Photoreaction of rac-Leucine in ice by circularly polarized synchrotron radiation: Temperature-induced mechanism switching from Norrish type II to deamination

Article abstract of DOI:10.1002/chem.201301831

The delivery of extraterrestrial organics to primitive Earth is considered to have triggered the origin and subsequent evolution of life. Indeed, enantiomerically enriched amino acids of nonterrestrial origin have been found in carbonaceous meteorites, and enantioselective photodecomposition by circularly polarized light (CPL) in outer space has been proposed to have played some role in the initial enantiomeric bias. To experimentally examine this possibility and elucidate the photoreaction mechanisms, we have studied the photolysis of racemic leucine (rac-Leu) in acidic and neutral ice/water media at 21-298 K with left- and right-CPL in an attempt to detect enantiomerically enriched D- and L-Leu, respectively. Comprehensive product analyses revealed that the CPL-induced deracemization of Leu proceeds in both acidic and neutral ice matrices even at 21 K, and that the main mechanism switches from Norrish-type II γ-hydrogen abstraction to S_Ni deamination on lowering the temperature. The potential role of the CPL-induced photodecomposition of amino acids as a source of the enantiomer imbalance in meteorites is discussed. ET arrives! Enantioselective photoreaction of rac-leucine by circularly polarized synchrotron radiation proceeds in both neutral and acidic ices at temperatures down to 21 K, with the main mechanism switching from Norrish type II γ-hydrogen abstraction to S_Ni deamination by lowering the irradiation temperature (see figure). Copyright

Full text of DOI:10.1002/chem.201301831

FULL PAPER
DOI: 10.1002/chem.201301831
Photoreaction of rac-Leucine in Ice by Circularly Polarized Synchrotron
Radiation: Temperature-Induced Mechanism Switching from Norrish Type II
to Deamination
Hideo Nishino,[b] Masahito Hosaka,[c] Masahiro Katoh,[d] and Yoshihisa Inoue*[a]
Abstract: The delivery of extraterres-
trial organics to primitive Earth is con-
sidered to have triggered the origin
and subsequent evolution of life.
elucidate the photoreaction mecha-
nisms, we have studied the photolysis
of racemic leucine (rac-Leu) in acidic
and neutral ice/water media at 21–
298 K with left- and right-CPL in an at-
tempt to detect enantiomerically en-
riched d- and l-Leu, respectively. Com-
prehensive product analyses revealed
that the CPL-induced deracemization
of Leu proceeds in both acidic and
neutral ice matrices even at 21 K, and
that the main mechanism switches
from Norrish-type II g-hydrogen ab-
straction to SNi deamination on lower-
ing the temperature. The potential role
of the CPL-induced photodecomposi-
tion of amino acids as a source of the
enantiomer imbalance in meteorites is
discussed.
Indeed, enantiomerically
enriched
amino acids of nonterrestrial origin
have been found in carbonaceous
meteoACHTUNGTRENNUNGrites, and enantioselective photo-
decomposition by circularly polarized
light (CPL) in outer space has been
proposed to have played some role in
the initial enantiomeric bias. To experi-
mentally examine this possibility and
Keywords: amino acids · chirality ·
photolysis · reaction mechanism ·
temperature effects
Introduction
In the 1980s, Rubenstein, Bonner et al. proposed a cosmic
scenario in which circularly polarized synchrotron radiation
from electrons orbiting neutron stars enantioselectively pho-
todecomposed racemic amino acids in an icy comet to leave
enantiomerically enriched antipodes that were subsequently
delivered to the primitive Earth and triggered biomolecular
homochirality.[4] However, circularly polarized light (CPL)
has never been detected in the vicinity of a neutron star and
thus the emission of CPL from neutron stars was ques-
tioned.[5] However, the star-forming region of the Orion mo-
lecular cloud was found to emit CPL in the IR region,[6] pre-
sumably provoked by Mie scattering on aligned interstellar
dust grains.[6,7] The proposal and discovery of such interstel-
lar CPL sources stimulated the investigation to elucidate the
plausibility of the cosmic scenario,[4,6,8,9] including the photo-
chemical enantiomer enrichment of amino acids by CPL in
outer space.[1a,b,9] Indeed, a variety of (racemic) aliphatic
amino acids are known to be produced upon vacuum ultra-
violet (VUV) irradiation of a matrix of H2O, CH3OH, CO2,
CO, and NH3 at 12 K under high vacuum.[10] Nevertheless,
the subsequent process, that is, the enantioselective photo-
decomposition of racemic amino acids by CPL, has not been
experimentally proven or disproven under interstellar condi-
tions.
The origin of homochirality in the biosphere is one of the
most crucial issues in chemical evolution, for which a variety
of biotic and prebiotic theories and chemical and physical
mechanisms have hitherto been proposed.[1–10] Recent mete-
orite studies have revealed that carbonaceous meteorites
carry biologically important organics, including aliphatic
amino acids.[2,3] Crucially, both the natural and unnatural
amino acids found in meteorites are not racemic but en-
riched enantiomerically as well as isotopically in 13C and 15N,
which indicates a nonterrestrial origin of the enantiomer im-
balance.[3]
[a] Prof. Dr. Y. Inoue
Department of Applied Chemistry, Osaka University
2-1 Yamada-oka, Suita 565-0871 (Japan)
Fax : (+81)6-6879-7923
E-mail: inoue@chem.eng.osaka-u.ac.jp
[b] Dr. H. Nishino
Entropy Control Project (Japan) Science and Technology Agency
Current address: Osaka Municipal Technical Research Institute
1-6-50 Morinomiya, Joto-ku, Osaka 536-8553 (Japan)
[c] Prof. Dr. M. Hosaka
Synchrotron Radiation Research Center, Nagoya University
Furo-cho, Chikusa-ku, Nagoya 464-8603 (Japan)
The principle of enantiomer enrichment upon CPL pho-
tolysis of a racemic substrate[11] rests on the relative differ-
ence in molar extinction coefficients (e) of an enantiomer
pair under left- or right-CPL (l- or r-CPL), which leads to
unequal rates of excitation and subsequent reaction. The
anisotropy factor (g), a measure of preferential excitation, is
defined as the normalized difference in e of an enantiopure
[d] Prof. Dr. M. Katoh
Institute for Molecular Science
National Institutes of Natural Sciences
Myodaiji, Okazaki 444-8585 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301831.
