Reaction of amino acids in a supercritical water-flow reactor simulating submarine hydrothermal systems

Article abstract of DOI:10.1246/bcsj.76.1171

A novel supercritical water flow-reactor was constructed in order to simulate submarine hydrothermal systems. The temperature of fluid inside the reaction tube could be monitored with thermocouples, which was proved to be different from the temperature outside the reaction tube. Oligomers of glycine up to tetraglycine were formed when a 100 mM glycine solution was heated at 200-350 °C for 2 minutes. None of glycine peptides were produced at 400 °C. It was suggested, however, that the formation of glycine condensates at higher temperature, including supercritical conditions of water. The stability of some amino acids under hydrothermal conditions was examined. ω-Amino acids and glutamic acid, which can form intramolecular condensates, showed higher stability than other α-amino acids at higher temperature, including supercritical conditions.

Full text of DOI:10.1246/bcsj.76.1171

Ó 2003 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 76, 1171–1178 (2003) 1171
Reaction of Amino Acids in a Supercritical Water-Flow Reactor
Simulating Submarine Hydrothermal Systems
ꢀ
Md. Nazrul Islam, Takeo Kaneko, and Kensei Kobayashi
Department of Chemistry and Biotechnology, Yokohama National University,
79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501
(Received October 28, 2002)
A novel supercritical water flow-reactor was constructed in order to simulate submarine hydrothermal systems.
The temperature of fluid inside the reaction tube could be monitored with thermocouples, which was proved to be dif-
ferent from the temperature outside the reaction tube. Oligomers of glycine up to tetraglycine were formed whenꢁa 100
ꢁ
mM glycine solution was heated at 200–350 C for 2 minutes. None of glycine peptides were produced at 400 C. It
was suggested, however, that the formation of glycine condensates at higher temperature, including supercritical condi-
tions of water. The stability of some amino acids under hydrothermal conditions was examined. !-Amino acids and
glutamic acid, which can form intramolecular condensates, showed higher stability than other ꢀ-amino acids at higher
temperature, including supercritical conditions.
Amino acids together with nucleic acid bases and sugars are
the building blocks for the present life system on earth. The
formation processes of those components are among the prime
questions about life. Oparin,1 and then Urey,2 proposed that
the primitive atmosphere consisted of reducing materials.
Their hypothesis was first examined by the historic experimen-
tal work of Miller,3;4 who demonstrated that amino acids and
many other organic compounds could be readily produced
by spark discharges under highly reduced conditions. Other
investigators showed that heat,5;6 UV radiation,7;8 shock
waves9 and laser-induced plasma energy10 could be used to
synthesize bioorganic compounds from a reducing starting
mixture.
Rubey11 challenged the Oparin–Urey model, which suggest-
ed that the early atmosphere was less reducing, but had carbon
dioxide and carbon monoxide as major carbon compounds.
Rubey’s model was also supported by Levin,12 Kasting,13
and Matsui.14 It has been reported that the formation of organ-
ic compounds from non or mildly-reducing atmospheres is
more difficult than from highly reducing ones. Only cosmic
rays could form bioorganic compounds from a very mildly re-
ducing atmosphere effectively.15
ture zone of SHSs from H2, CH4, NH3, H2S, CO, HCN etc.17
A number of both theoretical20{25 and experimental26{33 stud-
ies have been performed using closed systems (autoclaves) to
simulate the SHSs.
Recently, several experiments have been performed to sim-
ulate reactions in SHSs by using flow reactors, since SHSs are
not closed systems, as in autoclaves, but open flow systems.
Imai et al.19 examined the reactions of glycine in flow reaction
systems. They reꢁported peptide formation up to the glycine tri-
mer at 200–250 C without adding metal ions and control of
the pH, where an electric heater was used and the reaction tem-
perature was controlled solely by the temperature at a heater.
Submarine hydrothermal systems might consist supercritical
conditions at some points. Thus, it is needed to examine the
reactions at and over supercritical conditions. Ikushima et
al.34 reported on the formation oligoglycines at supercritical
conditions of water. They also used an electric heater; the re-
action temperature was controlled in the same way as in the
above-mentioned experiment. Preheaters were used in both
cases mentioned above. Alargov et al.35 also reported on the
oligomerization of glycine, where the reaction temperature
ꢁ
was set at 250–400 C. In this experiment, water was also
In the late 1970’s ‘‘local reducing’’ environments, called
submarine hydrothermal systems (SHSs), were discovered.16
SHSs are dynamic vent systems, and are thought to be ‘‘life re-
actors,’’17 which have non-equilibrium reaction systems.18;19
These include gradients of the temperature-pressure, pH and
concentration of various chemical components with the pres-
ence of many mineral catalysts, like montmorillonite clays,
iron hydroxide oxide, manganates, sulfides and zeolites. Wa-
ter that circulates through the systems transports dꢁissolved sub-
stances from a high-temperature zone (350–400 C) to an in-
termediate zone and then to a low-temperature zone (ca. 0–2
ꢁC). It has been theoretically proposed that organic mono-
mers, such as amino acids, are produced in the high-tempera-
heated with an electric heater, and was then mixed with the re-
actants of room temperature. Although the temperature at the
injection point was monitored, the actual temperature during
the reactions was not monitored.
