Continuous Radio Amplification by Stimulated Emission of Radiation using Parahydrogen Induced Polarization (PHIP-RASER) at 14 Tesla

Article abstract of DOI:10.1002/cphc.201901056

Nuclear Magnetic Resonance (NMR) is an intriguing quantum-mechanical effect that is used for routine medical diagnostics and chemical analysis alike. Numerous advancements have contributed to the success of the technique, including hyperpolarized contrast agents that enable real-time imaging of metabolism in vivo. Herein, we report the finding of an NMR radio amplification by stimulated emission of radiation (RASER), which continuously emits ¹H NMR signal for more than 10 min. Using parahydrogen induced hyperpolarization (PHIP) with 50 % para-hydrogen, we demonstrated the effect at 600 MHz but expect that it is functional across a wide range of frequencies, e.g. 10¹–10³ MHz. PHIP-RASER occurs spontaneously or can be triggered with a standard NMR excitation. Full chemical shift resolution was maintained, and a linewidth of 0.6 ppb was achieved. The effect was reproduced by simulations using a weakly coupled, two spin-1/2 system. All devices used were standard issue, such that the effect can be reproduced by any NMR lab worldwide with access to liquid nitrogen for producing parahydrogen.

Full text of DOI:10.1002/cphc.201901056

Accepted Article
Title: Continuous Radio Amplification by Stimulated Emission of
Radiation using Parahydrogen Induced Polarization (PHIP-
RASER) at 14 Tesla
Authors: Andrey Pravdivtsev, Frank D. Sönnichsen, and Jan-Bernd
Hövener
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPhysChem 10.1002/cphc.201901056
Link to VoR: http://dx.doi.org/10.1002/cphc.201901056
A Journal of
www.chemphyschem.org

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
Continuous Radio Amplification by Stimulated Emission of
Radiation using Parahydrogen Induced Polarization (PHIP-
RASER) at 14 Tesla
[a]
[b]
[a]
Andrey N. Pravdivtsev,* Frank D. Sönnichsen and Jan-Bernd Hövener*
Abstract: Nuclear Magnetic Resonance (NMR) is an intriguing
quantum-mechanical effect that is used for daily life medical
diagnostics and chemical analysis alike. Numerous advancements
have contributed to the success of the technique, including
hyperpolarized contrast agents that enable real-time imaging of
metabolism in vivo. Here, we report the finding of an NMR radio
amplification by stimulated emission of radiation (RASER), which
three seconds after a hyperpolarized sample was poured into a
cavity.[8] Continuous emission of the NMR signal is not feasible
with this approach because of relaxation and long sample
preparation time.
Closest to a continuous emission at high-frequency was
likely a 129Xe RASER, where 11 bursts were observed at 11.7 T
(
139 MHz) for 8.5 min. Because the polarization was not
1
continuously emits H NMR signal for more than 10 min. Based on
refreshed, the amplitude was continuously decaying. Interestingly,
the magnetization of the sample was so strong that its Larmor
frequency was elevated by 3 Hz at the beginning of the
experiment due to strong distant dipolar fields.[9]
parahydrogen induced hyperpolarization (PHIP, 50 % para-
enrichment), PHIP-RASER is expected to be functional in a wide
1
3
range of frequencies (10 – 10 MHz) and was demonstrated at
00 MHz. PHIP-RASER occurred spontaneously upon supply of pH
6
2
In a recent breakthrough, Süfke et al.[10] demonstrated a
to a solution containing a hydrogenation catalyst and unsaturated
molecule, or was triggered with a standard NMR excitation pulse. Full
chemical shift resolution was maintained and a linewidth of 0.6 ppb
continuous NMR RASER based on a combination of innovative
hardware[11] and the continuous hyperpolarization[12] provided by
Signal Amplification By Reversible exchange[13] (SABRE):
SABRE-RASER.[14] Using a sophisticated setup, a spectral
resolution of 0.6 Hz was achieved at ≈ 3.8 mT (≈ 4 ppm). This
unprecedented resolution of J-couplings was reported, but
chemical shift resolution was not possible because of the low
magnetic field.[10,14] A transfer of this method to higher fields is not
was achieved, despite apparent inhomogeneities from supplying pH
The effect was reproduced by simulating a weakly coupled, two spin-
system with superoperators. All equipment used was standard
2
.