Chem. Eur. J. 2013, 19, 13929 – 13936
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13929

compound towards l- and r-CPL at a given wavelength: g=
(elꢀer)/e=De/e, in which e=(el+er)/2 and hence 0ꢁg<
2.[11,12]
ly enriched Leu has been found in carbonaceous meteorite-
s,[3a] 2) VUV irradiation of a mixture of simple starting ma-
terials (H2O, CH3OH, CO2, CO, and NH3) produces rac-
Leu,[10] and 3) Leu photodecomposes efficiently, at least at
ambient temperatures.[12] The photolysis of rac-Leu was per-
formed in neutral and acidic water and ice media at temper-
atures ranging from 298 to 21 K to elucidate the effects of
temperature and pH on the photoreaction mechanism of
amino acids by using unpolarized UV light (214 nm) from a
Y-line lamp and circularly polarized synchrotron radiation
(215 nm) from a polarizing undulator[18a] installed in the
electron storage ring UVSOR-II.[18b]
It is crucial that the Kuhn–Condon zero-sum rule[13] does
not apply to the CPL photolysis of aliphatic amino acids in
water/ice media simply because the media absorb the VUV
CPL (<200 nm) and only the UV CPL of 200–250 nm is ab-
sorbed by amino acids.[12]
In 1977, Bonner and co-workers reported that CPL pho-
tolysis of rac-Leu in 0.1m HCl at 212.8 nm led to 2–2.5%
enantiomeric excess (ee) in the remaining Leu at 59–75%
conversion, although no photodegradation mechanism was
proposed.[14]
In our recent study,[12] rac-Leu and rac-alanine (Ala) were
photolyzed with CPL at 215 nm in acidic and neutral aque-
ous solutions at room temperature to give enantiomerically
enriched Leu and Ala, irrespective of the solution pH.
However, the mechanism was significantly pH-dependent,
switching from Norrish-type II g-hydrogen abstraction in
acidic solutions to the less efficient deamination to an a-lac-
tone by intramolecular nucleophilic substitution (SNi) in neu-
tral solutions,[12,15] as a result of the change in the excited
chemical species from carboxylic acid to carboxylate; for
the detailed mechanisms in acidic and neutral solutions, see
Scheme 1.
Meierhenrich et al. performed the CPL photolysis of a
thin film of pure rac-Leu at 182 nm to obtain enantiomeri-
cally enriched l- and d-Leu in 0.88ꢂ0.28 and 2.60ꢂ0.16%
ee, respectively, upon l- and r-CPL irradiation, presumably
at ambient temperatures.[16a] More recently, the same group
reported the results of the CPL photolysis of a 2:1:1 mixture
of H2O, CH3OH, and NH3.[16b] The mixture was first irradiat-
ed at 80 K by CPL at 187 nm to produce radical species
trapped in the low-temperature matrix, which was allowed
to thaw and react to afford a mixture of stable products con-
taining amino acids. The resulting mixture was subjected to
a second CPL irradiation at room temperature to enantiose-
lectively photodecompose the racemic amino acids produced
in the first irradiation; l- and d-enriched Ala were obtained
in 0.71ꢂ0.30 and 1.34ꢂ0.40% ee upon l- and r-CPL irradia-
tion at 187 nm, respectively.[16b] Takano et al. reported that
an aqueous extract of high-molecular-weight complex organ-
ics (of up to 3000 Da), prepared by irradiating a gaseous
mixture of CO, NH3, and H2O with 3 MeV protons, gave l-
and d-Ala in 0.65ꢂ0.23 and 0.44ꢂ0.31% ee upon l- and r-
CPL irradiation at >200 nm and subsequent hydrolysis by
acid.[17]
Results and Discussion
Temperature- and pH-dependent anisotropy factor: In the
CPL photolysis experiment, the anisotropy factor (g=De/e)
of the chiral substrate plays a decisive role in determining
the efficiency of photochemical enantiomer enrichment.
Hence we first examined the chiroptical properties of l-Leu
in acidic and neutral aqueous solutions at 268–318 K.
Aliphatic l-amino acids are known to show weak n,p*
transitions (e=45–70mꢀ1 cmꢀ1) with a positive Cotton effect
(De=1–2mꢀ1 cmꢀ1) at around 210 nm and an extremely
weak negative Cotton effect (jDej ꢁ0.001mꢀ1 cmꢀ1) at 250–
255 nm in aqueous solutions at pH 1.[12,19] This is exactly the
case with l-Leu. As shown in Figure 1, the anisotropy spec-
trum (Figure 1c) is broader than the UV and CD spectra
(Figure 1a,b), affording a relatively large positive g factor of
around 0.03 at 210–220 nm and a much smaller negative g
factor of around ꢀ0.002 at 255 nm.
CD spectral analysis and product study: The Leu samples
photolyzed by circularly polarized and unpolarized light in
acidic and neutral media at various temperatures were sub-
jected to CD spectral analysis and then to qualitative and
quantitative product analyses.[16] The CD spectral changes
caused by CPL irradiation at 21 and 81 K are shown in
Figure 2 and Figure S3 (in the Supporting Information), re-
spectively, and the nonvolatile and volatile products (see
Figures S4–S16) and their yields obtained upon photolysis in
the temperature range 21–298 K are summarized in Tables 1
and 2, respectively.
