APPENDIX
Pushing Electrons
"pushing electrons",
The manner in which organic chemists write mechanisms is known as
is extremely
electrons
at
pushing
proficient
Becoming
flow".
or"electron
"arrow pushing" ,
us from
can
keep
mechanisms,
communicate
to
pictorially
us
it
allows
becluse
important,
*ri,ir-rg.rrrreasonable mechanisms, and can guide us in the choice of experiments. Teaching
electhe fuidamentals of this notation is the goal of this appendix. Hereiry the concepts of
examples
and
many
with
resonance/
along
covered,
will
be
sinks
electron
and
tron sources
of correct and incorrect arrow pushing'
A5.L
The Rudiments of Pushing
Electrons
"
When drawing reaction mechanisms, chemists pictorially represent the movement of
the
bonds, electrons, or electron pairs using an electron-pushing notation. Understanding
compounds
the
of
stability
basic principles behind pushing electrons a1lows us to analyze
.ri^ r"'rorrut-t"e (see Chapter 1), ind to communicate conceptually how reactions are occurring. The arrows used tb denote how electrons are moving in chemical reactions are drawn
witiin a chemical structure or between chemical structures. An arrow with a fulI head on
end is used
one end denotes the flow of two electrons. An arrow wilh only one slash on one
curved or
arrows
are
often
these
to denote the flow of only one electron (see margin). Both of
A
reaction
that
is
straight.
S-shaped, and therefore are totally different than a reaction arrow
(or
chemical
step
several
arrow connects a reactant to a pioduct, and denotes a chemical
is one
steps), and does not imply a me_chanism. A fourth kind of arrow that chemists use
in
ways
two
different
with iutl heads on both sides. This is a resonance arrow denoting
two
notation
involves
arrow
which to draw the bonding in a chemical structure. One other
the reactant
arrows pointing in opposite directions. This represents an equilibriumbetween
/-\
Flow of two electrons
a-.
Flow of one electron
Reaction arrow
<..-.-.--------->
Resonance arrow
and
- product.
not
pructice of electron pushing is solely a bookkeeping method-a notation. It does
arrow
headed
,"pr"r"r,jt the real movement of eleittotrt. This means that the use of a double
io'rno* the flow of two electrons d.oes not literally mean that electrons are actually moving
Nevertheless, the notation is
uro"r"ra, within and between molecules in the matter drawn.
indicates how discrete bonds
useful within a valence bond theory (VBT) contex! because it
;;;i;"" pairs have been rearranged when comparing the reactant to the product. the drawIn
As stated above, the full headed arrow represents the flow of two electrons.
and the head of the
ing of this notation the tail of the arrow is plaied near the electron source
th"
Equilibrium arrows
_
to the electron sink (seveial examples of sources and sinks are givenbelow)'
is always some form of two electrons, such as a lone pairof electrons or a
source
An electron
accept a
o/o.Uorra. An electron sink is always an atom or bond within a molecule that can
pair of electrons in the form of a lone pair or an additional bond.
urio* i, pointed
i
/n
Tail emanates
from an electron
source
Head terminates
at an electron
sink
7067
1062
APPENDIX 5: PUSHING El,ECTRONS
A5.2 Electron
Sources and Sinks for Two-Electron Flow
Each and every arrow must start at an electron source and end at an electron sink. Therefore,
before analyzing electron-flow procedures, a list of common electron sources and sinks is
useful. Table A5.1 shows several electron sources and sinks. The sources are listed first. They
often consist of lone pairs of electrons on heteroatoms, and the atoms can be either negative
or neutral. The neutrals include, but are not limited to, alcohols, water, amines, and thiols.
The anionic examples include, but are not limited to, alkoxides, amide anions, hydroxide,
and thiolates. In all of these examples the lone pairs of electrons are the actual electron
source, not the entire chemical structure itself.
Table A5.1
Common Electron Sources and Sinks
I. Sources (R
:
H, alkyl, aryl)
Nonbonding Electrons
A. The lone pairs on the heteroatoms on the following strucfures: X ; RO-,
RrO, N&, NR;, RCO;, RCOr& R(CO)& R(CO)NRr, RrS, RS-, CN , Nr-,
R.P, RzSe, and RSe-
Electron Rich or Strained o Bonds
A. Organometallics: RMgX, RLi, RrCuLi, and R2Zn
B.
C.
