INTRODUCTION. The principles of which are to:

Download (0)

Full text



Peak detection and integration are fundamental tasks in chromatog-raphy, most often done using chromatography software. Enabling software to detect and integrate the peaks as desired (or as required by laboratory rules) is challenging and time-consuming. Common chal-lenges in peak detection include:

• Distinguishing peaks from noise

• Correctly identifying the underlying baseline

• Maintaining correct peak and baseline detection throughout a sequence of chromatograms

• Correctly handling rider peaks and other unresolved peaks Ideally, these challenges should be addressed using detection pa-rameters so that the same treatment can be applied across multiple chromatograms automatically, thereby minimizing variations introduced by different operators. However, finding parameter combinations that produce the desired results has often been a tedious process, causing many chromatographers to give up and resort to manual integration, which is subjective and labor-intensive.

In this presentation we look at a new peak detection algorithm that addresses these issues by separating peak detection from baseline determination. It uses second-derivative signal analysis, providing reliable and consistent peak detection even if the underlying baseline-shape changes. An easy to use interface that guides the user through the correct set-up of peak detection parameters is also shown. Examples of how the tools quickly solve problems with real-world chromatograms will be given.

With the new algorithm and assistant, chromatographers can get more accurate, reproducible results much faster and more reliably.

Taking the Pain Out of Chromatographic

Peak Integration

Taking the Pain Out of Chromatographic

Peak Integration

Shaun Quinn,1 Peter Sauter,1 Andreas Brunner,1 Shawn Anderson,2 Fraser McLeod1 1Dionex Corporation, Germering, Germany; 2Dionex Corporation, Sunnyvale, CA, USA

Shaun Quinn,1 Peter Sauter,1 Andreas Brunner,1 Shawn Anderson,2 Fraser McLeod1

1Dionex Corporation, Germering, Germany; 2Dionex Corporation, Sunnyvale, CA, USA


Cobra™ is Chromeleon® 7’s new peak detection algorithm that in

essence simplifies integration for all chromatographers. In designing Chromeleon 7, Dionex introduced the concept of Operational SimplicityTM . The principles of which are to:

• Minimize the number of steps needed to perform any task • Make all steps easy to understand and easy to use • Minimize time needed perform any task

Cobra is the new algortihm that adheres to these principles providing consistent and reliable peak detection across multiple chromatograms. With the introduction of Cobra, Chromeleon 7 provides the Cobra Wizard; an easy to use interface that guides the user through the correct set-up of peak detection parameters required by Cobra.

Define the integration range, specify the smoothing width, and then identify the smallest peak to be integrated. With these settings the Cobra peak detection algorithm instantly integrates the peaks of each chromatogram within a sequence accurately and concisely (Figure 1). In three easy steps Cobra has enough detail to integrate all chromatograms within a sequence.


2 Taking the Pain Out of Chromatographic Peak Integration 1. Define the integration range.

2. Select the narrowest peak.

3. Select the smallest peak to be integrated.

Figure 1. Cobra Wizard.

Cobra Peak DeteCtion algorithm

Reliable chromatographic quantification depends upon accurate and reproducible peak integration. Integration of chromatographic peaks determines the area under the peak, the height of the peak and the peak’s retention time. This information is used for all subsequent calculations, such as calibration or analysis of unknowns.

In simplistic terms, integration involves summing the detector output from peak start to peak end. To achieve this, integration is composed of two distinct events: it is started and then terminated. Peak start and end points define where integration commences and terminates and are identified by significant changes in detector output or by the rate of change of the detector output. Defining peak boundaries however is extremely challeng-ing. In an ideal world chromatographic peaks (Figure 2) would be perfectly symmetrical with pronounced detector signal changes that clearly identify peak start and end.

Figure 2. Idyllic chromatographic peak.

