biosensors

Review
PDMS Bonding Technologies for Microfluidic Applications:
A Review
Alexandra Borók, Kristóf Laboda and Attila Bonyár *

Department of Electronics Technology, Faculty of Electrical Engineering and Informatics,

Budapest University of Technology and Economics, H-1111 Budapest, Hungary; [email protected] (A.B.);
[email protected] (K.L.)
* Correspondence: [email protected]

Abstract: This review summarizes and compares the available surface treatment and bonding
techniques (e.g., corona triggered surface activation, oxygen plasma surface activation, chemical
gluing, and mixed techniques) and quality/bond-strength testing methods (e.g., pulling test, shear
test, peel test, leakage test) for bonding PDMS (polydimethylsiloxane) with other materials, such
as PDMS, glass, silicon, PET (polyethylene terephthalate), PI (polyimide), PMMA (poly(methyl
methacrylate)), PVC (polyvinyl chloride), PC (polycarbonate), COC (cyclic olefin copolymer), PS
(polystyrene) and PEN (polyethylene naphthalate). The optimized process parameters for the best
achievable bond strengths are collected for each substrate, and the advantages and disadvantages of
each method are discussed in detail.

Keywords: PDMS; polymers; bonding technologies; bond strength



1. Introduction


Polydimethylsiloxane (PDMS), a silicon-based elastomer, is considered a fundamental
Citation: Borók, A.; Laboda, K.; building block in microfluidics due to its many desirable material properties (e.g., elas-
Bonyár, A. PDMS Bonding ticity, optical transparency, biocompatibility, low autofluorescence, excellent thermal and
Technologies for Microfluidic
chemical stability, etc.) [1–4]. Additionally, due to its convenient patterning technologies,
Applications: A Review. Biosensors
such as replica molding or soft lithography, traditional fabrication methods centering on
2021, 11, 292. https://doi.org/
glass or silicon etching could be replaced with faster, less-expensive solutions [5–7]. Today,
10.3390/bios11080292
for a functional microfluidic device, microchannels are usually created in PDMS, then a
chip is formed by closing them with a solid or flexible substrate (e.g., glass, silicon, or
Received: 27 July 2021
various polymers) [8–11]. This closing method is referred to as bonding technology since
Accepted: 19 August 2021
Published: 23 August 2021
irreversible chemical bonds are formed between the activated or functionalized PDMS and
substrate surfaces. Silicon-based materials (e.g., glass or silicon) can be relatively easily
Publisher’s Note: MDPI stays neutral
bonded to PDMS via simple surface activation, and thus these are the most widespread
with regard to jurisdictional claims in
ways of sealing the microchannels.
published maps and institutional affil-
However, in several cases, another type of substrate is required for a given application
iations. area. A common requirement could be the flexibility of the substrate. With the spreading
of flexible electronics created by additive manufacturing [12,13] (e.g., electrodes on flexible
polymer substrates, such as polyimide, created by 2D or 3D printing of nanomaterials [14]),
functional elements such as sensors, biosensors, or microelectrode arrays can be integrated
onto the surface of the polymer substrate [15,16]. Besides microfluidics, PDMS is also often
Copyright: © 2021 by the authors.
used for the encapsulation of surface integrated electronics, or as protective coatings [17,18].
Licensee MDPI, Basel, Switzerland.
This article is an open access article
For complex Lab-on-a-Chip (LoC) or micro total analysis systems (Micro-TAS), PDMS is
distributed under the terms and
also often required to be bonded to other polymer-based materials, such as PET (polyethy-
conditions of the Creative Commons lene terephthalate), PI (polyimide), PMMA (poly(methyl methacrylate)), PVC (polyvinyl
Attribution (CC BY) license (https:// chloride), PC (polycarbonate), COC (cyclic olefin copolymer), PS (polystyrene) or PEN
creativecommons.org/licenses/by/ (polyethylene naphthalate) [19–21]. PDMS–polymer bonding is usually more complicated
4.0/). and requires the chemical functionalization of at least one bonded surface. These methods

Biosensors 2021, 11, 292. https://doi.org/10.3390/bios11080292 https://www.mdpi.com/journal/biosensors

Biosensors 2021, 11, 292 2 of 28

are called chemical gluing since chemical bonds between the applied molecular monolayers
and surface functional groups provide a strong bond between the attached surfaces. In
the past ten years, many techniques have been developed based on the combinations of
surface activation, chemical gluing, and adhesive-based methods to bond PDMS with
various substrates [22–24]. Since for most microfluidic applications a good quality bond
with high strength is critical (especially where high-pressures or flow rates are applied),
it is imperative to select the most appropriate bonding method for our target application.
Additionally, the optimization of the technological process parameters is essential since, as
demonstrated later, the quality of the bond greatly relies on it.
Our paper provides a review of these various technologies to bond PDMS with other
surfaces. It pivots around three comprehensive tables, and the sections are structured
accordingly. Table 1 and Section 2 compare the testing methods which are commonly used
to quantify bond strengths. The differences between these methods should be understood
in order to be able to evaluate and compare our bond quality with others. Tables 2 and 3
compare the bonding methods for various substrates, which are discussed in detail in
Sections 3 and 4. While Table 2 focuses on the achievable bond strengths for a given
substrate (based on the test method), Table 3 provides the optimized process parameters to
reach these results. It is important to emphasize that both tables contain only the optimized
process parameters. For bond quantification in a wider parameter range, please see the
references. Our review is also content with focusing only on these aspects of PDMS-based
microfluidic fabrication. Recent comprehensive reviews on the general application of
PDMS for microfluidics [2], biomedicine [25], or fabrication technologies [5–7,26] can be
recommended for those interested in other aspects.

2. Bond-Strength Testing Methods

Quality check of the prepared bond and the quantification of bond strength are im-
portant for all applications. Many different testing methods are used for this purpose; as
shown in Table 1, 12 different functionally different techniques could be distinguished.
Although a high bond strength—quantified by either tensile strength or shear strength—is
a general requirement, many applications have special needs, e.g., long-term channel
integrity [27,28], high working pressures inside the channel [29,30], or high inlet veloc-
ity [31–33]. Thus, understanding the differences between these testing methods and their
provided information is essential for picking the right test for a given application. Compre-
hension of the testing methods is also crucial for interpreting the different bond strengths
that characterize the bonding technologies.

2.1. Manual Peeling/Delamination Test

In this simple qualitative test, the bonded PDMS piece is forcibly removed from the
substrate by hand or with the help of some basic tools like a scalpel. If the PDMS cannot be
easily removed, or upon removal, the PDMS block tears/breaks (cohesive failure) instead
of detaching cleanly from the substrate (adhesive failure), the formation of chemical bonds
can be assumed. This test is the cheapest and easiest to implement: no special equipment or
sample preparation is needed. On the other hand, the test does not yield any quantitative
data, the results are quite subjective and have poor reproducibility. Usually, the main
purpose of this test is to verify bond formation and/or filter out faulty workpieces before
quantitative bond-strength testing with another method [19–21,34–37].

2.2. Tensile Strength Measurements

The quality of the prepared bond is best characterized by its tensile strength; thus,
this is one of the most commonly used quantitative testing methods. Several different
approaches exist to measure this property.
With a standard tensile testing machine, grips can be used to pull the bonded materials
apart from each other and measure the force required to detach the PDMS from the
substrate. The general advantage of this test is that the experimental conditions are well
Biosensors 2021, 11, 292 3 of 28

defined, only normal forces act on the materials, and thus the results can be easily compared
with other similarly executed measurements. The three approaches in Figure 1a–d differ in
how the samples are fixed to the pulling grips. Although using adhesives to fix the upper
(non-bonded) site of the PDMS block to the grip (or a specifically machined stump) might
seem straightforward, this might cause inconvenience for some equipment. Additionally,
the adhesive strength between the PDMS and the grip head (stump) must be higher than
the characterized PDMS–substrate bond, otherwise, it will break prematurely (Figure 1a).
Sunkara et al. used a silicone sealant (LC909N, Henkel, Germany) to bond the upper part
of the PDMS to an aluminum jig, which held up to 490 kPa [38].
To avoid using adhesives, it is possible to create special jaws/grips to hold the PDMS
and the substrate in the heads of the tensile testing machine. For this, the PDMS part should
be explicitly molded for the purpose, e.g., in a cylinder shape (illustrated in Figure 1b,
referred to as cylinder-based tensile strength measurement in Table 1). The protruding
part of the cylinder is bonded to the substrate, while the flat part of the PDMS can be
easily clamped to the equipment [39,40]. The substrate is similarly clamped to the bottom
jaw. A disadvantage of this technique is that the used PDMS shape (which requires a
separate molding form) is quite special and most possibly significantly differs from the
design of the functional element, which cannot be directly used for this test. By contrast,
the adhesive-assisted tensile testing could be performed even on the functional elements,
e.g., bonded microfluidic cells, and thus combined with other testing methods (leakage
test, burst test). The cylinder’s geometry should also be optimized: a too-long cylinder or a
cylinder with a too-small base diameter can be torn down from the PDMS base. Too thin or
fragile substrates (e.g., glass, silicon) are also vulnerable to breaking during the pulling.
The third solution is illustrated in Figure 1c. In this case, two layers of the substrate
are bonded to both sides of the PDMS block. The substrate is fixed to the tensile strength
testing machine with either clamps, screws, or adhesives [41,42]. The substrates should
be thick enough to avoid breaking them during pulling. Additionally, if screws are used,
holes are needed to be drilled into the substrates. The areas of the two bonds should be
controlled precisely and taken into account upon the calculation of the tensile strength.
An alternative way for testing tensile strength is shown in Figure 1d. Here two
twines are used to pull the PDMS/substrate assembly apart. One fixes the substrate to a
pulling head via a drilled hole. A second twine is inserted into the PDMS pre-polymer and
permanently cured into it. After bonding the PDMS with the substrates, the assembly is
pulled apart. Although this pull test is performed in several works (e.g., by the team of Nae
Yoon Lee [43–47]), it might have some drawbacks. Although cohesive failures can be easily
distinguished from adhesive failures, the tensile strength cannot be adequately calculated
in the latter case. Due to the nature of the assembly and the twine distribution inside the
PDMS block, the pulling force cannot be considered perfectly normal and homogenously
distributed along the bonded area. The bond between the twine and the PDMS is also
crucial. The twine might be torn out from the block without detaching the PDMS from
the substrate.
Biosensors 2021, 11, 292 4 of 28

Figure 1. Illustrations of different bond-strength testing methods. (a) Tensile strength measurement
with adhesive. (b) Cylinder-based tensile strength measurement with a specially designed PDMS
piece. (c) Double-substrate bonding-based tensile strength measurement. (d) String-based tensile
strength measurement. (e) Pushing-based shear strength measurement. (f) Pulling-based shear
strength measurement. (g) Peeling test. (h) T-peeling test. (i) Leakage test. (j) Burst test.
Biosensors 2021, 11, 292 5 of 28

2.3. Shear Strength Measurement

Although shear forces might be a bit less relevant in microfluidic applications, in some
instances, shear strength measurements are also performed on the bonded structures. In a
classical shear testing machine, a force parallel to the bonding plane is applied with a given
shear rate, as illustrated in Figure 1e. The shear force and displacement are measured,
which can be used to calculate shear strength (failing load divided by the bond area) and
shear strain [48].
A general problem with the shear test in Figure 1e is that the PDMS detaches gradually
from the substrates, which requires monitoring the propagation of peeling to calculate the
delamination area, as demonstrated in [49]. Pushing a partially peeled PDMS can also
modify the direction of the acting forces, distorting the results.
To tackle this issue, Wang et al. bonded two rigid substrates to either side of the PDMS,
and their shear tester pushed the upper substrate until the joint pair failed in one step. In
this way, the shear strength and shear strain of the bonds could be characterized more
conveniently and precisely [48].
An alternative approach could be using a tensile testing machine in shear mode, as
illustrated in Figure 1f. Although in some works the results of these tests are referred to as
tensile strength [50], due to the parallel forces, it is more appropriate to refer to them as lap
shear strength, which is defined as the failure load divided by the bond area [51]. There
are two ways to perform this test: either two PDMS blocks can be bonded to one substrate
(as in [50]), or two substrates can be bonded to one PDMS block to avoid the difficulty
of clamping PDMS. The bond areas should be precisely controlled and considered in lap
shear strength calculation.