Hydride reagents: LiAlH4, NaBHa, NaH, and BH3
Cyclopropropyl or cyclobutyl
n Bonds
Alkenes, dienes, alkynes, allenes, and arenes
Electron Rich n Bonds
Enols, enamines, enolates, anilines, and phenols
II. Sinks
(R: H, alkyl, aryl)
Species
with Empty Orbitals
Carbocations, aluminum and boron containing Lewis acids, and transition
metal (mercury, cadmium, and zinc) reagents
Acidic Hydrogens
Mineral acids, carboxylic acids, watet, alcohols, amines, and terminal
acetylenes
Weak Single Bonds
Peroxides (ROOR), molecular halogens (X), bieach (HOX), and disulfides
(RSSR)
The Carbons in Polarized o Bonds
Alkyl halides, alkyl tosytates, protonated alcohols (ROHr-), and protonated
amines (RNH3-)
The Carbons in Polarized Multiple Bonds
R2C:O, nitriles, ct,B-unsaturated carbonyl compounds, acyl halides,
anhydrides, esters, and amides
Sigma bonds between carbons are not normally electron sources, because the energy of
the electrons in these orbitals is too low for them to be involved in common chemical reactions. Examples of o bonds that can be electron sources are highly strained bonds such as in
A5.2 ELECT,RON SOURCES AND SINKS FOR TWO-ELECTRON FLOW
cyclopropane, and o bonds that are highly polarized and have. a partial or full negative
charge on carbon. These include lithium and Grignard reagents. However; if the sink is electrophilic enough, an unstrained o bond can be a source (see the carbenium ion arrange-
mentbelow).
\'--ro
-/- +
o /
-T-
In contrast to o bonds, simple n bonds from alkenes, alkynes, conjugated alkenes, allenes, and aromatic rings are often electron sources. Pi bonds are significantly better sources
when they are electron rich due to an attached electron donating group. These donating
groups are heteroatoms with lone pairs of electrons for which resonance structures can be
written showing negative character on a carbon of the n bond (see the enamine below and
see the next section on the details of denoting resonance structures).
a
6
<------------>
o
.il.r
^
o'"
Note that all the electron sources in Table A5.1 are also nucleophiles. As discussed in
Chapter 10, nucleophiles can donate electrons to a positive or partially positive atom. All nucleophiles are electron sources.
However, not all electron sources are considered to be nucleophiles. For example, consider the deprotonation of hydronium by ammonia. One of the arrows used to depict this reaction involves the ammonia lone pair as an electron source and a hydrogen of hydronium
as a sink. The other arrow denotes that the H-O o bond is an electron source to create a lone
pair on the positive oxygen, which is the sink for this arrow. We would not consider the H-O
o bond to be a nucleophile, although it is a source in this example.
H
u/:
tt
N:
\\\
H
"I{ot, -
t8-,
u('
H
:o
H
Another example is the deprotonation of a carbon ct to a carbonyl by ethoxide. In this reaction the ethoxide is an electron source, but so is the C-H o bond and the C-O a bond. In this
example the C-H bond would not be considered to be nucleophiiic.
(\)
€\ ^trAll
*
,^d,'
\
"\.,/
-^O-H
o
:d
I
Chemists do not always draw all the lone pairs on heteroatoms, and it is important for
the reader to realize that the lone pairs are there. When the lone pair is used as a source we
always draw it, and when a lone pair is created we will draw it. This is the convention used
in this textbook. Very oftery however, chemists stop drawing a1l lone pairs, even when they
are sources/ and you should recognize what is meant by the arrows.
Electron sinks are more varied in nature than electron sources (several are shown in
Table A5.1). Anything with a positive charge and/or an empty orbital is an electron sink.
Carbocations, Lewis acids, and metal cations are good electron sinks. Atoms involved in o
bonds can also be electron sinks. The proton in a Bronsted acid or weak single bonds between pairs of heteroatoms often act as sinks. Atoms on the partial positive end of polarized
bonds are common electron sinks-namely, the carbons in carbonyls, nitriles, conjugate acceptors, and alkyl halides. All of these sinks are also electrophiles.
