A Sensor Which Can Be Varied in Humidity Sensitivity

A First Experience Paving the Way to New Chemical Sensors?

Giovanni Saggio

, Arnaldo D’Amico

, Vito Errico

, Giovanni Costantini

Giorgio Pennazza

, Alessandro Zompanti

and Marco Santonico

Dept. of Engineering, University of Rome Tor Vergata, Italy

Dept. of Engineering, Campus Bio-Medico University of Rome, Italy

Keywords: Resistive Flex Sensor, Humidity Sensor, Humidity Sensitivity, Sensitivity.

Abstract: During last decades, a number of different sensors have been developing for different analytics to detect. A

key aspect of those sensors is that each of them results with a fixed particular sensitivity. Consequently, at

occurrence, it is necessary to use a plurality of sensors to arrange measures with different levels of sensitivity.

This work intends to investigate the possibility to obtain different sensitivity, in particular with respect to

humidity, from one sensor only. To this aim we investigated the resistive flex sensor, which has been already

used for other applications but, as far as we know, never investigated for its potential properties as a chemical

sensor. Results demonstrated how the resistive flex sensor behaves with different sensitivity values and

different sensitivity curves for different bend conditions.

1 INTRODUCTION

Humidity refers to the amount of water vapour in the

air. It is so important that can affect meaningfully a

number of different aspects of manufacturing

(processes in industries), of building structural

integrities, even of human comfort. Therefore, its

measure can be, somewhat, strategic.

Humidity sensors which have been developed

during last decades mainly rely on thermal, acoustic,

capacitive or resistive effects (Awang, 2014). The

latter, in particular, is commonly based on a comb-

shaped thick metal film, with underneath a polymeric

film, which “furnishes” a different number of ion

current carriers according to the surrounding

humidity.

The choice of the humidity sensor depends on the

specific application, and has to take into account

some parameters related to boundary conditions, such

as size, packaging, cost, interchangeability, and some

others related to the measurement effectiveness, such

as accuracy, repeatability, long-term stability,

recovering time, sensitivity (Chen and Lu, 2005).

Our work regards the investigation of a humidity

sensor which, with respect to the usually adopted

sensors based on the resistive effect, differs in type

and offers a changeable sensitivity.

In particular, we investigated the response to

humidity of a resistive flex sensor (RFS), with the

possibility to change its sensitivity, within a certain

range, as desired.

As far as we know, the RFS has never been

adopted for humidity measurement purposes and, in

addition, it is the first proposal of a humidity sensor

able to change its sensitivity of the measure.

2 MATERIALS AND METHODS

2.1 Resistive Flex Sensors

A resistive flex sensor (Figure 1) converts its

mechanical bending into a proportional electrical

resistance variation. This is due to “isles” of carbon

particles (engineered on top of a plastic support and

drowned in a binder) which increase their distance

with RFS bending, so that the resistance

proportionally increases too. This effect is almost

reversible, so that the RFS furnish its base resistance

when returned to flat conditions (Saggio et al., 2016;

Saggio et al., 2009).

Up to now, RFSs have been used in a number of

different applications.

Saggio, G., D’Amico, A., Err ico, V., Costantini, G., Pennazza, G., Zompanti, A. and Santonico, M.

A Sensor Which Can Be Varied in Humidity Sensitivity - A First Experience Paving the Way to New Chemical Sensors?.

DOI: 10.5220/0006546001130118

In Proceedings of the 11th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2018) - Volume 4: BIOSIGNALS, pages 113-118

ISBN: 978-989-758-279-0

113

(a)

(b)

Figure 1: (a) Top and (b) lateral view of a resistive flex

sensor, manufactured by Flexpoint Inc.

The most investigated application consists in

laying RFSs on parts of human skins, so to measure

flexions/warpings of joints/segments of the human

body. The focus has been for the hand, in measuring

fingers (Gentner and Classen, 2009; Saggio, 2012;

Saggio, 2014), palm (Rossel et al., 2009; Dalley et al.,

2012) and wrist (Howcroft et al., 2011; Kushsairy et

al., 2015; Yu et al., 2016). Anyway, RFSs have been

used for other body parts too, such as neck (Al-

Rahayfeh and Faezipour, 2014), shoulder (Kushsairy

et al., 2015), elbow (Kushsairy et al., 2015), torso

(Saggio et al., 2016), belly (for foetal movements of

pregnant woman) (Borges et al., 2009), ankle

(Mazhar and Bari, 2015; Resendiz et al., 2016), lower

limb (with a deep vein thrombosis cuff) (Qidwai et

al., 2016), and foot (Patil et al., 2016).

Other fields of RFS application involves human-

computer interface (Berlia and Santosh, 2014),

human-machine interface (Asgar et al., 2013),

actuators (Elgeneidy et al., 2016), robots (Mutka et

al., 2014), automotive (Persson, 2002) and, even,

plants (Shanmugam et al., 2016).

