Development of Embedded Fiber-Optic Evanescent Wave Sensors for

Subscriber access provided by READING UNIV

Article

Development of Embedded Fiber Optic Evanescent Wave Sensors for Optical Characterization of Graphite Anodes in Lithium-Ion Batteries AbdulRahman Ghannoum, Patricia M. Nieva, Aiping Yu, and Amir Khajepour ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b13464 • Publication Date (Web): 07 Nov 2017 Downloaded from http://pubs.acs.org on November 9, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Development of Embedded Fiber Optic Evanescent Wave Sensors for Optical Characterization of Graphite Anodes in Lithium-Ion Batteries AbdulRahman Ghannoum†, Patricia Nieva*,‡ Aiping Yu†, Amir Khajepour‡ †

Department of Chemical Engineering, University of Waterloo, 200 University Avenue West, Waterloo, Ontario

N2L 3G1, Canada ‡

Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, Ont., Canada N2L

3G1

KEYWORDS: fiber-optic sensor, evanescent waves, fiber-optic etching, fiber-optic sensor characterization, lithium ion batteries, graphite anode, state of charge

ABSTRACT The development, fabrication and embedding of fiber-optic evanescent wave sensors (FOEWS) to monitor the state of charge (SOC) and the state of health (SOH) of lithium ion batteries (LIB)s is presented. The etching of the FOEWS is performed using a solution of 6:1, 40 wt % ammonium fluoride (NH4F) and 49 wt % hydrofluoric acid (HF), and found to be superior to an etching solution containing just 49 wt % HF. FOEWSs were characterized using glycerol and found to have the highest sensitivity in a lithium ion battery when they lose 92 % of their

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 20

transmittance in the presence of glycerol on their sensing region. The physical effect that the FOEWS has on the graphite anode is also investigated and found to be much more significant in Swagelok cells compared to in-house fabricated pouch cells, mainly due to pressure variation. The FOEWS was found to be most sensitive to the changes in the LIB when it was completely embedded using a slurry of graphite anode material within a pouch cell. The optimized fabrication process of the embedded FOEWS demonstrates the potential of using such sensors commercially for real-time monitoring of the SOC and SOH of LIBs while in operation.

INTRODUCTION Lithium-ion batteries (LIB)s with their relatively high energy density have supported the development of plug-in hybrid/electric vehicles1. A concern of every driver is the distance they can travel before the depletion of their energy source, promoting research on more effective and accurate battery monitoring systems (BMS). The main goal of a vehicle’s BMS is to estimate the state of charge (SOC) and state of health (SOH) of individual batteries in a battery pack to effectively control the flow of current to achieve a specified optimum, such as maximum battery life. Sensors are utilized to measure variables (e.g. voltage, temperature and current flow) that can be correlated to the SOC and SOH. In addition to measuring voltage and current flow, research has extended towards developing new sensing systems to measure internal material property changes to improve the accuracy in estimating SOC and SOH and allow for early detection of cell failure2–5. A recent study observed that acoustic ultrasonic waves traveling through an operating battery experiences a shift in both amplitude and time of flight due to changes in the mechanical properties of the materials within a battery4,5. Coupling the acoustic measurements with voltage measurements in a machine learning model, the error of estimating SOC was reduced to about 1


