1.

0 Introduction

Sweat is a very important biofluid that consists of 99% of water and mixes with
the ions such as sodium, potassium, calcium, and others[1][2]. Sweat also
contains metabolites such as glucose, lactate, ethanol, etc[2]. The hormones, proteins, and
peptides that provide a wealth of physiological and metabolic chemical details can be found in
the sweat composition[3]. Most accessible sweat is typically secreted from the eccrine sweat
gland which is almost all over the body surface at densities > 100 gland / cm2 [4]. The sweat
gland is concentrated mainly in the skin of palms, soils, axillae, and forehead for body
temperature control and allowing numerous accessible sampling sites[4]. Sweat analysis is
desirable in non-invasive medical inspections and diagnostics because of its rich composition,
non-invasive availability, and connection to blood chemistry relationships [5][6]. For example,
the chloride content in the sweat can diagnose cystic fibrosis disease[7]. Sweat sodium and
chloride loss is critical for athletes to direct intake strategies and maintain an acceptable balance
between hydration and electrolyte[8][9], while excessive sodium loss in sweat increases the
risk of hyponatremia[10]. Urine acid and sweat creatinine levels can be used to provide early
warning against kidney disease[11][12]. Besides, sweat glucose analysis can be used to
measure blood glucose levels for clinical screening of diabetes[13].

Figure 1:The under-skin structure and the composition inside the sweat.

A wearable sweat sensor consists of two major parts which is sweat collection part and
sensing part. For sweat collection part, it involves the sweat sampling. Conventional sweat
collecting methods include physical exercise, heat or humidity exposure, or
iontophoresis[14][7]. The handling of the sample is very important as it is the first step of the
biofluid analysis. The most common collection materials are absorbent pads or sponges, under
which extraction requires ex-situ examination by laboratory instruments[15]. There are more
sophisticated methods focus on hydrogel[16], paper[17], or textile materials[18] to be used for
in situ measurements in an electrochemical sensing package. For the sensing part,
electrochemical sensors are a very promising technique in which local biological changes in
sweat are continuously tracked due to inherent advantages of miniaturization and not invasive
sample collection. The need for physical contact and continuous research screening is one
challenge in designing wearable biosensors. Chemical sensors are the part used to collect
important details such as the concentration of one substance in a sample by utilizing a
molecular chemical receiver as well as a physicochemical detector feature (transducer)[19].
The bioreceiver configured for the chemical detection device depends on the chosen analyte.
It must be able to produce the analyte concentration with a given sensitivity as a physical or
chemical signal. The transducer must also be chosen according to the bioreceptor and the
measuring technique needed[20]. The type of transducer is typically used as a biosensor
category such as optical, impedance and piezoelectric sensors[21]. The description of the
different components in a biosensor is shown in Figure 2.

Figure 2:The mechanism of the biosensor[20].

Conventional sweat sampling methods have few problems when integrated into
wearable sweat sensor. The problems can be overcome by the recent invention of flexible, skin-
friendly poly (dimethyl siloxane) (PDMS) microfluidic technology. The integration between
biosensor and microfluid chip make the ease of sweat handling, flowing and subsequent for
chemical analysis[22][23]. The challenges of the sweat sensor included contamination of
sample, sample evaporation, low sweat rate, and require large volume of sweat. By exploiting
some of the most complicated approaches, the recent advance wearable devices have potential
to save and store small amounts of sweat in isolated chambers to prevent marginal mixing or
contamination of the sample. Continuous monitoring involves continuous sweat flow
management involves preventing the accumulation of sweat. Past biosensors have a few
drawbacks as they can only track one analyte and have no circuit and sensor configuration for
on-site signal processing. With the coupled of microfluidics and MEMS
(Microelectromechanical systems), the integration of wearable sensor achieved the fluid
processing, biosensing, and incorporation capability of many practical units[24]. Microfluidics
and MEMS have made a significant contribution to the production of local sweat sensing tools
in recent years. Surface tension and capillary strength have been commonly implemented with
gazettes, heavily absorbed pads and well-designed microchannel setup for moving and
collecting sweat[25][26]. By inspired by transpiration in plants[27], some researchers reported
the use of a mixture of capillary and evaporation effects to actively produce a prolonged and
constant stream of sweat, which shows advantages of easy processing, great durability and low
manufacturing costs[28]. However, the sensing and flowing capacities of these sweat collectors
for applications of sweat sensing systems did not integrate completely. The development of the
microfluid device in the biosensor and bioelectronic, especially in the wearable devices become
the new challenges in both engineering and biomedical field. In conclusion, there are several
advantages of using the microfluid technology in the biomedical field because it can monitor
in a small volume of sweat. In addition, a real-time profile of sweat substance secretion is not
available for the separation of sampling from analysis. Thus, the actual tendency addressed
with the integration of wearable sweat sensor by functionalizing and miniaturizing the
materials and devices based on conventional sampling methods in real time and continues
sweat sensing devices analysis.

