Trace Level Quantification of Derivatized Boronic Acids by LC/MS/MS

Article abstract of DOI:10.1021/acs.oprd.8b00377

Boron-containing compounds are often controlled as potentially mutagenic impurities based on ICH M7 guidance in drug substances. Herein, we describe a simple method to derivatize a subset of organoborons, non-nitrogen-containing aryl boronic acids, in order to quantify trace levels remaining in a drug substance using LC/MS/MS. Through this derivatization we are able to increase the sensitivity, demonstrate linearity, and spiked analyte recovery in matrix down to low parts per million (ppm) levels. The feasibility of this method is proven as three different examples are described, two in which one boronic acid may be present in a drug substance and one where two boronic acids may be present in the final product.

Full text of DOI:10.1021/acs.oprd.8b00377

Article
pubs.acs.org/OPRD
Cite This: Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Trace Level Quantification of Derivatized Boronic Acids by LC/MS/
MS
Aaron F. Baldwin,*,§,† Robert North,§,† and Shane Eisenbeis§,‡
†
‡
§Pharmaceutical Sciences Small Molecule Analytical Research & Development and Chemical Research & Development, Pfizer,
Eastern Point Road, Groton, Connecticut 06340, United States
S
* Supporting Information
ABSTRACT: Boron-containing compounds are often controlled as potentially mutagenic impurities based on ICH M7
guidance in drug substances. Herein, we describe a simple method to derivatize a subset of organoborons, non-nitrogen-
containing aryl boronic acids, in order to quantify trace levels remaining in a drug substance using LC/MS/MS. Through this
derivatization we are able to increase the sensitivity, demonstrate linearity, and spiked analyte recovery in matrix down to low
parts per million (ppm) levels. The feasibility of this method is proven as three different examples are described, two in which
one boronic acid may be present in a drug substance and one where two boronic acids may be present in the final product.
KEYWORDS: boronic acid, LC/MS/MS, trace analysis, derivatization, MIDA
demonstrated at <1 ppm.20 However, the instrument itself
INTRODUCTION
■
had to be modified to allow for use of DMSO solutions.
Suzuki and Miyaura first introduced their palladium catalyzed
Coupling the ICP-MS with LC is another option, but this has
cross-coupling reaction between an organoboron and halide in
the added challenge of the organic modifiers suppressing the
1979.1,2 Since then, the breadth of the reaction has grown
analyte signal.21
exponentially to include various boron entities, such as
Herein, we describe a means of quantifying, via LC/MS/MS,
boranes, boronic acids, and esters or borate salts as well as
trace levels of boronic acids through derivatization. We set out
different catalysts, most notably being nickel.3−8 As a result of
to create a derivatization method that could be utilized in any
the versatility of the reaction and the ability to promote C−C
analytical laboratory without the need for substantial equip-
coupling of two aryl compounds, it has found itself used
ment setup or isolation of the analyte(s) in question. We also
commonly in small and large scale synthesis in the
evaluated the feasibility of reducing the API burden on the
pharmaceutical industry.9,10
instrument, as the need for higher API concentrations for
Boron itself has benefits to such biological functions as
quantification can be problematic to the ionization potential of
metabolism, bone development, etc.11,12 However, in 2011
the analyte. In order to improve the method sensitivity and
O’Donovan et al. reported that 12 of 13 of boronic acids and
quantification of the boronic acids, we chose to derivatize the
esters evaluated resulted in a positive signal for mutagenicity in
boronic acid with methyliminodiacetic acid (MIDA), which
AMES testing.13 Hansen et al. built upon this study by
forms a water stable cyclic boronic ester suitable for analysis
evaluating 44 commercially available boron-containing com-
using reversed-phase liquid chromatography and has a nitrogen
pounds with the goal of trying to build a model relating
atom which theoretically should improve the ionization
structure to mutagenicity.14 Unfortunately, in many cases, the
potential of the aryl boronate ester formed.
