Employing a robustness screen: Rapid assessment of rhodium(III)-catalysed C-H activation reactions This paper is dedicated to Professor Paul A. Wender, a great scientist, teacher and mentor

Article abstract of DOI:10.1016/j.tet.2013.05.068

Following the discovery of new synthetic methodology, an assessment of its potential utility in real synthetic problems is highly desirable to facilitate its application. Herein, we describe an assessment of two contemporary rhodium catalysed C-H activati

Full text of DOI:10.1016/j.tet.2013.05.068

Tetrahedron 69 (2013) 7817e7825
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Employing a robustness screen: rapid assessment of rhodium(III)-
catalysed CeH activation reactions
Karl D. Collins, Frank Glorius *
Westf a€ lische Wilhelms-Universit a€ t M u€ nster, Organisch-Chemisches Institut, Corrensstrasse 40, 48149 M u€ nster, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Following the discovery of new synthetic methodology, an assessment of its potential utility in real
synthetic problems is highly desirable to facilitate its application. Herein, we describe an assessment of
two contemporary rhodium catalysed CeH activation methodologies using our recently reported
Received 18 March 2013
Received in revised form 15 May 2013
Accepted 17 May 2013
‘robustness screen’. The results of this screen provide an evaluation of the tolerance of each reaction to
Available online 23 May 2013
specific chemical functionalities and structural motifs, as well as the stability of these functionalities and
motifs to the reaction conditions. The results of this screen provide valuable data that can be directly
applied in the design of synthetic routes, and thus can facilitate the application of the evaluated
reactions.
This paper is dedicated to Professor Paul A.
Wender, a great scientist, teacher and
mentor
Ó 2013 Elsevier Ltd. All rights reserved.
Keyword:
Scope
Screening
CeH activation
Rhodium
Functional groups
1. Introduction
syntheses, follow up methodological papers are published, or per-
sonal experience is developed. Moreover, even when this in-
Contrary to the rapid publication of new scientific break-
throughs, the application of new knowledge is often slow, and we
believe this to be particularly true of synthetic organic methodol-
ogy. Whilst new synthetic methodologies often report to provide
more widely applicable, increasingly efficient and more environ-
mentally benign reactions, or provide routes to previously in-
accessible chemical motifs, employing such methodologies with
any regularity to ‘real’ synthetic problems often takes several
years. In our initial disclosure of this robustness screen meth-
odology we proposed that the lack of information regarding
the utility of a given reaction outside the idealised conditions of
the seminal report contributes significantly to the delay in its
application.3
It would be of significant benefit to the synthetic chemistry
community should information relating to the application of new
methodologies to real synthetic problems be available at the time
of the initial publication, though generally this is not the case.
Currently, this information is typically developed over a number of
years as the reaction is gradually implemented within total
formation has evolved, it is often fragmented and difficult to locate.
In an attempt to both reduce the time period between the discovery
and application of new synthetic methodology, and to facilitate the
rapid location of relevant information, we recently reported a con-
ceptually simple method to assess the potential utility of new
synthetic methodology. The assessment is rapid, experimentally
simple, and when appropriate could readily be reported alongside
substrate scope in new synthetic methodologies.
It should be noted that this screen should in no way be con-
sidered a substitute for the traditional substrate scope. As a conse-
quence of the intentional decoupling of the ‘additives’ functionality
from the reactive centre (vide infra), a limitation of this method is
that it is not possible to assess the direct steric/electronic impact of
a given functionality on the reaction centre (e.g., an ortho sub-
stituent on an aromatic ring), unlike the traditional scope. This
screen is a fast and simple method to generate additional in-
formation of use to synthetic chemists applying the given meth-
odology, about the likely tolerance of the reaction conditions
assessed to the functionalities/chemical motifs investigated, and
whether these additives are stable to the reaction conditions. Fur-
thermore, like results obtained within a traditional substrate scope,
a ‘failed’ reaction does not preclude the possibility of optimisation
1,2
*
Corresponding author. E-mail address: glorius@uni-muenster.de (F. Glorius).
0
040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.05.068

7818
K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
to tolerate the specific functionality. Unlike a traditional scope, we
propose by having a standardised set of additives screened, omis-
sion of unfavourable results would be discouraged and conse-
quently more information relating to the reaction would be
available: this is highly beneficial. For additional discussion on
these points see Ref. 3.
Table 1
Chemical functionalities and structural motifs to be evaluated using the robustness
screen
The metal catalysed direct functionalisation of CeH bonds has
emerged over the past decade as a highly promising tool for organic
synthesis.4 This contemporary approach enabling the direct cou-
pling of CeH bonds without the need for pre-functionalisation
represents a significant step forward in synthetic organic chemis-
try, enabling not only more efficient transformations due to
avoidance of pre-functionalisation, but also opening up new dis-
connection pathways. Consequently, the application of methodol-
ogy employing a CeH activation strategy to real synthetic problems
is highly desirable. To facilitate the application of such methodol-
,5
ogies, we herein provide an assessment of two reactions previously
5
reported within our group:
a
rhodium catalysed tetrahy-
6
droisoquinolinone formation, and the direct ortho-bromination of
arenes. Using the methods previously described in our robustness
7
screen we simultaneously assess both the tolerance of these re-
actions to chemical functionality, and the tolerance of the specific
3
chemical motifs to the reaction conditions.
