water passivated in the hydrogel damper that
is laminated on the different skin spots (chest,
neck, wrist, and forehead) does not change
significantly (figs. S9 and S11). First, to sepa-
rate target mechanical signals from low-
frequency dynamic noise, we assembled the
hydrogel damper with an ultrasensitive crack-
based strain sensor (Fig. 1I and fig. S12) ( 27 , 28 ).
One example would be to acquire speech
information without dynamic noises deployed
ontheneck(Fig.4A).Thefrequencyofneck
vibration during speech is usually at least 100 Hz,
whereas physical noise arises from motions such
as swallowing and skin tension due to move-
ment, which usually has frequencies under
30 Hz (Fig. 4B) ( 22 , 23 ). A male volunteer spoke
“Viva la vida”after attaching the integrated
sensor to his neck (Fig. 4C, yellow arrows),
and then the volunteer performed a swallow-
ing motion to create dynamic noise (Fig. 4C,
black arrows). Even if the sensor has low-
frequency dynamic noise induced by the tension
of the skin or movement, it contributes over
awidefrequencyrangebecausethenoisecan
create various scales of friction ( 29 ). A Morelet
wavelet transforms analysis shows that using
the control sensors with PDMS and bandpass
filtering with 80 to 200 Hz, the vocal signals
and swallowing motion could barely be iden-
tified (Fig. 4C, fig. S15, and movie S3). By con-
trast, the spectrogram of the sensor with the
hydrogel damper shows a relatively clear sepa-
ration of acoustic signals from the swallowing
motion signal; even at 40°C with the inte-
grated heater, entire noise signals due to mo-
tion disappear (fig. S16). The reference, even
with the bandpass filtering at 80 to 200 Hz,
shows a broad frequency range with the lowest
signal-to-noise ratio (SNR), which means that
the target signal acquisition is inaccurate,
making it hard to discriminate signal from
noise (Fig. 4D). With the hydrogel damper,
the peak frequency ranges become narrow,
and the SNR increases with the temperature
(fig. S17). Another example of acquisition, heart
rate signals (0.3 to 4 Hz) on the wrist under
the lifting move or tapping (<0.4 Hz), supports
that the selective mechanical signal target can
be exploited regardless of the mechanical
noises (fig. S18).
Because the hydrogel damper can be used as
a conductive electrode as well as a damping
material, we next explored whether it can be
used to obtain electrophysiological signals
(ECG and EEG) (Fig. 1H and fig. S13). The
hydrogel can be expected to transmit the electro-
physiology while damping the external me-
chanical noises. With the lower impedance
of the hydrogel (fig. S13), one demonstration
involves ECG signal detection (0.05 to 150 Hz
range) measurement by attaching each electrode
to the chest of a volunteer (Fig. 4E) while the
electrode is tapped (7 to 8 Hz) and the parti-
cipant is breathing and walking (1 to 2 Hz)
(Fig.4F).Whentherepresentativesignals
before and after mechanical noise application
were analyzed, the mechanical noise obscured
the ECG waveforms, whereas the ECG signal
was detected in the hydrogel case regardless
of mechanical noise (Fig. 4G). Even band-
pass filtering of 0.5 to 40 Hz and additional
arbitrary bandstop filtering were ineffective
in each case because the noise frequency was
within the frequency ranges of the target sig-
nals. By contrast, the hydrogel damper results
show representative signals that were stable
and clear regardless of external mechanical
noise (fig. S19). During continuous biophys-
iology detection over a full day, the mechanical
noise-damping capability was not degraded
(fig. S20).
We also measured EEG alpha wave signals
(8 to 12 Hz) when the eyes are closing by at-
taching the electrodes to a volunteer’s fore-
head, together with an actuator that generates
10 Hz mechanical noise (Fig. 4, H and I). The
wavelet spectrogram shows that both the
hydrogel damper and commercial electrodes
detect electrical signals at ~9 to 12 Hz when
the volunteer closes their eyes (Fig. 4J). When
the eyes are open, in which case no signals
should appear at 9 to 12 Hz, continuous 10-Hz
noise was detected in the commercial electrode
cases, and even bandstop filtering of 10 Hz
was inefficient because it distorted the signals.
However, the hydrogel damper damps the
10-Hz noise selectively, and intermittent signals
at 9 to 12 Hz are detected by the hydrogel dam-
per during natural eye closing, indicating that
stable electrical signal acquisition is achieved
regardless of the noise (fig. S21).
Selective frequency damping with the visco-
elastic material minimizes mechanical noise
and enables the detection of biophysiological
signals with a high SNR under noisy condi-
tions. Rather than signal processing after the
interruption by the mechanical noises, selec-
tive frequency damping by the material itself
would be more effective to acquire clear sig-
nals. The strategy for the engineering of relax-
ation time on demand becomes essential under
the various environmental conditions and needs
to be further improved in the future work—as
well as the temperature controls, which may
require additional form factors in whole de-
vices, and the skin can suffer from the tem-
perature controls. In this way, we propose that
the real-time application of soft bioelectronics
that does not require a signal-processing step
can be accelerated with viscoelastic soft mate-
rials compared with that of rigid wearable
electronics.REFERENCES AND NOTES- D.-H. Kimet al.,Science 333 , 838–843 (2011).
- S. Leeet al.,Science 370 , 966–970 (2020).
