chip resistor array

2020-2027 ultra-small chassis size passive electronic components market outlook The global ultra-small chassis size passive electronic components market research report provides in-depth analysis of databases across industries and economies. It can provide development and profitability for participants in this market […]]

The global ultra-small case size passive electronic components market research report provides in-depth analysis of databases across industries and economies. For business management that may provide development and profitability for market participants. This is the latest report covering the current impact of COVID-19 on the market. The coronavirus (COVID-19) pandemic has affected lives all over the world. This has brought some changes to the market situation. The report covers early and future assessments of rapidly changing market scenarios and their impact. It provides important information related to current and future market growth. It focuses on technology, quantity, and in-depth analysis of materials and markets. This research has a dedicated section that introduces the main players in the market and their market share.

This report

AVX

Kemet

Coa

Murata

Nichicon

Matsushita

SEMCO

TDK

Visa

Kunio

…

SMD plastic film capacitors (PEN, PET and PPS)

Thick film and thin film chip resistors

Chip arrays, networks and integrated passive devices

Ferrite beads

Ferrite bead array

other

electronic product

Automobile industry

aerospace

The report also includes global markets, consumption tables, facts, main statistics and statistical data.

Historical year: 2015-2020

Base year: 2020

Expected year: 2020

Expected year: 2020-2027

b>

Industrygrowthinsights set the benchmark. By providing customers with joint and tailored research reports, the company's database in the market research industry is updated daily to provide customers with the latest trends and in-depth analysis of the industry. Our database pool covers a wide range of industries, including IT and telecommunications, food and beverage, automotive, medical, chemical and energy, consumer food, food and beverage. Each report has undergone appropriate research methods and verified by experts and analysts.

Alex Matthews

500 East E Street, Ontario

CA 91764, United States.

United States: +1 909 545 6473

Required fields are

shown as

Image Credit: Chris Lee/Dreamstime

by

Published

Researchers have demonstrated that the individual elements that make up the filter are integrated into a single electronic component. This can greatly reduce the space occupied on the computer chip.

The device manufacturing process includes depositing metal by electron beam evaporation and photolithography to define the metal pattern and etching process. Then, the final etching step triggers the self-winding process of the stacked film.

Image source: image provided by Li Xiuling

The researchers said that the team tested the performance of the rolled parts and found that under the current design, the filter is suitable for applications in the 1-10 GHz frequency range. Although these design goals are for radio frequency communication systems, the team said that based on past research, they can also achieve other frequencies, including frequencies in the megahertz range, because of their ability to implement high-power inductors. 

An electron microscope image of a series of new chip components that integrate inductors (blue) and capacitors (yellow) needed to make electronic signal filters in phones and other wireless devices.

Image source: Li Xiuling

"We used several simple filter designs, but in theory, we can use the same processing steps to combine any filter network,"

Graduate students lead research. "We use what already exists to provide a new, easier platform, integrating these components more closely than ever before."

Register to

Deliver such wonderful stories to your inbox every day.

Cheltenham, Gloucestershire

Safe and reliable resistors can expand surge protection and provide green performance in medical contact equipment.

Medical designers are faced with the complex fields of various applications and performance environments of the devices they develop. The aging population, more healthcare facilities, and the increasing demand for smarter and more convenient devices keep designers in a dilemma because they are faced with faster, smarter, unparalleled performance and precision devices. Continued competitive pressure.

These complex devices must also work closely with the human body, fusing complex digital components with the reality of the simulated patient. Therefore, resistors-passive components that resist the flow of current-play a vital role in medical design. In medical device design, each active integrated circuit (IC) requires up to 20 resistors. Off-the-shelf resistors are often insufficient, prompting designers to use specialized components specially manufactured to deal with various medical challenges, such as high voltage and electromagnetic interference.

Understanding the choice is critical, because various medical applications have different requirements for resistor performance. For example, in imaging, X-ray systems require ultra-high voltage resistors. Magnetic resonance imaging (MRI) scanners require non-magnetic resistors; ultrasound equipment requires integrated resistor arrays. In instrumentation and analysis, accuracy and stability are the key. In addition, some of the most challenging resistor applications have been found in contact devices. These contact devices are directly connected to the human body and are usually responsible for safely delivering high-energy pulses or detecting and monitoring biological signals.

Automated external defibrillators (AED) are widely used life-saving tools today, but their success is based on the ability to provide electrical pulses to ensure patient safety. Connected medical devices (such as electrocardiogram (ECG) monitors) also require the same security. Protection resistors with high rated energy support this application. This can be achieved using pulse-resistant thick-film components and a double-sided chip design, which can provide twice the energy of traditional components in the same footprint. The most important thing is that the defibrillation energy must reach the patient's body, which is usually achieved by designing a pulse resistance resistance directly in the lead set of the monitor's input circuit. The proportion of the defibrillation surge energy dissipated by the protective resistor decreases as the ohm value increases, which means that the designer must select the maximum value that is consistent with the functional requirements of the monitor. The designer must also consider the parameters of the test circuit itself, and consider the number of leads in a lead set based on electromagnetic compatibility (EMC) standards (such as IEC601).

In this case, the protective resistor (applying an appropriate surge to the patient and protecting the connected monitor from electrical damage) must reduce the rated energy from 25.00J at 1kO to 0.25J at 100kO. Although this has traditionally relied on synthetic technology, thick film solutions are being developed. Professional resistors complying with IEC 60601/61000 standards are also provided to minimize risks and speed up product development.

Consider that the total energy in the defibrillation pulse may be as high as 350J. The protection resistor is designed on the printed circuit board (PCB) in the display cable group or in the display itself to absorb part of the energy to prevent the defibrillation energy from entering the display. For small electronic components, even 1% of energy is important.

Other options that have been proven in industrial applications are also adding new value to medical design. Pulse tolerant chip (PWC) and high pulse tolerant chip (HPWC) provide higher energy capacity; for example, as an untrimmed resistor with 5% tolerance, HPWC can provide a capacity of about 1J during the duration of a defibrillation surge .

of

) Is an advancement of the high-density signal carrier (HDSC) option. It is an unfinished double-sided version that doubles the surge capability through thick film materials on both sides of the chip. This compact option is optimized for monitor input to ensure the pulse tolerance of the defibrillator. Surge protection can be maximized in two ways: against intentional surges (such as from a defibrillator) and against random surges (such as surges caused by electrostatic discharge (ESD)). DPCR has been pre-qualified to achieve maximum energy and voltage performance. DPCR is pre-certified IEC 60601-2-27 for defibrillation surge and IEC 61000-4-2 level 4 for ESD certification. It provides powerful design options to reduce risk and quickly bring the design to market.

Green options are also increasing, which can help designers meet the environmental requirements of medical devices. The use of lead-free resistors can meet the non-exempt standards for lead-free designs-with the continuous improvement of lead-free design requirements, the design has a longer service life and reduces the need to replace components.

Special resistors play an important role in medical design. These pre-qualified options have value far beyond mass-produced commercial components, providing features such as high electrical ratings, small size, high surge performance, and environmental value, enabling medical designers to provide safe and stable equipment To save lives.

About the author: Stephen Oxley is a senior application and marketing engineer at TT Electronics plc. He can be contacted by

After the US election and the COVID-19 vaccine – what will happen in the future?

When preparing for 2021, do we dare to reflect on 2020? Considering COVID-19 and political drama, we are not willing! There will be books and movies telling the story of 2020, so we will hold Hollywood in charge.

instead,

In the United States and around the world.

Some expectations in 2021:

– As of this writing, Pfizer, Moderna and AstraZeneca have all reported positive results of their vaccine clinical trials. Let us hope that the final result will continue to produce very positive results-other vaccines will also be released with similar results.

– We don’t want but now we want to get your attention. Let’s see what the Biden administration thinks about the reorganized Obamacare and what they choose to pay for.

– Combining the COVID-19 experience with the revised NAFTA (now known as USMCA) should yield positive results in reproduction/near-production activities in North America. COVID-19 tells us that certain medical supplies should be considered vital to the United States, so there should be a certain percentage of these key supplies manufactured in the United States, Canada or Mexico. This should be an important move, and USMCA should make this goal easier to achieve.

– China’s supply chain may become more stable and more competitive for two reasons: (1) China seems to have recovered from COVID-19; (2) the new US government may abandon the trade of the previous government war. Although there will be some withdrawal/approaching measures, China will remain a strong and reliable part of the global supply chain. (Note: We do recognize that some manufacturing industries in China are moving to Vietnam, Thailand and other places, but this development will not be as fast as some people think. For example, if a company wants to sell products in China, it Will still need to be made in China, otherwise it may lose market share.)

-We hope it will stay here. Since the Centers for Medicare and Medicaid Services (CMS) and insurance companies finally become logical for modern healthcare during COVID-19, it is difficult to predict this topic. Let us hope that they now understand that telemedicine is safer (in most cases) and very efficient.

We cannot guarantee this, but we hope. Although we are very happy to see you online, we also like to meet you in person.

I wish you all safety, prosperity and success in 2021! During this period, happy holidays, please be safe!

About the author: CEO Florence Joffroy-Black is MedTech's long-term M&A and marketing expert. You can contact her in the following ways

. Managing Director Dave Sheppard (Dave Sheppard) is a former medical OEM Fortune 500 company executive and an experienced MedTech M&A expert. He can be contacted by

Researchers have developed clinical-grade wireless implants that can operate without batteries.

A kind

Epilepsy, Parkinson's disease, chronic pain and other diseases may soon be treated. The miniature device is powered by magnetic energy and generates the same high-frequency signal as a clinically approved battery-powered implant.

Developed by

, The implant is about the size of a grain of rice.

The implant is a layer of magnetoelectric material that converts magnetic energy into voltage. This method avoids the shortcomings of radio waves, ultrasonic waves, light and electromagnetic coils, all of which have shown interference to living tissue or harmful heat.

Although battery-powered implants are often used to treat epilepsy and Parkinson’s disease, studies have shown that nerve stimulation can be used to treat depression, obsessive-compulsive disorder (OCD) and chronic, intractable pain, which can cause anxiety, depression and Opioid addiction.

Manufacturing wireless devices that are easy to implant and do not require batteries can make neurostimulation therapy more widely used. The researchers' equipment is small enough that it can be implanted almost anywhere in the body through minimally invasive surgery, similar to the type of implant surgery that places a stent in a blocked artery.

In order to prove the feasibility of magnetoelectric technology, the researchers implanted the device in rodents. During the test, rodents can roam freely in the enclosure. Studies have found that rodents prefer to be placed in a part of their shell, and the magnetic field will activate the stimulator, thereby providing a smaller voltage to the reward center of the brain.

Jacob Robinson, a member of the Rice Neural Engineering Initiative, said: "This proof-of-principle demonstration is important because it is a huge technological leap from a desktop demonstration to a treatment that can be used to treat people.