2
. Results and discussion
Our goal is to demonstrate the application of the semi-
quantitative robustness screen as previously reported, as well as
providing a rapid assessment of contemporary CeH activation
synthetic methodologies. The robustness screen itself is concep-
tually and experimentally very simple: a standard reaction is un-
dertaken in the presence of 1 molar equivalent of a given functional
group or structural motif, termed an ‘additive’, and the reaction
then analysed by an appropriate method (LC-MS, HPLC, GC). We
have employed gas chromatography (GC) analysis to enable de-
termination of the yield of the reaction product, and both the
starting material and the additive remaining after the pre-
determined reaction time. This allows determination of the toler-
ance of the reaction to the given functionality, whether the additive
either inhibits product formation or retards the rate of reaction, and
of the stability of the additive to the reaction conditions.
a
Retention times are for GC analysis using method A
min, increment
O
[RhCp*Cl2]2 2.5 mol%
CsOPiv 30 mol%
O
O
N
H
Ph
NH
Ph
PivOH 20 mol%
EtOH, 80 °C, 16h
O
86%
tR = 7.12
1
tR = 4.99
2
tR = 9.74
3
For an initial assessment of a given reaction we have proposed
2
0 common functional groups/chemical motifs. Additives were
Scheme 1. Tetrahydroisoquinolinone formation to be evaluated using the robustness
screen. t is retention time for GC method A.
selected to cover a range of typical properties, e.g., basic, ligating,
unsaturated etc., though it is important to note that the additives
were not evaluated with regard to possible influence on reactivity,
or their stability to the reaction conditions. In addition, commercial
availability and sufficient difference in retention times to allow for
batch analysis (vide infra) using GC analysis were additional pre-
requisites. We feel that a more critical selection process for the
additives should be avoided as it could potentially lead to a bias in
the screen, with additives likely to work being chosen in preference
to those predicted to fail. The 20 functional groups/chemical motifs
are split into two ‘groups’, with Group A containing common
functional groups, and Group B, common heterocyclic motifs
R
to establish any signal overlap with the additives or the analytical
standard (mesitylene, retention time 5.55 min). Should signal
overlap be detected, a change in the analytical method, or exclu-
sion/substitution of the given additive from the screen can be un-
dertaken if deemed necessary.
Using GC analysis,
a
batch calibration technique (see
Experimental section for details) inspired by Hartwig’s analysis of
9
complex mixtures was employed for the simultaneous calibration of
all Group A additives, the starting material and the product against
the analytical standard. This was then repeated for Group B additives.
This technique allows for a rapid semi-quantitative analysis of all
components of the screen. Only two GC experiments are performed,
one for each group of additives. The relative ratios of the additives,
starting material and product to the standard were subsequently
determined and used as ‘single point’ calibrations for further analysis
(see Experimental section for GC traces). Although absolute values
cannot be determined, this semi-quantitative analysis provides
a valuable measure of how a reaction proceeds in the presence of
a given additive, and the stability of the additive to the reaction
conditions.
(Table 1).
2
.1. Application of the robustness screen to a tetrahy-
droisoquinolinone formation reaction
The robustness screen was first applied to our recently reported
preparation of tetrahydroisoquinolinone 3 from hydroxamic ester 1
and styrene 2 (Scheme 1). This reaction has also been reported by
Fagnou using similar reaction conditions.
Prior to experimentation, GC (or selected alternative) analysis of
the starting materials and the product of the reaction are required
6
,8

K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
7819
With calibration complete, experimentation was undertaken. To
minimise the time required for undertaking the screen, reactions
were run in parallel. The standard reaction was prepared on scale,
and subsequently distributed to reaction vessels containing a given
additive. The experiments were performed in two runs, one with
group A and one with group B. The results are given below (Table 2).
Evaluation of the results for Group A clearly demonstrates that
the reaction is highly tolerant of all functional groups assessed with
the exception of the terminal alkyne (Table 2, A4). The reaction
typically proceeds in excellent yield, and all functional groups show
very high tolerance to the reaction conditions. This pleasingly
suggests that the tetrahydroisoquinolinone formation would be
suitable for application to more complex molecules containing any
of the above functional groups. The validity of this extrapolation
was investigated within our initial report by direct comparison of
the results generated from the screen when applied to the
BuchwaldeHartwig reaction, and results obtained following the
though it is apparent that reaction with styrene is significantly
more efficient.
Group B (Table 2) also shows high tolerance of the reaction to the
presence of a range of heterocycles. Very good yields for the reaction
are obtained in the presence of indole, protected pyrroles and
thiophene and reactions in the presence of furan, benzofuran,
benzoxazoleand quinoline derivatives proceed insynthetically useful
yields. In contrast, reactions in the presence of N-methylimidazole
(Table 2, B2) and 3,5-lutidine (Table 2, B3) proceeded in low yield,
suggesting complex substrates containing such motifs would be un-
suitable for this transformation. The yield of the additive after the
reaction was typically excellent, further highlighting the mild nature
of the reaction conditions. For Group B, the formation of carbamate 5
(Scheme 2) (determined by GCeMS analysis) was often observed. It is
likely 5 is formed via Lossen rearrangement of the hydroxamic ester 1
to give intermediate isocyanate 4, followed by addition of the solvent
10
(EtOH). As would be expected, 5 is observed in quantities inversely
proportional to the product formation, and is more notable in re-
actions containing basic additives, i.e., imidazole and pyridine de-
rivatives. Trace amountsof 5 were alsoobservedwhenanilinewasthe
additive (Table 2, A2).
