Current Directions in Biomedical Engineering 2016; 2(1): 329–332
Open Access
Matthias B. Schuh*, Michael Kirsch, David Dillmann, Franziska Breuer and Markus Eblenkamp
Climate retainment in carbon dioxide incubators
DOI 10.1515/cdbme-2016-0073
Abstract: Ensuring optimal climate conditions during cul-
tivation is essential for successful and reproducible cell
culture based investigation. Being the gold standard, car-
bon dioxide incubators fulfill this demand in various ge-
ometries and sizes to suit diverse cultivation applications.
A door opening results in a climate breakdown followed
by a restoration period during which optimal conditions
cannot be guaranteed. The following paper investigates
the influence of incubator door design to optimize climate
retainment during the above mentioned event.
Keywords: cell culture; CFD; DIN 12880; flow visualiza-
tion; incubator; light section; UDF.
1 Introduction
Cell culture assays have become standardized methods
in the toolbox also of medical engineers e.g. to test new
materials for cytotoxicity as an alternative for animal test-
ing [1]. Therefore, a broad range of established cell lines
are available, which are comfortable to handle. Neverthe-
less, biological systems like cell cultures are sensitive to
temperature and CO2 fluctuation. Thus the employment of
a suitable incubator is important [1], because the probe
has to reside for a long time in an optimal thermal and
gaseous environment e.g. 37°C, 10 Vol-% CO2 , >90% rh.
Though it is not required to ensure the absolute niveau
of the parameters rather than keeping them as constant
as possible within the limits [1]. The inspection of atmo-
sphere constancy of an incubator is regulated in the DIN
12880 [2].
During access to the incubator due to door opening the
climate is exchanged with the surrounding atmosphere.
Based on our experience a duration of more than 30 s the
*Corresponding author: Matthias B. Schuh, Institute of Medical
and Polymer Engineering, Boltzmannstraße 15 85748 Garching,
E-mail: matthias.schuh@tum.de
Michael Kirsch, David Dillmann, Franziska Breuer and Markus
Eblenkamp: Institute of Medical and Polymer Engineering, Boltz-
mannstraße 15 85748 Garching, E-mail: michael.kirsch@tum.de
(M. Kirsch); Dillmanndavid@web.de (D. Dillmann);
franziska_breuer@web.de (F. Breuer); markus.eblenkamp@tum.de
(M. Eblenkamp)
© 2016 Matthias B. Schuh et al., licensee De Gruyter.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 License.
distinct gaseous components have emanated to the exter-
nal environment. Furthermore the inside of the incubator
is cooling down. Improvement of the door opening kine-
matic may hold potential to reduce said exchange and
also induced turbulence i.e. air movement during door
opening. This may lead to lowering impact on the samples,
saving valuable and costly resources and lower the risk
to introduce potential harmful particles as carrier for bac-
teria [3] into the incubator’s atmosphere. Investigations
via experimental light section flow visualization [4, 5] and
CFD simulation have been conducted to compare an ad-
justed door opening concept to the conventional incubator
design.
2 Materials and methods
The typical carbon dioxide incubator relies on a two-fold
swing door concept consisting of an outer temperature
regulated metal door and an inner glass door. In contrast
to the standard design a sliding door concept, as shown
in Figure 1, is suggested to replace the inner door. It is
expected to reduce turbulence and induced suction to the
inner gas volume when being operated. The new door con-
cept is compared to the state of the art by CFD-simulation
and an experiment of visualization of fluid flow via light
section. The simulation is performed with ANSYS Fluent® .
The geometry of an incubator prototype is used to define
the inner volume with three perforated shelves (modelled
as pressure jump boundary conditions) in place. Both door
concepts are modelled to comply with the real dimensions.
The sliding door model is derived from the working proto-
type, used for the light section investigation. Its opening
area is slightly reduced compared to the swing door. The
movement of both doors is realized by using dynamic
meshes with user defined functions (UDF) applied to the
moving boundaries. The k-ε turbulence model is applied.
The setup of the light section experiment is shown in
Figure 2. A mock-up of an incubator is used enclosing
an inner volume of approx. 180 l which is temperature
regulated (37°C) and filled with effect fog for every run
of the experiment. The aperture of the mock-up allows
mounting of either a swing door or the prototype of the
sliding door concept. The inner volume is also equipped
with three perforated shelves. During door opening the
Unauthenticated
Download Date | 3/23/17 8:56 PM
330 | M.B. Schuh et al.: Climate retainment in carbon dioxide incubators

