DE102017131085A1 - Plasma polymer solid, in particular plasma polymer layer with hydrocarbon network formation, their use and process for their preparation - Google Patents

Abstract

The invention relates to a plasma polymer solid, in particular a plasma polymer layer, wherein the lower limit of the modulus of elasticity of the solid or the layer is determined by a specific function, which applies to certain areas of the modulus and determined by means of certain C / O molar ratios XPS. The invention further relates to the use of a corresponding plasma polymer solid or a plasma polymer layer and a process for their preparation.

Description

The invention relates to a plasma polymer solid, in particular a plasma polymer layer, wherein the lower limit of the modulus of elasticity of the solid or the layer is determined by a specific function, which applies to certain areas of the modulus and determined by means of certain C / O molar ratios XPS. The invention further relates to the use of a corresponding plasma polymer solid or a plasma polymer layer and a process for their preparation.

For the production of both easy-to-clean coatings, ie in the cleaning behavior of improved coatings, as well as dry chemical deposited release layers, prepared by plasma polymerization or PE-CVD is despite all efforts and developments as before the need, the mechanical resistance of such coatings and in order to increase their usability without losing the repellent character (a low surface energy (OFE) with the least possible polar character) of the layer.

The DE 102013219331 shows that it is possible to find plasma-polymer layers that are defined by their C / O ratio and have a high modulus of elasticity. There are only a few hints about the developing network. All indications point to the known Si-O network of organosilicon compounds. There are no indications about the network density or the physical density of the coatings.

Plasma polymers Thin films can be used not only as a dry release layer on forming tools, but also for the equipment of utility articles or machine parts, to facilitate their cleaning or. Of particular interest are such properties for sensor products, sanitary ware products, food and pharmaceutical processing articles, as well as e.g. for shaping tools.

DE 100 34 737 discloses a process for producing a permanent mold release layer by plasma polymerization. The coating is characterized by its gradient structure.

DE 102006018491 A1 discloses a flexible plasma polymer product with PDMS-like layer construction and thus a soft and relatively sensitive product. Here a concrete layer composition is claimed. Information on the determination of the hydrogen content is not made, as well as no information on the crosslinking situation or density determination are included. However, there are general information on density and hydrogen content. For the density, a range of 0.9 to 1.15 g / cm ^{3 is} given. From this the expert concludes on a rather weakly networked layer system. The H / C ratio ranges from 2.25 to 3 to 1. In addition, information on the shift of the Si 2p peak in the XPS spectrum can be found. This value provides information about the close bond conditions of the silicon. Strong shift from PDMS is an indication of a high number of Si-O bonds. The mechanical properties are described as elastic and crack-free extensible up to 50%. This is another indication that the coating is soft and could be referred to as a plasma polymer PDMS. For the production of the layer, it is pointed out that the total external leak rate should be less than 1% of the amount of oxygen supplied to the process, and care must be taken to keep the internal leak small due to residual water. A value for the total leak rate is not found. The power required for the process is about 7.25 W / sccm. It is kept small in order to be able to achieve the mechanical properties of the layer and to ensure that the working regime moves in the vicinity of the precursor excess.

WO2015044247 A1 discloses a further development in terms of mechanical properties. It describes a plasma polymer solid, which can be used primarily as a release layer in a mold. The plasma-polymer solid is characterized by a high modulus of elasticity, the modulus of elasticity being dependent on the C / O ratio. The surface energy and, as a consequence, the polar fraction thereof are low, but also clearly dependent on the modulus of elasticity.

Also in this application, the starting point is that the hardness of plasma polymer layers is influenced by the ratio of oxygen to organosilicon precursors in the plasma process and, as a result, the ratio of oxygen to carbon in the coating. Accordingly, the teaching of this document is that it is possible, with a ratio of organosilicon precursor to oxygen of 2 to 1 by the variation of the power to increase the modulus of elasticity or the hardness without taking too much influence on the surface energy. One recognizes this circumstance by the fact that the Maximum position of the Si2p peak with increasing modulus of elasticity shifts to higher energy levels. This is a clear indication of a more pronounced Si-O network.

For the production of the coating, it is stated that it is preferable to use plasma systems with RF coupling designed in such a way that the self-bias remains close to zero. The total leak rate is given as less than 0.3 mbar l / s. This corresponds to a limit of 17.75 sccm and is thus about 50% of the amount of oxygen supplied in the example. The power required for the process can be deduced from the examples with values from 10.5 W / sccm.

It is known from the prior art that the crosslinking network of organosilicon PE-CVD layers is a Si-O network which can be continuously modified starting from a PDMS-like layer so that an amorphous SiO ₂ film is formed at the end. Thus, more and more oxygen is incorporated into the coating and, in turn, hydrocarbons are removed. Starting from silicon atoms, which have two oxygen atoms in the immediate vicinity (secondary Si, PDMS) over an increasing number of tertiary and quaternary Si compound (3 or 4 oxygen atoms in the immediate vicinity) towards coatings with purely quaternary Si compounds ( SiO ₂ ).

A review can be found in the document The Hererogenie Nature of Deposits from Hexamethyldisiloxanes / Oxygen Plasmas "; Alexander, MR, Short, RD, Jones, FR, Stollenwerk, M., Mathar, G., Zabold. The dissertation "Separation and Characterization of Fluorocarbon- and Siloxane-Based Plasma Polymer Layers"; B. Jacoby, Dissertation University Mainz, 2007 In principle, it points in the same direction: It shows again how the layer composition can be changed by the increased addition of oxygen to quartz glass-like coatings. The increased incorporation of oxygen leads to an increase in the network density and, consequently, to an increase in the hardness and the modulus of elasticity, but also in the surface energy and its polar fraction. The C / O ratio drops significantly with increasing modulus of elasticity.

The increased incorporation of oxygen into the network is achieved either by an increased supply of oxygen (in particular with a constant total amount of gas) and / or by increasing the plasma power, so that a stronger fragmentation takes place and the oxygen supply can be used by the silicon.

If the surface energy increases and with it the polar fraction, then the easy-to-clean properties or the separation capacity are successively lost because soiling or generally materials which come into contact with the surface increasingly adhere to the surface polarity.

In the literature, there are only a few indications that in addition to the Si-O network, in principle, hydrocarbon crosslinks are possible. For example, CH ₂ networks are named here. However, these considerations are still rather theoretical in nature and are derived from the fragmentation mechanism of hexamethyldisiloxane (HMDSO), since there are no concrete indications that these linkages are actually produced in relevant numbers. Consequently, there are no indications as to what changes in the chemical-physical behavior result from this, and in particular which technical measures are to be taken to promote such a networking mechanism.

Against this background, it was an object of the present invention to provide a coating which allows high E-modules with a low polar content of surface energy. In particular, an improvement over that in the WO2015044247 A1 disclosed layers can be achieved. Preferably, the modulus of elasticity should only be dependent to a small extent on the C / O ratio and achieve particularly high values.

mit
x = C/O Stoffmengenverhältnis ermittelt mittels XPS
E = E-Modul [GPa]
für E = 1,5 bis 20 GPa und
x ≥ 1,3 und ≤ 2,0.This object is achieved by a plasma polymer solid, in particular a plasma polymer layer wherein the lower limit of the modulus of elasticity of the solid is determined by the following function (1): $$\text{e}=\mathrm{2,172.85}-\mathrm{4,716.04}\cdot \text{x}+\mathrm{3,854.96}\cdot \text{x}2-\mathrm{1,405.04}\cdot \text{x}3+192.54\cdot \text{x}4$$

With
x = C / O molar ratio determined by XPS
E = modulus of elasticity [GPa]
for E = 1.5 to 20 GPa and
x ≥ 1.3 and ≤ 2.0.

Surprisingly, the inventors have found working conditions which make it possible to produce a plasma-polymer coating with a novel, second crosslinking mechanism or to specifically influence this. This novel networking is independent of the existing Si-O network. The additional crosslinking takes place via hydrocarbon bridges, primarily via CH ₂ -CH ₂ bridges. Thus, the Si total crosslinking level can be controlled independently of the Si-O network and controlled by the power coupled into the plasma during deposition. It can be achieved a total degree of crosslinking of over 75%, as it is otherwise only achieved by a polar reference material such as polymethylsilsesquioxane. This secondary crosslinking mechanism surprisingly provides for a significant increase in modulus of elasticity without the need for additional oxygen in the layered network. Due to the additional crosslinking, the density of the coating increases significantly. It can reach values greater than 1.2 g / cm ³ and depends on the degree of crosslinking and thus on the coupled-in plasma power.

Due to this novel network structure of the coating not only the E-modulus increases strongly, but the coating has a lower density than non inventive layers of the same modulus of elasticity, which essentially have only one Si-O network. Accordingly, in the case of the same Si-O network, the layers according to the invention have a higher density than those from the prior art, which do not have cross-linking via hydrocarbon bridges. Furthermore, the refractive index increases. Furthermore, it can be observed that the surface energy increases, but with only a very, very moderate increase in the polar fraction, since hardly any more OH groups or other polar groups are generated. Due to the increase in density, however, the disperse fraction of the surface energy increases: More molecules are available for the interaction with the test liquid.

The range of the inventive feature window is in 1 shown as non-hatched area (1). The hatched area (2) does not represent feature combinations according to the invention. The significance of the features of the article according to the invention is only indirectly apparent to the person skilled in the art, therefore further explanations should be given here: it is surprising that the novel crosslinking mechanism allows the coatings according to the invention to to adjust the modulus of elasticity almost independently of the C / O ratio. In addition, the surface energy and its polar content can be kept surprisingly small ( 2 + 3). Despite the high hardness of the layers / solids, it is possible to produce largely non-polar coatings. This is a significant advance over the prior art, especially if one additionally takes into account the position of the Si 2p peak maximum, which is now largely independent of the modulus of elasticity.

The position of the Si 2p peak (maximum) provides information about the oxygen-containing near bond conditions at the Si atom within the plasma polymer layer, which contains information about the Si-O network generated in the layer. Low levels of Si 2p pollution indicate a low-oxygen Si environment. If it increases, the Si-O network is increasingly being formed, as far as a SiO ₂ -like network. In the context of the invention, when the plasma polymerization process is conducted such that, on the one hand, the formation of the Si-O network is dependent solely on the ratio of the organosilicon precursor to oxygen and, at the same time, hydrocarbon bridges form a second independent network. The latter manifests itself in a high C / O ratio and consequently low surface energy with very low polar contents.