3
preparation and reaction of multi-functional substrates. Such
a high tolerance to a wide range of ‘types’ of functionality, e.g.,
protic, ligating, etc. would suggest that this reaction is generally
very tolerant, and as such confidence in the application of this
methodology in the presence of functional groups not specifically
evaluated would also be significant. In the case of the terminal al-
kyne, we observe significant competitive reaction of the alkyne
with 1 to give the corresponding isoquinolinone (determined by
8
GCeMS analysis). This reactivity has been reported by Fagnou. In
the case of the terminal alkene (Table 2, A7) we also observed what
is thought to be trace amounts of the corresponding tetrahy-
droisoquinolinone formed from reaction of 1 and the additive,
Scheme 2. Lossen rearrangement of hydroxamic ester 1.
Table 2
Results of the robustness screen for the formation of tetrahydroisoquinolinone 3 in the presence of a given additive
Group A - Functional Groups
Group B - Heterocycles
Yield of
Additive
SM
Yield of
Additive
SM
Entry
A1
3 %
remaining % remaining %
Entry
B1
3 %
remaining % remaining %
Cl
O
45
ndb
0
92
>95
>95
0
-
nBu
NH
CN
2
A2
76
92
23
B2
N
N
x
x
-
10
16
54
>95
0
0
0
N
A3
A4
>95
8
0
0
B3
B4
-
59
x
31
x
>95
6
O
O
N
O
A5
OMe
>95
>95
87
0
B5
-
37
>95
0
S
A6
A7
OH
82
91
0
0
B6
B7
70
68
>95
81
0
0
7
9
nBu
Piv
nda
N
N
A8
Ph
NH
Cl
-
65
>95
0
B8
69
>95
0
N
H
O
Bn
A9
2
91
>95
>95
0
0
B9
N
63
52
>95
>95
0
0
1
0
-
-
A10
>95
B10
10
N
Cl
Table shows the yield of 3 when the standard reaction (Scheme 1) is undertaken in the presence of one molar equivalent of the given additive. Colour coding
facilitates the ready assessment of the data: green (above 66%), yellow (34-66%), red (below 34%). The additive and starting material remaining after
a
reaction is also given. All yields were determined by GC analysis. The additive remaining could not be determined due to signal overlap with small
b
amounts of double bond isomers of the additive. The additive remaining could not be determined due to signal overlap of the additive with styrene.
SM is starting material. nd is not determined.

7820
K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
This assessment of the tetrahydroisoquinolinone formation was
selectivity for halogenation of the substrate over the additive sug-
gests that a benzamide is a better directing group than an acetan-
ilide in this instance. It should be noted that the relative directing
group ability is reaction dependant, and is typically an unexplored
undertaken in only 3 days (calibration, experimentation and anal-
ysis) and has shown that the reaction is tolerant of a wide range of
chemical functionality and structural motifs. Futhermore, condi-
tions are shown to be sufficiently mild to avoid the degradation of
the majority of the additives screened. Although this data does not
guarantee success in any given reaction, it does provide significant
confidence that the application of this chemistry to more complex
synthetic problems, and ergo more complex substrates, is likely to
be successful. This screen provides a significant amount of data that
is directly relevant to the application of this methodology to real
synthetic problems that was not apparent in the original publica-
tion. We consequently hope that this evaluation will now lead to
more rapid application of this methodology.
4
,5
area of CeH activation chemistry as a whole. Reaction in the
presence of nonanol (Table 3, A6) is moderate and significant
starting material remains, though degradation of the additive is
apparent. In the case of aniline, the terminal alkyne, and the ter-
minal alkene, bromination of the additive is apparent in all in-
stances (Table 3, A2, A4, A7). In the case of the dodecylamine (Table
3, A9), degradation of the additive is observed. Group B demon-
strates that heterocycles are typically detrimental to the reaction.
Bromination of the heterocycle is often observed, though in some
instances (Table 3, B3, B5, B10) the reaction is simply inhibited by
the presence of the additive. Pleasingly though, the reaction shows
very high tolerance to the presence of benzofuran (Table 3, B4), and
the additive itself is stable to the reaction conditions.
2
.2. Application of the robustness screen to the ortho-bro-
mination of N,N-diisopropylbenzamide
2
The results for reaction in the presence of Cu(OAc) in place of
PivOH are broadly comparable, though some notable exceptions are
observed (Table 4). In the case of benzonitrile and dodecylchloride
Following the application of the screen to the formation of tet-
rahydroisoquinolinone 3, we sought to apply the screen to what we
believed to be a more challenging transformation with regard to its
broader applicability. We selected the ortho-bromination of arenes
containing a directing group, using rhodium catalysis and N-bro-
(Table 4, A3, A10) retardation of the rate of reaction is not observed,
and the reaction in the presence of phenylchloride (Table 4, A1)
proceeds in significantly increased yield. Of particular note is
the increased tolerance of N-methyl-N-phenylacetamide and
7
mosuccinimide as the bromine source; we believed that this re-
2-butylthiophene to the reaction conditions when compared to
action was likely to prove challenging for application to more
complex substrates than those reported in the initial publication
due to the highly electrophilic nature of NBS, and thus propose that
evaluation using the robustness screen would be of significant
value. In the initial publication a range of directing groups were
employed, and both bromination and iodination with the corre-
sponding N-halosuccinimides were investigated. Typically PivOH
was employed to enhance the reaction, though examples using
reaction in the presence of PivOH (Table 4, A8, B6). In both instances
the reaction proceeds in excellent yield and the additives prove
significantly more stable to the reaction conditions.
Although overall the bromination of N,N-diisopropylbenzamide
7
is not particularly tolerant of additional functionality, the screen
has clearly demonstrated which functional groups and heterocycles
are tolerant of the reaction conditions and do not inhibit reactivity.
