Chemical enhancement by nanomaterials under X-ray irradiation

Article abstract of DOI:10.1021/ja210239k

We report here a new phenomenon of dynamic enhancement of chemical reactions by nanomaterials under hard X-ray irradiation. The nanomaterials were gold and platinum nanoparticles, and the chemical reaction employed was the hydroxylation of coumarin carboxylic acid. The reaction yield was enhanced 2000 times over that predicted on the basis of the absorption of X-rays only by the nanoparticles, and the enhancement was found for the first time to depend on the X-ray dose rate. The maximum turnover frequency was measured at 1 × 10^-4 s^-1 Gy^-1. We call this process chemical enhancement, which is defined as the increased yield of a chemical reaction due to the chemical properties of the added materials. The chemical enhancement described here is believed to be ubiquitous and may significantly alter the outcome of chemical reactions under X-ray irradiation with the assistance of nanomaterials.

Full text of DOI:10.1021/ja210239k

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communication
pubs.acs.org/JACS
Chemical Enhancement by Nanomaterials under X-ray Irradiation
Neal N. Cheng, Zane Starkewolf, R. Andrew Davidson, Arjun Sharmah, Changju Lee, Jennifer Lien,
and Ting Guo*
Department of Chemistry, University of California, Davis, California 95616, United States
S
* Supporting Information
chemistry, chemical synthesis, radiotherapy, catalysis, sensing,
nanotoxicity, and nanomedicine.
Figure 1 shows several nanomaterials synthesized and
employed here, including 2.0 0.4 nm platinum nanoparticles
ABSTRACT: We report here a new phenomenon of
dynamic enhancement of chemical reactions by nanoma-
terials under hard X-ray irradiation. The nanomaterials
were gold and platinum nanoparticles, and the chemical
reaction employed was the hydroxylation of coumarin
carboxylic acid. The reaction yield was enhanced 2000
times over that predicted on the basis of the absorption of
X-rays only by the nanoparticles, and the enhancement
was found for the first time to depend on the X-ray dose
rate. The maximum turnover frequency was measured at 1
× 10−4 s−1 Gy−1. We call this process chemical enhance-
ment, which is defined as the increased yield of a chemical
reaction due to the chemical properties of the added
materials. The chemical enhancement described here is
believed to be ubiquitous and may significantly alter the
outcome of chemical reactions under X-ray irradiation with
the assistance of nanomaterials.
Figure 1. TEM images of (A) 1.7 nm PtNPs; (B−D) 3, 7, and 30 nm
AuNPs; (E) silica NPs (40 nm; and (F) silica-coated AuNPs (72 nm
Au core, 29 nm thick silica shell).
-ray absorption by materials has been broadly used in
X
imaging, lithography, and treatment since the discovery of
X-rays. Nanomaterials, which were widely used as catalysts
decades ago, are being intensely explored in many fields,
especially biology. The use of previously considered inert
nanomaterials such as gold nanoparticles to increase the
absorption of X-rays began a few years ago, and many chemical
and biological responses have been used to quantify the
enhancement.1,2 Because gold nanoparticles can be catalytically
active under suitable conditions,3−9 it is likely that these
nanomaterials may do more than simply enhance the
absorption of X-rays in a highly reactive environment such as
those created by X-ray radiation. However, all of the observed
enhancements to date have been attributed to physical
properties of the nanomaterials, i.e., high atomic numbers,
leading to increased X-ray absorption and subsequent increased
generation of reactive oxygen species (ROS), even though the
observed enhancements could be much higher than the values
predicted on the basis of the physical enhancement at low
loadings (<0.1 wt %) of nanoparticles1,10 or much lower at high
loadings (∼1 wt %) of nanoparticles.11 These disagreements
indicate that physical enhancement alone, even taking into
account reabsorption of emitted secondary photons and
electrons,12 which is negligible, cannot explain the observed