Because the g factor is dimensionless and hence inde-
pendent of the volume change caused by varying the solu-
tion temperature, it is convenient to examine precisely the
effects of temperature on the chiroptical properties. On
cooling the acidic Leu solution, the g factor was appreciably
enhanced at 210–220 nm at the expense of that at 255 nm
(Figure 1c), which has been attributed to the shift in the
equilibrium of amino acid conformers and hence is reversi-
ble in nature.19b Upon further cooling, the aqueous Leu solu-
tion solidified to give an ice matrix, which was glassy but
not completely transparent and hence slightly scattered inci-
dent light. Although the frozen Leu samples were not suita-
ble for spectral analysis, it is likely that the conformer com-
However, none of our experiments,[12] nor those of Meier-
henrich[16] or Takano,[17] has directly proven the occurrence
of the enantioselective decomposition of racemic amino
acids or identified the products derived therefrom under
conditions similar to interstellar ice (10–20 K).
In this study to elucidate the feasibility and the detailed
mechanism of the photoreaction of amino acids in ice matri-
ces and hopefully to obtain some insight into the cosmic sce-
nario, we chose rac-Leu as a representative aliphatic amino
acid substrate for the following reasons: (1) Enantiomerical-
13930
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13929 – 13936

Photoreaction of rac-Leucine
FULL PAPER
Figure 2. CD spectra of rac-Leu solutions recorded at 296 K before and
after CPL irradiation at 215 nm in an ice matrix at 21 K. (a) Neutral sam-
ples ([rac-Leu]=7.0 mm, pH298K =7), in which the conversions upon irra-
diation of l- and r-CPL (dose: 860 mAh) were 21 and 17%, respectively.
(b) Acidic samples ([rac-Leu]=30.5 mm, pH298K =1), in which the conver-
sions upon irradiation of l- and r-CPL (dose: 400 mAh) were 6 and 7%,
respectively.
Figure 1. UV (top), CD (middle), and anisotropy spectra (bottom) of
acidic (left) and neutral (right) aqueous solutions of l-Leu (10.6 mm for
the acidic solution and 106 mm for the neutral solution) at 318–268 K.
The UV and CD spectra at 318 and 293 K are corrected for the volume
changes caused by temperature variation, whereas the anisotropy factor
is inherently free from the errors arising from volume change as g=De/e.
consistent in shape with those of d- and l-Leu, respectively,
which indicates preferential photodecomposition of the anti-
pode (l- and d-Leu), as anticipated by the CD sign at
215 nm (Figure 1d). Naturally, no appreciable CD signals de-
veloped upon irradiation with linearly polarized light (LPL).
The optical purity (which is normally equivalent to the ee
but determined from the optical rotation) of the enantio-
merically enriched Leu was determined from the induced el-
lipticity to be ꢀ0.3 and +0.3% (at 17 and 21% conversion)
for l- and r-CPL, respectively. The acidic rac-Leu samples ir-
radiated at 21 K with l- and r-CPL also exhibit mirror-image
CD spectra with extrema at 200–210 nm (Figure 2b). From
the ellipticity induced, the optical purity was determined to
be ꢀ0.1 and +0.1% (at 6 and 7% conversion) for l- and r-
CPL irradiation, respectively.
position, and therefore the g factor, near the freezing point
(268 K) are preserved or at least not significantly altered in
the ice matrix.
In a neutral solution at pH 7, zwitterionic l-Leu shows
only a structureless absorption tail at 200–250 nm in both
the UV and CD spectra (Figure 1d,e), whereas the anisotro-
py spectrum (Figure 1f) shows a broad positive peak at 210–
230 nm that is greatly enhanced up to 0.02 on cooling the
solution to near the freezing temperature, that is, 273.5 K.
From a technical point of view, it was convenient that the
anisotropy spectra are relatively flat and the maximal values
of 0.02–0.03 are reached in the same wavelength region for
both the acidic and neutral solutions. This allowed us to ex-
ploit the optimal conditions for CPL photolysis without
fine-tuning the irradiation wavelength (which is achieved
only by altering the energy of electron acceleration in the
synchrotron).
Upon irradiation of Leu in an acidic aqueous solution at
298 K, Gly was produced as the major product in 47% yield
(based on the consumed Leu), together with a small amount
of ammonia (13% yield). In contrast, the photolysis of a
neutral solution of Leu afforded a-hydroxyisocaproic acid
(38% yield) and ammonia (73% yield) as the major prod-
Figure 2a illustrates the CD spectra of the neutral Leu
samples irradiated at 21 K with l- and r-CPL; the spectra are
Chem. Eur. J. 2013, 19, 13929 – 13936
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13931

Y. Inoue et al.
Table 1. Nonvolatile products detected in the rac-Leu samples irradiated at 214 nm with a Y-line lamp (YL) or at 215 nm with left-circularly polarized
synchrotron radiation (l-CPL) in acidic and neutral aqueous solutions and ice matrices at various temperatures.[a]
T
[K]
Light
source
pH298K
Irradiation
Time Dose
[h]
[mAh][b]
Leu
Absorbance[d]
Conversion
[mm (%)]
Product yield [mm (%)][e]
[mm][c]
G
Gly
NH3
iBuCH2NH2
iBuCH(OH)CO2H
AHCTUNGTRENNUNG
298
YL
1
7
1
1
7
1
7
1
7
1
7
1
1.5
5
1.5
4
5
4
5
4
5
28.1
8.12
28.1
29.4
6.75
29.4
6.75
29.4
6.75
30.5
7.08
29.4
29.4
6.75
30.5
6.95
0.24[e]
0.05[e]
0.24
0.28
0.05
0.28
0.05
0.28
0.05
0.29
0.05
0.28
0.28
0.05
0.29
0.05
11.8 (41.9)
1.24 (14.8)
10.6 (37.8)
7.48 (25.5)
0.84 (12.4)
4.54 (15.5)
0.78 (11.6)
2.76 (9.4)
0.77 (11.4)
1.75 (5.7)
1.41 (19.9)
0.72 (2.5)
1.54 (5.2)
0.89 (13.2)
1.75 (5.7)
1.30 (18.7)
5.48 (46.5)
1.5 (13)
0.9 (73)
0.9 (8)
0.1 (1)
0.1 (8)
0.1 (1)
<0.1
<0.1
<0.1
0.1 (13)
<0.1
0.1 (13)
0.1 (6)
0.1 (7)
<0.1
<0.1
0.47 (38)
0
268
238
YL
YL
3.86 (36.3)
1.11 (14.8)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1.8 (24)
0.4 (47)
0.9 (20)
0.3 (38)
0.5 (18)
0.3 (39)
0.6 (34)
0.6 (43)
0.2 (28)
0.5 (32)
0.2 (22)
0.6 (34)
0.5 (38)
0
183
123
81
YL
0.82 (18.1)
0
YL
0.19 (6.9)
0
l-CPL
YL
200
860
0.14 (8.0)
0
0
0
0
0
0
77
4
12
10
<0.1
<0.1
0.1 (6)
0.1 (8)
7
1
7
21
l-CPL
400
860
[a] Irradiated at 298–123 K in a Unisoku USP-203 cryostat, at 81 K in a continuous-flow liquid N2 cryostat, at 77 K in a Suprasil Dewar vessel filled with
liquid N2, and at 21 K in a continuous-flow liquid helium cryostat. [b] Light intensity defined as the accumulated electron beam current of the electron
storage ring UVSOR-II. [c] Lower concentrations were employed for neutral samples, due to the lower solubility of zwitterionic Leu. [d] Absorbance cal-
culated from the concentration and the effective pathlength (1.8 mm), unless noted otherwise. [e] Product yield based on the consumed Leu. In all irradi-
ated samples, trace amounts of CH3CO2H and HCO2H were detected. Effective pathlength: 1.5 mm.
Table 2. Gaseous and ether-extracted products detected upon gas chromatographic analyses of the rac-Leu samples photolyzed at 215 nm with left-circu-
larly polarized synchrotron radiation (l-CPL) in acidic and neutral ice matrices at 81 and 21 K.[a]
T
[K]
pH298K
Dose
[mAh][b]
Conc.
[mm]
Absor
G
Conversion
[mm (%)]
Product yield [mm (%)][d]
G
bance[c]
N
CO
CO2
iBuCHO
CH3CH=CH2
ACHTNUGRTNE(NUGN CH3)2C=CH2
[e]
[e]
[e]
[e]
81
1
7
1
7
200
860
400
860
30.47
6.95
30.47
6.95
0.28
0.05
0.28
0.05
1.75 (6.4)
1.23 (17.7)
2.05 (6.8)
1.38 (19.9)
–
–
–
–
0.07 (4)
0.09 (7)
0.12 (6)
0.16 (12)
–
–
–
–
[e]
[e]
[e]
[e]
21
>0.04 (>2)
>0.04 (>3)
>0.07 (>3)
>0.12 (>9)
>0.01 (>0.1)
>0.01 (>0.1)
<0.01
<0.01
[a] Irradiated in a continuous-flow liquid N2 (81 K) or He (21 K) cryostat. [b] Light intensity defined as the accumulated electron beam current of
UVSOR-II. [c] Absorbance calculated from the concentration and the effective pathlength (1.8 mm). [d] Product yield based on the consumed Leu.
[e] Yield not determined.
ucts. The rate of photodecomposition, or conversion, in gen-
eral, was reduced by lowering the irradiation temperature,
but the effect was more significant for the acidic samples
(Table 1). As shown in Figure 3a, a dramatic change was ob-
served for the Gly yield, which suddenly dropped at the
freezing point (indicated by the vertical broken line in Fig-
ure 3a) and gradually faded out upon further cooling, finally
disappearing at 77 K. The sudden drop in Gly yield at the
freezing point may be ascribed to the reduction of confor-
mational freedom upon freezing, which hinders the access of
the excited carbonyl oxygen to the g-hydrogen and subse-
quent abstraction (Scheme 1). It is interesting, however, that
some g-hydrogen abstraction still occurs even in ice matri-
ces, at least at temperatures higher than 77 K, for which two
independent mechanisms (one static and one dynamic) may
be responsible. It is reasonable to assume that (1) part of
the Leu molecules in the ice matrix are frozen in conforma-
tion(s) that are suitable for g-hydrogen abstraction and
(2) there still remains some partial mobility available for the
excited carbonyl moiety to approach and abstract the g-hy-
drogen atom. It is likely that both of these mechanisms may
operate in ice matrices at higher temperatures but become
structurally and/or energetically unfeasible at temperatures
below 77 K.
In contrast, the ammonia yield is much less sensitive to
temperature and in fact increases at lower temperatures at
the expense of Gly (Figure 3a). This indicates that the SNi
attack of the excited carboxylate moiety on the a-carbon
atom, leading to deamination, is conformationally more tol-
erant and energetically more feasible.
Leu in neutral ice matrices shows very different photobe-
havior. As shown in Figure 3b, the conversion and the am-
monia yield only moderately decrease on lowering the tem-
perature from 298 (solution) to 238 K (ice), whereas the for-
mation of a-hydroxyisocaproic acid, the main product of the
solution-phase irradiation, is completely suppressed in ice,
even at 238 K. This result is not unexpected because the
attack of water on the intermediate a-lactone, leading to a-
hydroxyisocaproic acid, is not feasible in ice matrices, as dis-
cussed below (Scheme 1).[12,19a,20a]
13932
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13929 – 13936

Photoreaction of rac-Leucine
FULL PAPER
in 4–6% yield for the acidic sample (Table 2). Crucially, the
formation of isovaleraldehyde was not proportional to the
period of irradiation, but showed an induction period, which
is in sharp contrast to the linear dependence of the ammo-
nia yield (Figure 4). This indicates that isovaleraldehyde is a
~
*
Figure 4. Yields of ammonia ( ) and isovaleraldehyde (iBuCHO, ) as a
function of the photodecomposition of neutral Leu (6.63–6.95 mm) upon
irradiation by l-CPL of varying doses (100–860 mAh) at 21 K.
*
*
Figure 3. Chemical yields of ammonia ( ) and Gly ( ) obtained upon ir-
radiation of (a) acidic and (b) neutral rac-Leu samples in aqueous solu-
tions at 268–298 K and in ice matrices at 21–238 K. The vertical broken
line indicates the freezing point of each solution.
product derived from a secondary photoreaction of the ini-
tial product. We propose isobutyl-a-lactone as the most
plausible intermediate for the formation of isovaleraldehyde
upon irradiation in ice; this lactone, which is derived from
the photochemical SNi deamination of Leu, is highly reactive
and immediately attacked by water to give a-hydroxyisocap-
roic acid in aqueous solutions.[12,20] In low-temperature ice
matrices, the a-lactone produced is likely to be trapped
intact and suffer secondary photodecomposition to isovaler-
Gas chromatographic (GC) analysis revealed that the gas-
eous products collected from the irradiated neutral and
acidic Leu samples contained CO in >2–3% yield and CO2
in >3–9% yield, whereas the diethyl ether extract contained
isovaleraldehyde in 7–12% yield for the neutral sample and
Scheme 1. Photochemical reactions of rac-Leu in acidic and neutral aqueous solutions and ice matrices.
Chem. Eur. J. 2013, 19, 13929 – 13936
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13933

Y. Inoue et al.
aldehyde and CO (Scheme 1, bottom row). Experimentally,
the detection of propene, a photodecomposition product
from isovaleraldehyde, in the photolyzed solution (Table 2)
may support this mechanism (Scheme 1, middle right).
On the other hand, we are less certain about the potential
roles of different ice phases in controlling the photobehavior
as we cooled the sample solutions at a constant rate to the
designated temperatures and no literature information is
available on the phases of ice matrices containing 7–30 mm
rac-Leu (and HCl). However, the experimental results ob-
tained in this study (i.e., no clear jump is observed in con-
version or product yield at the phase transition tempera-
tures) suggest that the photobehavior is not significantly af-
fected by the phase transitions from metastable phase I
cubic (Ic) to phase I hexagonal (Ih) and then to phase XI ob-
served upon cooling pure ice,[22] or more likely the ice matri-
ces containing Leu and/or HCl are amorphous in nature and
hence experience no clear phase transition.
From the point of view of the chemical evolution, it is cru-
cial that not only the CPL photoreaction of rac-amino acids
in ice has been experimentally proven to proceed under in-
terstellar conditions, but also that the major photoreaction
mechanisms and products have been elucidated. These re-
sults substantiate the cosmic scenario in which interstellar
ice analogues containing simple starting materials irradiated
by VUV light produce a mixture of racemic amino acids[10]
that subsequently photodecompose enantioselectively by
CPL radiated from the star-formation region en route to
primitive Earth.
Conclusion
In this study, we have identified the major volatile and non-
volatile products obtained upon direct irradiation of Leu in
aqueous solutions and also in ice matrices. Based on the re-
sults of the product study, we have discussed the photo-
chemical behavior of Leu (as a representative aliphatic
amino acid) in acidic and neutral media at temperatures
ranging from 298 to 21 K to reveal that the photoreaction
mechanism and the major products derived therefrom are
critical functions of the phase, temperature, and pH of the
media.
In acidic media, the overall photoreaction is greatly decel-
erated by lowering the irradiation temperature, particularly
below the freezing point, and the main photoreaction mech-
anism switches from the Norrish-type II g-hydrogen abstrac-
tion, which affords Gly and isobutene, to SNi deamination to
give isobutyl-a-lactone, which is trapped in situ and subse-
quently photodecarbonylated to isovaleraldehyde in ice ma-
trices.
However, the enantiomeric excesses obtained in the CPL
photolysis of aliphatic amino acids are generally low (0.1–
2.6% ee),[12,16a,b,17] including the present case. It has been
shown that the enantioselectivity obtained by CPL photode-
composition of a racemic mixture is highly conversion-de-
pendent and can be enhanced to 90% ee and even to 99%
ee at 88 and 96% conversion, respectively, if the g factor is
1.0.[11] In reality, the g factors of aliphatic amino acids, in-
cluding Leu (Figure 1), are typically 0.02 or slightly higher
at the irradiation wavelength (215 nm).[23] This means that
an ee of 4.6% will be achieved at 99% conversion and fur-
thermore that the last single molecule will have statistically
48% ee on average even when one mole of an amino acid
with g=0.02 is subjected to CPL photolysis.[11] Hence, the
unexpectedly high ee values of up to 50–60% reported re-
cently by Pizzarello and co-workers[3] for nonterrestrial iso-
leucine and allo-isoleucine extracted from the Murchison
and other meteorites, and also by Glavin et al.[3m] for nonter-
restrial proteinogenic aspartic acid and glutamic acid from
two out of the three examined fragments of the Tagish Lake
meteorite, are evidently unexplainable by the CPL photode-
composition mechanism alone and indicate the operation of
other mechanisms such as preferential crystallization to con-
glomerates[24] during the aqueous alteration in a meteorite
parent body.[3m] Nevertheless, it is also worth noting that es-
sentially all of the proteinogenic amino acids found in the
Tagish Lake meteorite are l-enriched,[3m] which suggests a
certain seeding or biasing mechanism to trigger the enantio-
meric imbalance in which the CPL photodecomposition
would have played some role.
In neutral media, the photoreaction proceeds consistently
by the SNi deamination mechanism to give the a-lactone,
which is followed by the nucleophilic attack of water leading
to a-hydroxyisocaproic acid in aqueous solutions at ambient
temperatures or alternatively by photodecarbonylation to
isovaleraldehyde in ice matrices at cryogenic temperatures.
The overall efficiency of the photoreaction is surprisingly in-
sensitive to the irradiation temperature (and medium),
maintaining the same level of conversion throughout the
temperature range examined (21–298 K). This suggests that
the SNi deamination survives as the major photoreaction
process under interstellar conditions (10–20 K).
Intriguingly, the photoreaction mechanism converges to
the deamination process in both acidic and neutral media at
temperatures below 77 K. This indicates that the amino acid
exists in its zwitterionic form in ice matrices at lower tem-
peratures, irrespective of the original pH at 298 K, because
the deamination occurs only upon excitation of the carboxy-
late moiety of a zwitterionic amino acid.[12,19a,20a] In this con-
nection, it is crucial to point out that the dissociation of HCl
adsorbed on an ice surface is discouraged at low temperatur-
es[21a,b] and HCl on amorphous ice at 70 K is totally molecu-
lar[21a.c] whereas around 85% of HCl is molecular in ice
nanocrystals at 60 K.[21d] This means that HCl becomes more
molecular, or less acidic, at lower temperatures and that it is
difficult to protonate zwitterionic amino acid in ice matrices.
This shift in equilibrium rationalizes the switching of the
major photoreaction mechanism from the g-hydrogen ab-
straction by an excited carboxylic acid to the SNi deamina-
tion by an excited carboxylate (Scheme 1).
13934
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13929 – 13936

Photoreaction of rac-Leucine
FULL PAPER
CD spectral analysis of the CPL-irradiated samples was carried out to de-
termine their optical purity (op) at 296 K in a cylindrical cell (13 mm
i.d.ꢁ2 mm thickness) on a JASCO J-725 spectropolarimeter with a modi-
fied optical system for higher light intensity and sensitivity. The optical
Experimental Section
Materials: rac-Leu (ꢃ99% pure), d-Leu (99% pure, 97% ee), and l-Leu
(ꢃ98% pure, 99% ee) were purchased from Wako and used without fur-
ther purification.
purities were calculated by using the equation op=q215/ACHUNTGRENU(NG 33De215ACHTUNGTRENNUNG[Leu]), in
which q215 is the ellipticity at 215 nm of a CPL-irradiated sample, De215 is
the molar CD of enantiopure Leu at the same wavelength, and [Leu] the
concentration of Leu remaining in the photolyzed solution.[12]
Sample preparation: Acidic Leu solutions (28.1–30.5 mm) were prepared
by dissolving rac-, d-, or l-Leu in an aqueous 0.1m HCl solution
(pH298K 1.03) for photoirradiation or in an aqueous 0.5m HCl solution for
UV and CD spectral measurements, whereas neutral Leu solutions (6.8–
8.1 mm) at pH298K 6.78 were prepared by dissolving rac-, d- and l-Leu in
distilled water. The acidic and neutral Leu solutions, with a thickness of
1.5–1.8 mm, were relatively transparent at 215 nm, showing absorbances
of 0.24–0.29 and approximately 0.05, respectively. Immediately after their
preparation, each Leu solution of 0.7 mL volume was placed in a regular
(10ꢁ10ꢁ40 mm) or thin (2ꢁ10ꢁ40 mm) quartz cell, degassed by three
freeze–pump–thaw cycles, sealed under high vacuum (<10ꢀ5 Pa), and
then subjected to spectral examination and photoirradiation at ambient
and lower temperatures.
Acknowledgements
This work was supported by the Joint Studies Program of the Institute
for Molecular Science and the Japan Society for the Promotion of Sci-
ence through
a Grant-in-Aid for Scientific Research (A, grant no.
21245011) both of which are gratefully acknowledged. We thank the
members of the UVSOR facility, Institute for Molecular Science, in par-
ticular Mr. Eiken Nakamura and Mr. Takayuki Murakami, for the opera-
tion of the linear accelerator and the continuous flow liquid He/N2 cryo-
stat. We also thank Ushio Inc. for the use of a Y-line lamp.
Spectroscopy: UV and CD spectra of the acidic and neutral l-Leu solu-
tions prepared above were recorded at 268–318 K on a JASCO V-550
spectrophotometer fitted with an HMC-358 temperature controller and
on a J-720WI spectropolarimeter fitted with a PMH-354WI temperature
controller. Spectral analysis at temperatures lower than the freezing
point was not feasible or the result was not reproducible due to surface
roughness and scattering of the frozen samples.
[1] a) W. A. Bonner, Top. Stereochem. 1988, 18, 1; b) W. A. Bonner,
Origins Life Evol. Biosphere 1991, 21, 59; c) J. Alain, C. Corinne,
Origins Life Evol. Biosphere 2002, 32, 129; d) U. J.; Meierhenrich,
G. M. MuÇos Caro, W. A. Schutte, B. Barbier, A. A. Segovia, H.
Rosenbauer, W. H. P. Thiemann, A. Brack, European Space Agency
Special Publication 2002, SP-518, 25; e) C. Meinert, J. H. Bredeo-
hoft, J.-J. Filippi, Y. Baraud, L. Nahon, F. Wien, N. C. Jones, S. V.