We have also identified that using Cu(OAc) in place of PivOH has
2
Cu(OAc)
screen we evaluated the bromination of N,N-diisopropylbenzamide
to give brominated arene 7 (Scheme 3). We were also interested
in the impact of PivOH relative to Cu(OAc) , and as such performed
an evaluation of both sets of reaction conditions.
2
as an alternative were also reported. For the robustness
a significant impact on the efficiency of the reaction, and can in-
crease the tolerance of certain additives to the reaction conditions.
Following the initial analysis of the bromination reaction, we
sought to further demonstrate the modular nature of the robust-
ness screen. To this goal we have determined two additional groups
of additives suitable for batch analysis, with the retention times
reported below (Table 5). The identification of compounds and
determination of the grouping to enable batch analysis requires an
initial investment of time, but once established, these groups can be
subsequently used for the evaluation of any given reaction. We
envisage the development of further groups directed towards
specific aims, e.g., evaluation of protecting groups to specific re-
action conditions, or the evaluation of specific structural motifs
common to a medicinal chemistry project. This would allow an
evaluation of a given set of reaction conditions towards relevant
chemical functionality or structure without the need to prepare
numerous complex substrates.
6
2
11,12
O
RhCp*Cl2(MeCN)3[SbF6]2 2 mol%
NBS 1.1 eq
Br
O
N
N
PivOH 1.1 eq or Cu(OAc)2 0.55 eq
DCE, 60 °C, 16h
1
00%a
tR = 7.79
6
tR = 8.46
7
Scheme 3. Bromination of N,N-diisopropylbenzamide 6 to be evaluated using the
a
robustness screen. t
R
is retention time for GC method A. GC yield.
Pre-experimental preparation and experimental procedure
are identical to that described for the evaluation of the tetrahy-
droisoquinolinone formation. The results for group A and group B
using reaction conditions employing PivOH are given in Table 3.
Analysis of Group A shows that phenyl- and dodecylchloride,
benzonitrile and methyl benzoate are tolerant to the reaction
conditions (Table 3, A1, A3, A5, A10). The bromination proceeds in
high yield for reactions in the presence of the aryl halide and aryl
ester, though in the case of benzonitrile and the alkyl halide, the
yields are moderate. Importantly though, significant starting ma-
terial still remains suggesting that the additive only retards the rate
of reaction rather than inhibits product formation. Therefore,
extending the reaction time in these instances is likely to result in
increased yields. Reaction in the presence of the tertiary amide
For groups C and D, we again evaluated reaction conditions
employing both PivOH and Cu(OAc)
2
for the bromination of
N,N-diisopropylbenzamide 7 (Tables 6 and 7).
Pleasingly we have identified that reaction proceeds efficiently
and the additive is tolerant of the reaction conditions in the case of
phenyl bromide and iodide, benzaldehyde, an aliphatic and an aro-
maticketone and notablychromone (Tables6 and 7, D2, D4, C2, C5, C8,
D7). Reaction in the presence of decanenitrile also proceeds, though
the rate of reaction is retarded (Tables 6 and 7, C6). Unsurprisingly, the
reaction was not successful in the presence of any of the additional
aromatic heterocycles investigated. Furthermore, electron-rich aro-
matic systems as well as aliphatic unsaturated systems also proved
intolerant tothe reaction conditions, with bromination of the additive
observed in most instances (determined by GCeMS). Overall, it is
(Table 3, A8) proceeds in very good yield, though bromination of the
additive is observed (determined by GCeMS analysis), presumably
via an analogous CeH activation mechanistic pathway. The
2
again shown that reaction in the presence of Cu(OAc) rather than
PivOH is generally more efficient.

Table 3
Results of the robustness screen for the bromination of N,N-diisopropylbenzamide 6 using PivOH to promote reaction, in the presence of a given additive
Group A - Functional Groups
Group B - Heterocycles
Yield of
Additive
SM
Yield of
Additive
SM
Entry
A1
7 %
remaining % remaining %
Entry
B1
7 %
remaining % remaining %
Cl
O
81
71
<5
88
x
x
x
<5
<5
x
x
<5
<5
90
nBu
NH
CN
2
N
N
A2
x
<5
x
12
>95
55
B2
B3
B4
>95
>95
14
N
A3
A4
-
61
<5
34
73
<5
82
82
90
x
-
6
O
O
O
N
A5
OMe
91
>95
<5
B5
x
15
94
85
S
A6
A7
OH
-
49
19
-
59
28
46
80
B6
B7
nBu
-
40
<5
-
-
35
34
58
90
7
9
Piv
x
x
N
x
N
A8
Ph
NH
Cl
70
-
58
15
B8
x
<5
x
<5
>95
N
H
O
Bn
<5
50
x
0
86
50
B9
N
x
x
<5
<5
-
38
>95
>95
A9
2
x
1
0
A10
-
>95
B10
>95
10
N
Cl
Table shows the yield of 7 when the standard reaction (Scheme 3) is undertaken in the presence of one molar equivalent of the given additive.
See Table 2 for additional explanation of the data.