Mechanical and biocompatibility characteristics

of RFSs have been already investigated (Saggio,

2012; Saggio et al., 2014), and the interested reader

can find a comprehensive review in (Saggio et al.,

2016).

Our hypothesis concerns the possibility to adopt

RFS as a chemical sensor too, an idea that, as far as

we know, was never investigated. This hypothesis

relies on the fact that mechanical changes with

bending induce physical changes of the “isle” of

carbon particles distribution (Saggio, 2012) that can

induce different chemical interaction with the

surroundings (Saggio et al., 2014), so that the RFS

with different bending can behave differently with

respect to external analytics.

The RFS we adopted is manufactured by

Flexpoint Inc. (Draper, UT, USA) In particular, we

selected the 0.005x0.3x3 inch (thickness, width,

length) type. Three different versions are available,

with overlamination (polyamide or polyester) as a

protective layer on-top of the sensible part but, for our

purposes, we adopted the bare-one, that is, with no

protective layer.

2.2 Electronic Circuitry

We configured the RFS as a variable resistor in the

feedback path of an OpAmp (TL082 by Texas

Instruments), as schematized in Figure 2, so to obtain

a sensitivity changed by the RFS alone, without any

influences due to the electronics (the latter behaving

with a constant sensitivity).

Figure 2: Reading electronics for the resistive flex sensor’s

(R2) outputs. The RFS behaves in feedback of an OpAmp.

Supply was sourced by a constant voltage batteries

(E=1.2V; Vcc=+/-8.4V). We provided a series

limiting current resistor R1 (10KΩ) too.

2.3 Measuring Setup

In order to investigate differences in sensitivity of the

RFS under different bending curvature, we realized

three different mechanical arrangements for the RFS,

fixing it on a planar, 35mm and 20mm in diameter

structures, respectively (Figure 3 shows the case of

20mm in diameter arrangement for the RFS), so to

investigate more and more bending conditions.

(a) (b)

Figure 3: (a) Lateral and (b) side view of the RFS placed on

a 20mm diameter support.

The measuring set-up was realized as shown in Figure

4. The humidity was generated by means of two-

channel mass flow controllers (MFCs) (SLA5850S

by Brooks Instruments), specifically calibrated for

nitrogen, obtaining a 0%-100% of humidity range.

One MFC fluxed different concentration of nitrogen,

the other channel contributed to form the humidity by

-

+

R1

+Vcc

-Vcc

Vout

R2

BIOSIGNALS 2018 - 11th International Conference on Bio-inspired Systems and Signal Processing

114

Figure 4: Set-up to investigate the different sensitivity of

the RFS for three different bending conditions.

bubbling water.

The RFS was placed inside of a climatic chamber,

time by time in a different bending condition as

detailed before, at a constant temperature of 25°C.

Inside the chamber the temperature and humidity was

monitored using a commercial temperature-humidity

sensor (SHT75 by Sensirion), with operating range

0RH%100 and -40°T123.8°, for relative

humidity and temperature, respectively.

3 RESULTS AND COMMENTS

Figure 5 illustrates the sensor’s output, as a dynamic

voltage responses in the case of flat configuration,

when different RH% values are flown into the sensor

cell.

With this example of result, we can worth

remarking three important evidences (obtained in all

the other occurrences):

▪ measurements have been randomly performed;

▪ sensor reproducibility is satisfactory, as shown

by highlighting couples of measurements for

the same values (50% and 100% of RH);

▪ the proportion between 100% and 50%

responses does not account for linearity.

Figure 5: Sequence of dynamic responses in the time

(seconds) domain, registered for the output voltage of the

sensor system (flex resistance in flat configuration +

electronic interface) when different RH% values are flown

into the sensor cell.

Figure 6 summarizes all obtained results, as three

outputs of the OpAmp voltage versus the relative

percentage humidity, Vout vs. RH%, one output for

each of the three mechanical configurations imposed

to the RFS: flat condition or bending conditions

around a 35mm and a 20mm diameter support,

respectively.

The three outputs can be summed by three V

out

vs.

RH% fitting curves as:

▪ flat setting:

V(RH%)

= p1

RH2 + p2

RH + p3

(with

= 0.00028; p2

= -0.055 ; p3

= 4.60,

Figure 5a);

▪ 35mm setting:

V(RH)

35S

= p1

35S

RH + p2

35S

(with p1

35S

= -0.035; p2

35S

= 4.39, Figure 5b);

▪ 20mm setting:

V(RH%)

20S

= p1

20S

RH + p2

20S

(with p1

20S

= -0.033; p2

20S

= 4.72, Figure 5c).