Howeve{, sinks are not necessarily the same as electrophiles. The sink is always an atom
that can accept a negative charge and still be relatively stable. To explain what is meant by
this, consider the nucleophilic attack of cyanide on the carbonyl of acetone. The source for
the first arrow is the lone pair of electrons on the carbon of cyanide, and the sink is the par-
r063
L064
APPENDIX 5: PUSHING ELECTRONS
tially positive carbon of the carbonyl.
o\
v
- ,t'
o)
N=C:-
''6:
.j)
+
,.r\
*
N'
Howevet we cannot stop there. Since the drawing of the arrow from the cyanide source to
the carbonyl sink implies that a bond is being formed. between the source atom and the sink
atom, in this example five bonds to carbon would be formed. Hence, another arrow is
needed to show thai some bond to this carbon must also be breaking. In this case the bond
that is breaking is a n bond (one of the traditional sources), and the sink is the oxygen of the
carbonyl, *heie a ione pair is being created. Chemists would not consider the oxygen as an
electrophiie, although ii is an electrbn sink in this case. This dichotomy in the terminology is
the reaion that the tJrms "source" and "sink" are associated with electron pushing, whereas
the terms "nucleophile" and "electrophiie" are used to describe reactivity patterns.
The sink is always the place where the head of the arrow terminates. Confusion can arise
as to the exact place that the arrow should terminate. When the sink is a heteroatom that is
accepting a lone pair, place the head of the arrow near that heteroatom. Similarly, when the
sink is ai elect.ophilic atom, just place the arrow head near this atom. However, when the
sink leads to the fbrmation of a new bond, there are two acceptable conventions on where to
place the end of the arrow. You can either place the arrow head between the two atoms where
the new bond will be drawn (A, below), or place the arrow head pointing towards the atom
where the bond is being formed (B, below). Both methods are acceptable.
B.
A.
-H
O1
t :O\
/^k
O--v:
I
H-d.
U
,coo
),,
In all mechanisms that we write, the total charge from one step to the next never
to a
changes. This is called the conservation-of-charge rule. If an anionic nucleophile adds
a
leaving
neut[l compound, the productmustbe negative. If an anionic intermediate expels
If
gro.rp'urrd U"comes neutral, the leaving gtorlp must be departing with a negative charge'
a
negaor
i tr",rt.ul compound fragments, it must do so to cleate either two neutral species
to
tive and u poriti.r" o.r". fhit rule of maintaining the net charge from one set of structures
ale
plausible
the next is very useful for checking if the steps being proposed in a mechanism
or not. If the charges are not maintained, then something is wrong'
A5.3 How to Denote Resonance
Resonance was defined in Chapter 1. We commonly accompany a collection of resonance
structures with electron flow uiro*, to describe how two electrons were moved around,
thereby creating the new bonding arrangement. As an example of using arlows to depict
and
resonance, .orrr]id", acetate. An arrow starts at a lone pair on the single-bond oxygen
to
formed
be
will
bond
terminates near the carbonyl carbon. This arrow states ihat a second
pair
a
lone
becomes
C:O
the carbonyl carbon. A sectnd arrow d,enotes that a bond in the
on oxygen.
O:o
'-\*i!/ €
o.
\-
c,'d
A5.4 COMMON ELECTRON-PUSHING ERRORS
We noted in Chapter 1 that dravi'ing resonance structures often gives insight into the re-
activity of a molecule. For example, consider the resonance structure of the cation shown below. In the first depiction there is a positive charge on the carbon and this carbon has an
en;rp|r p orbital. Therefore, we predict nucleophilic attack at the carbon is possible.
**X
A
However, directly adjacent to the empty orbital is an oxygen atom with lone pairs of electrons. Since the carbon is lacking an octet of electrons, a neighboring electron pair can donate
into the adjacent orbital to give a ri bond, thereby stabilizing this structure. We denote this
donation using a double headed arrow as shown. The arrow starts at the lone pair and ends
in between the carbon and oxygen, showing that a n bond is formed. Thus, the electron
source was the lone pair and the sink was the empty carbon p orbital. Since no atomic movement has occurred and only electrons have been moved within the molecule, we have created a resonance structure.
Resonance structures most often involve p orbitals that are in conjugation. For example,
the molecule acrolein has its'n"bonds in conjugatiory and the resonancJstructure shown indicates that there is some n bond character between the two central atoms.
o-P'
,o\
rr"
an
A5.4 CommonElectron-PushingErrors
There are several errors that are common for students when they are first learning to use
electron-pushing notation. It is instructive to cover many of the common mistakesio as to
spotlight them.
Backwards Arrow Pushing
Likely the most common mistake is pushing the arrow backwards. In other words, the
arrow is started at a sink and ended at a source. Three examples are givenbelow. The easiest
way to avoid this mistake is to remember that the arrow must start from an electron rich
region of a molecule. Most important, the arrow always starts with two electrons-namely,
lone pairs, o bonds, or t bonds. Do not use the positive regions of a molecule to start an
arrow. The vast majority of the time the arrow will terminate at a center with some positive
charge or a center that can accept a lone pair.
