak-baseline is drawn in some cases. For such
cases, it is also possible to increase only the shoulder peak detection sensitivity by adjusting the
[Detection Threshold of Unseparated Small Peak] setting (on the [Peak Unification] tab page in the
[Integration Advanced Settings] window).

vi
Basically the same approach is also used for LCMS, with the settings as described below.
 [Peak Baseline Height] = 0.01 (this setting also disables the baseline estimation function)
 [Detection Threshold] = 0
 [Noise Estimation Range] = Range of zero
However, because the [Tailing Off] checkbox is cleared in default settings, [Valley/Height Ratio] is set to 5,
rather than setting [Baseline Type] to [Vertical Division], which would collectively perform the [Tailing Off]
checkbox. In addition, several default settings are specified for newly created methods on the [Peak
Unification] tab page.

14 i-PeakFinder
3. Detecting and Unifying Peaks
Though already addressed in the previous chapter, this chapter describes when to use the peak detection
function and peak unification function and how to determine when to use those functions and peak filters.

3.1 Peak Detection

Peak detection is functionality used to separate chromatograms into peak and non-peak (baseline)
areas. Note that the functionality vii is also used to detect baseline areas.

Peak detection includes both functionality that uses information based on the signal intensity direction
(peak S/N ratio) and functionality that uses information based on the time direction (peak half height
width).

After detecting peaks, it automatically determines the starting and ending points of the peaks based on
the specified multiple of either the estimated baseline or estimated noise height, which are used to
determine the characteristic values of peaks and the remaining baseline areas.

3.2 Peak Detection Sensitivity and Estimated Noise Range

[Detection Threshold]
Given the S/N ratio used as the threshold value, this parameter is used to determine the baseline by
treating plot line bulges with a signal intensity greater than the threshold value as a peak and those with
a lower signal intensity as noise.

[Noise Estimation Range]

The noise value in the S/N ratio is automatically estimated by default, but this parameter allows the user
to fix the estimated noise level with this time range based on a user-specified plot line, which has noise.
(Normally it is baseline.)

Consequently, it can be set to a value more similar to the value the user desires, even when specifying
a [Detection threshold] setting.

Example: Chromatograms with High Noise Near Peak Tops and Excessive Peak Division
The [Noise Estimation Range] setting is normally set to target the highest noise range over the
baseline areas, but if the noise is higher for peaks than the baseline areas, then the range can
also be set based on noise over peak areas.

i-PeakFinder 15
Detecting and Unifying Peaks

Usually, the process can be performed based only on the [Detection Threshold] setting, but in this
case, due to the excessive peak division by noise within the blue box (where the peaks appear like a
red blur), a [Noise Estimation Range] setting can be specified for that area. サ 名
uV
140000

130000

120000

110000

100000

90000

80000

70000

60000

50000

40000

30000

20000

10000

5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0 85.0 90.0 95.0 100.0 105.0 110.0 115.0 120.0 125.0 min

The [Noise Estimation Range] was specified near the peak tops of the excessively divided peaks
(104.35 to 104.6 min).
To avoid the overall peaks from being detected as noise, the flattest possible point (with minimal
curvature) is selected.
uV

15000

12500

10000

7500

5000

2500

102.75 103.00 103.25 103.50 103.75 104.00 104.25 104.50 104.75 105.00 105.25 105.50 105.75 106.00 106.25 min

サ 名
uV
140000

130000

120000

110000

100000

90000

80000

70000

60000

50000

40000

30000

20000

10000

5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0 75.0 80.0 85.0 90.0 95.0 100.0 105.0 110.0 115.0 120.0 125.0 min

After looking down the overall chromatogram and confirming that the required peaks are detected,
make any necessary fine adjustments using the [Detection Threshold] or other settings.

Due to the narrow peaks in the beginning and end of this chromatogram and the broader peaks in the
center, attempting to resolve the excessive peak division using the [Minimum Half Width] or peak
unification function described below will affect other peaks. Therefore, if the procedure described
here is not successful, consider using the time program mode to process only the range required.

16 i-PeakFinder
 Detecting and Unifying Peaks

3.3 Minimum Half Height Width and Maximum Half Height Width

[Minimum Half Width]

This setting specifies that any plot line bulges narrower than this setting are not detected as peaks. As
the level of chromatogram smoothing is gradually increased, so that bulges become broader and lower,
this setting determines whether undulations satisfy the specified width, until their height eventually falls
below the [Detection Threshold] setting.

The function successively identifies each bulge that satisfies the specified conditions as a peak and
treats the rest as baseline. It functions similarly to the function that unifies adjacent fused peaks until
they reach the specified half height width, but it differs from the peak unification function that any area
that is not successively identified as a peak based on satisfying the specified conditions is treated as
the baseline.

