Spectrophotometer - Determining Copper Concentration

Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

kadriver

Well-known member
Joined
Oct 25, 2010
Messages
1,830
Location
United States
I've got a new toy.

It is a Milton Roy spectronic 20D spectrophotometer.

I bought it for experimentation purposes, and to try and find the concentration of dissolved copper in my silver cells.

It came with a generic manual, but I am clueless on how to use it - youtube has been helpful.

I need to make standard solutions to set up a graph of KNOWN concentrations of copper.

I will then use this graph as a refernece to determine UNKNOWN concentrations of copper from samples of electrolyte drawn from my silver cell.

I am thinking that the standard solutions will need to be prepared using virgin crystal-clear silver nitrate electrolyte - 100 grams of pure silver dissolved in 1 liter of liquid. Then adding a solution of a known quantity of dissolved copper to the electrolyte - but in what proportions?

Or, can I just use dissolved copper in solution and not bother with the copper being in the electrolyte.

I am stumped on how to make these standard solutions, has anyone else ever tried this experiment?

Thanks - kadriver
 

Attachments

  • Spectrophotometer57.jpg
    Spectrophotometer57.jpg
    310.7 KB · Views: 522
well :D

if you dont have tabs with highest absorbance of those electrolytes, you have to make a characterisation

you will measure absorbance- adjust your machine to this

make a standard solution of AgNO3 ( best way would be use similiar concentration as you use in cell electrolyte )
now you have to find out spectral characteristics - measure absorbance in whole spectra in cca 10 nm steps ( dont know spectral properties of your spectrophotometer)
now make a curve and find out the wavelenght with highest absorbance ( agno3 is colourless, i think peak is in UV, not in VIS )

prepare standard solution of Cu(NO3)2 and repeat spectral characteristics

now you have to find out the maximum absorbation wavelenght shift ( because of mixing two electrolytes )

mix those two electrolytes - make maximum Cu 2+ concentration that is allowed for silver cell
make spectral characteristics- you get the wavelenght of maximum absorbance of this mixture ( if it will be out of scale, use lower conc of Cu 2+ )

adjust your spectrophotometer to a wavelenght with maximum absorbance of this mixture

prepare few standard solutions with similiar Ag+ conc. and raising Cu 2+ concentration ( 6 different sols would be enough)
measure at adjusted wavelenght
as a zero value- blank solution use pure agno3 sol
measure all your prepared sols and make a graph: X axis- concentration , Y axis - measured absorbation( in excel, make a graph, absorbation - concentration is linear function - use linear function calculation to find out the line equation - this is very important to know because with this equation you will calculate the concentration)

// agno3 is decomposing at light, this has to be supressed in some way :lol: , solutions has to be crystalclear with no turbidity or precipitate
// for very low concentrations of Cu 2+, use ammonium complexes of Cu - much more colourful in very low concentrations

maybe i forgot something, it is time ago i worked with spectrophotometer :roll:

EDIT : check something about it here

http://teaching.shu.ac.uk/hwb/chemistry/tutorials/molspec/beers1.htm
 
Wow! Thank you very much.

Using fresh AgNO3 to zero the machine is very helpful.

All good stuff, thanks again.

Kadriver
 
AgNO3 is colorless for a reason--it's not in the visible region. It can be used as the blank for subtraction purposes.

Make certain that your solutions are crystal clear (finely filtered) and you use pure water. You don't want particulates to scatter or absorb light!
Take your quartz cuvette and fill with 0.5 M Cu (II) nitrate and just scan the wavelengths...given the color, you should have a good place to start, or you can determine it empirically and learn a bit. It's also just in the literature.

For the data to be meaningful, you must make a standards curve that is in the linear region. As all measurements will be determined at the wavelength of peak absorbance, you can set up a spreadsheet to make a graph. This will mean making a series of 6 or 7 samples and doing serial dilution; start with say, 0.05 moles of Cu dissolved in say 100 mL of nitric acid and dilute from there. Be precise in weighing the metal! If you want to extend the range down by several decades, you can make the solution ammoniacal, as Cu(NH3)4+2 has a much higher molar extinction coefficient.

So, you've got these solutions lined up like ducks in a row. To what end?

Now you need to review the Beer-Lambert Law:

https://en.wikipedia.org/wiki/Beer_lambert_law

You'll see that absorbance is linearly related to the path length the beam travels (often 1 cm, but watch your units!) , the concentration (M), and the molar absorptivity coefficient (M^-1cm^-1). Absorbance is the ratio between power detected and power produced and the inverse of transmittance.

Kevin, depending on how crazy you want to get with this, you can also determine the Pd+2 and Ni+2 as well as the Cu+2 and can use MS Excel and Solver to simultaneously determine these concentrations.