The main objective of this paper is to review the current methods of sweat analysis and
the challenges of the sweat biosensor devices. The microfluidic systems specifically built for
the accomplishment of the processing of biofluids, biofluid transfer and sensing. The analytes
are primarily handled and then analyzed dynamically because of the accessibility. We will
discuss the methods to perform the sensing (sweat analysis) including, microfluidic-based, low
limit ions detections and dynamic continuous sweat flow to improve the reliability of the sweat
sensors.
2.0 Conventional sweat collection technique

Body-wash-down technique is one of the early methods of sampling for composition

and sweat analysis. The whole body-washdown technique is known as the gold standard to
assess entire body sweat electrolyte loss because all sweat runoff is gathered, and the usual
evaporative sweating cycle is undisturbed [29]. Both devices and the subject are thoroughly
rinsed with deionized water before exercise. Subjects wearing lightweight clothes operate a
bike ergometer in a cardboard isolation case [30]. Body-wash-down technique can avoid
foreign contamination with electrolytes, and it does not interfere with the regular sweating
cycle. The experiment will be carried out by doing exercise, and all the machines and the
apparatus are thoroughly rinsed with deionized water. Then, supply of minimal clothes run in
a cardboard storage tub with a running ergometer. The topics, the tube, the tools, the garments
and any item affected are then thoroughly rinsed again with deionized water to extract all the
electrolytes deposited on the body. Finally, the electrolyte loss in sweat will be examined
before and after exercise[29]. This method allocates the experiments in physiological
conditions of sweat composition and electrolyte loss. Nevertheless, this approach is constrained
by a regulated laboratory environment, complicated steps and single practice testing. This
method is not suitable for field experiments and procedures of treadmills.

Figure 3: The whole-body wash down technique [29].

Next, most history studies used conventional raw clinical techniques as the collection in
bags or textiles of large quantities of sweat. Arm bags or plastic bag were used for collecting
the sweat for analysis by the past researchers [31][32]. The entire arm and hand were covered
entirely by the dry oilskin irrigation envelope and fixed tightly around the top of the arm just
below the axil. Then the sweat was collected into the bags and drained into a drain at the bottom
of the container [31]. This system can ensure that the sweat does not evaporate during the
sweat collection [33]. The device included a sticky rubber on the skin on the lower back and a
sweat-accumulation polypropylene copolymer tube. After some period of use, no sweat
composition was affected or contaminated by all machine components. The arm bag or plastic
bag method for sweat collection was presented and was shown to reduce the evaporation
problem but it required large volume of sample.

Figure 4: The arm and plastic bag in sweat collection [33].

The next technique is by integration of patch in sweat analysis. Patches possessing

hydrophilic and porous structure are usually sandwiched between the skin and an outer
membrane in sweat collection devices. It has obvious advantages of natural abundance, light
weight, flexible structure, low cost and facile surface-modification. Besides, it is extremely
versatile to apply patches in one or more different locations including forearm, neck, back, and
calf for sweat accumulation over hours [27]. Then the collected sweat is carefully weighed,
extracted and diluted for further analysis [28]. Comparing with whole body-washdown
technique, the regional sweat collection by patches generally overestimates the analyte
concentration like [Na+] and [K+]. However, there are significant correlations between the
results of the two methods [25]. Careful skin preparation like skin cleaning before the sweat
collection is necessary to reduce the contamination from cosmetics or bacterial on skin. Despite
the simplicity of this method, during the tedious collecting and processing steps, volume losses
can be as high as 2% per minute [29]. The results are also significantly affected by few factors
such as different sweating regions, individual health level, environmental conditions and
exercise intensity. The limitation of patches is time-consuming to apply when collect the sweat
and easy to detach during exercise. It may also have an hydromeiosis effect on sweat rates due
to sweat pools formation on skin surfaces, suppressed sweat evaporation, and increased local
skin temperatures [30,31].

However, most of the above-mentioned conventional sweat sampling methods involve

only repetitive procedures, large equipment and qualified professionals. These methods also
limited by insufficient sample volumes, non-negligible sample evaporation from the skin
surface, potential analytical degradation between sampling and testing, and uncontrollable
sweating rate and area. Therefore, a real-time profile of sweat substance secretion cannot be
given by isolation of sampling from study. The current trend to be discussed in the next chapter
is therefore developing the materials and miniaturizing devices based on conventional sampling
procedures in real time and continuous sweat sensing devices that integrate sampling and
analysis.

2.2 Recent modification of sweat sensor

There are few limitations and problems in past works such as contamination of
sample(sweat), not continuous flow of sweat, the sensor accuracy and integration issues.
Therefore, several new strategies are proposed to improve the reliability of sweat sensor in
order to make it happen in our real life. The next sub chapter will discuss the recent
modification of sweat sensor by the recent researcher.

