Common Mistakes: How not to draw resonance curved arrows

by James

in Organic Chem Study Tips, ResCC, Understanding Electron Flow, Where Electrons Are

No discussion of resonance structures would be complete without mention of how to royally screw them up. This isn’t something to feel bad about, by the way: there isn’t a chemist alive who hasn’t made one of these mistakes at some point. Think of it as a rite of passage. The trick is to make the mistakes while doing problems, not while doing an exam.

There are at least three common categories of mistakes regarding resonance structures:

  • Unbalanced equations
  • Moving atoms around
  • Incorrect drawing of resonance arrows

Let’s first talk about unbalanced resonance equations, where something (either an atom or electrons) has been added or subtracted. Remember that in drawing resonance forms we’re only allowed to move electrons, and nothing more. That means that the two resonance forms can neither differ in the number of their electrons nor can they differ in the number of atoms.

Moving atoms around is a second category of common mistake. Although the two structures shown below have the same number of atoms and electrons, they are not resonance forms because we have broken single bonds (as opposed to π bonds) and thus moved the location of one or several atoms. The easiest way to screw this up is to move hydrogens. While these molecules are related, they are actually pairs of constitutional isomers, not resonance structures.

One way to avoid making these types of mistakes is to try to interconvert the structures using curved arrows. There are only three legal arrow-pushing moves for drawing resonance structures. Double check to make sure you aren’t breaking the rules.

The last – and by far the most common class of mistake in drawing resonance structures is to screw up the curved arrows. There is a seemingly infinite number of different ways to do this. They fall into a number of sub-categories.

First, there’s arrow-pushing moves that are wrong and cannot be redeemed. Examples A-D each depict different ways of breaking the octet rule. In A, B, and C the resonance form that would result from these arrows would have five bonds to carbon. Example D would have five bonds to nitrogen. Inconceivable! 

Examples E and F  are wrong for a different reason: remember that the curved arrow depicts the movement of a pair of electrons. In example E, the “tail” of the leftmost arrow is shown at a positive charge – a big no-no, since there isn’t a lone pair of electrons here. Likewise for F, where the positively charged nitrogen also lacks an electron pair.

Then there’s arrow pushing “moves” that are also illegal, but can be made legal through drawing an additional arrow. See if you can draw an arrow to make it work (answers at the bottom).

Then there’s the sloppy mistakes, where these arrow pushing forms are missing something important. I guess you could say this entire post is devoted to sloppy mistakes but these examples are particularly egregious because they are just one tiny little detail away from being correct. In these two cases, there is neither a lone pair of electrons (or a formal negative charge) at the tail of one of the electron-pushing arrows, which make them incorrect. Neglecting to draw the formal charge of an atom is another common sloppy mistake (albeit not unique to resonance). Note that when I say sloppy I’m not making a moral judgement here. I’m just saying it makes for imprecise and ambiguous chemical structures, which are not useful.

Finally, there are resonance structures which are not illegal, per se, but  won’t make a significant contribution to the resonance hybrid.

In both examples we have very electronegative elements (oxygen and nitrogen) with less than a full octet. Recall that electronegativity is a rough measure of the ability of an atom to stabilize negative charge? Well, the converse is true – that is, the greater the electronegativity, the more positive charge will be destabilized on that atom (clarification: by “positive charge” here I am specifically referring to having less than a full octet of electrons (like a carbocation), not the common situation where O or N with a full octet bears a formal charge of +1.)

Avoiding all of these mistakes requires careful attention to detail, bordering on paranoia. The number of atoms and electrons on the left side of the resonance arrow should balance the number of atoms and electrons on the right side of the resonance arrow. Furthermore, the changes in bonding (and charge)  of the molecule on the left side of the arrow should be accurately mapped by the appropriate curved arrow(s).

If it sounds like I’m making a case for organic chemistry being a lot like accounting, you’re right!  In the final analysis, organic chemistry equations are not unlike accounting transactions. The two sides need to balance.

Next Post: How To Apply Resonance and Electronegativity To Understand Reactivity

P.S. Here’s the answers for the example above:




Related Posts:

{ 8 comments… read them below or add one }


Don’t mean to be a gadfly, but for the acylium ion in “close but no cigar,” doesn’t that depict illegal atomic movement? ;-) Great post, James. MANY instructors have been driven nuts by these mistakes (as I’m sure you’re aware). Be nice to your orgo professor next semester, kids, and avoid these common pitfalls!



Thanks for the spot! You’re right, need to fix this.



So far I’m finding this to be a good site to recommend to my students. Good job!

[Your answer to “C” is missing a charge in the right-hand structure.]



Oh snap. Thanks for the spot. Fixed!



Hi! This site has been very helpful, and I appreciate the time you took to make these summaries.

I just have a question about A though. Wouldn’t the resonance structure on the right have a positive charge on the bottom carbon?



Aw shoot! A mistake on the “common mistakes” post. Thanks – fixed



Hi much appreciation on your work!!

But, just wanna clarify on your Example 3, I think they are technically ok for these two. Why you still say CLOSE but not quite? Do the N and O assumingly have lone pairs in these two cases? Would you clarify it? Thanks a bunch!



Aha, I figured it out. You DID answer it in the context afterward. Never mind. Thanks, Mi


Leave a Comment