In order to examine the possible reactions of amino acids in
SHSs, we constructed a novel supercritical water flow reactor
(SCWFR). The SCWFR was equipped with an IR gold image
furnace, and the temperature of the fluid inside the reaction
tube could be monitored with thermocouples. In this study,
we examined the polymerization of glycine and the stability
of some amino acids at high temperatures, including supercri-
tical conditions of water.

1172 Bull. Chem. Soc. Jpn., 76, No. 6 (2003)
Hydrothermal Reactions of Amino Acids
Analytical Methods:
An Ion-pair-HPLC system (pump:
TOSOH DP 8020) was used to analyze peptides of glycine.
The column used was an Inertsil ODS-80A (4:6 ꢃ 250 mm; GL
Science), where an aqueous solution of 50 mM KH2PO4 and 7.2
mM C6H13SO3Na (pH 2.5) was used as a mobile phase in an iso-
cratic elution mode. The flow rate of the mobile phase was 0.5
dm3 minꢄ1 (1 M = 1 mol dmꢄ3). A UV-detector (TOSOH UV-
8020) was used, where the detection wavelength was 195 nm.
Amino acids were analyzed with a cation-exchange HPLC sys-
tem (pumps: Shimadzu LC-6A), which was equipped with a
Shim-pack ISC-07/S1504 (sodium type, 4.0 mm i.d. ꢃ 150
Fig. 1. Schematic Diagram of a Supercritical Water Flow
Reactor (SCWFR). Abbreviations— GC: gas cylinder,
R: reservoir, P: HPLC pump, IV: injection valve, IRGF:
infrared gold image furnace, RT: reaction tube, TC: tem-
perature controller, TD: temperature display, tc1 and tc2
thermocouples, CB: cold bath (0 C), CJ: cooling jacket,
UT: outlet tube, BPR: back pressure regulator, ST: sam-
pling tube.
ꢁ
mm). The temperature of the column was maintained at 55 C.
Gradient elution was performed by using the following eluents:
A) 0.07 M sodium citrate-perchloric acid (pH 3.20) with 7% eth-
anol, and B) 0.2 M sodium citrate–boric acid–NaOH (pH 10).36
The flow rate of the carrier was 0.3 dm3 minꢄ1
. Amino acids
ꢁ
eluted from the column were detected with a Shimadzu RF-535
spectrofluorometric detector (excitation wavelength, 355 nm;
emission wavelength, 435 nm) after post-column derivatization
with o-phthalaldehyde and N-acetyl-L-cysteine. A sodium hypo-
chlorite solution was used as the second derivatization reagent
for the detection of imino acids, such as sarcosine.
Experimental
Apparatus: A schematic diagram of SCWFR is shown in
Fig. 1. A sample solution was pumped into the reaction tube with
a HPLC pump (JASCO PU-1580). The fluid pressure was main-
tained at 25 MPa with a back-pressure regulator. The accuracy of
the temperature controller (ULVAC/SINKU-RIKU, TPC-1000)
Anhydrides of glutamic acid and !-amino acids were analyzed
with a reversed-phased-HPLC system (pump: TOSOH DP 8020).
The column used was a Develosil RP AQUEOUS C 30-U
(4:6 ꢃ 250 mm, Nomura Chemicals). The mobile phase used
was a 25 mM KH2PO4–H3PO4 buffer (pH 3.0). The flow rate
of the mobile phase was 1.0 dm3 minꢄ1. Eluents were monitored
with a spectrophotometric detector (TOSOH UV-8020; wave-
length, 205 nm).
Oligomerization of Glycine at Elevated Temperatures: We
examined the possible formation of glycine oligomers and their al-
teration in SHSs. An aqueous solution of 100 mM glycine was
pumped into the SCWFR at a flow rate of 0.5 dm3 minꢄ1 to heat
at 200–400 ꢁC under 25 MPa (over the critical pressure of water).
The sample solution was quenched externally at 0 ꢁC just after
being heated. The resulting products were analyzed by cation-
exchange HPLC and ion-pair HPLC.