2.4. Peel Test

Peel tests are also frequently used to assess the bond quality of adhesive joints exposed
to peel forces. Peel testing requires at least one flexible component, which refers to the
ability of the adherend to bend through 90◦ without breaking or cracking [52]. For example,
Hoang et al. used a Kapton foil bent at 90◦ angle, as illustrated in Figure 1g [53]. As the
upper clamp roses and pulls the Kapton with a constant speed, at the same time, the lower
platform shifts to maintain the 90◦ . Samples were glued onto glass slides using a silicone
adhesive to prevent the PDMS from lifting during peeling. The resulting peel strength is
calculated as the constant load per the bonded area’s width required to continue peeling the
joint after initiation, determined from the flat portion of the force–extension curve [52,53].
Another method for peel strength measurement is the so-called T-peel test, illustrated
in Figure 1h, which is also adopted by standard bodies (e.g., ISO 11339 or ASTM D
1876) [52]. A drawback of this method is that it requires both parts of the bonded structure
to be flexible. It is most commonly used to detach two bonded PDMS blocks, as in [54]
and [55], but a PDMS block bonded with a flexible foil could also be investigated this way.
Overall, peel tests are harder to be performed correctly and could be done on a limited
number of substrate types. The resulting peel strength is also a bit less relevant than the
more widely used tensile strength.

2.5. Leakage Test

The most practical way to test the functionality of a microfluidic chip is leakage
testing. In this family of methods (illustrated in Figure 1i), different approaches are used
to test different qualities. In the simplest static leakage test (sometimes also referred to
as durability test), a fluid is injected into the microchannels and kept there for a given
amount of time, from the shortest test durations of minutes–hours [40,42] to even as long as
months [20], and channel integrity is tested afterward. For these tests, no flow is applied on
the fluid after injection (static conditions), and for extended tests, the ports are hermetically
sealed. Besides colored water, other solutions are sometimes also used to test the long-term
durability and swelling of PDMS (e.g., acids, i.e., 1M HCl [36], bases, i.e., 1 M NaOH [36],
or organic solvents, i.e., tetrahydrofuran (THF) [43]).
Biosensors 2021, 11, 292 6 of 28

By contrast, in the case of dynamic leakage tests, a constant flow is applied to evaluate
the hydrodynamic stability of the device. Usually, the flow rate is gradually increased
to a point where channel integrity fails. For the sake of better comparability between
different structures/geometries, besides the absolute maximum flow rates (expressed in
mL/min), the ratio of the liquid volume that flows through the channel within 1 min
and the channel volume is also frequently given (expressed as flow-volume/channel-
volume/min). Without a doubt, this is one of the most widespread methods for testing
microchannel functionality [19,35,36,38,43–47,56,57].

2.6. Burst Test

Although technically this method could be considered to be a special type of leakage
test, it is commonly referred to as burst pressure test (illustrated if Figure 1j). In this case,
fluids or gases are introduced into either the microchannels or a blister specifically designed
for this test. The pressure is slowly increased until the integrity of the investigated structure
is compromised, either by delamination, leakage, or burst. This test requires appropriate
instrumentation, including pressure control (pump and pressure sensor). Although usually
the burst pressure level can be easily identified as a rapid drop in the measured pressure,
sometimes the continuous optical monitoring of the structures might be needed to detect
finer damage (e.g., slight leakage might happen before the burst of the cell). It has to be noted
that—depending on the bonding protocol—there might be a huge difference between the
burst pressures measured with air/nitrogen and water. For example, Pečar et al. measured a
5-fold difference between the burst pressures obtained with air and water, respectively [37],
explained by the hydrolysis of covalent bonds [58]. (For this reason, in Table 2, the medium
used for the test is also indicated.) Burst testing is probably the most used functional method
to quantify the strength of the prepared bonds [21,36–41,46–48,57,59–62].
Biosensors 2021, 11, 292 7 of 28

Table 1. The most frequently used bonding strength measurement methods and test.

Required Special Sample

No Name of the Test Measured Value Complexity Restrictions/Possible Problems Ref.
Equipment Preparation Requirements
-Only as a first attempt to
Manual Bond verify bond formation.
Performed
1 peeling/ formation + - -Results are subjective. [19–21,34–37]
by hand
delamination test (yes/no) -A successful bond might be too strong to
peel manually.
Adhesive- Tensile testing -The tensile strength of the adhesive bond
Tensile strength An adhesive must be
2 assisted tensile strength machine, + must be higher than the quantified bond [38]
[Pa] applied to the samples
measurement adhesive between PDMS and substrate.
-The size of the cylinder should be
optimized to avoid rupture, and thin
Cylinder-based substrates are vulnerable to breaking.
Tensile testing Tensile strength Requires a special
3 tensile strength ++ -Due to the special PDMS geometry, it [39,40]
machine [Pa] molding form for the PDMS.
measurement cannot be used on functional
elements or combined
with other test methods.
Two substrates are required
Double- to be bonded to the PDMS. If -The areas of the two bonds should be
substrate-bonding-based Tensile testing Tensile strength screws are used to fix the precisely controlled, the tensile strength of
4 ++ [41,42]
tensile strength machine [Pa] sample to the machine, holes the bond should be
measurement must be drilled into calculated accordingly.
the substrate.
-The bond between the string and PDMS is
The PDMS block and a crucial. The string might be torn out from
String-based tensile string should be the PDMS block without detaching the
Tensile testing Tensile strength
5 strength measurement +++ cured together. PDMS from the substrate. [43–47]
machine [Pa]
(Pull test) A drilled hole is needed in -The obtained tensile strength might not be
the substrate. directly comparable
with other methods.
Biosensors 2021, 11, 292 8 of 28

Table 1. Cont.

Special Sample
No Name of the Test Required Equipment Measured Value Complexity Restrictions/Possible Problems Ref.
Preparation Requirements
-Parallel forces should be maintained,
The width of the PDMS
Shear strength Shear testing Shear strength PDMS bending is a problem.
6 ++ block should not be wider [48,49]
measurement machine [Pa](shear strain) -The obtained shear strength should be
than the peeling tool.
treated with reservations.
-The sizes of the bonded areas are crucial
Either two PDMS blocks are and should be properly controlled.
Lap shear Lap shear
Tensile testing bonded to one substrate or -The shear strength result has to be
7 strength strength ++ [50]
machine two substrates to one calculated accordingly.
measurement [Pa]
PDMS block. -The clamping of the PDMS blocks might
be difficult.
-The pulling angle should be kept 90◦
A partially bonded
during the experiment to ensure
flexible foil, is required. The
Tensile testing normal forces. The movement of the
Peel strength PDMS is bonded to a
8 Peel test machine, glass +++ supporting glass slide should be [53]
[N/m] fixed support
slide, adhesive controlled accordingly.
(e.g., a glass slide)
-The strength of the adhesive bond should
with adhesive.
be higher.
A partially bonded
Peel strength formation is required. The forces acting on the bond should be
9 T-peel test Tensile testing machine ++ [54,55]
[N/m] A thin foil or a second PDMS kept normal.
substrate can be used.
Time until Microfluidic channels are -Proper ink leakage detection.
Static leakage test Ink, tubes, syringe,
10 leakage + required inside the -Might be time-consuming for [20,40,42]
(durability test) microfluidic cell
[h] PDMS block good-quality bonds.
Ink, tubes, syringe, Max. flow rate Microfluidic channels are
Dynamic leakage [19,35,36,38,43–
11 microfluidic cell, computer before leakage ++ required inside the Proper ink leakage detection.
test 47,50,56,57]
-controlled pump [mL/min] PDMS block.
Compressed air or fluid, -The size of the blister is important and
Burst
microfluidic cell, syringe A specific blister should be should be properly controlled. [21,36–41,46–
12 Burst test pressure ++
pump, high-pressure tank, formed in the PDMS block. -PDMS might explode from the substrate at 48,57,59,60,62]
[Pa]
pressure sensor high pressures.
Biosensors 2021, 11, 292 9 of 28

Table 2. Bond strengths measured with different test methods on PDMS–substrate bonds, categorized by the substrate type. In the table only the test results corresponding to the optimized
process parameters are given.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
11 >60 µL/min max. tested flow rate, ~1500 × min−1 internal volume [19]
Corona treatment
12 290 kPa average burst pressure (air), variation: 227–380 kPa [60]
5 91 kPa tensile strength [45]
9 190 N/m peel strength, for optimized plasma power [55]
11 10 mL/min max. flow rate, ~30,000 × min−1 internal volume [56]
O2 plasma 12 300 kPa average burst pressure (air), variation: 180–515 kPa [60]
burst pressure (N2 ), with optimized parameters
12 400 kPa [59]
(range: 69–400 kPa)
PDMS
12 105–180 kPa burst pressure (air) [62]
Air plasma 9 400 N/m max. peel strength, with optimized parameters [54]

O2 plasma + APTES / GPTMS 5 184 kPa tensile strength [45]

surface treatment 11 5 mL/min no leakage was observed, 2000 × min−1 internal volume [45]
UV illumination 12 87–95 kPa burst pressure (air) [62]
DMPMS pre-polymer gluing 11 330–500 µL/min max. flow rate, ~6600 × min−1 , depending on the preparation [35]
12 671 kPa average burst pressure (air), variation: 545–690 kPa [60]
PDMS pre-polymer gluing
12 >500 kPa burst pressure (air) [62]
Corona discharge 7 94 kPa max. lap shear strength [50]
6 125–245 kPa max. shear stress, for bonding areas between 10–90 mm2 [48]
6 70–113 kPa max. shear stress [49]
11 4 mL/min max. flow rate before leakage ~12000 × min−1 internal volume [56]
Glass
O2 plasma 12 185–270 kPa burst pressure (air) [62]
12 190–610 kPa burst pressure (ink), for bonding areas between 10–90 mm2 [48]
burst pressure (N2 ), with optimized parameters
12 510 kPa [59]
(range: 276–510 kPa)
Biosensors 2021, 11, 292 10 of 28

Table 2. Cont.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
UV illumination 12 270–314 kPa burst pressure (air) [62]
Glass DMPMS pre-polymer gluing 11 500 µL/min max. flow rate, ~6600× min−1 , with optimized preparation [35]
PDMS pre-polymer gluing 12 >500 kPa burst pressure (air) [62]
3 400 kPa tensile strength [39]
PDMS pre-polymer gluing
Parylene coated 12 36 kPa burst pressure (water) [39]
silicon 3 1.4 MPa bond tensile strength with optimized parameters [39]
Plasma treatment with N2 and
SF6 mixture 12 145 kPa burst pressure (water) [39]
UV illumination + MPTMS
Gold-coated glass 7 120 kPa max. lap shear strength [50]
surface treatment
PDMS pre-polymer gluing 4 15 kPa tensile strength [42]
Corona discharge 5 336 kPa tensile strength [43]

Corona discharge 5 306 kPa tensile strength [47]

+ APTES surface treatment + 11 45 mL/min no leakage for 3000× min−1 internal volume [47]
epoxy term. PDMS linker
12 586 kPa burst pressure (air) [47]
2 385 kPa tensile strength [38]
PMMA 4 1.6 MPa max. tensile strength, with optimized APTES functionalization [41]

O2 plasma + APTES surface treatment 11 60 mL/min no leakage at 24,000 × min−1 internal volume [38]
12 100 kPa delamination pressure in water, >500 kPa in air [37]
average burst pressure (water),
12 260 kPa [41]
with optimized APTES functionalization
4 2.5 MPa max. tensile strength, with optimized APTES functionalization [41]
O2 plasma + APTES surface treatment
+ Corona discharge average burst pressure (water),
12 300 kPa [41]
with optimized APTES functionalization
Biosensors 2021, 11, 292 11 of 28

Table 2. Cont.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
5 259 kPa tensile strength [46]
O2 plasma
+ APTES/TESPSA surface treatment 11 30 mL/min no leakage at 3000× min−1 internal volume [46]
12 345 kPa burst pressure (air) [46]
PMMA
5 180 kPa tensile strength [44]
O2 plasma
+ APTES/GPTMS surface treatment 11 30 mL/min no leakage at 2000× min−1 internal volume [44]
12 510–538 kPa burst pressure (air) [44]
Corona discharge + MPTMS surface treatment 5 511 kPa tensile strength [43]

Corona discharge + APTES surface treatment + 5 220 kPa tensile strength [47]
epoxy terminated PDMS linker 12 620 kPa burst pressure (air) [47]
2 430–490 kPa tensile strength [38]
O2 plasma 11 60 mL/min no leakage at 24,000 × min−1 internal volume [38]
+ APTES surface treatment 12 >228 kPa burst pressure (water) [58]
PC 12 100 kPa delamination pressure in water, >500 kPa in air [37]
5 477 kPa tensile strength [46]
O2 plasma 11 30 mL/min no leakage at 3000× min−1 internal volume [46]
+ APTES/TESPSA surface treatment
12 413 kPa burst pressure (air) [46]
5 178 kPa tensile strength [44]
O2 plasma 11 30 mL/min no leakage at 2000× min−1 internal volume [44]
+ APTES / GPTMS surface treatment
12 524–579 kPa burst pressure (air) [44]

Corona discharge + APTES surface treatment + 5 476 kPa tensile strength [47]
PS
epoxy terminated PDMS linker 12 620 kPa burst pressure (air) [47]
Biosensors 2021, 11, 292 12 of 28

Table 2. Cont.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
no leakage with 1 M HCl for 3 days,
10 - [36]
1 M NaOH for 1 week, water for 1 month
Air plasma
11 15 mL/min no leakage at 9000× min−1 internal volume [36]
12 >500 kPa burst pressure (air) [36]
2 388 kPa tensile strength [38]

PS O2 plasma + APTES surface treatment 11 60 mL/min no leakage at 24,000 × min−1 internal volume [38]
3 12 kPa tensile strength [40]
5 520 kPa tensile strength [46]
O2 plasma
11 30 mL/min no leakage at 3000× min−1 internal volume [46]
+ APTES / TESPSA surface treatment
12 448 kPa burst pressure (air) [46]
11 500 µL/min no leakage at 1000× min−1 internal volume [57]
O2 plasma + epoxy adhesive
12 >414 kPa burst pressure (air) [57]
Corona discharge
+ APTES surface treatment + epoxy terminated 5 189 kPa tensile strength [47]
PDMS linker
Corona discharge
5 476 kPa tensile strength [43]
+ MPTMS surface treatment

O2 plasma 1 - successful bonding [45]

+ APTES / GPTMS surface treatment 12 579 kPa burst pressure (air) [44]
PET
5 458 kPa tensile strength [46]
O2 plasma
11 30 mL/min no leakage at 3000× min−1 internal volume [46]
+ APTES/TESPSA surface treatment
12 379 kPa burst pressure (air) [46]
successful bonding—
KOH activation + MPTMS surface treatment + 1 - [20]
no delamination with optimized parameters
O2 plasma
10 - no leakage during 1 month storage [20]
Biosensors 2021, 11, 292 13 of 28

Table 2. Cont.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
epoxy adhesive 8 1.7 N/m peel strength [53]
UV/ozone + silicone adhesive 8 72 N/m peel strength [53]
UV/ozone treatment + APTES/GPTMS
8 2.7 N/m peel strength [53]
surface treatment
UV/ozone treatment + MPTMS/GPTMS
8 200 N/m peel strength [53]
PI surface treatment
UV/ozone treatment + MPTMS by liquid
8 470 N/m peel strength [53]
deposition + epoxy adhesive
successful bonding—
KOH activation + MPTMS surface treatment + 1 - [20]
no delamination with optimized parameters
O2 plasma
10 - no leakage for 1-month storage [20]
no leakage with 1 M HCl for 3 days,
10 - [36]
1 M NaOH for 1 week, water for 1 month
PP Air plasma
11 15 mL/min no leakage at 9000× min−1 internal volume [36]
12 >500 kPa burst pressure (air) [36]
12 >500 kPa >500 kPa burst pressure (air) [36]
11 15 mL/min no leakage at 9000× min−1 internal volume [36]
Air plasma
no leakage with 1 M HCl for 3 days,
10 - [36]
1 M NaOH for 1 week, water for 1 month
1 - the bond was not permanent [21]
COC O2 plasma
12 150 kPa burst pressure (air) [21]
1 - strong bond [21]
O2 plasma 12 380 kPa burst pressure (air) [21]
+ APTES surface treatment 2 432 kPa tensile strength [38]
11 60 mL/min no leakage at 24,000 × min−1 internal volume [38]
Biosensors 2021, 11, 292 14 of 28

Table 2. Cont.

Bonding Technology Test Method Test Results

Substrate Type Ref.
(see Table 3) (see Table 1) Value Parameter/Condition
O2 plasma
1 - strong bond [21]
+ GPTMS surface treatment
COC 1 - stronger bond than with only APTES or GPTMS [21]
O2 plasma
+ APTES/GPTMS surface treatment 12 >800 kPa burst pressure (air), after 6 months ≥ 700 kPa [21]
O2 plasma
ABS 12 100 kPa delamination pressure in water, >500 kPa in air [37]
+ APTES surface treatment
successful bonding—
KOH activation + MPTMS surface treatment + 1 - [20]
PEN no delamination with optimized parameters
O2 plasma
10 - no leakage for 1-month storage [20]
Corona discharge
PVC 5 467 kPa tensile strength [43]
+ MPTMS surface treatment
Biosensors 2021, 11, 292 15 of 28

3. Bonding Methods
In this section, the different strategies used to bond PDMS with various substrates
are introduced in a general way. The specific protocols will be discussed in greater detail
concerning the different substrate materials in Section 4. The general bonding strategies
are illustrated in Figure 2. Silicon-based materials (e.g., glass or silicon) can be relatively
easily bonded to PDMS. The process only requires surface activation (Figure 2a), which can
be achieved either using corona discharge treatment [63,64] or oxygen plasma treatment
on the surfaces [65–67]. For other substrate materials (e.g., thermoplastics), the bondable
functional groups need to be created via surface functionalization. This method is often
referred to as chemical gluing (Figure 2b,c). Sometimes these surface functionalization
protocols are mixed with microscopic amounts of adhesives, as in Figure 2d.

Figure 2. Illustrations of the process steps of (a) surface activation with oxygen plasma or corona treatment, (b) chemical
gluing. (c) The most often used chemical gluing methods. (d) Adhesive-based or combined gluing methods.

3.1. Surface Activation by Oxygen Plasma Treatment

The general aim of surface activation is to remove contaminants and generate reactive
chemical groups for covalent bonding. The convenient bonding of silicon-based materials
through silanol groups (–Si–OH) is well-known in MEMS technologies (e.g., silicon wafer-
level bonding) [68]. Through surface activation, the terminal methyl groups (–CH3 ) in
PDMS (generally comprised of repeating units of –O–Si(CH3 )2 –) can be replaced by silanol
groups. These can form covalent siloxane bonds (Si-O-Si) with a similar silanol group on
another activated surface after the loss of a water molecule (Figure 2a) [59]. The processes
taking place upon plasma/corona exposure can be quite complex since the material is
simultaneously subjected to a mixture of energetic particles and radiation, e.g., electrons,
ions, UV, and ozone, which can cause at least 12 different interactions (for details see [69]).
The surface oxidation of PDMS and the increased concentration of hydroxyl groups can
also lead to the formation of strong intermolecular bonds [69,70]. These reactions may
change the surface properties significantly [71], either by the formation of a thin SiOx
layer, by increased cross-linking, or on the contrary, by the degradation of the network
Biosensors 2021, 11, 292 16 of 28

structure—all depending on the experimental conditions (e.g., plasma power, exposure

time, etc.). Most of the proposed interactions upon plasma/corona exposure make the
surface hydrophilic [69,72], and thus the quality of surface activation is generally monitored
by water drop tests (contact-angle measurements). Exposed to normal ambient conditions,
the treated surfaces quickly regain their hydrophobicity, mostly due to the migration of low
molar mass chains from the bulk to the surface, the reorientation of polar groups on the
surface into the bulk, or simply due to contamination [69,73,74]. Mechanical deformation
(even compressive strains of less than 1%) were confirmed to speed up the recovery
process [73], and thus treated PDMS samples should be handled with great care during
alignment. Overtreated PDMS was also found to regain hydrophobicity faster, emphasizing
the optimization of experimental conditions [73].
Plasma treatment is by far the most commonly used method for PDMS surface mod-
ification [74]. A large variety of gases can be used (e.g., oxygen, argon, nitrogen, hy-
drogen, fluorinate, etc.) from medium vacuum to atmospheric pressure for different
purposes [75,76]. To create a silanol rich surface for PDMS bonding, oxygen plasma is
used with pressures in the low–middle vacuum range. Besides gas content and pressure,
the two other main process parameters are the plasma power and exposure time, which
need to be optimized for successful bonding [20,49,54,55]. A significant advantage of
plasma treatment is that with a common plasma chamber (reactor), these conditions can be
fine-tuned and controlled precisely, yielding reproducible results and overall better bond
quality. Low-pressure plasma is also less harmful to functional elements (e.g., thin-film
metallic sensors) that might be integrated/embedded onto the treated surface, compared
to corona treatment. Oxygen plasma surface activation is also frequently used as the first
step in chemical gluing procedures [41,46].

3.2. Surface Activation by Corona Treatment

A corona treater is a device that generates a high voltage across an electrode at the tip
of the unit. This handheld device is usually supplied with electrodes in different shapes for
various applications [77,78]. The high potential of the electrode ionizes the surrounding air,
creating a localized plasma called corona discharge. Corona treatment is a simple, cheap,
and fast method which does not require expensive equipment or a special environment
(the discharge can be generated at room temperature and atmospheric pressure, without
the need for a vacuum system); it is portable, safe, and easy to be used [47,60,79]. A
disadvantage of the technique is that it is usually performed manually. Although the
discharge can be expanded to cover larger areas (with appropriate electrodes), the sample-
electrode distance, movement speed, treatment time are all factors that need to be precisely
controlled [19]. It has to be noted that the discharge could be harmful to metallic structures,
e.g., it has been reported to have damaged a 35 nm thick plasmonic gold thin film on
glass beyond applicability [50]. Corona surface activation can also be used combined with
chemical gluing [43,47].

3.3. Surface Activation by UV/Ozone Treatment

The third possibility for enriching the PDMS surface with silanol groups is UV/ozone
treatment [80,81]. Although this technique is significantly slower compared to both plasma
and corona treatment, due to the greater penetration of photons, deeper surface modifi-
cation (i.e., several 10 µm) could be achieved without any surface damage [74,82]. The
process parameters (UV line, lamp power, exposure time) should be precisely selected to
control the surface properties and hydrophobicity recovery [83,84]. The formation of a
glassy, brittle SiOx layer [85] might not be desirable for some applications that rely on the
flexibility of PDMS and a matching substrate. UV/ozone is also used for surface activation
in chemical gluing [50].
Biosensors 2021, 11, 292 17 of 28

3.4. Chemical Gluing

As discussed before, PDMS can be conveniently bonded to glass or other silicon-based
materials through siloxane bonds. To bond PDMS with other materials, e.g., thermoplastics,
functional groups should be created either only on the target substrate or on both materials.
These techniques are often referred to as chemical gluing, where molecular monolayers
(acting as coupling agents) are anchored on the surfaces with specific terminal functional
groups [36,45,86–88]. The most commonly used molecules are all organosilanes, namely
APTES ((3-aminopropyl)triethoxysilane), MPTMS ((3-mercaptopropyl)trimethoxysilane),
GPTES ((3-glycidyloxypropyl)triethoxysilane), GPTMS ((3-glycidyloxypropyl)trimethoxy-
silane), and TESPSA ((3-triethoxysilyl)propylsuccinic anhydride). Figure 2c illustrates the
different bonding schemes between PDMS and a general polymer substrate by using these
functional molecular layers. As can be seen, the silanol groups of the PDMS can be directly
used to form bonds with the amino groups (–NH2 ) of APTES or the thiol groups (–SH) of
MPTMS functionalized polymers. APTES can also be used paired with GPTES, GPTMS, or
TESPSA, depending on the substrate, while MPTMS is usually paired with GPTMS.
During surface activation, the carbon backbones of thermoplastic substrates
(e.g., PMMA or PC) are broken with corona or oxygen plasma treatment. These free
carbons react with inorganic silicons of the silane coupling agent, which is used to function-
alize substrate surfaces to form Si–O–C bonds [89]. It is important to emphasize that the
process parameters of chemical gluing (e.g., concentration and temperature of the solution,
thorough rinsing, drying, and subsequent thermal treatment to stabilize the self-assembled
layers before bonding) need to be controlled for optimal results, similarly to the surface
activation [41]. The hydrolytic stability of Si–O–C bonds and various chemical gluing
protocols were also investigated in detail [89]. The exact protocols and achievable bonding
strengths are given in Tables 2 and 3, and discussed in the next section.

3.5. Adhesive-Based Gluing

Adhesive-based gluing—when a microscopic amount of adhesive materials is applied
between the two surfaces instead of molecular layers—is usually not considered bonding.
Depending on the target substrate, applying epoxy or silicone-based adhesives might prove
significantly weaker than chemical gluing methods [89]. However, since adhesives are
sometimes used in combination with chemical gluing methods to improve bond quality,
these hybrid techniques are worth mentioning. A few examples are given in Figure 2d, using
different types of epoxy adhesives (e.g., Norland Optical Adhesive (NOA74) [61], LePage
Gel epoxy adhesive [53], Biocompatible Epoxy 301-2 [57]). MPTMS [53] and APTES [61] can
both be used to covalently bind the epoxy adhesive layer (with thickness in a few µm range)
with the PDMS. In a different technique, a silicone-based adhesive (PrimeCoat, ~1 µm) and
an epoxy adhesive (3–4 µm) were spin-coated on top of the substrate [57].
Biosensors 2021, 11, 292 18 of 28

Table 3. Optimized process parameters for bonding PDMS with different substrates. For the bonding strength test results see Table 2. Abbreviations: APTES, 3-aminopropyl)-
triethoxysilane; CNC, Computer Numerical Control; COC, Cyclic Olefin Copolymer; DMPMS, Dimethyl-MethylPhenylMethoxy Siloxane; DRIE, Deep Reactive Ion Etching; GPTES,
3-glycidyloxypropyl)triethoxysilane; GPTMS, 3-glycidyloxypropyl)trimethoxysilane; KOH, Potassium hydroxide; MPTMS, 3-mercaptopropyl)trimethoxysilane; NOA74, Norland Optical
Adhesive; PC, Polycarbonate; PDMS, Polydimethylsiloxane; PEN, Polyethylene Naphthalate; PET, Polyethylene Terephthalate; PI, Polyimide; PMMA, Poly(methyl methacrylate); PS,
Polystyrene; PVC, Polyvinyl Chloride; RT, Room Temperature; TESPSA, 3-triethoxysilyl)propylsuccinic anhydride; UV, Ultraviolet; Physical quantities: R–PDMS curing agent/base polymer
ratio, t—time, P—RF plasma power, p—pressure, f—gas flow rate. Conc.—concentration. The label in the bracket refers to the used medium as a—aqueous, e—ethanol, m—methanol.
PDMS Curing Surface Activation Parameters Chemical Gluing Parameters Treated Side Bonding Conditions
Substrate Molding Form and
R T t t P p f Conc. T t Adhesives T t p Ref.
Type Fabrication Technology Type Type PDMS Substrate
[[–] [◦ C] [min] [s] [W] [mTorr] [sccm] [%] [◦ C] [min] [◦ C] [min] [MPa]

PDMS NA 1:10 60 60 corona 30–40 - - - - - - - - activation activation RT 302-60 - [19]

O2 plasma KOH
PDMS: one-step
PEN, (PDMS)3M KOH activation +
PET,
photolithography and deep
1:10 90 60 activation +
20 0.4/cm2 100
- MPTMS
5.5
RT 120 - activation MPTMS + - - - [20]
PI
reactive ion etching
O2 plasma
10 0.4/cm2 100 (m)
plasma
(DRIE) plastics: -
(plastics) treatment

PDMS: APTES 5 (a) 50 60 activation

COC 1:10 80 60 O2 plasma 20 500 8 - - activation +APTES 80 120 - [21]
-substrate: hot embossing GPTMS 5 (a) 50 60 +GPTMS

PDMS: SU-8 pre-polymer gluing

PDMS,
soft-lithography 1:10 75 60 - - - - - - - - - with DMPMS pre-polymer glue pre-polymer glue 70 240 0.02 [35]
glass
substrate: - (7 µm, spin coated)

PDMS: soft-lithography, glass

COC, 150 (COC)
mold fabricated by
PP, 1:10 80 60 air plasma 30 250 (PP) 75 14 - - - - - activation activation - - - [36]
microlithography
PS 150 (PS)
plastic: -

PDMS: one-step
ABS,
photolithography and deep 5 activation +
PC, 1:10 60 60 O2 plasma 60–300 50 1500 - APTES 60 20 - activation 80 60 - [37]
reactive ion etching (DRIE) (m) APTES
PMMA
plastics: -

PDMS: SU-8
COC, PC, activation +
soft-lithography 1:10 80 30 O2 plasma 60 60 - - APTES 1 (a) RT 20 - activation RT 60 - [38]
PMMA, PS APTES
plastics: -

PDMS: SU-8 pre-polymer glue

parylene C 50 (N2 ) pre-polymer gluing pre-polymer glue
soft-lithography 1:10 RT 720 mixed plasma 60 300 5 - - - - + plasma RT 720 - [39]
on glass 30 (SF6 ) with PDMS + plasma treatment
substrate: - treatment

PDMS: SU-8
activation +
PS soft-lithography 1:13 65 120 O2 plasma 60 30 320 - APTES 1 (a) RT 20 - activation 65 60 - [40]
APTES
plastic: -

PDMS: SU-8 activation (O2 ) +

O2 plasma 60 200 - 50 activation
PMMA soft-lithography 1:10 65 240 APTES 5 (a) 85 1 - APTES + 65 120 - [41]
corona 60 - - - (corona)
plastic: - corona

PET, PI, PP, MPTMS +

PDMS:
PS, PVC, - - - corona 120 - - - MPTMS 2 (a) RT 1-5 - activation corona RT 10 - [43]
-substrates: -
metals treatment

PC,
PDMS: SU-8
PET, APTES 1 (a) RT 20 activation + activation +
soft-lithography 1:10 80 30 O2 plasma 60 50-60 - - - RT 60 - [44]
PI, GPTES 1 (a) RT 20 APTES GPTES
plastics: hot embossing
PMMA

PDMS: SU-8
PDMS, APTES 1 (a) RT 20 activation + activation +
soft-lithography 1:10 80 30 O2 plasma 60 - - - - RT 60 - [45]
PET GPTMS 1 (a) RT 20 APTES GPTMS
plastic: -

PC, PDMS:
PET, -plastics: CNS APTES 1 (a) RT 30 activation + activation +
1:10 80 120 O2 plasma 60 - - - - RT - - [46]
PMMA, milling, TESPSA 1 (a) RT 30 TESPSA APTES
PS engraving
Biosensors 2021, 11, 292 19 of 28

Table 3. Cont.

Substrate PDMS Curing Surface Activation Parameters Chemical Gluing Parameters Treated Side Bonding Conditions
Molding Form and
Type R T t t P p f Conc. T t Adhesives T t p Ref.
Fabrication Technology Type Type PDMS Substrate
[[–] [◦ C] [min] [s] [W] [mTorr] [sccm] [%] [◦ C] [min] [◦ C] [min] [MPa]

monoglycidyl ether
PC, PDMS: SU-8
terminated, activation +
PET, soft-lithography
1:10 80 60 corona 60 - - - APTES 5 (a) 80 20 low-molecular-weight activation APTES + 25 15 0.1 [47]
PMMA, plastics: CNC milling,
PDMS was adhesive
PS engraving
added at 80 ◦ C for 4 h

35
PDMS: SU-8 60 45 10.5 60 (glass)
(PDMS) -
glass soft-lithography 1:10 + + O2 plasma (PDMS) 40 - - - - - activation activation - - - [48]
120 -
substrate: - 100 135 18 (glass) (PDMS)
(glass)

glass - 1:10 100 60 O2 plasma 300 200 - 60 - - - - - activation activation 75 300 0.5 [49]

PDMS: SU-8
activation +
gold on glass soft-lithography 1:10 70 60 UV/ozone 300 - - - MPTMS 2 (e) RT 60 - cleaning 60 60 - [50]
MPTMS
substrate: -

PDMS: SU-8
glass soft-lithography 1:10 70 60 corona 120 - - - - - - - - activation activation 80 720 - [50]
substrate: -

1 epoxy adhesive
activation +
MPTMS (m) RT 60 (LePage Gel epoxy activation +
PI - 1:10 60 120 UV/ozone 600 - - - GPTMS or RT 720 0.03 [53]
GPTMS 1 RT 60 adhesive) MPTMS
MPTMS + epoxy
(m) —optionally

PDMS - 1:10 200 8 air plasma 50 18 200 - - - - - - activation activation RT 5 0.03 [54]

PDMS - 1:10 80 60 O2 plasma 300 300 atm 15 - - - - - activation activation 160 20 1.4 [55]

PDMS: SU-8
PDMS,
soft-lithography 1:10 65 60 O2 plasma 12 150 - - - - - - - activation activation - - - [56]
glass
substrate: -

silicone adhesive
PDMS: SU-8 30
(PrimeCoat, 1 µm) activation +
PS soft-lithography 1:10 60 180 O2 plasma (PDMS) 18 45 100 - - - - activation 60 180 - [57]
epoxy adhesive adhesives
plastic: - 75 (PS)
(Epoxy 301-2, 3-4 µm)

PDMS: SU-8
glass - - - O2 plasma 20 20 1000 - - - - - - activation activation - - - [59]
soft-lithography

mold fabricated by
PDMS 1:10 60 720 corona 30 - - - - - - - - activation activation - - - [60]
Xurography

mold fabricated by
PDMS 1:10 60 720 O2 plasma 20 20 700 - - - - - - activation activation - - - [60]
Xurography

epoxy adhesive (NOA74),

COC, deposited by spin coating activation +
PDMS: SU-8
PET, (6000 rpm 1 min), cured APTES +
soft-lithography 1:10 80 90 O2 plasma 60 25 - - APTES 1 (a) RT 20 - - 90 - [61]
PMMA, PS, with UV lamp NOA74 epoxy
plastics: -
glass, metals glue
(20 mW/cm2 )

PDMS:
Aluminium mold, O2 plasma 180–300 - -
PDMS, glass 1:10 70 180 - - - - - - activation activation 80 15 - [62]
CNC milled UV/ozone 180–300 - -
substrate: -
Biosensors 2021, 11, 292 20 of 28

4. Bonding Strategies for Various Substrates

In this section, the exact protocols used to bond PDMS with various substrates will be
discussed in detail. Table 2 collects the applicable protocols for each substrate type and
presents the achievable bonding strength, systematically ordered along with the testing
methods of Table 1. Table 3 gives each protocol’s optimized process parameters to achieve
the best results for a given substrate type.

4.1. PDMS
One of the most straightforward approaches to close a PDMS microchannel with a
substrate is PDMS–PDMS bonding since a simple surface activation is sufficient, which
can be performed in many ways. The most comprehensive studies on the optimization of
oxygen plasma and corona treatment process parameters for the best achievable bonding
strengths were performed by Bhattacharya et al. [59,60]. They mapped a wide range of RIE
(Reacting Ion Etching) power (5–150 W), chamber pressure (20–1000 mTorr), and time of
exposure (5–60 s) for both an inductively coupled high-density (ICP) plasma system, and a
plasma-enhanced chemical vapor deposition (PECVD) system. The optimal parameters
were found to be 700 mTorr, 20 W, and 30 s exposure, with an ICP system, which resulted
in 400 kPa (58 psi) burst pressure [59]. In a newer work, they managed to reproduce
these results with the same experimental conditions, also demonstrating huge variation
(average burst pressure: 300 kPa, range: 180–715 kPa), and obtaining slightly smaller
values with corona treated surfaces (average: 290 kPa, range: 227–380 kPa) [60]. The better
reproducibility of corona treatment (handheld discharge unit, 15 kV output voltage for 30 s)
is admittedly against expectations. It has to be noted that these burst pressures, obtained
with optimized process parameters, are 2–3× better compared to other similar available
data, e.g., Yousuff et al., who measured 105–180 kPa after oxygen plasma, and 87–95 kPa
for UV/ozone treatment, respectively [62]. However, as also shown in Table 3, in many of
these cases, the exact process parameters were not elaborated in the papers [45,62]. The
parameters of atmospheric RF glow discharge plasma were optimized by Jung et al. [55],
and similarly good results—in terms of peel test—were demonstrated with low-pressure
RF air plasma as well [54].
As for chemical gluing, Lee et al. demonstrated a 2× increase in tensile strength
(from 91 to 184 kPa) when using APTES/GPTMS gluing compared to normal oxygen
plasma activation—although the plasma parameters are not given and presumably were
not optimized either [45].
Another important technique that needs to be noted here is pre-polymer gluing (also
referred to as ‘uncured PDMS adhesive’ or ‘partially cured adhesive’). In this approach,
uncured/partially cured PDMS is placed between two fully cured substrates, or one of the
two bonded pieces is only partially cured before making contact with the other substrate.
Here, the pre-polymer adhesive or partially cured piece is cured during/after the bonding
process. As an example, for partial cure bonding, Bhattacharya et al. pre-cured PDMS at
60 ◦ C for 35 min before bonding and then allowed the piece to fully cure overnight following
the bonding [60]. For both groups who tested this approach, the measured maximum
burst pressures were at least 2× higher, compared to the same structures bonded with only
either oxygen plasma or corona treatment (e.g., an average of 671 kPa), making it superior
compared to other techniques [60,62]. Similarly good results were reported by Cao et al.
with DMPMS-based (dimethyl-methylphenylmethoxy siloxane) pre-polymer gluing [35].

4.2. Glass (Silicon)

PDMS–glass and PDMS–silicon are also regularly used pairs to close PDMS microchan-
nels. In the latter case, PDMS is bonded to the native oxide of silicon, so although glass is
more frequently used, the protocols for bonding PDMS to silicon can be considered to be
the same.
All of the approaches listed in Table 2 are based on surface activation similar to
PDMS–PDMS bonding. The optimized oxygen plasma parameters with an ICP equipment
Biosensors 2021, 11, 292 21 of 28

were found to be 1000 mTorr chamber pressure, 20 W plasma power, and 30 sec exposure
time [59]. The resulting 510 kPa (74 psi) burst pressure is more than 25% higher compared
to PDMS–PDMS bonds also prepared with optimized oxygen plasma parameters [59].
For identical process parameters, others also reported significantly higher bonds between
PDMS–glass than PDMS–PDMS, and the difference was even higher for UV/ozone treat-
ment (95 to 314 kPa, respectively) [62]. Prepolymer gluing was also successfully used with
glass substrates [35,62].
In Table 2, two application-specific examples are also given. Bonding PDMS to gold-
coated glass [90] could be important for several sensor applications (either electrochemical
or optical). For this purpose, surface activation + MPTMS functionalization of the PDMS
could be used, relying on the terminal thiol groups to bond covalently to the gold surface.
Bakouche et al. successfully demonstrated this method with an SPR (surface plasmon
resonance) chip and even obtained higher lap shear strength values than for PDMS–glass
bonding with simple corona surface activation [50]. Parylene, an organic polymer with
a para-xylylene backbone, has many favorable properties considering microfabrication
or microfluidics [91]. To irreversibly bond parylene with PDMS, Rezai et al. tried both
pre-polymer gluing and plasma treatment with success [39]. By optimizing the process
parameters of the latter (using a mixture of SF6 and N2 ), they obtained a very high bond
tensile strength of 1.4 MPa and 145 kPa burst pressure (tested with water) [39].

4.3. PMMA
Similar to PDMS, poly(methylmethacrylate) (PMMA) is also a popular material for
micro-TAS or LoC fabrication [92]. PMMA is an optically transparent thermoplastic (one of
the least hydrophobic materials used in microfluidics), and its low price and compatibility
with several microfabrication methods (e.g., injection molding, milling, hot embossing, etc.)
makes it an ideal choice for disposable microfluidic devices. Since its direct bonding with
PDMS is not possible [93] with an adequate bonding strength (e.g., with pre-polymer gluing,
only 15 kPa tensile strength could be obtained [42]), a chemical gluing method has to be
selected. There are numerous possibilities: nearly all of the illustrated approaches presented
in Figure 2b,c can be used for PDMS–PMMA bonding, combined with either oxygen plasma
or corona surface activation. Most of the protocols listed in Table 2 uses only one molecular
adhesive. In these cases, the PMMA is functionalized with either APTES [37,38,41] or
MPTMS [43] then bonded to a surface-activated PDMS. Others functionalize both surfaces
with APTES/GPTMS [44] or APTES/TESPSA [46] pairs. By comparing the resulting tensile
strength and burst pressure values, no direct improvement can be observed in favor of the
double functionalization. With optimized plasma treatment and APTES coating parameters,
high tensile strength (1.6 MPa [41]) and high burst pressures (>500 kPa, measured in air [37])
were obtained by coating only the PMMA piece.
An interesting approach is presented by Zhang et al., who used APTES functionaliza-
tion on PMMA, followed by the addition of monoglycidyl ether terminated, low-molecular-
weight PDMS [47]. In this way, PDMS-like terminals were placed onto the substrate (via the
amine–epoxy bond), which can be subsequently bonded with bulk PDMS through classical
siloxane bond formation. The aminosilane + monoglycidyl ether terminated PDMS com-
bined functionalization thus leads back to PDMS–PDMS bonding, which can be used with a
variety of substrates (e.g., PS, PC), with good results (580–620 kPa burst pressures) [47].

4.4. PC
Similar to PMMA, PC is another optically transparent thermoplastic with a high glass
transition temperature (145 ◦ C), low moisture absorption, and durability that makes it
ideal for high-temperature applications. All of the chemical gluing protocols previously
discussed for PMMA-PDMS bonding can be used for polycarbonate (PC) as well, including
surface activation and APTES [37,38,58] or MTPMS [43] treatment, using APTES+ mono-
glycidyl ether terminated, low-molecular-weight PDMS functionalization [47], or using
APTES/TESPSA [46] or APTES/ GPTMS [44] linker pairs. A notable difference is that with
Biosensors 2021, 11, 292 22 of 28

the identical bonding techniques, the resulting bond tensile strength or burst pressure is
10–50% higher for PDMS–PC than PDMS–PMMA bonds, which makes PC a good choice
for high-pressure or high flow-rate applications.

4.5. PS
Polystyrene (PS) is another optically transparent material that is frequently used
material for cell cultures and biosensor applications, thanks to its good biocompatibil-
ity [57,94,95]. For PDMS–PS bonding, three previously discussed techniques were also
successfully applied, namely, surface activation combined with either APTES [38,40],
APTES/TESPSA [46], or APTES + low molecular weight PDMS linker [47] treatment.
A one-step bonding technique, which stands out from the other chemical gluing
methods, was also presented by Xu et al. [36]. They used air plasma (a mixture of oxygen
and nitrogen) to activate the surface of both PS and PDMS substrates. Oxygen is known to
create a large number of hydroxy (–OH) groups with a small portion of carbonyl (–C=O)
groups on the surface of carbon-based plastics. The presence of nitrogen in the plasma
enables the formation of imine groups ((R1, R2)C=NH) and their free radical forms, which
is the essential contribution for irreversible PDMS–plastic bonding [36,96]. This effect
of nitrogen was confirmed by both XPS performed on the treated surfaces and negative
controls with oxygen/argon mixed plasma. Although this one-step bonding process was
reproduced with COC and PP polymers, and they all performed well at dynamic leakage
and burst tests (e.g., burst pressures above 500 kPa, measured with compressed air), this
technique was still not tested for other mainstream polymers (PMMA, PC, PET, etc.).
Li et al. used an adhesive-based gluing technique to attach PDMS to PS by subse-
quently spin coating a silicone-based adhesive (PrimeCoat, ~1 µm) and an epoxy adhesive
(3–4 µm) onto the substrates [57]. The resulting maximum burst pressure was comparable
to other chemical gluing methods (see Table 2). The same adhesive-based technique was
successfully implemented to bond PDMS with glass and PET as well [57].