As already indicated, experts have so far regularly tried to produce an increased modulus of elasticity or an increased hardness by increasing the oxygen content during the layer formation. Although this had the desired consequences for the modulus of elasticity and the hardness, but the layers were simultaneously changed adversely, since the addition of oxygen at the same time hydrocarbon groups were degraded and the C / O ratio decreased. Finally, the polar fraction of surface energy increased significantly.

Surprisingly, it has now been found in the context of the invention that it is possible to produce even more low-energy, nonpolar, plasma-polymer coatings with an adjustable modulus of elasticity almost independent of the C / O ratio. For this approach, there has hitherto been no indications in the prior art and accordingly it was in fact not possible to produce the layers according to the invention in accordance with the feature window described above.

In other words, the feature profile of the present invention shows that the layer properties are by no means based solely on the material composition, but are also essentially due to the nature of the layer crosslinking (and thus by the plasma polymerization conditions). However, it has also been shown that the C / O ratio and, parallel to this, the polar fraction of the surface energy in plasma polymer layers has a significant influence on the Surface properties (hydrophilicity, hydrophobicity) have and at the same time largely independent of this, the modulus of elasticity is adjustable.

Preferred according to the invention is a plasma polymer solid according to the invention, in particular a plasma polymer layer according to the invention, the maximum of the Si 2p peak at 102.5-102.8 eV being measured on the surface of the layer by XPS.

The complete overall atomic composition of the plasma polymer coating is determined by XPS (ESCA) spectroscopy and microelement analysis. In addition, FTIR spectroscopy can be used. XPS spectroscopy must be operated at a high energy resolution to allow peak fitting of the Si 2p peak. In the process, the four energy states of silicon for different oxygen states are conventionally fitted and evaluated. The layers of the invention are characterized by a high content of secondary Si compounds. For definition and determination see measurement example 6.

Since it is possible to generate higher moduli of elasticity with the new knowledge on which the present invention is based, at a given C / O ratio, largely independently of this, than has hitherto been described in the prior art, it is also possible given E -Moduli produce particularly favorable surface energy ratios. This, too, was unpredictable from the state of the art.

Thus, the new approach according to the invention makes it possible, in addition to the control of the Si-O network, to influence the Si-C network in a targeted manner so that layers are produced which have a high modulus of elasticity for a given CO ratio without any loss of polarity Surface energy, or the surface energy in general to suffer. In this case, the relatively high modulus of elasticity is desirable to ensure a mechanical load capacity.

oder $$\mathrm{\sigma }\left(\text{p}\right)=\mathrm{0,25}$$

je nachdem welcher Wert der Funktionen (2) und (2a) der größere ist, mit
σ(p) = polarer Anteil der Oberflächenenergie [mN/m] und E = E-Modul [GPa]
für E = 1,25-20 GPa
und/oder
die Oberflächenenergie der Oberfläche hinsichtlich ihrer Obergrenze durch folgende Funktion (3) bestimmt ist: $$\mathrm{\sigma }\left(\text{p}\right)=\mathrm{0,4}\cdot \text{E}+\mathrm{22,0}$$

und die Oberflächenenergie der Oberfläche hinsichtlich ihrer Untergrenze durch folgende Funktion (4) bestimmt ist: $$\mathrm{\sigma }=\mathrm{0,17}\cdot \text{E}+\mathrm{21,5}$$

mit
σ = Oberflächenenergie [mN/m] und E = E-Modul [GPa].Accordingly, a preferred plasma polymer solid according to the invention or a preferred plasma polymer layer in the context of the invention is one such or such, wherein the maximum polar fraction of the surface energy of the surface is determined by the following functions (2) or (2a): $$\mathrm{\sigma }\left(\text{p}\right)=.0947\cdot \text{e}+0.106$$

or $$\mathrm{\sigma }\left(\text{p}\right)=0.25$$

depending on which value of the functions (2) and (2a) is the larger, with
σ (p) = polar fraction of surface energy [mN / m] and E = modulus of elasticity [GPa]
for E = 1.25-20 GPa
and or
the surface energy of the surface with respect to its upper limit is determined by the following function (3): $$\mathrm{\sigma }\left(\text{p}\right)=0.4\cdot \text{e}+22.0$$

and the surface energy of the surface with respect to its lower limit is determined by the following function (4): $$\mathrm{\sigma }=0.17\cdot \text{e}+21.5$$

With
σ = surface energy [mN / m] and E = E modulus [GPa].

The surface energy and the polar fraction of the surface energy are determined according to Measurement Example 4.

2 represents, as hatched area (1), the feature window due to the equations (3) and (4) as an E-modulus versus surface energy. The areas (2) represent non-preferred feature combinations. 3 represents as hatched area (1) that by the equations (2) and (2a) conditional feature window modulus against maximum polar portion of the surface energy.

Also preferred according to the invention is a plasma-polymer solid or a plasma-polymer layer which has a density in the range of 0.8 and 1.6 g / cm ³ , preferably between 1.0 and 1.5 g / cm ³ . The density is determined according to measurement example 5.

According to the invention, in particular for use as an easy-to-clean layer or as a release layer is a plasma polymer layer, wherein the layer has a layer thickness of 5 nm to 20 .mu.m, preferably 200 nm to 10 .mu.m and more preferably 400 nm to 5 microns. With these layer thicknesses, production-related surface roughness can be covered particularly well.

The measurement of the modulus of elasticity of the layers according to the invention takes place in case of doubt as described in measuring example 2.

According to the above, a plasma polymer solid according to the invention, in particular a plasma polymer layer according to one of the preceding claims, is preferred according to the invention, wherein the molar ratio H: C 2.0-3.0 is preferably 2.0-2.8, more preferably 2.2-2 , 7, measured by microelemental analysis.

The microelement analysis is performed in case of doubt, as described in Example 3. The preferred H: C ratio for the solid bodies or layers according to the invention makes possible a Si-C network which is preferred in the context of the invention, since valencies which are not saturated by hydrogen are available for other (network-forming) bonds.

Preferred according to the invention is a plasma polymer solid according to the invention, in particular a plasma polymer layer, the molar ratios on the surface of the layer being measured by means of XPS C: O 1,3 - 2,5 and / or O: Si 0.9-1.5, preferably 0.9-1.3 and / or C: Si 1.5 - 2.5 and / or H: C 2.0-3.0 (measured by microelemental analysis) be.

In addition to the C / O ratio, the other ratios of molar ratios described have proven to be particularly suitable. Particularly preferred is a plasma polymer solid according to the invention, in particular a plasma polymer layer according to the invention, where the following applies to the molar amounts on the surface of the layer measured by XPS: O: 20 - 38 at%, preferably 23 - 33 at% Si: 22 - 28 at%, preferably 23 - 27 at% C: 40 - 65 at% in each case based on the total number of atoms contained in the layer without H. The respective amounts of substance determined by means of XPS, in case of doubt as described in Example 3.

Preferred according to the invention is a plasma polymer solid according to the invention, in particular a plasma polymer layer according to the invention, wherein the Si H band can be recognized in the FTIR spectrum at about 2165 cm -1. This band is an indication of good crosslinking by means of additional hydrocarbon bridges.

Preferred according to the invention is a plasma polymer solid according to the invention, in particular a plasma polymer layer according to the invention, wherein the total silicon cross-linking degree is between 50 and 85%, preferably between 55 and 80%, and at the same time the silicon cross-linking via hydrocarbon bonds is between 5 and 50%, preferably between 5 and 30%, based on the total number of silicon atoms contained.

The overall degree of silicon crosslinking or the degree of silicon crosslinking via carbon bridges is determined in case of doubt as described in Example 6.

These degrees of crosslinking make it possible in a special way to achieve the positive layer properties of the plasma-polymer solids or plasma-polymer layers according to the invention.

According to the invention, a plasma polymer solid according to the invention, in particular a plasma polymer layer according to the invention, wherein the density measured by the SAW method is between 0.8 and 1.6 g / cm ³ , preferably between 1.0 and 1.5 g / cm ³ .

The density is determined as described in Example 5.

These density ratios are only possible for solids or layers according to the invention - in combination with their other properties - by controlling the hydrocarbon bridging.

• - die Gesamtleckrate ≤ 0,1 mbar l/s, bevorzugt ≤ 0,075 mbar l/s ist;
• - das Verhältnis von Gesamtleckrate zum zugeführten O₂-Fluss ≤ 0,12, bevorzugt ≤ 0,09, weiter bevorzugt ≤ 0,07 ist;
• - ein so hoher Gesamtgasfluss realisiert wird, so dass die rechnerische Gesamtaufenthaltsdauer im Plasma ≥ 10 s und ≤ 30 s (bevorzugt ≤ 20 s) beträgt;
• - ein Arbeitsdruck (mit Plasmaentladung) von ≤ 0,03 mbar, bevorzugt ≤ 0,02 mbar und ≥ 0,01 mbar verwendet wird;
• - das Flussverhältnis von siliziumorganischem Precursor zu Sauerstoff ≥ 1, bevorzugt ≥ 1,25 ist;
• - der ausgewählte siliziumorganische Precursor über ein CH₃/Si Verhältnis von ≥ 2,7, bevorzugt ≥ 3 verfügt und gleichzeitig dessen O/Si Verhältnis bezogen auf die Gesamtgasmenge ≤ 1,5, bevorzugt ≤ 1,1 beträgt;
• - der ausgewählte siliziumorganische Precursor möglichst nicht über Silazanverbindungen verfügt;
• - die Elektroden so ausgebildet sind, dass keine freiliegenden Elektrodenkanten vorhanden sind, so dass die Plasmaentladung im gesamten Raum visuell gleichmäßig und gleich stark erkennbar ist;
• - Lokale, intensivere Entladungen vermieden werden und/oder
• - insbesondere lokale Entladungen im Absaugflansch vermieden werden.
• - die Anlage großvolumig ausgelegt ist so, dass
  1. a.) die auf der Elektrode angeordnete Probe mind. 15 cm, bevorzugt 20 bis 25 cm Abstand von der nächsten Wand haben kann
  und
  • b.) der lichte Abstand der Kammerwände mind. 50 cm beträgt;
• - großflächige, ebene Elektroden eingesetzt werden (mit kapazitiver Einkopplung). Deren frei zugängliche Fläche gegenüber dem Plasma entspricht bevorzugt näherungsweise der freien Fläche der elektrischen Masse, so dass sich im Betrieb ein Self-Bias nahe Null ergibt. In diesem Zusammenhang spricht man von einer geometrisch symmetrischen Entladung.