½
issue, such that the effect can be reproduced by any NMR lab
worldwide with access to liquid nitrogen for producing parahydrogen.
feasible because a level anticrossing (LAC) is required for
spontaneous polarization transfer to occur (for 1H, ≈ mT).[15,16]
Introduction
Spontaneous SABRE was reported at higher fields, too, but only
with a very small polarizations.[17,18] Continuous high-field Radio
The quest for a continuously emitting, high-frequency, liquid-state
radio amplification by stimulated emission (RASER) at MHz
frequencies and above is ongoing for more than a decade.
Pioneering work demonstrated RASERs at low fields with
Frequency (RF) driven SABRE polarization transfer techniques
still provided only moderate and semi-continuous polarization.[19–
21]
3
resonance frequencies from 10 Hz to 50 kHz, based on He or
129Xe gases polarized with Spin Exchange Optical Pumping
(
SEOP),[1,2] for hours[3,4] or days. These techniques were used,
Results and Discussion
among others, for probing fundamental symmetries.[5]
A 400 MHz RASER was accomplished by Dissolution
Dynamic Nuclear Polarization (dDNP)[6,7], although only for a
single shot experiment: RASER signal bursts were observed for
Here, we present a liquid-state NMR RASER that
continuously emits RF-signal for any given time at room
temperature and 600 MHz (Fig 1 and 2) and under the conditions
described below. Using parahydrogen-induced hyperpolarization
(
PHIP), the RASER begins spontaneously or can be triggered
with a standard NMR pulse.
Similar to SABRE-RASER,[10] this new effect is based
[
a]
Dr. A. N. Pravdivtsev and Prof. Dr. J.-B. Hövener
Section Biomedical Imaging, Molecular Imaging North Competence
Center (MOIN CC), Department of Radiology and Neuroradiology,
University Medical Center Kiel, Kiel University,
Am Botanischen Garten 14, 24114, Kiel, Germany
E-mails: andrey.pravdivtsev@rad.uni-kiel.de
jan.hoevener@rad.uni-kiel.de
Prof. Dr. F. Sönnichsen
Otto Diels Institute for Organic Chemistry, Kiel University,
Otto Hahn Platz 5, 24098, Kiel, Germany
on the spin order of parahydrogen (pH
2
) (eq. 1),
1
ꢁ
ꢁ
휌̂ 푝퐻2
=
− 퐈 ∙ 퐈 ,
(1)
1
ꢂ
which is continuously supplied to a liquid sample (Fig 1B, 2B).[22]
Elegantly, pH has spin 0, hence does not have
[
b]
2
magnetization and is impervious to any excitation in its molecular
state. Thus, the reservoir of spin order is not affected by pulses or
the RASER effect, which takes place in the same physical location.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
The production of pH
more than H gas flowing through a porous catalyst at low
temperature.[23–25] Stored appropriately, pH
is stable for days to
weeks.[26]
can be purchased or produced on-site chemically or
by electrolysis.[27]
In contrast to SABRE-RASER,[10] here, pH
2
is well established, requiring no
Importantly, we expect that PHIP-RASER is functional
at all (high) magnetic field strengths, where parahydrogen and
synthesis allows dramatically enhanced nuclear alignment
(PASADENA)[33] - although the spectral appearance depends on
the radiation damping rate, polarization level and build-up rate
(Fig S3, S9, S10).[14] PASADENA conditions are met if the J-
coupling between the added hydrogens is weak compared to their
chemical shift difference; for J ≈ 10 Hz and chemical shift
difference of 1 ppm, this condition is fulfilled at fields > 0.2 T, i.e.
2
2
H
2
2
is
permanently incorporated into target molecules via homogeneous
hydrogenation (Fig 1A and 2A). The catalytic activity was chosen
such that only a small fraction of all precursor molecules is
hydrogenated at a time. This way, the reaction was upheld for an
extended period of time, and PHIP-RASER occurred.
Here, we demonstrated PHIP-RASER for two
molecules: ethyl cinnamate and methyl acrylate, for our purposes
approximately an AX, 2 spin-½ system and an AMX, 3 spin-½
system.
1
at H frequencies above 10 MHz.
Interestingly, the frequency difference between the
RASER lines (휁) was found to differ from their nominal chemical
shift difference (Δ훿). At the same time, several equidistant satellite
resonances with the same spacing 휁 were found (Fig 1E and
2E).[14] Their location depends on the radiation damping rate
relative to the chemical shift difference (Fig S8).[14] Another type
of satellites exhibited a more complex structure that strongly
depended on experimental settings, not unlike the spectral
clustering effect caused by a distant dipolar field.[34] The high level
of magnetization also caused a change of the local magnetic field
and thus a frequency drift of the resonances (Fig S9).[9,34] All these
findings were well reproduced by simulations, as described below.