But let's keep it simple for now. Go into Excel and make a sheet like this:

[Sample number] [Cu concentration] [ Absorbance @ ___ nm] [blank Absorbance @ ___ nm] [corrected absorbance @ ___ nm). You probably don't need to correct given AgNO3 absorption profile...

This link will help. http://www.chemguide.co.uk/analysis/uvvisible/beerlambert.html
The concentration will be the data on the x-axis in the graph you make, and the absorbance will be dependent and on the y-axis. The slope, when linear, is the molar absorptivity coeff. You can then make an equation that will solve for unknowns and do it multivariable with simple matrix/linear algebra. This is the same principle that guides AA, ICP, etc. It is very easy to implement if you follow the rules!

I can't post attachments any more on the forum due to unresolved technical issue or I'd just make you all a sheet for this that generates the graphs.

EDIT: sucho beat me to it, didn't see his post! His solid advice stands :)
 
Good to know the machine can't tell the difference between clean pure water and silver nitrate solution.

I'll dissolve 60 grams of clean copper then add water to 1 liter and take a reading (after getting the correct wavelength for max absorbtion).

Then I'll serial dilute in 10 gram intervals and build the refernce graph.

I can then draw a sample from the operating cell, set the proper wavelength, get an absorbtion reading from the sample and compare it to the graph to determine copper concentration in the cell.

I'm still learning about 0.05m - got an old text from 1930 and an excellent book entitled "The Organic Chem Lab Survival Manual".

Thank you both for the excellent information.

kadriver
 

Attachments

  • P1240187.jpg
    P1240187.jpg
    749.5 KB · Views: 454
  • P1240186.jpg
    P1240186.jpg
    672.9 KB · Views: 454
OK, today I finally had some time to play with the Spectronic 20D, Spectrophotometer.

I read the previous posts several times from Sucho and lou plus watched youtube tutorials.

One of the tutorials have a nice graphic on the internal workings of the machine:

[youtube]http://www.youtube.com/watch?v=xHQM4BbR040[/youtube]

There was also an excellent video lecture given by a Professor Fink. It explained Beers Law quite nicely:

[youtube]http://www.youtube.com/watch?v=EX_kyHg60tg[/youtube]

I have never used one of these, so I just took my time.

I began my experiment by putting 6 grams of pure copper (from some copper wire I had laying around) in a 250ml beaker and dissolving it with minimum amount of nitric acid.

I did the same thing with 20 grams of pure silver crystal from my silver cell, in a 400ml beaker.
 

Attachments

  • P1240256.jpg
    P1240256.jpg
    555.5 KB · Views: 410
  • P1240255.jpg
    P1240255.jpg
    651.3 KB · Views: 410
  • P1240254.jpg
    P1240254.jpg
    543.4 KB · Views: 410
  • P1240253.jpg
    P1240253.jpg
    395.4 KB · Views: 410
  • P1240251.jpg
    P1240251.jpg
    543.8 KB · Views: 410
I dissolved the metals slowly with only small incremental nitric acid additions so there was very little, if any, nitric acid left after the reactions.

I should have recorded the nitric additions but I did not. The copper took about 18ml nitric and the silver was about the same.

They were both on low to medium heat for about an hour, maybe a little longer.

The 250ml beaker was etched from a previous experiment that used NaOH, molten NaOH will dissolve glass.

The silver nitrate solution was crystal clear like water.
 

Attachments

  • P1240267.jpg
    P1240267.jpg
    547.6 KB · Views: 406
  • P1240264.jpg
    P1240264.jpg
    331.6 KB · Views: 406
I poured each solution into a seperate 100ml graduated cylinder, rinsing all the solution out of the beaker and into the cylinder with distilled water.

Once each solution was thoroughly rinsed into the cylinders, I carefully added distilled water until I had exactly 100ml in each one.

I hope I did this part correctly, I was shooting for 60 grams per liter dissolved copper because this is what has been described as the tolerance limit for dissolved copper in an operating silver cell. Six grams of copper dissolved in 100ml of liquid equals 60 grams per liter.

I then did the same with the 20 grams of silver dissolved on about 50ml of liquid. I added it to the cylinder, then added distilled water to produce a solution of 20 grams of silver dissolved in 100ml of liquid. This would equal 200 grams of silver dissolved in one liter of liquid.

I will be using an electrolyte concentration of 200 grams per liter while running the cell in a stainless steel bowl. The higher silver concentration will allow greater current flow through the cell and increase the speed at which the pure silver crystals are formed.
 

