DAD Reference Wavelength: do we need a separate channel?

Discussions about HPLC, CE, TLC, SFC, and other "liquid phase" separation techniques.

18 posts Page 1 of 2
I find the article on Reference Wavelength a little confusing. It talks about PDA/DAD software having to subtract another part of the spectrum, but it mentions that we need a separate channel:

Set up the detector to collect a second channel of data (2nd wavelength signal)
But DAD/PDA already gather the whole spectrum. Shouldn't this sentence refer only to MWD/SWD/VWD detectors?

And if so, this would mean that in case of DAD/PDA, this CDS feature is just software-based - nothing in the detector itself does anything special. It's just when extracting a particular wavelength chromatogram, we subtract one part of the spectrum (averaged) from another part of the spectrum (averaged).
Software Engineer at elsci.io (my contact: stanislav.bashkyrtsev@elsci.io)
If you used a DAD, you probably know that the reference wavelength is set as an option of a particular signal (channel), which is stored by the software, not as a separate signal (channel). If you choose to store the spectral data, it is generally possible to extract any signal (with or without a reference) from these data.

The DAD and MWD (Agilent) are physically the same devices. The only difference is the software. The DAD has extended software features that allow you to store and use the whole wavelength range. The MWD allows you to store only a limited number of signals.
vmu wrote:
If you used a DAD, you probably know that the reference wavelength is set as an option of a particular signal (channel), which is stored by the software, not as a separate signal (channel)
So what you're saying is that Reference WL can be set for each WL we want to ask the software to extract, right? Which means we're using DAD as MWD (even though we could optionally store spectra as well).

vmu wrote:
The DAD and MWD (Agilent) are physically the same devices. The only difference is the software.
If we look at the docs on Agilent 1290 Infinity II Variable Wavelength Detector, it doesn't seem to be true:
Optical System Overview section wrote:
Wavelength selection is made by rotating the grating, which is driven directly by a stepper motor.
Which sounds like a typical design for MWD. PDA/DAD don't have rotate the grating.
Software Engineer at elsci.io (my contact: stanislav.bashkyrtsev@elsci.io)
sbashkyrtsev wrote:
So what you're saying is that Reference WL can be set for each WL we want to ask the software to extract, right? Which means we're using DAD as MWD

Yes.

sbashkyrtsev wrote:
Agilent 1290 Infinity II Variable Wavelength Detector...

The VWD is a different device. It has only one light sensor. It can acquire the signals at different WLs only in sequence. A motor rotates the diffraction grating to select the WL that reaches the light sensor. The VWD switches frequently from one WL to another WL and back during the run when you set the VWD to acquire the signals at the two WLs.

DAD and MWD are designed identically. The diffraction grating is motionless. All the WLs are detected simultaneously, each WL by a single photodiode in the array. The difference is only in the software.
No, no, no..... You have the terms and concepts all mixed up.... We are not talking about scanning spectra or spectra in general. "IT DOES NOT SUBTRACT another part of the spectrum "! The DAD/PDA collects single wavelength signals (several discreet signals) and optionally can store entire spectral ranges too. NO extraction of any signal. Not using "DAD as MWD". We are looking at two separate wavelengths, one main signal.

I find the article on Reference Wavelength a little confusing. It talks about PDA/DAD software having to subtract another part of the spectrum, but it mentions that we need a separate channel:

Set up the detector to collect a second channel of data (2nd wavelength signal)

But DAD/PDA already gather the whole spectrum. Shouldn't this sentence refer only to MWD/SWD/VWD detectors?

And if so, this would mean that in case of DAD/PDA, this CDS feature is just software-based - nothing in the detector itself does anything special. It's just when extracting a particular wavelength chromatogram, we subtract one part of the spectrum (averaged) from another part of the spectrum (averaged)
.


The "Reference Wavelengthh" feature IS a built-in SOFTWARE feature. A DAD/PDA is an optical TOOL, the software programs it so we are always referring to the setup and use of the DETECTOR. DAD setup and use is an advanced topic so I encourage to study up on how settings offered can change the data you collect as part of your method.

REFERENCE WAVELENGTH is a feature specific to a SIGNAL CHANNEL, not a spectrum. A CHANNEL is a single wavelength, BUT all "single" wavelengths are in fact a range of many wavelengths. This is what BANDWIDTH defines, the range of the wavelengths which DEFINE the selected wavelength of analysis.