All simulations were performed using the MOIN spin library; the
source code is available online.[35,36]
The 600 MHz emission of the RASER began
spontaneously approx. 1.5 min after injection of pH
instantaneously after a single “trigger” excitation pulse of 45° (Fig
, 2). The “trigger” RF-pulse converted hyperpolarization into
2
, or
1
transverse magnetization, which, in turn, induced continuous
radiation in a few seconds. The “spontaneous” onset of PHIP-
RASER required a finite hydrogenation time for the accumulation
of a hyperpolarized product.
NMR-RASER data was acquired for a much longer time
period than the usual NMR signal. The duration of the emission
The interaction between the highly polarized system and
(
here > 10 min) is limited only by pH
catalyst and the reservoir of receiver molecules. In the current
implementation, only pH was continuously renewed, but
2
supply, the hydrogenation
RF-cavity is essential for the RASER emission.[3–5,10,11] This
interaction reveals itself as radiation damping. The radiation
휇0
ꢃ1
ꢂ
damping rate is defined as
(
휏푅퐷
)
푠 ꢅ
= 4 ℏ훾 휂푄푐 |푃|, where ꢄ is the
2
refreshing all constituents can be easily accomplished using a
flow setup.[28,29]
vacuum permeability, ℏ is the reduced Planck constant, 훾 is the
gyromagnetic ratio, Q is the quality factor of the coil with a filling
After the Fourier transformation of ≈ 10 min RASER
signal, two narrow lines were observed at the positions of the
added hydrogens, in addition to less pronounced satellites
factor 휂, 푐 is the concentration of the nuclear spin and 푃 is the
푠
longitudinal polarization.[37] This definition of radiation damping
rate is very convenient for experiments with large magnetizations
in thermal equilibrium that can be described simply with modified
Bloch equations, or, in a more advanced way, with single- or multi-
mode LASER/RASER equations.[14]
(
details of signal processing are given in methods). The line
shapes were irregular with a varying full width at half maximum
FWHM) of e.g. ≈ 0.6 ppb (≈ 0.4 Hz, Fig 1D) and ≈ 7.8 ppb (≈
.7 Hz, Fig 2D). The reason for the difference in line width may
be attributed to varying field homogeneity during pH injection and
(
4
All experiments in this paper were conducted with a
2
cryogenically cooled coil with high Q ≈ 500 (see SI). The question
arises, if the effect can be reproduced with a conventional coil,
too. According to the theory of radiation damping, only the filling
factor, the Q-factor of the coil and size of the magnetization are
important. Thus, it appears plausible to observe PHIP-RASER
with a high resolution NMR coil at room temperature with a Q >
100, but the experimental proof has yet to be established. Note
the different spin systems, but will have to be investigated
elsewhere. Note, that no information about the J-coupling network
in both 2 and 3 spin-½ systems of ethyl cinnamate and methyl
acrylate were revealed in these RASER experiments (Fig 1D, 2D).
At the same time, the resonances of the thermally polarized
solvent exhibited an inhomogeneously broadened line with
FWHM > 20 Hz, (Fig S4). Thus it remains unclear, how much of
the sample actually emitted RASER signal. In the absence of
these inhomogeneities and magnetic field drift (no lock was
applied), the lower limit for the RASER line width is the nominal
spectral resolution of the order of 1 mHz for 11 min. A better
homogeneity may be achieved in the future by using more
that no RASER was observed when continuous SABRE was
reported at ≈ 5 mT, ≈ 10 % filling factor and a Q ≈ 101
.
-2 [12,38]
To elucidate this remarkable effect further, we set out to
simulate PHIP-RASER in a coupled two spin-½ system using the
density matrix approach and Liouville-von Neumann equation
(LvN, eq. S8). It is the most convenient and precise method to
simulate liquid state NMR-experiments with coupled multi-spin
systems.
sophisticated techniques to dissolve pH [30–32]
and is currently
2
under investigation. A pH supply that does not disturb the
2
homogeneity of the magnetic field will be instrumental to elucidate
PHIP-RASER effect further (compare Fig 1E and Fig S3).