ꢁ
ꢁ
ꢁ
was ꢂ1:5 C (ꢂ0:15%) bellow 200 C, and improved to ꢂ1 C
(ꢂ0:1%) at > 200 ꢁC (up to 1700 ꢁC). The fluid could be heated
rapidly up to such a high temperature as 400 ꢁC within a few sec-
onds with an IR gold image furnace (ULVAC RHL-E 410 P) with-
out preheating. The furnace was equipped with four 26.5 cm long
quartz-tungsten lamps, which were situated inside a gold-plated
body, which equally distributed heat to the reaction tube by
radiation. The reaction tube (1.74 mm i.d., 3.3 mm o.d. ꢃ 43
cm) used was made of Hastelloy C-276, and connectors used in
the high-temperature area of the system were made of Inconel
(INC-200-3). The heating time could be altered by changing
the flow rate: the heating time at a flow rate of 0.5 dm3 minꢄ1
was 2 min. Both the temperature outside and inside the reaction
tube could be monitored with thermocouples. The thermocouple
used to monitor the fluid temperature was sheathed with Inconel
and was inserted into the reaction tube 15.5 cm from the inlet
of the reaction tube. In the present experiments, the fluid tempera-
ture was monitored only at one point due to an equipment limita-
tion.
Chemicals: Glycine, L-threonine, DL-serine and 2-oxohexa-
methyleneimine were purchased from Wako Pure Chemicals.
DL-Aspartic acid, triglycine and tetraglycine were purchased from
Sigma Chemicals. Diglycine and 6-aminohexanoic acid were
from Nippon Rikagaku Yakuhin. Diketopiperazine, ꢁ-alanine,
sarcosine, ꢀ-aminobutyric acid, pyroglutamic acid, DL-glutamic
acid and ꢂ-aminobutyric acid were purchased from Tokyo Kasei
Kogyo. 5-aminovaleric acid, 2-piperidone and 2-pyrrolidinone
were purchased from Aldrich Chemicals. All of the reagents used
were of analytical grade. A Milli-Q Labo and a Millipore Simpli
Lab-UV were used to eliminate organic compounds and metal
ions from water. The solution in the reservoir was degassed by
ultrasonic vibration under reduced pressure, and then bubbled
with a gas mixture of nitrogen (99%) and hydrogen (1%) pur-
chased from Takachiho Kagaku, pure hydrogen gas from Syowa
Denko, or pure nitrogen gas from Taiyo Toyo Sanso.
Stability of Amino Acids at Elevated Temperatures: An
aqueous solution of a mixture of ten amino acids (aspartic acid,
threonine, serine, sarcosine, glutamic acid, ꢀ-aminobutyric acid,
ꢁ-alanine, ꢂ-aminobutyric acid, 5-aminovaleric acid and 6-
aminohexanoic acid; 10 mM each) was heated at four different
temperatures (250, 300, 350 and 400 ꢁC) at 25 MPa for 2 min.
An aliquot of each of the product solutions was hydrolyzed with
ꢁ
6 M HCl at 110 C for 24 hours in a sealed tube. Products both
before and after hydrolysis were analyzed by cation-exchange
HPLC and by reversed-phase HPLC.
Results and Discussion
Temperature Profile of the Flow Reactor: Fig. 2 shows
the temperature profile of the flow reactor, where the fluid tem-
perature monitored inside the reaction tube is plotted against
the temperature monitored on the surface of the reaction tube
(‘‘the set temperature’’). The flow rate of tꢁhe water eluent was
ꢁ
0.5 dm3 minꢄ1. In the ranges of 150–250 C and 425–450 C
of the set temperature, the fluid temperature was higher than
ꢁ
the set temperature; at 100 C and in the range of 300–420
ꢁC, on the contrary, the set temperature was higher than the
fluid temperature. It seems to have been due to a change of
the physicochemical properties.37{50 The physical properties

M. N. Islam et al.
Bull. Chem. Soc. Jpn., 76, No. 6 (2003) 1173
Table 1. Effect of Pressure on the Supercritical Temperature of Water When Flow Rate Was 0.5 dm3 minꢄ1
Set temp/ꢁC Actual temperature inside the reaction tube/ꢁC
At press. 25 MPa At press. 21 MPa
(Outside of the reaction tube)
406
ꢁ
ꢁ
374.5 C (Supercritical fluid phase)
409 C (Vapor/liquid phase)
Table 2. The Effect of the Flow Rate on the Temperature in
ꢁ
the SCWFR When the Set Temperature Was 300 C
Flow rate/
dm3 minꢄ1
Fluid heating
time/s
Temp. of fluid inside
the reaction tube/ꢁC
0.5
1
2
3
4
5
6
7
8
120
60
30
20
15
12
10
8.6
7.5
6.7
6
298
294
290
287
284
281
278
258
236
211
182
9
10
Pressure: 25 MPa.
sures when the set temperature was 406 ꢁC. As shown in
Table 1, a big difference in the fluid temperature was
monitored. When the pressure is lower than the critical pres-
sure, water has the properties of vapor/liquid phases, and has a
low dissociation constant (Kw) and a dielectric constant. When
the pressure is over the critical pressure, it behaves like the su-
percritical phase, except that its dielectric constant and viscos-
ity are much lower, because there are less hydrogen bonds
among the water molecules.51 The dissociation constant of
water decreases as the density decreases.52 The pressure ef-
fects the fluid temperature under the supercritical condition.
