Sciencemadness Discussion Board - Predicting Reactions?

Predicting Reactions?

Xenon1898 - 29-1-2013 at 19:45

How wouuld a chemist go about predicting the products from competing rxns? For example if I have found these two reactions:

NaNO2 + NH4Cl -> NaCl + 2H2O + N2

and

NaNO2 + NH4Cl -> NH4NO2 + NaCl

Should I expect to get NaCl, N2, plus NH4NO2? Of course, performing the rxn and analyzing the products is one way, but for the generic case how would a chemist go about predicting rxn products in general? Do you need to guess all the possible rxns or is there another, perhaps more robust method (because you might miss some rxns when you guess)?

woelen - 30-1-2013 at 00:10

Both are ionic solids and hence you cannot make any difference between NaCl + NH4NO2 and NaNO2 and NH4Cl in aqueous solution, unless the concentration is so high that the least soluble precipitates from the solution.

An easier example is:

Person A dissolves exactly 1 mole of NaCl and exactly 1 mole of KBr in a liter of water.
Person B dissolves exactly 1 mole of NaBr and exactly 1 mole of KCl in a liter of water.

The two solutions are given to person C and this person has to tell which solution is made from NaCl and KBr and which is made from NaBr and KCl. Whatever advanced chemist he is and what advanced equipment and reagents he has, he will not be able to tell the difference between the two solutions. This is because both solutions simply contain Na(+), K(+), Cl(-) and Br(-) ions, which exist independently of each other.

-------------------------------------------------------------------------------

So, with NH4Cl and NaNO2 the same happens. You get NH4(+) ions in solution and NO2(-) ions, Na(+) ions and Cl(-) ions.

The first reaction is of a different nature. NH4(+) ions and NO2(-) ions react with each other, giving water and N2:

NH4(+) + NO2(-) --> N2 + 2 H2O

This reaction easily occurs with simple weak heating of the solution.

blogfast25 - 30-1-2013 at 11:46

What you expect to get will depend largely on which reaction leads to the biggest drop in Gibbs Free Energy (G).

For instance: aluminium powder reacts with calcium sulphate. Two reactions can be imagined:

A) CaSO4 + 10/3 Al == > Ca + 1/3 Al2S3 + 4/3 Al2O3

or:

B) CaSO4 + 8/3 Al == > CaS + 4/3 Al2O3

Even if there is enough Al present for A), the reduction doesn't proceed to Ca + Al2S3 because B) has a greater drop in G than A)...

[Edited on 30-1-2013 by blogfast25]

How do chemists do it?

Xenon1898 - 30-1-2013 at 20:09

I am asking my question from the perspective of a chemical engineer, because the way we are taught we basically have the chemical reaction equations handed to us, then we go and design chemical reactors to carry out the process, often at industrial scales. So I am just wondering how chemists (more to the point, the chemists on SM) do it? Do you folks generally look around for data on existing reactions and then see what you can do to reproduce those reactions? I see alot of people looking up patents or other sources for reaction data and trying to see if they can reproduce those (which is cool). Is anyone else doing other work (e.g. original research or other inquisitive investigations)?

elementcollector1 - 30-1-2013 at 20:32

Quote: Originally posted by Xenon1898

I am asking my question from the perspective of a chemical engineer, because the way we are taught we basically have the chemical reaction equations handed to us, then we go and design chemical reactors to carry out the process, often at industrial scales. So I am just wondering how chemists (more to the point, the chemists on SM) do it? Do you folks generally look around for data on existing reactions and then see what you can do to reproduce those reactions? I see alot of people looking up patents or other sources for reaction data and trying to see if they can reproduce those (which is cool). Is anyone else doing other work (e.g. original research or other inquisitive investigations)?

It seems to me that a lot of our work is improving theoretical yield of the reactions and such.

kavu - 30-1-2013 at 23:15

As for organic compounds one can say much just by looking at the functional groups in the reacting molecules. If incompatible groups are present side reactions are very likely. More than often side products can be tracked with TLC and crude NMR can give some insight to their structure. There are naturally many ways of shifting the portions of forming products. Solvents, additives, temperatures, reaction times and so forth can be screened to find the optimal results. As stated by blogfast25 ΔG can be used to evaluate things but this does not take into account the activation energies and thus kinetics of the reaction. Some product might have a nice big negative ΔG but that can't simply be reached due to mechanistic constraints.

AndersHoveland - 30-1-2013 at 23:55

Quote: Originally posted by Xenon1898

For example if I have found these two reactions:

NaNO2 + NH4Cl -> NaCl + 2H2O + N2

and

NaNO2 + NH4Cl -> NH4NO2 + NaCl

The answer is simple. Solutions of ammonium nitrite decompose when raised to a boiling temperature.

NH4NO2 --> 2 H2O + N2

yoshiharu198 - 31-1-2013 at 04:23

In first reaction
NaNO2 + NH4Cl = NH4NO2 + NaCl

after boiling temperature and STP NH4NO2 it possible it decompose.

woelen - 31-1-2013 at 05:04

What you write is not correct. You do not first get NH4NO2 and NaCl. In solution you simply have NH4(+), NO2(-), Na(+) and Cl(-) and you cannot tell the origin of all these ions.