I II

Figure 1: Sliding door concept for a carbon dioxide incubator.

The door is horizontally split into an upper and lower section (I)
which are initially connected. The linked sections can be moved
up and down to the preferred the access access postion (A). If
the position of interest has been reached the link between both
sections may be severed and the access zone may be extended to
a height that is comfortable to retrieve a sample from the inside of
the incubator (II).

(3) (2)

(7) (1) (7)

(5)
(6) Figure 3: Comparison of resulting air flow induced by door opening
(4) in the vertical symmetrical plane of the incubator.
(4)
(5)

are taken with a camera system from door opening until

(6) after 30 s. Characteristic dimensions for simulation and
Top view Cross-section view experiment are listed in Table 1.

Figure 2: Sketch of the light section experiment in top and cross-

section view: (1) fog machine, (2) fog homogenizer, (3) fog injection
to the mock-up, (4) aperture with door system, (5) laser plane, 3 Results
(6) laser source with line generator (7) incubator mock-up.
3.1 Simulation results
Table 1: Boundary conditions of experimental and simulated setup.
As a result of the post-processing of the simulated data
Simulation Figure 3 shows the magnitude of the induced velocity
in the vertical symmetrical plane of the incubator in the
Inner volume of incubator (w × h × d) 590 mm × 590 mm × 538 mm
Inner atmosphere 37°C, 10 Vol-% CO2 , 95 % rh
relevant area during door opening with the respective
Outer atmosphere 20°C, room air opening kinematic. Both opening processes are simulated
Aperture of swing door (w × h) 590 mm × 590 mm to be finished after 1 s. Especially at the beginning of the
Aperture of sliding door (w × h) 480 mm × 480 mm swing door opening the induced velocity in the opening
Light section experiment gap between the door and the incubator is very high due
Inner volume of incubator (w × h × d) 590 mm × 590 mm × 538 mm to the suction of the opening door. The velocity magni-
Inner atmosphere ∼37°C, effect fog tude remains high throughout the opening process as a
room air mixture
result of the fluid displacement, thus there are also ed-
Outer atmosphere 20°C, room air
Aperture of swing door (w × h) 590 mm × 590 mm
dies disturbing the atmosphere inside the incubator. In
Aperture of sliding door (w × h) 480 mm × 480 mm contrast, the sliding door concept induces only little air
movement during door opening in the area surrounding
the moving part of the door, leading to much smaller
fog emanates and is visualized by a laser plane projected eddies. Figure 4 visualizes the average and extremal values
perpendicular to the aperture plane. Light section images of carbon dioxide concentration in the incubator over the

Unauthenticated
Download Date | 3/23/17 8:56 PM
 M.B. Schuh et al.: Climate retainment in carbon dioxide incubators | 331

10

8
ψCO2 (%)

2 Swing
Sliding
0
0 2 4 6 8 10 12 14
Time t (s)

Figure 4: Plot of CO2 volume fraction over a period of 15 s for the

swing and the sliding door system:
dotted: maximum inside the volume
drawn through: mean inside the volume
dashed: minimum inside the volume

time span between the closed door and 15 s after the be-
ginning of opening, determined at the coordinates within
the incubator defined in DIN 12880 [2]. The rapid decline
of the minimum curve of the swinging door during the
first second is due to the described high velocities close to
the door, resulting in a fast mixing between the inner and
surrounding atmosphere, whereas in the sliding concept
the suction is much smaller, resulting in much less mixing.
Exchange of atmosphere is then achieved only by natu-
ral convection and diffusion leading to a much smaller
amount of climate loss during door opening. After door
opening these mechanisms result in a steady decline of the
before established atmosphere. The atmosphere using the
sliding approach is declining slower.

