Simple Determination of Double-Bond Positions in Long-Chain
Olefins by Cross-Metathesis**
Yongseok Kwon, Seonwoo Lee, Dong-Chan Oh,* and Sanghee Kim*
The accurate determination of the double-bond positions in
unsaturated long-chain compounds remains a challenging
task, even in this advanced spectroscopic era.Although
NMR spectroscopy is the principal method of structural
determination, it does not provide adequate information
regarding the double-bond positions in unsaturated chains,
owing to highly overlapped signals in NMR spectra. Thus,
current analytical approaches for this purpose rely on mass
Direct analysis of double-bond positions in the analyte
with a conventional form of MS, such as electron impact (EI)
or chemical ionization (CI) MS, is inherently unreliable.[1,4]
This unreliability occurs because EI ionization causes rapid
isomerization of the molecular ions, and CI with conventional
proton-transfer reagents often fails to yield useful fragments.
These problems could be solved either by using advanced
mass spectrometric equipment or special CI reactant gases[2,5]
or both. However, these methods have limited potential for
non-specialist laboratories, and they are not adapted to
More detailed and specific mass information on the
double-bond position can be acquired from derivatives of the
target analyte. Various types of chemical reactions have been
employed for derivatization.Some representative reactions
past chemical derivatization approaches is that the above
reactions do not generally provide compounds with phys-
icochemical properties adequate for chromatographic mass
spectrometry. Thus, to acquire the desired properties of the
analyte, a second derivatization is necessary in most cases. An
additional drawback is the possible interference of other
functional groups with the analyte during the derivatizations,
which could cause failure in forming the expected derivatives
and thus lead to ambiguous mass information. Consequently,
there still remains a need for simple and reliable methods to
identify the double-bond positions with high accuracy and
Olefin cross-metathesis (CM) is a metal-catalyzed process
that yields a new carbon–carbon double bond by an inter-
molecular mutual exchange of alkylidene fragments between
two olefins.This versatile reaction has had a wide and
profound impact in diverse areas of chemistry. However, the
application of CM in the field of analytical chemistry is rare.
To the best of our knowledge, there is only one report
concerning an application of CM for analytical purposes. Gee
and Prampin used CM to estimate the positions of double
bonds in mixtures of linear olefins.Their GC/MS-based
method requires about a gram of sample and a specialized
computer algorithm, and it has some limitations because it is
only applicable to a mixture of unsaturated hydrocarbons
with one double bond.
We envisioned that a CM reaction of unsaturated long
chains with a simple olefin could afford derivatives suitable
for chromatographic mass spectrometry without the need for
an additional derivatization step. The exhaustive interpreta-
tion of the fragment ion mass spectra of CM products is not
required to determine the original double-bond position.
Instead, the position can be simply deduced from the values
obtained by comparing the mass changes between the starting
materials and the CM products and considering the structural
characteristics of the starting olefins. Compared to conven-
tional oxidative derivatization, the reaction conditions of CM
are mild enough to tolerate a variety of functional groups.
Thus, CM-based position determination could be applied to
complicated compounds. Herein, we present our studies on
the development of a simple and reliable CM-based method
for the determination of the double-bond position in long
chain compounds by either LC/MS or GC/MS.
For a successful application of the CM reaction to the
analysis of double-bond positions, the CM partner and CM
catalyst should fulfill the following criteria: 1) CM partner
and catalyst should be easily available; 2) they can be handled
on the bench without any special equipments or techniques;
3) they should maximize the formation of CM products,
without giving a mixture of geometrical isomers, for clear
interpretation of data; and 4) they must not impede mass
spectrometry analysis of the products or can be easily
Based on the criteria listed above, we surveyed several
metathesis catalysts and CM partners to discover an optimal
set of reaction components. This effort resulted in the
selection of two sets of reaction components (Scheme 1):
set A is methyl acrylate (1)/second-generation Hoveyda–
[*] Y. Kwon, S. Lee, Prof. Dr. S. Kim
College of Pharmacy, Seoul National University
San 56-1, Shilim, Kwanak, Seoul 151-742 (Korea)
Prof. Dr. D.-C. Oh
Natural Products Research Institute, College of Pharmacy
Seoul National University (Korea)
[**] This work was supported by the World Class University Program
(R32-2008-000-10098-0) and the MarineBio Research Program
(NRF-C1ABA001-2010-0020428) of the National Research Founda-
tion of Korea (NRF) grant funded by the Korean government
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2011, 50, 8275–8278? 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Grubbs catalyst 2, and set B is 2-methyl-2-butene (3)/ second-
generation Grubbs catalyst 4.The CM reaction of elaidic
acid (5) with set A in CH2Cl2led to the formation of CM
products 6 and 7 with the E configuration in high yields, and
the set B formed gem-dimethyl olefins 8 and 9.