2

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


% in a controlled testing environment with equivalent cycle conditions5. However, this preliminary study did not examine the effects of varying charge rates on the acoustic measurements and further investigations are needed to determine its viability in a commercial electric vehicle. The use of optical fibers to track volume/strain dynamics within a LIB has also attracted researchers. Optical fibers with Bragg gratings6 have been used as sensors for temperature and strain7, since Bragg gratings reflect a narrow-band of wavelengths based on the grating period. This mechanism has been used to monitor strain in LIBs during cycling in the hopes of also using it as a SOC sensor. However, it requires a costly sensor interrogator to detect shifts in the wavelength8. Optical fibers utilizing evanescent waves as a sensing mechanism have been used in various applications9–13 due to their fast response, selectivity and durability14. A recent study on the reflectance of commercial graphite anodes and fiber-optic evanescent wave spectroscopy of electrochemically lithiated graphite demonstrated a correlation between lithiation and the transmittance through a FOEWS15. The spectroscopy results demonstrated significant increase in transmittance in the near-infrared band (i.e. 750–900 nm) when graphite was lithiated electrochemically. This illustrated that changes in graphite’s optical properties during lithiation can be observed using a FOEWS15. This discovery resulted in the simplification of the FOEWS’s interrogation system, whereby a photodetector sensitive to near-infrared wavelengths can be utilized instead of a costly spectrometer to monitor a LIB16. The lithium content of a graphite anode is directly related to the SOC of LIB. The transmittance intensity of a FOEWS, which is affected by the lithium content in graphite, can hence be used to estimate a battery’s SOC. Capacity fade in a LIB results in a decrease in the change of graphite’s lithium content, this


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 20

results in a decrease in the amplitude of the FOEWS’s transmittance. This implies that the amplitude of the FOEWS’s transmittance can be correlated to a battery’s state of health (SOH). The LIB environment can be considered a delicate but also a chemically harsh environment containing an anode, cathode, separator and an electrolyte. Over charging, excessive external stress and contamination due to poor assembly can all lead to capacity fade by allowing lithium deposition, electrode detachment, electrolyte decomposition (side reactions) or separator pore size reduction17–20. Consequently, it is essential that the assembly of a battery is studied and optimized when introducing a new component to the battery environment such as a fiber optic sensor. In this paper, we report the studies for the development of a FOEWS, its embedding into a LIB and performance characterization. Different etching solutions are studied in their ability to produce durable fiber optic sensors. The method for embedding the FOEWS into the LIB and its characterization have been developed to ensure maximum battery cell life and improved sensor sensitivity. The minimal capacity fade observed in all studied LIBs using embedded FOEWS, also demonstrates its commercial feasibility.

EXPERIMENTAL SECTION Materials.

The LIB sensor was fabricated using a step index multi-mode optical fiber

(AFS105/125Y, Thorlabs) with a core and cladding diameter of 105 µm and 125 µm respectively. The fiber was etched using both 49% Hydrofluoric Acid (JT9564-6, VWR) and buffered oxide etchant (JT5569-3, VWR). The FOEWSs were connectorized using universal bare fiber terminators (BFTU, THORLABS) with connectors (B10125A, THORLABS) attached. Glycerol (GX0185-2, EMD Millipore) was used to characterize the FOEWS’s etched cladding prior to battery testing. Swagelok cells and pouch/prismatic cells were fabricated using graphite


4

Page 5 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and LiFePO4 electrodes (MTI Corp.) and a trilayer polypropylene-polyethylene-polypropylene membrane (Celgard®) as a separator. As an electrolyte for the batteries a mixture of 1:1 (vol%) ethylene carbonate (BASF) and dimethyl carbonate (BASF) was used, with 1 M lithium hexafluorophosphate (Purolyte®, Novolyte Technologies) as a salt. A graphite electrode slurry was prepared for embedding the sensors, which consisted of 26% solids and 74% liquid. The solid mixture consisted of 85% Graphite (43209, Alfa Aesar), 10% poly(vinylidene fluoride) (182702, Sigma-Aldrich) and 5% carbon black (Lib-SuperP, MTI), while the liquid component contained 100% N-methyl-2-pyrolidone. An ethylene-vinyl acetate polymer based hot melt adhesive (PLIB-HMA8, MTI) strip was attached to the sensors to ensure a proper seal during pouch cell fabrication. Sensor/Battery System Preparation. The optical fibers are initially stripped and etched in a polypropylene beaker (1201-0250, NALGENE) as seen in Figure 1-a and described elsewhere15,21. The sensors are then tested using glycerol to determine the amount of transmittance loss as described in detail elsewhere21. The Swagelok assembly configuration has been presented elsewhere15. Sensors used in pouch cells had two hot melt adhesive strips (MTI) attached about 3 cm from the center of the sensor to ensure a hermetic seal. The graphite/LiFePO4 electrodes are cut into 3 cm by 2 cm rectangles and 1 cm by 2 cm of active material is removed to obtain square electrodes (2 cm by 2 cm). The region with the removed electrode material are reserved for current collecting tabs, that are spot-welded as seen in Figure 1-b. Aluminum Tabs (PLiB-ATC4, MTI) are used for the positive electrode (LiFePO4) and Nickel Tabs (PLiB-NTA3, MTI) are used for the negative electrode (Graphite). The electrodes are then used to sandwich a separator, and the sensor is positioned between the graphite electrode and the separator. The stack is then sandwiched between two laminated aluminum sheets (alf-


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 20

100-210, MTI) as seen in Figure 1-c. Three sides are sealed using a heat sealer (MSK-140, MTI). The open end is used for electrolyte insertion inside a glove box and then sealed using a Compact Vacuum Sealer (MSK-155A, MTI). The pouch is then placed on a 3D printed case to be held for testing as seen in Figure 1-d. Pouch cells with embedded sensors were prepared by placing the sensor on the graphite electrodes, a slurry of graphite is then poured on top to envelop the sensor and then left to air-dry in a fume hood. The batteries are then cycled using an eight channel battery analyzer (BST8-WA, MTI) and the optical signal is recorded using a custom optical sensor interrogator with a narrow band LED concentrated at 850 nm connected to a computer as described elsewhere21.

Figure 1. An overview of the pouch cell with an optical fiber sensor preparation process starting with the a. optical fiber etching then b. spot welding of the current collectors to the electrodes and c. stacking the battery components (Laminated aluminum, cathode, separator, anode and second laminated aluminum layer). d. The final assembled pouch cell with a connectorized FOEWS.


6

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSION Fabrication of Fiber Optic Evanescent Wave Sensors. Etching of optical fibers using HF concentrations ranging from 48 to 52 wt % has been performed in the past to reduce the fiber`s diameter22. We first examined the use of a 49 wt % HF solution to etch the fluorine doped silica cladding off the sensing region of the FOEWS. However, the resultant sensor was fragile as it broke easily during handling and battery fabrication. In addition, the FOEWSs prepared using 49 wt % HF were found to have significant non-uniformity on their surface as seen by scanning electron microscopy (SEM) in Figure 2-a, circular pits with diameters ranging from 3-25 µm were also observed.

Figure 2. SEM micrograph of fiber optic evanescent wave sensors (FOEWS) etched using a) Concentrated hydrofluoric acid (49%) and b) Buffered hydrofluoric acid (6:1 ratio of ammonium fluoride to hydrofluoric acid) Etching solutions consisting of 6:1 to 10:1 mixtures of 40 wt % ammonium fluoride (NH4F) and 49 wt % HF are utilized extensively to etch silica in the semiconductor industry and are referred to as buffered HF (BHF)23. A 6:1 mixture (7.0 wt % HF and 34.3 wt % NH4F) was tested and found to produce FOEWS that can withstand the LIB fabrication process. As seen in


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 20

Figure 2-b, FOEWS prepared using BHF were observed to have smoother surfaces. The observed difference between the two fibers is believed to be due to the variance in the etching mechanism. The solution with higher concentration of HF etches quicker23 and favors HF reactions with silica24. The second solution containing NH4F however, had higher pH and favored

HF

reactions

with

silica

over

HF

during

the

etching

process24.

