Sciencemadness Discussion Board

How to determine the ratio of two chemicals mixed together?

Krauser - 17-1-2019 at 12:07

Hello guys, i'm new in the forum, so forgive me if i posted this in the wrong place. But here's the question:
Let's suppose that you have two chemical compounds mixed together. Let's say, two salts, NaCl and KCl in solid form (powder). And you don't know the ratio of one relative to other, and you want to know this. How much of each you have in the mix (in %).
So, how can i do that? What kind of procedure i could do to get the answer?
I know it is a trivial question but i'm new to chemistry so i need some help. I would be glad if someone give me a answer. Thanks.

Σldritch - 17-1-2019 at 13:11

It will vary between compounds but in this case the simplest way would be to compare the densities. I would do it by immersing some mixture in toluene and measuring the mass and volume change to calculate the density.

Krauser - 17-1-2019 at 13:24

Quote: Originally posted by Σldritch  
It will vary between compounds but in this case the simplest way would be to compare the densities. I would do it by immersing some mixture in toluene and measuring the mass and volume change to calculate the density.

Hmm that sounds a little complicated for me because i don't have toluene... Can't i do like a solution of the mixed salts and afterwards do a crystalization? Like, puting it in the freezer and taking the precipitated salt and weighting? (If the solubility of the two at low temperatures were quite different from each other). But it's a good answer. Thanks Σldritch.

unionised - 17-1-2019 at 13:26

There are so many possibilities it's hard to know where to start.

For example- albeit a strange one- you could make use of the fact that potassium is slightly radioactive.
The radioactivity would be proportional to the potassium content.

Or you could dissolve a known amount of the mixture in water and add an excess of silver nitrate solution.
That would form a precipitate of silver chloride.
You can filter it off, dry it and weigh it.
That will tell you how much chloride was in the mixture. (about 108/143 of the silver chloride is silver, so the rest is chloride).
Since the percentage of chloride is different in sodium chloride and potassium chloride, you can find how much of each is present.




clearly_not_atara - 17-1-2019 at 13:29

You don't need to use toluene. It can be any liquid that doesn't dissolve salt. Even olive oil would work here.

Krauser - 17-1-2019 at 13:46

Quote: Originally posted by unionised  
There are so many possibilities it's hard to know where to start.

For example- albeit a strange one- you could make use of the fact that potassium is slightly radioactive.
The radioactivity would be proportional to the potassium content.

Or you could dissolve a known amount of the mixture in water and add an excess of silver nitrate solution.
That would form a precipitate of silver chloride.
You can filter it off, dry it and weigh it.
That will tell you how much chloride was in the mixture. (about 108/143 of the silver chloride is silver, so the rest is chloride).
Since the percentage of chloride is different in sodium chloride and potassium chloride, you can find how much of each is present.


That's really neat. But that's way beyond my capabilities.. and i don't have any equipment related to detecting or measuring radiation.. but it's cool anyways. And the nitrate one seems more accessible to the amateur "chemist". Gonna keep that in mind. Thank you.

Krauser - 17-1-2019 at 13:51

Quote: Originally posted by clearly_not_atara  
You don't need to use toluene. It can be any liquid that doesn't dissolve salt. Even olive oil would work here.

Hmmm now i understand. Well. That's sounds relatively simple and easy now. Gotta try that :)
But here's what i didn't understand. How i would be able to found out the percentage of each salt in the mix using this procedure?
Because the two are mixed together so it might have more of a salt than other and i get the same weight. Sorry if i'm slow to figure it.. but thanks so far for helping out.

DraconicAcid - 17-1-2019 at 13:57

If I needed to do that in the lab, I'd titrate a sample with silver nitrate. That would give you the total number of moles of the salts, and you also have the total mass of the salts. This gives you a system of two equations and two unknowns, which can be solved to give the mass of each.

unionised - 17-1-2019 at 14:02

If I was doing it at work I'd dissolve some of the mixture in water and put it through the ICP OES machine- a sophisticated flame test.