We envisioned that set A would be useful for developing
an LC/MS-based analytical method because set A generates a
CM product with increased hydrophilicity that is preferable
for reverse-phase LC. Furthermore, the UV-active a,b-
unsaturated ester functionality makes it possible to unambig-
uously identify the CM product by LC/MS with a common
diode-array UV detector, even with interference from non-
specific signals. However, it was expected that set B would be
suitable for GC/MS analysis because set B generates, at least
in part, a CM product that is more volatile relative to the
After finding two sets of reaction components, we
explored the optimized reaction conditions and procedures
that are suitable for chromatography–mass spectrometry.
After extensive efforts, we could set up the standard
procedure A (PA) for LC/MS analysis:1) Dissolve a
target molecule (?0.5 mg) in CH2Cl2/methyl acrylate 10:1
mixture (0.5 mL); 2) add 50 mg (5–10 mol%) of second-
generation Hoveyda–Grubbs catalyst 2; 3) stir at room
temperature for 2–3 h; 4) inject 10 mL of the reaction mixture
into an LC/MS (conventional C18column, H2O/MeOH 0.1%
formic acid gradient solvent) without any treatment; and
5) analyze the data based on the mass values.
The LC/MS spectrum of the analyte, obtained from elaidic
acid (5) using procedure PA, is shown in Figure 1. The major
peak showed UVabsorbance at 218 nm, which is typical of the
a,b-unsaturated ester (Figure 1b). Its identity could be
deduced by the molecular ions: [M?H]?at m/z 227 in the
ESI negative ionization mode and [M+ +H]+at m/z 229 in the
positive mode (Figure 1c and Supporting Information). The
obtained molecular mass (228 Daltons) of this CM product
matched well with the theoretical value for 6, and this result
clearly indicated that the double-bond position is at C-9 of
elaidic acid. The homodimerized product was also detected as
a minor peak (Figure 1a). However, only low levels of the
other CM product 7 were observed, possibly as a result of
To validate procedure PA, we applied it to various
unsaturated fatty acids. As shown in Table 1, all of the
tested fatty acids successfully yielded the expected CM
products, regardless of their geometry or position of the
double bond. The fatty acid positional isomers gave CM
products with mass values that differed from each other
(entries 1–4, Table 1). Octadecenoic acid and erucic acid
produced the same CM fragment (entries 4 and 6) because
their double bond occurs at the same number of carbon atoms
away from the carboxylic group. However, this similarity is
not a hurdle in assigning double-bond positions because these
two fatty acids can be easily distinguished by molecular mass.
A long-chain fatty acid with two double bonds, namely
linoleic acid, led to the detection of the CMproduct that arose
from the first double bond from the carboxylic terminus
Scheme 1. The olefin cross-metathesis of 5 with set A and set B.
Figure 1. LC/MS analysis of the CM products from elaidic acid 5.
a) Liquid chromatogram; b) the UV spectrum of the peak at 16.4 min
in the LC; and c) the ESI negative ionization mode mass spectrum of
the peak at 16.4 min in LC.
Table 1: Cross-metathesis-based LC/MS analysis of various fatty acids.
EntryFatty acids MWMW of CM product[a]
[a] Standard procedure PA was employed.
? 2011 Wiley-VCH Verlag GmbH & Co. KGaA, WeinheimAngew. Chem. Int. Ed. 2011, 50, 8275–8278
(entry 8). The peak corresponding to the product derived
from the reaction at the second double bond was not
detected.A similar pattern was also observed for arachi-
donic acid, which has four double bonds (entry 9). These
results indicate that set A results in the predominant
formation of the CM product of the first double bond,
regardless of the number of double bonds in long linear
As mentioned above, CM set B would be useful for GC/
MS analysis because the CM products using 2-methyl-2-
butene (3) would be more volatile than those produced using
methyl acrylate (1). Furthermore, 3 can easily be removed
from the final products owing to its low boiling point (398 8C).
Therefore, the standard GC/MS procedure PB was developed
through an optimization process. One notable difference
between PA and PB is that extra manipulation is necessary to
remove the nonvolatile ruthenium byproducts that are
generated from the catalyst with activated carbon for the
protection of the capillary column.