2.2.1 Microfluid channel for sweat flow

Recent studies have shown modern, small, soft and skin-compatible microfluidic
systems that allow sweat collection directly from the skin via the pores and transport through
valves, separators and examination reservoirs. The inlets on the cutaneous side of these
microfluidic systems enable the sweat flow activity at different locations of the skin.
Microfluidic channel design removes the moisture from the skin as it rises from the surface to
create the samples for on-board review. The channel construction often improves and secure
sensors from biofouling because of pollutants and oils on the surface of the skin and
environmental contaminants. By utilizing advanced microfluidic technologies and adaptable to
lab-on-a-chip principles, researchers mostly used complicated pump, storage, mixing, and
routing techniques to channel sweat to multiple separate sensory areas to reduce the possibility
of cross-speak and sensing errors [34].

Besides, the skin-interfaced devices utilize conventional methods in the manufacture of

microfluidics. The wearable sweat sensor is designed by comprising entirely of small-module
elastomers that can accommodate easy and non-irritating close displays on the skin instead of
rugged plastics and glass materials. The ideal input and outlet dimension, channel
configuration, adhesive interfaces and substrate selection unique design techniques are
essential to ensure the normal sweat pressure of the eccrine gland is adequate to push moisture
fluid through the system [22]. The schematic adhesive structures designed specifically so that
compensatory sweating effects can be stopped or exacerbated, which will significantly increase
local sweat intake levels. For colorimetric or fluorometric examination, the top layers of the
equipment translucent enough to enable optical inspection [35][36]. The specific microfluidic
designs use clear channels for colorimetric or electrochemical sensing from the inlets through
single reaction reservoirs as shown in figure 5 (a – f) [22][37][38]. Many colorimetric assays
are subject to irreversible colour changes which restrict these instruments to one-time analysis.
Most of the sophisticated microfluidic systems with several reaction chambers segregated using
various passive valve technologies monitor time-based shifts in the concentrations of
biomarker with colorimetric approaches. These architectures allow irreversible chemical
sensing while minimizing cross-discussions among various studies. In addition, the sweat
dynamics are degenerated by instantaneous sweat intensity and complete sweat loss by
measured markers through microfluidic channels.

Figure 5 (a-f): The soft microfluid based chemical sensors [39].

With the development of materials and devices, microfluidics type lab-on-a-chip can
be designed for wearable purposes in thin, soft and "skin-like" format [40]. It can be placed on
nearly every part of the body. These portable epidermal microfluidic instruments use sweat
sequestration and capillary pressure to direct sweat flux from secretions across regulated
channels for continuous spatiotemporal sampling and improve sensing in a well-defined
capsular chamber, reducing external contamination and preventing sweat evaporation [6].
Thus, wearable microfluidic epidermal devices will define not only local sweat chemistry in
freshly secreted sweat but also average and instant sweat loss or levels for real-time or
subsequent analytics [6][36].

2.2.2 Ion selective ion in electrochemical sweat sensors

Various detection systems have different platform and system requirements.

Electrochemical techniques showed more widespread success with point-of-care biofluid
analysis to date [41] while colorimetric, optical and other methods had advantages in certain
applications. That's why the recent work more concentrate on electrochemical sweat sensors.
The electrochemical sensor units consist of two integral parts: a sensing component with
functional electrodes to translate analytical concentrations into electric signals and an
electronic element for easy reading. Electrolytes are of essential significance in sweat analysis.
Electrolyte imbalance can lead in extreme cases to symptoms such as coma, heart attacks and
convulsions. Therefore, electrolytes are the most targeted biomarkers in sweat analysis
research. For example, sodium [53], potassium [48], calcium [54], ammonium [55] and
hydronium ions [56] are cations that are observed through potentiometry approaches. These
sensors typically use polyvinyl chloride (PVC) drop-cast membranes, incorporating selective
ionophores and plasticizers to give the electrodes indicator selectivity as a conformal coating.
Solid-state Ag and AgCl reference electrodes covered with sodium chloride (PVB) polyvinyl
butyral [57] are commonly used. Others depend on chloride-containing, polyvinyl acetate [57]
and polyhydrogel (2-hydroximethyl methacrylate) printable ink [56] [57].