Amanda Singer, an applied physics student at Robinson Lab, solved the problem of wirelessly powering devices by joining layers of two different materials in a single film. The first layer is a magnetostrictive foil of iron, boron, silicon and carbon, which vibrates at the molecular level when placed in a magnetic field. The second is a piezoelectric crystal, which directly converts mechanical stress into voltage.

Singh said: "The magnetic field creates stress in the magnetostrictive material." "This material generates sound waves, some of which are at resonant frequencies, creating an acoustic resonance mode."

Acoustic reverberation activates the piezoelectric part of the film. Magnetoelectric membrane can obtain a lot of energy, but its working frequency is too high to affect brain cells.

Robinson said: "One of the main projects Amanda solved was to create circuits to regulate cell activity at a lower frequency." "It is similar to an AM radio-you have very high frequency waves, but they are so you can listen to them. Low frequency modulated."

It is a challenge to produce biphasic signals that can stimulate neurons without compromising the modulation of neurons, and so is miniaturization.

"When you need to develop something that can be implanted into small animals subcutaneously, your design constraints will change a lot," said Caleb Kemere, a member of the Neural Engineering Initiative. "Let it work on rodents in an unconstrained environment, forcing us to minimize size and volume."

Robinson and Kemere are associate professors of electrical and computer engineering and bioengineering. The research was supported by the National Science Foundation and the National Institutes of Health.

Flexibility and scalability allow electronic products to be added to a wider range of applications and products.

• Reach US$8.3 billion by 2030

• Market for healthcare products including flexible electronics

– A medical technology start-up company based in Evanston, Illinois has developed a wearable device for non-invasive monitoring of the ventricular shunt function of patients with hydrocephalus – was named the winner of the 2020 global competition at the Virtual Medtech Conference and won 35 Ten thousand dollar jackpot.

From discrete monitoring during treatment to continuous monitoring, it can improve diagnostic capabilities, preventive medical services, and provide:

MT3 can perform grinding, milling, turning, drilling and tapping operations on the workpiece with one setting.

MT3 can perform grinding, milling, turning, drilling and tapping operations on the workpiece with one setting. Vertical cylindrical grinder, supplement

, MT3 comes standard with 42-inch diameter T-slot table and precision grinding spindle with HSK-50A connection. The machine tool spindle can be interchanged through the HBK-200 clamping system, so that the spindle can be adapted to specific applications.

The machine can be expanded from a vertical grinder to a single-machine processing machine tool system, combining various spindles and tools into an optional unit, which is automatically replaced by the Fanuc R-2000 robot and Bourn & Koch's Alien Claw arm end tool, which can be quickly replaced Most tools and spindles. The unit is equipped with a spindle gear plate type tool changer, which can manage various tools and spindles.

Programming is done by using Bouuc & Koch's Grinding Human Machine Interface (HMI) and Fanuc manual guide-i, and is controlled by Fanuc 0i CNC. The virtual Y axis allows the machine to perform standard milling functions. The MT3 spindle has a Fanuc Beta-il 160M motor, which can generate 30kW of power from 2,000rpm to 10,000rpm, providing sufficient power and range for various grinding, milling, drilling and tapping applications.

The cast iron table frame is reinforced with polymer concrete to dampen vibration, thereby providing extra strength; heavy linear guides run on the X axis

Large diameter ball screw with heavy linear guide; supports grinding head and spindle on Z axis; optional scale feedback

Servo cylinder assembly, used for precise movement, X-axis positioning for feeding; protection from chips, chips, and contaminants; optional scale feedback

Fanuc servo motor, right angle gear box, mounted on the top of the machine column, driving the Z-axis ball screw

Direct-drive work spindle can realize high-speed machining, worktable positioning; 100rpm standard; 250rpm optional

A fully implantable neural interface with a large number of recording channels brings the gospel to patients with motor or speech loss. With the rapid increase in the number of recording channels, conventional complementary metal oxide semiconductor (CMOS) chips used for neural signal processing face severe challenges in terms of parallelism, scalability, computational cost, and power consumption. In this work, we propose a previously unexplored method to utilize the rich dynamic characteristics of the memristor array to process multi-channel neural signals in the memristor array in parallel. Extract the key information of the neural signal waveform and encode it in the memristor conductance modulation. Develop signal splitting scheme to adapt to equipment changes. In order to verify the fidelity of the processing results, and further prove the epileptic seizure prediction, compared with CMOS similar products, the epilepsy prediction has a high accuracy of more than 95%, and the power efficiency is increased by more than 1000 times. This work shows that the memristor array may be a promising multi-channel signal processing module for implantable neural interfaces in the future.

Electrical monitoring of brain activity has been proven to be an effective method to explore underlying neural principles, and has a wide range of applications in the field of biomedicine (

-

). Invasive neural probes, microelectrode arrays or electrode grids with huge recording channels are popular tools for recording brain activity (

). In recent years, with the exponential increase in the number of recording channels, the amount of recorded neural signal data has exploded (

,

), which poses a major challenge for real-time processing of multi-channel neural signals (

). Most existing signal processing hardware systems rely on multiplexing multiple recording channels into a single or several processing units to perform heavy computing tasks (

) (

). However, these systems have inherent scalability issues due to the delay and power consumption caused by multiplexing. These problems further limit many practical biomedical applications that usually require more recording channels, such as disease diagnosis and neural interfaces. Under these circumstances, new equipment and hardware systems are urgently needed to process multi-channel neural signals in parallel, especially those that can directly extract and store key information from neural signals in an efficient manner.

(

) An illustration of a conventional neural signal processing system that usually uses a multiplexer (MUX) to convert multi-channel neural signals into serial signals and then process them for biomarker calculations. (

) The proposed memristor-based system, which uses a memristor array to process multi-channel neural signals in parallel, in which biomarker extraction is achieved by the memristor conductance modulation. BL, bit line; WL, word line; SL, source code line.

Memristor (Memristor) is an emerging neuromorphic device, due to its ability to use inherent physical properties to perform calculations (

). In the literature, vector matrix multiplication based on the memristor array based on Ohm's law and Kirchhoff's law has been widely proven (

). In addition, the rich dynamic characteristics of memristors related to physical state changes can be used to implement logical operations (

) And time correlation detection (

), the calculation result is usually directly stored in the device. The latter method attracts low-power real-time neural signal processing. The early suggestion of this application is to use TaO's inherent super-threshold integration feature to detect neuronal peaks

Memristor (

). However, due to the inevitable differences between devices and the saturation behavior of device resistance (

), it is difficult to directly apply this method to multiple recording channels for parallel processing. In addition, processing another important neural signal-local potential (LFP) (

), which reflects the sum of postsynaptic potentials (

), for neuroscience and epilepsy diagnosis (

). Due to the relatively low sampling rate of LFP, it may be more suitable for low-power biomedical electronics than neuron spikes (

). It also has richer signal dynamics than neuronal spikes. Therefore, more complex calculations are needed to extract the key information, instead of amplitude threshold detection like neuron spikes (

). Intracranial electroencephalogram (iEEG) is a typical method of recording LFP by using electrodes on or inside the brain, and iEEG signals are usually collected from epilepsy patients (

), which is one of the most common neurological diseases (

). The prediction of seizures helps to develop new epilepsy treatment strategies, and it is necessary to be able to distinguish between pre-seizure and inter-seizure states in order to warn patients before seizures or provide early treatment (

). For this reason, the multi-channel parallel processing of LFP based on memristor dynamics may be an attractive method for predicting epileptic seizures. So far, it has not been experimentally proven.

Here, we propose a system that uses one transistor-one resistor (1T1R) memristor array for parallel processing of multi-channel LFP (

And figure. S1). The system uses inherent memristor conductance modulation to extract the energy and change information of the input nerve signal. The signal segmentation scheme in the memristor array is implemented through the 1T1R architecture to retain more input information and can also adapt to differences between devices. In order to verify the feasibility of the proposed system, a 16-channel iEEG signal was used as the neural signal to be processed. In addition, high-precision epileptic seizure prediction based on the processing results is demonstrated.

1k cell crosspoint array of memristor with TiN/TaO

/ HfO

/ TiN material stack is manufactured (for manufacturing details, please refer to materials and methods) as a platform for parallel processing of multi-channel neural signals. Here, TaO

The layer is used as a thermal enhancement layer to improve the analog conductance modulation behavior of HfO

Based on the memristor, its conductance state can be gradually adjusted by applying electrical stimulation. Each unit has a 1T1R structure, as shown in Figure 2. S2A, the memristor stack is located on top of the transistor drain terminal [bit line (BL)]. The gate and source terminals are connected to the word (WL) and source line (SL), respectively.

Shows the analog conductance modulation by SL under 15 identical RESET voltage pulses (

= 1.3 to 1.8 V,

= 5.0 V, and

= 0 V) ​​Start from the same initial conductance value. Similarly, figure. S2B displays the result of the SET process. Different colored lines represent different pulse amplitudes. It can be seen that the conductance modulation process of the memristor is highly related to the amplitude of the input pulse. When the pulse amplitude increases, the conductance decreases faster and tends to saturate at lower values, so the memristor can distinguish different input signal energies.

) The memristor conductance modulation process under the same RESET pulse starting from the same initial conductance value. Each data point is taken from the average of 128 devices.

= 5.0 V and

= 0V. The pulse width is 50 ns. (

) The waveforms of several designed input signal pulse trains. They have the same average amplitude of 1.5 V, but the amplitude of change (σ) increases from top to bottom. σ unit: volt. (

) The pulse train applied in (B) causes the evolution of the memristor conductance. Each data point is taken from the average of 128 devices. (

) The average change in conductance (Δ

, Left

Axis) After applying different pulse sequences, Δ is displayed

Increase with the change of input signal (right

axis).

In addition to signal energy, memristors can also distinguish between different input signal waveforms, such as changes. In order to verify this statement, four sets of 15 pulse sequences are used, which have the same 1.5 V average amplitude (ie, the same signal energy), but have different SDσ (ie, different signal changes) to modulate the memristor conductance (

). As we have seen, for different input signal waveforms, different electrical directing processes have been observed (

). Relatively large conductance change (Δ

) Is usually observed after the pulse with higher amplitude in each pulse sequence (indicated by the arrow), so a signal with a larger change will result in a larger Δ

In the final conductance, as shown in the figure

. This conductance modulation behavior is essential for neural signal processing because it helps to extract key information about changes in energy and input signals and encode them as changes in memristor conductance. It is worth noting that this important information is automatically stored in the non-volatile memristor and can be accessed as the conductance state of the device through a read operation for later feature extraction. This is the basis for subsequent parallel processing of neural signals.

In order to process multi-channel neural signals in parallel, a signal segmentation scheme was developed based on the memristor array, as shown in the figure.