No [Minimum Half Width] setting specified

uV

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

Bulges are successively identified as peaks based on a [Minimum Half Width] setting of 10 and any
remaining areas are treated as the baseline.
uV

13000

12000

11000

10000

9000

8000

7000

6000

5000

4000

3000

2000

1000

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 min

[Maximum Half Width]

This setting is used to identify any bulges wider than this setting as upward baselines, rather than
peaks.

If the baseline estimation function is functioning, it is possible to approximately track the baseline, but if
it does not track properly, non-peaks can be reliably changed to baseline by specifying a [Maximum
Half Width] setting based on assuming plot line undulations and bulges are peaks.

i-PeakFinder 17
Detecting and Unifying Peaks

No [Maximum Half Width] setting specified

mV
60

55

50

45

40

35

30

25

20

15

10

-5

-10

-15

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 min

[Maximum Half Width] = 20

mV
60

55

50

45

40

35

30

25

20

15

10

-5

-10

-15

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 min

Example: LCMS Chromatograms with Peak Profiles Divided into Multiple Segments viii
For LCMS, other than adjusting the [Minimum Half Width] setting, parameter settings for this
procedure are already configured by default, but in this case each process step is performed for
explanatory purposes.

For LCMS quantitative peak integration, profiles such as those shown in the figures below are often
obtained. Therefore, roughly the same approach is used as described above in
Example: Chromatograms of Hydrocarbons Analyzed by GC, except that in the GC example,
[Baseline Type] is set to [Vertical Division], assuming peaks are not processed with tailing process.
Therefore, other than specifying whether or not to process peaks with tailing, the LCMS settings are
the same as for GC.

18 i-PeakFinder
 Detecting and Unifying Peaks

 To maximize peak detection, set [Noise Estimation Range] to 0 to 0 min (which fixes estimated
noise at the minimum value of 1 µV) and set [Detection Threshold] to 0.
 By setting [Peak Baseline Height] to 0.01, detection of adjacent unseparated peaks (a straight
fused peak-baseline involving vertical division) can be induced, while also disabling the baseline
estimation function.
3750

3500

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

-250
3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

If multiply-divided peaks are presumed to be one peak, specify the half height width as the [Minimum
Half Width] setting.
3750

3500

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

-250
3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

3750

3500

3250

3000

2750

2500

2250

2000

1750

1500

1250

1000

750

500

250

-250
3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

Because i-PeakFinder does not plot a peak-baseline that intersects with the chromatogram plot line,
the bulge in the plot line near 4.49 min (on the far right) was not included in the peak. Nevertheless, it
resulted in detecting a single peak roughly in the desired location.

If detecting all bulges as a single peak and including it in the peak area value is desired, then use the
smoothing process. This can be understood by imagining a moving average, where the smoothing
process preserves the peak area value. Consequently, it can usually be used without any problem.

i-PeakFinder 19
Detecting and Unifying Peaks

In this case, [Method] is set to [Standard(Priority width)], [Counts] to 1, and [Width] to 10 in smoothing
settings.
1700

1600

1500

1400

1300

1200

1100

1000

900

800

700

600

500

400

300

200

100

-100

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

For this procedure, [Minimum Half Width] was set to 8 sec, which is the half height width determined
from this figure. However, because [Detection Threshold] is set to 0 (so that the peak is detected
regardless of the smoothing level), the same result can be achieved even if a larger [Minimum Half
Width] value (the approximate value that does not match the target peak profile) is set.

The setting simply needs to be large enough (the default setting for LCMS is 20) to prevent
accidentally including any separate peaks located in the same area. This approach can significantly
reduce the amount of trouble normally involved in specifying peak integration settings individually for
each target peak.

Note, however, that whether or not the cluster of bulges in the multiply-divided plot line should be
treated as one peak is determined as shown in the figures below.
3750

3500

3250

3000

2750 Peaks described above

2500

2250
for a concentration of
2000
0.05 ppb
1750

1500

1250

1000

750

500

250

-250
3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

1750000

1500000
Peaks for the same
1250000 components at a
concentration of 50 ppb
1000000

750000

500000

250000

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

When preparing calibration curves, if the [Minimum Half Width] value is approximated (on the high
side) based on peaks from a high concentration sample used as a standard sample for calibration
curve preparation, then calibration points on the low-concentration end can also be determined
accurately.

20 i-PeakFinder
 Detecting and Unifying Peaks

3.4 Peak Baseline Height

[Peak Baseline Height]

After i-PeakFinder detects a peak top, the starting and ending points are determined as the points
where the deviation of chromatogram at the rising and falling of the peak becomes less than the
specified multiple of standard deviations for the estimated noise (specified in [Peak Baseline Height]).