It is thus necessary to monitor the fluid temperature, particular-
ly in supercritical flow-reactor systems.
Effect of the Flow Rate on the Fluid Temperature:
Table 2 gives the variations in the temperature of the fluid in-
side the reaction tube when various flow rates were applied at
the same set temperature (300 ꢁC). At a lower flow rate of 0.5
dm3 minꢄ1, the fluid was heated for 120 seconds. When the
flow rate was higher, the heating time became shorter. At a
flow rate of 10 dm3 minꢄ1, for example, the fluid was heated
only for 6 seconds, and the difference between the fluid tem-
perature and the set temperature was increased. Thus, only
the set temperature cannot give the actual temperature of the
fluid inside the reaction tube. Evaluating Fig. 2, as well as
Tables 1 and 2, it should be emphasized that monitoring the
fluid temperature is essentially needed to obtain an accurate
temperature of the reaction.
Fig. 2. Temperature of fluid inside the reaction tube vs set
temperature (outside it). The flow rate of water eluent was
0.5 dm3 minꢄ1 under 25 MPa.
include thermal expansion, compressibility, heat capacity and
viscosity of water for the variation of the temperature aꢁnd
pressure. For example, as the temperature rises from 25 C
ꢁ
to 300 C, the density41;42 of water decreases from 0.997 to
0.713 g cmꢄ3, its dielectric constant43 decreases from 78.85
to 19.66 and its solubility parameter44 decreases from 23.4
to 14.5 (cal cm )
ꢄ3 1=2. Over the same temperature range, the
ionic product45 (dissociation constant) of water increases by
three orders of magnitude, from 1013:99 to 1011:30. Changes
of the above physical properties of water make water at 200
ꢁC to have solvation properties more similar to methanol or ni-
trobenzene than those of water at room temperature,46 and wa-
ter at 300 ꢁC shows solvation properties more similar to
acetone.47 These might be the reasons for the increase and de-
crease of the fluid temperature as well as causes for a fluid
temperature higher than the set temperature. The temperature
profile shown in Fig. 2 had an inflection point near to the su-
percritical point of water. The reason might be that the extent
and strength of hydrogen bonding decreases peculiarly near to
the critical point, and promotes the production of protons,48
and also that the heat mass-transfer coefficient for horizontal
flow becomes maximum at the supercritical point of water.49
ꢁ
Thus, a change of even 1 C from the set temperature caused
Oligomerization of Glycine in SCWFR: Figure 3 shows
ion-pair chromatograms of the prodꢁucts when a glycine solu-
drastic changes to the fluid temperature near the supercritical
points of water. High-temperature and pressure conditions af-
fect the chemical behavior of water.50 Thus, it is necessary to
monitor the fluid temperature to study reactions in water under
supercritical or subcritical conditions.
ꢁ
tion was heated at 350 C and 400 C. Diketopiperazine, di-
glycine, triglycine and tetraglycine were formed in the
SCWFR. All of the four-glycine oligomers were detected in
ꢁ
samples heated at 200–350 C. The formation of tetraglycine
Effect of Pressure on the Supercritical Temperature:
The fluid temperature was monitored under two different pres-
at 350 ꢁC was tentatively identified by the HPLC retention
time, and due to the lower yield GC/MS was not performed.

1174 Bull. Chem. Soc. Jpn., 76, No. 6 (2003)
Hydrothermal Reactions of Amino Acids
flow reactor at rather lower temperatures of 200–250 ꢁC. They
recycled the fluid, and the cycle time was 34 or 78 seconds ac-
cording to the flow rate. Diketopiperazine, diglycine and tri-
glycine were detected in the products without the addition of
any catalysts. On the other hand, Ikushima et al.34 reported
on the formation of oligoglycines up to the trimer over super-
critical conditions of water. Their heating time was as short as
0.2 to 1.1 seconds. The yields of glycine oligomers were high-
er in both of the experiments than our results. In both of the
experiments, electric heaters were used and the reaction tem-
perature was controlled by monitoring the temperature of the
heater. Our present results suggested that it is difficult to ob-
tain simple oligomers of glycine under supercritical conditions
in actual submarine hydrothermal systems, where the heating
time should be longer than minutes.
Stability of Amino Acids: Figure 4 shows the chromato-
grams of the hydrolyzed products when an aqueous mixture of
10 amino acids was heated at 400 ꢁC for 2 minutes in the
SCWFR. The reactants solution was bubbled with a mixture
of nitrogen (99%) and hydrogen (1%) to make more reducing
environments, like submarine hydrothermal systems.27 Here,
Fig. 4(a) was obtained by the injection of the product 50-times
more than in the case of Fig. 4(b). Glycine and alanine ob-
tained here were produced from other amino acids by thermal
decomposition. Threonine and sarcosine were totally absent in
the chromatograms because of their lower stability against
heat.
Fig. 3. Ion-pair chromatogramꢁs when 100 mM glycine so-
ꢁ
lution was heated at (a) 350 C and (b) 400 C under 25
MPa. The flow rate of the mobile phase was 0.5
dm3 minꢄ1
.
Abbreviation— DKP: diketopiperazine.
The unidentified peaks with asterisks were diminished
after hydrolysis.
Figure 5 shows the recovery ratio of amino acids after being
heated in SCWFR before and after acid hydrolysis when the
reactants solution was bubbled with pure (100%) nitrogen.
The recovery ratios of amino acids after hydrolysis were much
higher than those before hydrolysis, suggesting that the de-
crease in the concentration of amino acids was caused not only
by decomposition, but also by the formation of condensates of
amino acids that yielded amino acids after hydrolysis. Figures
6 and 7 represent the recovery ratios of amino acids when the
reactant mixtures were bubbled with a mixture of nitrogen
(99%) and hydrogen (1%), or pure hydrogen (100%),
respectively. From Figs. 5(b), 6 and 7 it can be seen that
the recovery ratio of amino acids was the highest when the re-
actants mixture was bubbled with a mixture of nitrogen (99%)
and hydrogen (1%), and lowest when bubbled with pure
nitrogen. Kohara et al.27 reported on the stability of amino
acids, where the reactants soꢁlution was heated in a closed sys-
tem at 8 MPa and 200–350 C; the recovery ratio was higher
ꢁ
None of the four peptides were, however, found at 400 C. It
is suggested that the glycine reactions in a supercritical fluid
are quite different from those in the liquid phase (at 200–350
ꢁC). There were many other peaks of unknown compounds
in all of the chromatograms of the unhydrolyzed products.
Most of those peaks disappeared after acid hydrolysis.
Table 3 summarizes the yields of glycine and its oligomers.
The yield of diketopiperazine was higher than any other oligo-
glycines, which implies that diketopiperazine is more stable
than peptides of glycine. It also shows, at all temperatures, in-
ꢁ
cluding 400 C, that the glycine amount increased after acid-
hydrolysis. It is suggested that condensates of glycine, which
yielded amino acids afteꢁr hydrolysis, formed even at such a
high temperature as 400 C.
Imai et al.19 reported the results of glycine oligomerization,
when 100 mM of a glycine aqueous solution was heated in a
ꢁ
Table 3. Products Obtained When 100 mM Glycine Aqueous Solution Was Heated at 200–400 C under 25
MPa
Products/mM
Temp./ꢁC
Before hydrolysis
After hydrolysis
Glycine (A)
DKP
Digly
Trigly
Tetragly
Glycine (B)
(B) ꢄ (A)
200
250
300
350
400
18.6
1.9
1.8
0.023
0.003
0.0178
0.0177
0.0082
0.0057
n.d.
0.0024
0.0017
0.0012
0.00053
n.d.
0.00014
0.00007
0.00009
0.00004
n.d.
trace
trace
trace
trace
n.d.
30.1
2.8
2.5
0.086
0.016
11.5
0.9
0.7
0.063
0.013
Abbreviation— DKP: diketopiperazine, digly: diglycine, trigly: triglycine, tetragly: tetraglycine. n.d.: not
detected. Flow rate of the reactant: 0.5 dm3 minꢄ1
.

M. N. Islam et al.
Bull. Chem. Soc. Jpn., 76, No. 6 (2003) 1175
Fig. 4. ꢁ Cation-exchange chromatograms of acid hydrolyzed products when an aqueous solution of ten amino acids was heated at
400 C at the flow rate of 0.5 dm3 minꢄ1 under 25 MPa. Injection volume in (a) was 50 times more than that in (b). Abbrevia-
tions— Asp: aspartic acid, Thr: threonine, Ser: serine, Sar: sarcosine, Glu: glutamic acid, Gly: glycine, Ala: alanine, ꢀ-ABA: ꢀ-
aminobutyric acid, ꢁ-Ala: ꢁ-alanine, ꢂ-ABA: ꢂ-aminobutyric acid, 5-AVA: 5-aminovaleric acid, 6-AHA: 6-aminohexanoic acid.
when a mixture of nitrogen (99%) and hydrogen (1%) was
used as pressurized gas than when pure nitrogen was used.
In both systems, the fugacity of hydrogen was proved to be im-
portant for the stability of amino acids. The difference in the
recovery ratios was less in the present flow reactor system,
ð3Þ
which seems to be due to the lower partial pressure of hydro-
gen in the present system.
Shock et al.52 theoretically discussed the stability of organic
compounds in hydrothermal systems using the following equi-
librium:
2CO2 þ 1=2N2 þ 9=2H2 ꢀ Glycine(C2H5O2N) þ 2H2O
ð1Þ
ð4Þ
Equation 1 suggests that the destruction of organic com-
pounds, such as amino acids, was suppressed by the high fuga-
city of hydrogen, which also agrees with our results.