As evidenced, we found a non-linear behaviour,

with a saturation trend for RH%>80%, when RFS was

in flat condition, and almost linear behaviours when

RFS was in both bending conditions. In fact, for

bending conditions we obtained R-square=0.985, for

35mm diameter of the support, and R-square=0.9855,

for the 20mm diameter of the support, respectively.

The sensitivity, expressed as dV/dRH%,

calculated on the fitting curve reported above, was

confirmed to be linked to the physic deformation of

Mass flow

controller 2

Mass flow

controller 1

RH(%)

Sensirion

SHT75

RFS in test chamber

A Sensor Which Can Be Varied in Humidity Sensitivity - A First Experience Paving the Way to New Chemical Sensors?

115

the RFS, since it increases in value with RFS’s

bending.

(a)

(b)

(c)

Figure 6: Vout vs. RH% for RFS in (a) flat, (b) 35mm

diameter-support (R2=0.9852), (c) 20mm diameter-support

bending condition (R2=0.9855).

According to the results, graphically summed in

Figure 7, we can argue that the flat configuration for

RFS furnishes higher selectivity, but for humidity

within 1%-40% RH% range only; differently, the

bending configurations offer a lower sensitivity but it

remains almost linear within the overall RH% range.

In particular, in flat condition the sensitivity can

be estimated by a linear approximation at low RH%

value (less than 10%) as -50mV/%, while in the two

bending conditions we obtained -35mV/% and -

33V/% for the 35mm and 20mm settings,

respectively, within the overall RH% range of

variation.

(d)

Figure 7: Overall sensitivity.

4 CONCLUSIONS

Each humidity sensor has its proper sensitivity, that

is, the slope of its output characteristic curve which

relates sensor’s output with respect to relative

humidity (RH). Different sensitivities are for different

sensors.

Differently, we investigated the possibility to

vary the sensitivity of a unique sensor with changing

its bending.

The idea was to investigate the characteristic of a

resistive flex sensor, which is made of a sensible

carbon-based material, engineered on-top of a plastic

substrate, which can be bent with no damage.

The resistive flex sensor was measured under

different grade of bending, in a climate chamber with

relative humidity which was fully-range varied (0%-

100%).

The results demonstrated how the resistive flex

sensor behaves with different sensitivity and different

output-value versus RH% curves, so that we can

select the optimal bending to obtain the most useful

sensitivity-behaviour, according to specific measure

we are interested in.

As far as we know, this work opens to the

possibility to realize a new type of chemical sensors

which, respect to the currently adopted ones, can offer

the added value to mechanically real-time and on-the-

use change its selectivity.

BIOSIGNALS 2018 - 11th International Conference on Bio-inspired Systems and Signal Processing

116

This approach could be of great interest for

applications needing low-power and reduced

dimension, such as bio-medical applications in

general and in particular wearable solutions. Also

food industry could benefit from this approach when

focused on solutions for smart-packaging oriented to

quality monitoring and shelf-life assessment.

Future works will investigate change of

sensitivity with respect other chemicals (rather than

humidity) and change in selectivity, with respect to

different chemicals, so to evidence the possibility to

paving the way to a new kind of sensor, of which we

can change the sensitivity and the selectivity

according to the time-by-time necessity.

REFERENCES

Al-Rahayfeh, A., Faezipour, M., 2014, August. Application

of head flexion detection for enhancing eye gaze

direction classification. In Engineering in Medicine and

Biology Society (EMBC), 2014 36th Annual

International Conference of the IEEE, pp. 966-969.

Asgar, M., Badra, M., Irshad, K., Aftab, S., 2013.

Automated innovative wheelchair. International

Journal of Information Technology Convergence and

Services, Vol. 3, Issue 6, pp. 1-8.

Awang, Z., 2014. Gas sensors: a review. Sensors &

Transducers, Vol. 168, Issue 4, pp. 61-75.

Berlia, R., Santosh, P., 2014, November. Mouse Brace: A

convenient computer mouse using accelerometer, flex

sensors and microcontroller. In Contemporary

Computing and Informatics (IC3I), 2014 International

Conference on, pp. 558-561.

Borges, L. M., Barroca, N., Velez, F. J., Lebres, A. S., 2009,

April. Smart-clothing wireless flex sensor belt network

for foetal health monitoring. In Pervasive Computing

Technologies for Healthcare, 2009. PervasiveHealth.

3rd International Conference on, pp. 1-4.

Chen, Z., Lu, C., 2005. Humidity sensors: a review of

materials and mechanisms. Sensor letters, Vol. 3, Issue

4, pp. 274-295.

Dalley, S. A., Varol, H. A., Goldfarb, M., 2012. A method

for the control of multigrasp myoelectric prosthetic

hands. IEEE Transactions on Neural Systems and

Rehabilitation Engineering, Vol. 20, Issue 1, pp. 58-67.