In reality, integration is extremely complex and diverse. Chromatographic data systems have to decipher many different variants and effects such as peaks of varying symmetries, overlapping peaks, valleys, varying sizes of peaks and size ratios, shallow peak rises and declines: extreme fronting or tailing, shifting apexes and valleys of unresolved peaks, baselines with large sloping background absorption and background noise, etc. (Figure 3).

Cobra is a new peak detection algorithm. It uses advanced signal processing to distinguish true peaks from noise and sophisticated curve fitting techniques to accurately locate peak maxima and inflection points.

Figure 3. Typical chromatography.

Cobra correctly integrates all types of chromatograms using just a few parameters.

A fundamental requirement of Cobra is having an optimal number of data points of the order of 10 to 40 to sufficiently characterize the chromatogra-phy detection signal of a component peak.

Cobra then uses an adaptive Savitsky-Golay smoothing function followed by transformation into a 2nd derivative, assessing the 2nd derivative against automatic thresholds based on signal characteristics.

Since the noise is amplified when differences are calculated, Cobra adopts an adaptive Savitsky-Golay smoother to remove the noise without losing valuable information.

The second derivative of a chromatographic signal can recognize compos-ite peaks. Assessing the second derivative of a peak also reduces the ef-fects of background and ensures that points of inflection and peak maxima are accurately and consistently identified if the background absorption has lower curvature than the analyte peak.

For example, the following two chromatograms show a peak of the same analyte, but one has a large sloping background absorption (Figure 4). The first derivative of these two signals is shown in the center (Figure 5). You can see that the difference between the pure analyte chromatogram (grey) and the chromatogram with interference (blue) is reduced. This ef-fect is considerably enhanced in the second derivative (Figure 6). In this case, the chromatogram of the pure analyte and the chromatogram with interference are almost identical.


Figure 4. Overlayed chromatograms.

Figure 5. First derivative.

Figure 6. Second derivative.

IDeNTIfyINg Peaks

In the second derivative of the chromatogram, noise thresholds are auto-matically determined (as represented by the red dashed lines in Figures 7 and 8 below). The local minimum below the thresholds lower limit is the peak apex. The points of inflection or local maxima above the thresholds are peak start and end (Figure 7). Baseline is interpolated between points where curvature crosses the upper noise threshold limit (Figure 8).

Figure 7. Local minimum and maximum.

Figure 8. Baseline interpolation. 3.5 1.0 –0.50 0.121 0 620 630 640 650 660 670 678 –0.18 7.0 1.28 –2.0 0.20 –0.1 209 300 400 500 600 700 730 –0.35 However, the baseline profile may not be a direct point to point interpola-tion if more than one peak is eluted on top of it. In a situaHowever, the baseline profile may not be a direct point to point interpola-tion with an unresolved peak group, analysts have several choices for determining the location of the chromatogram’s baseline profile. The most common options for drawing the baseline between two peaks: drop, valley, tangen-tial skim, exponential skim, and Gaussian skim.


4 Taking the Pain Out of Chromatographic Peak Integration

Figure 9. Baseline Profiles: 1. Drop Perpendicular, 2. Valley to Valley, 3. Tangential Skim, 4. Exponential skim, 5. Gaussian Skim

The drop method involves the addition of a vertical line from the valley between the peaks to the horizontal baseline. The vertical line is drawn between the start and stop points of the peak group. The valley method sets start and stop points at the valley between the peaks, thus integrating each peak separately. Skim procedures separate the small peak from the larger parent with separate baselines. The parent peak is integrated from its starting point to the apparent end of the peak group. The small peak’s baseline starts at the valley between the peaks and ends when the signal nears the baseline. The area “under” the skimmed peak is added to the parent peak, not the skimmed peak. This approach has been described also as a tangent integration method and the small peak variously labeled a skim, shoulder, or rider peak.

Several variations of the skim procedure are possible. Tangential draws a straight line from the valley to the end of the peak. Figure 9 shows an exponential skim baseline. An exponential function is used to create curva-ture in the skim line in an attempt to approximate the underlying baseline of the parent peak. Using an exponential function a curved baseline is drawn under the skimmed peak. Alternatively “Gaussian” skim tends to more accurately reproduce the Gaussian shape of the parent peak.