On heating you get the reaction NH4(+) + NO2(-) --> N2 + 2H2O. The Na(+) ions and Cl(-) ions take no part in this reaction. These ions are so-called spectator ions in this system.

If you want to understand inorganic chemistry (or even chemistry in general), please try to understand the concept of ionic compounds and what happens when these are dissolved.

Xenon1898 - 31-1-2013 at 22:36

Quote: Originally posted by woelen

What you write is not correct. You do not first get NH4NO2 and NaCl. In solution you simply have NH4(+), NO2(-), Na(+) and Cl(-) and you cannot tell the origin of all these ions.

On heating you get the reaction NH4(+) + NO2(-) --> N2 + 2H2O. The Na(+) ions and Cl(-) ions take no part in this reaction. These ions are so-called spectator ions in this system.

If you want to understand inorganic chemistry (or even chemistry in general), please try to understand the concept of ionic compounds and what happens when these are dissolved.

I am not saying the two reaction equations qouted are in an aqueous solution. In fact these two specific reactions are not the question - I was just perusing a very long list of reaction equations and happened to notice these two as examples. They start with the same reactants, so just by looking at that list I can't pick one and say "ah, that's the way it's going to turn out because that's the one I picked! Yay, lets do it!" I am thinking that something could go in a direction I didn't expect, just wondering what people do to help them predict the outcome.

That's all the info they had in the resource, it's just a chemical reaction equation list. I realize the reaction conditions (temp./press. profile), kinetics, possible catalysts, etc. could all affect reactions, so it's hard to predict based on just a couple of equations. I just happened to notice these two together and I thought I would ask this to see if any chemists on this might think about a given reaction going in a direction they don't expect, what do you do to predict things ahead of time? I suppose lots of research on previous work is the main way. I was kind of wondering if there are any calculation type steps chemists go through to help predict reactions. It can't all be based on what someone else did, how would there every have been any discovery of new reactions?

I happen to think there is a whole bunch of chemistry knowledge waiting to be discovered. I am trying to identify any tools and techniques chemists might employ to help them shine a flashlight on the hairy edge of current knowledge. I took a bunch of chem classes but not anything like experimental chem.

AndersHoveland - 31-1-2013 at 22:53

Some reactions have multiple products, which are not in whole integer ratios. For example, consider the reaction of methane in the presence of a limited quantity of fluorine. All sorts of different products will form. To some extent the ratio of the products is dependant on the reactant ratios and reaction conditions.

To predict what will form, one must understand what is happening in the reaction. In some instances it can be difficult or essentially impossible to know how a reaction will actually proceed without conducting an experiment and measuring the products. Usually, however, we can still make a good prediction, especially if one is familiar with the particular field of chemistry in question.

Nicodem - 1-2-2013 at 10:29

Quote: Originally posted by Xenon1898

I'm not sure I understand your question. Are you asking why members of SM don't (or almost don't) post any scientific research? There can be many reasons why members here focus on technique training, reproducing published procedures and optimizing methods, as opposed to doing scientific research, total syntheses or new method development. One of the most important reasons is surely due to limited resources: you can't really do much relevant scientific research in chemical synthesis unless you have some analytical equipment. Other reasons, varying from one individual to other, could be:
- no affinity toward science;
- shyness;
- focus on a specific product synthesis;
- lack of chemicals and glassware;
- reserving the best achievements for academic publications or posting results elsewhere on the net;
- lack of knowledge;
- poor skills in the scientific method;
- poor literature searching skills;
- beginner's mentality, etc.

But it must be said, that you are looking from the wrong perspective. There are truly some fantastic achievements posted by some members on this forum, especially so by considering the how, where and by whom they were made!

Quote: Originally posted by Xenon1898

One of the two equations is of a reaction that does not take place!

Quote:

They start with the same reactants, so just by looking at that list I can't pick one and say "ah, that's the way it's going to turn out because that's the one I picked! Yay, lets do it!" I am thinking that something could go in a direction I didn't expect, just wondering what people do to help them predict the outcome.

Reactions starting from the same reactants but giving different products are relatively common. The equations can thus be identical on one side of the arrow. I don't understand what's so strange about it. What would be strange is that you would obtain different products by following identical procedures. It is the procedure that determines the product, not just the reactants.

Quote:

That's all the info they had in the resource, it's just a chemical reaction equation list.

Such a list of equations is totally useless. Anyone can come up with a list of balanced equations that mean absolutely nothing. A balanced equation says absolutely nothing on the viability of a reaction it claims.

Quote:

I was kind of wondering if there are any calculation type steps chemists go through to help predict reactions. It can't all be based on what someone else did, how would there every have been any discovery of new reactions?