IF TURNED ON, the "Reference Wavelength" (note, WAVELENGTH = channel) will collect data at a second wavelength and bandwidth that YOU specify, then subtract all of the data collected, at that wavelength, from the original one you selected in your method. It destroys the original signal and the reference signals to show a third signal as the result (failing all known validation protocols!). To overcome this disaster, do not use it. However, the concept can be useful sometimes for analysis, but you do not want to carry out the analysis by automatically destroying data. All you need to do is to select your original wavelength as the first channel, signal #1, then choose a second related wavelength (that does not interfere) as the second channel, or signal #2 (TWO Separate wavelengths to collect). When done, you will have all of the original data for processing. You can not perform baseline subtraction using the second signal to create a third signal for processing. This will leave you THREE separate signals. No data is destroyed, everything is documented and the no special software needed.
vmu wrote:
The VWD is a different device. It has only one light sensor. It can acquire the signals at different WLs only in sequence. A motor rotates the diffraction grating to select the WL that reaches the light sensor.
...

DAD and MWD are designed identically. The diffraction grating is motionless. All the WLs are detected simultaneously, each WL by a single photodiode in the array. The difference is only in the software.
I see.. I was sure that VWD and MWD are synonyms like DAD and PDA. But I was wrong. I looked into Thermo Fisher - they use a similar marketing (?) strategy and allow upgrading from MWD to DAD.

To summarize, we have these synonyms:
- VWD/SWD (variable or single wavelength detector) or even just UV-detector (though this term is quite overloaded). Acquire only particular WL at a time. But allow switching between multiple ones.
- DAD/PDA
- MWD which is physically the same as DAD/PDA, but the software throws spectra away.

Multidimensional wrote:
The DAD/PDA collects single wavelength signals (several discreet signals) and optionally can store entire spectral ranges too. NO extraction of any signal. Not using "DAD as MWD". We are looking at two separate wavelengths, one main signal.
Not sure if I follow, but it seems like you're saying that when we ask DAD to gather signal on a particular WL, it gathers that WL separately from the spectra? If so, I don't think this is true - DAD cannot physically measure a single wavelength like VWD does. So it must be extracting it from the spectra.
Multidimensional wrote:
It destroys the original signal and the reference signals to show a third signal as the result (failing all known validation protocols!).
Yes, but the data is destroyed only if we don't gather spectra. If we do, then we can always reconstruct the "original" signal after the fact.

Let me know if I misunderstood you.
Software Engineer at elsci.io (my contact: stanislav.bashkyrtsev@elsci.io)
Please note that different manufacturers use different system and names, but most of the concepts are true for all of them. Yes, most brands that offer a MWD and DAD use the same hardware to do so and just change the firmware and software to enable/disable features (via firmware). BOTH can store individual wavelength signals (multiple). *You seem very confused about what wavelengths are (aka: discreet signals, channels).

You wrote: "To summarize, we have these synonyms:
- VWD/SWD (variable or single wavelength detector) or even just UV-detector (though this term is quite overloaded). Acquire only particular WL at a time. But allow switching between multiple ones.
" - <> Some, but not all brands do that. For example, Agilent's VWD can actually gather TWO separate wavelengths at the same time (unique and useful for creating a second reference).
You wrote: - "MWD which is physically the same as DAD/PDA, but the software throws spectra away".

<> Not really, most MWD and DAD (PDA) use diode arrays for detection so depending on how the firmware is configured, they can acquire (scan) an entire range of discreet wavelengths (i.e. 210-600nm) for the DAD (PDA) OR store several discreet wavelengths (i.e. 210[10]; 254[8]; 280[10] etc, often 5 to 8 wavelengths. Each discreet signal is stored alone, no spectra is collected (no need to). THE DAD (PDA) is unique in that it can both collect individual signals and optionally scan a range of wavelengths (these are separate data sets). THE MWD can not "scan" so it can only store the multiple discrete wavelengths. This is easy to do using the available array of diodes within the bandwidth of each selected wavelength. Again, no need to"extract". Please understand that ALL signals are not actually one wavelength, but a range of signals collected. This is why bandwidth must always be stated when you specify a wavelength.

Multidimensional wrote:
"It destroys the original signal and the reference signals to show a third signal as the result (failing all known validation protocols!)."

You Wrote: Yes, but the data is destroyed only if we don't gather spectra. If we do, then we can always reconstruct the "original" signal after the fact.

<> Yes, you misunderstood. Gathering Spectra has nothing to do with this topic. As noted before, the DAD/MWD actually acquires discreet wavelength signals that the user selects in the software. It is these discreet signals that can optionally have the "Reference Wavelength" feature activated (which is by itself, yet wavelength[bandwidth] signal). No scanning. We re discussing WL alone.
Multidimensional wrote:
For example, Agilent's VWD can actually gather TWO separate wavelengths at the same time (unique and useful for creating a second reference)[/i][/b].

No, it can't.
Yes, it can. In stopped flow mode you can have it store a reference signal, then run a FULL SCAN of the wavelength range entered. Pretty neat feature.
Multidimensional wrote:
In stopped flow mode you can have it store a reference signal, then run a FULL SCAN of the wavelength range entered.

The stopped flow mode is not a chromatography mode. The words "then run..." contradict the phrase "can actually gather TWO separate wavelengths at the same time".
vmu wrote:
The VWD switches frequently from one WL to another WL and back during the run when you set the VWD to acquire the signals at the two WLs

This is similar to a quadrupole MSD acquiring several signals for different m/z in a SIM mode. Actually, Agilent's VWD can store only one signal, either at one selected WL or at different WLs for different time segments of a chromatography run. Thermo's VWD can store several separate signals, which are acquired as described above.
sbashkyrtsev wrote:
DAD cannot physically measure a single wavelength like VWD does. So it must be extracting it from the spectra.

DAD can physically measure a "single" wavelength". Each "single" WL has its own photodiode for this purpose (or a range of neighboring diodes (averaged absorbance) for a "single" WL with the bandwidth of several nanometers). If you store the raw data from a relatively wide range of photodiodes in the array, you store the spectrum. Any "WL" (i.e., a spectral band of some width Y nm with a maximum at X nm) can principally be extracted as a separate signal from this spectrum. [This is similar to the extraction of a single-ion chromatogram from an MSD chromatogram obtained in a scan mode; however, the DAD doesn't need to "scan" since it can see all WLs in the spectrum simultaneously (in contrast to the VWD, which really "scans")]. At least OpenLab software can do such an extraction with the Agilent DAD. Moreover, the signal can be extracted with a reference (i.e., simply the signal at X nm (bandwidth Y nm) minus the signal at N nm (bandwidth M nm)).
Multidimensional wrote:
they can acquire (scan) an entire range of discreet wavelengths (i.e. 210-600nm) for the DAD (PDA) OR store several discreet wavelengths (i.e. 210[10]; 254[8]; 280[10] etc, often 5 to 8 wavelengths. Each discreet signal is stored alone, no spectra is collected (no need to). THE DAD (PDA) is unique in that it can both collect individual signals and optionally scan a range of wavelengths (these are separate data sets). THE MWD can not "scan" so it can only store the multiple discrete wavelengths. This is easy to do using the available array of diodes within the bandwidth of each selected wavelength. Again, no need to"extract".
You assume that collecting a WL in DAD happens separately from acquiring full spectra. May I ask what's the source of this information? I went through Agilent and Waters operators guides on DAD - none of them mention how they do it. So either you have some insider knowledge, or you're just assuming - same as me :)

It's entirely possible that DAD always gathers full spectra, then digitally extracts the WL + BW that we were interested in and then removes spectra. And I have several reasons to believe this is actually the way:
1. I have Agilent DAD files, where WL signal is stored separate in files dad1A.ch, dad1B.ch, etc. And none of these files contain Retention Time - just intensities. It means RT is stored somewhere else - and the only such place is the spectra file (dad1.uv). This fact alone means that there was no separate measurement.*
2. I also have Themo MWD files. Multiple "channels" are written with their Retention Times - some of them are identical. They don't differ even by a microsecond. Again this means there was only 1 measurement for some wavelengths. There are other channels that could differ though, but maybe it's possible to set up that different measurements save different WLs.
3. Here's a quote from Waters DAD docs:
The data the detector reports to the database (Empower or MassLynx) can be the average of a number of data points. After calculating absorbance, the detector averages absorbance values based on the requested spectral resolution.
This kinda implies that there's a stage of gathering data from diodes, and then additional post-processing depending on user settings. But this quote is quite weak, you can't deduce much from it.
4. If spectra is gathered anyway, it doesn't make any sense to make yet another measurement. That would simply gather a subset of the same data. So why measure the same thing twice?
5. Another a priory thought: it'd make the hardware design more complicated. While it's usually easier to make such things on the software side.

vmu wrote:
DAD can physically measure a "single" wavelength". Each "single" WL has its own photodiode for this purpose (or a range of neighboring diodes (averaged absorbance) for a "single" WL with the bandwidth of several nanometers).
Agree.. But still the fact that in principle it's possible - doesn't yet mean that this is what they're actually doing.

*It's also possible that someone extracted these after the fact - ChemStation allows it, but I don't know if the file format is the same. If you give me an example of a file with DAD + channels that you ran, I can check it as well.
Software Engineer at elsci.io (my contact: stanislav.bashkyrtsev@elsci.io)
18 posts Page 1 of 2

Who is online

In total there are 36 users online :: 0 registered, 0 hidden and 36 guests (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: No registered users and 36 guests

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry