Introduction

It can be expected that the progress in development of RFID technology and integrated circuits (IC’s), in particular micro–electro–mechanical system (MEMS) makes it possible to combine them to perform wireless physico-chemical property measurements using devices on the centimeter scale. These devices can be made increasingly smaller. An important property of interest is the viscosity. Polymer solutions around 1000 [ppm] are more viscous and therefore more efficient and require less energy or invested exergy than water to recover oil (Hassan and Bruining et al. 2018) thanks to faster recovery avoiding an ERoEI below one, i.e., low exergy return on high exergy investment, which frequently occurs towards the end of a recovery process. It is expected that the opportunities for EOR-polymer (polymer-based enhanced oil recovery) grow in the future (Forst and Sullivan 2012) avoiding an exergy return on exergy investment or ERoEI (recovered exergy divided by invested exergy) below one at the end of the project, as obtained with polymer free water injection. However, cost and environmental impact require a ERoEI well above one. Natural polymers based on plant products (Avwioroko et al. 2014) can be an alternative for synthetic polymers. In particular they have a lower environmental footprint because of their biodegradability (Hindi et al. 2017a, b). Polymers, in particular natural-polymers (e.g., Guar–Arabic gum) are susceptible to microbial degradation. It is therefore important to monitor its quality at the injection and production side for real-time quality control. One candidate polymer applicable in enhanced oil recovery is Guar–Arabic gum, which is available in large quantities in India and Sudan. In a separate contribution, it’s showed with an exergy analysis that the use of Guar–Arabic gum leads to more efficient oil recovery, which requires less exergy for its production (Hassan and Bruining et al. 2018). In recent years, the combination of RFID technology with sensors has been used in applications to measure a wide range of environmental parameters including temperature (Opasjumruskit et al. 2006; Shenghua and Nanjian 2007; Vaz et al. 2010; Law et al. 2010; Virtanen et al. 2011), pressure (DeHennis and Wise 2002; Mori et al. 2004; Beriain et al. 2012), humidity (Jia et al. 2008; Virtanen et al. 2010, 2011; Amin et al. 2014) and chemical compositions (Potyrailo et al. 2012; Fiddes and Yan 2013). RFID tags have benefited from recent innovative research. Novak in his thesis Novak (2009) describes the use of the RFID tag as chemical sensor for trace substances. Ong et al. (2001) show that RFID technology can monitor the real and imaginary part of the dielectric coefficient of the medium surrounding the tag. Grimes et al. (2002) give an overview of sensors based on remote resonance frequency detection, which makes it possible to measure a variety of physical parameters. Only very recently the possibility of using RFID technology for determining dielectric coefficients, has been completely worked out by Hassan and Bruining Hassan et al. (2017). Humberto Lobato-Morales and his coauthors (Lobato-Morales et al. 2011, 2013) use a substrate-integrated-waveguide (SIW) resonant cavity and a tunnel sensor allowing the determination of the complex permittivity. The complex dielectric coefficient depends on the frequency; measurements at various frequencies therefore expand the capability using RFID’s for measuring quantities that depend on more than two parameters (Warnagiris 2000). The paper of Karappuswami and his coauthors. Karuppuswami et al. (2016) is recently published and describes different sensor designs for measurement of dielectric properties via the response functions under far field conditions. The paper reports possible applications to fluid compositions, i.e., isopropyl alcohol–water and methanol–water mixtures. Karappuswami’s research, which has been carried out in parallel with our research shows some overlap with the results described in this paper. Bonnier et al. (2013), present the development of an in line viscometer that can measure the viscosity at low shear rate in the field. The equipment consists of a pump that circulates the solution through a tube of given length and diameter. The pressure drop is measured, which can be used based on calibration data to determine the viscosity in line. Miller et al. (2011) describe a “Tuning Fork Instrument”, which is capable of in line measurement of the density and viscosity of drilling fluids at 10-s intervals, but point out that the development is still at the early stage in particular as to the analysis of the readings. The advantage of using sensors is in their compactness, which facilitates its deployment at remote locations. The new aspect described in this paper is that it makes the applicability plausible of (relatively) small acoustic wave sensors (AWS) for real-time viscosity measurement, in combination with cutting edge wireless RFID technology to allow wireless and battery-less real-time monitoring operation of injected and production fluids (i.e., environmentally acceptable natural-polymer solutions) during enhanced oil recovery (EOR) processes. The advantage of using these sensors lies in its small size, wireless communication and its expected developments for using ever smaller size sensors and its expected development to affordable extensions as to fluids (i.e., Guar–Arabic gum) composition measurement. The organization of the paper is as follows: in second section, we describe the state of the art SeGenuity ViSmart\(\textcircled {R}\) acoustic wave viscosity sensor (AWS), which is a solid-state viscometer based on bulk acoustic wave (BAW) technology. Third section describes how to deduce the so-called acoustic viscosity and the density using the calibration curve developed by “SeGenuity ViSmart”. The calibration curve transforms the acoustic viscosity to the dynamic viscosity. The thus measured viscosity is compared to literature values of glycerol (calibration fluid) viscosity. After an interpolation step, the measured dynamic viscosity only differs by few percent from the literature data of the viscosity of glycerol. Fouth section shows the results of measurement of the sensor of the polymer solutions. The viscosity of the polymer solution is compared to the viscosity measured with the Anton Paar viscometer (MCR). A difference of only a few percent is obtained. We end with some conclusions in final section.

Design and methods

We combine a state of the art Radio Frequency Identification using the RFID CISC Xplorer200, and MEMS technology using the SenGenuity Vismart acoustic wave Sensor AWS set-ups. The CISC RFID Xplorer-200 is a state of the art UHF (ultra high frequency) RFID system that consists of a reader (interrogator), which is connected to two circular patch antennae (transmitting and receiving) and a tag (transponder) (see Fig. 1). As the system has a separate transmitting RHCP or right-hand circularly polarized antenna and receiving LHCP or left-hand circularly polarized antenna (clockwise rotation of the electromagnetic wave) it is possible to transmit and receive signals simultaneously called full-duplex radio (Dobkin and Wandinger 2005; Dobkin and Weigand 2006; Want 2006; Nikitin and Rao 2008; Novak 2009; Pursula et al. 2007; Dobkin 2012; Fiddes and Yan 2013). The Xplorer-200 allows to determine the scatter parameters as function of frequency and dielectric coefficients, and thus able to determine the dielectric permittivity of the medium surrounding the RFID tag (Hassan et al. 2017; Hassan 2016). The Xplorer-200 allows to connect to the Acoustic Wave Sensor AWS (MEMS) and transfer the data via an arduino-box from the (read/ write) tag to the receiving antenna (see Fig. 1).

Fig. 1
figure 1

RFID-IC communication system: the CISC RFID Xplorer-200 is a state of the art UHF RFID system that consists of a reader (interrogator), which is connected to two circular patch antennae, and the read write tag (transponder). The tag is able to modulate the backscattered signal, which can then be separately amplified. The tag-reader communication process is always initiated by the reader emitting antenna (RHCP), the (read write) tag can modulate the signal before transmitting it back to the receiver antenna (LHCP); as the system has separate transmitting and receiving antennae it is possible to transmit and receive signals simultaneously (full-duplex radio). The Xplorer-200 allows to connect with the acoustic wave sensor AWS (MEMS) and transfer the data via an arduino-box from the (read/ write) tag to the receiving antenna

Full size image

The idea here is to use a rewrite RFID tag, which is able to store information from an acoustic wave sensor and uses wireless transfer to send it to the RFID-reader. The rewrite RFID tag is able to modulate the backscattered signal, which can then be separately amplified. Such a procedure makes it possible to read measurements, e.g., viscosity, at distances of up to 20 meters without using wires. New technologies, such as the Internet of Things (IoT) are expected to allow even longer distance data transmission, which would allow field operation, but currently the focus of interest is on RFID technology. Here we confine ourselves to the operation of the SenGenuity Vismart acoustic wave sensor (AWS). Such a sensor determines the acoustic properties of the environment surrounding the sensor, from which properties such as the density and viscosity can be inferred. The operating principles are based on bulk acoustic wave (BAW) technology (Vellekoop 1998; Gardner et al. 2001; Setter 2005). The X-tal in the sensor is a piezoelectric substrate sensing element, which is excited by a high-frequency oscillator and operates in the thickness shear mode (TSM). In this mode, shear displacement occurs at the surface of the piezoelectric crystal (see Fig. 2). When a TSM-BAW device is placed in a liquid solution, a layer of fluid couples to the vibrating surface of the sensor and creates an acoustic wave in the surrounding fluid, which after transmission is received and back-transformed to an electric signal. The propagated velocity carries information on the fluid density and sound speed, whereas the attenuation carries information on the viscosity (Thompson et al. 1991; Ballantine et al. 1996; Vellekoop 1998; Drafts 2001). Using a thickness shear mode (TSM) resonator, the elastic properties of a polymer solution can be measured (Ballantine et al. 1996). We have not found any reference to measure shear rate dependence on acoustic wave sensors. We compare the acoustic viscosity measurements with the Anton Paar Rheometer (NCR-302). This is a laboratory rheometer that can measure viscoelastic properties of, e.g., environmentally acceptable polymer (Guar–Arabic gum) solutions. A complete description of the rheological behavior of polymer solution is well presented by the Carreau (1972) model (Carreau 1972; Chen and Bogue 1972; Agassant et al. 2017) i.e.,

$$\begin{aligned} \frac{\mu _w(c,s) - \mu _{\infty }(c)}{\mu _0(c)-\mu _\infty (c)}= \left( 1+\lambda (c)\dot{\gamma }(c,s))^2\right) ^{((n(c)-1/2)}, \end{aligned}$$
(1)

where (c) is the mass fraction, \(\mu _0(c)\) is the zero shear-rate viscosity, \(\mu _{\infty }(c)\approx 0\) is the infinite shear-rate viscosity, \(\lambda (c)\) is a time constant, and n(c) is the concentration dependent power law index. The shear rate \(\dot{\gamma }\) in two phase flow can be written as (see Sorbie et al. 1987; Sorbie 2013, pp 176)

$$\begin{aligned} \dot{\gamma }(c,s) = C \left( \frac{3n(c)+1}{4n(c)} \right) ^{n(c)/(n(c)-1)} \frac{u}{\sqrt{kk_{rw} s \;\varphi }}, \end{aligned}$$
(2)

where C=6 is a constant, \(\mu _\infty (c)=0\) and \(n(c) = -15.366*c+0.4704\), \(\lambda (c) = 0.0014*\exp (199.97*c)\). Moreover, s is the water saturation, c is the fractional concentration in mass per unit mass. Finally, u is the total Darcy velocity, \(\varphi =0.3\) is the porosity, \(k= 10^{-12} m^2\) is the permeability and \(k_{rw}\) is the relative water permeability. We will assume that the relative permeability is given by

$$\begin{aligned} k_{rw} = \begin{array} [c]{c} \left\{ \begin{array} [c]{l} 0 \\ k_{rwe} \left( \frac{s-s_{wc}}{1-s_{wc}-s_{or}}\right) ^{n_w} \\ k_{rwe} \left( \frac{s_{or}-1+ k_{rwe}}{s_{or}}+\frac{s(1-k_{rwe})}{s_{or}} \right) \\ \end{array} \right. \begin{array} [c]{l} \mathrm {for}\\ \mathrm {for}\\ \mathrm {for} \end{array} \begin{array} [c]{l} 0\le s \le s_{wc}\\ s_{wc}<s\le 1-s_{or}\\ 1-s_{or}<s \le 1 \end{array} , \end{array} \end{aligned}$$
(3)

where \(k_{rwe} = 0.5\), \(n_w=2\), \(s_{wc} = 0.2\) and \(s_{or} = 0.3\).

Fig. 2
figure 2

Sengenuity acoustic wave sensor (AWS), ViSmart\(\textcircled {R}\) VS-26xx series, based on Bulk Acoustic Wave technology with a viscosity range: 0–500 centistokes [cSt], Power supply: 5 [V], 35 [mA], Frequency: 5–2000 Hz. Crystal Sensor Surface showing the Thickness Shear Mode (TSM) of vibration. As shown in the bottom part of this figure 2, the displacement profile occurs through the entire thickness of the plate and is a maximum at the surfaces. Due to the fact that the displacement motion is parallel to the plate, the TSM keeps functioning in fluids, while protecting its surface, making it perfect for fluid sensing

Full size image

Calibration of sensor with the Paar viscometer

For calibrating the sensor we use the data in Table 1. Table 1 shows in the first two columns the glycerol mixture used to make the solutions displayed in column 3 and 4. Column 5 expresses the concentration of glycerol in w/w %. Column-6 gives the viscosities of these solutions measured using the Anton-Paar viscometer (MCR-302). Column-7 shows the “acoustic viscosity” inferred from the measurements with the SenGenuity ViSmart\(\textcircled {R}\) Acoustic Wave Sensor (AWS). Columns-8,9, 10 give the temperature, density and literature values of the glycerol solutions respectively. Column-11 and 12 give the difference between the measured glycerol viscosity with the Anton Paar Viscometer and the literature values and the error, i.e, the normalized difference between the measured Anton Paar viscosity (MCR-302) and the viscosity cited in the literature. The calibration is shown in terms of equations 1 and 2 used for the Carreau viscosity. The table summarizes our experimental data together with some literature data 1. The shear rate dependent viscosity of the Guar–Arabic gum that is manufactured in Sudan is measured using the Anton Paar “viscometer”. The data in Fig. 3 is used to obtain the regression expressions of n(c), and \(\lambda (c)\) using the solver in EXCEL.