*-3.n
nr ry'l
,\o,'
"9-*
o^
.At\
/< r^N\-o
\\
n,o.H
(+)
U
Common backwards electron flow (incorrect)
Not Enough Arrows
The second most common error is to not show enough arrows. As an example, consider
the electron pushing shoum below that is meant to indicate an E2 reaction. The base is abstracting the proton while the leaving group is departing, but there is no arrow to denote the
formation of the double bond. The easiest way to avoid problems like this is to remember
that each arrow leads to the formation of a bond or a lone pair of electrons. Hence, keeping
track of where the arrow starts and terminates defines either a bond between th" r"rpe"tirre
atoms or a lone pair localized on a specific atom. Using this analysis on the example given
1065
7066
APPENDIX 5: PUSHING ELECTRONS
shows that the base is forming a bond to hydrogen, but since no bond to hydrogen is shown
as breaking, the result is two bonds to hydrogen. Furthermore, since the bond to the leaving
group is breaking to form a lone pair onbromide, and there is no arrow shown to fiIl the void
left behind, a carbocation must be forming.
\ rBr
\_J
\(,
O:v
o
:Br
As drawn, this electron flow
denotes this product
r,^
H
I
\o
/-
.t-.YH
lmplausible
Losing Track of the Octet Rule
Another common mistake is showing arrows that create atoms with electron counts
above and beyond an octet. As examples of this, consider a resonance structure of nitrate
and the attack of cyanide on the oxygen of an oxycarbonium ion. Both examples indicate
that an electron source is quenching the positive charge of an electron sink. But, these positively charged atoms are not electron sinks capable of accepting two elect{ons in the form of
an additional bond. In each case the number of electrons on the respective sink atoms increases to ten, two beyond an octet.
oo <----->
il9
^,":O' -O
oo
-ilO- rc
lncorrect
N-o
'-C"^,
A'o.
A*
N
t'C-oll
.\lncorrect
This problem is quickly remedied when we remember how many bonds each atom can
form and that we do not routinely draw the lone pairs of electrons. inboth examples given
above, the number of bonds increases beyond what is allowed (that is, five to nitrogen and
four to an oxygen possessing an undrawn lone pair). Therefore, although nitrogen and oxygen are positivelyiharged in these examples, they are not electron sinks for formation of ad-
ditionalbonds.
Losing Tiack of Hydrogens and Lone Pairs
One of the easiest mistakes to make is to forget that hydrogens on carbons are not drawn
in stick structures. This can often lead to the movement of an arrow to a carbon that already
has four bonds, although at first glance the carbon may seem to be a reasonable electron sink.
In the example shown below, the first resonance structure given is simply incorrect because
a double bond is drawn to a carbon that already has four bonds. A second problem also exists. The oxygen is left without an octet, and should therefore have a plus-two charge; no
lone pair wis added to give the oxygen eight electrons. The real result of the electron pushing is also shown. It is an implausible structure. The only way to avoid these mistakes is to
mentally take note of the hydrogens that are not drawn. Moreover, if you draw all the lone
pairs when embarking on an electron-pushing exercise, the latter of these two problems will
be much easier to spot and avoid.
.o
r'oxl
\ -/" '----*
o
lncorrect
resonance
structure
.'.)zu)
\\___]L9.
A lmplausible
resonance
lFt
H Structure
.
A5.4 COMMON ELECTRON.PUSHING ERRORS
Not Using the Proper Source
A harder error to spot is the use of the wrong electrons
as the source. As an example, in
catalysis
is
covered.
In
some
cases
the
base deprotonates water while
general-base
Chapter 9
adds
to
a
carbonyl
carbon.
The
best
source of electrons for showsimultaneously
the water
to
the
carbonyl
sink
is
a
lone
pair
the oxygen of water. Howof
electrons
on
addition
ing the
could
alsobe
depicted
as
inMechanismA5.l.
shown
As showninthis
evel the electronflow
the
H-O
o
bond
is
the
scheme,
electron
source
for
forming
a bond to the
electron-pushing
carbonyl carbon sink. This is not so serious because electron pushing is just a bookkeeping
notation, and the number of bonds and lone pairs is correctly portrayed in this scheme.
Howevel it is not the best reflection of how one thinks about this nucleophilic attack.