Table 4
Results of the robustness screen for the bromination of N,N-diisopropylbenzamide 6 using Cu(OAc)
2
to promote reaction, in the presence of a given additive
Group A - Functional Groups
Group B - Heterocycles
Yield of
Additive
SM
Yield of
Additive
SM
Entry
A1
7 %
remaining % remaining %
Entry
B1
7 %
remaining % remaining %
Cl
O
>95
72
<5
0
x
<5
x
<5
34
88
nBu
NH
CN
2
N
N
A2
x
<5
x
>95
B2
x
x
<5
-
>95
N
A3
A4
>95
<5
>95
11
<5
B3
B4
<5
-
56
93
94
<5
x
x
>95
>95
6
O
O
N
O
A5
OMe
>95
>95
<5
B5
x
7
>95
93
S
A6
A7
OH
-
-
39
37
x
22
47
64
87
B6
B7
nBu
>95
<5
90
30
<5
7
9
Piv
N
x
x
>95
-
-
N
A8
Ph
NH
Cl
>95
81
<5
B8
<5
x
<5
>95
N
H
O
Bn
x
<5
x
<5
>95
<5
B9
N
x
x
<5
<5
-
38
>95
94
A9
2
1
0
A10
>95
>95
B10
>95
10
N
Cl
Table shows the yield of 7 when the standard reaction (Scheme 3) is undertaken in the presence of one molar equivalent of the given additive.
See Table 2 for additional explanation of the data.

7822
K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
Table 5
that this method will facilitate the application of new and existing
chemical reactions within both an academic and industrial
environment.
Groups C and D to be evaluated using the robustness screen
4
4
. Experimental
.1. General experimental
ꢀ
All reactions were carried out in oven dried (80 C) reaction
vessels with Teflon screw caps under argon, using anhydrous sol-
vents. Reaction solvents were stored over molecular sieves from
purchase. GC analysis was undertaken on an HP-5 quartz column
using flame ionisation detection. Method A: initial temperature
ꢀ
ꢀ
ꢀ
5
0 C, hold 3 min, increment 40 C/min, final temperature 280 C,
hold 3 min. Cu(OAc) , CsOPiv and [RhCp*Cl were stored and
weighed in a glove box. All other reagents were used as purchased.
2
2 2
]
4.2. Procedure for gas chromatography batch calibration
Sample procedure for calibration of Group A for the evaluation
of the tetrahydroisoquinolinone formation: using a volumetric flask
.125 mmol of each compound in Group A was dissolved in toluene
20 mL). A separate calibration solution containing 0.125 mmol of
0
(
the standard (mesitylene), starting material phenyl hydroxamic
ester 1 and product tetrahydroisoquinolinone 3 in toluene (20 mL)
was also prepared. A 1:1 mixture of the calibration solutions
(1.0 mL total volume) was prepared and analysed by GC analysis.
The relative ratios of all additives, 1 and 3 to the standard were then
determined and used as a ‘single point’ calibration for subsequent
analysis. Sample chromatograms are given below (Fig. 1).
a
Retention times are for GC analysis. Method (A): Initial temperature
4.2.1. Additional information and tips.
ꢁ
Calibration solutions can be prepared and stored for future
analyses.
3
. Conclusion
ꢁ
Volatile additives should be dissolved in solvent directly.
We have demonstrated the application of the robustness screen
ꢁ Equimolar amounts of additive, starting material, product and
standard are employed to facilitate rapid analysis, though can
be varied as required.
to two contemporary Rh(III)Cp* catalysed CeH activation meth-
odologies. The screen has demonstrated that the tetrahy-
droisoquinolinone 3 formation from hydroxamic ester 1 and
styrene 2 is highly tolerant to a broad range of chemical function-
alities and structural motifs, including many common heterocycles.
The application of the robustness screen to the ortho-bromination
of N,N-diisopropylbenzamide 7 identified several specific func-
tionalities and structural motifs, which are tolerated by the reaction
conditions, but also highlighted that this methodology may be
limited in its application to more complex substrates. We have also
ꢁ 0.125 mmol of each additive is equivalent to between 10 and
25 mg of each additive. We suggest that preparation of stock
solutions containing smaller amounts should be avoided as the
error in GC calibration will be increased. For example, for an
additive amount of 10 mg, an error of 0.5 mg in weight is
equivalent to 5% yield.
ꢁ For the preparation of additional calibration ‘groups’, we have
established that a difference in retention time of 0.03 min (using
Method A) is typically sufficient for separation of compounds,
though specific compound/compound interactions have been
shown to affect retention times/separation.
established that bromination reactions in the presence of Cu(OAc)
2
rather than PivOH are generally more efficient, and importantly
more selective in specific instances. We intend for the data gener-
ated by these screens to be directly applied in the design of syn-
thetic routes, and facilitate the application of the given reactions.
This data provides significantly more information pertaining to the
potential application of these reactions than was determined in the
substrate scope of the respective publications, and we believe this
highlights the benefit of utilising such an approach to evaluate any
given reaction. We suggest this data is highly complimentary to the
substrate scope, and thus could be reported concurrently in new
methodological papers when appropriate.
We believe that the retrospective application of this method-
ology will be of particular use in an industrial environment where
the preparation of structurally similar analogues is commonplace.
This method can be tailored to assess a given chemical reaction
with respect to specific functionality or motifs within a project,
determining the suitability of the reaction prior to the preparation
of complex molecules that may prove either intrinsically unstable
to the reaction conditions, or inhibit reactivity. We strongly believe
ꢁ We propose that no more than 10 compounds should typically
be evaluated in a given group. This minimises the likelihood
of signal overlap with starting materials, products and
by-products of any given reaction.