The suggested procedure PB is as follows:1) Dissolve a
target molecule (?0.5 mg) in CH2Cl2/2-methyl-2-butene 1:1
mixture (0.5 mL); 2) add 50 mg (5–10 mol%) of the second-
generation Grubbscatalyst; 3) stir at room temperature for2–
3 h; 4) add 50 mg of activated carbon and stir for 10 min;
5) filter the reaction mixture with the eluent (CH2Cl2/Et2O
5:1) through a pipette column filled with 300 mg of silica gel;
6) remove 2-methyl-2-butene and eluent at low vacuum
(approximately 60 mbar) on a rotary evaporator; 7) dissolve
the residue in 2 mL of CH2Cl2; 8) inject the dissolved material
into a GC/MS with a conventional silica column; and
9) analyze the data based on the mass values.
Figure 2 shows the GC/MS profiles of the CM products
that were obtained from elaidic acid (5) and its methyl ester
derivative using PB. The gas chromatogram of the analyte
obtainedfrom fatty acid5 showedonly one predominantpeak
corresponding to the CM product 9.The peak correspond-
ing to the other possible product 8, containing a carboxylic
acid group, was not observed during the monitored period
owing to its high polarity and boiling point. However, the
chromatogram derived from the methyl ester of 5 exhibited
two peaks of similar intensity (Figure 2b). A similar pattern
was also observed for other fatty acids and their ester
derivatives, including nervonic acid (see Supporting Informa-
It is notable that our CM-based GC/MS analytic approach
to unsaturated fatty acids, unlike most other oxidative
derivatization approaches, does not require prior esterifica-
tion of the acid group because the CM reaction of fatty acid
with 3 certainly affords a nonpolar product suitable for GC.
To further demonstrate the utility of our positioning
method, we applied the standard CM-based LC/MS proce-
dure (PA) to the natural product olvanil, which has a
molecular weight of 417 (Figure 3). In this experiment, we
pretended not to know the exact position of the double bond.
The LC/MS chromatogram of the analyte had two peaks: a
major peak of the molecular ion at m/z 362 ([M?H]?) and a
minor peak at 581. Because the mass value of the minor peak
is larger than the sum of the individual mass values of olvanil
and methyl acrylate. The minor peak is assigned as a
homodimerization product of olvanil. The double-bond
position could easily be deduced from the value obtained
from the major peak.The sum of the mass values of olvanil
and methyl acrylate is 503, and the molecular mass of the
product is 363. The difference between the two values is 140,
which corresponds to the mass of C10H20. Because acrylate
inherently contributes one carbon atom to the deleted ten
carbon atoms, nine carbon atoms out of ten originate from
olvanil. Consequently, it can be deduced that olvanil has a
double bond at the ninth carbon position from the methyl
In conclusion, we developed a new chemical derivatiza-
tion method for the determination of double-bond positions
in unsaturated long-chain compounds. The method is based
on the cross-metathesis reaction between a target compound
and a simple olefin, which results in the formation of
fragmented olefinic products. Depending on the cross-meta-
thesis partner used, the produced CM fragments have distinct
physicochemical properties that are suitable for LC/MS and
Figure 2. GC/MS analysis of the CM products from elaidic acid 5 and
its methyl ester. Gas chromatograms of a) elaidic acid and b) elaidic
acid methyl ester; mass spectra (EI 70 eV) of the peaks at c) 12.5 min
and d) 17.0 min in GC.
Figure 3. Determination of the position of the double bond in olvanil
Angew. Chem. Int. Ed. 2011, 50, 8275–8278 ? 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
GC/MS analyses. The position of the double bond can easily
be deduced by comparing and analyzing the molecular mass
changes. Both of the presented LC/MS and GC/MS
approaches are equally reliable, and either one can be
chosen depending upon the availability of the instruments.
approaches, our proposed CM-based approach could be
applicable to more complicated compounds because the
reaction conditions of CM are mild enough to tolerate a
variety of functional groups. Furthermore, the method is
operationally simple and applicable at a submilligram scale.
Thus, we believe that our CM-based approach would be
practically useful, especially for nonspecialized laboratories.
The presented method is applicable to pure olefinic com-
pounds, but there is some limitation on its use in olefin
mixtures, such as biological extracts. Further studies and
improvements to the method are needed for applications in
analyzing a complex mixture of unsaturated hydrocarbons.