enhancement. Here we introduce a new concept, chemical
enhancement, that is enabled by both the radiation-generated
ROS and the surface of the nanomaterials. The concept
discovered here may also be useful in applications such as
energy production, nuclear waste processing, radiation
(PtNPs), 3.8 1.0 nm (denoted as “3 nm” in the text), 7.0
1.3 nm (“7 nm”), and 35.2
4.9 nm (“30 nm”) gold
nanoparticles (AuNPs), 53 6.4 nm silica nanoparticles, and
nanoparticles having a 72.6 4.6 nm gold core and a 28.8
4.0 nm thick silica shell. Figure 2A shows the increased
production of highly fluorescent 7-hydroxycoumarin-3-carbox-
ylic acid (7-OH-CCA) from hydroxylation of weakly
fluorescent coumarin-3-carboxylic acid (3-CCA) as a function
of AuNP concentration for the three nanoparticle sizes shown
in Figure 1B−D (3, 7, and 30 nm). The increase is expressed as
the enhancement of the yield of 7-OH-CCA caused by AuNPs,
which is defined as the ratio of the fluorescence signal of 7-OH-
CCA with AuNPs to that without AuNPs minus 1. Thus, an
enhancement value of 0 means no increase to the yield, and an
enhancement value of 1.0 means a 100% increase. The lowest
concentration to observe a 10% enhancement was less than 0.5
ppm or 20 nM for the 3 nm AuNPs, 3 ppm or 2.4 nM for 7 nm
AuNPs, and 25 ppm or 0.15 nM for 30 nm AuNPs. Because the
reaction is used as a dosimetry reaction, naturally such an
enhancement would be interpreted as an increase in OH·
production, which would be incorrect because AuNPs actually
played an active and chemical role. As the amount of AuNPs in
Received: October 31, 2011
Published: January 19, 2012
© 2012 American Chemical Society
1950
dx.doi.org/10.1021/ja210239k | J. Am. Chem.Soc. 2012, 134, 1950−1953

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of the American Chemical Society
Communication
or other ligands covering the AuNPs or even just any type of
nanoparticle, we synthesized and employed pure silica nano-
particles and silica-shell-covered AuNPs (Figure 1E,F). There
was almost no scavenging or enhancement at already high
nanoparticle concentrations. PEG-covered silica nanoparticles
were also tested, and no enhancement was observed, proving
that enhancement does not happen for nanoparticles in general
[see Figure SI-1 in the Supporting Information (SI)]. This
result also reconfirmed that previously claimed re-emission or
absorption of secondary X-rays did not cause the enhance-
ment.12 On the other hand, for an equal amount of AuNPs,
there was measurable enhancement, as shown in Figure 2A.
This proves that the enhancement observed in Figure 2A−C
was due to the surface gold atoms of the AuNPs and not to the
bulk gold atoms or surface atoms of nanoparticles in general.
We also synthesized and employed several other nanoparticles,
including Ag, Pt, CdTe, and TiO2 nanoparticles. PtNPs (2 nm)
covered with poly(vinylpyrrolidone) (PVP) ligands showed
similar enhancement as AuNPs. On the other hand, no
enhancement was observed over a large span of concentrations
for 15 nm AgNPs, suggesting that a plasmonic phenomenon is
not the cause of the enhancement.4 CdTe nanoparticles (3 nm)
were also synthesized and used, and no enhancement was
detected at a dose rate of 3.3 Gy/min. Large-band-gap
semiconductor TiO2 nanoparticles alone under X-ray irradi-
ation did not cause enhancement either.
We employed excessive amounts of sodium azide, sodium
nitrate, superoxide dismutase (SOD), and ascorbic acid to
determine the chemical species responsible for the enhance-
ment. Sodium azide was used to scavenge singlet oxygen
preferentially, and the enhancement was unchanged with the
addition of up to 1 mM sodium azide, proving that singlet
oxygen was not responsible for the enhancement. On the other
hand, 0.5 mg/mL SOD or 0.5 mM ascorbic acid quenched the
enhancement. Ascorbic acid scavenges OH, superoxide, and
singlet oxygen, whereas SOD removes only superoxides
effectively. These results suggest that the enhancement relies
on superoxides. Sodium nitrate aqueous solution was employed
to test the role of solvated electrons, and no detectable changes
were found.
