WO2012001236A1

WO2012001236A1 - Process and apparatus for glass manufacture

Info

Publication number: WO2012001236A1
Application number: PCT/FI2011/050612
Authority: WO
Inventors: Paavo Timonen; Markku Rajala
Original assignee: Beneq Oy
Priority date: 2010-06-30
Filing date: 2011-06-27
Publication date: 2012-01-05
Also published as: FI20105753A; FI20105753A0

Abstract

Process and apparatus for heating the glass ribbon in the A0-section of the float line. The glass ribbon is coated in at least two coating phases. At least one of the coating sections uses a droplet-based precursor delivery. The glass ribbon is heated between the coating phases, the heating comprising for example heating by radiation.

Description

Process and apparatus for glass manufacture

Field of invention

The invention relates to a process and apparatus for heating glass during the glass coating process when manufacturing, in particular, coated glass within a float glass process. The coating is formed in the so called AO-section of a float line. The coating process comprises a coating phase using a precursor which is an aerosol comprising liquid droplets. The coating process further comprises another coating phase. Heating of the glass by radiation is carried out between the coating phases. A coating apparatus can be equipped to perform such a process.

Description of the state of the art

The float glass or tin bath process was developed by Pilkington in the late 1950's, and today the float process is the most used technique in flat glass manufacturing. In the float process, molten glass is poured into a tin bath where it floats on molten tin, forming a glass ribbon. The thickness of the float glass is adjusted typically from 25 mm down to 2 mm by pulling it off the bath at lower or higher speeds.

The glass ribbon leaves the tin bath at a temperature of about 600°C, and continues through the AO-section to the beginning of the lehr. The purpose of the lehr is to cool the glass ribbon to room temperature in annealed state and thus prevent formation of residual stresses in the glass.

In in-line coating process the coating apparatus or coating module is typically positioned in the latter part of the tin bath or in the AO-section. Basically, there are two reasons that motivate heat transfer control during a coating process: the thermal stresses depend on the thermal history of the glass and the coating process itself is temperature dependent. The deposition of coating material on a glass surface is limited either by diffusion rate or reaction kinetics. In order to obtain a satisfactory deposition rate, i.e., assure high reaction speed, the glass surface temperature must be sufficiently high. Also, the uniformity of the surface temperature field is essential for obtaining uniform coating thickness. On the other hand, the cooling rate at elevated temperatures governs the formation of thermal residual stresses in the glass. Further, asymmetric cooling results in asymmetric temperature profiles which, in turn, may give rise to asymmetric stress profiles. Eventually, this may lead to bending or breakage of glass.

Due to the factors mentioned above, it is of importance to be able to control and adjust the temperature and heat fluxes in the coating process. Summary of the invention

The invention solves the problems of the prior art by a process and apparatus for heating the glass ribbon in the AO-section of the float line. The glass ribbon is coated in at least two coating phases and at least one of the coating sections uses a droplet-based precursor delivery such as the aerosol-assisted chemical vapor deposition ("AACVD") process or in particular the nAERO process as discussed in more detail below. The glass ribbon is heated between the coating phases, for example by radiation. The surface temperature of the heater, heating the glass ribbon by radiation, is preferably less than 900°C. The process and apparatus may also or alternatively include heating by convection.

The glass ribbon in the float process is practically continuous. The glass ribbon speed is typically 8 - 15 m/min. The glass ribbon is coated in AO-section by the nAERO®-process, described in detail e.g. in a patent application

FI20071003, Beneq Oy, June 21, 2009. In the nAERO process, at least part of the precursors is fed into the deposition chamber as fine droplets. The local heat transfer coefficient under an impinging jet is proportional to the momentum of the jet. Equal momentums of jets produce equal heat transfer. As the glass ribbon cools during the coating phase, there is a need for additional heating of the glass ribbon in the AO-section, especially when two or more coating phases are carried out,

The invention solves the problem of the prior part by heating the glass ribbon especially by radiation in the AO-section. Brief description of the drawings

In the following, the invention will be described in more detail with reference to the appended principle drawing, in which

Fig. 1 shows the principle of the coating process;

Fig. 2 shows the test set-up for testing the glass ribbon cooling; Fig. 3 shows the measured heat transfer coefficient in the nAERO

coating;

Fig. 4 shows the glass ribbon surface cooling due to the nAERO coating;

Fig. 5 shows the temperature variation across the glass thickness;

Fig. 6 shows the stress variation across the glass thickness; and Fig. 7 shows the glass temperature variation across the glass thickness after radiative heating.

For the sake of clarity, the figures show only the details necessary for understanding the invention. The details which are not necessary for understanding the invention and which are obvious for a person skilled art have been omitted from the figures in order to emphasize the

characteristics of the invention.

Detailed description of preferred embodiments

Figure 1 shows the principle drawing of the nAERO-coating process. The nAERO coating unit 1 is placed on a moving glass ribbon 2 in the AO-section of a float line. Coating 3 is deposited on the glass ribbon 2 in the coating unit 1. An atomization device 4 produces small droplets 5 in the atomization chamber 6. The droplets 5 beneficially vaporize before hitting the glass ribbon 2. The vaporized precursors then form a coating 3 on the glass ribbon 2 in the deposition chamber 7. Non-reacted precursors and other exhaust gases are then exhausted by the exhaust channel 8.

