Mr. Fosi's drilled 20H

[lakshwadeep]

For the standardization, I think they mean that you will add the 60 mL of water to the 40 mL sodium carbonate, but you would calculate as if the acid is only reacting with the 40 mL of sodium carbonate solution.

[Mr. Fosi]

Why do you think so? After being so careful to get the concentration just right, why change it right before you do the analysis?

[lakshwadeep]

Here are some tips and more detailed instructions from my potentiometric experiment, where I standardized .4 N HCl with THAM, a different primary base standard (should not be dried):

 

-Prepare CO2-free water by boiling distilled water vigorously in a beaker for 5-10 minutes and then letting it cool. Prepare just before use, and store in 1 L bottle. You should use this water for diluting both the sodium carbonate and concentrated HCl.

 

-the weight of the sodium carbonate should be to .0001g (you probably would have done that ). Ideally, do one crude and two accurate measurements.

 

-HCl solution must be at room temperature during titration (it heats when the concentrated acid is added to water)

 

-titrate immediately with the HCl after preparing the base standard solution. For the first (crude) sample, record mLs and pH for each pH change of ca. 0.50. For the accurate samples, you'll need to record the mLs and pH for each pH change of 0.10, if possible, around the end points.

 

-when you plot pH vs. mLs HCl for end point determination, the curve must be well defined to provide accurate results. What is most important is getting many data points around the inflection point.

 

Why do you think so? After being so careful to get the concentration just right, why change it right before you do the analysis?

 

 

The calculation of "N = AB/53C" may be corrected for the dilution with 60 mL water, so you may want to either calculate the stoichiometry on your own to see if there is an assumption of dilution.

[Mr. Fosi]

So instead of titrating directly to the end-point (pH 4.5) and recording the amount of acid added, I should record the volume and pH of each addition near the inflection point (end-point) and calculate it using the equation of the regression line?

 

The calculation of "N = AB/53C" may be corrected for the dilution with 60 mL water...

 

Is there anything in the wording that makes you think this is what they mean? Would they would have stipulated that CO2-free water was to be used to do that final 60 mL dilution if a dilution was the aim?

 

By the time I finish baking/cooling the soda and boiling/cooling the water, I'll have enough to do the standardization twice just in case they wanted the 60 mL water to the 40 mL. If a dilution is what they want, than the calculated N of the acid should come out obviously wonky if I don't do it.

[Lalani]

And you have nothing to say about my DIY mag-rack?

The mag-rack is pretty too.

 

 

 

[Carinya]

aside from the outstanding frag rack, this reminds me of when my husband and his brother start talking in zeros & ones with each other.

[Needreefunds]

*peeks into room*

 

Nice rack Isaac.

 

everything else in here gives me a headache

 

yet, I try to follow along.

 

* ducks out *

 

[lakshwadeep]

So instead of titrating directly to the end-point (pH 4.5) and recording the amount of acid added, I should record the volume and pH of each addition near the inflection point (end-point) and calculate it using the equation of the regression line?

 

The inflection point may not be exactly 4.5. Your main focus is getting a lot data points around pH 4.5 (in my case, this was between pH 7 and 2.5). You could use the equation of the regression line, but it can be affected by inaccurate values.

 

You can be a little more nuanced by printing the scatter plot of pH vs. mL HCl (graph only the values near the apparent inflection pt.) and sketching the "curve of best fit". If the pH is on the Y axis, and HCl on X, the curve should look like a backwards "S". On one of the sideways "peaks" on the curve, draw a tangent line (this will have a negative slope). Then, draw a line parallel to the first that is tangent to the second peak. Finally, draw a third parallel line that is halfway between the other two, and the point where this crosses the pH curve is the inflection point, from which you would determine the mL of HCl used.

 

 

Is there anything in the wording that makes you think this is what they mean? Would they would have stipulated that CO2-free water was to be used to do that final 60 mL dilution if a dilution was the aim?

 

By the time I finish baking/cooling the soda and boiling/cooling the water, I'll have enough to do the standardization twice just in case they wanted the 60 mL water to the 40 mL. If a dilution is what they want, than the calculated N of the acid should come out obviously wonky if I don't do it.