4.6. PET
Poly(ethylene terephthalate) (PET) is optically transparent, strong, and lightweight
plastic that is one of the favored substrates to choose when the planned application requires
flexibility. Its chemical inertness, good gas permeability, and low cost can also be advan-
tageous for several applications. Most previously mentioned chemical gluing technique
can be selected for PDMS–PET bonding (namely, surface activation combined with either
MPTMS [43], APTES/TESPSA [46], APTES/GPTMS [44,45], or APTES + low molecular
weight PDMS linker [47] treatments. The resulting tensile strengths and burst pressures are
a bit mixed (e.g., compared to PDMS–PMMA higher [43] and lower [47] tensile strengths
are also reported with the same protocols), but all listed approaches can be successfully
used to create reliable bonding with this flexible substrate.
Here, another technique that stands out has to be noted. Baraket et al. used KOH-
based activation and hydrolyzed the surface of the tested polymers, which were then
functionalized with MPTMS. Subsequently, oxygen plasma was made to substitute the
propyl–thiol chain with hydroxyl groups by cleaving the terminal groups on the MPTMS
treated polymers. These silanol groups form the siloxane bonds with the surface-activated
PDMS [20]. Please note that this proposed approach is different from other chemical gluing
methods which use MPTMS, e.g., proposed by Wu et al. [43], where the organosilane is
bound to the corona treated surface of the PET substrate through thiol groups. The silane
layer is in turn bonded with the activated PDMS, as illustrated in Figure 2d. Although
Baraket et al. successfully applied this method to both PI and PEN besides PET [20], since
no tensile or burst tests were performed, its comparison with the method of Wu et al. is not
directly possible.
Biosensors 2021, 11, 292 23 of 28

4.7. PI
Polyimides (including DuPont’s favored commercial product, ‘Kapton’) are popular
substrate materials for flexible electronic devices [97,98]. One of their main advantages
compared to the similarly flexible PET is their high glass transition temperature and
stability over a wide temperature range (e.g., between −269 ◦ C and 400 ◦ C) [53]. Its
high dielectric constant, good chemical resistance and compatibility with microelectronics
packaging technologies also make it an excellent candidate to bridge the gap between
microelectronics and microfluidics. The most detailed investigation of PDMS–PI bonding
was performed by Hoang et al., who compared chemical gluing methods with adhesives
(epoxy, silicon) and with hybrid MPTMS-epoxy adhesive methods, and tested the resulting
bonds with peel tests [53]. Using adhesives only (e.g., LePage Gel epoxy adhesive), the
resulting peel strengths were quite low (1.7–72 N/m). Out of the two chemical gluing
methods, they only obtained a strong bond with the MPTMS/GPTMS pair (200 N/m),
the APTES/GPTMS pair yielded a 100-times smaller 2.7 N/m peel strength. This result
is somewhat surprising, since other groups successfully used the APTES/GPTMS pair
with similar procedures (functionalizing the PDMS with APTES and the plastic with
GPTMS) [21,44]. For PDMS–PI bonding, the best results were obtained by combining
MPTMS surface functionalization with the epoxy adhesive (470 N/m). Both the MPTMS-
epoxy and the MPTMS/GPTMS strategy yielded peel strengths that are comparable or
even better than some values reported for PDMS–PDMS pairs [54,55], which indicates a
strong and reliable bond.

4.8. Other Polymer Substrates

In Table 2, protocols that can be used for bonding PDMS with five more polymers are
also given. These strategies were previously discussed in accordance with other substrates.
In short, the air plasma surface activation used by Xu et al. was also successfully applied
to bond PDMS with polypropylene (PP) and cyclic olefin copolymer (COC) [36]. For COC,
Cortese et al. investigated other protocols from oxygen plasma treatment to oxygen plasma
+ APTES or GPTMS treatment, including oxygen plasma + APTES/GPTMS combined
functionalization [21], which yielded the best results. For polyethylene naphthalate (PEN),
the protocol of Baraket et al. was used with success, including KOH surface activation,
MPTMS treatment, and oxygen plasma [20]. For polyvinyl chloride (PVC), a simple
corona activation combined with MPTMS treatment yielded high bonding strengths [43],
comparable to PDMS–PMMA or PDMS–PET with the same protocol. Finally, oxygen
plasma combined with APTES was successfully used to bond PDMS with acrylonitrile
butadiene styrene (ABS) as well.

4.9. Metals
Although metals cannot be considered as typical substrates in microfluidic fabrication,
it has to be noted that by simple chemical gluing, PDMS can be conveniently bonded
with various metals. The thiol–metal bonds, which were discussed in accordance with
gold-coated glass [50], were successfully used to bond PDMS with iron, copper, aluminum,
or brass by using a simple surface activation + MPTMS treatment protocol [43].

5. Conclusions
Strategies and methods to bond PDMS with various rigid and flexible substrate mate-
rials were collected and evaluated. As it was shown, with optimized process parameters,
high-quality and high-strength bonds can be achieved. Commonly achievable strengths
with optimized process parameters—quantitatively, based on the three most widely used
testing methods, which are tensile strength tests, burst tests, and channel leakage tests—are
>400 kPa tensile strength, >500 kPa burst pressure, or flow rates equal to several 1000× of
the internal volume of the channels per minute. This could be achieved with simple surface
activation methods for silicone-based substrates (e.g., glass, silicon, or PDMS). For polymers
and metallic surfaces, chemical gluing is advised. In some exceptional cases, a combined
Biosensors 2021, 11, 292 24 of 28

adhesive + chemical gluing method could yield extra-strong bonds. It is hoped that the
collected optimized process parameters in Table 3 will help others in this field to quickly
tune their processes based on their substrates, available equipment, or application area.
Despite the growing trend and need for LoC and Micro-TAS devices, PDMS-based mi-
crofluidics remains an academic and small-scale R&D, rather than an industrial/commercial
success, mostly owing to the scalability problems related to the micromachining and proto-
typing technologies considering large scale manufacture [99]. However, as demonstrated
in the paper, high-quality bonds can now be formed between PDMS and many other
plastic materials, some of which are compatible with either large-scale fabrication methods
(e.g., PMMA), or electronics packaging technologies (e.g., PI). The combination of PDMS
with such materials might open the door for its larger commercial/industrial success in the
near future.

Author Contributions: A.B. (Alexandra Borók): literature survey, writing, visualization; K.L.: lit-
erature survey, writing; A.B. (Attila Bonyár): conceptualization, supervision, writing, review and
editing. All authors have read and agreed to the published version of the manuscript.
Funding: The research reported in this paper and carried out at the Budapest University of Technol-
ogy and Economics has been supported by the NRDI Fund (TKP2020 IES, Grant No. BME-IE-BIO)
based on the charter of bolster issued by the NRDI Office under the auspices of the Ministry for Inno-
vation and Technology. Prepared with the professional support of the Doctoral Student Scholarship
Program of the Co-operative Doctoral Program of the Ministry for Innovation and Technology from
the source of the National Research Development and Innovation Fund.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The research reported in this paper and carried out at the Budapest University
of Technology and Economics has been supported by the NRDI Fund (TKP2020 IES, Grant No.
BME-IE-BIO) based on the charter of bolster issued by the NRDI Office under the auspices of the
Ministry for Innovation and Technology. Prepared with the professional support of the Doctoral
Student Scholarship Program of the Co-operative Doctoral Program of the Ministry for Innovation
and Technology from the source of the National Research Development and Innovation Fund.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
APTES: 3-aminopropyl)triethoxysilane; CNC, Computer Numerical Control; COC,
Cyclic Olefin Copolymer; DMPMS, Dimethyl-MethylPhenylMethoxy Siloxane; DRIE, Deep
Reactive Ion Etching; GPTES, 3-glycidyloxypropyl)triethoxysilane; GPTMS, 3-glycidyloxyp-
ropyl)trimethoxysilane; ICP, Inductively Coupled High-Density Plasma System; KOH,
Potassium hydroxide; LoC, Lab-on-a-Chip; MEMS, Micro Electro-Mechanical Systems;
micro-TAS, micro Total Analysis Systems; MPTMS, 3-mercaptopropyl)trimethoxysilane;
NOA74, Norland Optical Adhesive; PC, Polycarbonate; PDMS, Polydimethylsiloxane;
PECVD, Plasma Enhanced Chemical Vapor Deposition; PEN, Polyethylene Naphtha-
late; PET, Polyethylene Terephthalate; PI, Polyimide; PMMA, Poly(methyl methacrylate);
PS, Polystyrene; PVC, Polyvinyl Chloride; R&D, Research and Development; RIE, Re-
acting Ion Etching; RT, Room Temperature; SPR, Surface Plasmon Resonance; TESPSA,
3-triethoxysilyl)propylsuccinic anhydride; THF, Tetrahydrofuran; UV, Ultraviolet.

References
1. Bowen, J.; Cheneler, D.; Robinson, A.P.G. Direct e-beam lithography of PDMS. Microelectron. Eng. 2012, 97, 34–37. [CrossRef]
2. Raj, M.K.; Chakraborty, S. PDMS microfluidics: A mini review. J. Appl. Polym. Sci. 2020, 137, 48958. [CrossRef]
3. Becker, H.; Locascio, L.E. Polymer microfluidic devices. Talanta 2002, 56, 267–287. [CrossRef]
4. Mata, A.; Fleischman, A.J.; Roy, S. Characterization of polydimethylsiloxane (PDMS) properties for biomedical mi-
cro/nanosystems. Biomed. Microdevices 2005, 7, 281–293. [CrossRef]
Biosensors 2021, 11, 292 25 of 28

