A biosensor is an analytical device that detects a biological or chemical target and converts this interaction into a measurable signal. To translate this signal into reliable quantitative data, the sensor must first be tested against known concentrations of the target analyte.
During development, controlled analyte concentrations are used to characterize the sensor response, establish a calibration curve and evaluate its performance. As biosensing technologies are increasingly used for continuous monitoring, wearable devices and compact diagnostic instruments, calibration and testing can involve more complex and dynamic protocols that are difficult to perform manually.
Precise fluid handling helps automate these processes. By controlling which sample reaches the sensor, at which concentration, flow rate and time, programmable pumps and valves can support biosensor development from initial testing and calibration to continuous sensing.
Biosensor calibration is the process of establishing the relationship between a known analyte concentration and the response generated by the sensor.
A sensor may produce an optical, electrical, mechanical or other measurable signal when interacting with a target analyte such as glucose, a protein, DNA or RNA. To convert that signal into a quantitative concentration, the sensor is typically exposed to several known reference concentrations.
The resulting measurements can be used to create a calibration curve, linking analyte concentration to sensor response. This relationship can then be used to determine the concentration of an unknown sample.
Calibration curve showing the relationship between target protein concentration and biosensor response. Adapted from Almeida N.B.F. et al., Beilstein J. Nanotechnol. 13 (2022), CC BY 4.0.
Calibration also provides valuable information about sensor performance, including:
A calibration curve is only as reliable as the conditions used to generate it.
If two reference solutions reach the sensing surface differently, if the previous sample is not completely removed, or if the actual concentration in the measurement chamber differs from the expected concentration, the sensor response may no longer represent the reference value accurately.
Controlled fluid delivery therefore becomes part of the calibration process itself.
In optical biosensing with microring resonators, for example, different concentrations of C-reactive protein can be sequentially delivered to a functionalized sensor chip to establish its concentration-response relationship. A rotary valve can automatically select between buffers, samples and regeneration solutions while a pump drives the liquids through the sensing chamber.
Microfluidic setup for C-reactive protein biosensing, using an AMF rotary valve to select between reagents and a syringe pump to deliver them through the microring sensor. From Boertjes N. et al., Measuring C-Reactive Protein Using Microring Resonators.
The biosensor performs the measurement. The fluidics ensures that the sensor measures the right sample under controlled conditions.
The exact calibration method depends on the sensing technology and application, but a typical workflow follows the same general principle:
Baseline → known concentration → measurement → fluid exchange → next concentration → calibration curve
The process usually starts with a reference or blank solution to establish a baseline. The sensor is then exposed to one or several known concentrations of the target analyte.
For each concentration, sufficient time must be allowed for the fluid to reach the sensing area and for the sensor response to stabilize. Between measurements, the fluidic path may need to be flushed, washed or regenerated depending on the sensing mechanism.
Biosensor response to sequential lactoferrin concentrations, with wash steps between measurements. From Michielsen C.M.S. et al., ACS Sensors 10 (2025), 2895–2905.
Repeating the process across different concentrations creates the data required to characterize and calibrate the sensor.
This becomes increasingly complex when many samples are involved.
A calibration experiment may require several standards, blanks, buffers, washing solutions and repeated cycles. Manual switching increases operator time and can introduce variability between experiments.
A programmable syringe pump combined with a multiport valve can automate much of this process:
Select reference → deliver → measure → flush → switch → repeat
The same sequence can then be reproduced across different sensors, cartridges or experimental conditions.
In continuous protein sensing research, automated calibration strategies have been used to quantify lactoferrin concentrations over extended periods. Reference measurements and signal normalization helped compensate for sensor variations and drift, while calibration curves converted the measured sensor response into concentration values.
Automated calibration strategies for accurate lactoferrin quantification in milk. From Michielsen C.M.S., Continuous Protein Sensing using Biosensing by Particle Motion: Inspired by Nature, PhD thesis, Eindhoven University of Technology, 2025.
The study also demonstrated that, under controlled conditions, a common calibration curve could be applied across multiple sensor cartridges, highlighting the value of reproducible sensor preparation and calibration protocols.
Calibration is often considered primarily a sensor or data-analysis challenge.
But before a signal can be analyzed, the target analyte must physically reach the sensing surface.
This makes fluid handling an important part of the measurement chain.
When switching from one concentration to another, the previous sample must be sufficiently displaced from the tubing, fluidic channels and sensing chamber.
Incomplete exchange can create an intermediate concentration that changes over time rather than the expected step change. The sensor may then be accurately measuring the fluid in front of it, but that fluid is not the concentration the experiment assumes it to be.
Residual analyte from a previous sample can affect the following measurement. This becomes particularly important when switching between widely different concentrations or working with highly sensitive biosensors. Low internal volumes and efficient washing protocols can help reduce the amount of residual sample remaining in the fluidic path.