 

Hmm, now I'm going back to the 60 mL doesn't matter, especially since they don't say exactly 60.00 mL. As an alternative, you can use indicators and do a non-pH dependent method to standardize the HCl:

http://pages.towson.edu/rsours/docs/210/Lab/Experiment_3.pdf

 

Here's an interesting and related experiment:

http://pages.towson.edu/larkin/210DOCS/exp4.pdf

[ribbie]

nice mag rack fosi!

[Mr. Fosi]

The mag-rack is holding up but the epoxy on the ceramic magnets sloughed off. I suspect that, given time, it will also come off the rare earth magnets. It looks like I will have to re-do it with a marine-grade epoxy or some DIY acrylic casting material.

 

Before I checked the ceramic magnets, I made a second rack so... I may have some fun taking it back apart.

[Mr. Fosi]

After much running around and frustration yesterday, I was able to finally run the titration.

 

Of course, I screwed it up 3-4 times before I finally got control of myself (and my frustration!) enough to do it nearly right. It's been a long time since I took a chem lab!

 

You wouldn't believe the funny looks I got as I wandered from lab to lab asking for a buret! People knew what it was but no one in biology had one. Since I don't know anyone in chemistry, I was afraid I was going to have to rig something with a pipet (decidedly not preferable!) until a colleague suggested I check with marine science. Duh!

 

I left my notebook at school so numbers are approximate where noted: I baked 15.000 g of NaHCO₃ (on foil) in an oven at 300F for 3 hours and cooled the remaining powder in a desiccator. After cooling, I re-weighed it and it came out to be a little more than 9 g; almost perfectly the 0.631 ratio predicted here. I put the newly anhydrous powder in a 15 mL scintillation vial, tightly capped it and put it and back into the desiccator.

 

I mixed up the solutions like the method stated, except that I ignored that strange comment about 60 mL of water.

 

For the titration, I skipped boiling the solution before the final endpoint determination... I know it's potentially an important step since I need to be a precise as possible but I had a reason for skipping it: I was already an hour late leaving for home.

 

Seriously though, I did the boiling step once during one of the practice runs and after I boiled it, the pH shot up ~4 units, however I was overeager when I started again, added too much acid and overshot the endpoint. I didn't want to leave without any data so I did the titration again, skipping the boiling/cooling step because it took so long.

 

Here is are a couple data plots for the non-boiled standardization. Don't worry, I am going to do it again without skipping the boiling step... Perhaps later tonight.

 

You can see that I bracketed the second inflection point (the only one of interest) pretty well.

 

 

 

 

Here's that second inflection point magnified and including a linear polynomial fit (regression line).

 

 

 

I hope lak will come back and tell me if I did the calculation correctly!

 

Using the equation from the regression above:

 

y = -1.448x + 53.085

 

y = pH

x = mL 0.05N Na₂CO₃ solution

 

Plug in y = 4.50 and you get x = 33.56 mL.

 

From post #347:

 

N = AB/53C

 

N = Normality (of the acid solution)

A = g Na₂CO₃ weighed into 1-L flask

B = mL Na₂CO₃ solution taken for titration

C = mL acid used

 

Plug in my values:

 

A = 2.500 g

B = 15.0 mL

C = 33.56 mL (calculated using the equation of the second inflection point)

N = 0.0211

 

Not bad! In post #335, I showed that the acid in the salifert kits is 0.0250N; pretty close. It looks like I am even closer to the established method than Salifert. Of course, these calculations were done with data created by skipping a potentially critical step but that doesn't change much about my statement regarding the accuracy in mixing up the acid.

 

By boiling the solution, I will have to add slightly more acid to reach the endpoint (pH 4.5) and if you look at the equation for calculating the N of the acid, an increase in C will decrease N, making my acid concentration even closer than it looks right now. Sweet!

 

I'm interested to see how much difference the boiling makes in the calculated acid N.

[lakshwadeep]

The calculations and mL of acid for the inflection point look good to me, but I'll wait for the boiled (CO2-free) titration.

 

Here's a neat Excel application for analyzing titration curves.:

http://www2.iq.usp.br/docente/gutz/Curtipot_.html

[Mr. Fosi]

Thanks for the link, lak. I haven't played with it yet but I will.

 

And now, here is are some additional results.

 

I re-ran the titration twice last night, once with the original Na₂CO₃ solution I made using non-CO₂-free water (as per Std Methods) and again with a Na₂CO₃ solution made with CO₂-free water (as per some other undergrad lab methods I have seen online). In both cases, I boiled the solution for 4 min (and cooled it to room temp) after I titrated down to pH ~5.0.