Ion selective electrons (ISE) is a kind of electrochemical potentiometric sensor, which

is depend by a heavy junction, rapid response, miniaturization and integration [42][43]. In
comparison to conventional liquid joint electrodes, solid ISEs may be used in lightweight,
portable ion sensing systems. The hydrophobicity of material (water layer effect) as well as
interface ability (electric double-layer and/or redox capacities) between the transducer and the
ion-selective membrane (ISM) are other significant factors [44][45]. Several characteristics of
sweat sensors are needed be designed to enable effective and precise quantification of the
analytes. The characteristics such as selectivity, flexibility, reliability, reaction time and
reproducibility are important. Hence, the optimization and focus work on ion-selective sensors
is important. Selectivity of the ISE sensor is to identify its target study preferentially in the
presence of certain ions and it is the most basic requirement of a biosensor. In order to achieve
this attribute, ion-selective electrodes (ISEs) use permselectable lipophilic ion exchangers and
ion-selective ionophores[46]. ISE selectivity was explored using ionic tabletop solutions, but
the complexity of sweat means their perforation for in-situ sweat sensing needs to be validated
further [46][47]. When it works within the wearable sensing community, it has shown that the
pH and Na+ selectivity of sweat remain remarkable[48] [49][50]. The sensors focused on redox
detection reactions. A high voltage is needed to be used to trigger the enzyme reaction even
though target-specific enzymes are added. Such voltages may lead to redox transitions and
interference signals in certain electrochemically active compounds in sweat. A mediator for
electron transfer may be applied to resolve this issue to raising the redox potential of the target
reaction [51].

In conclusion, electrochemical sweat sensors and microfluidics are combined in an

enticing interface for efficient sweat sampling. The wearable devices can contact adhesives
with the skin's surface and use flexible printed circuit boards to support packaged electrodes
and related electronic elements. The result will measure currents, potential or impedances at
functionalized electrodes to transfer analyte concentration and on-site molecular real time
monitoring. Figure c shows the lightweight microfluidic sweat sensing patches that interfaced
in a microfluidic channel with a circuit board printed on the electrochemical sensing and sweat
rate analysis in real time [37]. Because sweat was pumped spontaneously via sweat secretion
pressure and capillary action from secretion sites to the microfluid channel, the sensing patch
possessed efficient sweat plating and autonomous analysis of dynamic changes in sweat
compositions. A main criterion for wearable sensors is to increase versatility and comfort while
being immune to the mechanical stresses and distortions associated with the body use. This is
why sensing components are ideally made of soft, deformable substrates that do not delaminate
from the underlying skin. The development methods of electrodes which have proven effective
on flexible substrates include photolithography and printing techniques such as screen printing,
flexography and printing [52]. For the usage of versatile substrate-based sensing devices, it is
important to ensure that usable membrane layers deform without breakup alongside the
substrate, and to avoid any motion-related signal objects from significantly altering the sensor
working region.
3.0 Challenges

Although these new sampling methods considerably improve the wearable device
sensing performance in future applications. The real time, non-invasive sweat monitoring
remains several challenges need to be overcome. Firstly, it is important to sample with
improved sweat levels. Intensive exercise and iontophoresis are currently the main strategies
to promote the production of sweat. Nevertheless, even in the workout, sweat secretion levels
will not reach 20 nL / min / gland [2]. In physical active local intensify of sweating by
iontophoresis is better suited for invalids, but the current pressure of the system must be
carefully regulated because the repetitive application of iontophoretic current to the same site
the harm the skin [58]. In order to enhance sensing efficiency, it would be exciting to create
ultrasensitive sensors that can operate at low sweat levels or explore highly effective biomarker
extraction techniques.

In addition, the contamination from the skin or atmosphere is hard to remove when the
moisture is separated from the gland and skin contact, which has a large impact on the accuracy
of the sensor readings. Simple skin surface cleaning cannot reduce contaminants such as
bacteria, measured at 1010 𝑐𝑐𝑐𝑐2 , and can cause considerable concentration mistakes by
glucose, proteins or cellular metabolites in sweat [58][59]. To manufacture wearable sweat
sensors on a commercial scale, a number of considerations exist including large-scale sensor
manufacturing. Commercial production needs not only a highly viable manufacturing process,
but also control in low cost. In fact, adhesives used in various wearable devices cannot last
longer than a couple of days because of many reasons such as a modified stratum corneum,
skin oils, baths and discomfort. Sensors with long-term reusability and reliability in complex
biological fluid systems that rely on advanced materials and production processes are therefore
urgently needed. In addition, the sensitivity and accuracy of the sensor also one of the
challenges. In some cases, ultrasensitive sensors with preconcentration techniques are needed
and provided that sweat analyte concentrations are often diluted to very low levels.
Contaminant and non-specific binding can also result in a signal shift so that sensor specificity
is important.