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
Figure 2: Spontaneous 1H PHIP-RASER at 600 MHz. An NMR-RASER was
2
induced by supplying pH into an NMR tube with methyl propiolate and Rh-
catalyst (A) coupled to an NMR resonator in situ (B). Without RF-excitation,
spontaneous emission of NMR signal began about one minute after the onset
2 2
of pH supply and lasted until the pH supply was stopped 11 minutes later (C).
The Fourier transform of 9 min data (C) exhibited two narrow lines with a full
width at half maximum of 8 ppb (D, E; blue lines: smoothed magnitude
spectrum). The RASER signal was acquired without a frequency lock; instead,
the right RASER line was aligned to the negative right line in a PASADENA
spectrum measured with 2H-lock and detuned probe (D, red). Several less
prominent satellites were observed in a magnified spectrum (E). Similar results
were obtained for ethyl phenylpropiolate (not shown). Details are given in
methods.
Figure 1: Triggered 1H PHIP-RASER at 600 MHz. An NMR-RASER was
2
induced by supplying pH into an NMR tube with ethyl phenylpropiolate and Rh-
catalyst (A) coupled to an NMR resonator in situ (B). Upon one 45° RF-excitation,
NMR signal was observed for more than 12 min (C). The Fourier transform of
10.5 min data (C) exhibited two narrow lines with a full width at half maximum
of 0.6 ppb (D, E; blue lines: smoothed magnitude spectrum). The right RASER
line, which was acquired without frequency lock, was set to match the right
negative line of a 45° PASADENA spectrum, acquired with lock and detuned
probe (D, red, no radiation damping). The magnified spectrum (E) revealed
several types of equidistant satellites, some of them are separated by 휁 (E).
Similar results were obtained for methyl proprionate (not shown). Details are
given in methods.
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
First of all, it was necessary to introduce an additional
element into conventional liquid state Hamiltonian[37] to account
broadening of spectral lines (Fig S10, SI) and RASER bursts,
which may confound the interpretation of the spectra or causes
inefficient polarization transfer to heteronuclei. The effect of RD
on H-PASADENA and polarization transfer to 13C was recently
[39]
for radiation damping:
푟푑
푅퐷
1
ꢀ
(
)
∑
ꢁ
V
푡, 휌̂ = 흎
∙
퐈 =
푘ꢆ1,ꢂ
푘
훼푅퐷 ∑푘,푛ꢆ1,ꢂ(푚푘푌(푡) 퐼 − 푚푘푋 푡 퐼푛푌)
ꢁ
( )ꢁ
(2)
demonstrated experimentally by Korchak et al.
temperature at fields of 1 T and 7 T.
[41]
at room
=
푛푋
This interaction can be written in analogy to the modified Bloch-
Maxwell equation[40] and is similar to multi-mode RASER
equations 17 and 18 from Ref. [14]. Here, 푚푘푋,푌 = 푇ꢇ ꢈ 휌̂ ∙
1
†
푘푋,푌
1
1
†
푘푋,푌
ꢁ
ꢁ
ꢁ
퐼
ꢉ/푇ꢇ ꢈ 퐼
∙
퐼
ꢉ are the amplitudes of the transverse
푘푋,푌
ꢂ
ꢂ
ꢂ
polarization of spin ꢊ = 1 or 2 of the density matrix 휌̂ ; in general,
the amplitudes are time dependent. The magnetic field induced
by RD is 흎푅퐷 = 훼푅퐷(∑ 푘푁 ꢆ1 푚푘푌 , − ∑푁 푚푘푋(푡) , ꢋ) . A radiation
푘ꢆ1
휇0
ꢂ
damping rate without polarization factor, 훼푅퐷 = 4 ℏ훾 휂푄푐 , is
푠
constant if the chemical concentrations and electronic circuitry are.
On the other hand, (휏푅퐷)ꢃ1 depends on the polarization, P, which
changes continuously because of relaxation or re-
hyperpolarization.
We assumed to find the simplified, two spin-½ system
[22]
after hydrogenation with pH
2
in the following state (eq 1):
1
ꢁ
ꢁ
퐼
휌̂ ꢌ퐴푆퐴퐷퐸푁퐴
=
− 퐼
(3)
1푍 ꢂ푍
4
The supply of para-order to the system was
implemented with a source operator:
( ) ꢁ
ꢁ
ꢁ
ꢂ
ꢍ = −푊 푡 ꢎ퐈 ∙ 퐈 ꢏ
(4)
푖푛
1
No unity operator was used (see eq. 1) to keep the trace over the
density matrix equal to 1. 푊 (푡) is a time-dependent rate of the
푖푛
spin-order influx. Depending on the simulated experiment,
different source operators can be used.