 

I unfortunately didn't have a watch glass or anything that would function like it so I had to improvise... I cut off the end of a Pasteur pipet and jammed it in a one-hole rubber stopper. I then parafilmed a 2 cm plastic core tube to the stopper, around the pipet. The space between the tube and the pipet was filled with ice water.

 

It worked essentially like like the much nicer reflux setup I used when I took organic chem.

 

 

 

 

Here is a comparison between boiled and non-boiled, non-CO₂-free Na₂CO₃ solution. The non-boiled data is the same as I presented in post #361. You can see that the curves are virtually identical until you get down near the second inflection point and they are visibly different after the boil. You can see the pH shoot up after the boil right around 34 mL.

 

 

 

Here's a comparison between titrations done using CO2-free and non-CO2-free Na₂CO₃ solutions. You can see that they are identical at this magnification.

 

 

 

Here is a magnification of the second inflection points of all three curves. So, once the second inflection is enlarged, "slight" differences start to look a bit more significant. The differences between the two boiled curves could potentially just be normal variation between replicates but since this isn't replicated, we don't know for sure.

 

The difference in line slope between the non-boiled and the others is almost certainly more than just regular variation.

 

 

 

Now it's time to see how different the calculated acid volumes are:

 

non-Boiled: 33.56 mL

non-CO₂-free: 34.10 mL

CO₂-free: 34.03 mL

 

That translates to a 1.59% difference between the Boiled and non-Boiled and only a 0.20% difference between the Boiled CO₂-free and non-CO₂-free solutions.

 

So how does that translate to calculated acid N?

 

non-Boiled: 0.0211N

non-CO₂-free: 0.0207N

CO₂-free: 0.0208N

 

So it looks like I was pretty spot-on when I mixed up the H₂SO₄. I am going to use the non-CO2-free number since that is what is suggested in Std Methods.

 

Now it's time to compare my acid it against the one in the Salifert alk kit!

 

Oh, and for those of you who never had the privilege of see/using a titration setup, here's what mine looked like. I' better clear it off the bench on Monday before my advisor has a coronary!

 

[Urchinhead]

I may have missed this as I read through this but... Other than the very cool chem work going on in a bio lab and the associated fun from watching the senior faculty go batsh1t from the blasphemy of the act... what are you trying to achieve? More accurate testing results? Easier testing? Doing it because you can?

 

I ask because it seems allot of work for accuracy that isn't required in the hobby unless you have managed to simplify the testing process which would be a bonus... Or you are doing something with your tanks that requires that level of resolution...?

[Mr. Fosi]

I want to stop buying test kits and I want to make tests that some of my acquaintances can use instead of paying $22 for a kit that only does 100 tests.

 

Why not pay $10 for 5x more tests? Tests that are just as accurate and precise as a commercial kit?

 

I already have (personally) most of the glassware I need and the reagents are over-the-counter. Now that I have an idea of what the reagents cost to buy, the 1000x markup of the commercial kits makes me go .

 

Right now, I am working out the comparison between my reagents and those that come with the Salifert alk kit. No numbers yet but I hope to have some within the next 48 hours.

[Urchinhead]

If you decide to 'share the love' and its no harder than an API I would like to make a contribution to the Mr. Fossi Foundation for Wayward Aquarists in return for some of the magic then!

[ribbie]

I'm interested in Mr.Fosi's poor man test kits!

[Carlton'sTank]

I would be glad to contribute a nice chalice frag that I have been holding hostage for 2 months now because I am with you, $20 and up for a test is pretty lame.

[Mr. Fosi]

The good news is that I now have a method that is getting me within 4.8% of the Salifert alk kit. The bad news is that it takes a 100 mL sample and a pH probe to get there.

 

But the other good news is that when I use a 4 mL sample (like the Salifert kit), I can get within 10.1% using just the indicator dye. That's just as good as the results published in an Advanced Aquarist article from 2007. There were a lot more tests done in that article but I am glad to see that I am at the lower end of the difference.

 

What's more interesting is that the end-point for the salifert dye isn't pH 3.93 as stated in the article. It was very close to 4.50 in my tests. If you drop in acid until the solution turns bright pink, you have overshot the titration and your numbers will be higher than they actually are in the tank. If you want to be more accurate, you need to stop dripping acid in when the solution turns a mostly pink shade of purple.