Therefore, chromatograms with high noise levels have the characteristic of higher peak-baselines and
chromatograms with low noise result in lower peak-baselines.

However, specifying this setting disables the automatic baseline estimation function.

In the chromatograms below, the noise level is extremely low, given the size of peaks, but actually
magnifying the peak base area shows tailing. Even though default settings result in correctly plotting the
peak-baseline, if the peak-baseline appears to be too long at that scale, it can be adjusted with the
[Peak Baseline Height] setting.

[Peak Baseline Height] = 5 (default)

uS/cm

11

10

-1

-2

7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 min

[Peak Baseline Height] = 200

uS/cm

11

10

-1

-2

7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 min

i-PeakFinder 21
Detecting and Unifying Peaks

Example: Ensuring Area Repeatability by Raising the Peak-baseline Slightly

If the baseline undulates, includes noise spikes, or otherwise does not stabilize at a given level ix,
specifying a higher peak-baseline that avoids those unstable areas can help ensure adequate area
repeatability by not including the unstable peak base area.
mA U
6 .5

6 .0

5 .5

5 .0

4 .5

4 .0

3 .5

3 .0

2 .5

2 .0

1 .5

1 .0

0 .5

0 .0

- 0 .5

- 1 .0

- 1 .5

- 2 .0

- 2 .5

- 3 .0
0 .0 0 0 .0 5 0 .1 0 0 .1 5 0 .2 0 0 .2 5 0 .3 0 0 .3 5 0 .4 0 0 .4 5 0 .5 0 0 .5 5 0 .6 0 0 .6 5 0 .7 0 0 .7 5 0 .8 0 0 .8 5 0 .9 0 0 .9 5 1 .0 0

In this case, the default [Peak Baseline Height] setting of 5 was increased to 20.
mA U
6 .5

6 .0

5 .5

5 .0

4 .5

4 .0

3 .5

3 .0

2 .5

2 .0

1 .5

1 .0

0 .5

0 .0

- 0 .5

- 1 .0

- 1 .5

- 2 .0

- 2 .5

- 3 .0
0 .0 0 0 .0 5 0 .1 0 0 .1 5 0 .2 0 0 .2 5 0 .3 0 0 .3 5 0 .4 0 0 .4 5 0 .5 0 0 .5 5 0 .6 0 0 .6 5 0 .7 0 0 .7 5 0 .8 0 0 .8 5 0 .9 0 0 .9 5 1 .0 0

Noise can cause variability in the status of fused or baseline-resolved adjacent peaks. Therefore,
corrective measures should be implemented accordingly, such as by specifying [Baseline Length] for
a [Baseline Type] setting, for example, that results in separation that is closer to baseline resolution
and by disabling detection for shoulder peaks and other factors that the presence of which tends to
be detected with relative variability.

In this case, the peak profile was stabilized by selecting the [Shoulder Peak Detect Off] checkbox and
setting [Unify Peaks by Depth Ratio] to 5 (explained in the next section), and by setting [Valley/Height
Ratio] to 1 for [Baseline Type] to adjust the peak-baseline in the baseline-resolved direction.
mA U
6 .5

6 .0

5 .5

5 .0

4 .5

4 .0

3 .5

3 .0

2 .5

2 .0

1 .5

1 .0

0 .5

0 .0

- 0 .5

- 1 .0

- 1 .5

- 2 .0

- 2 .5

- 3 .0
0 .0 0 0 .0 5 0 .1 0 0 .1 5 0 .2 0 0 .2 5 0 .3 0 0 .3 5 0 .4 0 0 .4 5 0 .5 0 0 .5 5 0 .6 0 0 .6 5 0 .7 0 0 .7 5 0 .8 0 0 .8 5 0 .9 0 0 .9 5 1 .0 0

22 i-PeakFinder
 Detecting and Unifying Peaks

3.5 Unifying Peaks

This function merges detected adjacent peaks (with some exceptions) into a single peak. This requires
that peaks are detected, so it must be used after peak detection.

Example: Chromatograms with Peak Distortion for LCMS Quantitative Peak Integration
Given peaks for the same components as described above in Example: LCMS Chromatograms with
Peak Profiles Divided into Multiple Segments, but with sample concentration increased slightly from
0.05 ppb to 10 ppb, the peak separation was not to the extent of being divided, but that tendency still
remained, making the peak profile not very well shaped.

For LCMS, the peak unification setting is specified by default, so that peaks can be processed even if
the [Minimum Half Width] setting is not appropriate.

