Home	The author of this page is independent and has no commercial intention in mind !			Diese Seite in Deutsch.

Introduction

Type: Photo Diode Array Spectrophotometer / Photo Diode Array Spectrometer   HP/A 8453

What features make this instrument so famous, excepting it's very affordable prize ?^{[2, 3]} :

1. The combination of the best polychromator and the best diode array technology in a single equipment:
   •  The best polichromator for diode arrays you can by, is coming from “Carl-Zeiss”
      (It's not enough to have as many lines as possible on the light dispersive element! There is a lot more to do!)
   •  The best diode array technology you can get from “Hewlett-Packard/Agilent”.
      (the CCD technology for diode array of other suppliers have several disadvantages: ghost images and so on.
       And: Even Photo Diode Arrays are by far not Photo Diode Arrays at all, - too!!)
   •  The optimal and the absolutely stabile arrangement of all optical components on the so called "optical bench"!
2. An incredible stability of the whole instrument and it's electronic:
   (Already the hp 8452A showed us this unbelievable values, but it had a certain sensitivity to temperature changes)
   •  Wavelength  stability far better then ±0.25nm over 5 and more years!
   •  Absorbance stability far better then ±0.30 %  over 5 and more years!
   •  A very high stability against temperature changes!
      (That doesn't mean, your samples will have the same insensibility to temperature!!!!!)
3. Real measured values are much better then specified by “ Hewlett-Packard/Agilent”:
   - Specifications 8453 UV-Visible Spectrophotometer: _{(by HP/Agilent):} 
   

   Wavelength range	190-1100 nm
   Slit width	1 nm
   EP resolution test	>1.6         toluene in hexane, ratio abs. at 269 nm/266 nm
   Stray light	<0.03 %    at 340 nm (NaN02, ASTM) <0.05 %    at 220 nm (Nal, ASTM) <1 %         at 200 nm (KCI, EP)
   Wavelength accuracy	<±0.5 nm   - 0.5-second scan (NIST 2034) <±0.2 nm     (at 486.0 and 656.1 nm)
   Wavelength reproducibility	<±0.02 nm   - ten consecutive scans (NIST 2034)
   Photometric accuracy	<±0.005 nm   at 440.0, 465.1, 590.0, and 635.0 nm,  at 1.0 AU (NIST 930e)
   Photometric noise	<0.0002 A   - sixty 0.5-second scans at 0 A, 500 nm, rms
   Photometric stability	<0.001 A     - at 0 A, 340 nm after 1-hour warm up, measured over 1 hour, every 5 seconds, constant ambient temperature.
   Baseline flatness	<0.001 A     - 0.5 second blank; 0.5 second scan; rms.
   Typical scan time	1.5 seconds  - full range.

   (Some specialised spectroscopists will tell you, that at least the values for stray light are not acceptable, as they mean, that the instrument has only a useable linear rang from 0.0001 to 2.0 AU's and every scanning spectrometer will reach a linearity till 4.0 and more AU's [specially if you use a double monochromator!]. Be told, these are not specialists at all! I will argue a little bit later)
   