Fig. 3
figure 3

The viscosity [mPa s] of Guar—Arabic gum as a function of the shear rate [\(s^{-1}\)] (see Eq. 2), which has been measured using the Anton Paar viscometer (MCR-302)

Full size image

Results and discussion

Figure 4 shows the comparison between Carreau’s model (Eq. 1), where the concentration dependent power law index n(c) and a concentration dependent time constant \(\lambda (c)\) have been used with the experimental data. The results show the same trend, but also a difference between measured and the regression results around a shear rate of 0-20000. Above a shear rate of 20000, there is good agreement. Figure 5 shows a comparison between the Anton Paar viscosity and the literature data on the viscosity of glycerol solutions. The agreement is excellent. It shows the applicability of the Anton-Paar viscometer for high precision measurement of polymer solutions. Figure 6 compares the viscosity measured with the Anton Paar Viscometer with the viscosity measured by the acoustic sensor. There is a difference between the values showing that an intermediate calibration step is necessary. After this calibration the interpolated data of the sensor shows excellent agreement with the data obtained with the Paar viscometer. It shows that a comparison in a work flow that uses the Anton Paar viscometer and the acoustic wave sensor is a promising method to validate the wireless SenGenuity Vismart for real-time measurements of the viscosity and thus for an on line quality control for polymer enhanced oil recovery. In addition such a work flow can help with a proof of concept of combining the acoustic wave sensor with RFID. Still the integration of RFID or alternatively the (Internet of Things, IoT) with sensor technology needs to be further elaborated.

Fig. 4
figure 4

A comparison between the rheological behavior of polymer solution by Carreau’s model (1972) and experimental data

Full size image
Fig. 5
figure 5

The concentration of the calibration fluid (Glycerol) as function of dynamic viscosities from Anton Paar viscometer (MCR-302) and literature

Full size image
Fig. 6
figure 6

The concentration of the calibration fluid (Glycerol) as function of acoustic viscosity (AV) and dynamic viscosities, which are measured by Sengenuity Acoustic Wave Sensor and Anton Paar viscometer (MCR-302) respectively

Full size image
Table 1 Parameters of data calibration (T temperature, Lit. literature, Diff difference)
Full size table

Conclusions

  1. (1)

    Guar–Arabic gum is an environmentally acceptable natural product, which is available in large quantities in India and Sudan; it can increases the viscosity of an aqueous solution to about 800 [mPa s] at concentrations of 2 w/w%, thus beneficial for enhanced oil recovery (EOR) processes.

  2. (2)

    There is a difference between the glycerol acoustic viscosity values obtained with SenGenuity ViSmart\(\textcircled {R}\) Acoustic Wave Sensor (AWS) and the shear viscosity obtained with the Anton Paar viscometer, showing that an intermediate calibration step is necessary. After this calibration is done, the interpolated data of the sensor shows excellent agreement with the data obtained with the Paar viscometer.

  3. (3)

    A expected, the viscosity versus the shear rate measured with the Anton Paar viscometer follows Carreau’s model.

  4. (4)

    The optimal fitting requires a concentration dependent power law index \(n(c) = -15.366*c+0.4704\) and a concentration dependent time constant \(\lambda (c) = 0.0014*\exp (199.97*c)\). The optimal regression equations are obtained using the EXCEL solver to obtain the characteristic concentration dependent power index and the concentration dependent time constant, and using Eureqa \(^{TM}\) to obtain the regression expressions. Eureqa \(^TM\) does not require an a priori choice of the regression equation.

  5. (5)

    Some preliminary acoustic viscosity data with Arabic gum without Guar, which has a low viscosifying effect have been obtained, but makes it plausible that the sensor can be used to monitor in situ polymer enhanced viscosities

  6. (6)

    The integration of cutting edge RFID or alternatively the (Internet of Things, IoT) technology with Acoustic wave sensor (AWS) technology needs to be further optimized.