in silico and Q(SAR), quantitative structure−activity relation-
ship, modeling did not accurately predict the same
mutagenicity result found in the AMES testing. As a
consequence, in cases where AMES testing is positive or has
not been carried out, boron-containing compounds are
controlled as potentially mutagenic impurities within pharma-
ceuticals in accordance with ICH M7 guidelines.15−18
Within the guidelines, the maximum allowable intake of a
mutagenic impurity is based on the length of dosing. For a >10
years to lifetime usage of drug, the intake limit of the
mutagenic impurity is capped at 1.5 μg/day.19 This low
threshold necessitates the need for highly sensitive analytical
methods. Ideally, a chromatographic method could be
developed using common detectors, such as flame ionization
or UV, but these could be limited by the need for highly
concentrated API solutions to reach suitable sensitivity. Patel
et al. developed a process to detect boron at trace levels using
ICP-MS in which detection/quantiation of boron was
EXPERIMENTAL SECTION
■
Materials and Methods. Phenylboronic acid, trifluoro-
acetic acid, and ammonium bicarbonate were purchased from
Sigma-Aldrich. Methyliminodiacetic acid (MIDA) and N,N-
dimethylacetamide were bought from Acros Organics.
Dimethyl sulfoxide and tetrahydrofuran were purchased from
EMD Millipore. Acetonitrile and formic acid were procured
from Fisher.
Derivatization Procedure. There are numerous examples
of generating boronate esters from boronic acids and MIDA
described in the literature.22−24 A traditional method is to
perform the reaction in a toluene/DMSO mixture at reflux in
order to solubilize the MIDA and in which the water formed as
Received: November 14, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.8b00377
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a byproduct is azeotropically distilled off. The derivatization
reaction described here looks to perform the reaction in only
DMSO without the need to remove the water byproduct via
glassware setup, i.e. a Dean−Stark trap. Scheme 1 illustrates
RESULTS AND DISCUSSION
■
Initially the derivatization reaction was attempted using THF
or acetonitrile. However, 1.15 mmol of MIDA proved to be
sparsely soluble in 5 mL of these solvents without elevating the
temperature to reflux. Therefore, the derivatization was done
in a polar, high boiling point solvent. Using DMAc as a solvent
resulted in a conversion of phenylboronic acid to phenyl-
boronate ester of 80.6%:19.4%, (product:reactant) after 1 h at
80 °C. In an attempt to achieve higher conversions, DMSO
was used in place of DMAc. Conversion after 1 h was increased
to 94.7%:5.3%, product/reactant with the solvent switch. The
reaction was also evaluated after 1.5 and 3.5 h at 80 °C, and
there was no significant increase in conversion found, i.e. less
than 0.2%. NMR confirmed the LC results that the
phenylboronate ester was being produced, as the protons of
the methylene group of the MIDA boronate were two doublets
of singlets in contrast to a singlet for MIDA alone; see
Supporting Information for chromatogram and NMR spectrum
of derivatization reaction performed in DMSO.
A portion of the derivatization reaction was worked up to
isolate the phenylboronate ester. This isolated product was
then analyzed using LC/MS/MS positive ionization mode
(positive mode). A linearity range from 0.5 to 25 ng/mL was
prepared, and an S/N of 19.4 was obtained at 0.5 ng/mL with
an LOD of 0.1 ng/mL. In order to test the feasibility of doing
the quantification without the need to isolate the analyte first, a
second derivatization reaction was performed in which a
linearity curve ranging from 1.8 to 17.3 ng/mL was prepared
by serial dilutions from a stock solution made by diluting 1 mL
of the reaction solution to 50 mL. Linearity was achieved with
an r-value of 0.9911 for the regression. The S/N for the lowest
point in the linearity curve was only 9.3, so based on an ideal
S/N of 10 in order to do quantification; the LOQ would be 2
ng/mL with a calculated LOD of 0.6 ng/mL.