 

Right now, I have enough acid to do more than a thousand tests but I don't have any indicator dye other than what's left in my Salifert kit. I have a recipe for the dye though, I just have to find an inexpensive source. The dye is the only thing that isn't over-the-counter.

[fishdaddy0831]

I'm interested in Mr.Fosi's poor man test kits!

[Mr. Fosi]

Since I am using the industry-standard method, I am temped to say that my numbers are better than theirs.

 

After pricing out the the parts of a Salifert-style alkalinity kit, it looks like it would cost in the neighborhood of $6.00 to put together (straight cost and no profit). And this is buying components retail! It'd be so much cheaper if could buy/produce in bulk.

 

The most expensive parts are by far the plastics (bottles, syringes, etc.) at 77% the total cost, followed by the indicator ingredients at 23%.

 

So cutting out the plastics makes the cost of a strip-down kit ~$1.40 (again, straight cost).

 

This of course doesn't account for the time to make the kits (not insubstantial and I'm not sure yet how to account for it) or the equipment I need (~$1550) to do it at home without using university equipment (certainly frowned upon and unethical).

 

Would you be willing to pay ~$10 (plus shipping) for a Salifert-style alkalinity kit that did 3x the number tests and was at least as accurate? Would you be willing to provide the plastics needed and just pay ~$3.00 (plus two-way shipping) for the reagents? These are the questions that need answered.

 

Just a dream for now but I think I will make up several small kits and send them out (for the cost of shipping and materials) to some people to try. If they stand up to field testing maybe I'll be able to do something more profitable in the future.

[Urchinhead]

Labor cost and capital expenditures are pretty straight forward mate.

 

Take the yearly salary of someone in your profession (after graduation) and divide that by 2040. That will give you the hourly rate of pay. Add up the hours it takes you to create the regents, separate them, package them, etc and the multiply that number by the hourly rate. Then divide that number across the number of test kits produced in that timeframe and you have your labor cost per unit. Add that into the retail cost of the test kit.

 

Cap Ex is the $1,500 divided by 5 years which is the standard amortization timeline for equipment. Take that number and further divide it by the number of units you expect to manufacture per year and there is your cost for equipment. Again add that into the retail cost of the test kit.

[Mr. Fosi]

I'm no businessman and I've never taken an accounting class so these calculations are new to me. Thanks for posting them though! Makes my life easier.

 

I don't know how many of these things could be sold by me. Right now, no one aside from me has used one so there will have to be some field testing done. If I try a commercial angle (even a small-scale one) I may have to provide some kind of warranty about their accuracy.

 

I'm no marketing man and I don't have the time for it anyway... It's baby steps for now.

 

I guess I'll first send out a couple kits to some reliable peeps (you're at the top of the short-list, UH) and see what they think... That means a small initial production run, probably of fewer than 20 kits all to be sent at cost + shipping. I have 1/3rd of the work done with standardizing the acid (and keeping track of how I did it). I'll need to shell out for a small amount of the indicators, figure out what to package the liquids in and I'm good to go.

[Urchinhead]

No worries mate. Its the economic game theorist in me coming out. Some things to think about...

 

Why offer a warranty? I don't believe anyone else does.

 

I will be happy to email you sample articles of incorporation, a sample business plan along with P/L spreadsheet that you can work your way through to come up to speed on what you need to make this "legitimate", limit your liability, and most important of all turn this into a nice way to generate profits while taking "business expense deductions" on your taxes.

 

Outsource things that are not critical path. Meaning things that are not part of coming up with test kit materials like web site design, marketing plan (think about here. The cost is low when weighed against the benefits gained from reaching your target demographic coupled with your reputation here), and product packaging design etc. John Maeda comes to mind as a good general resource for this by the way.

 

I would think that regent containers are also pretty straight forward and cheap as well as available from most chemistry supply houses. All you would want to do is slap a company logo on the bottle and you are good to go.

[mkregs]

Just stopping by to see how your tank is looking....

 

? - how would someone become aquainted with the person who provided these frags to you?? I am loving those blue zoas and have had a spot in my tank waiting for a nice frag of blues.

 

Of course if you're getting bored of them, I would be willing to babysit them in my tank for an indefinite period of time

 

You're tank is looking good!