A compact integrated microfluidic oxygenator with high gas exchange efficiency and compatibility for long-lasting endothelialization.

We have developed and tested a novel microfluidic device for blood oxygenation, which exhibits a large surface area of gas exchange and can support long-term sustainable endothelialization of blood microcapillaries, enhancing its hemocompatibility for clinical applications. The architecture of the parallel stacking of the trilayers is based on a central injection for blood and a lateral injection/output for gas which allows significant reduction in shear stress, promoting sustainable endothelialization since cells can be maintained viable for up to 2 weeks after initial seeding in the blood microchannel network. The circular design of curved blood capillaries allows covering a maximal surface area at 4 inch wafer scale, producing high oxygen uptake and carbon dioxide release in each single unit. Since the conventional bonding process based on oxygen plasma cannot be used for surface areas larger than several cm2, a new "wet bonding" process based on soft microprinting has been developed and patented. Using this new protocol, each 4 inch trilayer unit can be sealed without a collapsed membrane even at reduced 15 μm thickness and can support a high blood flow rate. The height of the blood channels has been optimized to reduce pressure drop and enhance gas exchange at a high volumetric blood flow rate up to 15 ml min-1. The simplicity of connecting different units in the stacked architecture is demonstrated for 3- or 5-unit stacked devices that exhibit remarkable performance with low primary volume, high oxygen uptake and carbon dioxide release and high flow rate of up to 80 ml min-1.

solubility and diffusivity of O 2 averaged over the range of O 2 partial pressures (PO 2 ) in blood between the inlet and the outlet of the capillaries.The equations and settings presented below are explained in detail in the article of Potkay 2013 [1].Please do not adjust margins Please do not adjust margins where is the diffusivity of oxygen in blood plasma, is the solubility of oxygen in blood plasma, = 1,39 Hct/3 is the oxygen binding capacity of hemoglobin (ml O 2 /ml blood), Hct is the hematocrit, is the slope of the oxygen-hemoglobin dissociation curve (dSatO 2 /dPO 2 ) for a given partial oxygen pressure (PO 2 ).
Used settings: Please do not adjust margins Please do not adjust margins Where is the diffusivity of carbon dioxide in blood plasma, is the solubility of oxygen in blood plasma, is the slope of the CO 2 volume percentage curve (dC B /dPCO 2 ) for a given CO 2 partial pressure.
Used settings:

SI.1C. Calculating the O 2 uptake and CO 2 release
The O 2 uptake for different flow rates was calculated by summing the amount of dissolved oxygen in blood and the amount of oxygen molecules bound to hemoglobin, based on the equation [5,6]: where is the O 2 uptake (ml O 2 /min); is the solubility coefficient of O 2 in plasma (3.14 10 haemoglobin in blood (g Hb/ ml blood): for swine, is estimated to be 0,10 g/ml [8,9]; is the oxygen saturation variation between the inlet and the outlet of the device (unitless).
The CO 2 release for different flow rates was calculated by summing the amount of carbon dioxide in plasma (dissolved and in the form of bicarbonate) and in erythrocytic fluid, based on the equation [6,7]: = 7,19 + 0,77 ( -7,40) + 0,035 (1 - 2 ) = 6,125 -log ( 1 + 10   -7,84 -0,06  ΔPO 2 is much smaller with the mixture of air/oxygen than with pure oxygen for all blood flow rates.At the maximum blood flow rate of 15 ml/min, ΔPO 2 is about 5.6 mmHg with the mixture of air/oxygen, while it reaches 54.5 mmHg with pure oxygen.ΔPCO 2 takes the same value for both gas oxygenation at very low flow rates (from 0.5 ml/min to 1.5 ml/min) and high flow rates (from 10 and 15 ml/min).However, in the range of 2 -8 ml/min, one can observe different ΔPCO 2 values.This is surprising since the percentage of oxygen in the oxygenation gas should not change the carbon dioxide partial pressure.Indeed, the concentration of carbon dioxide is negligible (0.04%) in air.It would be interesting to repeat this experiment on several devices to quantify measurement errors.
As expected, the oxygen saturation at the outlet is lower with the mixture of air/oxygen than with pure oxygen at flow rates higher than 1.5 ml/min.Above 5 ml/min, SatO 2 with the mixture of air/oxygen is less than 95% despite a very high inlet value of 91%.For comparison, SatO 2 with pure oxygen is 100% for all blood flow rates up to 15 mL/min.
We can conclude that the oxygen percentage of the gas oxygenation is one of the most important parameters for oxygenation capacity.For this reason, we decided to work with pure oxygen in order to increase the rate of oxygen transfer to the extent possible.

SI.3 -Predictive theoretical calculations
In order to identify the influence of membrane thickness and exchange surface area on O 2 uptake and CO 2 release as function of blood flow rate, theoretical calculations of O 2 uptake and CO 2 release (see part.2.1 in the article and SI.1 in supplementary information) have been done based on the mathematical model of Potkay [4].