For the preparation of the coating according to the invention, the person skilled in the art preferably prefers some or all of the procedural instructions in US Pat DE 102013219331 observe and / or observe one or more or all of the following measures, namely that

• - the total leakage rate ≤ 0.1 mbar l / s, preferably ≤ 0.075 mbar l / s;
• the ratio of total leakage rate to the supplied O ₂ flow is ≦ 0.12, preferably ≦ 0.09, more preferably ≦ 0.07;
• - such a high total gas flow is realized, so that the calculated total residence time in plasma is ≥ 10 s and ≤ 30 s (preferably ≤ 20 s);
• - A working pressure (with plasma discharge) of ≤ 0.03 mbar, preferably ≤ 0.02 mbar and ≥ 0.01 mbar is used;
• the flow ratio of organosilicon precursor to oxygen is ≥ 1, preferably ≥ 1.25;
• - The selected organosilicon precursor has a CH ₃ / Si ratio of ≥ 2.7, preferably ≥ 3 and at the same time its O / Si ratio based on the total amount of gas ≤ 1.5, preferably ≤ 1.1;
• - If possible, the selected organosilicon precursor does not have silazane compounds;
• - The electrodes are formed so that no exposed electrode edges are present, so that the plasma discharge in the entire space is visually uniform and equally strong;
• - Local, more intense discharges are avoided and / or
• - In particular, local discharges are avoided in Absaugflansch.
• - The plant is designed large volume so that
  1. a.) The arranged on the electrode sample may have at least 15 cm, preferably 20 to 25 cm distance from the next wall
  and
  • b.) the clear distance of the chamber walls is at least 50 cm;
• - large, flat electrodes are used (with capacitive coupling). Their freely accessible area relative to the plasma preferably corresponds approximately to the free area of the electrical ground, so that a self-bias close to zero results during operation. In this context one speaks of a geometrically symmetrical discharge.

In order to control the preferably required plant and electrode structure, the determination of the self-bias can first be used. It should be noted that preferably the self-bias is not changed by other measures (additional investments of DC or AC voltages). Since commercial matchboxes usually only display the amount of self-bias, it is preferable to ensure by control measurements that no self-bias voltage ≥ + 10 V is generated during the plasma discharge. Preferably, a self-bias is in the range of -25 to +10 volts, more preferably in the range of -10 to 0 volts.

In case of doubt, the determination of the bias can be carried out either directly by measurement by means of a high-voltage probe and oscilloscope directly at the electrode, whereby the safety measures are to be observed when handling RF transmitters, or by measurement by means of a multimeter within the range of the control electronics of the corresponding matchbox, provided the connections of the DC measuring unit are accessible. The signal picked up at the DC measuring unit is generally correct in sign but does not correspond in magnitude to the true self-bias value.

If the self-bias is very low, the plasma potential for layer formation in a capacitively coupled high-frequency plasma is of crucial importance. The conductivity of the plasma and its contact with the electrodes results in a voltage drop across the electrode surfaces (both at the ground electrode and at the RF electrode). In English, we speak of the sheath formation (dark space or boundary layer), cf. e.g. http://www.chm.bris.ac.uk/~paulmay/misc/msc/msc4.htm. In the case of a self-bias close to zero, the plasma potential is not only positive over time but also that electric field which accelerates the positive ions to the electrode or to the sample arranged on it. The dark room is very small locally. As a result, on the one hand, the positive ions can contribute to the formation of layers and are nevertheless only shot with low energy onto the surface. The energy distribution of the ions is kept smaller than with self-bias. The forming coating is disturbed as little as possible by the ion bombardment in this way. The absolute size of the plasma potential is dependent on various quantities, e.g. the pressure, the gas mixture, the total gas flow and the power.

In case of doubt, the geometric symmetry of the discharge can be measured by measuring the plasma potential (the plasma potential is a voltage between the plasma and the surface, for example, it can be determined with a Langmuir probe, see, for example, RSC Adv., 2013 , 3, 13540-13557.). If the plasma potential follows a pure sinusoidal signal, there is a geometrically symmetrical discharge. Deviation from the sinus indicates asymmetric discharge. The surface layers in front of the HF electrode and the ground electrode behave differently in time.

Which of the cited instructions the expert follows depends inter alia on the actual plasma system used. It is preferred to take at least three, more preferably at least five, and particularly preferably all of these measures in order to arrive at a plasma polymer solid according to the invention or a plasma polymer layer according to the invention. Preferably, the method is to be performed so that a zero bias is ensured.

Preference is given to a homogenized gas extraction, z. As described in the thesis "Design and Evaluation of gas supply system, electrodes and Gasabsaugungssystemen for low-pressure plasma chambers", Gustavo Simiema de Freitas Barbosa, University of Bremen), and a uniform Precursorgaszuführung. This aspect becomes more important the larger the plant is built.

• - ein Doppel U-Aufbau (Elektrode gegen Masse);
• - Elektroden eingelassen in eine geeignete Isolation gegen die Kammerwand von mind. 50mm. Bevorzugt wird Polyethylen (PE) verwendet. Polytetrafluourethylen (PTFE) und ähnliche fluorhaltige Isolationsmaterialien sind zu bevorzugt vermeiden, zumindest für die gegenüber dem Plasma frei liegenden Flächen und/oder
• - eine rechteckige Vakuumkammer

In addition, the following operative measures for the plasma apparatus are additionally used:

• a double U structure (electrode to ground);
• - Electrodes embedded in a suitable insulation against the chamber wall of min. 50mm. Preferably, polyethylene (PE) is used. Polytetrafluoroethylene (PTFE) and similar fluorine-containing insulating materials are preferred to avoid, at least for the areas exposed to the plasma and / or
• - a rectangular vacuum chamber

The skilled person will adjust the discharge power of the size, the electrode area and the total gas flow of the working gases.

In case of doubt, the total leak rate is determined according to the pressure rise method over a period of at least 1 hour.

Part of the invention is also the use of a layer according to the invention for improving the cleanability of an uncoated solid (easy-to-clean applications). The cleaning can be carried out with solvents, aqueous cleaners and / or dry ice, with soda rays, because due to the favorable surface properties (see above) it is possible to positively influence the cleaning behavior of surfaces with the layers according to the invention. Preferred cleaning agent is water and / or Sodastrahlen. Furthermore, it is possible to use typical cleaners, as in the field of Find food processing use. These are distinguished, for example, by their basicity, their acidity or else by the use of ozone or chlorine.

Part of the invention is also the use of a layer according to the invention or a solid according to the invention as a release layer.

The layers according to the invention are mechanically particularly resistant and can therefore preferably be used for casting or injection molding processes or other particularly abrasive plastics processing methods, e.g. in pipes or on conveyor such as e.g. Snails. In principle, the layers according to the invention are also suitable as release layers for other processes in which a separation of layers / solids is required.

1. a) Bereitstellen eines Substrates,
2. b) Abscheiden eines plasmapolymeren Festkörpers, insbesondere einer plasmapolymeren Schicht auf dem Substrat, wobei ein erfindungsgemäßer plasmapolymerer Festkörper und/oder eine erfindungsgemäße plasmapolymere Schicht entsteht.

Part of the invention is also a method according to the invention for producing a coating according to the invention. In particular, this method comprises the steps

1. a) providing a substrate,
2. b) depositing a plasma polymer solid, in particular a plasma polymer layer on the substrate, wherein a plasma polymer solid according to the invention and / or a plasma polymer layer according to the invention is formed.

Furthermore, in the method according to the invention, the person skilled in the art will take into account one, several or preferably all of the above-described instructions / steps / method configurations / plasma deposition apparatus designs.

Of particular importance is the selection of precursors, the amount of precursor gas, the maintenance of the distances and / or the use of a structure which implements the requirement of self-bias close to zero. Of course, the person skilled in the art will of course also take into account the area which additionally arises as the electrode surface through the component to be coated, the components to be coated. It is expressly pointed out that it is very advantageous to follow the construction instructions of Gustavo Simiema de Freitas Barbosa.

In summary, it can be said that the present invention provides the possibility of producing plasma-polymer solids, in particular plasma-polymer layers, which in comparison to the one disclosed in US Pat WO 2015044247 A1 have a remarkably limited C / O ratio over an exceptionally high modulus of elasticity.

Accordingly, part of the invention is a solid comprising a substrate and applied to this substrate a plasma polymer layer according to the invention. In general, but also in particular in this context, it is part of the invention that the plasma polymer layer according to the invention is also used as a corrosion protection layer, especially in application requiring a high base and alkali resistance. It can also be used in combination with layers that have particularly good barrier properties.

Finally, part of the invention is the use of such coatings having an E-modulus of about 5 to 20 GPa, preferably with moduli of elasticity in the range of 8-16 GPa as scratch and abrasion protective layers, in particular for plastics and soft metals.

Thus, the coating serves product requirements with regard to low-energy, chemically inert and mechanically stable surface properties.

The invention will be explained in more detail by way of examples.

Examples

Exemplary Embodiment 1 Production of Layers According to the Invention (with a Non-inventive Counterexample)

The plasma polymer coatings used in this example differ in their production only in the power coupled into the plasma process. Accordingly, they were prepared by means of a constant gas mixture of hexamethyldisiloxane (HMDSO) and oxygen. For HMDSO a gas flow of 92 sccm and for oxygen of 53 sccm was used. The external plasma parameters pressure (0.016 mbar), gas mixture ratio and total gas flow are kept constant for all coatings. There were five different powers from 1000W to 3400W, in the 600W distance, set. For all processes the self-bias was <-10V.

The plasma reactor used to make the layers is a large volume reactor of about 1.2m ^{3 operated} with capacitive radio frequency (13.56 MHz) excitation. This is a self - made construction of the Fraunhofer IFAM, Bremen. The special feature of the system used is that for many processes based on hexamethyldisiloxane (HMDSO) a self-bias close to zero is reliably achievable and a very homogenous plasma is formed throughout the free reactor space. The low self-bias is determined by the geometry of the plasma system, with areas of the electrodes and the grounded surfaces being approximately equal. In 4 the plasma system used is shown schematically with its most important components.

The figure shows a scheme of the plasma system used (the electrode forms a U, the remaining, not provided with the electrode wall represents for the plasma, the electrical mass, which is also correspondingly U-shaped).