The

reasoning/mechanism in which the non-uniformity is formed is outside the scope of this paper but we simply conclude that using BHF solutions produced FOEWS with higher durability, which at the same time increased the throughput of the sensor/battery fabrication process. Fiber Optic Evanescent Wave Sensor in LIBs. Two LIB configurations were used with the developed FOEWS and tested: (1) a Swagelok and (2) an in-house pouch cell, as shown in Figure 3.

Figure 3. The FOEWS in two LIB configuration: a) Swagelok and b) pouch cell. Both the illustrated cells contained three layers, 1. Graphite electrode 2. Separator 3. Lithium Iron Phosphate (LiFePO4), with the fiber optic sensor positioned on the graphite electrode. The Swagelok configuration allows for repeated testing of the same FOEWS, since the cell can be disassembled and the electrodes can be replaced. We found that with careful handling, a sensor can be tested up to three times within one Swagelok cell. Theoretically the FOEWS should not break but since assembly takes place in a glove box handling can be difficult. The


8

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Swagelok cell has been reported to have a low first cycle efficiency (FCE) of 72%, attributed to the higher pressure conditions within the cell when compared to other configurations such as coin cells15. It is difficult to control the applied pressure, since it depends on the tightness of the Swagelok nuts. However, decreasing the degree of turning would increase the risk of contamination by air. Hence, a second generation of Swagelok cells were fabricated by reducing the applied pressure and keeping them in an inert environment to avoid air contamination. With these precautions, a FCE of 83.2% ± 3.00% was achieved, which is comparable to the reported FCE of coin cells tested with the same electrodes15. Graphite electrodes were extracted from the Swagelok cells after cycling and an SEM was used to observe the structural effect that the FOEWS has on the graphite electrode. As seen in Figure 4, a “trench” was formed on the graphite electrode in the region where the FOEWS was positioned covering about 1.7% of the entire area. The increased strain in this trench is expected to cause a variation in the performance of the electrode in the region below the sensor compared to the bulk region20. This variation may cause the sensor signal to only represent the local region but can be correlated to the general SOC as the local strain is constant. For preliminary testing, the Swagelok configuration is considered ideal but not sufficient for long term testing due to observed capacity fade after multiple cycles.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 20

Figure 4. SEM of a graphite anode extracted from a Swagelok cell that has completed two complete cycles, demonstrating the trench formed by the presence of a FOEWS on the graphite electrode. For improved pressure control, long term testing and packing efficiency, the pouch cell configuration shown in Figure 3-b was chosen over prismatic and cylindrical cells. The incorporation of a FOEWS requires the use of a polymer based hot melt adhesive around the optical fiber’s exit points from the pouch cell to avoid air from seeping into the pouch21. Humidity in the air forms PF3O and HF from the electrolyte’s salt LiPF6, and HF would cause battery degradation25. As a battery cycles, the applied pressure on the electrodes increases during charge and decreases during discharge due to volume change20. The assembled pouch cells were hence placed in custom housings to apply an initial pressure of 4 psi promoting uniform current distribution and minimizing rippling of the electrodes resulting in lower capacity fade20. Using this in-house pouch cell configuration, a FCE of 87.2% ± 2.5% was achieved, which exceeded the performance of both the coin cells and second generation Swagelok cells with the same electrode material. Electrodes from the in-house pouch cells were also examined using an SEM


10

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to determine the degree of deformation caused by the presence of the FOEWS. As seen in Figure 5, the graphite electrode in the pouch cells did not undergo significant deformation as compared to the Swagelok cells, which demonstrates that the Swagelok cells were cycling at a larger applied pressure promoting an increase in the rate of capacity fade. However, the area of the FOEWS interfacing with the graphite anode in the pouch cell is smaller, since the deformation and indentation occurring in the electrode is smaller as seen in Figure 5.

Figure 5. SEM micrographs of the region where the FOEWS was positioned (between the dotted lines) on the graphite anode extracted from a) an in-house pouch cell and b) a Swagelok cell. To maximize the interaction between the FOEWS and the graphite electrode a slurry of graphite was prepared to completely embed the FOEWS (see Figure 6).