3.2 Light section results

Figure 5: Comparison of swing and sliding concept over the duration
The resulting movement of fluid from the start (t = 0 s) (1s) of active door movement by light section photography.
until finish (t = 1 s) of the door opening process is com-
pared for the swing and the sliding door in Figure 5 . The
inner atmosphere (visualized by fog) seems to adhere to
extent than for the sliding door setup. At t = 20 s the
the moving swing door (t = 0.5 s, opening angle approx.
inner atmosphere is nearly depleted whereas emanating
45°). Furthermore two major eddies can be recognized on
atmosphere from the mock-up with the sliding door is
the upper and the lower edge. After door movement has
still clearly visible. Just before closing the door after 30 s
completed (t = 1 s, opening angle 90°) two air drafts on
the atmosphere has been exchanged completely with the
the upper and the lower level can be seen that reach out
environment for both concepts.
by the length of the swing door.
The upward movement of the sliding door generates
small delicate eddies. After completion of opening proce-
dure the primary movement direction of the atmosphere 4 Discussion and conclusion
is upwards due to natural convection which remains the
dominating effect as shown in Figure 6. After 10 s the The comparison of the suggested sliding with a standard
atmosphere of the swing door has emanated to a higher swing door concept shows that there is a high influence

Unauthenticated
Download Date | 3/23/17 8:56 PM
332 | M.B. Schuh et al.: Climate retainment in carbon dioxide incubators
Figure 6: Comparison of resulting atmosphere exchange after active
door movement is completed.
during the opening process on fluid movement of door
kinematic which is obvious in the simulation as well as
in the experiment. The suction of the swing movement
induces higher fluid velocity and bigger eddies, which
may favor the introduction of particles from the outside
and thus increase the risk for contamination. Furthermore
the atmosphere exchange or depletion occurs faster for
the swing than for the sliding door though this outcome
might be biased due to the reduced aperture area. Re-
garding the performance of both concepts over the stan-
dardised duration of 30 s according to DIN 12880 [2] one
can state that the atmosphere depletes completely for
both concepts. An improvement of retainment might be
possible by using the option to individually define the
aperture position and area by adjusting the sliding door
to the position and aperture size required for access to
the inside (cf. Figure 1). A clear advantage is the reduced
turbulence of air induced by the sliding mechanism, since
this is favorable for clean working conditions as well as
clean room environments.
Author’s Statement
Conflict of interest: Authors state no conflict of interest.
Material and methods: Informed consent: Informed con-
sent has been obtained from all individuals included in
this study. Ethical approval: The conducted research is not
related to either human or animal use.
References
[1] Gstaunthaler G, Lindl T. Zell- und Gewebekultur. 7th ed.
Springer Spektrum; 2013.
[2] Deutsches Institut für Normung eV. Elektrische Laborgeräte und
Brutschränke DIN 12880. Beuth Verlag Berlin; 2007.
[3] Keller M. Biokontaminationskontrolle. In: Gail L, Gommel
U, Hortig HP, editors. Reinraumtechnik. Springer; 2011.
p. 461–82.
[4] Donges A, Noll R. Springer Series in Optical Sciences: Laser
Measurement Technology. Vol. 188. Springe; 2015.
[5] Eckelmann H. Einführung in die Strömungsmeßtechnik.
Wiesbaden: Springer Fachmedien Wiesbaden GmbH; 1997.
Unauthenticated
Download Date | 3/23/17 8:56 PM