1

Geerdete Reaktorkammer

3

Druckregelventil

5

Absaugung zum Pumpenstand

67

Baffleplate

9

HF-Elektrode

11

Isolierung (PE)

13

Gaszuführung

15

HF-Einspeisung

17

Match box

19

HF-Generator

In the 4 the reference symbols mean:

1

Grounded reactor chamber

3

Pressure control valve

5

Extraction to the pump stand

67

Baffleplate

9

RF electrode

11

Insulation ( PE )

13

gas supply

15

HF feed

17

Match box

19

RF generator

To regulate the process pressure, the suction is used with an adjustable butterfly valve. During the process, gas is continuously supplied via the gas feeds ( 13 ) and the controllable suction ( 5 ) ensures a constant process pressure. The power is generated by the HF generator and coupled via the Matchbox in the plasma. The Matchbox ( 17 ) serves to compensate for the unsteady ohmic resistance of the plasma. At the same time, the HF generator ( 19 ) protected against backscattered power.

The self-bias is a DC value which is between the RF feed ( 15 ) and the grounded reactor. For this purpose, via a coil of the RF component, at the feed ( 15 ), filtered. The remaining DC component is set in relation to the grounded reactor wall. The value of the self-bias largely characterizes the plasma process in addition to the process pressure and the radiated power. At a high bias value, the fast electrons have migrated to the grounded reactor surfaces. As a result, the ions are accelerated toward the electrodes and impinge on the substrate with increased energy. This increases the deposition rate, but also influences the film formation. The fact that plasma processes also have a self-bias of almost zero volts leads to a layer deposition is due, among other things, to the plasma potential. For energetic reasons, this plasma potential must be the most positive in the equilibrium plasma, electrode and grounded wall. This results in a voltage difference of the order of typically 10V. This difference causes a net drift of the ions towards the electrode.

The total leak rate was determined to be 0.065 mbar l / s. The pressure rise method was used over 1 h. The base pressure chosen was 8 * 10 ^-03 mbar.

In order to be able to demonstrate the same initial conditions in the plasma system as possible, a conditioning of the plasma system was carried out before the production of the layers. This is the process with 2200W power under the above conditions over a period of at least one hour. This conditioning ensures that all surfaces within the plasma system are already provided with a plasma polymer separation layer and that no uncontrolled contamination of the samples by preceding processes occurs.

All samples prepared lay on the base electrode. In order to obtain sufficient adhesion of the plasma polymer layer for the following analytical tests on the test substrates, a 10-minute oxygen activation (400 sccm oxygen, 1500 W, 0.02 mbar of the substrates was carried out Silicon wafers, aluminum plates or glass slides.

Only in the case of the samples prepared for the microelement analysis, activation was completely dispensed with since the plasma polymer layer must be present in powder form for the analysis without substrate. Without prior activation, the peeling of the coating from the glass slides used is greatly facilitated.

• Als nicht erfindungsgemäße Referenzschicht wurde eine Beschichtung auf der gleichen Anlage hergestellt, für welche ein HMDSO-Gasfluss von 75 sccm und Sauerstoffgasfluss von 300 sccm bei einer Leistung von 2000 W und einem Arbeitsdruck von 0,026 mbar, verwendet wurde.

Noninventive Example (Reference):

• As a reference layer not according to the invention, a coating was produced on the same plant for which an HMDSO gas flow of 75 sccm and oxygen gas flow of 300 sccm at a power of 2000 W and a working pressure of 0.026 mbar was used.

Measuring Example 2: Nanoindentation of the Layers from Example 1

To determine the modulus of elasticity of the layers of Example 1, the method of nanoindentation was used. Nanoindentation is a registered intrusion into the sample material. This is a peak in the form of a Berkovichpyramide with the force F, in 5 named Load P, pressed into the sample surface. Both the force and penetration of the tip are recorded. This results in a force-penetration curve, as in 5 you can see.

5 Figure 4 is a schematic representation of a force-penetration curve (see WC Oliver and GM Pharr, "Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinement to methodology," Journal of Materials Research, 2003).

From this measurement curve, the elastic modulus and the hardness can be determined. Important in the determination of these material constants of thin layers is not to exceed a maximum penetration depth of about 10% of the layer thickness, otherwise the substrate properties distort the measurement result (see. J. Gersten and F. Smith, The Chemistry of Materials, New York: John Wiley & Sons, Inc., 2001 ).

The evaluation of this force-penetration curve is carried out as generally customary according to the method of Oliver and Pharr, see above. According to this theory, the effective elastic modulus E _eff is Equation 1. $$e_{eff}=\frac{s}{2\beta }\sqrt{\frac{\pi }{A}}$$

Here, β is a geometry factor which is one for a Berkovichpyramide. Furthermore, S is the contact stiffness which is determined by deriving the maximum force at maximum penetration depth, see 5 , The contact surface A is also in the effective modulus of elasticity, short modulus of elasticity. From this effective modulus of elasticity, the E-modulus of the sample E can be calculated according to equation 2. $$\frac{1}{e_{eff}}=\frac{1-v^2}{e}+\frac{1-v_i^2}{e_i}$$

In this case, the variables indicated by i are based on the indenter and thus constants. The transverse contraction number v can not be measured, but is assumed to be v = 0.17 in all measurements (see Oliver and Pharr, supra, and A. Clausner, "Evaluation of the Determination of the Flow Stress in Nanoindentation," 2013 ).

Since the modulus of elasticity of the plasma polymer coatings to be investigated is within an expected value range of approximately 1 to 20 GPa, it is necessary to use samples having a minimum layer thickness of 2 μm manufacture. For itself thinner layers, a sample of the same composition is created with minimum layer thickness.

For nanoindentation measurements, a Berkovich indentor was used. The device is a "Universal Material Tester", UMT 1 Bruker. The sensor is a nanohead 2.

The measurement results for the layers from Example 1 are in 6 played. The E-modulus increases significantly with the supplied plasma power. The reference example has an E-modulus of about 13 GPa and thus lies in the size range which is claimed for the layers according to the invention.

6 1 shows the modulus of e ectivity of the plasma polymer coatings from Example 1 as a function of the power, determined by nanoindentation with Berkovich indentor.

Measuring Example 3: Microelementar and XPS analysis of the layers from Example 1

To characterize the coatings according to the invention, it is of great importance to determine the elemental composition. Since hydrogen is in principle not accessible by means of XPS analysis, supplementary microelement analysis was used.

The microelemental analysis is a combustion taking place under defined conditions. The sample powder is first dried to exclude any stored water as a source of interference. Subsequently, the combustion of the sample takes place under pure oxygen. The resulting combustion gases are passed through various reagents (e.g., caustic soda), each of which exhibits a measurable response to the presence of specific gas molecules. The measured quantity for determining the carbon content is the change in the electrical conductivity of caustic soda due to absorbed carbon dioxide. For this purpose, the sample was previously burned at 1200 ° C in an oxygen stream. To determine the hydrogen content, the sample was also burned at 1050 ° C in an oxygen stream and the combustion water formed detected by IR measurement.

The values obtained are compared with the carbon content determined by means of XPS. In this way, the hydrogen content H _{z of} the coating with the composition SiO _x C _y H _z can be determined.

As already mentioned, the coatings were scraped off glass slides to produce enough sample powder. For this purpose, a plurality of glass slides were distributed almost on the entire base electrode of the reactor. A layer thickness of about 2 μm was deposited on the glass slides.

X-ray photoelectron spectroscopy, XPS for short, is based on the external photoeffect. The process can be approximated as a three step process. First, a photon with a wavelength in the X-ray region transfers its energy to a trunk electron of the sample, whereupon it leaves its previous orbital. Then it moves through the sample to the surface. Finally, the electron exits the sample and can be detected. The photoelectrons are detected energy-selectively. The evaluation of the spectra is carried out fundamentally according to equation 3. Hv is the energy of the photons and thus the energy of the characteristic X-ray line used. $$\text{hv}={\text{e}}_{\text{kin}}+{\text{e}}_{\text{B}}{}^{\text{F}}+{\mathrm{\phi }}_{\text{sp}}$$

The kinetic energy of the electrons E _kin is measured in the photoelectron spectrometer. The work function of the spectrometer φ _Sp is almost constant and depends on the device. Thus, from the spectrum directly on the binding energy, based on the Fermi level, E _B ^{F are} closed (see HKS Brümmer, solid state analysis with eletrons, ions and X-rays, Berlin, 1980).

The binding energy is orbital and element dependent. In different chemical bonding environment, there is a shift in the binding energy, the chemical shift, the "chemical shift". The presence of neighboring atoms has an influence on the binding energy of the core electrons of an atom. Specifically, this means that, for example, oxygen atoms bonded to a silicon atom result in a shift in the binding energy of the Si atom, and the oxidation state of the silicon has changed. Thus, not only statements about the elements of the sample, but also about make their bonds. The strength of the shift depends on the different electronegativities of the binding partners and their number (cf. M. Cardona and L. Ley, "Photoemission in Solids I - General Principles," in Topics in Applied Physics Vol. 26, Springer-Verlag, 1978 ). Therefore, XPS is well suited for analyzing the chemical composition.

For all measurements a KRATOS AXIS Ultra from the company Kratos Analytical was used. Here, a monochromatized Al Kα source was used. The pass energies of the different spectra are given in the respective sections.

Table 1 shows the elemental contents of the coatings of Example 1. The values and errors given are the mean of the three measurement positions of a sample and the associated standard deviation. In addition, the position of the maximum of the Si 2p peak is indicated. It is nearly constant for the power variation. At about 103 eV, it lies between secondary and tertiary bond ratios. Only the non-inventive reference shows a significant deviation. Table 1: Elemental content from XPS measurements, averages of three measurement positions of a sample Power of the plasma process Si content Si 2p maximum charge C-content O content [W] [at-%] [EV] [at-%] [at-%] 1000 24.99 ± 0.12 102.99 ± 0.1 47.26 ± 0.11 27.75 ± 0.22 1600 25.18 ± 0.06 103.01 ± 0.1 47.01 ± 0.12 27.81 ± 0.07 2200 25.43 ± 0.14 102.99 ± 0.1 46.60 ± 0.12 27.97 ± 0.11 2800 25,44 ± 0,16 102.95 ± 0.1 46.06 ± 0.13 28.50 ± 0.05 3400 25.86 ± 0.09 102.96 ± 0.1 45.53 ± 0.17 28.61 ± 0.13 Reference (not according to the invention) 28.20 ± 0.23 104.00 ± 0.1 28.32 ± 0.30 43.48 ± 0.15

Peak charge corrected to 285eV for the aliphatic hydrocarbon

From the values, different element ratios can be derived. Normalized on silicon they are in 7 shown.

7 represents: Element ratios from XPS measurements, normalized on silicon, data in at%.

It turns out that the content of the elements in all samples is almost constant. A slight decrease in carbon content is observed with increasing power. In comparison, the reference has a drastically increased O / Si ratio. The ratio of C to Si, however, is significantly lower. The proportion of ternary and especially of quaternary silicon structures is significantly higher in the reference than in the investigated layers.

