Checking for a software and IoT oriented solution that helps to monitor environmental conditions....
Public Lab is an open community which collaboratively develops accessible, open source, Do-It-Yourself technologies for investigating local environmental health and justice issues.
All topics »
If you cannot use the ReCaptcha to verify you are not a bot, use this alternative verification.
As an open source community, we believe in open licensing of content so that other members of the community can leverage your work legally -- with attribution, of course. By joining the Public Lab site, you agree to release the content you post here under a Creative Commons Attribution Sharealike license, and the hardware designs you post under the CERN Open Hardware License 1.1 (full text). This has the added benefit that others must share their improvements in turn with you.
sign up to join the Public Lab community
Forgot your password? Reset it here
by falbriard |
February 11, 2019 17:41 |
Checking for a software and IoT oriented solution that helps to monitor environmental conditions. Use spectrophotometry is one of those topics.
First, using a spectrometer in the visible range ,by itself, normally isn't possible. You normally have to add chelating agents to make the metals visible and adjust the ph . Chromium might be an exception to this, depending on the detection limits.
There are standard methods for detecting metals in liquids. Check with the epa for the specific method. But they are rarely continuous. Usually, they require sampling the water and adding reagents for testing. Many also require moderately expensive equipment.
@Ag8n , Thanks for the reply. There might exist different view angle, about the challenge and ways to tackle the problem. Your answer is right, mentioning the standard process for chemical analysis, which certainly gives more precision about composition and dilution of contaminants. At the other side, lets think about a one step process, simply by measuring a water sample with precision and loading its data into a digital infra. Two ideas here:
. If we add a mixture of chelants or other reagents prepared to match the known fingerprint of the contaminants, so that its spectral lines get visible in the VIS/UV spectrophotometer image?
. If we use more advanced algorithms for detection of emission & absorbance lines of chemical elements based on a database, much in a ways as practiced in astrochmeistry? Machine learning might also be helpful to raise precision.
Any thoughts on this how to improve the detection by use of contaminants in a simple IoT oriented process?
Is this a question? Click here to post it to the Questions page.
You might be able to find some thing. Let's take it step by step.
First, the only one of those metals that has any color in the Vis is chromium. Even then, you will probably need to get it to the same oxidation state. Cr(3) and Cr (6) absorb at different wavelengths. It might still be better to go with a cheating agent, even with chromium.
The public labs spectrometer is visible only. That's not to say you couldn't extent the range. But it will take some extra money to do. For exsmple, the lamps commonly used are deuterium lamps, which are quite expensive. Having said that, there was an article on public labs about using tesla coils as a plasma source for other ga s lamps. That might be a way to generate a source of uv light.
Now, as to the astro stuff. The common methods on earth are atomic absorption(AA), graphite furnace AA, or a variation on inductively coupled plasma (ICP). These are usually ICP mass spec or ICP optical emission. These are abbreviated ICP/MS and ICP/OES. They are quite expensive. But, prior to the fourier transform revolution, most instruments were. Now, FTIR instruments have dramatically reduced the cost of those instruments. Anything I can do to help, let me know.
Checking several chemical measurement processes, I found an active reagent that applies, it is named: "Diphenylthiocarbazone (or Dithizone)". It reacts with most of heavy metal ions, binding to Cu, Zn, Cd, Hg, Pb, Mn, Co, Ni. I believe that we will get a color indicator in the water sample that could be measured by colorimetry (or spectrophotometry by reflectance).
Not sure how it behaves under our condition at Rio Paraopeba, where we find multiple contaminants in the same water sample. Hopefully able to separate it using analytics comparing it with standard measurements by element.
What is your opinion, does this idea of using this chemical reagent sound like a feasible solution. Some documentation I found here [ https://core.ac.uk/download/pdf/78563164.pdf ]
Looking forward to hear your opinion.
Reply to this comment...
Log in to comment
I'm downloading it now. Give me a day or two to go through it. Thanks!
I've used a version of this method. We used it for lead and just for visual comparison. There was a link for it but it wouldn't work. It was the USP, monograph 251, in the 2012 edition. if you Google it, it should show up.
From the operator standpoint, it took a fair number of reagents and operat or time. It's a good starting point!
More work to do on my end.
First off, spectrometers are much more common than spectrophotometers, these days. And cheaper too.
T he method you referenced relies heavily on liquid/liquid extraction. So you will need separately funnels ( and please read up on how to vent them), a hood for using the chloroform, as well as some good gloves that are chloroform resistant. The method also used some cyanides, which are hard to get in the US, although the paper lists other alternatives.
For a first pass, sample the water and let the residue settle out. You may have to let it sit overnight. If you still have residue, you can filter the water through a coffee filter.
The base outline of the method is water sample in Sep funnel. Do multiple seps with dithizone(save dithzone) until dithizone is green. Then you do the saved dithizone / water reagent Sep funnel extract ( again remember to vent). This time, save the water reagent. Then it's the water reagent into the proper dithizone reagent in a Sep funnel and shake again. This time save the dithizone. It should be discolored. For lead, they say the color is violet. To me, it was more like a deep blue.
With proper pH and buffer, you can probably do more than one element at a time. Then the spectral data processing takes place. But it takes a lot of time to get there! Please double check me, but I think that's correct ( from USP 251).
You might want to look into ANDalyze. They use a new technology based on DNA for heavy metal analysis in water samples. I'd need to see validation data and a bunch of other data to be convinced, but it could fill your needs.
@Ag8n , thanks so much for support and answers. Meanwhile I also found a summary of reagents and methods based on UV/VIS spectrophotometry [ https://www.hach.com/dr6000-uv-vis-spectrophotometer-with-rfid-technology/product-parameter-reagent?id=10239244800 ]. Counting with support of a University in São Paulo I hope we can focus on a few, but important process methods for heavy metal detection. Most important is to keep the cost down for the kit and have portable spectrophotometer equipment to collect samples at the remote locations, recording the data in our central database, using proper reagents and doing absorption readings. Thanks for the feedback about the ANDalyze Fluorimeter process, thats a new instrument kit with a very different method. It looks interesting.
This is part of: