**3. The MagneChip platform: construction and operation**

MagneChip is a microfluidic platform centered on a chip consisting of several reaction chambers enabling accumulation (and release) of MNPs. This magnetic microreactor chip can utilize the benefit of excellent separation ability of MNPs in magnetic field. In various applications, the MNPs covered by biologically active molecules (e.g., bioreceptors) are immobilized on their surfaces may be used. Magnetic techniques enable anchoring the particles inside certain compartment of microreactors, where the accumulated magnetic particles can form a dense layer. Afterfilling (in a consecutive step), reagents can flow through the chip, while bioreaction occurs inside the microchambers and the resulted product flows through the chip. The outflow can be collected and/or quantified outside the chip, for instance by absorbance method. Because the enzyme to be immobilized on the MNP surfaces can be chosen freely, a wide variety of applications are possible (**Figure 2**). Taking the advantage of the continuous-flow operation, product formation can be monitored for a long time under various conditions over the same anchored *ec*MNP layer. MagneChip can be reinitialized periodically which enables multiparameter experiments, and therefore, reaction kinetics can be characterized in a fully automated way. Because the flow control system of the platform allows changing the actual substrate over the *ec*MNP layer, reactions can be screened even with unexplored substrates (**Scheme 1**). This feature renders MagneChip as a tool for substrate discovery as well.

**Figure 2.** Possible applications of MagneChip platform. Reaction chambers are filled with bio-functionalized magnetic nanoparticles, product formation is measured by in-line UV detector.

**Scheme 1.** Ammonia elimination from different amino acids (**1a-f**) catalyzed by *Pc*PAL immobilized onto MNPs with‐ in the MagneChip.

#### **3.1. Basic aims and principles**

**Figure 2.** Possible applications of MagneChip platform. Reaction chambers are filled with bio-functionalized magnetic

**Scheme 1.** Ammonia elimination from different amino acids (**1a-f**) catalyzed by *Pc*PAL immobilized onto MNPs with‐

nanoparticles, product formation is measured by in-line UV detector.

164 Lab-on-a-Chip Fabrication and Application

in the MagneChip.

A microfluidic test bench was developed for carrying out microreactor experiments with MagneChip (**Figure 3**). The test bench consisted of two syringe pumps for dispensing reagents, a thermostable chip holder and a zoom microscope for the optical inspection of the chip. The chip holder had four magnet drawers enabling to push permanent magnets under reaction chambers of the chip and also pull them out as the magnetic field is no longer required.

**Figure 3.** Schematic diagram of the fluid control system of MagneChip platform [22].

MagneChip reaction chambers (volume of ~1 μl) were designed to accomplish the following requirements:


A four reaction chamber MagneChip layout is presented in **Figure 4a**. CFD simulations revealed that the flow velocity distribution inside of the chambers varied in a scale of two (**Figure 4b**). Depending on the typical flow rates used in MagneChip, reaction residence time in the chambers may vary from 1 to 10 s (**Figure 4c**).

**Figure 4.** (a) Layout of the four chamber MagneChip; (b) flow velocity distribution inside the reaction chambers (units are in mm s−1); (c) residence time vs. flow rate in MagneChip reaction chambers.

#### **3.2. Construction method of MagneChip**

**Figure 4a** depicts the arrangement of a four-chamber chip used for testing the enzymatic reactions.

The chip was constructed by PDMS molding technology. SU-8 photoresist structures were prepared as a molding master, resulting in a channel height of 110 μm. PDMS was poured on the master and was kept on room temperature for 1 day. After cross-linking, the PDMS replica was released and the PDMS channel bodies were bonded to standard microscope glasses after oxygen plasma treatment. Some of the chips were equipped with a resonant coil magnetome‐ ter placed under the chambers for MNP quantity measurement [23]. The coil was embedded in an intermediate PDMS layer. For further construction details, see [23].

#### **3.3. Method of MNP quantification in the reaction chambers**

The magnetic behavior of MNPs initiated the development of an inductive method to quantify the nanoparticles. The measurement is based on the resonance frequency shift of a passive electrical resonant circuit, where a flat inductor coil integrated in a silicone elastomer film acts as a sensor. From the suspension of MNPs flowing into the chip, MNPs were anchored within the reaction chambers by external permanent magnets. The *ec*MNP amount inside the chamber affected the inductance; therefore, the resonance frequency was changed. The method also enabled on-line monitoring of the actual *ec*MNP quantity in the chamber. This test arrange‐ ment enabled to study the effect of particle size and arrangement on the chamber filling MNP mass and also on the catalytic activity of the PAL bound to the *ec*MNPs [23].