Phenylboronate ester was chosen as a “model” compound to
develop the derivatization conditions, as it is a simple non-
nitrogen-containing aryl boronic acid. Our next goal was to do
quantification of an aryl boronic acid, BA-1, used in the
synthesis of an API, API-1. BA-1, which is electron deficient,
was measured in API-1 as either the native acid or derivatized
ester in both positive and negative ionization mode (negative
mode). Table 1 shows the results of each LC/MS/MS run,
while Figure 1 compares the TIC, total ion chromatogram, of
either single ion or MRM transition of underivatized and
derivatized BA-1 at the lowest concentration in the measured
linear range. Measuring the level of the boronic acid proved to
be more challenging when the acid itself was used as the
analyte. In the negative mode, BA-1 could only be measured
Scheme 1. General Derivatization Reaction Scheme of Aryl
Boronic Acids with MIDA
the derivatization reaction performed on all boronic acids
described in the paper. In general 1.0 mmol of the boronic acid
and 1.15 mmol of MIDA are weighed and transferred to a
scintillation vial. To the vial was pipetted 5 mL of DMSO. The
contents of the vial are then heated to 80−90 °C and stirred
for 1 h. The reaction solution is then evaluated for LC
conversion and purity.
For NMR analysis the derivatization is performed in a
similar manner, but on a scale of 0.20 mmol of boronic acid
and 0.23 mmol of MIDA in 1 mL of DMSO-d6.
Instrumentation. Purity and conversion were evaluated on
a Waters H-Class LC system. Separation of boronic acid and
boronic ester is performed on a Waters Acquity UPLC HSS T3
2.1 mm × 100 mm, 1.8 μm column at 45 °C with a 0.1%
trifluoroacetic acid in water and acetonitrile gradient at a flow
rate of 0.5 mL/min. Mass spectra were collected on a Waters
single quadrupole spectrometer to confirm esterification. NMR
spectra were collected on a Bruker 400 MHz instrument, and
data were processed using an ACD/Lab’s Spectrus Processor.
Quantification of boronic acids and boronate esters was
perfomed using an Agilent 1200 LC with a Sciex 6500 triple
quadrupole mass spectrometer. Separation of the boronate
ester(s) from the API is achieved using a Waters Acquity
UPLC BEH Shield RP18 2.1 mm × 100 mm, 1.7 μm column
at 45 °C with either 0.1% formic acid in water or 10 mM
ammonium bicarbonate in water and acetonitrile gradient at a
flow rate of 0.5 mL/min. The gradient was adjusted for each
case in order to achieve separation between analyte(s) and
APIs. A valve was employed to divert the flow at points in the
gradient to ensure that DMSO and API did not reach the triple
quadrupole.
Standard and Sample Preparation. In general, a 1 mg/
mL stock solution of boronate ester is prepared from the
derivatization reaction mixture based on the reaction purity/
conversion found by LC. From the stock solution, serial
dilutions are made with acetonitrile to create a linearity curve
for each boronate ester. An unspiked API sample is prepared
by dissolving 50 mg of API and 170 mg of MIDA in DMSO
and heating the mixture to 80−90 °C for 1 h. After the
contents have re-equilibrated to room temperature, 1 mL of
the reaction solution is diluted to 10 mL with acetonitrile. The
solution is then filtered through a 0.45 μm PTFE filter to
remove any API or MIDA that may precipitate. An API spiked
solution is prepared in a similar fashion to the unspiked
solution, with the exception that the API and MIDA are
solubilized in 5 mL of a 100 ng/mL solution of the boronic
acid in DMSO.
Table 1. Quatitation Results of BA-1 and BA-1 Ester
LOD/
a
Matrix
Recovery
[%]
System
Precision
[%]
SIM/
LOQ
Analyte Mode MRM
[ng/mL]
r-value
BA-1
BA-1
BA-1
Neg
Pos
Neg
SIM
MRM
MRM
4.8/15.9
>50/>50
0.8/2.5
63.9
ND
96.7
0.9983
ND
0.9980
33.1
ND
0.43
ester
BA-1
ester
Pos
MRM
1.4/4.7
99.8
0.9999
6.9
a
LOD and LOQ are calculated based on the S/N found for the lowest
point in the calibration curve and what corresponding concentration
would give an LOD with an S/N of 3 and an LOQ with an S/N of 10.
B
DOI: 10.1021/acs.oprd.8b00377
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 1. TIC of single ion or MRM transition of BA-1 and BA-1 ester. (a) BA-1 SIM in negative mode at 25 ng/mL. (b) BA-1 MRM in positive
mode at 50 ng/mL. (c) BA-1 ester MRM in negative mode at approximately 5 ng/mL. (d) BA-1 ester MRM in negative mode at approximately 5
ng/mL.
using SIM, as BA-1 did not give consistent fragmentation in
Q2. Although BA-1 could be detected at 25 ng/mL, some
matrix effects were observed as the recovery at 25 ng/mL was
only 63.9%. Evaluating BA-1 under positive mode allowed for
fragmentation which in theory would result in greater
sensitivity. However, a 50 ng/mL solution of BA-1 did not
exhibit any discernible peak. By derivatizing BA-1, detection
and quantitation down to trace levels were found to be more
practical; see Table 1 for quantification results summary. In
positive mode, linearity was achieved over a range of 5.6−55.7
ng/mL with an r-value of 0.9999 and an S/N of 12 at the
lowest level in the curve. Recovery of the BA-1 ester was done
in two ways. Recovery of the BA-1 ester in which the API was
spiked with an 11.5 ng/mL solution of the ester averaged
86.0%. Recovery of the BA-1 ester, in which a 100 ng/mL
solution of BA-1 was used and derivatization was carried out
with the API-1 present, averaged 99.8%. Unspiked samples of
API-1 did not show any detectable level of BA-1 ester.
Evaluation of BA-1 ester using negative mode gave similar
results to the positive mode. A linear range from 5.3 to 52.8
ng/mL had an r-value of 0.9980 with an S/N at the lowest
point of 20.8. Injection precision was also evaluated in this
case, and five injections at 5.3 ng/mL gave an RSD of 0.43%.
Recovery in matrix was only done using a 100 ng/mL solution
of BA-1 and averaged 96.7%. Solution stability was also
evaluated by measuring the recoveries of two different 100 ng/
mL solutions of BA-1 ester prepared 15 days apart versus the
linear regression. Recoveries were 103.5% for the older
solution and 102.5% for the newer solution. As with the
positive mode, no detectable BA-1 ester was found in the
unspiked API-1 solution.
methylene-carbonyl bond; see Figure 2 for proposed fragment
structures. Tuning the mass spectrometer on this fragmenta-
Figure 2. Common ion fragmentation for boronate esters in positive
mode.
tion resulted in a linear calibration curve with an r-value of
0.9995, an injection precision of 4.4%, and an S/N of 25 at the
lowest concentration (4.87 ng/mL) in the calibration curve.
The spiked recovery of BA-2 in the presence of the
intermediate was 97.8%.
Our final test aimed at quantitfying two boronic acids, BA-3
and BA-4, used in the synthesis of API-3 in one sequence. BA-
3 and BA-4 both have one electron-withdrawing and one
electron-donating group attached to the phenyl ring. As with
the previous experiment for BA-1, both positive and negative
modes were used to attempt to quantify either the boronic
acids or boronate esters. The derivatizations of BA-3 and BA-4
were carried out in the same scintillation vial, and standard
solutions for the linearity curves were prepared from stock
solutions containing either the two boronic acids or boronate
esters. The LC gradient was altered to ensure separation of
BA-3 and BA-4 from API-3. When measuring levels as the
boronic acid, the sensitivity was on par with BA-1 in that no
discernible peak for either acid could be seen at 50 ng/mL. As
the derivatized species, a lower detection/quantitation thresh-
old was obtained. Table 2 illustrates the quantitative data for
BA-3 and BA-4. Quantitation using the negative mode gave
The residual levels of a second boronic acid, an electron-rich
BA-2, was analyzed in an intermediate in the synthetic pathway
to a different API, API-2, as a means of challenging the
derivatization and detection methods. A common ion fragment
in positive mode seen in both phenylboronate ester and BA-1
was 102. This fragment is attributed to the cleavage of the
eight-member ring system at a boron−oxygen bond as well as a
C
DOI: 10.1021/acs.oprd.8b00377
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 2. Quatitation Results of BA-3 and BA-4 Esters
Analyte
Mode
SIM/MRM
Linearity Range [ng/mL]
Matrix Recovery [%]
r-value
System Precision [%]
BA-3 ester
BA-3 ester
BA-4 ester
BA-4 ester
Neg
Pos
Neg
Pos
MRM
MRM
MRM
MRM
1.79−179
8.95−179
1.54−154
7.72−154
121
131
113
107
0.9998
0.9922
0.9997
0.9987
8.74
8.08
3.70
4.90
in the Presence of Palladium Catalyst. J. Chem. Soc., Chem. Commun.