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 20

Figure 6. SEM micrograph of prepared graphite slurry for embedding the FOEWS sensor.

Embedding the sensor allowed for easier pouch cell assembly since the sensor was fastened to the graphite electrode. In an industrial process, this would ensure higher throughput in a production line. The three tested configurations are illustrated in Figure 7, where the sensor in the Swagelok cell sits deeper into the graphite electrode when compared to the pouch cell as observed by the larger trench in Figure 5. While the embedded sensor is positioned on top of the original electrode with a layer of added slurry to encompass the sensor’s diameter. The results from testing the three configurations are discussed in the next section.

Figure 7. Schematic illustration of the FOEWS positioning within the graphite anode in a) a Swagelok cell (based on Figure 5-b), b) a pouch cell (based on Figure 5-a) and c) a pouch cell with added slurry (Figure 6) to embed the sensor. Fiber Optic Evanescent Wave Sensor Transmittance and Glycerol Testing. The fabrication of FOEWS for LIBs requires a validation step to ensure the etched optical fibers are functional and sensitive to changes within the battery. Many sensors were fabricated and tested


12

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with 100% glycerol to ensure that the optical fibers are not over/under-etched and to correlate the loss in glycerol to the sensors sensitivity in a LIB. Transmittance through an optical fiber relies on the refractive index difference between the core of the fiber and the cladding (i.e. the layer around the core), the core’s refractive index needs to be higher than the cladding to allow total internal reflection to occur6. Adding a temporary layer of glycerol on the etched sensors would indicate if the cladding still exists on the fiber and the condition for total internal reflection still exists, since glycerol’s refractive index (i.e. 1.4656 at 850 nm26) is higher than the refractive index of the sensor’s core (silica) (i.e. 1.4525 at 850 nm27). If a cladding does not exist 100 % loss in transmittance is observed when glycerol is placed on the etched fiber, which translates to a sensor with no signal in a battery. Etched FOEWS with about 85 to 96 % loss in glycerol were tested in LIBs. To compare the various cells, the LIBs were cycled from 0 to 100 % SOC based on the battery’s voltage limits (i.e. 2.5 to 3.7 V), while recording the optical transmittance through a FOEWS. The sensor’s baseline (i.e. transmittance at 0 % SOC) differed between the various batteries due to variation in the contact area, optical fiber coupling and cladding thickness. To compare the results, the sensors’ sensitivities were determined using the following expression, [(Transmittance at 100% SOC – Transmittance at 0% SOC)/ Transmittance at 0% SOC], and correlated to the FOEWSs’ loss in glycerol (see Figure 8).


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 20

Figure 8. Fiber optic evanescent wave sensor sensitivity in relation to the loss in 100% glycerol. The presented results include all three configurations, Swagelok cells, pouch cells and sensors embedded within the graphite electrode in a pouch cell.

Based on the results in Figure 8, the Swagelok and pouch cells with and without embedded sensors demonstrated the highest sensitivity when the sensor’s transmittance loss in glycerol was about 92.0, 93.5 and 92.0 %, respectively. However, since only two pouch cells with embedded sensors were tested, based on the two other configurations we conclude that FOEWS with about 92.0 % loss in glycerol achieve the highest sensitivity in a LIB. In the same figure a scatter was observed in the Swagelok cells’ results, which can be attributed to the variation in pressure


14

Page 15 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between the cells, since the depth of the trench is dependent on the pressure applied during assembly. The deeper the trench, the larger the interaction between the FOEWS and the graphite electrode, increasing the sensitivity of the sensor. This also explains the reason why the sensors in the Swagelok cells had larger sensitivity than the pouch cells as seen by the trench size difference in Figure 5. Embedding the sensor completely within the graphite electrode as illustrated in Figure 7-c within a pouch cell resulted in the highest sensitivity, which can also be seen when comparing the full cycle transmittance of all three FOEWS positions (Swagelok cell, pouch cell and pouch cell with an embedded FOEWS) (see Figure 9). The relative sensor signal change was determined by shifting all transmittance measurements to start at 0 at 0 % SOC and dividing the transmittance measurements by the initial transmittance at 0 % SOC as a point of reference to permit comparison between all three configurations.