4.3. Experimental procedure for robustness screens
4.3.1. Procedure for the screening of the formation of tetrahy-
droisoquinolinone 3. To a solution of N-(pivaloyloxy)benzamide 1
(1.375 mmol, 304 mg, 1 equiv), CsOPiv (0.413 mmol, 97 mg,
0.3 equiv), PivOH (0.275 mmol, 28 mg, 0.2 equiv) and styrene 2
(2.07 mmol, 237
[RhCp*Cl (0.034 mmol, 21 mg, 2.5 mol %). The reaction mixture
was distributed into 11 reaction vessels (0.8 mL, 0.125 mmol of N-
(pivaloyloxy)benzamide 1 per reaction), 10 containing one of each
additive (0.125 mmol) and one control reaction containing no ad-
ml, 1.5 equiv) in EtOH (8.8 mL), was added
2 2
]
ꢀ
ditive. The reactions were then heated at 80 C for 16 h before

K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
7823
Table 6
Results of the robustness screen for the bromination of N,N-diisopropylbenzamide 6 using PivOH to promote reaction, in the presence of a given additive
Group C
Group D
Yield of
Additive
SM
Yield of
Additive
SM
Entry
C1
7 %
remaining %
remaining %
Entry
D1
7 %
remaining % remaining %
S
x
<5
84
-
45
90
80
15
x
<5
92
nda
91
<5
N
N
O
Br
C2
H
D2
>95
Cl
OH
C3
C4
C5
x
x
9
x
21
91
79
83
>95
20
D3
D4
D5
N
x
<5
94
<5
>95
>95
<5
>95
<5
S
N
I
<5
80
O
x
x
x
x
94
N
4
4
O
C6
-
51
95
40
D6
<5
x
16
>95
7
N
N
O
H
N
C7
C8
Ph
x
10
x
9
81
<5
D7
D8
>95
23
>95
9
<5
72
O
O
O
>95
>95
x
x
Table shows the yield of 7 when the standard reaction (Scheme 3) is undertaken in the presence of one molar equivalent of the given additive.
a
See Table 2 for additional explanation of the data. Yield of additive not determined due to signal overlap with an unidentified by-product.
Table 7
2
Results of the robustness screen for the bromination of N,N-diisopropylbenzamide 6 using Cu(OAc) to promote reaction, in the presence of a given additive
Table shows the yield of 7 when the standard reaction (Scheme 3) is undertaken in the presence of one molar equivalent of the given additive.
See Table 2 for additional explanation of the data.

7824
K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
Fig. 1. Chromatograms used for the single point calibration of the gas chromatograph.
cooling, the addition of mesitylene (0.125 mmol, 17.5
sequent GC analysis. For GC sample preparation, the reaction mix-
ture was filtered through a short plug of silica or Celite (vide infra).
m
l) and sub-
a solution of N,N-diisopropylbenzamide 6 (1.375 mmol, 282 mg,
equiv), NBS (1.51 mmol, 269 mg, 1.1 equiv), Cu(OAc)
(0.756 mmol, 110 mg, 0.55 equiv) in DCE (11.0 mL) was added
RhCp*(MeCN) [SbF (0.028 mmol, 23 mg, 2.0 mol %). The reaction
1
2
3
6 2
]
4
.3.2. Procedure for screening of the ortho-bromination of N,N-dii-
mixture was distributed into 11 reaction vessels (1.0 mL,
0.125 mmol of N,N-diisopropylbenzamide 6 per reaction), 10 con-
taining one of each additive (0.125 mmol) and one control reaction
sopropylbenzamide 6 using PivOH to promote reaction. To a solution
of N,N-diisopropylbenzamide 6 (1.375 mmol, 282 mg, 1 equiv), NBS
ꢀ
(
1.51 mmol, 269 mg,1.1 equiv), PivOH (1.51 mmol,155 mg,1.1 equiv)
in DCE (8.8 mL) was added RhCp*(MeCN) [SbF (0.028 mmol,
3 mg, 2.0 mol %). The reaction mixture was distributed into 11
reaction vessels (0.8 mL, 0.125 mmol of N,N-diisopropylbenzamide
per reaction), 10 containing one of each additive (0.125 mmol)
containing no additive. The reactions were then heated at 60 C for
y
3
6
]
2
16 h before cooling, the addition of mesitylene (0.125 mmol,
2
17.5 ml) and subsequent GC analysis. For GC sample preparation, the
reaction mixture was filtered through a short plug of silica or Celite
(vide infra).
6
and one control reaction containing no additive. The reactions were
ꢀ
then heated at 60 C for 16 h before cooling, the addition of
4.3.4. Additional information and tips.
y
mesitylene (0.125 mmol, 17.5
GC sample preparation, the reaction mixture was filtered through
a short plug of silica or Celite (vide infra).
ml) and subsequent GC analysis. For
ꢁ In preparation of samples for GC analysis, reactions containing
dodecylamine, N-methylimidazole, 4-methylthiazole and pyr-
idine-N-oxide should be filtered through Celite and not silica to
avoid removal of the additive from the analytical sample.
ꢁ
For GC sample preparation 30 ml of each reaction was filtered
through a plug of silica w2 cm in depth in a Pasteur pipette,
washing through with 2 mL of EtOAc.
4
.3.3. Procedure for screening of the ortho-bromination of N,N-dii-
sopropylbenzamide
2
6 using Cu(OAc) to promote reaction. To
ꢁ
Sonication can be used to facilitate dissolution of all reagents
prior to distribution of the reaction mixture to individual re-
action vessels. If complete dissolution of all reagents is not
observed after sonication, the reaction mixture should be
stirred vigorously when aliquots are taken to ensure uniform
distribution of all reagents.
y
For reactions containing benzo[d]oxazole, benzo[d]thiazole, decanenitrile,
,5-lutidine, 2-chloroquinoline and 2-methyl pyridine-N-oxide as additives, the
3
4
reactions were quenched with NaBH (3 equiv in EtOH) prior to the addition of
mesitylene.