On the basis of these investigations, we hypothesize that
weakly electronegative metal surfaces free of oxides, such as
those of AuNPs or PtNPs, may be necessary for the
enhancement observed here.4 Superoxides are also required;
their role may be to transfer electrons to the AuNPs or PtNPs
to make them anionic, allowing OH radical-adduct inter-
mediates 3-OH-CCA· to react on the surface to form 7-OH-
CCA either sequentially or simultaneously. If these hypotheses
are true, then the enhancement should increase as a result of
simultaneously increasing the concentration of intermediates
and the total surface area of the nanomaterials. This could be
done with more intense X-ray sources and greater nanoparticle
concentrations. Figure 2D shows the results of enhancement
measurements using a more intense microfocus X-ray source.
The dose rate measurements showed that the enhancement was
much improved at higher dose rates and high AuNP
concentrations, eventually reaching 200% or 2-fold enhance-
ment at 20 Gy/min with 0.1 wt % 7 nm AuNPs (square
symbols for experimental data). The solid lines are theoretically
predicted responses (see below). These results show that
enhancement is dose-rate-dependent at high AuNP concen-
trations and suggest that the enhancement processes must
Figure 2. Chemical enhancement results. (A) Enhancement as a
function of AuNP concentration for the three sizes of AuNPs.
Enhancement was observed below 1 ppm. (B) TOF for these three
sizes of AuNPs. (C) TOF as a function of the total surface area. (D)
Dose rate dependence of the absolute enhancement for 7 nm AuNPs.
The experimental data are shown as symbols and the theoretical
simulations (see the text) as solid lines. The dose rate dependence
reached saturation above 20 Gy/min. The simulations were based on a
model proposing that the activity of the AuNPs comes from
superoxide produced by X-ray radiation.
solution increased to >0.1 wt %, which is the value needed to
generate ∼10% physical enhancement (PE) (see below), the
experimentally measured enhancements started to decrease and
even became negative (i.e., antienhancement; data not shown),
which implies that these nanoparticles or their surfactants begin
to scavenge OH· at high enough AuNP or PtNP concen-
trations.13 This scavenging process may be the cause of the
observed low enhancement at high loadings of nanoparticles.
However, scavenging is negligible at sufficiently low concen-
trations (<0.1 wt %) of AuNPs, as shown in Figure 2. The
enhancements reached a maximum of 0.6 (60%) at an X-ray
irradiation dose rate of 3.3 Gy/min. If this is a catalytic reaction,
then on the basis of Figure 2A, the traditionally defined
parameter of turnover frequency (TOF), which is the number
of chemical reactions catalyzed by a surface atom in
nanoparticles per second, can be calculated. Figure 2B shows
that the TOF reached the highest values at the lowest
concentrations for each of the three sizes of nanoparticles
and then gradually decreased as the concentration of
nanoparticles increased. Such a decrease in TOF with
increasing total surface area suggests that there is a limiting
reagent other than the surface area of AuNPs. On the basis of
the data shown in Figure 2A,B over a large range of
concentrations, it appears that there is a size dependence.
The prominence of this feature subsided when the TOF data
were plotted as a function of the total surface area (Figure 2C),
although the 3 nm AuNPs still seemed to be better than the 30
nm AuNPs by a factor of 2 at the lowest surface areas. The
TOF reached a plateau at the maximum value of nearly 1 ×
10−4 s−1 Gy−1 at a dose rate of 3.3 Gy/min for 3 nm AuNPs
with minimum total surface areas. This weak size independ-
ence, which exists for several catalytic systems,14 is character-
istically different from the catalytic properties of small AuNPs,
indicating that the mechanism of enhancement is different from
that causing the oxidation of CO by small AuNPs on
substrates.3
To test whether the observed activity truly originated from
the surface of gold and not from poly(ethylene glycol) (PEG)
1951
dx.doi.org/10.1021/ja210239k | J. Am. Chem.Soc. 2012, 134, 1950−1953

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of the American Chemical Society
Communication
involve species such as superoxides that are generated by X-ray
radiation.
ence and nanoparticle size effect were also duplicated using this
model (see Figures SI-3 and SI-4).