Figure 9. The full charge and discharge transmittance for all three configurations, Swagelok, Pouch and Pouch with an embedded sensor. Relative sensor signal change is equal to [(Transmittance –Transmittance at 0% SOC)/ Transmittance at 0 % SOC].


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 20

The transmittance through the FOEWS was observed to increase when the battery is charged in all three configurations following a similar trend. The slope in the charging signal significantly decreased as the battery approached 62.5 % SOC, this may indicate the sensor’s limit of interaction with the graphite particles, due to the sensing depth28, or the decrease in lithiation activity in the local region29 of the FOEWS. During discharge the transmittance was observed to decrease steadily as lithium de-intercalated, indicating a steady change in the graphite particles surrounding the FOEWS during discharge unlike charging. To investigate this further, a method to embed multiple fibers within one graphite electrode in different locations/depths is required for complete characterization of the graphite anode. Nevertheless, the current observed trend in the signal from an embedded optical fiber within a pouch cell can be used to estimate SOC based on the optical changes within a LIB21. In addition to investigating multiple embedded FOEWSs within one battery, an optical model of the FOEWS and a model for lithium transport into graphite are currently being coupled and fitted to the experimental results and will be presented in future work. Further development in the embedding process of FOEWS will allow the use of such sensors in cylindrical cells and further promote their commercial use in many more applications.

CONCLUSIONS The process of fabricating fiber optic evanescent wave sensors and integration of the sensors into Swagelok and pouch battery cells have been presented. A method for characterization to achieve higher sensitivity and higher efficiency in the battery performance has been developed. The experimental results demonstrate that a buffered hydrofluoric solution of 6:1, 40 wt % ammonium fluoride (NH4F) and 49 wt % hydrofluoric solution, is superior to a solution of 49 wt


16

Page 17 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


% hydrofluoric acid for the cladding etching process during sensor fabrication. A qualitative analysis of the graphite electrodes used in both Swagelok and pouch cells demonstrate the significant reduction in applied pressure when using pouch cells allowing for battery’s with longer life. The quantitative analysis of the first cycle efficiency of the two battery configurations demonstrate the superiority of using a pouch cell with an initial applied pressure of about 4 psi. The fiber optic sensor was tested within batteries and found to have highest sensitivity when the transmittance loss in 100 % glycerol approached 92 %. It was also demonstrated that increasing the area that the FOEWS contacts the graphite electrodes by embedding the sensor enhanced the sensors sensitivity. Future studies will incorporate models of the battery/sensor system to analyze the signal and further optimize its integration into commercial lithium ion batteries.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Author Contributions All authors have contributed to the completion of the manuscript and have given approval to the final version of the manuscript. ACKNOWLEDGMENT This work was financially supported by the “Green Intelligent Transportation Systems (GITS)” – Ontario Research Fund, Canada, the “Next Generation Electric Vehicles: Development of Key Technologies and Full Vehicle Testing” – Automotive Partnership Canada, and General Motors.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 20

Special thanks to Liliana Zdravkova and Krishna Iyer for their contribution on the etching and the characterization of the optical fibers. REFERENCES (1)