. The original signal collected from the brain is first linearly converted (with amplification and offset) to the desired voltage range (ie ~1 to 2 V), and then used as the input signal of the memristor array, and then

Illustration of a signal segmentation scheme in a memristor array.

) Typical 16-channel waveforms of seizure and seizure signal clips. Blue, the middle. Red, paroxysmal. The sampling rate is 400 Hz. (

with

) Input voltage histogram after linear transformation of the original interictal and pre-ictal neural signals. Gain = 10

= 1.5V. (

) The conductance changes after processing the inter-wall and inter-wall signal segments in the memristor array respectively. (

) Histogram of conductance change (Δ

) Are located in (E) and (F) respectively.

The converted multi-channel signals (each lasting about 2.4 s) are fed into multiple rows of the memristor array. Then the WL turns on these 64 columns in turn to divide the input signal in each channel into segments with the same length (37.5 ms) to be applied to the corresponding memristor column, whose conductance changes record energy and change information. input signal. Here, the signal sampling rate is 400 Hz, so each segment corresponds to a 15-pulse sequence, similar to

. This signal segmentation scheme in the memristor array with 1T1R architecture allows it to store information about the evolution of the input signal over time. In fact, the size of the required memristor array can be determined by the number of recording channels, the length and the sampling rate of the neural signal to be processed and the working conditions of the memristor.

An example of a set of 16-channel beat and beat iEEG signals from the Kaggle epileptic seizure prediction dataset (for details on the dataset, see Materials and Methods) is shown in

(B and C). In order to obtain more recording channels and longer signal duration in practical applications, the size of the memristor array can be increased proportionally. Distribution of input voltage (linear conversion from original data, same gain, gain = 10)

And offset

= 1.5 V), compare the signals before and after the two groups

.

Plot Δ

The two signals are processed in parallel in the memristor array and then mapped. Distribution of output Δ

show on

. It is found that the average value of the inserted signal (~1.5 V) is similar, but the SD signal (0.24 V vs. 0.15 V) is much larger, resulting in a larger change in the conductivity of the memristor device (for example |Δ)

|> 15μS).

In order to further optimize the neural signal processing system based on the memristor array, a systematic study was carried out using different experimental conditions. For example, figure. When using input signals for SET and RESET operations, S3 compares system performance. It was found that the RESET operation resulted in a much larger contrast between intra-signal and inter-signal correlations, partly because the RESET process was more gradual than the SET process.

And figure. S1B. We should mention that the system performance may also be affected by the characteristics of the memristor (such as the on/off ratio, the number of states, and the linearity of the conductance modulation; see the discussion in Figure S3) and linear transformation conditions (such as the deviation of the neural signal). And amplification gain). For example, figure. S4A shows the correlation as a function of amplification gain.

(

= 1.5 V). Within 0.7×10

To 1.7×10

, The greater the gain, the greater the correlation contrast, and therefore the better the performance. Similarly, figure. S4B shows the correlation with

(Earnings = 10

). In the range of 1.2 to 1.5 V, the smaller the offset, the better the performance. All these results show that there is a lot of room for optimization to improve the system performance in practical applications. Further research is needed to further study the influence of memristor conductance modulation characteristics and neural signal linear transformation conditions on system performance.

For practical applications, the robustness of the multi-channel neural signal processing system based on the memristor array will be further analyzed on the time and space scale. The robustness of the time scale has been verified by conducting 200 trials, in which the first 100 wakes use the same seizure signal, and the other 100 wakes use the same seizure signal. All signals are processed in the same memristor array.

Display the correlation diagram between the corresponding Δ

Figure (For calculation of correlation coefficient, please refer to Materials and Methods). The inter-signal correlation is very low (<0.2), while the intra-signal correlation is very high (> 0.7). This result confirms that the processed neural signal in the memristor array retains sufficient information about the resulting Δ

Can be used to distinguish between seizures and inter-seizure signals. In practical applications, it is also important that the memristor maintain its characteristics during repeated operations to process a large amount of neural signal data from the brain. For this reason, we investigated the durability characteristics of the memristor, and the results showed that it can reach more than 2×10

Cycle (Figure S5). Our previous work also confirmed that the durability of analog switches can be further extended to ~10

By adjusting the switch conductance range (

). Then, we can roughly estimate that our memristor can support continuous neural signal processing for at least 5 years (for details on estimation, see Materials and Methods). For most typical applications, this lifetime is sufficient, and the lifetime can be further extended by using a memristor array with a larger number of columns.

) The output correlation matrix Δ

In different trials. The interictal and pre-ictal signals were used for the first 100 trials and the last 100 trials, respectively. (

) The shaded error bar graph of Δ

The waveform of the corresponding signal. (

) Distribution of Δ

The difference between interval and period trials. (

)Δ

The picture is a map drawn from different experiments, where the same suite and suite neural signal sets are applied to two different memristor arrays. The path from top to bottom and left to right is labeled T

To T

Correspondingly. (

) Correlation matrix between different experiments. T

And T

: Apply the same interval signal segment to the experiments of array #1 and #2 respectively. T

: Apply the same assertion signal clip to the trials of array #1 and #2, respectively.

Please note that there are differences in experimental results between experiments, which are mainly due to the periodic changes between memristors (

).

Display Δ from test to test

The waveform of the processing result.

The changes in the interocular signal and the frontal signal show similar distributions under small SD (3.74μS vs. 3.93μS). along with

, Which indicates that the difference between trials is very small, and therefore does not prevent the system from distinguishing between interictal and pre-ictal signals. In addition, the differences between the devices in the memristor array may also affect system performance. To study this effect, we processed the same neural signal segment on 16 channels of the memristor array, as shown in Figure 2. S6. The results show that the different channels in the memristor array show similar output Δ

Graph of the same input signal (Figure S6, E and F). In order to further verify the spatial scale of the memristor used for multi-channel processing of neural signals, that is, the robustness of the array level, two sets of new 16-channel iEEGs were processed on two different memristor arrays, and the results were summarized as follows:

. Similarly, it can be seen from the correlation diagram that the room and room signals can be distinguished. Future equipment optimization to reduce the variability of the memristor array is expected to further improve system performance.

After processing the input signal on the memristor array, the conductance (or equivalent Δ

It is possible to read the same initial conductivity value of all memristor devices) for further analysis, such as feature extraction and classification, which can be done by general machine learning algorithms. For proof of concept, here we can simply use the number of devices in a specific Δ

The range can justify the prediction of seizures. Reasons for the number of devices with a specific Δ

The input as the classification is the processed result, namely Δ

The distribution can represent signal energy and variation, as described above. as the picture shows

, The distribution of Δ

Although the change has similar signal energy, it can be measured by the periodic change that the signal from the periodic signal is more scattered than the signal from the periodic signal.

) Average Δ

Relative to the signal energy of eight consecutive segments. (

The relationship with the signal changes of eight consecutive segments. (

) The training results of the LDA classifier for 54 16-channel signal segments and the test results of the trained LDA classifier for 6 16-channel signal segments are plotted. F

, F

And F

Use Δ to indicate the number of equipment

Respectively in the specific range of (-35μS, -27μS), (-16μS, -9μS) and (-7μS, 0μS). (

) Training and testing accuracy from 10-fold cross-validation.

In order to prove the feasibility of using the proposed method to predict seizures, as described above, 60 sets of pre-seizure and inter-seizure iEEG signals were processed in our memristor array. Number of devices with Δ

∈ (-35 microseconds, -27 microseconds) (F

),Δ

∈(−16μS,−9μS) (F

) And Δ

∈(−7μS, 0μS) (F

) Is used as the extracted feature. Use linear discriminant analysis (LDA) to design a linear classifier (

) To distinguish the room and the signal in front of the room. To be fair, we use 10-fold cross-validation to train and test the classifier.

The results of 1 out of 10 trials are shown. The average accuracy of the 10 training and test trials were 95.33 and 95.00% (

), just use simple linear classifiers, which can be easily implemented in hardware. The average sensitivity and specificity of all test tracks are 93.33 and 96.67% respectively (for detailed calculations, please refer to Materials and Methods). It should be noted that more features can be extracted from the processing results of multi-channel neural signals to process more complex data sets and achieve higher classification accuracy.

Finally, the performance of the memristor-based hardware system is compared with the related literature work using memristors, and the performance of complementary metal oxide semiconductor (CMOS) application specific integrated circuit (ASIC) chips for neural signal processing is also compared. Benchmarks. This comparison is compared in detail in "Materials and Methods". The data and methods are used to process a segment of 16-channel neural signal segment with a sampling rate of 10 kHz (duration about 1-s). Compared with the CMOS ASIC reported in the literature about 80μW per channel, the power efficiency of 60.81 nW per channel in this work shows an improvement of about three orders of magnitude. Therefore, our system shows substantial advantages in the parallel processing of low-power multi-channel neural signals.

In addition to power consumption advantages and high scalability, our memristor array-based system also has the potential for real-time processing by using two memristor arrays to alternate conductance modulation and reading steps. Similarly, in our method, the input signal is segmented and directly applied to the corresponding memristor, essentially eliminating the need for a data buffer for temporarily storing neural signal segments. In addition, our system performs a lot of calculations and compresses the results into device conductance, thereby reducing the need for complex circuit modules for calculations and additional storage of intermediate and final data. These advantages can help significantly reduce the chip area, which is also a serious challenge for implantable biomedical electronics.

In summary, we have designed a highly scalable low-power hardware system for parallel processing of multi-channel neural signals based on memristor arrays. In addition, based on the results of experimental processing, it has been proved that the accuracy of epileptic seizure prediction is as high as 95% or more to verify the feasibility and efficiency of the system. The analog conductance modulation behavior of the memristor allows us to extract the critical amplitude and change information of the input signal and encode it in the conductance change of the memristor. With the signal segmentation scheme developed in the memristor array, the system has demonstrated excellent robustness and resistance to device changes. Our work highlights the huge potential of using memristor arrays to process multi-channel neural signals in parallel with high fidelity and power efficiency, which is essential for future fully implanted neural interfaces.

In the 1T1R structure, the transistor is used as a selector, and the resistive switch layer is fabricated on top of the drain terminal of the transistor. For our 1k cell memristor array, the transistor array is manufactured by a commercial foundry in a 0.13μm CMOS process. The remaining process steps are completed in the laboratory. For more detailed information about equipment manufacturing, please refer to our previous work (

).