In this case, the following explains how to use that approach by specifying each successive step.
500000

450000

400000

350000

300000

250000

200000

150000

100000

50000

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

Given the above, 0.8 was specified for the [Unify Peaks by Separation Width] x setting.
This unified the peaks within the specified separation factor, with the highest peak in the center.
500000

450000

400000

350000

300000

250000

200000

150000

100000

50000

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

Next, the [Shoulder Peak Detect Off] checkbox xi was selected (to switch OFF the function).

i-PeakFinder 23
Detecting and Unifying Peaks

This shows the shoulder peaks on the peak base area unified.
500000

450000

400000

350000

300000

250000

200000

150000

100000

50000

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

There are small peaks remaining on the right side of the peak base, but an enlargement of the peak
shows a valley, indicating that it is not a shoulder peak. That is why it was not unified in the process
above.

However, this area can be unified, if desired, by setting [Unify Peaks by Depth Ratio] to 20.
(The default setting for LCMS is 5.)
500000

450000

400000

350000

300000

250000

200000

150000

100000

50000

3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.0

24 i-PeakFinder
 Detecting and Unifying Peaks

3.6 Peak Filters

This function is used to eliminate detected peaks and return them to the baseline.

Just like the peak detection function, it can be used to retain large peaks and avoid detecting small
baseline bulges as peaks, but because peak filtering is used as the final step, after peak detection and
peak unification, simply deleting the peaks can prevent affecting the baseline confirmed up to that point.

Result of peak detection function Result of peak

サ 名 filtering function
mV mV

19
18
18
17
17

16
16

15
15

14
14

13
13

12
12

11
11

10
10

9 9

8 8

7 7
18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min 18.25 18.50 18.75 19.00 19.25 19.50 19.75 20.00 min

Example: Process for When a Downturn at the Peak End Point Due to Baseline Undulations
Causes a Long Narrow Region That Is Detected as a Peak
In some cases, baseline undulations can cause an area that is not a peak to be detected as a peak.
mV

15.0

12.5

10.0

7.5

5.0

2.5

0.0

-2.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 min

i-PeakFinder 25
Detecting and Unifying Peaks

To avoid effects from the negative peak near 1.7 min, set [Integration Range] from 2 min and set
[Maximum Half Width] to 10, so that bulges in the baseline are not incorrectly detected as a peak.
mV

15.0

12.5

10.0

7.5

5.0

2.5

0.0

-2.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 min

The [Maximum Half Width] setting is disabled if a bulge in the baseline does not have the
characteristics of a peak (i.e. does not have a half height width area). Consequently, broad short
areas can remain incorrectly detected as peaks.

In such cases, try setting the [Peak Filter] - [Min. Aspect] setting to a somewhat larger value, such as
1000 or 10000, and then determine the setting based on that result.

Unnecessary areas can be eliminated by setting [Min. Aspect] to 1000.

mV

15.0

12.5

10.0

7.5

5.0

2.5

0.0

-2.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 min

26 i-PeakFinder
 Detecting and Unifying Peaks

This page is intentionally left blank.

i-PeakFinder 27
Detecting and Unifying Peaks

Chapter Endnotes
vii This function is used to determine the peak-baseline by determining whether the tail of the main peak is
treated at baseline, with corresponding peaks treated as tailing peaks, based also on whether or not tailing
processing is specified, or whether the baseline is determined assuming the tailing peaks are simply fused
rider peaks.

viiiBasically the same approach is also used for LCMS, with the settings as described below.
 [Peak Baseline Height] = 0.01 (this setting also disables the baseline estimation function)
 [Detection Threshold] = 0
 [Noise Estimation Range] = Range of zero
However, because the [Tailing Off] checkbox is cleared in default settings, [Valley/Height Ratio] is set to 5,
rather than setting [Baseline Type] to [Vertical Division], which would collectively perform the [Tailing Off]
checkbox.
In addition, several default settings are specified for newly created methods on the [Peak Unification] tab page.
ix
Conversely, if the baseline noise is constant, and peaking tailing is long, lowering the peak-baseline height
can increase stability in some cases.
mV
5.0

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5

-2.0

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 min

mV
5.0

4.5

4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5

-2.0

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 min

x This is the only peak unification function that unifies peaks based on the resolution, but does not have a
prerequisite of peaks being adjacent.

xi To reduce the detection level of shoulder peaks, rather than not detecting them at all, the detection level can
be adjusted by selecting the [Unify Peaks by Shoulder Ratio] checkbox and specifying the tangent line
height ratio (in terms of percent) for shoulder peaks on the main peak. Conversely, it is also possible to
increase the detection level by resetting the [Detection Threshold] setting or increase only the shoulder peak
detection sensitivity with the [Detection Threshold of Unseparated Small Peak] setting (on the [Peak
Unification] tab page in the [Advanced] window).

28 i-PeakFinder