The addition of RD into the LvN equation makes the
simulation of polarization transfer in the presence of RD[41]
straightforward. Using the published SABRE models including
chemical exchange, the same approach can be used to simulate
SABRE-RASER.[35,42] A more detailed description of the theory
and more simulation examples are given in SI. The conditions of
the RASER threshold are discussed in more detail elsewhere.[14]
With these additions (eq. 2-4), we were able to
reproduce all essential features observed experimentally (Fig 3
and SI): (i) two PHIP-RASER lines (Fig 3D), (ii) a distance
between the RASER lines 휁 that is not equal to the chemical shift
difference  and which depends on RD parameters (Fig 3D and
Fig S9), (iii) an equidistant frequency-comb (Fig 3E) and (iv) a
frequency shift of resonances (the latter was also reported in Refs
[9,34],
Fig S9).
We found that the experimental signals exhibited much
Figure 3: Simulation of 1H PHIP-RASER at 600 MHz. A two spin-½ system
was continuously hyperpolarized by means of PHIP (A, 퐽 = 10 Hz, Δ훿 = 1.023
ppm, relaxation times 5 s). The evolution of the system after a 2° RF-excitation
was simulated using the LvN equation (C, 퐵ꢅ = 14.1 T, 훼푅퐷 = 1500 s-1 and 푊푖푛
stronger amplitude variations than the simulated ones. We
attribute this effect to the magnetic field inhomogeneity induced
by the pH
2
supply, which was not considered in the simulations.
-
1
=
50 s ). Radiation damping was found to induce zz-spin order (C, violet),
The same simulations were used to shed light into the
longitudinal (푚1푍, red) and transverse magnetization, the RASER signal (푚1푋 +
푚ꢂ푋 , black line, with magnified view). The last 10 s of the transverse
magnetization (C, blue square) were Fourier transformed and exhibited two
lines (D, E; blue lines; magnitude, apodized with 푒ꢐꢑ(−푡/ꢒꢓ)). A simulated
PASADENA spectrum was added for reference (D, red line, no radiation
damping, FWHM = 1 Hz). Several equidistant but smaller satellites at  were
found in the enlarged RASER spectrum (E, compare with Fig 1E,2E). The right
classic PASADENA experiment, too. For example, an asymmetric
line broadening and emission of “spontaneous”, echo-like RASER
2
bursts were observed up to 30 s after the pH supply was stopped
(
see Fig S7, S11). Simulations revealed that radiation damping
results in conversion of PASADENA spin order (eq 3) into
longitudinal 푚 − and transversal 푚푥푦 − magnetization (Fig S10,
푧
S11). Experimentally, this is manifested in an asymmetric
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
RASER resonance was shifted by +18 Hz to match the negative right line in
PASADENA spectrum.
2
experiments, some convection and diffusion occurred. pH bubbling was
stopped only after approx. 12 minutes of signal acquisition. Unfortunately,
the NMR spectrometer did not allow for continuous data acquisition.
Instead, the data was saved in 9 (Fig 1B) or 8 (Fig. 2B) blocks of 90 s and
1730768 points each. A 1-3 s delay in between the blocks caused a loss
of phase coherence such that the spectra had to be calculated for each
block individually. For the figures, the spectra of the first and last block in
each set were omitted to exclude transient effects at the onset and end of
the pH2 injection. Magnitude RASER spectra were smoothed after
averaging using a Savitzky-Golay filter with a first-order polynomial for a
window of 50 points (~0.22 Hz or °0.36 ppb). No other post-processing
Conclusion
These findings establish PHIP-RASER as the first, long-
lasting 1H NMR-RASER at high field (and high frequencies),
operating at room temperature and in the liquid state. Using
standard commercial equipment, the method can be easily
implemented in any NMR laboratory. The key element is a strong
coupling between the resonator and hyperpolarized sample.
Moreover, with a frequency range of 10 MHz and above,
PHIP-RASER is very flexible. The maximum frequency is
currently limited only by the static magnetic fields available (≈
methods such as apodization or others to improve the magnetic field
stability/homogeneity (as was done in Ref. [14]) were applied.
Simulation. Simulation parameters for Fig 3: initial density matrix 휌̂ (−ꢋ) =
ꢔꢕ /ꢖ − Iꢁ 1푍 Iꢁ ꢂ푍 , equilibrium state 휌̂ ꢗ푞 = ꢔꢕ /ꢖ , J = 10 Hz, chemical shift
-
1
1
.2 GHz).