We illustrate our detailed system overview in Figure 5. S1. In this picture, we propose a

X

A memristor array with peripheral circuits, where

Represents the number of channels and segments respectively. The design supports four operating modes, including FORM, SET, PROCESS and READ. The first three modes are used for electroforming, setting and resetting the memristor. In this work, the PROCESS mode is selected to process multi-channel neural signals in parallel through the RESET operation. In the reading mode, a current sense amplifier (CSA) with a resolution of 3 bits is used as the reading circuit. Figure S7 illustrates the programming scheme in PROCESS mode. Activated BLs and modules used to process specific parts of the neural signal are color-coded, while those that are not used are drawn in gray. Parallelism can be achieved in this mode by applying multi-channel input signals to the memristors in a column (ie segment) to simultaneously modulate their conductance. The shift register controls and selects which memristor column the input signal is applied to, and only one WL is turned on at a time. Similarly, figure. S8 shows the reading scheme of the memristor array in the reading mode. When sensing and digitizing the current, the CSA clamps the SL voltage to 0.15 V (ie, the read voltage), and then compares the current with the reference current. Figure S9 shows the FORM and SET schemes of the array. In these two modes, the memristor can be programmed in parallel or sequentially.

The 16-channel neural signal comes from the widely used open-access data set Kaggle seizure prediction data set (

). The data set includes multi-channel neural signals from four dogs and two human patients. In our work, a neural signal sampled from "dog 2" at 400 Hz is used. We randomly selected 60 groups of onset and onset signals for classification. Here, the pre-seizure signal refers to a brain signal segment recorded tens of minutes before the onset of the epileptic seizure, and the interictal signal was recorded from the brain at least 1 week before the next seizure occurred.

Correlation coefficient, use

It is a statistic used to measure the linear relationship between two sets of data. the value of

Between -1 and 1. The closer its absolute value is to 1, the better representation of these two data sets can be achieved by linear equations. There is almost no linear relationship between them, or

Close to 0. For any given two vectors

Have the same length

, Their correlation coefficient

It can be calculated using the following formula

where

Yes average

respectively.

Figure S5 shows that the typical durability of our memristor can reach more than 2×10

cycle. Because the memristor has such a long service life, we did not observe any significant drop in system accuracy in the experiment. In this work, we use the neural signal sampled at 400 Hz as the input of the memristor array, so when processing one segment (64 segments) of the neural signal segment, each device has to withstand 15 RESET voltage pulses (Ie a fragment). About 2.4 seconds. Then, we can roughly estimate this endurance as 2×10

The loop can continuously process 2×10 neural signals

/ 30×2.4 seconds = 1.6×10

s> 5 years, assuming that the same number of SET pulses (15) are needed to initialize the memristor conductance after each segment. Therefore, the lifetime of the memristor array is sufficient to meet the needs of most applications.

In the prediction, the early samples are considered positive samples, and the interphase samples are considered negative samples. The results of the LDA classifier can be divided into four categories: true positive, false negative, true negative and false negative. Here, "true" and "false" indicate correct and incorrect predictions, respectively. In order to evaluate the prediction performance, three statistical indicators are calculated as follows, including accuracy, sensitivity and specificity.

To estimate the power efficiency, we used a task to process the 16-channel neural signal sampled at 10 kHz. The whole task can be divided into two parts: multi-channel signal processing, and then read the conductance state of the memristor. For comparison with CMOS ASIC, 16 consecutive input pulses are applied to each memristor, which is similar to the 15-pulse experiment in this demonstration. Then, a signal segment with 10,240 samples in each channel can be processed in a 16×640 memristor array to estimate power efficiency.

The module used for multi-channel signal processing steps is shown in Figure 2. S7. In order to handle the tasks defined above, the energy consumed in the array operation can be calculated as (2 V)

×30μS×50 ns / pulse×16×640×16 = 983.04 nJ, where 2 V is the upper limit of the input voltage amplitude, because the input voltage pulse is linearly transformed to the range of 1 to 2 V, as shown in Figure 2.

(C and D). Here, the initial memristor conductance is 30 μS, and the pulse width is 50 ns. The energy consumed by the peripheral circuit is calculated as follows: For 640 BL multiplexer (MUX), 0.6μW×50 ns/pulse×16×16 pulse×640 = 4.92 nJ; 0.6μW×50 ns/pulse×16×16 pulse ×For 16 SL MUXs, it is 640 = 4.92 nJ, where in PROCESS mode, the power consumption of BL / SL MUX is 0.6μW. In the programming scheme of PROCESS mode, the dynamic power consumption of the shift register is only negligible, which is estimated to be 10 kHz / 16×(5 V)

×10 fF×(10,240 samples / 10 kHz) = 0.16 nJ, where 5 V is the source voltage and 10 fF is the transistor gate capacitor. Therefore, the total energy of this step is estimated to be 993.04 nJ.

The module used for reading mode is shown in Figure 2. S8. During the above tasks, the energy consumed in the memristor device can be estimated as (0.15 V)

×30μS×50 ns/pulse×16×640×16 = 0.35 nJ, where 0.15 V is the reading voltage. The energy consumed by the peripheral circuit is calculated as follows: For 640 BL MUX, 0.045μW×50 ns/pulse×16 pulse×640×16 = 0.37 nJ and 0.045μW×50 ns/pulse×16 pulse×640×16 = 0.37 16 NJ of SL MUX, where 0.045μW is the power consumption of BL/SL MUX in READ mode. The 16 CSAs consume 4μW×50 ns×16×640 = 2.05 nJ of energy to sense and digitize the output conductance, among which the 4μW CSA power consumption is estimated through simulation. Similarly, the dynamic power consumption of the shift register is only 0.16 nJ. Therefore, the total energy for this step is 3.30 nJ.

To calculate the power efficiency, we first estimate the total energy consumption to be 996.34 nJ. Therefore, the power efficiency is estimated to be 996.34 nJ / (10,240 samples / 10 kHz) / 16 channels = 60.81 nW per channel. For similar products based on CMOS, (

) Reported an ASIC chip for epileptic seizure control, in which the biological signal processing unit consumes a total of 2.5 mW of energy to extract features based on 128-point fast Fourier transform, and perform entropy calculation based on 16-input ridge regression model for classification . Therefore, its power efficiency is calculated as 2.5 mW / 16 channels = 156.2μW/channel. Cheng

) Reported another ASIC chip with a similar power efficiency of 163.2μW per channel. According to the amount of calculation used for feature extraction and classification, it is estimated that the power consumed by the feature extraction of these two ASIC chips exceeds 50%. Therefore, compared with similar products based on CMOS, the neural signal processing system based on memristor shows more than 1000 times advantage in power efficiency (60.81 nW per channel versus ~80μW per channel).

For supplementary materials for this article, please visit:

.

This is an open access article distributed under the following terms

, It allows use, distribution and reproduction in any medium, as long as the final use is

For commercial interest, and provide the original works appropriately cited.

Volume 6, Number 41

October 7, 2020

Thank you for your interest in promoting the term "scientific progress".

Note: We only ask you to provide your email address so that the person you recommend to the page knows that you want them to see the email and that it is not spam. We will not capture any email addresses.

This question is used to test whether you are a visitor and prevent automatic spam submission.

Please log in to add an alert to this article.

by

: Eabc4797

The memristor array provides a scalable and efficient platform for the parallel processing of multi-channel neural signals.

Volume 371, Issue 6526

©2021

. all rights reserved. The American Association for the Advancement of Science is 

, 

 with 

ISSN 2375-2548.

These authors contributed equally to this work.

Current address: ABB Research Center, 5405 Segelhofstrasse 30-34, Baden, Switzerland.

Current address: University of Zurich and ETH Zurich Institute of Neuroinformatics, Zurich 8057, Switzerland.

Microfluidics technology is essential for many laboratory-on-chip applications, but it is still challenging to implement a portable and programmable device that can autonomously perform measurement protocols when used by people with minimal training. Here, we propose a general concept towards this goal by implementing a programmable liquid circuit, in which the liquid in the capillary-driven microfluidic channel can be controlled and monitored from a smartphone to perform various advanced Liquid handling tasks. We achieve this by combining electronically controlled valves (e-gates) with passive capillary valves and self-venting channels. We demonstrate the concept by implementing a microfluidic clock with a diameter of 5 mm, a chip that uses 100 electronic gates with electronic feedback to control four liquids, and is designed to deliver in any order and combination sequence or in parallel And combine multiple liquids. This concept is scalable, compatible with high-throughput manufacturing, and can be used in many microfluidic-based analyses that will benefit from the precise and convenient handling of liquids.

The use of microfluidic technology in laboratory-on-chip equipment has proven its advantages in a wide range of applications, including diagnostics, biochemistry, preclinical research, and cell biology. It reduces the time to obtain results and the consumption of reagents, while Increase the throughput of analysis (

-

). Despite these advantages, most microfluidic implementations still rely on cumbersome methods to replace liquids, including complex components such as valves, actuators, pumps, and responsive materials (

,

), and involves labor-intensive sample and chip preparation steps. These practical challenges will limit the scalability, portability, adoption, and commercialization of these devices, especially if they are developed for point-of-care (POC) diagnostic applications (

). After the pioneering work of the Quake team (

), the integrated pneumatic valve has become a popular strategy for fine-tuning and programmable control of liquid in microfluidic channels. However, this kind of manipulation has usually been solved by developing complex microfluidic devices with mobile structures and using bulky, dedicated and expensive peripheral equipment, which usually also requires skilled users. Many other integrated liquid control strategies, including manual (

),electric(

), Chemistry (

), optical (

),magnetic(

) And heat (

Although drives have been developed, they often lack programmability and real-time feedback, and still suffer from manufacturing complexity and known issues related to high-throughput manufacturing and polydimethylsiloxane (PDMS) and hydrogel integration .

Digital microfluidic technology has become an attractive alternative, which can accurately program biochemical reactions without the use of moving parts to accurately manipulate droplets through various actuation principles (

), such as electrowetting on dielectric (EWOD), dielectrophoresis, surface acoustic waves and magnetism. Among these principles, EWOD has become the most widely used technology, but it has some practical problems, such as the need to use an immiscible oil phase to separate aqueous samples, high pressure sources, hydrophobic coatings, transparent electrodes and dozens of (if No) hundreds of external electrical contacts to address the individual electrodes in the high-density array (

). For most applications covered by microfluidics, a continuous flow of samples, reagents or particles through a specific area of ​​the device is preferred or required (

). Especially for POC equipment, since no peripheral equipment is needed (

). However, because the flow path and flow rate are defined by design and materials, and optimized for specific analysis, this advantage cannot be achieved due to the limited flexibility of flow control. After the work of the Whiteside Group (

) And Yager (

), many research groups have also made great efforts to improve the analysis performance of paper microfluidics by adding new functions to control the flow rate (

). We and other teams use passive valves by precisely constructing hard materials such as silicon, glass and plastic,

). In general, it is challenging to implement precise, high-density and programmable valves in paper microfluidics. Closed microfluidic channels implemented on other substrates are complex to manufacture, which is different from capillary-driven fluids (such as hydrophobic polymers). Compatible and difficult. Handling liquids without generating bubbles (for example, controlled consolidation or sequential delivery of liquids).