5. De Jong, J.; Lammertink, R.G.H.; Wessling, M. Membranes and microfluidics: A review. Lab Chip 2006, 6, 1125–1134. [CrossRef]
6. Gale, B.K.; Jafek, A.R.; Lambert, C.J.; Goenner, B.L.; Moghimifam, H.; Nze, U.; Kamarapu, S.K. A Review of Current Methods in
Microfluidic Device Fabrication and Future Commercialization Prospects. Inventions 2018, 3, 60. [CrossRef]
7. McDonald, J.C.; Duffy, D.C.; Anderson, J.R.; Chiu, D.T.; Wu, H.K.; Schueller, O.J.A.; Whitesides, G.M. Fabrication of microfluidic
systems in poly(dimethylsiloxane). Electrophoresis 2000, 21, 27–40. [CrossRef]
8. Becker, H.; Gartner, C. Polymer microfabrication methods for microfluidic analytical applications. Electrophoresis 2000, 21, 12–26.
[CrossRef]
9. Sia, S.K.; Whitesides, G.M. Microfluidic devices fabricated in poly(dimethylsiloxane) for biological studies. Electrophoresis
2003, 24, 3563–3576. [CrossRef] [PubMed]
10. Brown, L.; Koerner, T.; Horton, J.H.; Oleschuk, R.D. Fabrication and characterization of poly(methylmethacrylate) microfluidic
devices bonded using surface modifications and solvents. Lab Chip 2006, 6, 66. [CrossRef] [PubMed]
11. Sackmann, E.K.; Fulton, A.L.; Beebe, D.J. The present and future role of microfluidics in biomedical research. Nature
2014, 507, 181–189. [CrossRef] [PubMed]
12. Wu, J.W.; Wang, R.D.; Yu, H.X.; Li, G.J.; Xu, K.X.; Tien, N.C.; Roberts, R.C.; Li, D.C. Inkjet-printed microelectrodes on PDMS as
biosensors for functionalized microfluidic systems. Lab Chip 2015, 15, 690–695. [CrossRef]
13. Alkhalaf, Q.; Pande, S.; Palkar, R. Review of Polydimethylsiloxane (PDMS) as a Material for Additive Manufacturing, Innovative
Design, Analysis and Development Practices in Aerospace and Automotive Engineering. Lect. Notes Mech. Eng. Springer Singap.
2021, 7, 265–275.
14. Magdassi, S.; Kamyshny, A. Nanomaterials for 2D and 3D Printing; Wiley: New York, NY, USA, 2017.
15. Cheng, S.; Wu, Z.G. Microfluidic stretchable RF electronics. Lab Chip 2010, 10, 3227–3234. [CrossRef] [PubMed]
16. Siegel, A.C.; Bruzewicz, D.A.; Weibel, D.B.; Whitesides, G.M. Microsolidics: Fabrication of three-dimensional metallic microstruc-
tures in poly(dimethylsiloxane). Adv. Mater. 2007, 19, 727–733. [CrossRef]
17. Eduok, U.; Faye, O.; Szpunar, J. Recent developments and applications of protective silicone coatings: A review of PDMS
functional materials. Prog. Org. Coat. 2017, 111, 124–163. [CrossRef]
18. Gupta, S.; Vilouras, A.; Dahiya, R. Polydimethylsiloxane as polymeric protective coating for fabrication of ultra-thin chips.
Microelectron. Eng. 2020, 221, 7. [CrossRef]
19. Yang, C.; Wang, W.; Li, Z. Optimization of Corona-triggered PDMS-PDMS Bonding Method. IEEE Int. Conf. Nano/Micro Eng. Mol.
Syst. 2009, 5, 319–322.
20. Baraket, A.; Zine, N.; Lee, M.; Bausells, J.; Jaffrezic-Renault, N.; Bessueille, F.; Yaakoubi, N.; Errachid, A. Development of a
flexible microfluidic system based on a simple and reproducible sealing process between polymers and poly(dimethylsiloxane).
Microelectron. Eng. 2013, 111, 332–338. [CrossRef]
21. Cortesea, B.; Mowlemb, M.C.; Morgan, H. Characterisation of an irreversible bonding process for COC–COC and COC–PDMS–
COC sandwich structures and application to microvalve. Sens. Actuators B Chem. 2011, 160, 1473–1480. [CrossRef]
22. Thompson, C.S.; Abate, A.R. Adhesive-based bonding technique for PDMS microfluidic devices. Lab Chip 2013, 13, 632–635.
[CrossRef]
23. Tsao, C.W.; Syu, W.C. Bonding of thermoplastic microfluidics by using dry adhesive tape. RSC Adv. 2020, 10, 30289–30296.
[CrossRef]
24. Li, J.M.; Liang, C.; Zhang, H.; Liu, C. Reliable and high quality adhesive bonding for microfluidic devices. Micro Nano Lett.
2017, 12, 90–94. [CrossRef]
25. Souza, A.; Ribeiro, J.; Araújo, F.S. Study of PDMS characterization and its applications in biomedicine: A review. J. Mech. Eng.
Biomech. 2019, 4, 1–9.
26. Hwang, Y.; Candler, R.N. Non-planar PDMS microfluidic channels and actuators: A review. Lab Chip 2017, 17, 3948–3959.
[CrossRef] [PubMed]
27. Zheng, W.F.; Wang, Z.; Zhang, W.; Jiang, X.Y. A simple PDMS-based microfluidic channel design that removes bubbles for
long-term on-chip culture of mammalian cells. Lab Chip 2010, 10, 2906–2910. [CrossRef] [PubMed]
28. Wengler, J.; Ognier, S.; Zhang, M.X.; Levernier, E.; Guyon, C.; Ollivier, C.; Fensterbank, L.; Tatoulian, M. Microfluidic chips for
plasma flow chemistry: Application to controlled oxidative processes. React. Chem. Eng. 2018, 3, 930–941. [CrossRef]
29. Sollier, E.; Murray, C.; Maoddi, P.; Di Carlo, D. Rapid prototyping polymers for microfluidic devices and high pressure injections.
Lab Chip 2011, 11, 3752–3765. [CrossRef] [PubMed]
30. Tiggelaar, R.M.; Benito-Lopez, F.; Hermes, D.C.; Rathgen, H.; Egberink, R.J.M.; Mugele, F.G.; Reinhoudt, D.N.; van den Berg, A.;
Verboom, W.; Gardeniers, H.J.G.E. Fabrication, mechanical testing and application of high-pressure glass microreactor chips.
Chem. Eng. J. 2007, 131, 163–170. [CrossRef]
31. Chen, H.W.; Ran, T.; Gan, Y.; Zhou, J.J.; Zhang, Z.; Zhang, L.W.; Zhang, D.Y.; Jiang, L. Ultrafast water harvesting and transport in
hierarchical microchannels. Nat. Mater. 2018, 17, 935. [CrossRef] [PubMed]
32. Van Steijn, V.; Kreutzer, M.T.; Kleijn, C.R. Velocity fluctuations of segmented flow in microchannels. Chem. Eng. J.
2008, 135, S159–S165. [CrossRef]
33. Abkarian, M.; Faivre, M.; Stone, H.A. High-speed microfluidic differential manometer for cellular-scale hydrodynamics. Proc.
Natl. Acad. Sci. USA 2006, 103, 538–542. [CrossRef] [PubMed]
Biosensors 2021, 11, 292 26 of 28

34. Sofla, A.Y.N.; Martin, C. Study of the vapor-assisted method for bonding PDMS and glass: Effect of the vapor source. J.
Micromechanics Microeng. 2010, 20, 125024. [CrossRef]
35. Cao, H.H.; Dinh, T.H.N.; Hamdi, F.S.; Couty, M.; Martincic, E.; Woytasik, M.; Dufour-Gergam, E. Reversible bonding by dimethyl-
methylphenylmethoxy siloxane based stamping technique for reusable poly(dimethyl)siloxane microfluidic chip. Micro Nano Lett.
2015, 10, 229–232. [CrossRef]
36. Xu, B.Y.; Yan, X.N.; Xu, J.J.; Chen, H.J. One step high quality poly(dimethylsiloxane)-hydrocarbon plastics bonding. Biomicrofluidics
2012, 6, 16507–165078. [CrossRef] [PubMed]
37. Pečar, B.; Možek, M.; Vrtačnik, D. Thermoplastic—PDMS polymer covalent bonding for microfluidic applications. J. Microelectron.
2017, 47, 147–154.
38. Sunkara, V.; Park, D.K.; Hwang, H.; Chantiwas, R.; Soper, S.A.; Cho, Y.K. Simple room temperature bonding of thermoplastics
and poly(dimethylsiloxane). Lab Chip 2011, 11, 962–965. [CrossRef]
39. Rezai, P.; Selvaganapathy, P.R.; Rwohl, G. Plasma enhanced bonding of polydimethylsiloxane with parylene and its optimization.
J. Micromech. Microeng. 2011, 21, 065024. [CrossRef]
40. Song, K.Y.; Zhang, H.B.; Zhang, W.J.; Teixeira, A. Enhancement of the surface free energy of PDMS for reversible and leakage-free
bonding of PDMS-PS microfluidic cell-culture systems. Microfluid. Nanofluidics 2018, 22, 135. [CrossRef]
41. Kim, K.; Park, S.W.; Yang, S.S. The optimization of PDMS-PMMA bonding process using silane primer. BioChip J. 2010, 4, 148–154.
[CrossRef]
42. Chow, W.W.Y.; Lei, K.F.; Shi, G.; Li, W.J.; Huang, Q. Micro Fluidic Channel Fabrication by PDMS-Interface Bonding. Smart Mater.
Struct. 2005, 15, 141–148.
43. Wu, W.; Wu, J.; Kima, J.H.; Lee, N.Y. Instantaneous room temperature bonding of a wide range of non-silicon substrates with
poly(dimethylsiloxane) (PDMS) elastomer mediated by a mercaptosilane. Lab Chip 2015, 15, 2819–2825. [CrossRef]
44. Tang, L.; Lee, N.Y. A facile route for irreversible bonding of plastic-PDMS hybrid microdevices at room temperature. Lab Chip
2010, 10, 1274–1280. [CrossRef] [PubMed]
45. Lee, N.Y.; Chung, B.H. Novel Poly(dimethylsiloxane) Bonding Strategy via Room Temperature “Chemical Gluing”. Langmuir
2009, 25, 3861–3866. [CrossRef]
46. Sivakumar, R.; Lee, N.Y. Chemically robust succinimide group-assisted irreversible bonding of poly(dimethylsiloxane)-
thermoplastic microfluidic devices at room temperature. Analyst 2020, 145, 6887–6894. [CrossRef] [PubMed]
47. Zhang, H.N.; Lee, N.Y. Non-silicon substrate bonding mediated by poly(dimethylsiloxane) interfacial coating. Appl. Surf. Sci.
2015, 327, 233–240. [CrossRef]
48. Wang, J.; Wang, S.; Zhang, P.; Li, Y. Easy-Disassembly Bonding of PDMS Used for Leak-Tight Encapsulation of Microfluidic
Devices. In Proceedings of the 18th International Conference on Electronic Packaging Technology (ICEPT), Harbin, China, 16–19
August 2017; pp. 1051–1055.
49. Xionga, L.C.; Chena, P.; Zhoua, Q. Adhesion promotion between PDMS and glass by oxygen plasma pre-treatment, J. Adhes. Sci.
Technol. 2014, 28, 1046–1154. [CrossRef]
50. Bakouche, M.T.; Ganesan, S.; Guérin, S.; Hourlier, D.; Bouazaoui, M.; Vilcot, J.P.; Maricot, S. Leak-free integrated microfluidic
channel fabrication for surface plasmon resonance applications. J. Micromech. Microeng. 2020, 30, 125003. [CrossRef]
51. Campbell, F.C. Chapter 8—Adhesive Bonding and Integrally Cocured Structure: A Way to Reduce Assembly Costs through
Parts Integration. In Manufacturing Processes for Advanced Composites; Elsevier Science: Amsterdam, The Neatherlands, 2004;
pp. 241–301.
52. Duncan, B. 14–Developments in testing adhesive joints. In Advances in Structural Adhesive Bonding; Woodhead Publishing:
Sawston, UK, 2010; pp. 389–436.
53. Hoang, M.V.; Chung, H.J.; Elias, A.L. Irreversible bonding of polyimide and polydimethylsiloxane (PDMS) based on a thiolepoxy
click reaction. J. Micromech. Microeng. 2016, 26, 105019. [CrossRef]
54. Chen, C.F. Characterization of fracture energy and toughness of air plasma PDMS-PDMS bonding by T-peel testing. J. Adhes. Sci.
Technol. 2017, 32, 1239–1252. [CrossRef]
55. Jung, K.H.; Kim, D.G.; Jung, S.B. Adhesion of PDMS substrates assisted by Plasma Graft Polymerization. Surf. Interface Anal.
2016, 48, 597–600. [CrossRef]
56. Abidin, U.; Daud, N.A.S.M.D.; Le Brun, V. Replication and leakage Test of Polydimethylsiloxane (PDMS) Microfluidics Channel.
AIP Conf. Proc. 2019, 2062, 020064.
57. Li, X.; Wu, N.; Rojanasakul, Y.; Liu, Y. Selective stamp bonding of PDMS microfluicdic devices to polymer substrates for biological
applications. Sens. Actuators A Phys. 2013, 193, 186–192. [CrossRef]
58. Aran, K.; Sasso, L.A.; Kamdar, N.; Zahn, J.D. Irreversible, direct bonding of nanoporous polymer membranes to PDMS or glass
microdevices. Lab Chip 2010, 10, 548–552. [CrossRef]
59. Bhattacharya, S.; Datta, A.; Berg, J.M.; Gangopadhyay, S. Studies on Surface Wettability of Poly(Dimethyl) Siloxane (PDMS) and
Glass Under Oxygen-Plasma Treatment and Correlation With Bond Strength. J. Microelectromechanical. Syst. 2005, 14, 590–597.
[CrossRef]
60. Eddings, M.A.; Johnson, M.A.; Gale, B.K. Determining the optimal PDMS–PDMS bonding technique for microfluidic devices. J.
Micromech. Microeng. 2008, 18, 067001. [CrossRef]
Biosensors 2021, 11, 292 27 of 28