Elgeneidy, K., Lohse, N., Jackson, M., 2016. Data-driven

bending angle prediction of soft pneumatic actuators

with embedded flex sensors. IFAC-PapersOnLine, Vol.

49, Issue 21, pp. 513-520.

Gentner, R., Classen, J., 2009. Development and evaluation

of a low-cost sensor glove for assessment of human

finger movements in neurophysiological settings.

Journal of neuroscience methods, Vol. 178, Issue 1, pp.

138-147.

Howcroft, J., Fehlings, D., Zabjek, K., Fay, L., Liang, J.,

Biddiss, E., 2011. Wearable wrist activity monitor as an

indicator of functional hand use in children with

cerebral palsy. Developmental Medicine & Child

Neurology, Vol. 53, Issue 11, pp. 1024-1029.

Kushsairy, A. K., Malik, M. A. A., Zulkhairi, M. Y., Nasir,

H., Khan, S., 2015, November. Real time monitoring

system for upper arms rehabilitation exercise. In Smart

Instrumentation, Measurement and Applications

(ICSIMA), 2015 IEEE 3rd International Conference on,

pp. 1-5.

Mazhar, O., Bari, A. Z., 2015, May. Real-time gait phase

detection using wearable sensors. In Control

Conference (ASCC), 2015 10th Asian, pp. 1-4.

Mutka, A., Kočo, E., Kovačić, Z., 2014. Adaptive control

of quadruped locomotion through variable compliance

of revolute spiral feet. International Journal of

Advanced Robotic Systems, Vol. 11, Issue 10, pp. 1-15.

Patil, J., Nandur, D., Mellikeri, M., Naik, K., Kulkarni, P.,

2016, March. Integrated sensor system for gait analysis.

In Electrical, Electronics, and Optimization Techniques

(ICEEOT), International Conference on, pp. 2298-

2301.

Persson, L., 2002. Bältespåminnelse-system för baksäte:

inventering och utvärdering University West,

Department of Technology (Trollhättan: University

West Trollhättan)

Qidwai, U., Kamran, S., Al-Sulaiti, S., Ahmed, G., Hegazy,

A., 2016, March. Monitoring DVT cuffs for long-term

operation: A fuzzy approach. In Signal Processing &

Its Applications (CSPA), 2016 IEEE 12th International

Colloquium on, pp. 41-45.

Resendiz, A., Odicho, D., Gabrielian, V., Nahapetian, A.,

2016, June. Edemeter: Wearable and continuous fluid

retention monitoring. In Wearable and Implantable

Body Sensor Networks (BSN), 2016 IEEE 13th

International Conference on, pp. 153-158.

Rosell, J., Suarez, R., Rosales, C., García, J. A., Pérez, A.,

2009, May. Motion planning for high DOF

anthropomorphic hands. In Robotics and Automation,

2009. ICRA'09. IEEE International Conference on, pp.

4025-4030.

Saggio, G., 2012. Mechanical model of flex sensors used to

sense finger movements. Sensors and Actuators A:

Physical, Vol. 185, pp. 53-58.

Saggio, G., 2014. A novel array of flex sensors for a

goniometric glove. Sensors and Actuators A: Physical,

Vol. 205, pp. 119-125.

Saggio, G., Bianchi, L., Castelli, S., Santucci, M. B.,

Fraziano, M., Desideri, A., 2014. In vitro analysis of

pyrogenicity and cytotoxicity profiles of flex sensors to

be used to sense human joint postures. Sensors, Vol. 14,

Issue 7, pp. 11672-11681.

Saggio, G., Bocchetti, S., Pinto, C.A., Orengo, G., Giannini,

F., 2009. A novel application method for wearable bend

sensors. 2nd International Symposium on Applied

Sciences in Biomedical and Communication

Technologies, ISABEL, IEEE.

Saggio, G., Riillo, F., Sbernini, L., Quitadamo, L. R., 2015.

Resistive flex sensors: a survey. Smart Materials and

Structures, Vol. 25, Issue 1, pp. 1-30.

Shanmugam, M., Ramasamy, A., Paramasivam, S.,

Prabhakaran, P., 2016. Monitoring the Turmeric Finger

A Sensor Which Can Be Varied in Humidity Sensitivity - A First Experience Paving the Way to New Chemical Sensors?

117

Disease and Growth Characteristics Using Sensor

Based Embedded System—A Novel Method. Circuits

and Systems, Vol. 7, Issue 08, pp. 1280-1296..

Yu, L., Xiong, D., Guo, L., Wang, J., 2016. A remote

quantitative Fugl-Meyer assessment framework for

stroke patients based on wearable sensor networks.

Computer methods and programs in biomedicine, Vol.

128, pp. 100-110.

BIOSIGNALS 2018 - 11th International Conference on Bio-inspired Systems and Signal Processing

118