- J. Leeet al.,Nat. Electron. 4 , 291–301 (2021).
- T.-i. Kimet al.,Science 340 , 211–216 (2013).
5. J. Denget al.,Nat. Mater. 20 , 229–236 (2021).
6. C. S. Bolandet al.,Science 354 , 1257–1260 (2016).
7. E. Song, J. Li, S. M. Won, W. Bai, J. A. Rogers,Nat. Mater. 19 ,
590 – 603 (2020).
8. S. P. Lacour, G. Courtine, J. Guck,Nat. Rev. Mater. 1 , 16063
(2016).
9. M. Baumgartneret al.,Nat. Mater. 19 , 1102–1109 (2020).
10. O. A. Araromiet al.,Nature 587 , 219–224 (2020).
11. A. Pena-Francesch, H. Jung, M. C. Demirel, M. Sitti,Nat. Mater.
19 , 1230–1235 (2020).
12. I. Youet al.,Science 370 , 961–965 (2020).
13. B. C.-K. Teeet al.,Science 350 , 313–316 (2015).
14. K. Leeet al.,Nat. Biomed. Eng. 4 , 148–158 (2020).
15. C. J. De Luca, L. D. Gilmore, M. Kuznetsov, S. H. Roy,
J. Biomech. 43 , 1573–1579 (2010).
16. F. G. Barth,A SpiderÕs World: Senses and Behavior(Springer,
2002).
17. S. L. Younget al.,Acta Biomater. 10 , 4832–4842 (2014).
18. J. F. Vincent,Compos., Part A Appl. Sci. Manuf. 33 , 1311– 1315
(2002).
19. L. Xuet al.,Sci. Adv. 5 , eaau3442 (2019).
20. K. Liuet al.,J. Am. Chem. Soc. 143 , 1162–1170 (2021).
21. E. K. Antonsson, R. W. Mann,J. Biomech. 18 , 39– 47
(1985).
22. K. L. Venkatachalam, J. E. Herbrandson, S. J. Asirvatham,
Circ. Arrhythm. Electrophysiol. 4 , 965–973 (2011).
23. S. Brage, N. Brage, P. W. Franks, U. Ekelund, N. J. Wareham,
Eur. J. Clin. Nutr. 59 , 561–570 (2005).
24. T. McLeish,Adv. Phys. 51 , 1379–1527 (2002).
25. M. Tanget al.,ACS Omega 2 , 2214–2223 (2017).
26. J. A. Plaizier-Vercammen, E. Lecluse, P. Boute, R. E. De Neve,
Dent. Mater. 5 , 301–305 (1989).
27. D. Kanget al.,Nature 516 , 222–226 (2014).
28. B. Parket al.,Adv. Mater. 28 , 8130–8137 (2016).
29. T. Cristianiet al.,Tribol. Lett. 66 , 149 (2018).
30. A. P. Unwinet al.,Sci. Rep. 8 , 2454 (2018).
ACKNOWLEDGMENTS
The authors thank S. J. Kwon (SKKU) for helpful discussion on the
theoretical modeling; D. S. Hwang (Postech) for the chitosan
hydrogel; D. Kim (SKKU) for the high-speed camera; and
H.-J. Chung (University of Alberta), H. Lee (KAIST), B. Jin, and
S. Jo for helpful discussion on the hydrogel.Funding:This
work was supported by National Research Foundation of
Korea (NRF) grants funded by the Korean government (MSIT)
(NRF-2019R1I1A2A01061966, NRF-2020M3C1B8016137, and
NRF-2019M3C7A1032076) and by the Technology Innovation
Program (20013794, Center for Composite Materials and
Concurrent Design) funded by the Ministry of Trade, Industry &
Energy (MOTIE, Korea). B.P. was partially supported by Korea
Institute for Advancement of Technology (KIAT) grant funded by
the Korea Government (MOTIE) [P0017305, Human Resource
Development Program for Industrial Innovation (Global)].Ethics
statement:All human subject experiments conducted with the
volunteers were preapproved by the human subject review board
of Sungkyunkwan University (ISKKU 2019-06-015).Author
contributions:B.P. and T.-i.K. designed the experiments and
led this work. B.P. and S.P. synthesized the hydrogel damper. B.P.,
J.O., and S.P. investigated the mechanical properties of the
hydrogel damper. B.P., W.J., C.J., S.C., and Y.J.J. investigated
the properties of gelatin and chitosan. B.P. and J.H.S. implemented
the detection of biophysiological signals with the hydrogel damper.
All authors discussed and shared the results and ideas. B.P.
and T.-i.K. wrote the paper.Competing interests:T.-i.K, B.P., and
S.P. are inventers on a patent application related to this work
(KR 10-2020-0061065, 12 May 2021). The authors declare no other
competing interests.Data and materials availability:All data are
available in the main text or the supplementary materials.SUPPLEMENTARY MATERIALS
science.org/doi/10.1126/science.abj9912
Materials and Methods
Supplementary Text
Figs. S1 to S21
Table S1 and S2
References ( 31 – 43 )
Movies S1 to S4Submitted 16 June 2021; accepted 31 March 2022
10.1126/science.abj9912SCIENCEscience.org 6 MAY 2022•VOL 376 ISSUE 6593 629
RESEARCH | REPORTS