Here, we introduce a new method of controlling the capillary to drive the liquid in the microfluid by using a set of key functions to implement a programmable liquid circuit (

). Especially in these circuits, (i) low-cost, portable peripherals and smart phone application protocols can be used to stop and guide the liquid where and when needed, (ii) the position of the liquid can be monitored electronically, and (iii) can be generated The co-current or sequential flow of liquids, and (iv) the simultaneous and bubble-free merging of liquids can be completed in closed and hydrophilic microchannels manufactured using high-throughput technology. Stop-and-go flow control is based on e-gate (

), which is based on Gibbs' inequality (

) And low voltage (<5 V) electrowetting effect (

) Restore it. By providing multiplexing and smart phone control, and combining it with passive capillary valves and self-exhausting channels, we can extend this concept in a programmable liquid circuit to control multiple in a compact and portable system. Combination and sequential delivery of various liquids.

The concept of using a series of electric microfluidic gates (e-gates) and protocols from smartphone applications to control, monitor, sequentially deliver and integrate multiple liquids flowing in the microfluidic chip. The liquid sucked into the capillary-driven microfluidic chip flows autonomously in the microfluidic channel by capillary action, until the liquid meniscus is pinned to the groove orthogonal to the flow path to stop the flow. The flow is restored by applying a potential difference (<5 V) between the liquid (positive electrode) and the electrode patterned on the trench (ground). Use capacitance measurements as feedback to monitor the position of the meniscus. Using capillary valves with vents and self-venting channels and electronic gates, the pipetting liquid can flow through the same channel in any order and combination sequence, and the liquids flowing in the opposite direction can be combined without generating bubbles.

In the simplest implementation, the programmable liquid circuit uses capillary force to automatically transfer the liquid in the microchannel and the capillary pinning position at the electronic gate to stop and guide the liquid. Capillary pinning does not require external energy, and its size is determined by the surface tension of the liquid, the channel geometry and the contact angle between the liquid and the wall of the microchannel (

). This phenomenon has been used by other research groups in centrifugal microfluidics to make burst valves (

), passive microfluidics for capillary trigger and shut-off valve (

), the microfluidic phase guide used to control the filling and emptying of the microfluidic structure (

), and a microfluidic chip with a virtual wall (

). In the electronic gate, the flow is restored by applying a local potential difference between the meniscus and the pinned geometry through an integrated electrode, so that an electrically controlled and independently addressable valve array can be realized in a compact chip layout. The design of the programmable liquid circuit does not require specific rules. The placement of electronic gates and passive capillary valves can be done like adding on/off switches or transistors to electronic circuits. For example, the e-gate can be placed in a specific location that is essential for analysis, for example, in a mixing chamber to control the reaction time, reagents are selected along parallel microchannels, or a more general-purpose array can be placed in a regular array format. Chip architecture. The layout and operating principle of the e-gate, exemplary chip design, and the layout of passive microfluidic diodes, capillary shut-off valves, and capillary trigger valves are shown in Figure 1. S1. In addition to the flow control element, the microfluidic chip usually includes a loading pad that can hold about 10 μl of liquid, a microfluidic channel and a capillary pump to draw liquid by capillary force, and a vent hole.

In the manufacturing process, we prefer to pattern silicon wafers by photolithography, because the process is compatible with high-volume manufacturing, supports capillary-driven flow without post-processing, and allows electrodes and drying reagents to be easily and accurately integrated in In the sealed microchannel (

). We use moderately hydrophilic materials (for example, epoxy SU-8 and dry film resist (DFR)) (which form a forward contact angle of approximately 80° and 70° with water, respectively) as the side and top channel walls, and The e-wall is patterned. The gate trench with sharp edges (~90° angle) on the hydrophilic bottom SiO

Because this structure makes the wettability of the channel reach a good balance, it not only satisfies the Gibbs inequality condition to fix the liquid-air meniscus, but also generates a sufficiently high capillary pressure to replace the liquid to other places. The details of chip manufacturing involving three photolithography masks are provided in "Materials and Methods" and Figure 2. S2.

In our previous work, we demonstrated the basics of electronic gates that stop biological buffers, human serum and artificial urine in a semicircular groove geometry (

). Here, we further analyze the operation of the electronic gate array and its use in more complex microfluidic networks with different channel widths and hydraulic resistance. An exemplary behavior of the meniscus at the electronic gate is shown in FIG. 1. S3 and film S1 are used for channels 200μm wide and 15μm deep. The incoming meniscus is first fixed in the center of the channel, following the semicircular groove, and stabilized within 200 milliseconds by pivoting on the groove and following the contact angle of the other three walls. After applying a potential difference between the liquid (3 to 5 V) and the electronic grid electrode (ground), the meniscus will loosen from the center of the channel within 500 milliseconds, similar to releasing a bowstring. Electronic gates implemented in narrower channels (50, 100, and 150μm) exhibit similar behavior (Figure S4); however, due to the reduced radius of curvature, they may be activated at lower voltages and have faster Response time, but at the cost of lower stability (for example, an electronic gate in a 50μm channel cannot hold the liquid for more than 1 minute). And the corresponding Laplace pressure applied by the side wall increases. On the other hand, wider channels (>500μm) increase the risk of DFR sagging when sealing the microfluidic chip and filling it with liquid. In addition, our experiments show that there is no correlation between the operation of the electronic gate and the upstream hydraulic resistance. The upstream resistance determines the liquid flow rate before reaching the electronic gate. For the chip tested in this work, the range is 0.5 to 3 nl/s . After long-term storage (16 months) of the "ready-to-use" chip, we also tested the stability of the capillary drive flow and the operation of the electronic gate. Specifically, we observed that when these chips are stored in the dark in an opaque wafer box, their functions are preserved (Figure S5). Here, we also show that the electronic gate is compatible with whole blood (Figure S6A). However, they were unable to fix the blood-dissolving meniscus containing high concentrations of surfactants (Figure S6B). In order to obtain a stronger fixation effect on this challenging sample, we replaced the hydrophilic DFR with a hydrophobic adhesive film (Figure S6, C and D).

Unlike the classic EWOD technology developed to manipulate droplets (

) Or continuous flow (

), the electronic gate does not use insulated electrodes or hydrophobic coatings, because it does not require reversible and large dynamic range surface wettability adjustment. This implementation greatly simplifies the manufacturing process and reduces the voltage required to restore the flow of the stationary liquid (<5 V). This advantageous feature allows us to implement battery-powered peripherals that interface with a microfluidic chip and control via Bluetooth via a smartphone application (app). Displayed device

And detailed description in Figure 2. S7 uses a standard Arduino-based microcontroller with a size of only 70×12×25 mm

A voltage of 0 to 5 V can be generated in 1.2 mV steps, and the resulting potential difference can be applied to 2 of the 16 electrical contacts on the microfluidic chip in any combination. This can bring another kind of flexibility to chip design, because the polarity and voltage level of each electrode can be programmed via a smartphone. We have developed two applications that communicate with the device. One allows manual control of parameters (for example, the applied voltage and the amplitude and duration of the selected electrode pair), the other works autonomously based on a programmable protocol that can be uploaded to the application. In a typical workflow, you only need to press a button on the app to establish Bluetooth communication between the phone and the device. Insert a microfluidic chip with 16 electrical contacts into the device, just like connecting a universal serial bus ( USB) memory is inserted into the computer just like. Then, pipette a liquid sample with a volume between 2 to 10 μl onto the chip, control its flow conditions by telephone, and discard the chip after use (movie S2).

(

) Insert a capillary-driven microfluidic chip with the same design into the peripheral device, and pipette 3μl of PBS containing blue dye onto the chip (left), use the smartphone application to manually activate the electronic gate (middle), and click Microscope images were captured in the process of activating 12 electronic gates (right) row by row (chip 1) and column by column (chip 2). The first microscope image taken before pipetting is subtracted from the other liquids to highlight the position of the liquid in the flow path. (

) A "microfluidic clock" running autonomously based on the protocol of a mobile phone application. The picture on the left is a picture of the chip next to the pencil before drawing 3μl of PBS solution with red dye and inserting the chip into the device. The liquid flows through a 100μm wide and 15mm long channel and fills the center of the clock within 20 s (inset: an oblique macro photo of the meniscus fixed at the 200μm wide electronic gate). On the right is a microscope image of the chip, and the smart phone automatically activates the electronic gate every 5 minutes, representing the minutes of the 1-hour clock (see also movie S4). Image credit: Yuksel Temiz, IBM Zurich Research Center.

In the following, we demonstrate the flexibility of programmable liquid circuits and the precise flow control achieved using a smartphone.

The figure shows the application interface and a microscope image of a chip design with three parallel channels, each channel having four electronic gates connected in series between the areas of the simulated reaction chamber. Here, two chips with the same design are used to generate two different process sequences by manually activating the electronic gate from the application (the complete sequence is provided in movie S3). We also used the same strategy to implement a more complex microfluidic circuit, which includes multiple options for flow paths, bypass channels, and microfluidic diodes (Figure S8). Relying on the reproducible and reliable operation of the programmable liquid circuit, we took advantage of this concept by designing a “microfluidic clock” with a diameter of only 5 mm, which automatically runs for about 1 hour without user intervention, and takes 5 minutes as Interval display time (

). Here, a 22-step protocol (including 3-s e-gate activation duration at 3 V and 297 s waiting time of 11 e-gates) is constructed as a form and uploaded to the application, and then executed automatically These steps. Except for the last electronic gate, the microfluidic chip and peripherals work according to the protocol. The last electronic gate fails after holding the phosphate buffered saline (PBS) solution for 50 minutes instead of 55 minutes (instead of 55 minutes) (movie S4). For many POC diagnostic applications, this retention time is long enough, but if necessary, multiple electronic gates connected in a cascade configuration can be used, wider channels for higher stability and lower hydrophilicity DFR or hydrophobic treatment to extend retention time. Palladium electrode

).

In the experiments shown so far, we manually activated the electronic gate or applied the protocol while monitoring the liquid position using a microscope. However, in potential end-user applications, optical surveillance may not be available. In addition, multiple electronic gates connected in different configurations, such as multiple electronic gates connected in parallel and other electronic gates connected in series, may require different voltage levels or activation times. Therefore, it is ideal to continuously monitor the flow conditions and have a feedback mechanism to synchronize the activation of the electronic gate with the actual position of the liquid. Based on our previous work using capacitance measurement for flow monitoring (

), we can detect the liquid reaching the electronic grid and its activation by measuring the capacitance of the electronic grid electrode, without using other components, manufacturing steps or external tools. The principle is based on measuring the change in double layer capacitance, which is proportional to the area of ​​the electrode in contact with the liquid. The use of the above-mentioned peripheral devices for capacitance measurement only requires the addition of two resistors and an analog switch. After a 5 V pulse is applied to the selected electrode pair through a 10 MΩ external resistor and when the voltage on the electrode pair reaches 1 V, the peripheral device calculates the capacitance based on the resistor (RC) time constant. , The capacitor is discharged to 0 V through the 1 kohm resistor for the next measurement. Keeping the maximum voltage of the electrode at 1 V minimizes the risk of undesirable electrochemical effects and activation of the electronic gate during capacitance measurement. Since the activation of the electronic gate and the measurement of the capacitance require different voltage levels to be applied to the same pair of electrodes, we use an analog switch controlled by a microcontroller to decouple its operation.