From these concentration data, a preliminary stoichiometry, initially without regard to the hydrogen content can be determined. For this purpose, the following is normalized to silicon, since this represents the structure-forming element of all considered coatings. This results in the stoichiometries in Table 2. Table 2: Stoichiometry from XPS measurements, based on silicon Power of the plasma process O: Si C: Si C: O [W] 1000 1.11 ± 0.01 1.89 ± 0.01 1.70 ± 0.02 1600 1.10 ± 0.01 1.87 ± 0.01 1.69 ± 0.01 2200 1.10 ± 0.01 1.83 ± 0.01 1.67 ± 0.01 2800 1.12 ± 0.01 1.81 ± 0.02 1.62 ± 0.01 3400 1.11 ± 0.01 1.76 ± 0.01 1.59 ± 0.01 Reference (not according to the invention) 1.54 0.01 1.00 0.02 0.65 ± 0.01

Table 2 makes it clear that the oxygen to silicon content remains unchanged under the error for all five plasma outputs. Only the carbon content decreases slightly with increasing power. In comparison, the reference not according to the invention has a significantly higher O to Si ratio.

Table 3 shows the results of the microelement analysis and the resulting ratio of hydrogen per carbon atom. Table 3: Results of microelemental analysis, means of duplicate determination Power of the plasma process C H H: C [W] [Wt%] [Wt%] 1000 27.55 ± 0.15 6.67 ± 0.01 2.88 ± 0.01 1600 27.50 ± 0.10 6,52 ± 0,02 2.82 ± 0.01 2200 27.10 ± 0.01 6.28 ± 0.01 2.76 ± 0.01 2800 26.35 ± 0.05 5.84 ± 0.01 2.64 ± 0.01 3400 26.12 ± 0.10 5.26 ± 0.01 2.40 ± 0.01 Reference (not according to the invention) 16.61 ± 0.04 4.01 ± 0.04 2.88 ± 0.01

A duplicate determination (two measurements with different sample material) was performed for each plasma polymer coating. The results show, just like the results from the XPS studies, a slight decrease in the carbon content with the plasma power.

Since microelemental analysis is a method of determining bulk properties, the results of XPS measurements can be verified. In addition, the results of the microelement analysis suggest a very homogeneous coating process because the samples were distributed throughout the reactor for analysis. Furthermore, from the combined results of XPS and microelement analysis, as indicated in Table 4, the complete stoichiometry of the layers can be determined. Table 4: determined stoichiometry of the five plasma polymer coatings, based on silicon, combined results of XPS (Si, O, C content) and microelement analysis (H content) Performance of the plasma process Si O C H [W] 1000 1.00 1.11 ± 0.01 1.89 ± 0.01 5.46 ± 0.04 1600 1.00 1.10 ± 0.01 1.87 ± 0.01 5.27 ± 0.03 2200 1.00 1.10 ± 0.01 1.83 ± 0.01 5.06 ± 0.04 2800 1.00 1.12 ± 0.01 1.81 ± 0.02 4.78 ± 0.05 3400 1.00 1.11 ± 0.01 1.76 ± 0.01 4.22 ± 0.04 Reference (not according to the invention) 1.00 1.54 ± 0.01 1.00 ± 0.02 2.89 ± 0.04

It can be stated that the plasma polymer coatings according to the invention are very similar in terms of their elemental composition. In particular, the ratio of oxygen and silicon is the same for all considered coatings. The Si-O network is not changed by the power change. This result is unexpected and surprising in view of the previously known increase in elastic modulus with increasing plasma power.

Overall, it is noticeable that the chemical composition measurements are very consistent overall. This means that the results from the XPS and from the microelementaranalysis fit together very well, even though they are completely different analytical methods. The most striking feature of this is the decreasing content of carbon with increasing power. This high quality of the analyzes can be achieved only with a sufficient homogeneity and reproducibility of the plasma process. The samples for the XPS measurements, as well as the micro elemental analysis come from different batches of the plasma process. Thus, there is a very high reproducibility from process to process. At the same time, the samples for the microelemental analysis were distributed almost on the entire base electrode. This also indicates a high homogeneity of the coating process within the chamber. This conclusion is based on comparison with the XPS results, which also show a decrease in carbon, but were measured on a single silicon sample. Thus, both the values averaged over the entire base area of the reactor and the position-dependent XPS measurement show the same tendency. This result can be achieved only by the modern and extensive process control, as well as by the construction of the plasma chamber itself. It was created in the entire, effective reactor space homogeneously burning plasma.

On closer examination, valuable information can be obtained from the H: C ratio. The ratio depending on the plasma power is in 8th shown.

Figure 8 illustrates: H: C ratio in at% from microelemental analysis

As previously described, there is a decrease in the H: C ratio with increasing plasma process performance. This means that with increasing power other structural elements must be present which produce a lower H: C ratio. For example, a linear PDMS has a H: C ratio of exactly 3. However, a CH ₂ group has a ratio of 2. The resulting crosslinking structure is discussed in Example 6.

Measuring Example 4: Surface Energy Measurement of the Layers of FIG

Easy-to-clean coatings, as well as release coatings, must have a low surface energy, in particular a low polar fraction of the surface energy. Only then are they able to minimize the interaction forces with substances on their surface and to meet the technical requirements.

The surface free energy of a solid can be calculated from the measurable three-phase contact angle. The measurement of the contact angle and its mathematical description goes back to Thomas Young. Young described the measurable contact angle Θ (Equation 4). $$\text{cos}\mathrm{\Theta }\text{=}\frac{{\mathrm{\sigma }}_s-{\mathrm{\sigma }}_{\text{sl}}}{{\mathrm{\sigma }}_{\text{l}}}$$

Here, σ _s and σ _{l are} the surface energies of the solid and the test liquid used. For the interaction σ _sl there are different approaches. This work uses the approach of Owens-Wendt-Rabel-Kaelble, see Equation 5. $${\sigma }_{sl}={\sigma }_s+{\sigma }_l-2\left(\sqrt{{\sigma }_s^D*{\sigma }_l^D}+\sqrt{{\sigma }_s^P*{\sigma }_l^P}\right)$$

und polaren Anteile ${\sigma }_s^P,{\sigma }_l^P$

${\sigma }_l^P$

beschreiben lässt. Weiterhin gilt für die Oberflächenenergien bzw. Spannungen von Festkörper und Flüssigkeit die Gleichung 6. $${\sigma }_{s,l}={\sigma }_{s,l}^D+{\sigma }_{s,l}^P$$

This approach to solve the surface free energy of a solid assumes that the interaction between the solid and the liquid phase σ _sl is above the geometric mean of the disperse fractions ${\sigma }_s^D.{\sigma }_l^D$

and polar parts ${\sigma }_s^P.{\sigma }_l^P$

${\sigma }_l^P$

can describe. Furthermore, equation 6 applies to the surface energies or voltages of solid and liquid. $${\sigma }_{s.l}={\sigma }_{s.l}^D+{\sigma }_{s.l}^P$$

aufweist (vgl. C. D. Krüger, „Experimentelle Untersuchungen zur Oberflächenveränderung metallischer Implantatmaterialien durch Plasmabehandlung,“ 2009).In order to solve the equation system resulting from Equation 5 and 6, measurements of the contact angle with at least two liquids with known polar and disperse portions are to be carried out. Care must be taken to ensure that at least one of the two liquids has a polar component ${\sigma }_l^P>0$

(see CD Kruger, "Experimental studies on the surface modification of metallic implant materials by plasma treatment," 2009).

In principle, two different measuring methods are used to determine the contact angle. On the one hand, the dynamic method, in which the droplet is expanded from a cannula during the measurement of the contact angle, progressive contact angle measurement. On the other hand, the static measurement method is used. Here, a defined drop volume is placed on the sample surface by means of a cannula, and then the contact angle is measured without the presence of the cannula. Here the static measuring method was used. For this purpose, the contact angle of 10 drops of water and diiodomethane were measured on a silicon wafer. From the mean values of the measured contact angles, the surface energy is determined according to Owens-Wendt-Rabel-Kaelble.

The measurements were carried out with the "contact angle measuring device G2" from the company "Krüss". For this purpose, drops with a volume of 2 μl were deposited on the sample surface and the contact angle was measured for 10 seconds at a measuring rate of one second. This process was performed at 10 different positions of a sample with two liquids each (water and diiodomethane). The samples used were coated 2-inch silicon wafers.

It turns out that the surface energy of the reference not according to the invention is significantly increased compared with the layers according to the invention. In particular, the polar portion has a much higher value of about 5.25 mN / m, see 9 , This can be explained by the increased oxygen content of the reference compared to the examined coatings. Since the Si-O bond is a polar bond, an increase in this structural group leads to an increased polar fraction of the surface energy.

9 represents: free surface energy of the plasma polymer separation layers as well as the reference.

Measuring Example 5: Density Measurement of the Layers of Example 1

The density of thin layers can be determined by measuring the dispersion of surface acoustic waves, in short SAW, in coated media. Rayleigh waves are generated at the sample surface with a laser. Due to the coating, a dispersion of the propagating wave occurs. The dispersion function can be fitted on a known substrate with the parameters layer thickness, density, modulus of elasticity and Poisson number. Thus, it is possible to measure non-destructively these material properties. The generation of the surface acoustic wave is based on a short-term, localized warming. The heating by the pulsed laser beam generates a thermal expansion on the sample surface and thus a surface acoustic wave. This wave contains a large frequency spectrum, which leads to a dispersion in the presence of a thin surface coating. The phase curve of the surface wave over the frequency is shown as a measuring curve. According to the theory of Farnell and Adler (cf. GW Farnell and EL Adler, "Elastic wave propagation in thin layers," in Physical Acoustics, Volume 9, 1972 ), a fit procedure is performed to determine the elastic parameters of the layer.

In a homogeneous, isotropic material, the relationship in Equation 7 applies to the phase velocity c of a surface acoustic wave. $$c=\frac{0.87+1.12\mathrm{<figure-callout\; id="1"\; label="\nu "\; filenames="DE102017131085A1\_0068.png,DE102017131085A1\_0071.png"\; state="\{\{state\}\}">\nu </figure-callout>}}{1+\nu }\sqrt{\frac{e}{2\rho \left(1+\nu \right)}}$$

Here, E is the modulus of elasticity of the homogeneous, isotropic material, v is the Poisson's number, which indicates the transverse contraction ratio, and ρ is the density. This simple relationship of the propagation velocity with the elastic parameters of the material applies only to uncoated substrates. If there is a thin surface coating, it is due to the different elastic parameters and densities to a dispersion of the propagation velocity, c becomes frequency-dependent. At higher frequencies, the penetration depth of the wave sinks into the sample material. Thus, with sufficient layer thickness and higher frequencies, the coating mainly contributes to the propagation velocity.