In Mechanism A5.2, the lone pair on oxygen is correctly shown as the source for the carbonyl sink, but the deprotonation of the water is incomplete. Instead, two bonds are being
formed to hydrogen because there is no arrow showing the breakage of the O-H o bond. The
best method is shown in Mechanism A5.3.
\^
. .-?',-G''- nlo -
H-N:zNrH
\
\-U
.A
q1
H
H
(Mechanism A5.1)
lncorrect source for the arrow
depicting the nucleophilic attack
/ \^
u-N. .-N:
\
Ch
--z'
,l
H-d'-----/
I
Not enough arrows; this
indicates the formation
of two bonds to hydrogen
(Mechanism A5.2)
H
tn
\
H- N-.ZN'
--,2/zt
+
g161'-----/
I
H
/ \o
.(.
.. _o-
H- N.,r-N-H ,o4
"I\
(Mechanism A5.3)
H
Best electron-pushing notation
Mixed Media Mistakes
An important rule to remember with arrow pushing, much like on the bench top, is to
not mix strong acids and bases. If the reaction is performed in acidic media, it does not make
sense to show the creation of a strong base in your mechanism. Similarly, if a reaction is run
with added base, it is unreasonable to create a strong acid in the mechanism. Significant concentrations of strong acids and bases cannot co-exist in the same medium. For example, any
medium that is acidic enough to protonate a carbonyl reactant would not have any appreciable hydroxide. Two examples of this mistake are given below.
Hydration of a ketone
l'3''
9"il+l'*
,\
,e?*
lncorrect: Usage ol a
strong acid along with
a strong base
,€i'
1,2-Elimination
,,o'H
\-l
.?o'\
@o-H
ll ,,
,t'o'"
.H
Too Many
qol
lncorrect: Explusion of a
very basic leaving group
with the concurrent formation
of a strong acid
Arrows-Short Cuts
The iast error that we cover is the use of too many arrows in too complex of a scenario. It
is sometimes tempting to combine several steps together into one step as a means of taking a
short cut to the product. For example, consider the pinacol rearrangement, shown below.
1067
1068
APPENDIX 5: PUSHING ELECTRONS
The electron-pushing notation does indeed lead to the product, but ihis proposed pathway
is chemically unreasonable. It involves the breaking and forming of several bonds simultaneously. The entropy disadvantage of such a reaction would be quite large. Such a combination of steps is not what chemists observe when studying these reactions (for the correct
scheme see Section 11.8).
n-Q-H
LY
H
rl
.9-H
)o
*)1 ,/\
H-o) oi H\ /*'
7--\
The second example involves an acyl transfer from chloride to water. The addition and
elimination reactions are combined into one step. Although the arrows do keep track of the
electrons involved in the reaction, such steps are known not to occur simultaneously, and
thus the electron-pushing notation does not reflect what is known about the mechanism.
rol
'.|/^
.'\cr'
+
..)
O
A&"
H-QH
H
A5.5 ComplexReactions-Drawinga
Chemically Reasonable Mechanism
writing a reasonable mechanism when a chemist is first
confronted wiih the producl of a reaction. All the chemical intuition that the chemist has,
built upon past experience, is used to create the mechanism. Any available experimental
data, any knowledge as to the feasibility of intermediates, and just a "gut" intuition are often
the stariing points. In actuality, for most chemists, the "gut" intuitive feeling is based on
either a conscious or unconscious recognition of a logical electron-pushing pathway. In
other words, when first considering a reactiory most chemists apply the rules of electronpushing notation to visualize a logical sequence of chemical reactions that can lead to the obserrred pr6duct. These steps are then examined in light of the experimental data. If there are
data, the electron-pushing analysis is used to create hypotheses thatcanbe
,-ro
"*p"ri*"ntal
Hence, being able to apply the electron-flow rules to completely new
experimentally.
tested
a sophisticated organic chemist'
is
one
mark
of
scenarios
into words ihe mental process that a chemist goes through
to
translate
is
difficult
It
for a new reaction, especially since undoubtedly every
with
a
mechanism
up
when coming
we attempt to do just that here. A few simple rules
However,
does
it
differently.
chemist
There are many factors that go into
will
assist.
1. Find a1:'J. correspondencebetrneen all atoms in the reactsnts and theproducts.
This may lead you to ask a question such as "Where does this oxygen come
from?"
2. Keep your rnind on where you are going.
In other words, iookior a pith ihut -ltt lead to the product. To do this, note which
groups have added to or left from the reactant, and make sure that such steps are included in the mechanism.