#### **3.4. Operation methods of MagneChip**

#### *3.4.1. Fluid handling steps*

A four reaction chamber MagneChip layout is presented in **Figure 4a**. CFD simulations revealed that the flow velocity distribution inside of the chambers varied in a scale of two (**Figure 4b**). Depending on the typical flow rates used in MagneChip, reaction residence time

**Figure 4.** (a) Layout of the four chamber MagneChip; (b) flow velocity distribution inside the reaction chambers (units

**Figure 4a** depicts the arrangement of a four-chamber chip used for testing the enzymatic

The chip was constructed by PDMS molding technology. SU-8 photoresist structures were prepared as a molding master, resulting in a channel height of 110 μm. PDMS was poured on the master and was kept on room temperature for 1 day. After cross-linking, the PDMS replica was released and the PDMS channel bodies were bonded to standard microscope glasses after oxygen plasma treatment. Some of the chips were equipped with a resonant coil magnetome‐ ter placed under the chambers for MNP quantity measurement [23]. The coil was embedded

The magnetic behavior of MNPs initiated the development of an inductive method to quantify the nanoparticles. The measurement is based on the resonance frequency shift of a passive electrical resonant circuit, where a flat inductor coil integrated in a silicone elastomer film acts as a sensor. From the suspension of MNPs flowing into the chip, MNPs were anchored within the reaction chambers by external permanent magnets. The *ec*MNP amount inside the chamber affected the inductance; therefore, the resonance frequency was changed. The method also enabled on-line monitoring of the actual *ec*MNP quantity in the chamber. This test arrange‐

in the chambers may vary from 1 to 10 s (**Figure 4c**).

166 Lab-on-a-Chip Fabrication and Application

are in mm s−1); (c) residence time vs. flow rate in MagneChip reaction chambers.

in an intermediate PDMS layer. For further construction details, see [23].

**3.3. Method of MNP quantification in the reaction chambers**

**3.2. Construction method of MagneChip**

reactions.

The experiments in MagneChip (**Figure 3**) involved four steps: (1) filling up the chip with MNPs, (2) absorbance calibration, (3) experiment cycles, and (4) chip cleaning.

*Chip filling.* In the chip-filling step, an MNP suspension was driven through the chip by applying a slight air pressure (0.2–0.3 bar) to the vial containing the MNP suspension and connected to the inlet of MagneChip via a PTFE tube (**Figure 3**) at 25°C. During the filling process, the MNPs were accumulated in the reaction chambers due to the permanent mag‐ nets placed in moveable drawers enabling "on/off" switching of the magnetic field. Once the chamber most distantfrom the inlet (**Figure 3**, *Chamber 4*) was saturated, the permanent magnet of the chamber at preceding position was turned on (**Figure 3**, *Chamber 3*). The same proce‐ dure was repeated (**Figure 3**: *Chambers 2* and *1*) until all chambers were filled up. Each chamber of the MagneChip device could capture ca. 250 μg of *ec*MNP biocatalyst [23].

*Calibration and experiment cycles*: During the forthcoming steps, the valve at the inlet of the MagneChip (**Figure 3**) was switched to the substrate (reagent) circuit. The flow controller performed the dosage of the substrate and other chemicals as dictated by the programmed sequence.

*Chip cleaning*: At the final, chip-cleaning step, the magnetic drawers of the MagneChip (**Figure 3**) were drown out and a washing solution was driven through the chip to remove *ec*MNPs.

The individual steps in a series of experiments, called experiment cycle (**Figure 3**, *Experiment cycles*), involved a *Reaction step* and a *Re-initialization step*. A series of experiments could consist of several experiment cycles depending on the number of parameters to be changed.

*Reaction step.* In a *Reaction step,* a substrate-containing solution was flowing through the chip at a constant flow rate, and the specific absorbance of the product was continuously moni‐ tored in the outflow of the chip.