In order to determine the dispersion, at least two signals of the wave are recorded with different spatial distance between the laser excitation and signal receiver. From the Fourier transform of the signals, the amplitude and phase spectrum can be determined. The amplitude spectrum is used to control the signal-to-noise ratio and from the phase spectrum Φ (f) the dispersion is calculated according to Equation 8 (cf. D. Schneider, S. Frühauf, S. Schulz and T. Gessner, "The current limits of the laser-acoustic test method to characterize low-k films," Microelectronic Engineering 82 (2005) 393-398, 2005 ). $$c\left(f\right)=\frac{2\pi f*\left(x_2-x_1\right)}{\left[{\mathrm{\Phi }}_2\left(f\right)-{\mathrm{\Phi }}_1\left(f\right)\right]}$$

Here, f is the frequency of the wave and the parameter x describes the distance between the excitation line of the laser and the detector. The indices designate two different excitation positions with the associated distances x and phase functions Φ (f).

The data processing is carried out as in "A photoacoustic method for characterizing thin films", ( Dieter Schneider, Thomas Schwarz, Surface and Coatings Technology 91 (1997) 136-146 ).

The measured dispersion curves may have different degrees of curvature depending on the material parameters, as well as the attenuation which the coating material has with respect to the surface acoustic wave. Thus, the dispersion curve obtained can correspond to nearly a straight line or be curved several times.

In this study, the goal was to determine the density of the layer. The values for the Poisson number and the layer thickness had to be determined in another way due to the low curvature of the dispersion. The Poisson's number was set to 0.17 based on the nanoindentation measurements, the layer thickness was determined by reflectometric three-point measurements.

The measurements were carried out with an LAwave ^® , a development of the Fraunhofer IWS.

The measurements show that the reference not according to the invention has a mass density which is significantly above the maximum mass density of the layers according to the invention ( 10 ). This can be interpreted as an indication that the reference tends more in the direction of the density of amorphous SiO ₂ , due to the significantly higher oxygen content and to about 25% present quaternary crosslinking.

10 represents: mass density of the plasma polymer separation layers and the reference

Example 6 Determination of a Structural Model and the Degree of Crosslinking of the Layers of Example 1

From the results of the XPS, IRRAS (Example 7) and microelement analysis, a structural model of the chemical structure of the plasma polymer coating can be created. The results of the IRRAS measurements are only considered with regard to the functional groups found. There is no semiquantitative evaluation of the absorption spectra. It was estimated that approximately 5% of a functional group must be present to produce a detectable band in the absorption spectrum. Specifically, this means that it has been assumed that 5% of the oxygen atoms result in an OH group. A semiquantitative evaluation was omitted, as it was shown in preliminary experiments that the IRRAS spectrum is highly dependent on the layer thickness. Therefore, different values in a semiquantitative evaluation could be pure artifacts of this circumstance. In addition, the band assignment, especially in the fingerprint range (750-1500 1 / cm) is very difficult for plasma-polymerized samples. This type of sample contains a multitude of similar structural elements which give rise to very broad bands which can not be readily decomposed into their constituent parts.

With the help of the XPS data a peak fitting at the Si 2p peak was carried out in addition to the determination of the element concentrations. Since the p-orbital energetically in 2p _1/2 and 2p _3/2 splitting up the four possible oxidation states of silicon were fitted with doublets. To calibrate the energy scale, the aliphatic C1s peak was set at 285.0 eV. Here, no shift of the peak was considered by possible OH groups, but the peak as measured to 285 eV laid. The error caused by the disregarded shift of the peak maximum through the OH groups is very small, since the percentage of OH groups in the total peak area of the C1s is only a few percent.

The microelement analysis provides an indication of the ratio of CH ₃ to CH ₂ groups in the different plasma polymer coatings. From the set of results, a chemical structural model can be created.

In order to be able to create a structural model from the measured data of the chemical composition, the peak fitting is used. For this purpose, assumptions are made about the existing oxidation states of the silicon. Each oxidation state has a different binding energy. Thus, the measured Si 2p peak can be decomposed into its constituents.

An example of the peak fitting of the Si 2p peak is in 11 shown. The four oxidation states are expressed in different binding energies and are given in Table 5 with their binding energies and literature references.

11 shows: Fit example of Si-2p peak fitted with doublets

Benennung Si prim. Si sek. Si tern. Si quart. Bindungsenergie Si 2p 101,8 102,7 103,4 104,1 [eV] In 11 the trace and the different fitted peaks can be seen. The corrected binding energies in eV are shown on the X-axis and the intensity in counts per second (cps) on the Y-axis. The quaternary silicon peak was fitted with a zero area and therefore can not be seen in the figure. The positions of the oxidation states were set to allow a comparable evaluation of all spectra. Table 5: Oxidation states of silicon with bonding energies of Si 2p, R denotes radicals which do not represent oxygen atoms structure

designation Si prim. Si sec. Si tern. Si quart. Binding energy Si 2p 101.8 102.7 103.4 104.1 [EV]

The peak fitting results in the percentages of the various silicon oxide structural elements. A graphic representation of these proportions can be found for the coatings of Example 1 in 12 ,

12 represents: Shares of the different oxidation states of the silicon, determined from the peak fitting of the Si2p, plotted against the plasma power.

It can be seen here that the percentages of the oxidation states are constant for all powers used within the error range. The error range of the fiter results was estimated using a Monte Carlo algorithm implemented in the software. The constant percentages of the oxidation states indicate that the fundamental cross-linking does not change. This finding is consistent with the constant O: Si ratio, which was obtained from the overview spectra, cf. Example 3.

In summary, it can be stated that the chemical structure of the five different plasma polymer coatings according to the invention is very similar. It is only a slight increase in OH groups with increasing performance, as well as a decrease in the H: C ratio to recognize. This surprisingly means that the coatings chemically change to a much lesser extent with a power variation than initially thought due to the large change in the modulus of elasticity. With regard to changing other plasma parameters, the coupled power has a relatively small effect on the chemical composition. These findings should apply to all layers / solids according to the invention.

In the following, an estimate for the determination of the carbon-containing bridging groups as well as the overall degree of cross-linking shall be made. For this purpose, no new measurement results are used, but an empirical, mathematical estimation with respect to the results from XPS and microelementaranalysis is performed. In this case, a structural model is to be created on the percentage of different structural elements to be determined as well as their characteristic values, which best represents the measurement results.

First, structural elements as suggested by the precursor and IRRAS studies are selected. The atomic ratios (O: Si, C: Si and H: C) for these possible structural elements are calculated below. In addition, the proportions of the silicon oxidation states, according to the values determined during the peak fitting of the XPS results, are taken into account. Taking into account the XPS results and the microelemental analysis, the best possible approximation of the possible structural elements to the analysis results takes place. This procedure results in a well-defined range of values for each plasma polymer coating.

The possible structural elements discussed in this context are basically divided into two categories, the network-forming structural groups and those which terminate with respect to the hydrocarbon bridges. The discussed structural groups are shown in Table 6 and Table 7.

It should be noted that the structure groups shown for the sake of simplicity were each drawn with silicon in the primary oxidation state. Each of the illustrated structural groups contains several structural elements which differ with respect to the oxidation states of the silicon. As a result of this structure of the structural elements or structural groups, a linking of the individual structural elements via the Si-O network can take place later. The terminations shown actually belong to structure group A, but carry other functional groups, such as OH, SiH or ethylene groups. The structural groups classified as terminating with respect to the hydrocarbon bridges occur only in lower concentrations within the coating, they do not form the actual coating network.

The choice of structures was made for simplicity and IR results. All structures used can not be excluded via the IRRAS spectra. Some, such as the OH groups and the Si-H groups, even show a performance-related tendency in the IRRAS spectrum. Some other structural groups, such as methoxy groups (Si-O-CH ₃ ), which are generally a have pronounced bands in the IR, could be excluded via the IRRAS spectra and therefore remain unconsidered in the structural model.

In Table 7, four network-forming structure groups are exemplified as primary structures with respect to the conventional Si-O network. Structure group A is pure siloxanes. The structural group B, however, has a hydrocarbon bridge. This is a methylene bridge. Structure group C includes an oxymethylene bridge. Finally, structural group D represents an ethylene bridge.

In the calculation, all possible combinations of the oxidation states of the silicon were considered. A more detailed description of this procedure including the calculated atomic ratios of the different structural elements using the example of structure group D can be found in Table 8.

primär - primär 2,67 3,0 0,50

primär - sekundär 2,60 2,5 0,75

sekundär - sekundär 2,50 2,0 1,00

sekundär - ternär 2,33 1,5 1,25

primär - ternär 2,50 2,0 1,00

ternär - ternär 2,00 1,0 1,50 Table 8: Structural group D with all considered structural elements and the respective atomic ratios Si-O network H: C C: Si O: Si

primary - primary 2.67 3.0 0.50

primary sekundary 2.60 2.5 0.75

secondary - secondary 2.50 2.0 1.00

secondary - ternary 2.33 1.5 1.25

primary - ternary 2.50 2.0 1.00

ternary - ternary 2.00 1.0 1.50

3 3 0,5

2 1 1,5

3 2 1

2,33 1,5 1,25

3 1 1,5

2,5 2 1

3 0 2

2,5 2 1

2 0,5 1,5

2,6 2,5 0,75

2,5 1 1,25

2,67 3 0,5

2,67 1,5 1

3,5 2 0,75

2,67 1,5 1

4 1 1,5

2,75 2 0,75

3 3 0,5

2,8 2,5 0,5

3 2 1

2,5 1 1,5

3 1 1,5

2,67 1,5 1,25

2,5 2 1,5

2,75 2 1

2,67 3 1

2,75 2 1

2,75 4 0,5

2,8 2,5 0,75

4 1 1

2,83 3 0,5

3,5 2 0,5 Table 9: all considered structural elements as well as the respective atomic ratios H: C C: Si O: Si H: C C: Si O: Si

3 3 0.5

2 1 1.5

3 2 1

2.33 1.5 1.25

3 1 1.5

2.5 2 1

3 0 2

2.5 2 1

2 0.5 1.5

2.6 2.5 0.75

2.5 1 1.25

2.67 3 0.5

2.67 1.5 1

3.5 2 0.75

2.67 1.5 1

4 1 1.5

2.75 2 0.75

3 3 0.5

2.8 2.5 0.5

3 2 1

2.5 1 1.5

3 1 1.5

2.67 1.5 1.25

2.5 2 1.5

2.75 2 1

2.67 3 1

2.75 2 1

2.75 4 0.5

2.8 2.5 0.75

4 1 1

2.83 3 0.5

3.5 2 0.5

The overall overview of the atomic ratios of the respective structure, sorted according to structural groups, can be found in Table 9. As can be seen in Table 8, in the calculation of the structural components of the layers according to the invention, no structural elements are considered which have quaternary silicon. This is due to the results of Si2p peak matching from the XPS data. See Example 3, the plasma polymer layers do not have quaternary silicon.

