Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
experimental efficiency
Tom Haggen's question about "most time wasted on an experiment" got me to thinking about a related subject: experimental efficiency.
Most of the projects I have done at work have been carefully researched and aren't too much of a stretch from what is already known. Therefore
they have usually been successful, or at least not much time was lost. I couldn't keep my job if that were not the case. The amateur chemist
however may want to do something clear out in left field like making sodium. I'm wondering what would make such quests most efficient.
So far my list looks like this:
1. having general chemical knowledge
2. having specialized knowledge in the area of the proposed synthesis
3. careful research of previously successful methods
4. careful planning (on paper) including a flowchart with a logic tree
5. acquiring proper materials and equipment
6. frequent consults on MadScience forum
7. intermediate very small scale experiments to verify the chosen approach
I realize that we learn and have fun from our mistakes too, but homing in efficiently would give a lot of satisfaction.
What are your thoughts on this?
|
|
|
The_Davster
A pnictogen
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
Well I have used all those points that you suggested(except the flowchart one), never all in the same experiment though. I've never used the
flowchart one, what is it? Depending on what experiment is being done affects which points to use. For example; when all I did in chemistry was
pyro, for my first AP synth I used points #'s 1,3(had about 10 different peoples procedures),5,(no #6 because this was about 2 years before I
joined MSDB), 7. On the other hand when I first got into "pure" chemistry there were many instances of "I wonder what happens if I mix
X and Y?" which obviously only uses number 1. But for most of my experiments so far it has been all numbers but number 4 and sometimes without
number 2.
|
|
|
avtr01
Harmless
Posts: 19
Registered: 10-4-2004
Location: california, united states
Member Is Offline
Mood: curious i think
|
|
Homing in on efficiency would also make the chance for some kind of screw-up less prevalent. But then again, to much planning takes the fun out of
actually making the chemical. For some reason, it is that rush that we all get from seeing a reaction occur that motivates us to do more. An hour or
two of research followed by a succesful implementation would be very satisfying and enjoyable but days and days of research followed by a succesful
experiment might just fuel ones ego...maybe
|
|
|
DDTea
National Hazard
Posts: 940
Registered: 25-2-2003
Location: Freedomland
Member Is Offline
Mood: Degenerate
|
|
Lets not forget something extremely important--something you all learned in your basic chem classes: GRAPHING!
To illustrate my point, I will refer to a simple synthesis...Chloropicrin. Imagine we are trying to find the amount of excess Nitromethane that
would be best on the scale at which we are working. The patent says anything up to 8% will work. So here's one way to do it:
We do 4 reactions to produce Chloropicrin, each using the same amount of NaOCl and the same temperatures (don't add a second variable, as you
know). The first reaction uses 2% excess NM, the second 5% excess, the third 8%, and the fourth 12%. The percent excess will be the X-variable on
your graph, and the yields produced from the reaction will be the y-axis. In theory, it should form a parabola, since once you leave the sphere of an
efficient excess, the yields will decrease. Now all you have to do is calculate the maximum point on the parabola, representing the highest yield,
and find the theoretical "best excess." Then, try the reaction again with that excess and see how your yields compare.
The same process can be followed when find best temperature or best pressure or whatever else. In this way, you have plenty of experiments to do and
you are still doing your own research as opposed to following someone else's "recipe."
|
|
|
avtr01
Harmless
Posts: 19
Registered: 10-4-2004
Location: california, united states
Member Is Offline
Mood: curious i think
|
|
Good comment. Just like making ammonia industrially. Hauber and Bausch (i think those are the two scientists) found that the only way to make
artificial ammonia efficiently was at high temperature and high pressure using the same technique
|
|
|
Magpie
lab constructor
Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline
Mood: Chemistry: the subtle science.
|
|
flowchart
rogue chemist:
By flowchart I simply meant a logic chart something like you would construct before writing a computer program. This lays out the steps sequentially
with yes/no questions, limits, etc, and where you go from any given intermediate result. It could have loops that take you back to earlier steps with
only slight parameter change. During most experiments I suspect this chart would be under constant revision as indicated by results obtained.
Samosa:
A variant on your method is the "parametric sudy" where the minimum number of "experiments" can be determined based on the number
of variables involved.
I don't think these methods are anything we don't do intuitively. They are just a more formal, disciplined approach that might save time,
effort, and frustration. But they do take away spontaneity and require more "brain" work up front.
One clear advantage is that potential hazards may become obvious earlier in the planning stage.
|
|
|
The_Davster
A pnictogen
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
Oh thanks Magpie, I guess I am using that for some experiments. Ex, if I am successful making "x" then I can use "x" to make
"y". Or "if the reaction starts by itself good, if not I'll try heating it. But very rarely do I actually do it on paper, most of
its in my head.
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