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

clearly_not_atara - 17-1-2019 at 15:10

In this particular case, if sodium perchlorate is available, you can simply dissolve some of the sample in a minimal amount of water and add a slight molar excess of saturated sodium perchlorate solution. The precipitate consists of pure KClO4 accounting for nearly all of the potassium in the sample. This also works with sodium bitartrate which is rare.

Krauser - 17-1-2019 at 17:47

Quote: Originally posted by unionised  
If I was doing it at work I'd dissolve some of the mixture in water and put it through the ICP OES machine- a sophisticated flame test.

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


Well... not everyone has access to a + $30,000 piece of equipment at home lol

Krauser - 17-1-2019 at 17:58

@clearly_not_atara I have Potassium Nitrate. It does the job? And can you explain more about the first procedure you have suggested? (That with the toluene/oil) like a simple tutorial for me to try it lol. And sorry for making so many questions. You guys are being of so much help. I might not have all these resources but are all valid ways of doing it and good source of info. Thanks so far people.

clearly_not_atara - 17-1-2019 at 19:15

What does potassium nitrate have to do with anything? The perchlorate isn't supposed to be an oxidizer.

You calculate the density as follows:

1. Put some oil in a graduated cylinder. Note the volume, call it V1
2. Measure out a quantity of the mixed chlorides on the scale. Note the mass, call it M.
3. Pour the salt into the graduated cylinder carefully. Note the volume at the top of the oil, call it V2.

The density is rho = M / (V2 - V1). To obtain the mixing ratio:

A. Set up a linear equation for rho in terms of the density of potassium chloride, the density of sodium chloride, and the mixing ratio.
B. Solve the equation to get an expression for the mixing ratio in terms of rho and the densities of the chlorides.

Krauser - 20-1-2019 at 10:56

Quote: Originally posted by clearly_not_atara  
What does potassium nitrate have to do with anything? The perchlorate isn't supposed to be an oxidizer.

You calculate the density as follows:

1. Put some oil in a graduated cylinder. Note the volume, call it V1
2. Measure out a quantity of the mixed chlorides on the scale. Note the mass, call it M.
3. Pour the salt into the graduated cylinder carefully. Note the volume at the top of the oil, call it V2.

The density is rho = M / (V2 - V1). To obtain the mixing ratio:

A. Set up a linear equation for rho in terms of the density of potassium chloride, the density of sodium chloride, and the mixing ratio.
B. Solve the equation to get an expression for the mixing ratio in terms of rho and the densities of the chlorides.


Thank you man. I did the 3 first steps carefully. And now i got this value for the density:
ρ = 2,362.

How do i setup the equation now?
Sorry, i'm a bit dumb, but i'm trying my best to understand it.
And thanks again for your patience :)

CobaltChloride - 20-1-2019 at 11:25

That is a bit too high. The density of KCl is 1.984 g/cm^3, while the density of NaCl is 2.17 g/cm^3, so a mixture of these salts cannot have a density of 2.362 g/cm^3.

Krauser - 20-1-2019 at 11:27

Quote: Originally posted by CobaltChloride  
That is a bit too high. The density of KCl is 1.984 g/cm^3, while the density of NaCl is 2.17 g/cm^3, so a mixture of these salts cannot have a density of 2.362 g/cm^3.

Hmm maybe it has some impurites then i think. Gonna try again with a mix that im gonna make.
But anyway.
How do i proceed with the equation?
Can you tell how i setup it?
Thanks!

CobaltChloride - 20-1-2019 at 11:42

Well you can do it like this, where x is the mass fraction of NaCl, y is the mass fraction of KCl and ρ is the measured density of the mixture:
x*2.17+y*1.984=ρ
x+y=1
Now you just need to solve this system of 2 equations and 2 unknowns. First of all, from the second relation you can see that x=1-y, so the first relation becomes (1-y)*2.17+y*1.984=ρ => y=(2.17-ρ)/0.186. Knowing y, x can be easily calculated by using x=1-y.

In my opinion, the silver nitrate titration which was suggested previously on this thread should give more accurate results. Your error most probably comes from the fact that the graduated cylinder you used isn't very accurate.