Changes in flow conditions may also influence the sensor response or the time required to reach equilibrium. Stable and repeatable flow makes it easier to compare measurements performed at different concentrations or on different sensors.
Automating sample selection and delivery reduces the number of manual interventions required during calibration and helps reproduce the same protocol over multiple experiments.
For biosensor calibration, precise fluidics is therefore not only about moving liquid. It is about creating repeatable measurement conditions.
Continuous protein sensing experiments have shown that incomplete fluid exchange can contribute directly to signal drift. Slow equilibration after a sample change can create time-dependent concentrations inside the measurement chamber and increase measurement variability.
Repeated lactoferrin measurements showing signal variability over time in buffer and milk samples. The observed drift was partly attributed to incomplete fluid exchange and slow diffusive equilibration. From Michielsen C.M.S. et al., ACS Sensors 10 (2025), 2895–2905.
Optimizing the fluid exchange protocol can therefore improve not only automation, but the quality of the sensor data itself.
A traditional calibration curve typically measures a series of fixed concentrations.
But many real applications are not static.
A wearable biosensor, continuous monitor or process sensor may encounter concentrations that rise, fall or fluctuate over time. Testing only isolated concentration points does not necessarily show how the sensor will behave under these conditions.
Dynamic sensor testing can reproduce more realistic concentration profiles, including:
This allows developers to study response time, reversibility, hysteresis, drift and recovery in conditions closer to the final application.
Generating such profiles manually becomes difficult very quickly.
Each transition must happen at the correct time while maintaining controlled flow conditions and avoiding contamination between samples.
Programmable fluid handling makes it possible to automate these sequences and repeat exactly the same profile across several sensor prototypes.
For wearable sensor development, for example, known glucose concentrations can be delivered according to predefined profiles to compare the expected concentration with the sensor response.
The same principle can be applied to other biomarkers or to sensors integrated into life science instruments: instead of asking only “Does the sensor detect this concentration?”, engineers can ask “Can the sensor accurately follow how this concentration changes over time?”
Continuous biosensing introduces another challenge: calibration is no longer necessarily a one-time operation performed before measurement.
A sensor operating for hours, days or longer may experience changes in its response over time.
These can originate from the sensing surface, molecular interactions, environmental conditions, matrix effects or the sensor itself.
For quantitative continuous sensing, developers therefore need to understand whether the original calibration remains valid and whether reference measurements or recalibration strategies are required.
Known reference samples can be periodically introduced to evaluate the evolution of the sensor response. Normalization strategies can then compensate for some forms of signal drift and improve the relationship between the measured signal and the actual analyte concentration.
The frequency of these reference measurements depends on the stability of the sensor and the accuracy required by the application.
Research on continuous protein monitoring has demonstrated automated calibration over more than 12 hours of sensing. By combining known reference concentrations with signal normalization, lactoferrin concentrations could be quantified across multiple sensors with deviations from reference measurements below 10% in the studied concentration range.
Six consecutive calibration cycles over 13 hours, using reference samples for signal normalization during continuous lactoferrin sensing. From Michielsen C.M.S., Continuous Protein Sensing using Biosensing by Particle Motion: Inspired by Nature, PhD thesis, Eindhoven University of Technology, 2025.
This illustrates an important transition: fluidics can support not only initial biosensor calibration, but also the maintenance of reliable quantitative measurements during continuous operation.
Once the sensor has been characterized and calibrated, fluidics can continue to play an active role in the biosensing instrument. Depending on the sensing principle, this can include:
sample delivery → reagent switching → washing → regeneration → reference measurement → continuous monitoring
Some biosensors require washing or chemical regeneration between measurements.
Others are designed with reversible molecular interactions, allowing the sensor to respond continuously to increasing and decreasing analyte concentrations without an active regeneration step.
In both cases, controlled sample delivery remains essential.
A programmable pump can sequentially expose the sensor to different samples or concentrations while the sensing system continuously records its response.
A multiport valve or integrated valve-pump architecture can extend this approach to several samples, buffers and reagents without requiring manual reconnection.
A miniaturized single-molecule biosensing platform developed at Eindhoven University of Technology integrated automated microfluidics to continuously track changing concentrations of a cancer-associated nucleic acid target.
A programmable pump was connected to several analyte concentrations, buffer, waste and the sensing substrate. Automated switching allowed a complete concentration profile to be delivered while the optical system continuously monitored the sensor response.
Automated microfluidic integration for continuous monitoring of an EGFR Exon 19 deletion concentration profile. An LSPone programmable syringe pump delivers changing analyte concentrations to the sensing substrate while the biosensor response is monitored over time. From Valk K., Lamberti V. & Zijlstra P., npj Biosensing 2, 42 (2025).