Besides that, the wearable sensor needs multiple sweat analyte extensions. Present
sensors are capable of tracking minimal sweat form of electrolytes and metabolites. It is
therefore the future trend in manufacturing wearable sensors that can simultaneously monitor
multiple sweat analytes. The sensor should have the ability of self-calibration. Self-calibration
schemes based on large-scale measurements should be used in wearable devices as the results
of sensors are easily influenced by humidity or temperature in the atmosphere and by human
dietary and sweat collecting regions.
 In conclusion, these challenges suggest that the realistic application of wearable sensors
still has a lot of room and potentials which prompting more researchers to make greater effort
to improve the sweat wearable sensors. The revolutionize physiological diagnosis can be more
customized and accurate health care after addressing these obstacles by using non-invasive
wearable sweating instruments.
4.0 Conclusion

In summary, this paper discussed the development of sweat analysis in past few years.
This paper present and analyse the conventional sampling techniques and adapted sampling
strategies for sweat measurement in depth in wearable devices. The conventional sweat
sampling methods have few limitations such as need repetitive procedures, large equipment
and qualified professionals. The recent technologies have the advantages to solve all these
problems. However, there are a lot of research still need to be done to integrate the sweat
sensing tools in wearable devices.

Extensive work was conducted in recent years to provide a non-invasive, in situ, real-
time analysis of the amounts of sweat samples. The development of microfluid in industrial
technologies and material science has made it possible to combine multiple sensors into a
single, mechanically robust, multiplexing device with on-site signal processing and wireless
data transfer circuitry. These instruments are ideal for better calibrating analytes, which rely
on certain parameters. More specifically, the development of methods to stimulate sweat
production in sedentary conditions is necessary for the expansion of opportunities for athletic
sweat-based health monitoring beyond sweat applications. Iontophoresis has proved a promise
for one-time calculation and allow continuous monitoring. It is important to minimize current
density demands and create highly miniaturized sensors with low secretion ion concentrations
for biomarker extraction.

Finally, a microfluid technology would be addressed on existing problems in the

realistic implementations of wearable sweat sensors with the possible development and future
work directions. Many areas of sweat sensing must be improved further to advance the field.
In addition to the major problems involved with understanding sensor data for predictive
healthcare, system level enhancements would also lead to the precision and utility of sweat
sensor measurements. Further research and development will be improved ways of packaging
sensors in durable and easy-to-wear systems that are less prone to strain or movement noise,
especially at the interface between soft sensing components and stiffened electronics.
Reference:

[1] M. Bariya, H. Y. Y. Nyein, and A. Javey, “Wearable sweat sensors,” Nat. Electron.,
vol. 1, no. 3, pp. 160–171, 2018, doi: 10.1038/s41928-018-0043-y.
[2] Z. Sonner et al., “The microfluidics of the eccrine sweat gland, including biomarker
partitioning, transport, and biosensing implications,” Biomicrofluidics, vol. 9, no. 3,
2015, doi: 10.1063/1.4921039.
[3] Y. Yang and W. Gao, “Wearable and flexible electronics for continuous molecular
monitoring,” Chem. Soc. Rev., vol. 48, no. 6, pp. 1465–1491, 2019, doi:
10.1039/c7cs00730b.
[4] K. Sato, “The physiology, pharmacology, and biochemistry of the eccrine sweat
gland.,” Rev. Physiol. Biochem. Pharmacol., vol. 79, pp. 51–131, 1977, doi:
10.1007/bfb0037089.
[5] J. Kim, A. S. Campbell, B. E. F. de Ávila, and J. Wang, “Wearable biosensors for
healthcare monitoring,” Nat. Biotechnol., vol. 37, no. 4, pp. 389–406, 2019, doi:
10.1038/s41587-019-0045-y.
[6] J. Choi, R. Ghaffari, L. B. Baker, and J. A. Rogers, “Skin-interfaced systems for sweat
collection and analytics,” Sci. Adv., vol. 4, no. 2, pp. 1–10, 2018, doi:
10.1126/sciadv.aar3921.
[7] S. Emaminejad et al., “Autonomous sweat extraction and analysis applied to cystic
fibrosis and glucose monitoring using a fully integrated wearable platform,” Proc.
Natl. Acad. Sci. U. S. A., vol. 114, no. 18, pp. 4625–4630, 2017, doi:
10.1073/pnas.1701740114.
[8] R. J. Maughan and S. M. Shirreffs, “Development of individual hydration strategies
for athletes,” Int. J. Sport Nutr. Exerc. Metab., vol. 18, no. 5, pp. 457–472, 2008, doi:
10.1123/ijsnem.18.5.457.
[9] M. N. Sawka, L. M. Burke, E. R. Eichner, R. J. Maughan, S. J. Montain, and N. S.
Stachenfeld, “Exercise and fluid replacement,” Med. Sci. Sports Exerc., vol. 39, no. 2,
pp. 377–390, 2007, doi: 10.1249/mss.0b013e31802ca597.
[10] S. J. Montain, S. N. Cheuvront, and M. N. Sawka, “Exercise associated
hyponatraemia: Quantitative analysis to understand the aetiology,” Br. J. Sports Med.,
vol. 40, no. 2, pp. 98–105, 2006, doi: 10.1136/bjsm.2005.018481.
[11] Y. Y. Al-Tamer, E. A. Hadi, and I. E. I. Al-Badrani, “Sweat urea, uric acid and
creatinine concentrations in uraemic patients,” Urol. Res., vol. 25, no. 5, pp. 337–340,
1997, doi: 10.1007/BF01294662.
[12] K. Kohagura et al., “An association between uric acid levels and renal arteriolopathy
in chronic kidney disease: A biopsy-based study,” Hypertens. Res., vol. 36, no. 1, pp.
43–49, 2013, doi: 10.1038/hr.2012.135.
[13] J. Kim, A. S. Campbell, and J. Wang, “Wearable non-invasive epidermal glucose
sensors: A review,” Talanta, vol. 177, no. July 2017, pp. 163–170, 2018, doi:
10.1016/j.talanta.2017.08.077.
[14] S. B. Kim et al., “Soft, Skin-Interfaced Microfluidic Systems with Wireless, Battery-
Free Electronics for Digital, Real-Time Tracking of Sweat Loss and Electrolyte
Composition,” Small, vol. 14, no. 45, pp. 1–9, 2018, doi: 10.1002/smll.201802876.
[15] K. Mitsubayashi, M. Suzuki, E. Tamiya, and I. Karube, “Analysis of metabolites in
sweat as a measure of physical condition,” Anal. Chim. Acta, vol. 289, no. 1, pp. 27–
34, 1994, doi: 10.1016/0003-2670(94)80004-9.
[16] J. Kim et al., “Wearable smart sensor systems integrated on soft contact lenses for
wireless ocular diagnostics,” Nat. Commun., vol. 8, 2017, doi: 10.1038/ncomms14997.
[17] J. Kim et al., “Wearable temporary tattoo sensor for real-time trace metal monitoring
 in human sweat,” Electrochem. commun., vol. 51, pp. 41–45, 2015, doi:
10.1016/j.elecom.2014.11.024.
[18] W. Jia et al., “Electrochemical tattoo biosensors for real-time noninvasive lactate
monitoring in human perspiration,” Anal. Chem., vol. 85, no. 14, pp. 6553–6560, 2013,
doi: 10.1021/ac401573r.
[19] F. W. Scheller, R. Hintsche, D. Pfeiffer, F. Schubert, K. Riedel, and R. Kindervater,
“Biosensors: Fundamentals, applications and trends,” Sensors Actuators B. Chem., vol.
4, no. 1–2, pp. 197–206, 1991, doi: 10.1016/0925-4005(91)80198-S.
[20] D. R. Thévenot, K. Toth, R. A. Durst, and G. S. Wilson, “Electrochemical biosensors:
Recommended definitions and classification,” Biosens. Bioelectron., vol. 16, no. 1–2,
pp. 121–131, 2001, doi: 10.1016/S0956-5663(01)00115-4.
[21] S. Vigneshvar, C. C. Sudhakumari, B. Senthilkumaran, and H. Prakash, “Recent
advances in biosensor technology for potential applications - an overview,” Front.