Attachments

  • P1240275.jpg
    P1240275.jpg
    940.1 KB · Views: 404
  • P1240273.jpg
    P1240273.jpg
    822.7 KB · Views: 404
  • P1240272.jpg
    P1240272.jpg
    945 KB · Views: 404
  • P1240269.jpg
    P1240269.jpg
    922.3 KB · Views: 404
Once I had the proper concentrations, I filtered each solution to remove any particulate.
 

Attachments

  • P1240281.jpg
    P1240281.jpg
    542.8 KB · Views: 403
  • P1240278.jpg
    P1240278.jpg
    731.4 KB · Views: 403
I did NOT rinse the filters after the filtering operation as this would alter the concentrations.

I did rinse the silver nitrate filter into my silver jar to recover any remaining silver from the filter.
 

Attachments

  • P1240283.jpg
    P1240283.jpg
    631.4 KB · Views: 403
  • P1240282.jpg
    P1240282.jpg
    642.1 KB · Views: 403
With the solutions prepared, I began the experiment with the spectrophotometer.

It had been running for the last 45 minutes to warm it up thoroughly. The instructions said the warm it up for 15 minutes.

I arbitrarily picked the number 500 as a starting point and adjusted the wavelength knob (number 3) until the digital readout (number 1) displayed 500nm.

Then, with no cuvette in the well and the mode selector (number 2) set to "transmittance", I used the zero adjust knob (number 4) to set the transmittance to 0.0 (zero point zero).

Once this numer is set, it should stay at zero, but it kept changing. After about 45 minutes I got it to stay at zero. I concluded that for this machine, a minimum of 45 minutes is required before the unit is warmed up and ready to use.
 

Attachments

  • P1240292a.jpg
    P1240292a.jpg
    946.8 KB · Views: 401
The unit came with a box of about 40 cuvette tubes - tiny test tubes with an aligment mark on each one.

I selected and thoroughly cleaned four of the cuvettes and put them in the rack.

I filled the first one with distilled water, the second with the 200g/l silver nitrate solution, and the third with 60g/l copper nitrate solution. I left the fouth tube empty and ended up not using it.
 

Attachments

  • P1240304.jpg
    P1240304.jpg
    596.4 KB · Views: 401
I then took the distilled water cuvette and inserted it into the cuvette well.

Using the 100% transmittance knob (number 5), with the mode button (numer 2) set to "transmittance", I adjusted the 5 knob to get the number 100.0 (100% transmittance) on the display, the wavelenght number did not change and stayed at 500nm.

My first experiment was to see if there was a difference in transmittance between the pure water and the silver nitrate solution.

With the transmittance set to 100.0 (100%), I removed the DH2O cuvette and inserted the silver nitrate cuvette. The number changed from 100.0 (with the pure water cuvette) to 96.4 with the silver nitrate cuvette.

So there is definitly a difference in transmittance of light between a sample of pure water and a sample of 200g/l silver nitrate solution. I should have selected "absorbance" with the mode button (number 2) and checked both samples again, but I did not do it - lack of experience.

I decided to use the silver nitrate solution as my sample "blank" for the rest of the experiment.
 
With the silver nitrate cuvette in the well and serving as my "blank" I adjusted the 100% transmittence know (number 5) so that 100.0 was set on the display. The wavelength number did not change and stayed at 500nm.

I removed the silver nitrate "blank" cuvette and inserted the copper nitrate cuvette and observed a reading of 94.4, so I recorded this in the "T" column in my notebook.

With the copper nitrate cuvette still in the well I then selected "absorbance" with the mode selector button (number 2) and observed a reading of .025 on the display. I recorded this number in the "A" column in my notebook.
 
Using the mode button (number 2) I reset the mode to "transmittence".

I then adjusted the wavelength adjustment knob (numer 3) and set the wavelength to 520nm.

I inserted the "blank" silver nitrate cuvette and adjusted the 100% knob (number 5) to 100.0 on the display (the wavelength number stayed at 520 and did not change).

I removed the "blank" cuvette and inserted the copper nitrate cuvette and observed a reading of 86.4 on the display - recorded that number in the "T" column.

I then selected "absorbance" with the mode select button (number 2) and observed a reading of .063 on the display - recorded in the "A" column of the notebook.

Reset the mode to "Transmittence" with the mode selector (numnber 2) and repeated the process all the way up the wavelength to 680nm

After that, I could not get the 100% knob to adjust to 100 when I was at the 700nm setting on the display - so I figured the experiment was done - NOT.

Now I have to plot a graph of the data (I think). But I did establish that 640nm was the wavelength of light that had the greatest absorbance of light through the copper nitrate solution with a corresponding least amount of transmittence of light through the same sample.