This is where biosensor calibration and biosensing begin to converge: the same fluidic architecture used to characterize the sensor can become part of the automated sensing platform itself.
The need for controlled sensor testing and calibration extends across a wide range of sensing technologies and applications.
Wearable devices for glucose and other biomarkers need to convert sensor signals into reliable physiological measurements.
During development and validation, controlled reference concentrations can be used to characterize sensor sensitivity, response time, dynamic behavior and reproducibility.
Compact diagnostic instruments increasingly combine sensing, microfluidics and automated data analysis.
Precise fluid handling can support calibration during development and later manage samples, buffers and reagents within the instrument.
Continuous biosensors aim to follow concentration changes rather than provide a single measurement.
Dynamic calibration, reference measurements and controlled sample exchange can help evaluate and maintain their quantitative performance over time.
Optical, electrochemical and particle-based biosensors can detect proteins, DNA, RNA and other biological targets.
Automated delivery of known concentrations enables dose-response measurements, performance characterization and comparison between sensor designs.
Sensors are also increasingly integrated into experimental platforms to monitor or control biological processes.
Even when the sensor itself is not strictly a biosensor, the same principle applies: controlled reference conditions can be generated upstream to characterize and calibrate the sensing element before it is used in the biological experiment.
There is no universal fluidic architecture for biosensor calibration.
The appropriate solution depends on the number of fluids, required flow rates, sample volumes, calibration protocol and level of automation.
However, several parameters are particularly important.
Calibration relies on comparison. The fluid delivery system should reproduce the same conditions from one measurement or sensor to another.
Many biosensing chambers operate with small volumes. Controlled low-flow delivery helps manage sample consumption while maintaining predictable exchange conditions.
Reducing the volume between samples and the sensing chamber helps shorten fluid transitions and limit the amount of reagent required for flushing.
Low carryover becomes particularly important when alternating between different analyte concentrations.
Calibration frequently requires more than one sample.
A multiport rotary valve can provide automated access to several standards, samples, buffers or regeneration solutions without building a complex network of individual on/off valves.
From a simple dose-response curve to a dynamic concentration profile lasting several hours, programmable fluidics allows protocols to be repeated with limited operator intervention.
Together, these parameters define the fluidic architecture required to turn a calibration protocol into a reliable and repeatable automated workflow.
Biosensor development can start with a laboratory setup and progressively evolve towards a fully automated or embedded instrument. AMF provides fluidic components for both stages, from programmable benchtop testing to compact OEM integration.
From research and calibration…
For early sensor development, the LSPone laboratory programmable syringe pump combines precise liquid handling with multi-fluid selection, making it possible to automate calibration sequences, concentration profiles, washing steps and continuous sensing experiments.
This approach has already been demonstrated in biosensing research. In a miniaturized single-molecule biosensing platform developed at Eindhoven University of Technology, an LSPone automatically delivered a sequence of six analyte concentrations, from 0 to 250 nM, together with buffer and waste handling. The automated fluid exchange enabled the sensor to follow changes in the concentration of an EGFR cancer-associated marker over approximately 90 minutes.
In another continuous protein sensing setup, an LSPone combined with a 12-port AMF rotary valve was used to transport different samples to the sensor cartridge. The pump, valve and microscope were controlled together by a computer, enabling automated sample delivery and calibration experiments for lactoferrin sensing.
…to OEM integration
Once the fluidic protocol is established, the same principles can be transferred into an instrument using AMF OEM components. The SPM programmable syringe pump can provide automated pumping and dosing, while the RVM rotary valve can select between multiple reference solutions, samples, buffers or cleaning fluids. Together, they provide a compact fluidic architecture for automated sensor testing, calibration and biosensing.
For applications where footprint is critical, the RVM mini brings multi-fluid switching into an even smaller format, making it particularly suitable for compact and embedded sensing platforms, point-of-care instruments and other space-constrained devices.
This creates a direct path from development to integration:
LSPone for prototyping and validation → SPM + RVM for OEM integration → SPM + RVM mini when footprint is critical
A biosensor does not operate independently from the fluid that reaches it. Accurate calibration requires known concentrations, but also controlled delivery, complete sample exchange and reproducible experimental conditions.
From establishing a calibration curve to reproducing dynamic concentration profiles and automating continuous sensing, precise fluid handling helps create the controlled conditions required for reliable sensor data.
Developing or integrating a biosensor? Explore how AMF programmable syringe pumps and rotary valves an simplify calibration, sensor testing and automated fluid handling.
Receive the latest updates, insights, and advancements in microfluidics directly to your inbox Your email address is used exclusively for newsletter distributionand you can opt out at any time through the provided unsubscribe link.