The observed experimental data can be explained by
reactions of radical intermediates 3-OH-CCA· with super-
oxide-activated AuNPs or PtNPs. Figure 3 shows the new
The chemical enhancement (CE) of the effect of X-ray
radiation described here requires the activation of nanomateri-
als by superoxides produced under X-ray irradiation. Therefore,
we call it dynamic CE. CE is different from physical
enhancement (PE), which is defined as the increased
absorption of radiation that leads to increased generation of
ROS such as superoxides, OH·, and singlet oxygen as a result of
the introduced materials under irradiation. PE hence enhances
the X-ray absorption and therefore the energy deposition.19
Many examples exist. For instance, AuNPs were employed to
increase the cleavage of DNA strands through increased
absorption of X-rays.2 Nanoporous gold has been shown to
enhance radiolysis of water.20 Several recent experiments
employed AuNPs for their PE property and observed enhanced
damage to biological samples.21−23 Theoretical works have also
been carried out to explain the results in terms of the enhanced
energy deposition from the added nanomaterials.19,24−26 PE
can be further divided into two categories. Average or remote
PE, which we call type-1 PE, creates uniform enhancement in
solution. A general rule of thumb is that adding 1 wt % gold
(relative to water in the sample) creates ∼140% increase in
energy deposition. The observed CE for 3 nm AuNPs at 4 ×
10−5 wt % (Figure 2A) is hence 2000 times the predicted type-1
PE (Figure SI-5). Type-2 PE, a nanoscale or local physical
enhancement, can be effective only when two conditions are
simultaneously met: (1) the probe molecules (e.g., DNA) must
be placed within nanometers of the nanomaterial (e.g., AuNPs)
and (2) scavengers must be present to reduce the contribution
of OH· from surrounding water.19 However, neither condition
was met here. Both types of PE can depend on the X-ray
energy.27 Another possibility is that superoxides may be
converted to OH· and hydroperoxyl radicals by AuNPs (see
the SI). However, the amount of OH· produced this way is
considerably less than that produced from radiation of water,
and only a very small amount of hydroperoxyl radicals exist at
pH 7.0.28 Furthermore, this mechanism could not reproduce
the observed dose rate dependence shown in Figure 2D. As a
result, both types of PE could not explain the enhancements
measured here, and there is a negligible increase in ROS due to
the introduction of nanoparticles. We hence conclude that the
observed enhancement is solely caused by the increased
conversion of intermediates to the products occurring on the
surface of AuNPs or PtNPs.
Figure 3. Proposed mechanisms for chemical enhancement. The
proposed mechanism is a combination of the Michaelis−Menten (M−
M) and Eley−Rideal (E−R) mechanisms. Two possible reaction
pathways are shown: pathway 1 (dashed lines) is the previously
established mechanism of formation of 7-OH-CCA, and pathway 2
(solid lines) displays the proposed superoxide-activated AuNP
pathways. Pathway 2 employs OH· produced from AuNPs, but OH·
from water would also be possible. O2, superoxide (O2−), OH·, 3-CCA,
3-OH-CCA· (radical), 7-OH-CCA (the product), and AuNPs are
shown.
reaction pathway involving activation of the surface atoms in
AuNPs by X-ray radiation-generated superoxides (solid lines).
The originally proposed pathway in the literature is also shown
(dashed lines).15 The new pathway can be considered as a
combination of at least two well-known catalytic reaction
mechanisms. The radical intermediate 3-OH-CCA· can be
regarded as the substrate in the traditional enzyme kinetics
described in the Michaelis−Menten (M−M) framework, where
AuNP−superoxide (AuNP−O2−) would be the designated
enzyme. However, it is possible that an AuNP could become
negatively charged upon reaction with a superoxide, and the
negatively charged AuNP would enhance the reaction between
3-OH-CCA· and one of the O2 molecules around the negatively
charged AuNP (shown in Figure 3) to form 7-OH-CCA. This
deviates from the original M−M picture but resembles a
process described as the Eley−Rideal (E−R) mechanism