The long emission > 10 min allows to partially negate the
0
difference 1.014 ppm, B =14.1 T, 훼푅퐷 = 1500 s , relaxation time 5 s and
-
1
rate of polarization influx 푊푖푛 = 5ꢋ s . The simulation model is described
detrimental effects of the strong inhomogeneity present in the
sample, as narrow lines of <1 ppb were observed. Given a more
homogeneous delivery of pH2, it appears feasible to further
improve on these linewidths. Simulations suggest that extremely
narrow lines of a few mHz can be obtained at 600 MHz, effectively
in SI and source code is available online.[36]
Data availability. All data is available from the corresponding authors
upon reasonable request.
2
overcoming the T limit. Note that at the same time, the full, 14-T-
chemical shift dispersion is maintained. These and other
applications yet to discover and implement (like RASER
gyroscopes[14]) make PHIP-RASER a highly interesting effect for
NMR, physics, chemistry and more.
Methods
Materials.
The
sample
solution
contained
4 mM
1,4-
Bis(diphenylphosphino)
butane
(1,5-cyclooctadiene)
rhodium(I)
tetrafluoroborate (Strem Chemicals, CAS: 79255-71-3) and 60 mM ethyl
phenylpropiolate (EP, Figure 1A, Sigma-Aldrich, CAS: 2216-94-6) or
Scheme 1. Scheme of the experimental workflow (protocol 1): i) At a given
100 mM methyl propiolate (MP, Figure 2A, Sigma-Aldrich, CAS: 922-67-8)
time, the pH
excitation by hard 45o RF pulse to “trigger” RASER, and iii), signal
acquisition (ACQ) in multiple blocks of 90 each. Below, the
concentrations of pH (dotted blue line), total product ([product]=c , dashed
wine line) and hyperpolarized product ([HP-product]=mzz, solid red line)
2
푏
supply is initiated and upheld for for a period 휏 . ii) optional
dissolved in acetone-d 99.8% (Deutero GmbH, CAS: 666-52-4). Upon
6
hydrogenation, ethyl cinnamate (EC, Figure 1A) or methyl acrylate (MA,
Figure 2A) was formed.
s
2
s
Experimental setup. Experiments were carried out on a 600 MHz
spectrometer (Bruker Avance II) with a cryogenically cooled probe (TCI)
with Q=푣/Δ푣 ≅600.2 MHz/1.2 MHz ≅ 5ꢋꢋ (see SI, Fig. S3) and 5 mm
screw-cap NMR tubes (Wilmad). Tubes were filled with 500 μl of the
2
are plotted qualitatively. Note that pH is immune to excitations with RF
pulses.
2
sample solution. pH was prepared using a home-build liquid nitrogen
generator that provided a 50 % para enrichment. The gas was delivered
into the spectrometer by a 1/16” polytetrafluoroethylene (PTFE) capillary.
A hollow optical fibre was glued with epoxy resin to the end of the capillary
and inserted into the NMR tube and solution (Molex, part. num. 106815-
Acknowledgements
We acknowledge support by the Emmy Noether Program
metabolic and molecular MR” (HO 4604/2-2), the research
“
0026, internal diameter 250 μm and outer diameter 360 μm). A pH
2
training circle “materials for brain” (GRK 2154/1-2019), DFG -
RFBR grant (HO 4604/3-1, № 19-53-12013), the German Federal
Ministry of Education and Research (BMBF) within the framework
of the e:Med research and funding concept (01ZX1915C), the
Cluster of Excellence “precision medicine in inflammation” (PMI
1267). Kiel University and the Medical Faculty are acknowledged
pressure of approx. 1.2 bar (0.2 bar overpressure to atmosphere) was
used to achieve a steady bubbling.
2
Protocol (Scheme 1). pH was supplied (bubbled) into the sample solution
for 휏푏 = 10 s to hydrogenate EP (or MP) and generate PASADENA. After
that, an optional rectangular, 4 μs 45o RF-pulse was applied. All
o
experiments were carried out at 25 C and ambient pressure. During the
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
[
[
23] J.-B. Hövener, S. Bär, J. Leupold, K. Jenne, D. Leibfritz, J.
for supporting the Molecular Imaging North Competence Center
Hennig, S. B. Duckett, D. von Elverfeldt, NMR Biomed. 2013,
(
MOIN CC) as a core facility for imaging in vivo. MOIN CC was
2
6, 124–131.
founded by a grant from the European Regional Development
Fund (ERDF) and the Zukunftsprogramm Wirtschaft of
Schleswig-Holstein (Project no. 122-09-053).