Bioeng. Biotechnol., vol. 4, no. FEB, pp. 1–9, 2016, doi: 10.3389/fbioe.2016.00011.
[22] S. Sugama et al., “A soft, wearable microfluidic device for the capture, storage, and
colorimetric sensing of sweat Ahyeon,” Sci. Transl. Med., vol. 26, no. 4, pp. 39–46,
2017, doi: 10.1016/j.bbi.2015.08.015.Chronic.
[23] J. Choi, D. Kang, S. Han, S. B. Kim, and J. A. Rogers, “Thin, Soft, Skin-Mounted
Microfluidic Networks with Capillary Bursting Valves for Chrono-Sampling of
Sweat,” Adv. Healthc. Mater., vol. 6, no. 5, Mar. 2017, doi: 10.1002/adhm.201601355.
[24] T. Kaya et al., “Wearable Sweat Sensors: Background and Current Trends,”
Electroanalysis, vol. 31, no. 3, pp. 411–421, 2019, doi: 10.1002/elan.201800677.
[25] E. Berthier and D. J. Beebe, “Flow rate analysis of a surface tension driven passive
micropump,” Lab Chip, vol. 7, no. 11, pp. 1475–1478, 2007, doi: 10.1039/b707637a.
[26] Y. X. Guan, Z. R. Xu, J. Dai, and Z. L. Fang, “The use of a micropump based on
capillary and evaporation effects in a microfluidic flow injection chemiluminescence
system,” Talanta, vol. 68, no. 4, pp. 1384–1389, 2006, doi:
10.1016/j.talanta.2005.08.021.
[27] J. M. Li et al., “A bio-inspired micropump based on stomatal transpiration in plants,”
Lab Chip, vol. 11, no. 16, pp. 2785–2789, 2011, doi: 10.1039/c1lc20313d.
[28] Z. R. Xu, C. H. Zhong, Y. X. Guan, X. W. Chen, J. H. Wang, and Z. L. Fang, “A
microfluidic flow injection system for DNA assay with fluids driven by an on-chip
integrated pump based on capillary and evaporation effects,” Lab Chip, vol. 8, no. 10,
pp. 1658–1663, 2008, doi: 10.1039/b805774e.
[29] L. B. Baker, J. R. Stofan, A. A. Hamilton, and C. A. Horswill, “Comparison of
regional patch collection vs. whole body washdown for measuring sweat sodium and
potassium loss during exercise,” J. Appl. Physiol., vol. 107, no. 3, pp. 887–895, 2009,
doi: 10.1152/japplphysiol.00197.2009.
[30] A. Montagu, “Whole body sweat collection in humans: an improved method with
preliminary data on electrolyte content,” JAMA J. Am. Med. Assoc., vol. 179, no. 11, p.
887, 1962, doi: 10.1001/jama.1962.03050110055011.
[31] J. S. Weiner, “A COMPARISON OF ARM-BAG SWEAT AND BODY SWEAT BY
RUTH VAN HEYNINGEN AND J . S . WEINER From the Medical Research
Council Climatic and Working Efficiency Unit , Oxford In preliminary experiments
designed to determine the changes in the com- position and rat,” no. 1943, pp. 395–
403, 1951.
[32] B. Y. W. S. S. Ladell, “THE MEASUREMENT OF CHLORIDE LOSSES IN THE
SWEAT,” 1947.
[33] B. M. R. Appenzeller, C. Schummer, S. B. Rodrigues, and R. Wennig, “Determination
of the volume of sweat accumulated in a sweat-patch using sodium and potassium as
 internal reference,” J. Chromatogr. B Anal. Technol. Biomed. Life Sci., vol. 852, no. 1–
2, pp. 333–337, 2007, doi: 10.1016/j.jchromb.2007.01.037.
[34] P. N. Nge, C. I. Rogers, and A. T. Woolley, “Advances in Microfluidic Materials ,
Functions , Integration , and Applications,” 2013.
[35] S. B. Kim et al., “Super-Absorbent Polymer Valves and Colorimetric Chemistries for
Time-Sequenced Discrete Sampling and Chloride Analysis of Sweat via Skin-Mounted
Soft Microfluidics,” Small, vol. 14, no. 12, pp. 1–11, 2018, doi:
10.1002/smll.201703334.
[36] Y. Sekine et al., “A fluorometric skin-interfaced microfluidic device and smartphone
imaging module for: In situ quantitative analysis of sweat chemistry,” Lab Chip, vol.
18, no. 15, pp. 2178–2186, 2018, doi: 10.1039/c8lc00530c.
[37] H. Y. Y. Nyein et al., “A Wearable Microfluidic Sensing Patch for Dynamic Sweat
Secretion Analysis,” ACS Sensors, vol. 3, no. 5, pp. 944–952, 2018, doi:
10.1021/acssensors.7b00961.
[38] A. Martín et al., “Epidermal Microfluidic Electrochemical Detection System:
Enhanced Sweat Sampling and Metabolite Detection,” ACS Sensors, vol. 2, no. 12, pp.
1860–1868, 2017, doi: 10.1021/acssensors.7b00729.
[39] A. J. Bandodkar, W. J. Jeang, R. Ghaffari, and J. A. Rogers, “Wearable Sensors for
Biochemical Sweat Analysis,” Annu. Rev. Anal. Chem., vol. 12, no. 1, pp. 1–22, 2019,
doi: 10.1146/annurev-anchem-061318-114910.
[40] Y. Yu, H. Yin, Y. Nyein, W. Gao, and A. Javey, “Flexible Electrochemical
Bioelectronics : The Rise of In Situ Bioanalysis,” vol. 1902083, pp. 1–25, 2019, doi:
10.1002/adma.201902083.