Received: April 16, 2011
Published online: June 24, 2011
Keywords: double bonds · fatty acids · metathesis · olefins ·
 For a review, see: a) S. J. Blanksby, T. W. Mitchell, Annu. Rev.
Anal.Chem.2010, 3, 433–465; b) T. W. Mitchell, H.Pham,M. C.
Thomas, S. J. Blanksby, J. Chromatogr. B 2009, 877, 2722–2735;
c) N. J. Jensen, M. L. Gross, Mass Spectrom. Rev. 1987, 6, 497–
536; d) B. Schmitz, R. A. Klein, Chem. Phys. Lipids 1986, 39,
 For selected recent examples, see: a) V. Vrkoslav, M. H?kov?, K.
Peckov?, K. Urbanov?, J. Cvac ˇka, Anal. Chem. 2011, 83, 2978–
2986; b) H. J. Yoo, K. H?kansson, Anal. Chem. 2010, 82, 6940–
6946; c) Y. Xu, J. T. Brenna, Anal. Chem. 2007, 79, 2525–2536;
d) M. C. Thomas, T. W. Mitchell, S. J. Blanksby, J. Am. Chem.
Soc. 2006, 128, 58–59.
 For selected recent examples, see: a) V. V. Homann, M. Sandy,
J. A. Tincu, A. S. Templeton, B. M. Tebo, A. Butler, J. Nat. Prod.
2009, 72, 884–888; b) H. Liu, F. Zhao, R. Yang, M. Wang, M.
Zheng, Y. Zhao, X. Zhang, F. Qiu, H. Wang, Phytochemistry
2009, 70, 773–778; c) T. Ishii, T. Okino, Y. Mino, J. Nat. Prod.
2006, 69, 1080–1082.
 a) F. Borchers, K. Levsen, H. Schwarz, C. Wesdemiotis, H. U.
Winkler, J. Am. Chem. Soc. 1977, 99, 6359–6365; b) C. K.
Van Pelt, J. T. Brenna, Anal. Chem. 1999, 71, 1981–1989.
 a) R. E. Doolittle, J. H. Tumlinson, A. Proveaux, Anal. Chem.
1985, 57, 1625–1630; b) H. Budzikiewicz, E. Busker, Tetrahe-
dron 1980, 36, 255–266.
 A. B. Attygalle, E. D. Morgan, Angew. Chem. 1988, 100, 475–
494; Angew. Chem. Int. Ed. Engl. 1988, 27, 460–478.
 M. Beroza, B. A. Bierl, Anal. Chem. 1967, 39, 1131–1135.
 P. Capella, C. M. Zorzut, Anal. Chem. 1968, 40, 1458–1463.
 H.-R. Buser, H. Arn, P. Guerin, S. Rauscher, Anal. Chem. 1983,
 R. T. Aplin, L. Coles, J. Chem. Soc. Chem. Commun. 1967, 858–
 D. E. Minnikin, Lipids 1975, 10, 55–57.
 For a review, see: a) G. C. Vougioukalakis, R. H. Grubbs, Chem.
Rev. 2010, 110, 1746–1787; b) S. J. Connon, S. Blechert, Angew.
Chem. 2003, 115, 1944–1968; Angew. Chem. Int. Ed. 2003, 42,
 J. C. Gee, D. S. Prampin, Anal. Chem. 2009, 81, 1646–1651.
 Other low-molecular-weight olefins (such as ethylene, isobuty-
lene, 2,3-dimethyl-2-butene, acrylonitrile, and methacrylonitrile)
would not meet the criteria, especially in terms of their boiling
points and/or poor reaction efficacy.
 The reaction scale could be increased or reduced by a factor of
up to 5 with no adverse affects.
 When catalyst 2 in PAwas replaced by catalyst 4, we observed an
additional major peak that could be assigned to the product
resulting from the CM reaction on the second double bond.
 For the removal of ruthenium byproducts by activated carbon,
see: J. H. Cho, B. M. Kim, Org. Lett. 2003, 5, 531–533.
 The peak corresponding to the homodimerization product was
detected, but it was negligible.
 The mass value of the homodimer also provides useful informa-
tion regarding the position of the double bond. The mass
difference between the homodimer and two molecules of olvanil
is 252 Daltons. This value indicates that C18H36was completely
removed during the homodimerization. It also indicates that
nine carbon atoms were abstracted from olvanil. Thus, the
double-bond position of olvanil is ninth from the terminal
? 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2011, 50, 8275–8278