24] B. Feng, A. M. Coffey, R. D. Colon, E. Y. Chekmenev, K. W.
Waddell, Journal of Magnetic Resonance 2012, 214, 258–262.
[25] S. Tam, M. E. Fajardo, Rev. Sci. Inst. 1999, 70, 1926–1932.
[
[
26] S. Wagner, Magn. Reson. Mater. Phy. 2014, 27, 195–199.
27] A. Gamliel, H. Allouche-Arnon, R. Nalbandian, C. M. Barzilay,
J. M. Gomori, R. Katz-Brull, Appl. Magn. Reson. 2010, 39,
329–345.
28] P. M. Richardson, A. J. Parrott, O. Semenova, A. Nordon, S. B.
Duckett, M. E. Halse, Analyst 2018.
29] A. Friebel, E. von Harbou, K. Münnemann, H. Hasse, Ind. Eng.
Chem. Res. 2019, 58, 18125–18133.
30] N. Amor, P. P. Zänker, P. Blümler, F. M. Meise, L. M.
Schreiber, A. Scholz, J. Schmiedeskamp, H. W. Spiess, K.
Münnemann, Journal of Magnetic Resonance 2009, 201, 93–
99.
Keywords: Coherent emission • NMR spectroscopy • para-
hydrogen induced polarization • Magnetic properties • RASER
[
[
[
[
[
1]
2]
T. G. Walker, W. Happer, Rev. Mod. Phys. 1997, 69, 629–642.
J. P. Mugler, T. A. Altes, J. Magn. Reson. Imaging 2013, 37,
3
13–331.
[
[
[
3]
4]
5]
M. G. Richards, B. P. Cowan, M. F. Secca, K. Machin, J. Phys.
B 1988, 21, 665–681.
T. E. Chupp, R. J. Hoare, R. L. Walsworth, B. Wu, Phys. Rev.
Lett. 1994, 72, 2363–2366.
W. Heil, C. Gemmel, S. Karpuk, Y. Sobolev, K. Tullney, F.
Allmendinger, U. Schmidt, M. Burghoff, W. Kilian, S. Knappe‐
Grüneberg, et al., Ann. Phys. 2013, 525, 539–549.
A. Abragam, J. M. Winter, Phys. Rev. Lett. 1958, 1, 374–375.
S. J. Nelson, J. Kurhanewicz, D. B. Vigneron, P. E. Z. Larson,
A. L. Harzstark, M. Ferrone, M. van Criekinge, J. W. Chang, R.
Bok, I. Park, et al., Sci. Transl. Med. 2013, 5, 198ra108-
[
31] L. Buljubasich, M. B. Franzoni, K. Münnemann, Top. Curr.
Chem. 2013, 338, 33–74.
[32] J. Eills, W. Hale, M. Sharma, M. Rossetto, M. H. Levitt, M.
Utz, J. Am. Chem. Soc. 2019, 141, 9955–9963.
[33] C. R. Bowers, D. P. Weitekamp, J. Am. Chem. Soc. 1987, 109,
[
[
6]
7]
5
541–5542.
[34] H. Desvaux, Progress in Nuclear Magnetic Resonance
Spectroscopy 2013, 70, 50–71.
[35] A. N. Pravdivtsev, J.-B. Hövener, Chem. Eur. J. 2019, 25,
659–7668.
[36] A. N. Pravdivtsev, J.-B. Hövener, MOIN Spin Library,
Www.Moincc.De/Method-Development/Mr/Moin-Spin-
1
98ra108.
H.-Y. Chen, Y. Lee, S. Bowen, C. Hilty, J. Magn. Reson. 2011,
08, 204–209.
[
[
[
[
[
8]
9]
7
2
D. J.-Y. Marion, G. Huber, P. Berthault, H. Desvaux,
ChemPhysChem 2008, 9, 1395–1401.
(
10] M. Suefke, S. Lehmkuhl, A. Liebisch, B. Blümich, S. Appelt,
Nat. Phys. 2017, 13, 568–572.
11] M. Suefke, A. Liebisch, B. Blümich, S. Appelt, Nat. Phys.
Library) UKSH, University Kiel, 2018.
[37] A. Abragam, The Principles of Nuclear Magnetism, Clarendon,
1961.