because the reacting oxygen molecule comes to the surface of
the AuNP to initiate the reaction. This combination hence
represents a new mechanism that makes the enhancement
dose-rate-dependent. Although 3-OH-CCA· may react with O2
in water and the complex may migrate to the surface of AuNPs,
this is unlikely because the lifetime of 3-OH-CCA−O2 is fairly
short.16 The proposed mechanism is different from another
previously proposed mechanism that suggests oxygen may
interact with the gold surface to form superoxides;17 if AuNPs
could form superoxides without radiation as found in previous
studies,17 then the enhancement would be dose-rate-
independent. We theoretically modeled the enhancement by
establishing the rate equations for AuNPs, DMSO, O2, O2− and
the dose rate (see eqs 1−6 in the SI). The AuNP concentration
dependence shown in Figure 2A was reproduced, closely
resembling that obtained based on the Langmuir−Hinshelwood
formula with the modification that O2 can be any of those
around the AuNPs and not necessarily the adsorbed superoxide
itself.18 The solid lines in Figure 2D show the theoretically
modeled results of the dose rate calculation, which agree with
the experimental data. In addition, the concentration depend-
The presented results show that only a small amount of
AuNPs can cause significant changes in the outcome of
radiation experiments. The proposed radiation-activated
dynamic CE can also explain the previously observed
enhancement in several studies in which a ∼50% increase in
the damage to biological samples was observed when <0.01 to
0.1 wt % AuNPs (uptake) was employed.11,29,30 The observed
enhanced damage could not be explained by PE, which at best
could account for only a small fraction of the damage. On the
other hand, on the basis of the work presented here, CE should
be on the order of 50% at those AuNP concentrations, and the
CE discovered here remains high in the presence of radical
scavengers such as those abundant in cells. It is also possible
that CE may lead to more complicated biological enhancement
(BE), so the concept of CE may play an important role in
understanding reactive environments such as cells where
radiation-generated and naturally existing intermediates and
1952
dx.doi.org/10.1021/ja210239k | J. Am. Chem.Soc. 2012, 134, 1950−1953

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of the American Chemical Society
Communication
(22) Sicard-Roselli, C.; Brun, E.; Duchambon, P.; Blouquit, Y.; Keller,
G.; Sanche, L. Radiat. Phys. Chem. 2009, 78, 177.
(23) Hwu, Y.; Liu, C. J.; Wang, C. H.; Chen, S. T.; Chen, H. H.;
Leng, W. H.; Chien, C. C.; Wang, C. L.; Kempson, I. M.; Lai, T. C.;
Hsiao, M.; Yang, C. S.; Chen, Y. J.; Margaritondo, G. Phys. Med. Biol.
2010, 55, 931.
(24) Pradhan, A. K.; Nahar, S. N.; Montenegro, M.; Yu, Y.; Zhang, H.
L.; Sur, C.; Mrozik, M.; Pitzer, R. M. J. Phys. Chem. A 2009, 113,
12356.
ROS-activated nanomaterials are abundant. Both CE and BE
could be crucial for understanding nanotoxicity under radiation.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis details, experimental protocols, theoretical method-
ology for modeling the CE, and additional experimental and
theoretical results. This material is available free of charge via
the Internet at http://pubs.acs.org.
(25) Cho, S. H.; Jones, B. L.; Krishnan, S. Phys. Med. Biol. 2009, 54,
4889.
(26) Kobayashi, K.; Usami, N.; Porcel, E.; Lacombe, S.; Le Sech, C.
Mutat. Res., Rev. Mutat. Res. 2010, 704, 123.
(27) McMahon, S. J. J. Phys. Chem. C 2011, 115, 20160.
AUTHOR INFORMATION
Corresponding Author
tguo@ucdavis.edu
■
(28) Zafiriou, O. C. Mar. Chem. 1990, 30, 31.
(29) Roa, W.; Zhang, X. J.; Guo, L. H.; Shaw, A.; Hu, X. Y.; Xiong, Y.
P.; Gulavita, S.; Patel, S.; Sun, X. J.; Chen, J.; Moore, R.; Xing, J. Z.
Nanotechnology 2009, 20, 37.
(30) Kong, T.; Zeng, J.; Wang, X. P.; Yang, X. Y.; Yang, J.;
McQuarrie, S.; McEwan, A.; Roa, W.; Chen, J.; Xing, J. Z. Small 2008,
4, 1537.
ACKNOWLEDGMENTS
■
We thank Ms. Larissa Miyachi for her experimental assistance.
This work was supported by the National Science Foundation
(CHE-0957413).
REFERENCES
■
(1) Hainfeld, J.; Slatkin, D.; Smilowitz, H. Phys. Med. Biol. 2004, 49,
N309.