1979, 19, 866−867.
greater sensitivity, in agreement with what was observed for
BA-1, and linearity in comparison to the positive mode.
(3) Roy, D.; Uozumi, Y. Recent Advances in Palladium-Catalyzed
Cross-Coupling Reactions at ppm to ppb Molar Catalyst Loadings.
Adv. Synth. Catal. 2018, 360, 602−625.
CONCLUSION
■
In summary we have developed a simple derivatization
technique to determine as low as 1−5 ppm of non-nitrogen-
containing aromatic boronic acids in APIs or API inter-
mediates without the need to isolate the analyte(s) in question.
Boronate esters were quantified in three different cases, and the
method has proven to work in accurately determining a system
in which more than one boronic acid may be present in the
final API. Using MRM in the positive mode showed a common
ion fragment that could be used to quantify against, but across
systems, both positive and negative mode could be used in the
determination. The boronic acids used in these experiments
contain electron-donating and -withdrawing or both types of
groups as substituents. As a result of substituent variability a
more systematic study would be helpful to evaluate the effects
of these groups on the ionization efficiency of the parent as
well as the fragmentation potential of the MIDA portion of the
analyte. Ideally, the results of this study would then allow for
initial LC/MS/MS conditions to be set for new analytes based
on their structure. Other means of increasing sensitivity, i.e.
raising the source temperature or using smaller inner diameter
columns, as well as derivatizing other types of boron
containing species will also be explored.
(4) Len, C.; Bruniaux, S.; Delbecq, F.; Parmar, V. S. Palldium-
Catalyzed Suzuki-Miyaura Cross-Coupling in Continuous Flow.
Catalysts 2017, 7, 146.
(5) Willemse, T.; Schepens, W.; van Vlijmen, H.; Maes, B.; Ballet, S.
The Suzuki−Miyaura Cross-Coupling as a Versatile Tool for Peptide
Diversification and Cyclization. Catalysts 2017, 7, 74.
(6) Chatterjee, A.; Ward, T. R. Recent Advances in the Palladium
Catalyzed Suzuki-Miyaura Cross-Coupling Reaction in Water. Catal.
Lett. 2016, 146, 820−840.
(7) Maluenda, I.; Navarro, O. Recent Developments in the Suzuki-
Miyaura Reaction: 2010−2014. Molecules 2015, 20, 7528−7557.
(8) Ranjani, G.; Nagarajan, R. Insight into Copper Catalysis: In Situ
Formed Nanao Cu2O in Suzuki-Miyaura Cross-Coupling of Aryl/
Indolyl Boronates. Org. Lett. 2017, 19, 3974−3977.
(9) Almond-Thynne, J.; Blakemore, D. C.; Pryde, D. C.; Spivey, A.
C. Site-selective Suzuki-Miyaura coupling of heteroaryl halides-
understanding the trends for pharmaceutically important classes.
Chem. Sci. 2017, 8, 40−62.
(10) Torborg, C.; Beller, M. Recent Applications of Palladium-
Catalyzed Coupling Reactins in the Pharmaceutical, Agrichemical,
and Fine Chemical Industries. Adv. Synth. Catal. 2009, 351, 3027−
3043.
(11) Nielsen, F. H. Update on human health effects of boron. J.
Trace Elem. Med. Biol. 2014, 28, 383−387.
(12) Uluisik, I.; Karakaya, H. C.; Koc, A. The importance of boron
in biological systems. J. Trace Elem. Med. Biol. 2018, 45, 156−162.
(13) O’Donovan, M. R.; Mee, C. D.; Fenner, S.; Teasdale, A.;
Phillips, D. H. Boronic acids-A novel class of bacterial mutagen.
Mutat. Res., Genet. Toxicol. Environ. Mutagen. 2011, 724, 1−6.
(14) Hansen, M. M.; Jolly, R. A.; Linder, R. J. Boronic Acids and
Derivatives-Probing the Structure-Activity Relationships for Muta-
genicity. Org. Process Res. Dev. 2015, 19, 1507−1516.