The complexity and errors increase as resolution decreases, and the valley of unresolved peaks shifts adding to a chromatographer’s woes.

Understanding these methods and techniques and being able to assign the correct integration parameters are extremely difficult tasks, even for the most experienced chromatographer.

smaRTPeaks INTegRaTION assIsTaNT

Chromeleon 7 addresses these issues, providing chromatographers with the simplest of interfaces designed in accordance with Dionex’s Operational Simplicity principles. SmartPeaks™ integration assistant produces desired results quickly and intuitively. The user simply selects a region of the chromatogram and SmartPeaks graphically displays available integration options, such as valley-to-valley baselines or exponential skims (Figure 10).

When the user selects an option, SmartPeaks adds the corresponding parameters and values to the processing method so that the desired in-tegration is automatically applied to all chromatograms in the sequence. This process takes just a few seconds eliminating the trial-and-error process of adjusting integration.


Passion. Power. Productivity.

North america U.S./Canada (847) 295-7500 south america Brazil (55) 11 3731 5140 europe Austria (43) 1 616 51 25 Benelux (31) 20 683 9768; (32) 3 353 4294 Denmark (45) 36 36 90 90 France (33) 1 39 30 01 10 Germany (49) 6126 991 0 Ireland (353) 1 644 0064 Italy (39) 02 51 62 1267 Sweden (46) 8 473 3380 Switzerland (41) 62 205 9966 United Kingdom (44) 1276 691722 asia Pacific Australia (61) 2 9420 5233 China (852) 2428 3282 India (91) 22 2764 2735 Japan (81) 6 6885 1213 Korea (82) 2 2653 2580 Singapore (65) 6289 1190 Taiwan (886) 2 8751 6655 Dionex Corporation 1228 Titan Way P.O. Box 3603 Sunnyvale, CA 94088-3603 LPN 2297-01 06/09 Chromeleon is a registered trademark and Operational Simplicity, Cobra, and SmartPeaks are trademarks of Dionex Corporation.

New INTellIgeNT PaRameTeRs

Integration is extremely complex and challenging but critical to accurate quantitative analysis. Chromeleon 7’s new detection algorithm Cobra not only simplifies the peak detection and integration process but also provides new adaptable and flexible parameters to assist chromatogra-phers achieve the best results. Minimum Relative Area is one such parameter. Minimum Relative Area rejects peaks, if their minimum area is below a given threshold. The minimum area is measured, on a first pass basis, relative to the sum of all peaks (total area) in a chromatogram.

An example of its use is for an analytical method that generates a chromatogram with peaks of interest and other smaller non-related peaks. The requirement is to total the peak response throughout a chromatogram. Typically to ensure that the smaller unwanted peaks are not included the minimum area detection parameter is set to exclude non-related peaks from being integrated. However within a sequence the analyte concentration can vary and the main analyte peaks may drop below the prescribed minimum area threshold. Using the Minimum Relative Area parameter peaks of interest that drop below the prescribed mimimum area will still be detected and integrated because they are greater than the minimum threshold of the current chromatograms total area (Figure 11).

minimum Relative area

Figure 11. Minimum Relative Area. Even though area of main peak drops below level of non-detected peaks in previous chromatogram, the main peak is still integrated.


Chromeleon 7 is a next generation chromatography data system that delivers intelligent functionality while maintaining operational simplicity. Critical data analysis is achieved accurately and reliably using powerful features of Cobra. The Cobra Wizard and SmartPeaks integration assistant aid and guide the chromatographer to perform the required data manipula-tions quickly, easily and reproducibly. Using Chromeleon 7, laboratories gain a significant boost in overall efficiency and productivity. Chromeleon 7 is Simply Intelligent.




Related subjects :