*Re-initialization step*. In the *Re-initialization step,* the feed of the substrate-containing solution was stopped, and the chip was flushed with a washing buffer, while the magnetic particles were retained by the permanent magnets.

#### *3.4.2. Reaction step variants*

The substrate feed (with continuous monitoring of the absorbance in the outflow at a previ‐ ously selected wavelength) was accomplished according to one of the following variants. The cycle ended when the predesigned step time had been passed or when the reaction reached saturation.


#### **3.5. Quality assessment of the operations in MagneChip**

A series of subsequent measurements performed by the system were considered as reliable if all the following conditions were met [22]:


In order to test the fulfillment of the first group of conditions, a control measurement was performed after each series of experiments; that is, the first step of the sequence was repeat‐ ed in the last step under the same conditions, and the specific activity of the immobilized biocatalyst (*U*B) at saturation concentrations of L-phenylalanine (L-**1a** in **Scheme 1**) in the first and last cycles was compared.

#### *3.5.1. Reproducibility of the individual measurements*

*Reproducibility of the chip-filling process:* The first single chamber of the MagneChip was filled with MNP suspension. Biotransformation of L-**1a** to **2a** (**Scheme 1**) was performed in flowthrough mode and monitored by in-line UV–Vis. After reaching the stationary state (i.e., constant level of product formation), the magnet of the chamber was released and the *ec*MNPs were captured in the next chamber. The experiments performed in three consecutive cham‐ bers were repeated three times resulting in *U*B = 8.01 ± 0.14 μmol g−1 min−1 [22].

The filling–refilling results indicated that neither that the homogeneity of the MNP suspen‐ sion nor the filling procedure of the chambers had remarkable effect on the reproducibility of the measurements. The significant difference between the *U*<sup>B</sup> values of MNP biocatalyst in shake vials and in MagneChip indicated increased effectivity of the biocatalysts in MagneChip device [22].

*Reproducibility of independent measurements:* Biotransformation of L-phenylalanine (L-**1a**) to (*E*) cinnamic acid (**2a**) by MNP biocatalyst suspension (**Scheme 1**) was performed in shake vial as three parallel reactions and resulted in *U*B = 2.91 ± 0.08 μmol g−1 min−1 ensuring that the homogeneity of the MNP suspension was sufficient [22].

#### *3.5.2. Optical inspection of the reaction chambers*

**•** *Repeatibility test.* The feed of the substrate started (1st cycle) or continued at unchanged flow

**•** *Flow rate test.* The feed of the substrate started (1st cycle) or continued, while the flow rate

**•** *Substrate concentration test*. The substrate-containing solution and the washing buffer were feed in parallel at a predesigned ratio resulted in a predefined dilution of the substrate at

**•** *Substrate screening.* The actual substrate was loaded into the substrate syringe through a bypass valve from the actual container of the substrate stock (A-F, in **Figure 3**), and the feed of the substrate began at a predefined flow rate. To change to the next substrate, a *Reinitialization step* was performed, followed by loading the next substrate into the substrate

A series of subsequent measurements performed by the system were considered as reliable if

**•** independent measurements were reproducible using the same type of *ec*MNP biocatalyst, **•** the product of the enzyme reaction could be measured selectively in the UV–Vis range,

**•** the enzymatic activity of the *ec*MNP biocatalyst remained unchanged during the measure‐

**•** and last but not least, the *ec*MNP layer in the magnetic reactors remained unharmed during

In order to test the fulfillment of the first group of conditions, a control measurement was performed after each series of experiments; that is, the first step of the sequence was repeat‐ ed in the last step under the same conditions, and the specific activity of the immobilized biocatalyst (*U*B) at saturation concentrations of L-phenylalanine (L-**1a** in **Scheme 1**) in the first

*Reproducibility of the chip-filling process:* The first single chamber of the MagneChip was filled with MNP suspension. Biotransformation of L-**1a** to **2a** (**Scheme 1**) was performed in flowthrough mode and monitored by in-line UV–Vis. After reaching the stationary state (i.e., constant level of product formation), the magnet of the chamber was released and the *ec*MNPs were captured in the next chamber. The experiments performed in three consecutive cham‐

The filling–refilling results indicated that neither that the homogeneity of the MNP suspen‐ sion nor the filling procedure of the chambers had remarkable effect on the reproducibility of

bers were repeated three times resulting in *U*B = 8.01 ± 0.14 μmol g−1 min−1 [22].