   
4. The benefit to have a whole spectrum in "no time" —
   A spectrum is telling you more, than thousand single wavelength measurements!: Before you have really started your scanning instrument, you have measured the whole spectrum on the diode array spectrometer. It is absolutely documenting, that you have never thought about, when you are telling, that you are able to record a spectrum on a scanning spectrometer with 900 nm/minute^[2]:
   • First compare your spectrum recorded with 900 nm/min. with the "same" spectrum recorded with 10 nm/min.
   • Second, record your spectrum after a year again with 900 nm/min. and compare the spectra you got.
   Opposite, we are comparing spectra we have measured 15 years ago with spectra taken yesterday on the same instrument, and they fit in a manor, that you cannot distinguish one from the other, also not with the help of a computer (and there was never a service necessary all these years!)!
5. The wavelength resolution of 1 nm with a slit width of 1 nm is never the end of the reliable use of this instrument:
   • First, you can not measure the slit width of a diode array in the same kind as you will doing it on a scanning instrument!
     The opposite is true, that you must first find a new, more adequate definition of “slit width” for this type of instrument! No — I will not give my definition yet, I will first wait for the discussion about it.
   • As also the wavelength resolution needs a new definition, At latest yet, It will be absolutely clear, that this type of instrument is in no way, comparable with one of the scanning family!
   • Not enough, also the stray light needs to be new defined. For it, I will exceptionally give you my newer definition, as it is published in the paper: “New System Suitability Test for diode array spectrophotometers”:
     “Stray Light is light inside the polychromator/diode array housing of any wavelengths other than the one on which the particular diode is placed for measuring”. (It may be also a hint for the other missing definitions, too.)
   • Now as you are well introduced, you will understand, that the slit width is narrower as given in the specifications, and the light is much more divided as you believed. So it is now allowed to interpolate values on wavelength between the real measured ones, provided that you are correctly using a reliable interpolation algorithm!
     (We found, that it is easy possible to interpolate until 0.1 nm steps, without generating information, which is not there!)
6. Doing IQ, OQ, PQ, PV, and a real System Suitability Test (SST) every morning in less then 10 seconds:
   It needs more time to sign the printout of the SST and collect it in the document folder of your instrument, as time is required to carry out the SST. With the new designed SST you have not only validated the instrument for routine use, you have also validated — every morning — your computer and the program and the algorithms you are using during the day!
   The only 'thing' that cannot be validated by the SST, is your “specialist”, “your spectroscopist”   !!!!!!!!!!
7. With this instrument's precision, you can real accomplish reliable chemometrics^[5]:
   Forget it once and for all, this papers talking about chemometrics and reports at the same time only an absolutely minor precision and reproducibility! With this instrument, and the correct  ”celebrated” spectroscopy, you solve tasks, with a reliability and precision, that seems to be impossible to you before:
   •  Adjusting the stoichiometry of your production vessel, like using a micrometer and a scalpel at
      the same moment!
   •  Detecting the endpoint of your reaction in the production vessel, as if you are looking through a microscope!
   •  You will get the possibility, to produce the even higher quality as ever before, and
      then analyse your quality, and keep the high quality standard of your product over 15 and more years!
   •  Simplifying your quality control task and make it more robust and reliable at the same time!
   •  Have you ever thought about validating your measurings in the kind of
      Sample-Method Validation (SMV) and Sample-Reference Identification (SRI) ^[6]?
   •  Identify your products at the same moment, when you determine their quality and content^[6]!
      That means, as example, doing 'CU', 'Dissolution ratio', etc. in the pharmaceutical field, and
      identify your product for USP, EP, etc. in one task and at the same moment, also and special,
      if your product contains more then one active ingredients!
   •  All what is told about the production vessel, is also true for your "pharma lab", dyestuff mile, etc.!
   •  Detecting unknown components in your reaction, identify them by their
      real spectrum, and calculate their concentration, too!
   •  . . . . .
   •   Simplifying IQ, OQ, PQ, PV and SST, and the paperwork for it!
   •  . . . . .
   •  I can expand this list over several pages with my examples !!
      But I have not time to do it right now. So please be patient, till more will arrive.
8. Let it be enough for the moment. I will expand this list also, if I have time to do so.