3.
Measure yoLff progress at intermediate stages bnsed upon how many bonds stiT need to
be
formed
or broken.
4.
Note any
rearrangement of atoms utithin the chemicsl structure and ffiake sure appropriate steps
qre inclttded.
A5.6 TWO CASE STUDIES OF PREDICTING REACTION MECHANISMS
5. Do not push too many srro'Lt)s as a wiy to crehte a short cut to the proclttct.
To do this, always stick to common reaction steps such as those presented in Chapters 10
and
11.
5. Aaoid the common electron-pushing mistakes.
7. FinaIIy, do not form any intermediates'of unreasonably high energy.
Similarly, do not form high energy intermediates when other intermediates of lower
energy are possible. This is a more difficult analysis to make. Howeve4, the information
and examples given throughout this text should provide excellent guidance.
If you can remember all of these items, you should be able to write out mechanisms that
are both chemically sound and often prove to be correct after an experimental analysis. TWo
examples of complex chemical reactions, along with a discussion of how to proceed in writing a mechanism, are discussed here, and several practice problems are given at the end of
this appendix.
A5.6
TWo Case Studies of Predicting Reaction Mechanisms
Our first example is the acid-catalyzed hydrolysis of an enamine to give
a
ketone and a pro-
tonated amine.
a
0
an
H3O-/H20
?
a)
-iH
o
The second suggestion given in Section A5.5 is to note where you are going. In this case, the
amine has to leave the molecule and an oxygen has to add. With regard to the departure of
the amine, amide anions (RrN-) or neutral amines (NRr) canbe considered as possible leaving groups. However, since an amide anion is highly basic, it is a very poor leaving group.
Moreover, we want to avoid the creation of a strong base in the presence of acid. Both of these
points make it clear that the amine must depart as a neutral species. The nucleophile must be
water and not hydroxide since the reaction is run in acidic conditions. The next point to note
is that the double bond in the enamine is within the ring and between carbons, but the double bond in the product is exo to the ring and is to oxygen. Hence, the double bond must
change position at some point in the mechanism.
Once it is clear what bonds must change positions and what groups must add and leave
the reactant, the next step is to consider which reaction steps canbe used to accomplish these
tasks. In this case, because the reactions are being performed under acidic conditions, an
acid can be used to move the position of the double bond in a manner similar to an acid-catalyzed tautomerization reaction. The acid can also be used to protonate the leaving group so
that the amine can depart as a neutral species. Furthermore, the acid could be used to activate a polarized tl bond toward nucleophilic attack, if necessary.
Once the likely reactions have been identified, it is a matter of putting them in the correct
order. The reactant currently does not have a polarized r bond for nucleophilic attack by
water, and neither is the leaving group ready for departure as a neutral amine. Therefore, we
should start by either creating a polarized n bond, or making the leaving group ready for departure. Let's consider the latter first. Protonation of the amine can lead to its departure as
shown below. However, this creates a vinyl cation, an unreasonably high energy intermediate. Hence, this is not plausible.
H
[)H
'l'l:
.\\,
H.d@
r'\H
-*
a;)
illH
:o
H
O
o 6-
1069
r070
APPENDIX 5: PUSHING ELECTRONS
howevel the cation formed is an imminium, which is
much more stable than the vinyl cation shown above. In additiory this protonation leads to
the migration of the double bond, as is required in the mechanism. Moreover, there is now a
highly polarized n bond within the molecule. Due to the fact that this reaction contains several of the aspects of the mechanism that were identified as necessary, it is a good place to
If protonation occurs on the
B carbory
start.
f)
n-
H* (dH
'N'
do
'l)'-' 'H
!d
,
tt
\-,/
-\
tt
\.,'
'd-'H
Keeping in mind where we are going, leaving group departure and nucleophilic attack are
stiltlequired. The amine is not ready to depart since it is held to the structure via a double
bond. Therefore, addition of water to the polari zed n bond, followed by loss of a proton, is
the next logical sequence of events.
H
{;)
r-N-
X
\-l
H
-
,dH
{-'o'H
( ) nqN
\OY
V-H
a)
\-,
o
*N
OV-H
a\
(,
All that remains now is leaving group departure. Protonation of the amine foliowed by a 1,2elimination assisted. by the n"lghfoinglone pairs on oxygen gives an intermediate that is
simply a proton transfer away from the correct product.
OG
-or*--5-'-
"5-
?,qE
O
The next example involves base catalysis, and is known as the Robinson annulation.