[41] A. J. Bandodkar and J. Wang, “Non-invasive wearable electrochemical sensors: A
review,” Trends Biotechnol., vol. 32, no. 7, pp. 363–371, 2014, doi:
10.1016/j.tibtech.2014.04.005.
[42] A. Cadogan et al., “All-Solid-State Sodium-Selective Electrode Based on a Calixarene
Ionophore in a Poly(vinyl chloride) Membrane with a Polypyrrole Solid Contact,”
Anal. Chem., vol. 64, no. 21, pp. 2496–2501, 1992, doi: 10.1021/ac00045a007.
[43] M. Cuartero and G. A. Crespo, “All-solid-state potentiometric sensors: A new wave
for in situ aquatic research,” Curr. Opin. Electrochem., vol. 10, pp. 98–106, 2018, doi:
10.1016/j.coelec.2018.04.004.
[44] J. Janata, M. Josowicz, P. Vanýsek, and D. M. DeVaney, “Chemical Sensors,” Anal.
Chem., vol. 70, no. 12, pp. 179–208, 1998, doi: 10.1021/a1980010w.
[45] M. Bamsey, T. Graham, C. Thompson, A. Berinstain, A. Scott, and M. Dixon, “Ion-
specific nutrient management in closed systems: The necessity for ion-selective
sensors in terrestrial and space-based agriculture and water management systems,”
Sensors (Switzerland), vol. 12, no. 10, pp. 13349–13392, 2012, doi:
10.3390/s121013349.
[46] J. M. Zook, J. Langmaier, and E. Lindner, “Current-polarized ion-selective
membranes: The influence of plasticizer and lipophilic background electrolyte on
concentration profiles, resistance, and voltage transients,” Sensors Actuators, B Chem.,
vol. 136, no. 2, pp. 410–418, 2009, doi: 10.1016/j.snb.2008.12.047.
[47] M. Telting-Diaz and E. Bakker, “Effect of lipophilic ion-exchanger leaching on the
detection limit of carrier-based ion-selective electrodes,” Anal. Chem., vol. 73, no. 22,
pp. 5582–5589, 2001, doi: 10.1021/ac010526h.
[48] W. Gao et al., “Fully integrated wearable sensor arrays for multiplexed in situ
perspiration analysis,” Nature, vol. 529, no. 7587, pp. 509–514, 2016, doi:
10.1038/nature16521.
[49] B. Schazmann et al., “A wearable electrochemical sensor for the real-time
 measurement of sweat sodium concentration,” Anal. Methods, vol. 2, no. 4, pp. 342–
348, 2010, doi: 10.1039/b9ay00184k.
[50] S. Anastasova et al., “A wearable multisensing patch for continuous sweat
monitoring,” Biosens. Bioelectron., vol. 93, no. June 2016, pp. 139–145, 2017, doi:
10.1016/j.bios.2016.09.038.
[51] M. Mehrvar and M. Abdi, “Recent developments, characteristics, and potential
applications of electrochemical biosensors,” Anal. Sci., vol. 20, no. 8, pp. 1113–1126,
2004, doi: 10.2116/analsci.20.1113.
[52] G. Grau, J. Cen, H. Kang, R. Kitsomboonloha, W. J. Scheideler, and V. Subramanian,
“Gravure-printed electronics: Recent progress in tooling development, understanding
of printing physics, and realization of printed devices,” Flex. Print. Electron., vol. 1,
no. 2, pp. 1–23, 2016, doi: 10.1088/2058-8585/1/2/023002.
[53] A. J. Bandodkar et al., “Epidermal tattoo potentiometric sodium sensors with wireless
signal transduction for continuous non-invasive sweat monitoring,” Biosens.
Bioelectron., vol. 54, pp. 603–609, 2014, doi: 10.1016/j.bios.2013.11.039.
[54] H. Y. Y. Nyein et al., “A Wearable Electrochemical Platform for Noninvasive
Simultaneous Monitoring of Ca2+ and pH,” ACS Nano, vol. 10, no. 7, pp. 7216–7224,
2016, doi: 10.1021/acsnano.6b04005.
[55] T. Guinovart, A. J. Bandodkar, J. R. Windmiller, F. J. Andrade, and J. Wang, “A
potentiometric tattoo sensor for monitoring ammonium in sweat,” Analyst, vol. 138,
no. 22, pp. 7031–7038, 2013, doi: 10.1039/c3an01672b.
[56] V. F. Curto et al., “Real-time sweat pH monitoring based on a wearable chemical
barcode micro-fluidic platform incorporating ionic liquids,” Sensors Actuators, B
Chem., vol. 171–172, pp. 1327–1334, 2012, doi: 10.1016/j.snb.2012.06.048.
[57] M. Parrilla, J. Ferré, T. Guinovart, and F. J. Andrade, “Wearable Potentiometric
Sensors Based on Commercial Carbon Fibres for Monitoring Sodium in Sweat,”
Electroanalysis, vol. 28, no. 6, pp. 1267–1275, 2016, doi: 10.1002/elan.201600070.
[58] M. R. Prausnitz, “The effects of electric current applied to skin: A review for
transdermal drug delivery,” Adv. Drug Deliv. Rev., vol. 18, no. 3, pp. 395–425, 1996,
doi: 10.1016/0169-409X(95)00081-H.
[59] T. C. Boysen, S. Yanagawa, F. Sato, and K. Sato, “A modified anaerobic method of
sweat collection,” J. Appl. Physiol. Respir. Environ. Exerc. Physiol., vol. 56, no. 5, pp.
1302–1307, 1984, doi: 10.1152/jappl.1984.56.5.1302.