What have you to do to get all these benefits :

1. It is true, that the linear range of this diode array spectrophotometer will not far exceed the value of 2.00 AU's, because the relatively high stray light! So you are not able to measure reliable spectra, which raises above this limit!
2. In more than 99.99% of all cases, you are not observing a wavelength where you will read an absorbance of, let say, 4.1 AU's and an other wavelength with an absorbance of approximately 0.10 AU's at the same time. Why was you then buying a double monochromator spectrophotometer with such an expanded linearity and such a nice price? Think about, that you always have to change the wavelength between the both observations (not only a few nm), and what for a wavelength reproducibility your instrument has, and what this means to the correctness of your results.
3. You think it is nice to have this linear range, because you have then not so often to make an additional dilution, and so you can save a lot of solvent? Think about, “Are you not feeding the horse at it's wrong end?” I think so! I don't believe, you are an owner of a “super-micro extra ordinary analytical balance”, so you must normally prepare a so called 'stock solution'^[7] and then you make manually a known dilution, hopefully one that will fit the linear range of your spectrophotometer. Is it not more wise to think about a device, that is doing the dilution for you and using much less solvent as you do, when you are preparing the dilutions by hand?
   Now it is easy to make an additional dilution, and still saving a lot of solvent, anyway.
   This intelligent device is called “ Dynamic Dilutor”^[8], and 'he'^[8] fills the cuvettes for you, for no extra price!
   Even if my page: "Dynamic Dilutor" is not coplete yet, you may be interested in the possible results and in its fastness on the Linearity^{[8, 9]} page !
4. Are you thinking it is nice to have a large linearity range, because you are still filling the cuvettes for your spectrophotometer as it already Mr. A. Beer (1825 - 1863) and Mr. J. H. Lambert (1728 - 1777) did, manually, or even worse, are you using a peristaltic pump or a water-jet vacuum pump for filling/cleaning? Then you also didn't gain very much in doing spectroscopy at all, and you are frequently using the first till the fifth derivation on your recorded spectra to evaluate the concentration of your samples. Why are you always running the reference spectrum just before or just behind your samples?
   If you don't have the many for a “ Dynamic Dilutor"^[8], then you can fill your cuvettes with the “ Bar-Sipper”!
5. Do you need more arguments? - Please send me an e-mail. I will answer you as soon as possible!
6. As you have seen, there is no need to have a linearity range till 4.0 AU's, also if it would be nice to have. There is also no need to have a double monochromator spectrophotometer at all, except it's price! For the price of such an expanded linearity range you can buy the diode array spectrophotometer, including the cuvette changer, and almost the dynamic dilutor, too. And think, we have not started yet, to talk about all the “nice” accessories you have for your comfortable spectrophotometer!
   If you are doing as I told, you will have, of course calculated back from concentration to linearity, a linearity range^{[8, 9]} of your new spectrometer combination of over 20'000 AU's, on your finger tip!
   And that's realy meaning: on one of your finger tipps.
   And please don't think abaut, that behind AU's only logarithmic functions live!

Why were the constructors and the designers not knowing, what they have build for an instrument ?:

• When you have read the paper “New System Suitability Test for ....... (SST) ”, you will be able to answer this question by yourself, when you are reading the list of the features above.
• For all other I try to explain: If you have the features above in mind, specialty the required definitions, and now you take a look at the Agilent page with all the publications for DQ, IQ, OQ, PQ, and all the other test manuals they have written:
  

  Part Nr.:

  0.9957; 0.8351/1; 0.8351/2; 0.8169/1; 0.8169/2; 0.7967/1; 0.7967/2; 0.7967/3; 0.7967/4; 0.6967/5 0.7742/1; 0.7742/2; 0.7742/3; 0.7742/4; 0.7742/5; 0.7742/6; 5988-9647EN; 5963-5615E;

  and:
  1. )   Design qualification (DQ)
  2. )   Installation qualification (IQ) and operational qualification/performance verification (OQ/PV)
  3. )   Method validation
  4. )   Performance qualification (PQ)
    and all the additional written manuals, as for example:
      “Operational Qualification / Performance Verification for HP 8453 UV-Visible Spectrophotometers” and
      “Fundamentals of modern UV-visible spectroscopy (1996)” and so on.
  Now as you are knowing, that in all this manuals, still the traditional tests are applied to the instrument, and introduced to the customer, it cannot any longer be believed, that they were knowing what special instrument they have build.

Peter Forster: