Sciencemadness Discussion Board

nit pick with sucrose molecular weight

Little_Ghost_again - 29-8-2015 at 05:34

Hi,
Ok I know this might seem like a nit pick but I was wondering the following.

If your working with mols etc then surely you try to weigh out to 1mg? alot of small cheap scale (<300g) can weigh down to 1mg (ok accuracy is probably rubbish).
So when you work out how many grams are in X to make 1M then you work to 3 sig digits? Otherwise you would be a fair bit out?
For example I tried to work out how many grams in 1M sucrose, I rounded up and tried 2 sig digits, this gave me 342.01g.
Now wikipedia says it should be 342.3. So thats 290mg different!
Seems alot to me especially if doing test tube size stuff.

So I used the correct molecular weights and didnt round up or down I just plugged in, I got 342.29808
Seeing as mg seems a reasonable weight weigh too I figured it should be to 3 digits, so that would be 342.298g
I understand wiki has rounded up but I was wondering which is the correct way in the real world? Do you round up the elements weight then add together as I did at the start, or do you do it the way I did last of all and ignore wikipedia?
I know its a nit pick and not important for sucrose (in this case), but for the future how should I be doing it?

phlogiston - 29-8-2015 at 13:38

Add up the unrounded atomic weights, ignore wikipedia. Also make sure that you are not using monoisotopic atomic masses, but the standard atomic weights (unless, ofcourse, your compound is monoisotopic).

In practice, the accuracy with which you need to do weigh something depends completely on what you need it for. If you need to make a standard solution for an analytical method, try to be as accurate as you can (However, it is not useful to weigh something with extreme accuracy if you can not pipet your solvent with the same level of accuracy). Then, it is best to weight an amount that is in the mass range of your balance in which it is most accurate.
For synthetic chemistry, it usually matters much less.
If you are weighing a kilogram of something, you hardly ever need mg accuracy. When you need to accurately weigh a few mg of an extremely expensive or unique compound (so you can't simply take a little more just to be able to weigh it properly), you are going to need a very expensive balance.

aga - 29-8-2015 at 13:51

The bits past the decimal point matter a lot if you're dealing with tiny amounts.

Up in the 1000 kg range, a kg either way matters little.

Human error is by far the biggest factor when it comes to errors, as we tend to get it wrong somewhere quite often.

Those errors have led to some quite amazing discoveries at times, and rather a lot of explosions and deaths.

Little_Ghost_again - 29-8-2015 at 15:00

Its not that I am doing anything important etc but I am working on the test tube scale so using <1g. It was just that 290mg seemed alot when working at that scale.

I missed a decent toledo on ebay that trends around £300 and it went for £30!! I didnt have £30 but wish now I had sold a couple of toes or something :D.
Thanks for the answers. I asked because normally I dont weigh out too carefully but at the moment I am trying to prove something works so need to be a bit more accurate, my normal calibrated eyeball method is suitable for this :D.

unionised - 30-8-2015 at 04:13

" I rounded up and tried 2 sig digits, this gave me 342.01g."
Well, if you rounded to two digits or to two significant figures you would have got 340 g. What you did was round to two decimal places, but that's beside the point.

The issue is not the absolute error, but the relative error.
Being wrong by 290mg matters if you only have a gram of stuff to work with, but if you have a kilogram it'snot going to matter much if you have 1000 grams or 1000.29 grams.

Also, if you want to pretend that you are weighing something to 5 digits you need to ensure that your materials are 99.999% pure and that you allow for the effect of air buoyancy while weighing (or weigh them in a vacuum chamber).

Ozone - 30-8-2015 at 05:18

There is a limit to how accurate/precise you can be. Every value (measurements, compound purity, adsorbed moisture, etc.) that is used to calculate a quantitative value has it's own degree of error, and error propagates. Tracking down and accounting for this is difficult (sometimes impossible), so you usually have to define your own systematic range--via instrument replicates, full method replicates, and a linearity check that goes to a low/high-enough concentration to determine where linearity is no longer good (e.g. the LOQ and detector saturation/hyperbolic behavior).

In practice, I routinely handle 1mg amounts, often diluted into 1.5mL sample vials. The accuracy and reproducibility (in my lab and once tested in others) is generally excellent (if you call ±0.5% propagated over seven replicates of 12 analytes excellent--thats <0.1%/analyte vs. traceable standard materials). In order to do this, however, I:

1. prepare everything by mass, g/gtot. Eliminate as many sources of error as you can, e.g. density measurement/correction and volumetric error (±0.08% for a 100 mL Class-A volumetric flask, for example).

2. I use a balance precise to 0.00001 g which gives ample headroom for that 1 mg mass (e.g. within the greater overall propagated error). Even this may not be enough for very expensive or rare (e.g. 0.01 mg of some mAb, for example). The balance is verified with known masses daily and re-qualified quarterly. So, it's up to you to know if 1.00±0.01 -ish mg is good enough.

3. My balance is in a sealed box on a very heavy vibration-buffered table (and has internal vibration compensation), and an ionizer to kill static.

So, know what is "good enough" to meet your needs.

I will note, however, that if you would like to make mass balances composed of multiple quantitatve measurements (e.g. organic acids/glycerol, etc. via HPLC-RID, carbohydrates by HPAEC-PAD and then calculate the whole over, say refractive dry solids), close, you will be limited to (at the very least) the worst measurement in there...probably RDS at no better than ±0.01 % with a high-end instrument, ±0.1 % typical). So, on closing, you might get 98-102% overall, which is considered quite good--at least for my needs--yours? Maybe not.

O3

Little_Ghost_again - 30-8-2015 at 06:41

I appreciate that as pointed out other factors like purity etc play a part, however I am trying to do the best I can with what I have. I dont have 99.999% purity, I do have a vacuum chamber and I do have a reasonable (for a amateur) balance. So while other things matter it at least helps me to how the correct way to work things out on paper!
I dont need the accuracy as such for the practical side, but surely trying to work out the maths side correctly matters. Like in the pre pub section, no one hardly ever says I added 12.983 mg of X, nine times out of ten it will say I added 12g of g. so from that view point I see that it dosnt matter much in most situations, but what I am trying to do is get the calculation right then see how close I can get in the real world to what the maths said it should be.
As the title said I know its a nit pick but I also wanted to know how you guys did your working out etc. In the experiments I am doing at the moment there is a massive margin for error because I am using yeast, so even with access to high grade microscopes and haemocytometers the cell density measured will be a approximation anyway. Unless I actually stain and count each of the 20 billion cells I am pretty sure I am going to be out by at least a couple of hundred or so :) (yes I know the figure will massively higher than a couple of hundred!). then I have the problem of dealing with a living organism so they are not all going to be performing like 'ideal' cells.
but I at least wanted the theory side with the numbers correct, thanks alot for the input its given me some useful information. as an aside I have isolated a wild yeast strain that is performing really well at producing alcohols other than Ethanol. way to early to tell very much yet but its interesting to see how how different its behaving from the known strain I am using, I know its down to the yeast because it was plate isolated and the culture is in a proper reaction vessel that I plate weekly to make sure there is no contamination.
Again its not perfect but the plates are coming up clean and only showing the wild strain.
I have had alot of help from my dad with it though.

thanks again

macckone - 8-9-2015 at 08:45

Points to remember:

Significant digits are those that are not zero, not just the ones after the decimal.
Zeros between non-zeros are obviously significant and trailing zeros can sometimes
be significant and in lab notes that is usually marked with a line under the zero to
indicate significants (different labs may do that differently).

Error is relative and usually expressed as a percentage.
Your 290mg error is actually only .084%.

Another thing that is important in organically derived products is that
plants actually concentrate carbon and oxygen of different isotope ratios
from the naturally occurring carbon and oxygen as measured for the periodic
table. If you are doing 5 decimal place work then this is important.
For the weekend chemist 2 decimal places will suffice and it won't matter.

In anthropology those isotope ratios are important because different foods
have different ratios and the ratio in the human and animal remains can
be used to tell what those people and their pets/livestock ate.

In the case of sugar cane (where most sucrose is derived from) this may
skew the molecular weight of the sugar at 4 decimal places or it could just
be wiki being wiki. I would use merck for biologically derived compounds.
It is going to be more useful. And if you really need 4 decimal place accuracy
your compound needs to be 99.99% pure. Table sugar does not meet that
requirement.

Little_Ghost_again - 8-9-2015 at 09:27

I tried to download MERCK from the reference section, i had a working copy that ran off a CD with win7 but I cant find a version that will work with win 8.1 on CD

macckone - 8-9-2015 at 15:40

There are still paper copies out there.

And of course other sources like:

http://bookzz.org/md5/921c261160eff80455ce0d1ec2a39bf3

**edit other link didn't work


[Edited on 8-9-2015 by macckone]

j_sum1 - 8-9-2015 at 16:00

Quote: Originally posted by Little_Ghost_again  
As an aside I have isolated a wild yeast strain that is performing really well at producing alcohols other than Ethanol. way to early to tell very much yet but its interesting to see how how different its behaving from the known strain I am using, I know its down to the yeast because it was plate isolated and the culture is in a proper reaction vessel that I plate weekly to make sure there is no contamination.

Way to go LGA! This is seriously cool.
What alcohols are you producing? Have you been able to determine that?
It would be really amazing if you stumbled across something that produced a tertiary alcohol. I have no idea how likely that is, but it would be useful.

back on topic...

macckone - 8-9-2015 at 16:02

BTW Merck agrees with Wikipedia in the 13th edition software.
My calculation showed 342.30 and oxygen and carbon can easily differ
in that 5th place.

My calculation:
12 * 12.011 = 144.132
22 * 1.0080 = 22.176
11 * 15.999 = 175.989

144.132+22.176+175.989 = 342.297 = 342.30 after rounding.

Actual molecular weight is going to differ in the last decimal place due to
dueterium, carbon-14, oxygen-17 and oxygen-18 in the growing environment.

Little_Ghost_again - 8-9-2015 at 17:37

Thanks for the link!! while I miss the CD with quick search etc the paper copy is just as good. The yeast experiments are a bit hard to evaluate at the moment, if I had the gas for the GC I could run a tiny sample through that.
I am using several sugars and I appreciate table sugar isnt great, but its doing a reasonable job at the moment. So far depending on your view point honey (35% W/W) Fructose (15% W/W), Glucose (10% W/W), Heavy sugar (I think its a high heated molasses I got from far supplies as feed additive, very very dark and sticky) (20% W/W) and sucrose (table sugar 20% W/W), is producing well.
Yes some of the sugars are the same but it a slightly different form, also I have the same yeast feeding on each of the sugars alone, the fructose one is not doing well. I also have various combinations and using anaerobic (as little air as I can manage getting in) and heavily aerobic cultures of each.
I should probably mention I have been messing with this since before Christmas in one way or another, the best cultures are in the proper reactor vessels (I only have 5 various real reactors) the others are in 2 litre conical flasks with RTV bungs I made and on a large Jenway multi stir plate, I have these ones in an old fridge converted to an incubator while the reators have there own heating.
The variations I have tried and those in progress include alginate enclosed yeast (got that from a paper) and also normal yeast culture. Some have PH buffering (Calcium Carbonate with Mag Carbonate) and some dont), I currently have 30 different containers on the go.
The best one is in a reactor with immobilized yeast, I used H2O2 and bleach etc etc to clean the alginate balls to try and stop yeast on the surface of the balls from breaking off, mostly I failed but two seemed to have worked well and no yeast I can find so far is loose in the solution. BUT I am only taking 5-20Ml samples via a syringe and septum a day. The one above with all the different sugars and having pure oxygen fed into the reactor spurge pipe 4 times a day for 15 seconds smells the strangest, it also gives a couple of low boiling point fractions. BUT its hard to tell as I have only tried distilling with a tiny micro still (size 10) and havnt taken much from it sample wise.
I feed the sugar (15ml (ish)) once or twice a day depending on the gas flow (buble counter with food dye and IR sensor). Its reaching its end point though as its now taking a while to eat it up, at the peak it would stop making Gas after 6 hours or so, now 15ml last around 14 hours but somedays it lasts around 10 hours.
I dont have the reagents to test for tertiary,secondary or primary alcohols yet, but my most promising reactor smells a little like acetone but not as harsh and kinda 'buttery'. I think maybe two more weeks and I will stop it and let it stand for 3 weeks with no food, I will of course keep plating the yeast from each.

Although I have taken shed loads of notes the following is just some the things I have seen so far, IF I get anything worthwhile or unusual I will type my notes up.

Now some of the things I have observed............. The immobilized yeast with aerobic conditions and kept at 38c (while feeding, explain that in a bit) with the funny mix of sugars behaved very different from the other alginate yeasts, I got the yeast for this off a plum brought at a country fair. It took ages to separate the yeasts of it but eventually i got i definite pure strain on a plate and used that. I started with a 10% sugar solution (W/V) and it didnt do anything for around 4 days, then it suddenly sprang into life. I kept feeding it up until around 6 weeks later it stuck solid, I stopped adding sugar and left it at ~22c for a week. I then added more sugar and put back in the incubator, this started fermenting again.
The fermentation with this one has always been pretty clear except for the slight tinge from the added sugar, it has made a very thin <1mm white layer on the bottom of the vessel when I stop the stirer and let it sit for a bit, I thought at first it was dead cells but the scope dosnt show ANY dead cells and I have no idea at the moment what the substance is.
I should mention this reactor has a top stir bar with two stir paddles on, one at the top and one I set near the bottom. When its going well the balls float when you stop stirring it, when it stops fermenting the balls dont exactly sink but they no longer float and bob about at the very top. So I take out of the incubator and leave in the cupboard for a week or two then add a little more sugar (5ml) and put back in the incubator, it starts up again.
Several of the unbuffered solutions died off after 5 weeks with pretty anaerobic conditions and very low PH levels, one went very wrong and I think it was contaminated with bacteria, I have tried to be super aseptic but hey ho........
I have kept the solutions in sealed containers but not done anything to them yet (I am planning on doing all the work ups together).

I originally set about trying to ferment by killing the yeast and isolating the bacteria from any that fermented. Unfortunately I couldnt find a source of the anti microbial I needed at a price I could afford so went with the yeast instead. I tried the doctors for the antibiotic I needed but apparently it isnt used for humans but mainly for brewing etc, I dont get this as it appeared the stuff should be available everywhere but I had trouble finding a source.
I am pretty sure I have more than just ethanol in many of them, some just smell different but in a nice way.
I havnt checked the ABV content yet but I am sure at times its gone over the magic 14%, until I do some tests I have no idea what happens at this point, the fermentation seems to stop but if left and cooled a while it can start up again.
Current theory is somehow when the ABV reaches the normal cut off point the yeast spore (this is all guess work for now) and the ethanol is oxidized or used in some way.
Once the alcohol content or whatever it is changes enough the yeast can start back up again. All this is speculation at the moment.

Two of the reactors have loads of different configuration possibilities so the next round of messing I might change a few things. I have loads of small sealed vials of the different yeast samples in a fridge, I keep plating and also putting in a broth so I can keep them going.
Next job is to find some more methods of testing what sugars are present (if any) and what alcohols are present. until the GC is fully working I will have to use the wet test's for primary,secondary and tertiary alcohols and boiling points etc.
it was meant to be 6 weeks tops as an experiment but I kept finding different combinations I wanted to try :D.
Ultimately I want to build a electrophosphresis tank especially for the yeast's (14 channel) and see which ones are related or have changed genetically from what I started with. I wont have the resolution or equipment to be that accurate but it should be goog enough to at least see if there are major changes.
If anyone gives a shit I will write it all up once I start working up the solutions. Also any suggestions on what test I can perform are most welcome :D.
I have pretty detailed notes but in the end it might have all lead to nothing much interesting, I have at least had fun and enjoyed it :D. the next big experiment is hydroponics and cell culture of tomatoe plants :D, which is already kind of underway.

If anyone finds a CHEAP source of cycloheximide give me a shout :D.

[Edited on 9-9-2015 by Little_Ghost_again]


I might have found a free 100mg sample :D not sure if they will send me some though. http://www.alomone.com/FreeSample.aspx?Product=53&Type=N...

[Edited on 9-9-2015 by Little_Ghost_again]