Figure 1 shows the capacitance measurement results when three e-gates are activated in sequence. As the liquid moves along the channel from the loading plate, the portable peripheral device starts to operate in "flow monitoring" mode and records the measured capacitance at a sampling rate of 1 Hz. When the meniscus is pinned to the electronic gate, the capacitance increases by an average of 13±3 pF. Then, switch the system to run in the "electrochemical" mode, restore the flow by applying a voltage of 5 V for 1 second, and immediately return to the flow monitoring mode. Since the electrode is partially wetted during the pinning process and completely wetted after activation (Figure S9A), the restoration of the flow will result in an average increase of 5±3 pF, which can be used to detect the successful activation of the electronic gate. If necessary, larger electrodes can be used or averaged to improve the signal-to-noise ratio, but at the cost of a slower response time. It is expected that the measured capacitance value will vary according to the composition of the liquid, the geometry of the electrodes and the distance between the electrodes. However, this variability should not be a concern for this work, because we believe that the relative change of the measured value rather than the absolute value is an indicator of the wetting of the electronic gate. For example, in

, We have proposed a simple algorithm that can be used to automatically activate the electronic gate and adaptively adjust the applied voltage and its duration according to the change in the measured capacitance. We also demonstrated that capacitance measurement can be used to detect the insertion of a chip into a peripheral device, pipette samples, and monitor multiple electronic gates (

). Theoretically, this method can be used to measure from hundreds of e-gates at 1-Hz sampling rate using time division multiplexing, because the measurement usually takes less than 1 ms (for example, 446 μs for 200 pF). Figure 5 shows a situation with 15 electronic gates. S9B. Such a system is powerful and can correct user errors, operating failures or deviations, and automatically generate reports and test for each device. The system is powerful and can correct all actions and liquid positions of users over a period of time. These functions are especially important for POC diagnostic applications.

) Microscope image showing three electronic gates (left) when activated sequentially. The capacitance of each electronic gate (C

), that is, during the activation of the e-gate (right), the capacitance measured between the common electrode and the e-gate electrode in the loading pad is continuously measured. (I) The capacitance remains at 127 pF before the liquid reaches the electronic gate. (Ii) C

After activating e-gate 1 (not shown in the microscope image), the sample concentration rose to 140 pF and the sample reached e-gate2. (Iii) Then E-gate 2 is activated, and C

As the surface area of ​​the electrode in contact with the liquid increases, the current density increases to 145 pF. The capacitance of electronic gate 3 (C

) And e-gate 4 (C

) After activating (v) the electronic gate 3, the sample reaches the electronic gate 3 (iv), and the liquid reaches the electronic gate 4 (vi), the sample shows a similar behavior. (

) The algorithm flow chart that can be automatically used to activate the electronic gate based on the feedback of the capacitance measurement. (

) The capacitance measured from the two electronic gates shows the change in capacitance (ΔC) after the chip is inserted, the sample is drawn, and the meniscus reaches the electronic gate (i) and the electronic gate is activated (ii and iii). The top microscope image highlights the location of the meniscus. Scale bar, 500μm. Image credit: Yuksel Temiz, IBM Zurich Research Center.

Many microfluidic applications require several liquids to flow in parallel under well-defined flow conditions, such as multipath

) Or high throughput (

) Determination. The chip design shown so far allows the use of up to 15 electronic gates to control the stop-and-go flow control of a liquid, which is limited by the 16-contact pogo pin connector. Peripheral electronic devices can be easily scaled to handle more electronic gates, but this will increase the area of ​​the chip to accommodate more external electrical contacts. Instead, we tried to increase the number of e-gates by connecting multiple e-gates to the same contact (ground electrode), and increase the liquid by adding liquid selection electrodes (positive electrodes) to their respective loading pads quantity(

). Here, the electronic gates of all liquids are connected to each other at a given position. The liquid flow at this location is triggered by applying a potential difference between the liquid selection electrode and the common electron grid electrode, which is very similar to row/column selection in a matrix. Using microfluidic chip (1×2 cm

), with four loading plates, microchannels (similar to a racetrack configuration) and a total of 100 electronic gates (

). Each channel has 25 electronic gates connected to 12 external electrical contacts. As shown in the figure above, pipette the colored PBS solution to the corresponding loading pad (about 1.5μl for each pad).

. The liquid can be controlled individually by selecting the corresponding electrode pair from the smartphone app. In order to activate the second e-gate connected to the same contact, the flow of liquid can be restored by applying the same voltage level (5 V), but the activation delay will be slightly longer. We tested this delay using a design with 10 parallel e-gates (Figure S10). The activation delay of the 10th electronic gate gradually increased from 0.5 s to 10 s. This may be due to the voltage drop and the increase in capacitance on the previously activated electronic gate, resulting in a higher RC time constant. When the liquid passes through the electronic gate and moves to the next electronic gate in the channel, the measured capacitance increases (

) Can be used to track the position of each liquid and reduce activation delay. Moreover, for many applications involving cell biochemical processes or experiments that often take tens of minutes, these delays are unlikely to be a problem.

) By selecting the electrode in the liquid (E

) (Composition in the loading plate) and e-gate selection electrode (E

), which is common to all liquids at a given location and connected to the ground. Multiple electronic gates can be electrically connected to the same external contact, for example, two consecutive electronic gates are connected to the same electrode, as shown in the figure. (

) A photo of a tested chip that contains four independent channels of similar races, each with 25 electronic gates connected to 12 contacts. The upper part shows the configuration of four loading pads. The four loading pads provide the stained PBS solution to the corresponding channels and the liquid selection electrode (E

) The pattern is on the loading board. The inset shows an enlarged image of a stained sample controlled by a set of electronic gates and focused on a yellow solution: before reaching the electronic gate (frame A), stopping at the electronic gate (frame B), and after activation (frame C). (

) The above figure shows the indicated value of a solution before reaching the electronic gate (green, box A in group (B)), and when it reaches the solution, the indicated electrode at the electronic gate and the loading pad The capacitance measurement between the electrodes stops [red, (B) box B in the illustration]. The continuous vertical black line represents the voltage (5 V) applied to activate the electronic gate. The figure below shows the difference in capacitance measured before and after the meniscus reaches the electronic gate. Image source: Yulieth Arango, IBM Zurich Research Center.

Performing assays in microfluidic systems, such as enzyme-linked immunosorbent assay (ELISA), requires a seamless sequential flow of liquids with various reagents and analytes. Specifically, there should be no bubbles or uncontrolled liquid interface between two liquids that flow continuously in the same flow path. The use of active pumping and/or mechanical valves at the expense of manufacturability, portability, and user-friendliness has been extensively studied for the complex manipulation and route selection of this type of multiple liquids, and the passives that are being developed for POC diagnosis This is still a long-term challenge for microfluidic systems. Safavieh and Juncker proposed an outstanding example of the complicated flow control in capillary-driven microfluidics, the concept of "capillary" (

). Although the capillary valve has been used to demonstrate the sequential flow and reverse flow of pipetting reagents, the flow conditions have been pre-programmed in the design and manufacture of the chip, and the time required to precisely adjust the volume of the pipetting liquid, and the implementation process is consistent with PDMS Capillary-driven applications in microfluidics. In contrast, electronic gates use materials that are inherently compatible with capillary-driven flow, and flow conditions can be programmed after manufacturing. However, only the electronic gate cannot achieve liquid consolidation and sequential flow. This can be solved by combining an electronic gate with a capillary valve and partial electrolysis of water at a specific electronic gate to divide the liquid. Provides a remarkable example

, Where a series of input liquids can be selected to pass through a common flow path and be classified to a specific outlet channel (

). The concept is based on two key features: (i) the use of capillary valves with vents to incorporate liquid control into a channel already filled with another liquid; (ii) the use of electrochemical methods to generate liquid to prevent liquid flow using electronic grids The electrode generates bubbles (

The concept shows a variety of liquids, which are controlled by electronic gates and flow through common channels in any combination and in any order. (

The capillary valve with vent allows multiple liquids to be combined without generating bubbles, and the liquid flow is controlled by an electronic gate. In the diagram involving two liquids, liquid 1 (L1) flows in one channel and is connected to the second channel through a capillary valve (left). The flow of Liquid 2 (L2) is restored by activating its electronic gate (middle). Then, by pushing the air out through the vent, it merges into L1 at the capillary valve, which has another valve that prevents L2 from entering the vent. Both liquids flow through the common channel until the flow of L1 is blocked by the bubble, which is formed by hydrolysis by applying a sufficiently high potential difference on the electronic gate (right). (

) The experimental results show that the liquid flows sequentially through the common channel. (I) Pipette PBS solutions of four colors (transparent, red, green and blue) onto their respective loading pads and fix them with electronic gates. First, the blue PBS flow starts, fills the common channel, and stops at the valve connecting the other channels. The other electronic gate located behind the common channel is used to direct the liquid to one of the four outlet channels. (Ii) The flow of the blue PBS is completely blocked by the bubble, which is generated by applying +15 V DC between the electron grid electrode (positive electrode) of the loading electrode (ground) and the common electrode. Then, by activating its e-gate, merging it into the common channel by keeping the PBS blue capillary valve to start the red PBS flow. (Iii and iv) Similarly, the remaining two liquids (green and transparent) flow through the common channel in sequence after blocking the flow of the previous liquid and activating the corresponding electronic gate. The arrow that matches the color of the liquid highlights the flow direction, and the entire sequence can be seen in movie S5. Image credit: Yuksel Temiz, IBM Zurich Research Center.

We demonstrated a test chip that uses a test chip for controlled merging and sequential delivery of liquids. The chip has four loading pads, a common channel, four outlet channels, and two for each liquid before and after the common channel. Electronic gate (Figure S11, A and B).