Krauser - 20-1-2019 at 11:57

Ok. Now i mixed some NaCl and KCl, and i tried again. I got this value for density:

ρ = 0.80

Krauser - 20-1-2019 at 12:14

Quote: Originally posted by CobaltChloride  
Well you can do it like this, where x is the mass fraction of NaCl, y is the mass fraction of KCl and ρ is the measured density of the mixture:
x*2.17+y*1.984=ρ
x+y=1
Now you just need to solve this system of 2 equations and 2 unknowns. First of all, from the second relation you can see that x=1-y, so the first relation becomes (1-y)*2.17+y*1.984=ρ => y=(2.17-ρ)/0.186. Knowing y, x can be easily calculated by using x=1-y.

In my opinion, the silver nitrate titration which was suggested previously on this thread should give more accurate results. Your error most probably comes from the fact that the graduated cylinder you used isn't very accurate.


Yes, maybe isn't too accurate. But i think it's somewhat right now.
So, do i replace the ρ with my value?
Can you use my value (ρ = 0.80) and do the equation for me so i can understand it better? I like to have a example, you know.
I'm 12, so I'm not too good at mathematics. I know... sounds like a cheap excuse lol

Metacelsus - 20-1-2019 at 12:38

ρ = 0.80 doesn't make sense since the mixture won't be less dense than the least dense component (in this case, KCl). If that's the density value you measured, you've made an error somewhere. The density of the mixture should be between the densities of the two components.

(Note that for some non-ideal cases, mixtures can be less dense than the least dense pure component, or more dense than the densest one. But this isn't one of those cases.)

Edit: bonus question: can anyone explain why KCl is less dense than NaCl?

[Edited on 2019-1-20 by Metacelsus]

Krauser - 20-1-2019 at 12:47

Quote: Originally posted by Metacelsus  
ρ = 0.80 doesn't make sense since the mixture won't be less dense than the least dense component (in this case, KCl). If that's the density value you measured, you've made an error somewhere. The density of the mixture should be between the densities of the two components.

(Note that for some non-ideal cases, mixtures can be less dense than the least dense pure component, or more dense than the densest one. But this isn't one of those cases.)

Edit: bonus question: can anyone explain why KCl is less dense than NaCl?

[Edited on 2019-1-20 by Metacelsus]


Hmm. I'm gonna have to buy new graduated cylinders i guess..
Or I'm doing something wrong in the calculations. Hold on.
Oh, and a example would be really good. So i can keep trying again and again until i get it right.
I really wanna learn how to do this.

For the moment let's suppose that the valuefor density is 2.00

[Edited on 20-1-2019 by Krauser]

Assured Fish - 22-1-2019 at 01:06

Why is this thread not in beginnings.
What liquid are you using with the salt to determine its volume?
Also are you sure you did it like this d=m/v and not d=v/m.

To calculate from their you would likely have to plot a graph.
Concentration of sodium chloride on on the y axes and density on the x axes, from their you will want to write number 1-100% likely in intervals of ten along the y axes, and on the x axes you will want to write the densities starting from 1.984 to 2.17, in intervals of 1.
Then set a dot at 100% and 2.17 and a dot at 0% and 1.984, then draw a line from one dot to the other.
Next put your density (in this example 2.00) along the x axes and then line it up with the line.
Where ever that dot in relation to the y axes is, is your very rough concentration of sodium chloride, the other percent would be potassium chloride.

If you are having trouble with a solvent which doesn't dissolve the salt mixture, which is still dissolving a small amount of the mixture, then you can saturate the solvent by adding some of this mixture and then filtering off any residual salt.
Then add this to your graduated cylinder and put it on a scale, then record the volume and tar the scale.
Next you want to add the salt mixture to the solvent and give it a little bit of a swirl to dislodge any air pockets, then record the new volume and weight.

Toluene would be a preferred solvent, or any other hydrocarbon based solvent like naphtha or hell even mineral oil, anything that wont dissolve the salt will work.