Aspartic acid, threonine, serine and sarcosine showed low
stability against heat, compared to glutamic acid and the !-
amino acids. The recovery of hydrolyzed products of glutamic
acid and !-amino acids was much greater than the other ꢀ-
amino acids, and their amounts drastically increased after acid
hydrolysis. It seems to be due to the formation of intramole-
cular condensates of glutamic acid, ꢂ-aminobutyric acid (ꢂ-
ABA), 5-aminovaleric acid (5-AVA) and 6-aminohexanoic
acid (6-AHA) upon being heated (Eqs. 2–5) and the rings were
opened after acid hydrolysis.
ð5Þ
Figure 8 gives chromatograms of samples analyzed before and
after hydrolysis by reversed-phase HPLC: (a) is that of authen-
tic standards of pyroglutamic acid 1; 2-pyrrolidinone 2; 2-
piperidone 3; and 2-oxohexamethyleneimine 4. (b) and ꢁ(c)
are the unhydrolyzed product by heating at 250 and 400 C,
respectively. It shows the formation of pyroglutamic acid,
2-pyrrolidinone, 2-piperidone and 2-oxohexamethyleneimine.
All of the anhydrides were absent in the hydrolyzed products,
as shown in Figs. 8(d) and (e), which suggests that the pyroglu-
tamic acid, 2-pyrrolidinone, 2-piperidone and 2-oxohexameth-
yleneimine were hydrolyzed to amino acids. Pyroglutamic
acid decreased when heated at higher temperature. Actually,
as shown in Figs. 5(b), 6 and 7, that the recovery of glutamic
acid was around 95% at 250–300 ꢁC, while it was less than 5%
ð2Þ

1176 Bull. Chem. Soc. Jpn., 76, No. 6 (2003)
Hydrothermal Reactions of Amino Acids
Fig. 5. Recovery ratios of amino acids when reactants solution was bubbled with pure nitrogen (100%). The flow rate of reactant
solution was 0.5 dm3 minꢄ1 under 25 MPa. (a) Before hydrolysis, (b) After hydrolysis. Abbreviations— are shown in the caption
of Fig. 4.
Fig. 6. Recovery ratios of amino acids (after hydrolysis), when reactants solution was bubbled with a mixture of nitrogen (99%)
and hydrogen (1%). The flow rate of reactant solution was 0.5 dm3 minꢄ1 under 25 MPa. Abbreviations— are shown in the cap-
tion of Fig. 4.

M. N. Islam et al.
Bull. Chem. Soc. Jpn., 76, No. 6 (2003) 1177
Fig. 7. Recovery ratios of amino acids (after hydrolysis) when reactants solution was bubbled with pure hydrogen (100%). The
flow rate of reactant solution was 0.5 dm3 minꢄ1 under 25 MPa. Abbreviations— are shown in the caption of Fig. 4.
amino acids was possible at higher temperature, since the lac-
tams of these amino acids are quite stable against heat. We
can also say that the predominance of these !-amino acids
in some environments provides chemical markers of abiotic
hydrothermal synthesis.
Conclusions
The present SCWFR was proven to have the following char-
acteristics as a simulator of submarine hydrothermal systems:
1) Since an IR gold image furnace was applied to the present
system, it could heat the fluid more rapidly without preheating
than the previously reported reactors with electric furnaces.
Thus, an alteration of the reactants before being heated up to
the attempted temperature can be minimized.
2) The fluid temperature can be directly monitored with a ther-
mocouple dipped into the fluid where the reactions are
occurring. The monitoring of the fluid temperature was proved
to be very essential for high-temperature and pressure experi-
ments to obtain the actual reaction temperature. Monitoring of
the fluid temperature was performed only at one point due to
an equipment limitation. Thus, it remains one of our future
Fig. 8. Reversed-phase HPLC chromatograms of the prod-
perspectives to monitor the fluid temperature at multiple points
ucts when an aqueous solution of ten amino acids were
to evaluate the temperature profile in detail.
Both of the characteristics described above are essential
merits for the simulation of SHSs. It can therefore be used
heated at the flow rate of 0.5 dm3 minꢄ1 under 25 MPa.
(a) Standards: 1. Pyroglutamic acid, 2. 2-Pyrrolidinone,
3. 2-Piperidone and 4. 2-Oxomethyleneimine; (b) 250
ꢁ
to find new types of reactions that possibly occurred during
the course of chemical evolution in the primordial ocean.
Four types of oligoglycines, such as diketopiperazine,
diglycine, triglycine and tetraglycine, were formed at 200–
ꢁC, before hydrolysis; and (c) 400 C, before hydrolysis;
ꢁ
ꢁ
(d) 250 C, after hydrolysis; and (e) 400 C, after hydrol-
ysis.
at 400 ꢁC. The thermal stability of pyroglutamic acid is much
lower than that of lactams of !-amino acids, since pyroglu-
tamic acid still has a carboxylic group, which is decomposed
ꢁ
350 C, together with many other unknown compounds, from
a 100 mM glycine solution. The reaction of glycine in the
supercritical water was different from that in the sub-critical
water.
ꢁ
at a higher temperature (350 C).
Islam et al.53 reported on the formation of some !-amino
acids, such as ꢁ-alanine, ꢂ-aminobutyric acid, 5-aminovaleric
acid and 6-aminohexanoic acid, only at higher temperature
The stability of glutamic acid was higher than other ꢀ-ami-
no acids at 250–300 C due to the formation of an anhydride
ꢁ
ring (pyroglutamic acid). !-Amino acids, such as ꢂ-aminobu-
tyric acid, 5-aminovaleric acid and 6-aminohexanoic acid,
showed a much higher recovery than ꢀ-amino acids, even over
supercritical conditions of water. The higher recoveries were
ꢁ
(300–400 C), when a mixture of KCN (0.1 M), HCHO (0.1
M, with 10–15% methanol) and NH4HCO3 (0.05 M) was
heated. Hence, it is suggested that the formation of these !-

1178 Bull. Chem. Soc. Jpn., 76, No. 6 (2003)
Hydrothermal Reactions of Amino Acids
27 M. Kohara, T. Gamo, H. Yanagawa, and K. Kobayashi,
Chem. Lett., 1997, 1053.
28 H. Yanagawa and F. Egami, Biosystems, 12, 147 (1980).
29 H. Yanagawa and K. Kojima, J. Biochem., 97, 1521
(1985).
due to the formation of lactams. These !-amino acids could
be chemical markers of abiotic hydrothermal synthesis. It
was also shown that the stability of organic compounds de-
pends on the oxidizing-reducing conditions of the systems.
It is suggested that some types of chemical evolution might
require high-temperature and high-pressure environments
close to SHSs, it is important to study the chemical evolution
pathways in hydrothermal systems by using the SCWFR. Fur-
thermore, it is also important to use metal ions to simulate
SHSs in order to see what actually happening in natural
systems.19 These days, several projects have been conducted,
including the ‘‘Archaean Park’’ project,54 to study actual sub-
marine hydrothermal systems. It is of interest to compare the
results of simulation experiments with those observed in actual
submarine hydrothermal systems.
30 H. Yanagawa, Y. Kobayashi, and F. Egami, J. Biochem,
87, 359 (1980).
31 H. Yanagawa, Y. Kobayashi, and F. Egami, ‘‘Genesis of
amino acids in the primeval sea: formation of amino acids from
sugars and ammonia in a modified sea medium.’’ in ‘‘Origin of
Life,’’ ed by Y. Wolman, Reidel Publishing, Dordrecht (1981),
p. 181.
32 W. L. Marshall, Geochim. Cosmochim. Acta, 58, 2099
(1994).
33 R. J. C. Hennet, N. G. Holm, and M. H. Engel, Naturwis-
senschften, 79, 361 (1992).
34 Y. Ikushima, K. Hatakeda, and O. Sato, Sixth conf. On
supercritical fluid and their applications, ed by E. Reverchon,
Maiori, Italy, 525 (2001).
35 D. K. Alargov, S. Deguchi, K. Tsuji, and K. Horikoshi,
Origins Life Evol. Biosphere, 32, 1 (2002).
36 K. Kobayashi, T. Kaneko, T. Kobayashi, L. Hui, M.
Tsuchiya, T. Saito, and T. Oshima, Anal. Sci., 7 Suppl., 921
(1991).
This work was partly supported by Grant-in-Aid (No.
11640488) from MEXT and by special Coordination Fund of
‘‘Archaean Park’’ project from MEXT.
References
1 A. I. Oparin, ‘‘The Origin of Life on the Earth,’’ Academic
Press, New York (1957).
2
3
4
5
6
H. Urey, Proc. Nat. Acad. Sci. U.S.A., 38, 35 (1952).
S. L. Miller, Science, 117, 528 (1953).
S. L. Miller and H. Urey, Science, 130, 245 (1959).
K. Harada and S. W. Fox, Nature, 201, 335 (1964).
H. Yanagawa and K. Kobayashi, Origins Life Evol. Bio-
37 H. Arendt, ‘‘Washington’s problem-solver: Where they
went wrong,’’ New York Times, 5 April, (1972), p. 21.
38 P. Marris and M. Rein, ‘‘Dilemmas of Social Reform,’’
Aldine, Hawthrone, NY (1967), p. 245.
39 P. Moynihan, ‘‘Maximum Feasible Misunderstanding,’’
Macmillan, New York (1969), p. 170.
sphere, 19, 540 (1989).