Movies S5 and S5 show the experimental results of sucking 2 μl of colored PBS solution onto the chip and fixing it in place through their respective e-gates. Different from the classic capillary-driven microfluidic system, in the traditional capillary-driven microfluidic system, a variety of liquids should be introduced one after another according to the required procedures. Here, all liquids can be piped at the same time, thus completely controlling the smart phone procedures. A more flexible, synchronous and error-free operation is realized. After activating two specific electronic gates, the first liquid (blue in this case) starts to fill the common channel and the target outlet channel, but is still held by capillary valves in other positions. This design is symmetrical, you can choose any liquid to pass through first. Then, by applying a potential difference of 15V between the liquid (ground) and the electron grid electrode (positive electrode) for several seconds to generate bubbles, thereby stopping the flow of the first liquid. Start the flow of the second liquid (red) by merging the second liquid (red) through the capillary valve and the exhaust hole into the common channel, where the first liquid has been fixed. Repeat the same process for the rest of the liquid. By incorporating new liquids into channels that are already filled with liquid (movie S6 and Figure S11C), the concept can also be used to create a controlled downstream flow of multiple liquids in multiple combinations.

The flow inside the microchannel requires replacement of air unless the channel remains open (

) Or made of permeable material (

). Some microfluidic devices may require a small vent or a long path extending from the channel exit to the vent (for example, the chip in the chip

). For more complex microfluidic circuits involving multiple liquids and capillary shut-off valves with vent holes, arranging multiple vent holes to the edge of the chip may become a serious design challenge. for example,

Must be returned to the side of the loading plate. In some cases, improper placement of the vent may cause malfunctions (for example, liquid blocking the vent), or even worse, the body fluid sample may leak outside the chip and cause biosafety issues. Sometimes a vertical opening is formed by the sealing layer and the packaging, but this is troublesome in terms of manufacturing. Programmable liquid circuits provide excellent control and flexibility for manipulating liquids, but without a new exhaust strategy, they cannot be combined arbitrarily. To this end, we introduced the concept of "self-ventilation channel" and proposed a manufacturing process compatible with electronic gates, which is described in detail in "Materials and Methods".

The concept is based on partial wetting of the cross-section of the flow path, similar to liquid microchannels formed using wettability contrast (

) Or virtual electrowetting channel (

). We easily achieve this by creating side stripes of black silicon in the bottom layer of the closed microfluidic channel. Figure S12A shows the working principle and experimental results of the automatic exhaust mechanism. Movie S7 shows that the meniscus of two PBSs flow in opposite directions and merge without trapping bubbles. The self-exhausting channel can be combined with a trench containing black silicon, and the liquid is fixed at the electronic gate using a common deep reactive ion etching (DRIE) step (

). The electrical insulation of the liquid from the black silicon and silicon wafer substrate is achieved by depositing a conformal thin layer of Al

Ø

(Figure S12B). This insulation is important to ensure multiplexing capabilities because it can activate selected electronic gates without crosstalk with other electronic gates on the silicon substrate. The generated electronic gate is fully functional (activated by 4-V for about 1 s;

).

An exemplary design is shown in which the trench for the e-gate and the pinning structure for self-venting are patterned on black silicon and the wafer surface is passivated with Al

The patterns of the electrodes and microfluidic channels are as described above.

Movie S8 and Movie S8 show the experimental results of this configuration in which two liquid meniscuses are combined in different combinations using a smartphone application.

) Side view of a component that uses black silicon areas instead of trenches for e-gate to be electrified. (

) The microscope image shows the stop and stop control of the liquid flow. (

) A three-dimensional (3D) rendering showing that Liquid 1 and Liquid 2 fixed on the e-gate merge into Liquid 1, without generating air bubbles due to the side vents patterned on the black silicon. (

) An example of merging two liquids stopped at the e-gate (upper frame): The microscope image on the left shows the merging without bubbles after activating the e-gate of Liquid 2, while the image on the right shows two e -Use the common electrode to trigger the gate at the same time (from movie S8). Image source: Yulieth Arango, IBM Zurich Research Center.

The concept of programmable liquid circuit introduced here (i) can realize a universal, programmable and portable microfluidic system that can be manufactured using high-throughput technology; (ii) is simple to use; (iii) can quickly and reliably control a variety of Liquid; (iv) Provide real-time feedback on liquid flow conditions. This concept combines electronic gates with passive microfluidic valves and self-exhausting channels, which has many advantages over other liquid processing technologies. First, compared to microfluidic systems with active pumping of liquids, programmable liquid circuits do not require any labor-intensive preparations, such as preparing syringes and manually connecting tubing to the microfluidic chip, they can handle small amounts of samples or reagents . The capillary force is used to drive the liquid without using any peripheral equipment to generate the pressure gradient. Passively stop the flow by fixing the capillary in the groove. Therefore, there is no need for external energy to keep the liquid in place, which can allow the necessary time for the functions normally performed in microfluidics (such as dissolving reagents, reacting, separating analytes, stimulating cells, etc.). The activation of the flow can be achieved through application 3. The time from 5 V to the selected electronic gate is less than 1 s. This allows us to control many electronic doors using a battery-powered device that is about the size of a matchbox and communicates with a smartphone via Bluetooth. We also show that after 16 months of storage, the operation of the chip will not be affected. Second, the capacitance measurement can be used as a feedback to the user (for example, a warning about a malfunction) to monitor each electronic door individually in real time. In addition, this feedback can be used in algorithms to optimize electronic gate parameters and run the measurement protocol completely autonomously. Third, multiple electronic gates can be connected to the same external electrical contact to control the continuous flow of multiple liquids at multiple locations, which is difficult to achieve in a digital microfluidic system using EWOD.

Like many other technologies, programmable liquid circuits have limitations. First, when using solutions with surfactants, human serum or whole blood, the fixing ability of the electronic gate is challenged. The geometry of the channel (for example, a wider channel) can be changed to make the electrode hydrophobic (

), or use a lower hydrophilic film as the top sealing layer (Figure S6). Secondly, unlike mechanical valves, electronic gates can only be activated and deactivated once unless the bubbles formed to prevent flow are removed by electrochemical methods. If you need to stop and repeatedly restore the flow of liquid, you can pattern a series of electronic gates along the flow path (preferably in a capillary pump). In addition, these chips are not meant to be reusable, not only because it is impractical to dry the fluid-filled channels, but it is also undesirable to reuse these chips for POC applications that may cause cross-contamination between patient samples. one question. Third, although it takes less than 1 s to activate a single electronic gate, connecting more electronic gates to the same electrode will cause an additional delay in the activation time (for example, 10 electronic gates require 10 s). If fast response is critical to the application, this delay can be reduced by having electrodes bias the liquid near each electron gate instead of just placing it in the loading pad. By reducing the distance between the two electrodes and therefore the resistance, this will also help improve the accuracy of the capacitance measurement used for feedback. Fourth, we use silicon wafers and clean room processes to manufacture these circuits, which are generally considered "expensive" and not easily available to research groups engaged in microfluidics research. For high-end complementary metal oxide semiconductor (CMOS) chips (for example, microprocessors) or small-volume prototypes, this statement is usually true. However, there are many examples of microfluidic realizations that benefit from the unique characteristics of silicon and the sophisticated manufacturing infrastructure developed for processing silicon (

), this is Abbott’s i-STAT POC device, which is very successful commercially. In addition, our process is simple and efficient: it only requires three photolithography masks to manufacture and seal hundreds of chips at the same time, and uses precision advantages to install many electronic gates on a small footprint. All of these may help reduce costs. The process is also compatible with CMOS. Therefore, high-density electronic gate arrays can be post-fabricated on electronic chips, and row/column selection can be used for individual addressing. In addition, we believe that the concept shown here can be implemented with other materials, such as glass or hydrophilic polymers, as long as the manufacturing process can pattern the capillary pinning structure and the integrated electrode.

Programmable liquid circuits bring exciting possibilities for laboratory-on-chip applications: microfluidic devices can have a common architecture, can be controlled on demand with a smartphone, or can be fully automated through an easily changeable protocol. For example, keeping liquids for tens of minutes can be used to study cellular or reaction kinetics in immunoassays and molecular assays. Programming the delivery of samples, reagents, and rinsing buffers that flow through the same channel sequentially or in parallel can eliminate the need for manual driving or sliding mechanisms (for example, V-chip technology (

)], a kit with reagents and an advanced disk laboratory design. The programmable liquid circuit can also be combined with our newly invented self-coagulating module (

), in order to realize the programmable and precise time and space control of the dissolution of integrated reagents in the chip sample extraction/response device. A particularly interesting application can be multiple homogeneous analysis, such as analysis to measure multiple enzymatic reactions. Using a single reader to synchronize reagent flow and measure the reaction kinetics of multiple analytes is often challenging. Here, the programmable liquid circuit can ensure that all reagents reach the required area in time, and the reader can be used to quench specific enzymatic reactions when needed to obtain a stable endpoint signal. In addition, the concept can be used to pattern biomolecules on the surface using cross flow without manually manipulating the channels, such as the highly multiplexed micromosaic immunoassay (

). Self-emptying channels can be used to solve one of the main failure mechanisms in microfluidics, which is the process of filling microchannels, combining multiple liquids, or forming bubbles due to detection procedures (eg, electrochemistry, thermal cycling, etc.). PCR, gaseous by-products). The concept also provides interesting possibilities for counterflow involving programming different liquids. The liquid can be brought into contact with a well-defined location, and the initial time and reaction at the interface between the liquids can be accurately studied or implemented, such as crystallization process, diffusion process or kinetics (enzymatic reaction).

In summary, the versatility, simplicity and scalability of this concept provide compelling technology for the development of portable, smart and interactive lab-on-a-chip equipment that can automatically manipulate multiple liquids with minimal human intervention.

Use L-Edit software (Mentor Graphics) to design the layout of the microfluidic chip and convert it to the GDSII file format. The DWL-2000 (Heidelberg Instruments) laser lithography system was used to transfer the layout to the glass/Cr photomask (NanoFilm, USA). These chips are made on 4-inch Si wafers purchased from Si-Mat (Germany) (single-side polished, 525±25–μm thickness, N/Phos doping, 1 to 10 ohm·cm resistivity, < 100> direction). Mask aligner (MA6, SÜSSMicroTec AG, Germany) with i-line UV power density (13 mW/cm)

) Used to expose photoresist. Use AZ 4533 (MicroChemicals GmbH) positive photoresist with a thickness of 3.3μm as a mask for SiO etching

Layer and pattern the metal electrode using a lift-off process. The photoresist was spin-coated at a speed of 4000 rpm for 40 s, baked at 110°C for 1 minute, and then exposed to ultraviolet light (dose of 91 mJ/cm

), and develop with AZ 400K developers. The microfluidic channels were fabricated by patterning 15μm thick SU-8 (SU-8 3010, MicroChem Corp.) using an optimized formula to achieve the smallest feature size and spacing less than 5μm. The recipe includes spin coating (40 s at 1500 rpm), baking (65°C 2 minutes + 95°C 5 minutes), exposure (85 mJ/cm

), post-exposure baking (65°C for 1 minute + 95°C for 2 minutes), developing [use propylene glycol methyl ether acetate (PGMEA) for 90 s], and finally hard baking (150°C, 5 minutes) .