Once agair; the first item to note is where we are going.
OO
(^/
-4o :,^'. .'\l-1
11
(-\Ao
,,.
By counting carbons it is clear that no additional carbon atoms are required. To see how the
two reactants are put together, it is useful to letter or number some or all of the carbons of
the reactants and ilace ttose markers near the same carbons in the product. Note that in the
starting material carbons a and c are identical due to symmetry in the molecule.
o
*Y
I I
\-6O
"-4o +Base
+ //t\
d S
Once it is clear how the two pieces have to go together, we can make decisions as to the appropriate reactions to use to fuse the two reactants. First, note that carbonb has undergone a
conjugate addition forming a bond to carbon d. Under basic conditions a conjugate addition
would start via formation of an enolate nucleophile. Second, the bonds formed between carbons c and g require the loss of a molecule of water. The loss of water most likely arises from
the eliminaiionbf water from an alcohol, since no other scheme seems reasonable, given the
reactants and the experimental conditions. Elimination of water from an alcohol under ba-
45.7 PUSHING ELECTRONS FOR RADICAL REACT]ONS
sic conditions would require a 1,4-elimination reaction on a B-hydroxy carbonyl. Finally,
B-hydroxy carbonyl structures are formed via aldol reactions.
Once the reactions
lequired are knowry it is a matter of stringing them together in the
proPer sequence. Since this reaction is performed under basic conditions, it is logical to have
thebase startby abstractingthemost acidic proton. This wouldbe thehydrogenihatis alpha
to two carbonyls. Furthermore, since the resulting enolate carbon was found by the previous
analysis to be attached to carbon d, it makes sense to draw the conjugate addition ai the first
carbon-carbonbond forming sequence in the mechanism.
,
oo,'\
o:
Fl o^.
..' ^fro
o
,:o,"c'/1
', /</9
(?i"
-?lv
/l'_
\Ao
\Ao
\-,\o
/ ^.'o"
O ,-O"
,(r
rI$o
\
At this point the rest of the mechanism must consist of the aldol reaction followedby the loss
of water. The electron pushing indicates the formation of an enolate followed by an aldol reaction. The loss of hydroxide is simply a 1,4-elimination.
a)
At
Im\
vDo
^()oa
o^:.
o-rz
\t
v^O:t,i(_O.
^
/
H'v
o
-*lll_-,\t
"-..
+l
\-1'-lS.t
HO
o
t-"_\
(-) ..
-,\1,^.t
ll
\-\Ao
o
:OH
"ov/
In summary, the power of electron pushing is the ability to write chemically reasonable
mechanismsby combining sources and sinks using reactions that are well precedented inorganic chemistry. The electron pushing allows chemists to communicate their thoughts as
to steps involving nucleophile and electrophile combinations. ln atl the discussion to this
point, we have focused upon two-electron arrow pushing. None of these reactions involves
radical intermediates. F{owever, radicals are common intermediates in organic transformations, and therefore we also need an understanding of how to perform electron pushing for
radical reactions.
A5.7 Pushing Electrons for Radical Reactions
To denote the movement of single electrons we use arrows with a single slash head. The
arrow still starts at an electron source/ but the source can now be any bond, any single electron, or any lone pair. Unlike two-electron arrow pushing, we do not consider the source for
one-electron arrows to have any analogy to nucleophiles. The arrow still ends at an electron
sink, but now the sink is defined as any site that can accept a single electron. These are not
necessarily traditional electrophilic sites.
Two arrows starting at a bond and spreading apart are used to denote the homolytic
cleavage of a bond. Two arrows starting on separate radicals and coming together represent
the formation of a bond.
n.66H. *
z1 -.\*'.Cl
2
+-\
.cH3
cl
r071
t-
7072
APPENDIX 5: PUSHING ELECTRONS
Other common reactions are hydrogen abstraction and radical additions to double
bonds. These reactions are denoted withiingle-headed arrows as combinations of the two
kinds of electron pushing given just above. Combining the kinds of steps shown here, with
different radical reactants, will allow you to write the electron pushing for most radical
reactions.