The procedure for calculating the percentage of the different structural elements is described below. First, a table was created which contains all the structural elements with their respective parameters (H: C, C: Si and O: Si ratio). Furthermore, columns are needed for the calculation of the hypothetical atomic ratios. Now, a percentage frequency is assumed for each feature. From the proportion of a single structural element and the respective atomic ratio of the product is formed. It has been ensured that the total amount of the shares is 100%. Now the sum of the hypothetical atomic proportions can be formed over all structural elements and divided by one hundred. This results directly in the hypothetical atomic ratio of H: C, C: Si and O: Si as a function of the assumed percentage distribution of the structural elements. The hypothetical atomic ratios are adjusted via a variation of the assumed percentages of the values known from the XPS. At the same time, the hypothetical distribution of the oxidation states of the silicon must show a small deviation in the context of the error of the XPS results.

The hypothetical distribution of the oxidation states can be made by a classification into primary, secondary, ternary and quaternary structural elements. It should be noted that the structural groups B, C and D each consist of two Si atoms and therefore the proportions are to be weighted with the factors 0.5 or 2. The distribution thus found is normalized to a sum of 100% and compared with the results of Si2p peak fitting from the XPS data. If there is a good agreement with the distribution of the percentage of structural elements found, both with the measured atomic ratios and with the oxidation states, the composition thus found can be assumed to be the solution of this underdetermined system of equations.

By the boundary conditions, atomic conditions, oxidation states and total of the percentages, there are 8 equations. In contrast, there are 32 structural elements and thus 32 variables in the equation system. The system is thus clearly underdetermined, as a result of which there are several possible solutions. In order to verify the significance of the found result, a comparison with other measurement results was used. It is known from the IRRAS spectra that the coatings which are deposited at higher plasma power must have both more OH and Si-H groups. This tendency must be reflected in the structure distribution found. Furthermore, knowledge of the plasma process can be used to perform a consistency check of the found distributions. Again, only a comparison of all coatings is possible. This is the well-founded assumption that at higher plasma powers a stronger fragmentation occurs and thus the proportion of the structural group A, which is based on the precursor HMDSO decreasing, at least constant, across the different powers. Owing to the non-nominally determinable external restrictions which relate to the entire coating system and not independently to the respective coupled power (proportion of OH and Si-H groups, decrease of structure group A), the calculation should be carried out in a consistent order. Either from the low power, PDMS-like with a large amount of siloxane groups, to the high power or vice versa. It has proven to use the values of the previous performance as a starting point for the next higher performance.

The presented procedure scheme can be implemented in a least square optimization of the underdetermined system of equations. This makes it possible to carry out a more independent evaluation of the structure components by the user. Within this example, the implementation was implemented as a Python script. Python is a programming language that is well-suited for solving mathematical problems.

By adapting the distribution of the structural elements under the given boundary conditions, it is possible to find consistent results of the structural group distribution for all investigated plasma outputs. The results are in good agreement with the manually found solution. To verify the significance of the fit result, two different algorithms already implemented have been used. The algorithms do not differ in the basic minimization, but only in the parameter to which the minimization refers. Thus, one algorithm minimizes the absolute square of the vector resulting from the converted system of equations. The second algorithm, on the other hand, minimizes each individual vector component. In addition, the constancy of the structural group proportions was considered with varying total number of structural elements. The results are in 13 and 14 shown.

13 Fig. 2 shows comparison of two optimization algorithms; dashed line: minimization of | c | ² ; solid line: minimization of vector components c _i ..

14 represents: percentage of the structure groups as a function of the total number of structural elements (adaptation algorithm: minimization of the vector components of c). $$c=Ax-b$$

For optimization, the function in Equation 9 was used in each case. This is the established, transformed linear system of equations. A is the coefficient matrix which describes all atomic ratios as well as the contribution of the structural elements to the oxidation states. The vector b is the result vector which contains the nominal measured values from the XPS analysis. To solve the problem by means of a least squares algorithm, the vector c must be passed to the algorithm. This vector or its square of sums must be minimized.

The 13 shows the comparison of two optimization algorithms. On the one hand, this is an algorithm which individually minimizes the components of the vector c and, on the other hand, minimizes the magnitude square of the vector c. It can be stated here that given the total number of structural elements and a number of 100 adaptations carried out with arbitrary start vectors, both algorithms show a good agreement in the context of the standard deviation formed in the case of several of the selected plasma powers.

For both algorithms the same system of equations and the same boundary conditions were used. The system of equations consists of 10 equations, since the proportion of OH and (Si-H) groups per coating must be specified in order to arrive at a physically meaningful result. Furthermore, there are two constraints that further constrain the fit, on the one hand limiting the terminations to a maximum of 20% of the total set of features and on the other hand limiting the total sum of (C-O) bonds due to peak matching of the C1s in the XPS data.

Since the adjustments at the lowest plasma lead to slightly different results, the product A * x was considered after the optimization. It could be determined that the adaptation via the minimization of the vector components of c provides the required results much more accurately. Therefore, this algorithm was used below.

In order to check the dependence of the found structure group distribution on different total numbers of structural elements, the variation in 14 performed (compare there).

Since the variation in the number of structural elements shows a consistent result of the structural group distribution, a total of 1000 structural elements is used. At the same time, 100 adjustments are carried out, each with arbitrarily distributed start vectors. From these single results, the mean and standard deviation for each structure group can be determined. This adjustment is carried out for all coatings used. The results are in 15 and shown in Table 10.

15 represents: Comparison of the proportions of different structure groups for the plasma power used

In 15 it can be seen that the proportion of siloxane groups decreases with increasing plasma power. This can be correlated with the processes in the plasma. As performance increases, fragmentation increases. The structure group A ultimately reflects the precursor, HMDSO, and thus describes a structure-preserving plasma. Thus, with increasing fragmentation, the content of the structural group A must decrease.

Furthermore, the figure clearly shows that the ethylene bridges represent the dominant group at higher plasma powers.

The methylene bridges are formed only to a very small extent. The methoxy bridges are bounded by the already described boundary condition, the content of C-O bonds. The proportions of OH- and Si-H groups were fixed in each case on the basis of the IRRAS results and represent assumptions. The absolute proportion of these groups is not known. However, the exact value of (Si-H) and OH groups only represents an offset for the other structure groups. This means that the structure group dominating per plasma power will not change, but only all structural groups slightly changed proportions, if the setting has changed of (Si-H) - or OH content, respectively.

By means of the mathematical adaptation by means of the least squares method, the different crosslinking structures of the plasma polymer coatings can be estimated in compliance with the consistency criteria. The trends are summarized in Table 10. The specification of the standard deviation has been omitted here for the sake of clarity. This can be the 15 be removed.

Table 10: Summary of the structural groups for the investigated plasma polymer coatings; only mean values (without standard deviation) Proportions of different structure groups in at% 1000W [%] 1600 W [%] 2200 W [%] 2800 W [%] 3400 W [%] Structure group A (Si-O, methyl terminated) 58 41 24 6 0 (4 structural elements) Structure group B (Si-CH ₂ -Si network) 16 2 2 1 5 (6 structural elements) Structure group C (Si-O-CH ₂ -Si network) 6 6 8th 7 7 (6 structural elements) Structure group D (Si-CH ₂ -CH ₂ -Si network) 0 32 47 67 67 (6 structural elements) Sum of hydrocarbon-containing bridges 22 39 56 74 80 -OH termination 1 4 6 8th 10 (5 structural elements) -CH ₂ -CH ₃ termination 19 12 9 6 2 (3 structural elements) Si-H termination 0 4 5 6 8th (2 structural elements)

The mathematical adaptation, as well as the manual method, is used to estimate the percentage distribution of all 32 structural elements. These structural elements have hitherto been combined to form the structural groups. Now, the percentages of the individual structural elements themselves are to be considered.

From these percentages and an introduced index, the Si mesh number, a degree of Si crosslinking of the plasma polymer coatings can be determined. Si mesh number describes the number of network-forming bonds that a silicon atom has. In the case of SiO ₂ , in the case of an infinitely extended structure, the Si crosslinking degree is 100% and the Si network number is 4. Each Si atom has 4 oxygen bonds contributing to the network. However, the introduced Si network number is not related to a crosslinking structure, but takes into account all possible structure groups. Thus, each of the structure elements can be assigned such a Si network number. For example, a degree of crosslinking of 75% would be a structure in which each silicon atom has exactly three network-forming bonds, for example the polymethylsilsesquioxane already mentioned. A degree of crosslinking of 50% corresponds to linear chains with two crosslinking bonds per silicon atom. This can be, for example, PDMS.

16 represents: degrees of crosslinking of the plasma polymer coatings as a function of power; Degree of crosslinking of the Si-O-Si network determined by the Si network number (black); Overall level of crosslinking from distribution of the structural elements, determined by Si network number (dotted); Degree of crosslinking of the hydrocarbon network from the difference between the total degree of crosslinking and the Si-O-Si network (dashed line)

In 16 the total crosslinking degree of the various plasma polymer coatings calculated via the Si network number is plotted. In addition, the degree of crosslinking of the Si-O-Si network was also calculated via the Si mesh number. The proportions of different Si network numbers can be taken from the proportions of the oxidation states of the silicon. The difference between these two degrees of crosslinking results in the degree of crosslinking caused by the hydrocarbon network.

The results obtained are based purely on empirical, mathematical experiments and do not claim to be an exact structural model. In other words, the inventors are not bound by this theory. Nevertheless, tendencies can be noted. For example, that the carbon-containing crosslinking structures have to be ethylene groups rather than, for example, methylene groups. The specified distributions of the structural elements or structural groups were respectively determined via the adaptation by means of a least squares algorithm. The boundary conditions already described have been complied with.