Hoffmann, U. J. Meierhenrich, Angew. Chem. 2012, 124, 4562;
Angew. Chem. Int. Ed. 2012, 51, 4484.
[2] a) J. Podlech, Angew. Chem. 1999, 111, 501; Angew. Chem. Int. Ed.
1999, 38, 477; b) S. Pizzarello, Acc. Chem. Res. 2006, 39, 231; c) U. J.
Meierhenrich, J.-J. Filippi, C. Meinert, P. Vierling, J. P. Dworkin,
Angew. Chem. 2010, 122, 3826; Angew. Chem. Int. Ed. 2010, 49,
3738; d) J. E. Hein, D. G. Blackmond, Acc. Chem. Res. 2012, 45,
2045.
[3] a) M. H. Engel, B. Nagy, Nature 1982, 296, 837 (see also ref. [3f] for
further discussion); b) M. H. Engel, S. A. Macko, J. A. Silfer, Nature
1990, 348, 47; c) S. Pizzarello, J. R. Cronin, R. V. Krishnamurthy, S.
Epstein, Geochim. Cosmochim. Acta 1991, 55, 905; d) J. R. Cronin,
S. Pizzarello, Science 1997, 275, 951; e) M. H. Engel, S. A. Macko,
Nature 1997, 389, 265; f) S. Pizzarello, J. R. Cronin, Geochim. Cos-
mochim. Acta 2000, 64, 329; g) M. A. Sephton, Nat. Prod. Rep. 2002,
19, 292; h) S. Pizzarello, M. Zolensky, K. A. Turk, Geochim. Cosmo-
chim. Acta 2003, 67, 1589; i) S. Pizzarello, Y, Huang, M. R. Alex-
andrer, Proc. Natl. Acad. Sci. USA 2008, 105, 3700; j) D. P. Glavin,
J. P. Dworkin, Proc. Natl. Acad. Sci. USA 2009, 106, 5487; k) D. P.
Glavin, M. P. Callahan, J. P. Dworkin, J. E. Elsila, Meteorit. Planet.
Sci. 2010, 45, 1948; l) S. Pizzarello, D. L. Schrader, A. A. Monroe,
D. S. Lauretta, Proc. Natl. Acad. Sci. USA 2012, 109, 11949; m) D. P.
Glavin, J. E. Elsila, A. S. Burton, M. P. Callahan, J. P. Dworkin,
R. W. Hilts, C. D. K. Herd, Meteorit. Planet. Sci. 2012, 47, 1347.
[4] a) E. Rubenstein, W. A. Bonner, H. P. Noyes, G. S. Brown, Nature
1983, 306, 118; b) W. A. Bonner, E. Rubenstein, Biosystems 1987,
20, 99.
Photoirradiation: The Leu samples prepared above were irradiated at
214 nm by using unpolarized light from an Ushio Y-line lamp in solution
at 268–298 K and also in ice at 77–238 K.
The frozen samples for unpolarized light irradiation were prepared as fol-
lows. A quartz cell containing 0.7 mL of the Leu solution was gradually
cooled to 238 K by horizontally placing the cell in a cooling bath. Once
frozen, the icy sample was quickly transferred to a Unisoku USP-203
cryostat (with three optical windows for spectral examination and irradia-
tion) and further cooled to the desired temperature of ꢃ123 K at a rate
of 10 Kminꢀ1 or directly to 77 K by immersing the sample cell in liquid
N2, and then irradiated by a Y-line lamp.
For CPL irradiation, a quartz cell containing the Leu solution was fixed
on a brass sample holder, which was placed horizontally in a vacuum
chamber connected to the polarizing undulator beam-line and cooled to
21 or 81 K first at a rate of 2–5 Kminꢀ1 (298–220 K) and then at 10–
20 Kminꢀ1 (<220 K) in a continuous-flow helium or nitrogen cryostat
with temperature monitored by a thermocouple attached to the cell sur-
face. The quartz cell containing a frozen Leu sample of around 1.5–
1.8 mm thickness was rotated to the upright position and irradiated at
21ꢂ3 or 81ꢂ2 K with CPL at 215 nm generated by the polarizing undu-
lator installed in UVSOR-II at the Institute for Molecular Science (see
Figures S1 and S2 in the Supporting Information).[18]
Product analyses: The irradiated samples were warmed to ambient tem-
perature in a plastic bag with a septum cap. The volatile products were
sampled by using a gas-tight syringe, and an aliquot of the aqueous solu-
tion was extracted with diethyl ether, both of which were analyzed by gas
chromatography (GC). The rest of the solution was subjected to qualita-
tive and quantitative analyses as described below.[12]
The gaseous inorganic and organic products collected in a gas-tight sy-
ringe were analyzed by GC on a Shimadzu GC-14A instrument (with a
TCD detector) fitted with a column (2.1 mꢁ3 mm i.d.) of molecular
sieves (5 ꢂ, 60/80 mesh). The ether extract was subjected to GC-MS anal-
ysis on an HP-6890GC/5973MSD instrument fitted with a CP-Pora-
BOND Q column (25 mꢁ0.25 mm i.d.) to characterize and quantify the
organic products. The yields obtained were corrected for the distribution
coefficients of the analytes between the ether (2 mL) and Leu solution
(0.2 mL), which were determined on a Shimadzu GC-14A with a J&W
DB wax column (30 mꢁ0.25 mm i.d.). Nonvolatile products in the photo-
lyzed Leu samples were characterized and quantified by using the amino
acid analysis, ion chromatography, and capillary electrophoresis, as re-
ported previously.[12]
[5] J. A. Roberts, Nature 1984, 308, 318.
[6] a) J. Bailey, A. Chrysostomou, J. H. Hough, T. M. Gledhill, A.