The field of chemistry that deals with reaction outcome is called chemical synthesis. Contrary to what some believe, it is a science. As such it has theories. Synthetic chemists rely on several important theories, but in practice the one that is always used (usually in combination with others) for evaluating the possible reaction outcome is the theory of reaction mechanisms.

thanks

Xenon1898 - 4-2-2013 at 19:03

After poking around on the Internet for a couple weeks I have come to the conclusion that general chemical reactions are too complicated to predict, even with modern computers cranking away (e.g. ab initio methods). So much for that project!

amazingchemistry - 4-4-2013 at 20:08

Xenon,

A lot of times chemistry research is just 'lets throw this together and see what happens.' The only differences with the past (when chemists like Lavoisier and Proust really did just throw stuff together and watched) is that thanks to people like them we have gotten a bit smarter as to what to throw together. For example, elements with similar properties will react similarly in simple reactions. We have also developed amazing methods to analyze what happens after we throw stuff together. So it isn't just guessing, it's educated guessing. Chemical discoveries are filled with stories of chemists who were trying to make A (based on previously done reactions with similar compounds) but ended up with B instead. The part where both your basic chemistry training and your independent literature research come into play is when you're trying to figure out what is LIKELY to happen, and how to recognize when a 'mistake' is really a 'further research opportunity.' Let's say you were trying to make nitric acid from ammonia. You know you need to oxidize the ammonia somehow, so you need a source of oxygen. You also know how tightly bound the hydrogens are to the nitrogen, and if they are tightly bound you are probably going to need a lot of energy to break the bonds. This suggests high temperature, or catalysts, or both. After teasing out more or less what is likely to happen, you can ask "has this been done before?" Note that this doesn't necessarily mean that you're looking for the exact reaction you are attempting, but for the critical bits you determined earlier. You could say, for example, "okay, this paper says they tried this catalyst and it worked to break the bonds of compound X. Compound X is similar to what I have, so maybe the catalyst will work in my reaction" This is where many, if not most, of the chemical equations we know come from. Textbooks often say things like "the Haber process can be summarized by the following equations..." and then proceed to give you a neat little list of balanced reactions. While the finished equations may look simple, they say nothing about the amount of time and effort it took to come to them (Haber for one was awarded the Nobel Prize for his work). If what you're doing works, great! You have just accomplished original research. If it doesn't (and a lot of the time it doesn't) maybe something interesting happened while you were trying things out. Finding out why something DOESN'T happen can be just as useful as finding out why something happens.

AJKOER - 5-4-2013 at 05:19

I remember a paper which look at some twenty possible reactions involving the treatment of water with chlorine in the presence of NH3. Pressure, temperature, ionic strength and pH were all changing as the reaction chains progressed. Now, those expecting all the ducks are in a row based on Gibbs free energy are wrong. The author calculated it for each reaction, and even commented it is only a rough guide as he monitored the apparent actually paths taken. My recollection was that the best predictor, believe it or not, was the number of moles of gas produced for those reactions with pressure considerations. Those reactions producing fewer moles of gas in a pressure environment, despite unfavorable Gibbs free energy indications, were, in the majority of cases (not a perfect predictor), being observed. An eye opener for real world chemists.

In my opinion in these real world problems, the issue is more likely lack of understanding of the exactly how reactions are occurring (see some environmental studies on the formation of nitrates, NO2Cl, O3,..). The process seems to be first observe and then backtrack and postulate/verify a free radical path promoted most likely by uv radiation. And, if you still have faith in our current state of knowledge, look at those atmospheric studies noting some unique chemistry (one author refers to it as 'novel chemistry') that appears to be occurring between gases (including water vapor) on the surfaces of porous solids not observed in the usual aqueous environment.
----------------------------------------------------------

Now, if you really want to understand what could happen and save your body parts, forget the ionic argument, while true, it may make it harder to quickly assess what is about to happen to you. Take your example:

"NaNO2 + NH4Cl -> NaCl + 2H2O + N2

and

NaNO2 + NH4Cl -> NH4NO2 + NaCl"

In my opinion, what you need here is actualy more reactions, not ionic summarizes. More precisely:

NH4Cl (s) --Heat--> NH3 (g) + HCl (g)

NaNO2 (s) + HCl --> HNO2 + NaCl

2 HNO2 --Low Ph or Heat--> H2O + NO + NO2

2 NH3 + H2O + NO + NO2 --> 2 NH4NO2

NH4NO2 --Low Ph or Heat--> Boom, you're Dead, you have just detonated the High Explosive Ammonium Nitrite.

In fact, I recently mentioned in the Energetic section that NaNO2 plus NH4Cl is employed commercially as an explosive. Do you really think ionic summaries would have warned you not to heat these salts and the actual results about to befall you?

Expanding the reaction sequence and noting conditions for reactions (pH and temperature), and knowledge of some of the particular properties of these intermediate compounds, could have just saved your life.

By the way, if you are some arrogance educator with no real experience, and still profess something different from what I stated as it was some weird solid state reaction, you deserve to perform this reaction in an aqueous manner and concentrate the solution. Then, perhaps after an 'incident' you can (if so able) re-read my comments on the 'properties' of the possible products, and, in particular, on aqueous Ammonium nitrite behavior in acidic and/or concentrated aqueous solutions.

[Edited on 5-4-2013 by AJKOER]