H-Ct
,,-{+ .cr
,,^,,-,ct
One last reaction type that is commonly encountered is electron transfer to a rr or o bond'
of receiving a negative charge, and it is common for these bonds
This bond must be
"upuUt"
atoms. Since the electron transfer is to an intact bond, the newly
to have elgctronegative
an antibonding orbital for thatbond. In the case of an alkene or carinto
added electron go-es
bonyl, this can 6e denoted by the creatibn of a anion radical with the two species drawn on
separate atoms. For a o boni, however, it is difficult to designate where the electron went,
and we simply draw a radical anion next to a dashed bond.
ro
>il
--\
+
o
o
e
:O-
,,\
r. z^P
)-"'-
\.o
)-",
Given the above examples, we can write the mechanisms of common radical reactions
that involve multiple stept. et our only example of single-electron arrow pushing, let's^c-onsider the mechaniimof HBr addition to alkenlsunder iadical conditions (see Section 10.10)'
Under radical conditions, there is an initiation of the reaction to create bromine radicals,
hyoften by a peroxide. The peroxide first homolyzes, and'the resulting radical abstracts a
combination
a
is
pushing
drogen atorn from HBr. Tirese are the initiation steps. The electron
addiof h"omolysis and hydrogen abstraction steps. Propagation is a combination of radical
any
of
the
combination
is
tion to an alkene followJd by hydrogen abstraction. Termination
in
as
step
each
for
same
the
is
two radicals to create a obond. i.{ote.-that the electron pushing
the simp,le examples given above.
lnitiation steps:
n-65-n
+
/--\ ^
R-Ol +H-Br*
2 R-o.
RO-H+.Br
Propagation steps:
-K;tt' -* \.r,
,,lg;;in, * )uBr +.Br
A possible termination steP
)G\aBr + .)!r,
PRACTICE PROBLEMS FOR PUSHING ELECTRONS
Practice Problems for Pushing Electron
1.
Identify any atoms, bonds, or lone pairs in the following molectiles that could be considered
as
electron sources or sinks
for two-electron arrowsA.
B.
o"
L",
I
z\
2.
c.
D.
o'H
o
E.
@o'H
N,.
ll
-\
G.
F.
il
-\
Show mechanisms for the following transformations, along with all electron pushing. These reactions involve more than
one steP.
OH
A.
,4
+HBr
Br
B. ,. ,c|
-tr
I
o
cH3NH2
JI
JI
N"
I
H
H.oo
o
o
C. \AO,--- ,-'on .^O-\Ao.\
D.
.=A;
o^-
NaOEt
HOEt
C)
+ "oH
U
O
are more complicated reactions than those presented to this
following
.Nrra
o
E' .\-^-A"
3. The
n
o
-Ill
point. However, they all involve two-electron
steps that are-similar to those presented in this book, especially in Chapters 10 and 11. Use your best chemical intuition to
write reasonable mechanisms for these transformations. Draw all intermediates and show all electron pushing. The mechanisms that you write may not actually be the ones that have been supported by experiments. You cannot be expected to
know this. Howevel, your mechanism should be reasonable, and the electron pushing should be correct.
an
A. -( lsl- .[
B.
U
NaH''Ao^'
c. O
------4--
oo-
H.oHrO
C-
U
o
E (r.
l]
(/
-,,--
G.
-O
o
NaoEt,
(/ l-f
cr.-Ao^ -HoEt* a^.-A---"
1)Accr ,
+
D.
o
//\
NaOH/ H2O
^0- ,+
F.
;"',-p (&"
H.
ll
/.\r+/\/
\_l\\
A
(-,(
H
OH
O,rO
,,2-.,.,,o
o
I
N\/
nro9nro
H3o"/H2o Jl
\_1
tl
o
Ph
1073
1074
APPENDIX 5: PUSHING ELECTRONS
o
r. l])
o
1) R2NL|
,'\-.'\."
il
(YV
2) PhSeBr
3)
oo
+
\2
HprJ"*r*
H.o9n"o
J. -\Ao^
lt
o
Heat
+ HOEI +
PhseoH
CO2
1) CH3t
K.
-S_
2) NaH
+
3)
5
o
Ph. ,S. ,,S
L.
S
NH, s'P's'PlPh, ;'-:rAllH,
2)NaH
\2
M. O^*--:*
N.
"#tr- O
.'=.i- _U .^A.O
V
R*.
NacN/EtoH
A'
oo
o.
,,'\-.'\ AoA
lttt_
\*"V
T
'r'\Ao
\)
D
tlttOCl, tt.O@
\\-/-------.---:/ 2)n-Buli
sy
4.
H.oo/Hro
U
il
./-N
I
H
Write reasonable steps for the following radical reactions, showing all the proper electron pushing.
BooR
A. )-no*"
' iA
o
oAoM"
c. )#)-a'+HBr