[38] P. Rovedo, S. Knecht, T. Bäumlisberger, A. L. Cremer, S. B.
Duckett, R. E. Mewis, G. G. R. Green, M. Burns, P. J. Rayner,
D. Leibfritz, et al., J. Phys. Chem. B 2016, 120, 5670–5677.
2
015, 11, 767–771.
12] J.-B. Hövener, N. Schwaderlapp, T. Lickert, S. B. Duckett, R.
E. Mewis, L. A. R. Highton, S. M. Kenny, G. G. R. Green, D.
Leibfritz, J. G. Korvink, et al., Nat. Commun. 2013, 4,
ncomms3946.
[
39] J. Jeener, eMagRes 2007, 1–37.
[40] N. Bloembergen, R. V. Pound, Phys. Rev. 1954, 95, 8–12.
[
41] S. Korchak, M. Emondts, S. Mamone, B. Bluemich, S.
[
[
13] R. W. Adams, J. A. Aguilar, K. D. Atkinson, M. J. Cowley, P.
I. P. Elliott, S. B. Duckett, G. G. R. Green, I. G. Khazal, J.
López-Serrano, D. C. Williamson, Science 2009, 323, 1708–
Glöggler, Phys. Chem. Chem. Phys. 2019, DOI
1
0.1039/C9CP05227E.
1
711.
[42] S. Knecht, A. N. Pravdivtsev, J.-B. Hövener, A. V.
14] S. Appelt, A. Kentner, S. Lehmkuhl, B. Blümich, Progress in
Nuclear Magnetic Resonance Spectroscopy 2019, DOI
Yurkovskaya, K. L. Ivanov, RSC Adv. 2016, 6, 24470–24477.
1
0.1016/j.pnmrs.2019.05.003.
[
[
15] A. N. Pravdivtsev, A. V. Yurkovskaya, H.-M. Vieth, K. L.
Ivanov, R. Kaptein, ChemPhysChem 2013, 14, 3327–3331.
16] K. L. Ivanov, A. N. Pravdivtsev, A. V. Yurkovskaya, H.-M.
Vieth, R. Kaptein, Prog. Nucl. Magn. Reson. Spectrosc. 2014,
8
1, 1–36.
[
[
[
17] A. N. Pravdivtsev, K. L. Ivanov, A. V. Yurkovskaya, P. A.
Petrov, H.-H. Limbach, R. Kaptein, H.-M. Vieth, J. Magn.
Reson. 2015, 261, 73–82.
18] D. A. Barskiy, K. V. Kovtunov, I. V. Koptyug, P. He, K. A.
Groome, Q. A. Best, F. Shi, B. M. Goodson, R. V. Shchepin, A.
M. Coffey, et al., J. Am. Chem. Soc. 2014, 136, 3322–3325.
19] A. Svyatova, I. V. Skovpin, N. V. Chukanov, K. V. Kovtunov,
E. Y. Chekmenev, A. N. Pravdivtsev, J.-B. Hövener, I. V.
Koptyug, Chemistry – A European Journal 2019, 25, 8465–
8
470.
[
[
[
20] T. Theis, M. Truong, A. M. Coffey, E. Y. Chekmenev, W. S.
Warren, J. Magn. Reson. 2014, 248, 23–26.
21] A. N. Pravdivtsev, A. V. Yurkovskaya, H.-M. Vieth, K. L.
Ivanov, J. Phys. Chem. B 2015, 119, 13619–13629.
22] R. A. Green, R. W. Adams, S. B. Duckett, R. E. Mewis, D. C.
Williamson, G. G. R. Green, Progress in Nuclear Magnetic
Resonance Spectroscopy 2012, 67, 1–48.
This article is protected by copyright. All rights reserved.

ChemPhysChem
10.1002/cphc.201901056
RESEARCH ARTICLE
RESEARCH ARTICLE
PHIP-RASER. An effect of radiation
damping on hyperpolarized using
para-hydrogen induced polarization
Andrey N. Pravdivtsev*, Frank D.
Sönnichsen and Jan-Bernd Hövener*
(
PHIP) molecules was revealed.
Page No. – Page No.
Interaction between a highly polarized
system and an NMR observation coil
allowed continuous (more than 10
minutes) and coherent (less than 1
ppb line width) emission at 600 MHz
using standard NMR equipment.
Continuous Radio Amplification by
Stimulated Emission using
Parahydrogen Induced Polarization
(
PHIP-RASER) at 14 Tesla
This article is protected by copyright. All rights reserved.