Please do not adjust margins
Please do not adjust margins

SI.3A. Simulations: role of the membrane thickness (15µm < e m < 105µm)
Theoretical O 2 uptake and CO 2 release are presented below as a function of blood flow rate Q (1; 5; 10; and 15 ml/min) for different membrane thickness e m (15, 45, 75 and 105 µm).As expected, the higher the blood flow rate and the thinner the membrane thickness, the better are the O 2 uptake and the CO 2 release.
As observed in the resistance equation [4]  the membrane thickness allows reducing the resistance to diffusion.Concerning O 2 uptake, we observe that the effect of the membrane thickness is noticeable only at high blood flow rate in the range from 10 to 15 ml/min.Whereas at low flow rate (Q<5ml/min), blood has enough time to be almost fully oxygenated (Sat(O 2 )~ 100%) even for the thickest membrane of 105 µm, at high blood flow rate (Q>10ml/min), the retention time is sufficiently reduced to make the effect of thin membrane thickness observable.
Concerning CO 2 release, the effect of membrane thickness is more pronounced at every blood flow rate.Reducing the thickness of the membrane is therefore important for the O 2 uptake only at high blood flow rate and for the CO 2 release regardless every blood flow rate.

SI.3B. Simulations: role of the exchange surface area (10 cm² < S A < 40 cm²)
Please do not adjust margins Please do not adjust margins Theoretical O 2 uptake and CO 2 release are presented below as a function of blood flow rate for different exchange surface area S A (10, 20, 30 and 40 cm²).As expected, the higher the blood flow rate and the larger the exchange surface area, the better are the O 2 uptake and the CO 2 release.Concerning O 2 uptake, similarly as the previous simulation, the effect of exchange surface area is more pronounced at high blood flow rate and particularly between 10 and 20 cm².Concerning CO 2 release, the effect of the exchange surface area is significant at every blood flow rate.
To conclude, from these simulations (SI.3A and SI.3B), the exchange surface area has a larger impact on gas transfer efficiency than the membrane thickness, especially for the CO 2 release.Therefore, it appears very important to enhance as much as possible the exchange surface area.This can be done easily without altering the other properties (such as pressure drop) by realizing a device with a double-sided diffusion, with two integrated membranes at the bottom and the top of the blood capillaries similarly as in the work of Rieper [10].The device can also be expanded laterally but in this case the capillaries will be longer, so the pressure drop in the whole system should be reconsidered.
Please do not adjust margins Please do not adjust margins O 2 uptake and CO 2 release per unit layer as a function of blood flow rate per unit layer is shown here.These graphs allow to compare the oxygenation and decarbonation efficiency per unit layer for one single or more units stacked in parallel.
Concerning the first graph, O 2 uptake per unit layer is a bit enhanced for 3-and 5-stacked units at high blood flow rate in the range 10 to 15 ml/min.For the second graph, CO 2 release per unit layer is also enhanced for 3-and 5-stacked unit devices in the range 5 to 15 ml/min.
Concerning CO 2 release, one main factor is involved: carbon dioxide partial pressure at the inlet of the device (PCO 2,in ) was too low at the beginning of the first experiment (single unit with PCO 2,in = 54 mmHg) compared to the second experiment (stacked units with PCO 2,in = 63,2 mmHg).With higher PCO 2,in the second experiment allows a better determination of the maximum CO 2 transfer capability of the device.
Moreover, O 2 uptake and CO 2 release increase slightly with the number of stacked layers.This enhancement can be explained by the proximity of the adjacent parallel blood channels that can be oxygenated by the sandwiched gas channels.The results on CO 2 release is more significant as the CO 2 release is more dependent on the exchange surface area (see SI.3.B).These results confirm that realizing a double-membrane stacked device could be a good way to further increase O 2 and CO 2 efficiency per unit layer as the exchange surface area will be twice as large.

Fig SI. 2 .
Fig SI.2.In vitro performance of single-layer devices (with H = 85 µm) compared for the two gases studied: air or pure oxygen: (a) variation of the oxygen saturation level SatO 2 measured in the blood collected at the output (dotted lines give the input values), (c) variation of the oxygen partial pressure PO 2 and (b) carbon dioxide partial pressure CO 2 .

SI. 4 .SI. 5 .SI. 6 .
In vitro experiments: pictures and videos recorded for stacked devices during blood filling For videos: 3-stacked unit device: file "video1.mpeg"and 5-stacked unit device: file "video2.mpeg"In vitro experiments: ΔPO 2 and ΔPCO 2 between the outlet and the inlet of the device for 3 and 5 stacked units layers 3-units stacked device (25ml/min) 5-units stacked device (80m This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins In vitro experiments: comparison of experimental O 2 uptake and CO 2 release per unit layer as a function of blood flow rate for one single unit layer, 3-and 5-stacked unit devices , 2 specific PO 2 , and is the effective diffusivity of O 2 in blood at a specific PO 2 .
2 , -  2 ,where is the expected oxygen partial pressure at the inlet of the blood capillary, is the expected  2 ,   2 ,oxygen partial pressure at the outlet of the blood capillary, is the effective solubility of O 2 in blood at a

Calculating the effective solubility and diffusivity of CO 2 in blood (model of Potkay 2013 [1])
[4]m²/s An equation relating the volume percentage of CO 2 in blood to the CO 2 partial pressure (PCO 2 ) in blood was described by Mochizuki et al.[4], as given below: -5 ml