(2) Foley, E.; Carter, J.; Shan, F.; Guo, T. Chem. Commun. 2005,
3192.
(3) Haruta, M.; Kobayashi, T.; Sano, H.; Yamada, N. Chem. Lett.
1987, 405.
(4) Garcia, H.; Navalon, S.; de Miguel, M.; Martin, R.; Alvaro, M. J.
Am. Chem. Soc. 2011, 133, 2218.
(5) Mirkin, C. A.; Zhang, K.; Cutler, J. I.; Zhang, J. A.; Zheng, D.;
Auyeung, E. J. Am. Chem. Soc. 2010, 132, 15151.
(6) Tsukuda, T.; Tsunoyama, H.; Ichikuni, N.; Sakurai, H. J. Am.
Chem. Soc. 2009, 131, 7086.
(7) Cao, R.; Cao, R.; Villalonga, R.; Diaz-Garcia, A. M.; Rojo, T.;
Rodriguez-Arguelles, M. C. Inorg. Chem. 2011, 50, 4705.
(8) Lambert, R. M.; Turner, M.; Golovko, V. B.; Vaughan, O. P. H.;
Abdulkin, P.; Berenguer-Murcia, A.; Tikhov, M. S.; Johnson, B. F. G.
Nature 2008, 454, 981.
(9) Hutchings, G. J.; Hughes, M. D.; Xu, Y. J.; Jenkins, P.; McMorn,
P.; Landon, P.; Enache, D. I.; Carley, A. F.; Attard, G. A.; King, F.;
Stitt, E. H.; Johnston, P.; Griffin, K.; Kiely, C. J. Nature 2005, 437,
1132.
(10) McMahon, S. J.; Hyland, W. B.; Muir, M. F.; Coulter, J. A.; Jain,
S.; Butterworth, K. T.; Schettino, G.; Dickson, G. R.; Hounsell, A. R.;
O’Sullivan, J. M.; Prise, K. M.; Hirst, D. G.; Currell, F. J. Sci. Rep. 2011,
1, 18.
(11) Chithrani, D. B.; Jelveh, S.; Jalali, F.; van Prooijen, M.; Allen, C.;
Bristow, R. G.; Hill, R. P.; Jaffray, D. A. Radiat. Res. 2010, 173, 719.
(12) Misawa, M.; Takahashi, J. Nanomedicine 2011, 7, 604.
(13) Shirahata, S.; Hamasaki, T.; Kashiwagi, T.; Imada, T.;
Nakamichi, N.; Aramaki, S.; Toh, K.; Morisawa, S.; Shimakoshi, H.;
Hisaeda, Y. Langmuir 2008, 24, 7354.
(14) Qu, Y. Q.; Sutherland, A. M.; Lien, J.; Suarez, G. D.; Guo, T. J.
Phys. Chem. Lett. 2010, 1, 254.
(15) Louit, G.; Foley, S.; Cabillic, J.; Coffigny, H.; Taran, F.; Valleix,
A.; Renault, J. P.; Pin, S. Radiat. Phys. Chem. 2005, 72, 119.
(16) Bohn, B. J. Phys. Chem. A 2001, 105, 6092.
(17) Ionita, P.; Gilbert, B. C.; Chechik, V. Angew. Chem., Int. Ed.
2005, 44, 3720.
(18) Zhang, Z. Y.; Berg, A.; Levanon, H.; Fessenden, R. W.; Meisel,
D. J. Am. Chem. Soc. 2003, 125, 7959.
(19) Carter, J. D.; Cheng, N. N.; Qu, Y. Q.; Suarez, G. D.; Guo, T. J.
Phys. Chem. B 2007, 111, 11622.
(20) Renault, J. P.; Musat, R.; Moreau, S.; Poidevin, F.; Mathon, M.
H.; Pommeret, S. Phys. Chem. Chem. Phys. 2010, 12, 12868.
(21) Butterworth, K. T.; Coulter, J. A.; Jain, S.; Forker, J.; McMahon,
S. J.; Schettino, G.; Prise, K. M.; Currell, F. J.; Hirst, D. G.
Nanotechnology 2010, 21, 29.
1953
dx.doi.org/10.1021/ja210239k | J. Am. Chem.Soc. 2012, 134, 1950−1953