61. Agostini, M.; Greco, G.; Cecchini, M. Polydimethylsiloxane (PDMS) irreversible bonding to untreated plastics and metals for
microfluidics applications. APL Mater. 2019, 7, 081108. [CrossRef]
62. Yousuff, C.M.; Danish, M.; Ho, E.T.W.; Basha, I.H.K.; Hamid, N.H.B. Study on the Optimum Cutting Parameters of an Aluminum
Mold for Effective Bonding Strength of a PDMS Microfluidic Device. Micromachines 2017, 8, 285. [CrossRef]
63. Goldman, M.; Goldman, A.; Sigmond, R.S. The corona discharge, its properties and specific uses; Pure and Applied Chemistry.
Pure Appl. Chem. 1985, 57, 1353–1362. [CrossRef]
64. Hillborg, H.; Karlsson, S.; Gedde, U.W. Characterisation of low molar mass siloxanes extracted from crosslinked polydimethyl-
siloxanes exposed to corona discharges. Polymer 2001, 42, 8883–8889. [CrossRef]
65. Foerch, R.; Izawa, J.; Spears, G. A comparative study of the effects of remote nitrogen plasma, remote oxygen plasma, and corona
discharge treatments ont he surface properties of polyethylene. J. Adhes. Sci. Technol. 1991, 5, 549–564. [CrossRef]
66. Grace, J.M.; Gerenser, L.J. Plasma Treatment of Polymers. J. Dispers. Sci. Technol. 2003, 24, 305–341. [CrossRef]
67. Vesel, A.; Junkar, I.; Cvelbar, U.; Kovac, J.; Mozetic, M. Surface modification of polyester by oxygen- and nitrogen-plasma
treatment. Surf. Interface Anal. 2008, 40, 1444–1453. [CrossRef]
68. Najafi, K.; Harpster, T.J.; Kim, H.; Mitchell, J.S.; Welch, W.C. 1.09–Wafer Bonding. In Comprehensive Microsystems; Elsevier:
Amsterdam, The Netherlands, 2008; pp. 235–270.
69. Hillborg, H.; Gedde, U.W. Hydrophobicity changes in silicone rubbers. IEEE Trans. Dielect. Elect. Insul. 1999, 6, 703–717.
[CrossRef]
70. Garbassi, F.; Morra, M.; Barino, L.; Occhiello, E. Polymer Surfaces. In From Physics to Technology; Wiley: New York, NY, USA, 1994.
71. Waters, L.J.; Finch, C.V.; Bhuiyan, A.K.M.H.; Hemming, K.; Mitchell, J.C. Effect of plasma surface treatment of poly(dimethylsiloxane)
on the permeation of ppharmaceutical compounds. J. Pharm. Anal. 2017, 7, 338–342. [CrossRef]
72. Zou, J.; Wong, P.Y. Thermal Effects in Plasma Treatment of Patterned PDMS for Bonding Stacked Channels. MRS Online Proc. Libr.
2003, 782, A5.5. [CrossRef]
73. Hillborg, H.; Gedde, U.W. Hydrophobicity recovery of polydimethyl siloxane after exposure to corona discharges. Polymer
1998, 39, 1991–1998. [CrossRef]
74. Zhou, J.; Ellis, A.V.; Voelcker, N.H. Recent developments in PDMS surface modification for microfluidic devices. Electrophoresis
2010, 31, 2–16. [CrossRef]
75. Lopera, S.; Mansano, R.D. Plasma-Based Surface Modification of Polydimethylsiloxane for PDMS-PDMS Molding. Int. Sch. Res.
Not. (ISRN) Polym. Sci. 2012, 5, 767151. [CrossRef]
76. Kim, H.T.; Jeong, O.C. PDMS surface modification using atmospheric pressure plasma. Microelectron. Eng. 2011, 88, 2281–2285.
[CrossRef]
77. Sugawara, M. Generation of a highly uniform and large area corona discharge source adaptable to surface treatment. Surf. Coat.
Technol. 2001, 142, 290–292. [CrossRef]
78. Adamiak, K.; Inculet, I.I.; Castle, G.S.P. The Control of Corona Current Distribution Using Shaped Electrodes. J. Electrost.
1993, 30, 381–392. [CrossRef]
79. Haubert, K.; Drier, T.; Beebe, D. PDMS bonding by means of a portable, low-cost corona system. Lab Chip 2006, 6, 1548–1549.
[CrossRef]
80. Özçam, A.E.; Efimenko, K.; Genzer, J. Effect of ultraviolet/ozone treatment on the surface and bulk properties of poly(dimethyl
siloxane) and poly(vinylmethyl siloxane) networks. Polymer 2014, 55, 3107–3119. [CrossRef]
81. Kim, J.; Chaudhury, M.K.; Owen, M.J. Hydrophobic recovery of polydimethylsiloxane elastomer exposed to partial electrical
discharge. J. Colloid. Interface Sci. 2000, 226, 231–236. [CrossRef]
82. Berdichevsky, Y.; Khandurina, J.; Guttman, A.; Lo, Y.H. UV/ozone modification of poly(dimethylsiloxane) microfluidic channels.
Sens. Actuators B Chem. 2004, 97, 402–408. [CrossRef]
83. Fu, Y.J.; Qui, H.Z.; Liao, K.S.; Lue, S.J.; Hu, C.C.; Lee, K.R.; Lai, J.Y. Effect of UV-Ozone Treatment on Poly(dimethylsiloxane)
Membranes: Surface Characterization and Gas Separation Performance. Langmuir 2009, 26, 4392–4399. [CrossRef] [PubMed]
84. Efimenko, K.; Wallace, W.E.; Genzer, J. Surface Modification of Sylgard-184 Poly(dimethylsiloxane) Networks by Ultraviolet and
Ultraviolet/Ozone Treatment; J. Colloid. Interface Sci. 2002, 254, 306–315. [CrossRef]
85. Schnyder, B.; Lippert, T.; Kötz, R.; Wokaun, A.; Graubner, V.M.; Nuyken, O. UV-irradiation induced modification of PDMS films
investigated by XPS and spectroscopic ellipsometry. Surf. Sci. 2003, 532, 1067–1071. [CrossRef]
86. Li, C.H.; Wilkes, G.L. The mechanism for 3-aminopropyltriethoxysilane to strengthen the interface of polycarbonate substrates
with organic-inorganic sol-gel coatings. J. Inorg. Organomet. Polym. 1998, 8, 33–45. [CrossRef]
87. Vlachopoulou, M.E.; Tserepi, A.; Pavli, P.; Argitis, P.; Sanopoulou, M.; Misiakos, K. A low temperature surface modification
assisted method for bonding plastic substrates. J. Micromech. Microeng. 2009, 19, 015007. [CrossRef]
88. Sivakumar, R.; Lee, N.Y. Microfluidic device fabrication mediated by surface chemical bonding. Analyst 2020, 145, 4096–4110.
[CrossRef] [PubMed]
89. Lee, K.S.; Ram, R.J. Plastic-PDMS bonding for high pressure hydrolytically stable active microfluidics. Lab Chip 2009, 9, 1618–1624.
[CrossRef] [PubMed]
90. Casanova-Moreno, J.; To, J.; Yang, C.W.T.; Turner, R.F.B.; Bizzotto, D.; Cheung, K.C. Fabricating devices with improved adhesion
between PDMS and gold-patterned glass. Sens. Actuators B Chem. 2017, 246, 904–909. [CrossRef]
Biosensors 2021, 11, 292 28 of 28

91. Shin, Y.S.; Cho, K.; Lim, S.H.; Chung, S.; Park, S.J.; Chung, C.; Han, D.C.; Chang, J.K. PDMS-based micro PCR chip with Parylene
coating; J. Micromech. Microeng. 2003, 13, 768–774. [CrossRef]
92. Henry, A.C.; Tutt, T.J.; Galloway, M.; Davidson, Y.Y.; McWhorter, C.S.; Soper, S.A.; McCarley, R.L. Surface Modification of
Poly(methyl methacrylate) Used in the Fabrication of Microanalytical Devices. Anal. Chem. 2000, 72, 5331–5337. [CrossRef]
[PubMed]
93. Duffy, D.C.; McDonald, J.C.; Schueller, O.J.A.; Whitesides, G.M. Rapid protyping of microfluidic systems in poly (dimethylsilox-
ane). Anal. Chem. 1998, 70, 4974–4984. [CrossRef]
94. Liu, Y.; Feng, X.; Shen, J.; Zhu, J.J.; Hou, W. Fabrication of a Novel Glucose Biosensor Based on a Highly Electroactive
Polystyrene/Polyaniline/Au Nanocomposite. J. Phys. Chem. B 2008, 112, 9237–9242. [CrossRef]
95. Qian, W.P.; Yao, D.; Yu, F.; Xu, B.; Zhou, R.; Bao, X.; Lu, Z. Immobilization of Antibodies on Ultraflat Polystyrene Surfaces. Clin.
Chem. 2000, 46, 1456–1463. [CrossRef]
96. Fridman, A.A. Plasma Chemistry, 1st ed.; Cambridge University Press: New York, NY, USA, 2008.
97. Kaltenbrunner, M.; Sekitani, T.; Reeder, J.; Yokota, T.; Kuribara, K.; Tokuhara, T.; Drack, M.; Schwödiauer, R.; Graz, I.; Bauer-
Gogonea, S.; et al. An ultra-lightweight design for imperceptible plastic electronics. Nature 2013, 499, 458. [CrossRef]
98. Weber, J.; Potje-Kamloth, K.; Haase, F.; Detemple, P.; Volklein, F.; Doll, T. Coin-size coiled-up polymer foil thermoelectric power
generator for wearable electronics. Sens. Actuators A Phys. 2006, 132, 325–330. [CrossRef]
99. Convery, N.; Gadegaard, N. 30 years of microfluidics. Micro Nano Eng. 2019, 2, 76–91. [CrossRef]