MaCall, S. Clark, F. Mꢃnard, Science 1998, 281, 672; b) M. Buscher-
mçhle, D. C. B. Whittet, A. Chrysostomou, J. H. P. Hough, W. Lucas,
A. J. Adamson, B. A. Whitney, M. J. Wolff, Astrophys. J. 2005, 624,
821.
[7] a) K. Kobayashi, T. Kasamatsu, T. Kaneko, J. Koike, T. Oshima, T.
Saito, T. Yamamoto, H. Yanagawa, Adv. Space Res. 1995, 16, 21;
b) B. T. Draine, Astrophys. J. 2003, 598, 1017.
[8] J. Orꢄ, Nature 1961, 190, 389.
Chem. Eur. J. 2013, 19, 13929 – 13936
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13935

Y. Inoue et al.
[9] U. Meierhenrich, B. Barbier, R. Jacquet, A. Chabin, C. Alcaraz, L.
Nahon, A. Brack, European Space Agency Special Publication 2001,
SP-496, 167.
[10] a) G. M. MuÇoz Caro, U. J. Meierhenrich, W. A. Schutte, B. Barbler,
A. A. Segovia, H. Rosenbauer, W. H. P. Thiemann, A. Brack, J. M.
Greenberg, Nature 2002, 416, 403; b) M. P. Bernstein, J. P. Dworkin,
S. A. Sandford, G. W. Cooper, L. J. Allamandola, Nature 2002, 416,
401; c) C. Meinert, J.-J. Filippi, P. Marcellus, L. L. S. Hendecourt,
U. J. Meierhenrich, ChemPlusChem 2012, 77, 186.
[11] a) W. Kuhn, E. Knopf, Z. Phys. Chem., Abt. B 1930, 7, 292; b) G.
Balavoine, A. Moradpour, H. B. Kagan, J. Am. Chem. Soc. 1974, 96,
5152; c) H. Rau, Chem. Rev. 1983, 83, 535; d) H. Nishino, A. Naka-
mura, Y. Inoue, J. Chem. Soc., Perkin Trans. 2 2001, 1693.
[12] H. Nishino, A. Kosaka, G. A. Hembury, F. Aoki, K. Miyauchi, H.
Shitomi, H. Onuki, Y. Inoue, J. Am. Chem. Soc. 2002, 124, 11618.
[13] S. F. Mason, Nature 1997, 389, 804.
[18] a) S. Kimura, M. Kimada, H. Hama, X. Marꢅchal, T. Tanaka, H. Ki-
tamura, J. Electron Spectrosc. Relat. Phenom. 1996, 80, 437; b) M.
Katoh, K. Hayashi, T. Honda, Y. Hori, M. Hosaka, T. Kinoshita, S.
Kouda, Y. Takashima, J. Yamazaki, Nucl. Instrum. Methods Phys.
Res. Sect. A 2001, 467/468, 68.
[19] a) I. Mandl, A. D. McLaren, J. Am. Chem. Soc. 1951, 73, 1826; b) R.
Hꢆkansson in Chemistry of Acid Derivatives, Vol. 2 (Ed. S. Patai),
Wiley, Chichester, 1992, Pt. 1.
[20] a) O. L. Chapman, P. W. Wojtkowski, W. Adam, O. Rodriguez, R.
Rucktꢇschel, J. Am. Chem. Soc. 1972, 94, 1365; b) O. L. Chapman,
Pure Appl. Chem. 1974, 40, 511.
[21] a) L. Delzeit, B. Rowland, J. P. Devlin, J. Phys. Chem. 1993, 97,
10312; b) N. Uras, M. Rahman, J. P. Devlin, J. Phys. Chem. B 1998,
102, 9375; c) H. Kang, T. H. Shin, S. C. Park, I. K. Kim, S. J. Han, J.
Am. Chem. Soc. 2000, 122, 9842; d) J. P. Devlin, N. Uras, J. Sadlej,
V. Buch, Nature 2002, 417, 269.
[14] J. J. Flores, W. A. Bonner, G. A. Massey, J. Am. Chem. Soc. 1977, 99,
3622.
[22] A. Leadbetter, R. C. Ward, J. W. Clark, P. A. Tucker, T. Matsuo, H.
Suga, J. Chem. Phys. 1985, 82, 424.
[15] A. D. Mclaren, D. Shugar, Photochemistry of Proteins and Nucleic
Acids, Pergamon Press, Oxford, 1964.
[23] H. Nishino, A. Kosaka, G. A. Hembury, K. Matsushima, Y. Inoue, J.
Chem. Soc., Perkin Trans. 2 2002, 582.
[16] a) U. J. Meierhenrich, L. Nahon, C. Alcaraz, J. H. Bredehçft, S. V.
Hoffmann, B. Barbier, A. Brack, Angew. Chem. 2005, 117, 5774;
Angew. Chem. Int. Ed. 2005, 44, 5630; b) P. de Marcellus, C. Mei-
nert, M. Nuevo, J.-J. Filippi, G. Danger, D. Deboffle, L. Nahon, L.
Le Sergeamt d’Hendecourt, U. J. Meierhenrich, Astrophys. J. 2011,
727, L27.
[24] a) C. Viedma, Origins Life Evol. Biosphere 2001, 31, 501; b) A. Al-
varez Rodrigo, H. Lorenz, A. Seidel-Morgenstern, Chirality 2004,
16, 499; c) C. Viedma, J. E. Ortiz, T. de Torres, T. Izumi, D. Black-
mond, J. Am. Chem. Soc. 2008, 130, 15274.
Received: May 13, 2013
Revised: July 7, 2013
Published online: August 28, 2013
[17] Y. Takano, J. Takahashi, T. Kaneko, K. Marumo, K. Kobayashi,
Earth Planet. Sci. Lett. 2007, 254, 106.
13936
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 13929 – 13936