As a counter example, an attempt was made to adapt to the measured data only with the structural group of the siloxanes (A) and the terminations. It can be shown that it is not possible to map the measured values only with these two structure groups, see Table 11. Table 11: Comparison of the measured values and the values determined by the mathematical adaptation, considered structure group only A and the terminations O: Si C: Si H: C Si primary Si secondary Si ternary Si quaternary reading 1.10 1.83 2.77 12.3 61.8 25.9 0.0 Adaptation 1.07 1.90 2.98 12.1 61.7 25.9 0.1

The fitting algorithm can not accurately map the measured values. This is especially evident in the H: C ratio. In the previously considered adaptation, taking into account all the structural groups mentioned, the algorithm can simulate the measured values very accurately. A difference only occurs in later decimal places.

Overall, this calculation of the structural groups can be a clear indication of an increasing proportion of hydrocarbon-containing crosslinks at higher plasma powers, as well as an overall increase in the overall degree of crosslinking. The increase in hydrocarbon crosslinks is consistent with the increasing overall degree of crosslinking with increasing performance and explains the increasing elastic modulus, as well as the slightly increasing mass density.

In summary, it can be stated that for the layers according to the invention (and not only for those from Example 1), a structural model is developed successively, which contains a constant Si-O network for all considered plasma powers and is supplemented by a second, performance-dependent hydrocarbon network. This novel second cross-linking mechanism occurs via hydrocarbons, primarily via ethylene bridges (Si-CH ₂ -CH ₂ -Si). This new mechanism is ultimately responsible for the observed, large changes in layer properties. As the hydrocarbon crosslinking increases, the overall crosslinking density increases. This in turn increases the modulus of elasticity. Due to the increased cross-linking density increases at the same time, the layers become more compact. The disperse fraction of surface energy increases slightly due to the greater interaction between surface and droplet. In addition, the more crosslinked plasma polymer coatings have less free methyl groups. This requires a very small increase in the polar component of the surface energy.

Measuring Example 7: IRRAS Measurement of the Layers from Example 1

The reflection absorption spectroscopy in the infrared spectral range is a frequently used spectroscopic method for chemical structure analysis. Using infrared radiation sources, molecular vibrations are excited. The energy at which a vibration can be excited depends on the molecule or the moiety, so a CH ₃ group has a different oscillation frequency than a CH ₂ group. Furthermore, the frequency depends on the type of excited vibration. One distinguishes between different types of vibration. These are either Stretching vibrations where there is a change in the bond lengths, or deformation vibrations in which the bond angles change. Both basic modes of vibration can be further characterized by their symmetry behavior. In addition to the symmetrical oscillations, in which the molecular symmetry is retained, there are still antisymmetric oscillations associated with the loss of at least one symmetry element, as well as degenerate vibrations. The absorption bands of these different vibrations combine to form the IRRAS spectrum [8].

For this study, a Bruker Corperation Vertex 80 / XSA with a mid-infrared radiation source, MIR (500-4000 cm ^ -1), and a nitrogen-cooled mercury-cadmium-telluride (MCT) detector were used. The device was operated in reflection mode. This means that the sample was irradiated at an angle of 45 ° and the reflected intensity was measured. The samples are aluminum with the plasma polymer layer to be examined. An uncoated aluminum sample serves as a reference and atmospheric compensation and baseline correction are performed. Care must be taken to ensure that the samples to be compared have the same layer thickness, since the ratios of some IR bands change with different layer thicknesses. Thus, normalization of the spectra with different layer thicknesses would lead to false conclusions, as the peak heights would no longer be comparable. Therefore, all samples evaluated by IRRAS were produced with a layer thickness of approx. 500nm. Thus, the comparability of the spectra is given. In order to produce a better statistic of the spectra, four samples were measured per plasma power. The individual spectra were added and then normalized to the maximum peak on the Si-O-Si peak. Thus it is possible to compare the band heights with each other.

In Example 3 it was found that only the carbon content decreases slightly with increasing power. This can be attributed to an increase of Si-H and -OH groups by means of the IRRAS measurements. These bands grow with increasing power in the absorption spectrum, see 17 ,

17 shows: IRRAS measurements of the five coatings, band assignments according to J. Lambert, S. Gronert, H. Shurvell and D. Lightner, Spectroscopy - Structure Elucidation in Organic Chemistry, Pearson, 2012 and Smith, "Analysis of Silicones," in Chemical Analysis Volume 41, John Wiley & Sons, 1974 ,

Furthermore, although the presence of Si-CH ₂ -CH ₂ -Si groups can not be detected directly, it is possible to detect the decrease in Si-CH ₃ groups. This can be observed in the IRRAS spectrum, see 18 ,

18 Figure 1 shows a fragment of the IRRAS spectrum, average spectra of the different plasma polymer coatings, band assignments according to L. Smith, Analysis of Silicones, in Chemical Analysis Volume 41, John Wiley & Sons, 1974.

The spectrum shows that the intensity of the Si (CH ₃ ) _x peaks decreases with higher plasma power. This indicates a decrease in CH ₃ groups with increasing plasma power. Thus, indirectly, the developed structural model of Example 6 can be confirmed. Also, a slight growth of the Si-O-Si peak caused by the Si (CH ₂ ) _1,2 -Si bond can be observed. When coated with 3400 W plasma power, the band clearly shifts to lower wavenumbers. Thus, an increased proportion of ethylene bridges is present. Here is thus a more concrete indication of the increasing hydrocarbon cross-linking via ethylene bridges.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102013219331 [0003, 0045]
• DE 10034737 [0005]
• DE 102006018491 A1 [0006]
• WO 2015044247 A1 [0007, 0015, 0061]

Cited non-patent literature

• Alexander, M.R., Short, R.D., Jones, F.R., Stollenwerk, M., Mathar, G., Zabold. The dissertation "Separation and Characterization of Fluorocarbon- and Siloxane-Based Plasma Polymer Layers"; B. Jacoby, Dissertation University Mainz, 2007 [0011]
• J. Gersten and F. Smith, The Chemistry of Materials, New York: John Wiley & Sons, Inc., 2001 [0079]
• M. Cardona and L. Ley, "Photoemission in Solids I - General Principles," in Topics in Applied Physics Vol. 26, Springer-Verlag, 1978 [0093]
• G.W. Farnell and E.L. Adler, "Elastic wave propagation in thin layers," in Physical Acoustics, Volume 9, 1972. [0118]
• D. Schneider, S. Frühauf, S. Schulz and T. Gessner, "The current limits of the laser-acoustic test method to characterize low-k films," Microelectronic Engineering 82 (2005) 393-398, 2005 [0121]
• Dieter Schneider, Thomas Schwarz, Surface and Coatings Technology 91 (1997) 136-146 [0123]
• J. Lambert, S. Gronert, H. Shurvell and D. Lightner, Spectroscopy - Structure Elucidation in Organic Chemistry, Pearson, 2012 [0178]
• Smith, "Analysis of Silicones," in Chemical Analysis Volume 41, John Wiley & Sons, 1974 [0178]

Claims (11)

Plasma polymer solid, in particular plasma polymer layer, wherein the lower limit of the modulus of elasticity of the solid is determined by the following function (1): $$\text{e}=\mathrm{2,172.85}-\mathrm{4,716.04}\cdot \text{x}+\mathrm{3,854.96}\cdot \text{x}2-\mathrm{1,405.04}\cdot \text{x}3+192.54\cdot \text{x}4$$

with x = C / O molar ratio determined by XPS E = modulus [GPa] for E = 1.5 to 20 GPa and x ≥ 1.3 and ≤ 2.0. Plasma polymer solid, in particular plasma polymer layer according to Claim 1 , wherein on the surface of the layer measured by XPS the maximum of the Si 2p peak at 102.5-102.8 eV. Plasma polymer solid, in particular plasma polymer layer according to Claim 1 or 2 in which the maximum polar fraction of the surface energy of the surface is determined by the following functions (2) or (2a): $$\mathrm{\sigma }\left(\text{p}\right)=.0947\cdot \text{e}+0.106$$

or $$\mathrm{\sigma }\left(\text{p}\right)=0.25$$

depending on which value of the functions (2) and (2a) is the larger, with σ (p) = polar fraction of the surface energy [mN / m] E = modulus of elasticity [GPa] for E = 1.25 to 20 GPa and / or the surface energy of the surface is determined with respect to its upper limit by the following function (3): $$\mathrm{\sigma }=0.4\cdot \text{e}+22.0$$

and the surface energy of the surface with respect to its lower limit is determined by the following function (4): $$\mathrm{\sigma }=0.17\cdot \text{e}+21.5$$

with σ = surface energy [mN / m] E = modulus of elasticity [GPa]. Plasma polymer solid, in particular plasma polymer layer according to one of the preceding claims, wherein the molar ratio H: C 2.0-3.0 is preferably 2.0-2.8, more preferably 2.2-2.7, measured by means of microelemental analysis. Plasma polymer solid, in particular plasma polymer layer according to one of the preceding claims, wherein the molar ratios on the surface of the layer measured by XPS C: O 1,3 - 2,5 and / or O: Si 0.9-1.5, preferably 0.9-1.3 and / or C: Si 1.5 - 2.5 be, A plasma polymer solid, in particular a plasma polymer layer according to one of the preceding claims, wherein the following applies to the molar proportions on the surface of the layer measured by means of XPS: O: 20 - 38 at%, preferably 23 - 33 at% Si: 22 - 28 at%, preferably 23 - 27 at% C: 40 - 65 at% in each case based on the total number of XPS-determinable atoms contained in the layer. Plasma polymer solid, in particular plasma polymer layer according to one of the preceding claims, wherein in the FTIR spectrum at about 2165 cm-1, the Si-H band is visible. Plasma polymer solid, in particular plasma polymer layer according to one of the preceding claims, wherein the total silicon cross-linking degree is between 50 and 85%, preferably between 55 and 80% and at the same time the silicon cross-linking degree via hydrocarbon bonds between 5 and 50%, preferably between 5 and 30%. is based on the total number of silicon atoms contained. Plasma polymer solid, in particular plasma polymer layer according to one of the preceding claims, wherein the density measured by the SAW method is between 0.8 and 1.6 g / cm ³ , preferably between 1.0 and 1.5 g / cm ³ . Use of a plasma polymer layer, as in one of Claims 1 to 9 defined for improving the corrosion protection and / or for improving the cleaning ability and / or as a release layer and / or as an anti-scratch layer. Process for producing a plasma polymer solid, in particular a plasma polymer coating according to one of the Claims 1 to 9 comprising the steps of a) providing a substrate, b) depositing a plasma polymer solid, in particular a plasma polymer layer on the substrate, wherein a plasma polymer solid and / or a plasma polymer layer according to one of Claims 1 to 9 arises.

Images