K.D. Collins, F. Glorius / Tetrahedron 69 (2013) 7817e7825
7825
ꢁ
ꢁ
A control reaction ran in parallel to each group of additives to
be analysed is critical to confirm the absence of experimental
error.
For the bromination of N,N-diisopropylbenzamide 6 using
2
Cu(OAc) the concentration of the reaction was adjusted rela-
tive to the initial report to facilitate dissolution of the reagents.
A control reaction was run prior to the screen to ensure the
reaction was complete in the given time period.
The scale of reaction is operationally viable in terms of exper-
imental procedure, cost, efficiency and analysis, though can
readily be altered as appropriate.
45, 788; (o) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K. Angew. Chem., Int. Ed.
012, 51, 8960; (p) Ackermann, L. Acc. Chem. Res. 2013, http://dx.doi.org/10.
021/ar3002798; (q) Wencel-Delord, J; Glorius, F. Nat. Chem. 2013, 5, 369.
. For reviews on Rh(III)-catalysed CeH activation, see: (a) Satoh, T.; Miura, M.
Chem.dEur. J. 2010, 16, 11212; (b) Song, G.; Wang, F.; Li, X. Chem. Soc. Rev. 2012,
2
1
5
41, 3651; (c) Patureau, F. W.; Wencel-Delord, J.; Glorius, F. Aldrichimica Acta
2012, 45, 31 For selected recent progress of Rh(III)-catalysed CeH activations,
see; (d) Ueura, K.; Satoh, T.; Miura, M. Org. Lett. 2007, 9, 1407; (e) Umeda, N.;
Hirano, K.; Satoh, T.; Miura, M. J. Org. Chem. 2009, 74, 7094; (f) Stuart, D. R.;
Alsabeh, P.; Kuhn, M.; Fagnou, K. J. Am. Chem. Soc. 2010, 132, 18326; (g) Rakshit,
S.; Patureau, F. W.; Glorius, F. J. Am. Chem. Soc. 2010, 132, 9585; (h) Patureau, F.
W.; Glorius, F. J. Am. Chem. Soc. 2010, 132, 9982; (i) Mochida, S.; Hirano, K.;
Satoh, T.; Miura, M. J. Org. Chem. 2011, 76, 3024; (j) Patureau, F. W.; Besset, T.;
Glorius, F. Angew. Chem., Int. Ed. 2011, 50, 1064; (k) Li, H.; Li, Y.; Zhang, X.-S.;
Chen, K.; Wang, X.; Shi, Z.-J. J. Am. Chem. Soc. 2011, 133, 15244; (l) Morimoto, K.;
Itoh, M.; Hirano, K.; Satoh, T.; Shibata, Y.; Tanaka, K.; Miura, M. Angew. Chem.,
Int. Ed. 2012, 51, 5359; (m) Kim, J. Y.; Park, S. H.; Ryu, J.; Cho, S. H.; Kim, S. H.;
Chang, S. J. Am. Chem. Soc. 2012, 134, 9110; (n) Lian, Y.; Bergman, R. G.; Ellman, J.
A. Chem. Sci. 2012, 3, 3088; (o) Hyster, T. K.; Kn o€ rr, L.; Ward, T. R.; Rovis, T.
ꢁ
Acknowledgements
Generous financial support by the European Research Council
under the European Community’s Seventh Framework Program
FP7 2007-2013)/ERC Grant agreement no 25936 and the Deutsche
Forschungsgemeinschaft (DFG, Leibniz award to F.G.) is gratefully
acknowledged. Thank you to Christian Richter and Daniel Tang for
very helpful discussions.
Science 2012, 338, 500; (p) Ye, B.; Cramer, N. Science 2012, 338, 504; (q) Chen,
K.; Li, H.; Lei, Z.-Q.; Li, Y.; Ye, W.-H.; Zhang, L.-S.; Sun, J.; Shi, Z.-J. Angew. Chem.,
Int. Ed. 2012, 51, 9851; (r) Li, B.-J.; Wang, H.-Y.; Zhu, Q.-L.; Shi, Z.-J. Angew. Chem.,
Int. Ed. 2012, 51, 3948; (s) Xu, X.; Liu, Y.; Park, C.-M. Angew. Chem., Int. Ed. 2012,
(
51, 9372; (t) Jayakumar, J.; Parthasarathy, K.; Cheng, C.-H. Angew. Chem., Int. Ed.
2012, 51, 197; (u) Chan, W.-W.; Lo, S.-F.; Zhou, Z.; Yu, W.-Y. J. Am. Chem. Soc.
2012, 134, 13565; (v) Wang, C.; Chen, H.; Wang, Z.; Chen, J.; Huang, Y. Angew.
Chem., Int. Ed. 2012, 51, 7242; (w) Zhen, W.; Wang, F.; Zhao, M.; Du, Z.; Li, X.
Angew. Chem., Int. Ed. 2012, 51, 11819; (x) Wang, H.; Glorius, F. Angew. Chem., Int.
Ed. 2012, 51, 7318; (y) Wang, H.; Grohmann, C.; Nimphius, C.; Glorius, F. J. Am.
Chem. Soc. 2012, 134, 19592; (z) Kuhl, N.; Hopkinson, M. N.; Glorius, F. Angew.
Chem., Int. Ed. 2012, 51, 8230; (aa) Itoh, M.; Hirano, K.; Satoh, T.; Shibata, Y.;
Tanaka, K.; Miura, M. J. Org. Chem. 2013, 78, 1365; (ab) Lian, Y.; Huber, T.; Hesp,
K. D.; Bergman, R. G.; Ellman, J. A. Angew. Chem., Int. Ed. 2013, 52, 629; (ac)
Dong, J.; Long, Z.; Song, F.; Wu, N.; Guo, Q.; Lan, J.; You, J. Angew. Chem., Int. Ed.