**•** product and substrate could be completely removed through the washing steps,

the chip inlet. The dilution ratio could be different cycle by cycle.

rate.

ment

the measurement cycles.

and last cycles was compared.

*3.5.1. Reproducibility of the individual measurements*

changed cycle by cycle.

168 Lab-on-a-Chip Fabrication and Application

syringe from the substrate stock (A-F).

all the following conditions were met [22]:

**3.5. Quality assessment of the operations in MagneChip**

During the experiments, the chip was optically inspected by a zooming microscope and a monochrome hi-speed smart camera. Before evaluating the measurement sequence, the plan view of the chip was stored as a reference (∏ref). At the end of the step *i* of the measurement sequence, the plan view of the chip was sampled again (∏seq,i) and it was compared to the reference as follows [22]:

**Figure 5.** MagneChip device with four MNP-filled and external magnet-equipped microchambers (top left) and SEM image of the MNP layer (top right). The effect of air bubble passage through the reaction chamber [(a)–(f)]: (a) photo‐ graph, before passage; (b) difference image (difference score SC = 5073), after passage; (c) calculated flow velocity field before and (d) after the passage; (e) velocity profile in the middle cross section of the chamber before and (f) after the passage [22].

$$\Pi\_{d\circ\mathcal{Y}}\left(j,k\right) = \begin{cases} \Pi\_{r\circ\mathcal{Y}}\left(j,k\right), \Pi\_{r\circ\mathcal{Y}}\left(j,k\right) - \Pi\_{s\circ\mathcal{Y},i}\left(j,k\right) < 0\\ 0, \Pi\_{r\circ\mathcal{Y}}\left(j,k\right) - \Pi\_{s\circ\mathcal{Y},i}\left(j,k\right) \ge 0 \end{cases}$$

where (*j,k*) are the pixel coordinates of the plan view image; therefore, the changes in accord‐ ance to the reference image are indicated by white pixels. The total number of white pixels is defined as *chamber difference score* (*SC*) used as a marker for describing the changes of the MNP layer arrangement. Therefore, the changes compared with the image of the first cycle (reference) were indicated by white areas during the consecutive cycles of the measurement.

In practice, *SC* values under 2000 reflected to negligible changes. However, *SC* > 3000 indicated serious structural change of the *ec*MNP layer, for instance, the complete breakthrough of a bubble (**Figure 5b**) [22]. Air bubbles usually did not split at the channel entrance, rather passed at one side along the chamber wall. Numerical simulations revealed (**Figure 5c–f**) that the velocity profile became asymmetric due to the bubble passage and the overall mass flow rate through the porous MNP layer significantly decreased (28.6–20.7 μL min−1, roughly 72% of its original value), while the remaining fluid passed through the developed tunnel. The passing bubble could drift away particles which decreased the total mass of the biocatalyst in the reaction chamber. Therefore, the biocatalytic activity of the damaged chamber decreased and the consequent measurements were no longer reliable.

Reliability assessment of the measurements was based mostly on the following parameters:


Each of the experiments carried out by the platform was justified based on the above criterion.

**Figure 6.** Time plot of the periodic absorbance change during the cyclic measurement (attempt 1, stable layer). The chip is re-initialized between the reaction steps (reaching zero absorbance) by washing out substrate and product com‐ pletely [22]. The last measurement served as a control.

A crucial feature is the reproducibility of cyclic reactions performed by the system. To check the reproducibility of the test reactions, the MagneChip was filled with MNP biocatalyst, and biotransformation of L-**1a** to **2a** was performed in seven consecutive cycles, while the chip was re-initialized during the steps by washing out the substrate and product completely [22]. The absorbance plot at 290 nm in **Figure 6** with the aid of the previously measured extinction coefficient of the product (**2a**) indicated the concentration changes of **2a**.

The product quantity in cycle by cycle—calculated by taking the integral of the absorbance plot—clearly indicated that the chip was successfully re-initialized in every cycle through‐ out the experiment, and the reaction was repeated reproducibly seven times (average product quantity of *P* = 0.12 ± 1.5% μmol) [22]. The moderate mean value of the chamber difference score *SC* = 1322 (1609 max) reflected negligible changes in the MNP layer.