Gu, Y.-J.; Zeng, C.-S.; Wu, H.-K.; Cui, H.-Z.; Huang, X.-W.; Liu, X.-B.; Wang, C.-L.; Yang, Z.-N.; Liu, H. Enhanced Cycling Performance and High Energy Density of LiFePO4 Based Lithium Ion Batteries. Mater. Lett. 2007, 61, 4700–4702. (2) Raghavan, A.; Kiesel, P.; Sommer, L. W.; Schwartz, J.; Lochbaum, A.; Hegyi, A.; Schuh, A.; Arakaki, K.; Saha, B.; Ganguli, A.; Kim, K. H.; Kim, C.; Hah, H. J.; Kim, S.; Hwang, G.-O.; Chung, G.-C.; Choi, B.; Alamgir, M. Embedded Fiber-Optic Sensing for Accurate Internal Monitoring of Cell State in Advanced Battery Management Systems Part 1: Cell Embedding Method and Performance. J. Power Sources 2017, 341, 466–473. (3) Ganguli, A.; Saha, B.; Raghavan, A.; Kiesel, P.; Arakaki, K.; Schuh, A.; Schwartz, J.; Hegyi, A.; Sommer, L. W.; Lochbaum, A.; Sahu, S.; Alamgir, M. Embedded Fiber-Optic Sensing for Accurate Internal Monitoring of Cell State in Advanced Battery Management Systems Part 2: Internal Cell Signals and Utility for State Estimation. J. Power Sources 2017, 341, 474–482. (4) G. Hsieh, A.; Bhadra, S.; J. Hertzberg, B.; J. Gjeltema, P.; Goy, A.; W. Fleischer, J.; A. Steingart, D. Electrochemical-Acoustic Time of Flight: In Operando Correlation of Physical Dynamics with Battery Charge and Health. Energy Environ. Sci. 2015, 8, 1569– 1577. (5) Davies, G.; Knehr, K. W.; Tassell, B. V.; Hodson, T.; Biswas, S.; Hsieh, A. G.; Steingart, D. A. State of Charge and State of Health Estimation Using Electrochemical Acoustic Time of Flight Analysis. J. Electrochem. Soc. 2017, 164, A2746–A2755. (6) Kawasaki, B. S.; Hill, K. O.; Johnson, D. C.; Fujii, Y. Narrow-Band Bragg Reflectors in Optical Fibers. Opt. Lett. 1978, 3, 66–68. (7) Kersey, A. D.; Davis, M. A.; Patrick, H. J.; LeBlanc, M.; Koo, K. P.; Askins, C. G.; Putnam, M. A.; Friebele, E. J. Fiber Grating Sensors. J. Light. Technol. 1997, 15, 1442– 1463. (8) Sommer, L. W.; Kiesel, P.; Ganguli, A.; Lochbaum, A.; Saha, B.; Schwartz, J.; Bae, C.-J.; Alamgir, M.; Raghavan, A. Fast and Slow Ion Diffusion Processes in Lithium Ion Pouch Cells During Cycling Observed with Fiber Optic Strain Sensors. J. Power Sources 2015, 296, 46–52. (9) Polynkin, P.; Polynkin, A.; Peyghambarian, N.; Mansuripur, M. Evanescent Field-Based Optical Fiber Sensing Device for Measuring The Refractive Index of Liquids in Microfluidic Channels. Opt. Lett. 2005, 30, 1273–1275. (10) Nath, P. Enhanced Sensitive Fiber-Optic Sensor with Double Pass Evanescent Field Absorption. Microw. Opt. Technol. Lett. 2009, 51, 3004–3006. (11) Choudhury, P. K. On The pH Response of Fiber Optic Evanescent Field Absorption Sensor Having a U-Shaped Probe: An Experimental Analysis. Opt. - Int. J. Light Electron Opt. 2003, 114, 13–18. (12) Choudhury, P. K. On The Fiber-Optic Chlorine Sensor with Enhanced Sensitivity Based on The Study of Evanescent Field Absorption Spectroscopy. Opt. - Int. J. Light Electron Opt. 2004, 115, 329–333.