2013, 52, 580; (ad) Neely, J. M.; Rovis, T. J. Am. Chem. Soc. 2013, 135, 66; (ae)
Wang, H.; Schr o€ der, N.; Glorius, F. Angew. Chem., Int. Ed. 2013, 52, 5386; (af) Shi,
References and notes
1
. For discussions on the direction and goals of the field of organic synthesis see:
a) Seebach, D. Angew. Chem., Int. Ed. Engl. 1990, 29, 1320; (b) Newhouse, T.;
(
Baran, P. S.; Hoffmann, R. W. Chem. Soc. Rev. 2009, 38, 3010; (c) Trost, B. M.
Science 1991, 254, 1471; (d) Trost, B. M. Science 1983, 219, 245; (e) Trost, B. M.
Angew. Chem., Int. Ed. Engl. 1995, 34, 259; (f) Wender, P. A.; Miller, B. L. Nature
2
009, 460, 197; (g) Horvath, I.; Anastas, P. T. Chem. Rev. 2007, 107, 2169.
2
. For diverse applications of synthetic methodology see: (a) Nicolaou, K. C.;
Z.; Grohmann, C.; Glorius, F. Angew. Chem., Int. Ed. 2013, 52, 5393.
6. Rakshit, S.; Grohmann, C.; Besset, T.; Glorius, F. J. Am. Chem. Soc. 2011, 133, 2350.
7. Schr o€ der, N.; Wencel-Delord, J.; Glorius, F. J. Am. Chem. Soc. 2012, 134, 8298.
Vourloumis, D.; Winssinger, N.; Baran, P. S. Angew. Chem., Int. Ed. 2000, 39, 44;
b) M u€ ller, T. J. J.; Bunz, U. H. F. Functional Organic Materials: Syntheses, Strategies
(
and Applications; Wiley: New York, NY, 2007; (c) Li, J. J.; Johnson, D. S. Modern
Drug Synthesis; Wiley: New York, NY, 2010; (d) Lamberth, C.; Dinges, J. Bioactive
Heterocyclic Compound Classes: Pharmaceuticals and Agrochemicals; Wiley-VCH:
New York, NY, 2012.
. Collins, K. D.; Glorius, F. Nat. Chem. 2013, http://dx.doi.org/10.1038/NCHEM.1669
. For recent reviews on CeH activation: (a) Daugulis, O.; Do, H.-Q.; Shabashov, D.
Acc. Chem. Res. 2009, 42, 1074; (b) Giri, R.; Shi, B.-F.; Engle, K. M.; Maugel, N.; Yu,
J.-Q. Chem. Soc. Rev. 2009, 38, 3242; (c) Colby, D. A.; Bergman, R. G.; Ellman, J. A.
Chem. Rev. 2010, 110, 624; (d) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110,
8. Guimond, N.; Gorelsky, S. I.; Fagnou, K. J. Am. Chem. Soc. 2011, 133, 6449.
9. Robbins, D. W.; Hartwig, J. F. Science 2011, 333, 1423.
10. For seminal reports see: (a) Lossen, H. Liebigs Ann. Chem. 1869, 150, 314; (b)
Lossen, W. Liebigs Ann. Chem. 1872, 161, 347 For review, see; (c) Bauer, L.; Exner,
O. Angew. Chem., Int. Ed. Engl. 1974, 13, 376.
11. For the bromination of N,N-diisopropylbenzamide using PivOH as the additive,
a quantitative GC analysis of the additive remaining after reaction was also
undertaken. Calibration lines for all additives were generated and the yields
calculated accordingly. In all instances the yields determined for the additives
using either the calibration lines or the simplified ‘single point’ calibration are
within 3%. This demonstrates that the relative accuracy of the simplified screen
is high, and thus is readily applicable to this type of analysis.
3
4
1147; (e) Jazzar, R.; Hitce, J.; Renaudat, A.; Sofack-Kreutzer, J.; Baudoin, O. Chem.
dEur. J. 2010, 16, 2654; (f) Wencel-Delord, J.; Dr o€ ge, T.; Liu, F.; Glorius, F. Chem.
Soc. Rev. 2011, 40, 4740; (g) Newhouse, T.; Baran, P. S. Angew. Chem., Int. Ed. 2011,
5
0, 3362; (h) Ackermann, L. Chem. Rev. 2011, 111, 1315; (i) McMurray, L.; O’Hara,
2 2 6
12. The initial report employed a catalyst system of [RhCp*Cl ] and AgSbF . Prior
F.; Gaunt, M. J. Chem. Soc. Rev. 2011, 40, 1885; (j) Yeung, C. S.; Dong, V. M. Chem.
Rev. 2011, 111, 1215; (k) Sun, C.-L.; Li, B.-J.; Shi, Z.-J. Chem. Rev. 2011, 111, 1293; (l)
Cho, S. H.; Kim, J. Y.; Kwak, J.; Chang, S. Chem. Soc. Rev. 2011, 40, 5068; (m) Kuhl,
N.; Hopkinson, M. N.; Wencel-Delord, J.; Glorius, F. Angew. Chem., Int. Ed. 2012,
to application of the robustness we established that the reaction proceeds
equally well using the air/moisture stable pre-formed cationic rhodium
complex RhCp*Cl [(MeCN) [SbF ] . To facilitate rapid experimentation we
2 3 6 2
consequently employed the cationic rhodium complex for evaluation of the
reaction.
51, 10236; (n) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem. Res. 2012,