(15) Scott, H.; Walmsley, R. M. Ames positive boronic acids are not
all eukaryotic genotoxins. Mutat. Res., Genet. Toxicol. Environ.
Mutagen. 2015, 777, 68−72.
(16) Galloway, S. M.; Reddy, M. V.; McGettigan, K.; Gealy, R.;
Bercu, J. Potentially mutagenic impurities: Analysis of structural
classes and carcinogenic potencies of chemical intermediates in
pharmaceutical syntheses supports alternative methods to the default
TTC for calculating safe levels of impurities. Regul. Toxicol. Pharmacol.
2013, 66, 326−335.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.oprd.8b00377.
■
S
Chromatogram and NMR spectrum of derivatization
reaction performed in DMSO (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: aaron.baldwin@pfizer.com.
ORCID
Aaron F. Baldwin: 0000-0002-5621-0555
Notes
The authors declare no competing financial interest.
■
́
́
́
(17) Soriano-Ursua, M. A.; Farfan-García, E. D.; Lopez-Cabrera, Y.;
Querejeta, E.; Trujillo-Ferrara, J. G. Boron-containing acids:
Preliminary evaluation of acute toxicity and access to the brain
determined by Raman scattering spectroscopy. NeuroToxicology 2014,
40, 8−15.
ABBREVIATIONS
ppm, parts per million; API, active pharmaceutical ingredient;
MIDA, methyliminodiacetic acid
■
(18) McLaughlin, M.; Dermenjian, R. K.; Jin, Y.; Klapars, A.; Reddy,
M. V.; Williams, M. J. Evaluation and Control of Mutagenic
Impurities in a Development Compound: Purge Factor Estimates vs
Measured Amounts. Org. Process Res. Dev. 2015, 19, 1531−1535.
(19) ICH Guideline M7 (R1) Step 4 Version on Assessment and
Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals
to Limit Potential Carcinogenic Risk, 31 March 2017. http://www.
ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/
Multidisciplinary/M7/M7_R1_Addendum_Step_4_2017_0331.pdf
(accessed Sep. 7, 2018).
REFERENCES
■
(1) Miyaura, N.; Yamada, K.; Suzuki, A. A New Stereospecific Cross-
Coupling by the Palladium Catalyzed Reaction of 1-Alkenylboranes
with 1-Alkenyl or 1-Alkynyl Halides. Tetrahedron Lett. 1979, 20,
3437−3440.
(2) Miyaura, N.; Suzuki, A. Stereoselective Synthesis of Arylated
(E)-Alkenes by the Reaction of Alk-1-enylboranes with Aryl Halides
D
DOI: 10.1021/acs.oprd.8b00377
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(20) Patel, I.; Venkatramani, C. J.; Stumpf, A.; Wigman, L.; Yehl, D.
Trace Analysis of Potentially Mutagenic Boronic Acids and Esters in
Drug Substance by ICP-MS. Org. Process Res. Dev. 2017, 21, 182−
186.
(21) Forsgard, N.; Nilsson, E.; Andersson, M.; Pettersson, J.
Investigation of matrix effects in boron determination using organic
solvents as modifiers for liquid chromatography coupled to ICP-MS. J.
Anal. At. Spectrom. 2006, 21, 305−310.
(22) Wang, H.; Zeng, Y.-F.; Lv, W.-X.; Tan, D.-H. Synthetic
Transformations of Alkenyl MIDA Boronates toward the Efficient
Construction of Organoborons. Synlett 2018, 29, 1415−1420.
́
(23) Khanizeman, R. N.; Barde, E.; Bates, R. W.; Guerinot, A.;
Cossy, J. Modular Access to Triarylethylene Units from Arylvinyl
MIDA Boronates Using a Regioselective Heck Coupling. Org. Lett.
2017, 19, 5046−5049.
(24) Close, A. J.; Kemmitt, R.; Roe, S. M.; Spencer, J. Regioselective
routes to orthogonally-substituted aromatic MIDA boronates. Org.
Biomol. Chem. 2016, 14, 6751−6756.
E
DOI: 10.1021/acs.oprd.8b00377
Org. Process Res. Dev. XXXX, XXX, XXX−XXX