So, 640nm will be the wavelength I use to set up the next phase of the experiment - the serial dilution and measurement of the known samples of copper nitrate.

I am done, this whole deal took about 8 hours for me to complete because it was my first time doing it.

This is too cool - I really enjoyed doing this and can't wait to get the rest of the experiment done tomorrow.

kadriver
 

Attachments

  • P1240308.jpg
    P1240308.jpg
    731.5 KB · Views: 396
Hi, your new toy reminds me of school days, I have used this type of spectrophotometer ( as visible range colorimeter) for estimation of chemical parameters in water (Ammonia, nitrites, nitrates etc.) What I remember is that one has to follow some standard methods for accurate estimation of chemicals (elements and compounds) and the bible for it is the AOAC official methods (Association of Analytical Communities) http://www.aoac.org/.
In theory you are on the right path but I think variations in solution of different cells may bias the results. If your standard graphs is linear then its a success story.
Good Luck, Sal.
 
I got busy with other "stuff".

Sal, I am troubled by the way I used two different solutions to determine the wavelenth of highest absorbance.

I used 200g/l (one day I will figure out how to say this in moles) silver nitrate as my "blank" for setting 100% transmittence.

But then I used the 60g/l copper nitrate solution to get my readings - but there is no silver in with the copper, and this is not representative of what a sample of electrolyte from the silver cell would be - that sample would containe a mixture of silver nitrate and copper.

To be as accurate as possible, if I am using silver nitrate as my "blank" then I should be using it as a mixture with the copper nitrate solution to get a truely acurate absorbance or transmittence number for the copper in the cell.

I think this was pointed out earlier in the post by Sucho - but I have no clue on how to mix the solutions - what proportions to use.

More study is needed.

kadriver

Edited once to add words in red
 
I took a sample of the electrolyte from the cell and set it next to the sample of known concentration of copper.

P1240360.jpg

The tube labled "copper nitrate" is the standard of 60 grams copper per liter of liquid.

To the right of that is the sample of electrolyte from the silver cell labeled "Ag cell sample"

I warmed up the machine (it takes about 40 minutes for this thing to get warmed up where the zero setting stays steady).

Using the Spectronic 20D spectrophotometer, I then measured the tranmittence and absorbance of the silver cell electrolyte and recorded the numbers.

But I do not have the known concentrations established yet. I must sit down and figure out how to serial dilute the standard 60g/l solution, so that I have samples diluted down by 10 grams per liter. Then I can compare these readings to the silver cell sample and get an idea of the concentration of dissolved copper in the cell sample.

Visually, the silver cell sample (on the right) looks much lighter in color than the standard sample of dissolved copper (on the left).

P1240360a.jpg

kadriver
 
To get the proper dilutions for the set of standards I am going to do this;

Warm up the spectrophotometer with the wavelength set to 640nm (the setting that gave the lowest transmittence and the highest absorbance of the stock 60g/l solution from the previous experiement above)

I have about 96ml of stock solution with 6 grams of dissolved copper in 100ml of liquid in a stoppered 250ml erlenmeyer flask. This is my stock 60g/l copper solution.

Add 1ml distilled water to a 10ml graduated cylinder, then add 9 ml stock solution. This would yield a solution of 60g/l divided by 1.1 = 54.54 g/l dissolved copper. Analyze the dilution with the spectrophotometer and record - discard the solution.

Repeat by adding 2ml distilled water to the cylinder, then add 8ml stock solution to yield 60g/l divided by 1.2 = 50g/l, analyze with spectronic 20D and record. And so on.

If I go all the way to 9ml distilled water and 1ml stock solution there will be plenty of stock soluion left and I will get 10 sets of numbers to use as standards for comparisons

This may not be the correct way to do this dilution, and I don't know if the math is correct (it seems right).

If there are folks out there who can give some input I would be greatful.

kadriver
 
I got a chance to tinker with the machine today.

First I added 1ml distilled water to a 10ml graduated cylinder.

Then I added 9ml of the 60g/l stock solution of copper nitrate for the first dilution.

After inverting to mix, I added the diluted solution to the cuvette (a cuvette is a small test tube used to hold the sample for analysis with the spectrophotometer).
 

Attachments

  • P1240366.jpg
    P1240366.jpg
    628.9 KB · Views: 314
  • P1240365.jpg
    P1240365.jpg
    525.5 KB · Views: 314
  • P1240364.jpg
    P1240364.jpg
    631.8 KB · Views: 314
  • P1240362.jpg
    P1240362.jpg
    546 KB · Views: 314
  • P1240361.jpg
    P1240361.jpg
    653.5 KB · Views: 314
Back
Top