18

Page 19 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(13) Infrared Fiber Evanescent Wave Spectroscopy for In-Situ Monitoring Of Chemical Processes; Proc. SPIE: Los Angeles, CA, United States, 1989; Vol. 1048, pp 145–153. (14) Messica, A.; Greenstein, A.; Katzir, A. Theory of Fiber-Optic, Evanescent-Wave Spectroscopy and Sensors. Appl. Opt. 1996, 35, 2274–2284. (15) Ghannoum, A.; Norris, R. C.; Iyer, K.; Alemohammad, H.; Zdravkova, L.; Yu, A.; Nieva, P. Optical Characterization of Commercial Lithiated Graphite Battery Electrodes and in Situ Fiber Optic Evanescent Wave Spectroscopy. ACS Appl. Mater. Interfaces 2016, 8, 18763–18769. (16) Bae, C.-J.; Manandhar, A.; Kiesel, P.; Raghavan, A. Monitoring the Strain Evolution of Lithium-Ion Battery Electrodes Using an Optical Fiber Bragg Grating Sensor. Energy Technol. 2016, 4, 851–855. (17) Arora, P.; White, R. E.; Doyle, M. Capacity Fade Mechanisms and Side Reactions in Lithium‐Ion Batteries. J. Electrochem. Soc. 1998, 145, 3647–3667. (18) Peabody, C.; Arnold, C. B. The Role of Mechanically Induced Separator Creep in Lithium-Ion Battery Capacity Fade. J. Power Sources 2011, 196, 8147–8153. (19) Lu, M.; Cheng, H.; Yang, Y. A Comparison of Solid Electrolyte Interphase (SEI) on The Artificial Graphite Anode of The Aged and Cycled Commercial Lithium Ion Cells. Electrochimica Acta 2008, 53, 3539–3546. (20) Cannarella, J.; Arnold, C. B. Stress Evolution and Capacity Fade in Constrained LithiumIon Pouch Cells. J. Power Sources 2014, 245, 745–751. (21) Ghannoum, A.; Iyer, K.; Nieva, P.; Khajepour, A. Fiber Optic Monitoring of Lithium-Ion Batteries. In IEEE SENSORS 2016 Proceedings; Orlando, FL, USA, 2016; pp 868–870. (22) Lyons, E. R.; Lee, H. P. Demonstration of an Etched Cladding Fiber Bragg Grating Filter with Reduced Tuning Force Requirement. IEEE Photonics Technol. Lett. 1999, 11, 1626– 1628. (23) Spierings, G. a. C. M. Wet Chemical Etching of Silicate Glasses in Hydrofluoric Acid Based Solutions. J. Mater. Sci. 1993, 28, 6261–6273. (24) Proksche, H.; Nagorsen, G.; Ross, D. The Influence of NH4F on the Etch Rates of Undoped SiO2 in Buffered Oxide Etch. J. Electrochem. Soc. 1992, 139, 521–524. (25) Goodenough, J. B.; Kim, Y. Challenges for Rechargeable Li Batteries †. Chem. Mater. 2010, 22, 587–603. (26) Rheims, J.; Köser, J.; Wriedt, T. Refractive-Index Measurements in The Near-IR Using an Abbe Refractometer. Meas. Sci. Technol. 1997, 8, 601–605. (27) Malitson, I. H. Interspecimen Comparison of the Refractive Index of Fused Silica*,†. J. Opt. Soc. Am. 1965, 55, 1205–1209. (28) Golden, J. P.; Anderson, G. P.; Rabbany, S. Y.; Ligler, F. S. An Evanescent Wave Biosensor. II. Fluorescent Signal Acquisition From Tapered Fiber Optic Probes. Biomed. Eng. IEEE Trans. On 1994, 41, 585–591. (29) Maire, P.; Evans, A.; Kaiser, H.; Scheifele, W.; Novák, P. Colorimetric Determination of Lithium Content in Electrodes of Lithium-Ion Batteries. J. Electrochem. Soc. 2008, 155, A862–A865.

Table of Contents Graphic


19


Page 20 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

20

Development of Embedded Fiber-Optic Evanescent Wave Sensors for

Recommend Documents