7
8
9
C. Sagan and B. N. Khare, Science, 173, 417 (1971).
A. Bar-Nun and H. Hartman, Origin of Life, 9, 93 (1978).
A. Bar-Nun, N. Bar-Nun, S. H. Bauer, and C. Sagan, Sci-
40 Bureau of Labor Statistics, ‘‘The U. S. economy in 1980,’’
Monthly Labor Rev., April, (1970), p. 6.
41 ‘‘CRC Handbook of Chemistry and Physics,’’ 61 ed, CRC
Press, Boca Raton, FL (1981), p. E57.
ence, 168, 470 (1970).
10 T. W. Scattergood, C. P. Mckay, W. J. Borucki, L. P.
Giver, G. H. Van, J. E. Parris, and S. L. Miller, Lacrus, 81, 413
(1989).
11 W. W. Rubey, Geol. Soc. Am. Bull, 62, 1111 (1951).
12 J. S. Levin, J. Mol. Evol., 18, 161 (1982).
13 J. F. Kasting, K. J. Zahnle, and J. C. G. Walker, Precam-
brian Res., 20, 121 (1983).
14 T. Matsui and Y. Abe, Nature, 319, 303 (1986).
15 K. Kobayashi, T. Kaneko, T. Saito, and T. Oshima, Origins
Life Evol. Biosphere, 28, 155 (1998).
16 J. B. Corliss, J. Dymond, L. I. Gordon, J. M. Edmond, R. P.
von Herzen, R. D. Ballard, K. Green, D.Williams, A. Bainbridge,
K. Crane, and T. H. van Andle, Science, 203, 1073 (1979).
17 N. G. Holm, Origins Life Evol. Biosphere, 22, 5 (1992).
18 E-i. Imai, H. Honda, K. Hatori, and K. Matsuno, Viva Ori-
gino, 25, 289 (1997).
42 ‘‘NBC/NRC Steam Tables,’’ ed by L. Harr, J. S. Gallagher,
and G. S. Kell, Hemisphere, New York (1984).
43 G. C. Akerloff and H. I. Oshry, J. Am. Chem. Soc., 72,
2844 (1950).
44 Calculation from the specific internal energy of vaporiza-
tion in J. H. Keenan, F. G. Keyes, P. G. Hill, J. G. Moore, Steam
Tables, Wiley-Interscience, New York (1969).
45 W. L. Marshall, E. U. Frank, J. Phys. Chem. Ref. Data, 10,
295 (1981).
46 G. H. Brimhall and D. A. Crerar, ‘‘Mineralogy Chapt. 8,
ore fields: Magnetic or supergene,’’. ‘‘Thermodynamic modeling
of geological Materials: Minerals Fluids and Melts,’’ ed by I. S.
E. Carmiahael and H. P. Eugster, Mineralogical Society of
America, Washington, D. C. (1987), p. 235.
47 M. Siskin and A. R. Katritzky, Science, 254, 231 (1991).
48 Y. Ikushima, K. Hatakeda, N. Saito, and M. Arai, J. Chem.
Phys., 108, 5855 (1998).
49 K. Yamagata, K. Nishikawa, S. Hasegawa, T. Fuji, and S.
Yoshida, Internat. J. Heat Mass Transfer, 15-12, 2575 (1972).
50 J. P. Ferris, Origins Life Evol. Biosphere, 22, 109 (1992).
51 H. C. Helgeson, D. H. KirKham, and G. C. Flower, Amer.
J. Sci., 281, 1241 (1981).
52 E. L. Shock, Origins Life Evol. Biosphere, 22, 67 (1992).
53 M. N. Islam, T. Kaneko, and K. Kobayashi, Anal. Sci., 17
Suppl, 1631 (2001).
19 E-i. Imai, H. Honda, K. Hatori, A. Brack, and K. Matsuno,
Science, 283, 831 (1999).
20 E. L. Shock, Geochim. Cosmochim. Acta, 54, 1185 (1990).
21 E. L. Shock, Origins Life Evol. Biosphere, 20, 331 (1990).
22 E. L. Shock, Origins Life Evol. Biosphere, 22, 67 (1992).
23 E. L. Shock, Origins Life Evol. Biosphere, 22, 135 (1992).
24 M. Schulte and E. L. Shock, Origin of Life and Evol. Bio-
sphere, 25, 161 (1995).
25 S. L. Miller and J. E. Van Trump, ‘‘Origin of Life,’’ ed by
Y. Wolman (1981), p. 135.
26 K. Kobayashi and H. Yanagawa, Origins Life Evol. Bio-
sphere, 22, 147 (1992).
54 T. Urabe, N. Seama and J. Ishibashi, A. Maruyama, and K.
Marumo, Abstr., Earth and Planetary Science Conf., Chem. Soc.
Jpn., Cm-001, Tokyo, June (2001).