The production process of the chip is shown in the figure

Involves three photolithography steps, used for (i) etching semicircular trenches in SiO

(Ii) Pattern the electrode using a metal lift-off process, and (iii) Pattern the microfluidic channel in SU-8 (Figure S2). First, a Si wafer of 3μm thick SiO is used

Clean with O

Plasma (600 W, 600 sccm, 5 minutes) and treatment with hexamethyldisilazane vapor at 110°C. After patterning AZ 4533 using photomask SiO for trench

In an inductively coupled plasma etching tool (​​PlasmaPro 100, Oxford Instruments), anisotropically etch to a depth of 1.9μm at an etching rate of 250nm/min. After removing the photoresist mask in O

Plasma ashing and HMDS processing of the wafers patterned a new AZ 4533 layer for the metal lift-off process. Use electron beam evaporation (Evatec BAK 501) to deposit 5nm thick Ti (adhesion layer) and 80nm thick Pd (electrode layer), while tilting the wafer about 40° and rotating it at 20 rpm to ensure that the SiO etching Conformal coating of the metal layer above the trench

. The wafer was then immersed in acetone for 10 minutes to expose the electrodes (peeling). Gentle and short (5 to 10 s) ultrasonic stirring is performed every minute to break the continuous metal layer and accelerate the lift-off process. The wafer is rinsed with isopropyl alcohol (IPA) and dried with N air flow

. O after 2 minutes

Plasma treatment (200 W, 200 sccm) and dehydration at 150°C for 2 minutes, 15μm thick SU-8 patterned the microfluidic structure. Before cutting the wafer to half thickness, use AZ 4533 to protect the wafer surface. The wafer is then cleaned in acetone and IPA to remove the photoresist protective layer and residues from the cutting process. A rectangular strip of DFR with a thickness of 50 μm [DF-1050 of EMS Inc.] was aligned and laminated on the microfluidic structure at 45°C. Finally, the "ready-to-use" chips are cut into individual pieces by manually cutting the wafer from the dicing cut. After lamination, DFR is neither exposed to ultraviolet light nor temperature. For long-term storage experiments (Figure S5), the laminated chips were stored in an opaque wafer box and tested after 16 months. In order to electrochemically treat the dissolved blood (Figure S6, C and D), we replaced DFR with a hydrophobic adhesive film (ARseal 90880, Adhesives Research Inc.) to achieve a stronger capillary fixation at the electronic gate.

The microfluidic chip (Figure S12A) designed to demonstrate the self-venting mechanism was fabricated on a Si wafer with 200 nm SiO

Floor. First, wet etch the pattern to be converted into black silicon into SiO

The wafer is immersed in (1:10) buffered hydrofluoric acid (BHF) solution for 3 minutes to form a layer. Then, a "physical channel" was constructed in SU-8. The wafer part is cut to half the thickness, then cleaned in acetone, and then cleaned in IPA. Black silicon is formed by etching the Si area using DRIE (Alcatel AMS 200). Silica

The SU-8 layer serves as a hard mask for this process. DRIE is performed after partial cutting to protect the fragile needle-like black silicon structure (surrounded by hydrophobic C)

F

Polymer passivation). Since the wafer is still in a single-piece state after being partially cut, the DRIE and DFR lamination steps can be completed at the wafer level without using a supporting substrate or wafer. Finally, the sealed chips are cut into individual pieces by cutting the wafer, and can be used without further post-processing.

For microfluidic chips that combine electrochemical and self-venting (

), by creating a semicircular pattern of black silicon instead of etching trenches to achieve a fixed geometry of the electronic gate. The manufacturing process (Figure S12B) starts with a Si wafer without SiO

Layer, then (i) create a black silicon area in DRIE, (ii) remove the residue of photoresist mask and C

Polymer layer using O

Plasma ashing, (iii) 15 nm Al deposition

Use atomic layer deposition (FlexAL, Oxford Instruments; deposition rate, 0.12 nm per cycle) to passivate the entire wafer, (iv) pattern the metal electrode in the SU-8, and (v) perform the microfluidic structure Composition. Then, the wafer is protected with photoresist, cut into small pieces, cleaned, sealed and diced as previously described. Here, the cutting and subsequent cleaning process will not adversely affect the mechanical stability of the black silicon layer covered with Al

And photoresist protection.

Android applications use DroidScript (

)platform. The first application allows manual control of the parameters of the selected electrode pair as well as the applied voltage and its duration, while the second application autonomously executes a protocol containing a list of electrochemical and flow monitoring steps. In these two applications, after a Bluetooth connection is established between the application and the peripheral device, and when a button is pressed to activate or deactivate electrochemical or flow monitoring, the application connects all parameters in a string, and It is sent to Peripheral Device Object Notation (JSON) format in JavaScript. For example, to apply a potential difference of 3.5V between contacts #1 and #16 for electrification, the string {"PositivePin":"1","GroundPin":"16","Electro-gating":"1 ","FlowMonitoring":"0","Voltage":"3500"} to the peripheral device. It is preferred to use the JSON file format to easily add or delete the parameters of alternative versions or future versions of the platform, which may obtain parameters from the Cloud. For applications that automatically execute the protocol, according to the timing information defined for each step, the data is parsed from a CSV (comma separated value) file, indexed, and then sent to the peripheral device.

The circuit diagram and working principle of the portable peripheral can be found in Figure 2. S7. The key component of the device is a 1.27 mm pitch, 16-contact Pogo pin connector (855-22-016, Mill-Max), which can be easily and reliably electrically connected to the Arduino microcontroller chip (Pro Micro 5). V/16 MHz, SparkFun Electronics), Bluetooth module (RN-42, Microchip Technology) connected to the microcontroller via the Universal Asynchronous Receiver Transmitter (UART) bus, 12 digital-to-analog converter (DAC) (MCP4725 (Microchip Technology) Generate e-gate voltage, and two 16-channel analog multiplexers (74HC4067, Nexperia) to select an electrode pair in any combination. The device is powered by 3.7V, 110mAh rechargeable lithium polymer battery, MCP73831 charge management controller The chip (Microchip Technology) charges the battery through the USB port of the Arduino microcontroller. The boost converter XC9142 (Torex) is used to convert 3.7 V to 5 V, which is required for the microcontroller, DAC and multiplexer The Bluetooth module running at 3.3 V is powered by a linear regulator (MCP1700, Microchip Technology) and is connected to the microcontroller (5 V) through a TXB0102 (Texas Instruments) bidirectional voltage level converter. Due to For the different voltage levels involved, it is impossible to perform electrochemical and flow monitoring on the same electrode pair at the same time, so an analog switch (TS5A3166, Texas Instruments) controlled by a microcontroller is used to connect the positive pole of the microfluidic chip to the DAC or flow rate Monitoring circuit. The basis of flow monitoring is to measure the double-layer capacitance of the electrode pair by the following method: apply a 5V voltage pulse through a known resistor (1 Mohm), and measure the charging time, so that the voltage across the electrode pair reaches RC Time constant. To 1V. Then, the capacitor immediately discharges to 0V through a smaller resistance (1 kW). Because the maximum potential difference of the electrode pair is limited to 1 V, the electrochemical corrosion of the electrode is minimized during the long flow monitoring experiment. (For example, the 20-minute measurement in Figure S9). All components are assembled in a housing with a spring locking mechanism, which is designed in FreeCAD software and printed with a Stratasys Dimension Elite three-dimensional (3D) printer Finally, place the RGB LED (SK6812) inside the housing to display the status of the device by allowing light to diffuse through the translucent housing.

The code for the Arduino microcontroller uses the Arduino software (

). Five libraries are used: (i) Wire.h to implement I2C communication, (ii) CD74HC4067.h to control the multiplexer, (iii) Adafruit_MCP4725.h to communicate with the DAC, (iv) ArduinoJson.h in Transfer data between applications and peripherals in JSON format, and (v) Adafruit_NeoPixel.h to control the RGB status LED. The code uses a simple execution flow. After the variables and libraries are initialized, the code will wait until Bluetooth communication is established. The JSON format string passed in from the phone will be parsed into individual variables, which record the positive contact and ground contact of the microfluidic chip, the voltage of the electronic gate, and the operating mode (electricity, flow monitoring or idle). Then, the code sets up the multiplexer, DAC and analog switch. For example, the electrochemical function is performed by setting the DAC voltage to a given value and connecting the DAC output to the microfluidic chip through an analog switch. The code loops every 100 milliseconds to check if there is new input data from the phone and execute the selected function until it receives a string with {" Electrogating": "0, "FlowMonitoring": "0"} (Ie idle mode).). In order to record the data of the graph, all parameters and measured capacitance values ​​are sent back to the phone (Bluetooth) or computer (USB) every second.

Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich. The colored solution of PBS is prepared by dissolving disodium shikonate (blue), a vegetable red (red) and tartrazine (yellow) at a concentration of 1 mg/ml. Green is obtained by mixing the solution into blue and yellow. The blood sample used in the test chart. S6 was purchased from Transfusion Interregionale CRS in Switzerland, which provided research samples by collecting samples from anonymous donors using 9 ml test tubes coated with EDTA (anonymous number 92000). To lyse red blood cells (Figure S6, B to D), a solution of 10 mM Triton X-100 surfactant was prepared in PBS. Transfer 50μl of solution into a 200μl reaction tube. Open the lid and let the solvent dry completely. Add whole blood (50μl) and dry reagents to the reaction tube, vortex for 5 seconds, and then incubate for 5 to 10 minutes to achieve complete lysis of red blood cells.

A Nikon 1 J3 camera connected to a Leica MZ16 stereo microscope was used to record images and videos of the microfluidic chip. Time-lapse microscope image

with

And figure. S8B was post-processed using ImageJ software to enhance the contrast and highlight the location of the liquid. By color inverting all RGB images, subtracting the first image taken before pipetting the sample from the continuous image and inverting the resulting image, the best contrast can be obtained.

For supplementary materials for this article, please visit:

This is an open access article distributed under the following terms

, It allows use, distribution and reproduction in any medium, as long as the final use is

For commercial interest, and provide the original works appropriately cited.

Volume 6 Number 16

April 15, 2020

Thank you for your interest in promoting the term "scientific progress".

Note: We only ask you to provide your email address so that the person you recommend to the page knows that you want them to see the email and that it is not spam. We will not capture any email addresses.

This question is used to test whether you are a visitor and prevent automatic spam submission.

Please log in to add an alert to this article.

by

: Eaay8305

The capillary network controlled by a simple low-pressure gate provides new possibilities for fluid delivery for instant detection.

Volume 371, Issue 6526

©2021

. all rights reserved. The American Association for the Advancement of Science is 

, 

 with 

.

ISSN 2375-2548.