It is obvious for a person skilled in the art that multiple nAERO deposition chambers, each constituting one coating step, may be connected into series to produce multiple coatings or one or several coating layers consisting of several sublayers created each by a coating step. It is also obvious that by the same token, the nAERO-coating process may be combined in series with any number of other pyrolytic coating processes such as Chemical Vapor

Deposition (CVD) process or spray-coating process in some order best suited for the coating functionality to be achieved.

By the same token, heating can be provided between one or several coating steps in the coating process. For example, in case of four coating steps CI, C2, C3 and C4, two heating steps can be provided between C2/C3, and C3/C4, or alternatively three heating steps can be provided between coating steps C1/C2, C2/C3 and C3/C4, again depending on the coating functionality to be achieved. An experiment was made to reveal glass cooling during the nAERO deposition process. The test set-up is shown in Figure 2. The test setup consisted of a heated plate 25, copper plate 24 for temperature homogenization, glass plate 2, soft insulation material 23 and thermocouples 21. Half of the thermocouples 21 were placed on top of the glass plate 2 and positioned so that they are directly on top of the other half of the thermocouples 21 at the bottom of the glass plate.

To be able to calculate the convective heat transfer coefficient, the heat flux from the plate has to be known. This can be calculated by measuring the temperature difference ΔΤ across the glass plate. Heat flux can then be calculated from q = kAT/s, where k is the thermal conductivity of glass, and s is the glass plate thickness. The temperature difference across the glass plate 2 was measured by inserting 0.5 mm thick K-type thermocouples

symmetrically on both sides of the glass plate. As the thermal conductivity of glass is relatively low, the conduction in the horizontal direction should not pose a serious error source to measurements. The soft insulating felt 23 is soft enough so that the thermocouple wire will not cause formation of air gaps between the glass plate and the copper plate. This way one can be confident that the bottom surface thermocouple 21 actually measures the temperature of the glass 2 bottom surface, instead of that of the glass-copper interface. On the top side of the glass 2, the thermocouples 21 were covered with aluminum tape so that the jet would not impact them directly. It is noted, however, that optimally, the thermocouples 21 should be located inside the glass. If the material flow was in the laminar regime, inserting the thermocouples 21 on top of the glass plate might cause the flow or the boundary layer to become turbulent. This would have an effect to the heat transfer, and thus cause the measurement situation to differ from the actual case. However, the

experiment revealed that the flow is turbulent also without the measurement devices. Therefore, the disturbing effect of the thermocouples is negligible. The coating unit was placed directly on top of the plate assembly, and nitrogen gas was impinged via the atomizer 4 on the glass 2 with typical process flow parameters. No liquid was used since the use of low

temperatures in the test would result in droplets impacting the glass surface, which would have an effect on the heat transfer coefficient. In the actual coating process, the droplets evaporate before reaching the glass surface.

The results from the measurements are shown in Figure 3A. At a long lateral distance from the impingement point, the heat transfer coefficient tends to stabilize to a constant value, being h(x -> oo) = 42 W/m²K. That value can be used for the flow between two parallel plates, situation located further downstream of the impingement point. A function fitted to the measurement results enables us to derive a continuous function to describe the heat transfer coefficient as a function of location. The heat transfer coefficient as a function of the x-axis position in the deposition chamber is h(x) = -0.03x²+3.35x+94.09 for 0≤x<100 and h(x)=1767253/x²- 12398.8/X+64.80 for 100<x≤250., where x is given in millimeters. The curve is shown in Figure 3B.

The heat flux from the glass can then be readily calculated when the average gas temperature and glass surface temperature are known along the channel. The glass top surface temperature was calculated using a finite-difference method. The glass speed was 3 m/min, and the time step used in calculations was 0.01 s. The average gas temperature above the glass was assumed to be a constant 160°C. The bottom surface was assumed to be perfectly insulated, and radiation was not accounted for. The resulting top surface temperature variation as a function of position is shown in Figure 4. The top surface cools rapidly under the impingement region, and as the heat transfer coefficient decreases further downstream, conduction from the inner layers of the glass rises the top surface temperature.

Glass ribbon cools by convection and radiation in the deposition chamber. Since glass is not a good heat conductor, steep temperature gradients may form. This causes thermal stresses within the glass.

Glass is a viscoelastic material, and thus stresses in it have a tendency to relax with time. The time that it takes a stress to relax is strongly dependent on the temperature. At low temperatures glass behaves like a solid material, since the relaxation time is extremely long (in the order of decades). However, when temperature increases, the relaxation times become shorter and shorter. This feature of temperature-dependent stress behavior creates the

fundamental basis to quenching of glass.

Iterative methods can be used to calculate the stresses inside the glass plate. Figures 5 and 6 show the temperature and stress profiles in two-second intervals obtained for a 4 mm thick glass from a simulated coating process. The glass temperature was initially 630°C, and the heat transfer coefficients were 150 W/m²K for the top surface and 25 W/m K for the bottom surface. Corresponding ambient temperatures were 250°C for the both sides. The process lasted for 6 seconds.

Initially a tensile stress is formed on both surfaces. As cooling continues the top surface stress turns into compressive stress. Due to the relatively high temperature, the stresses stay at low levels as they relax in a short time. It is worth noting that the residual stresses in the glass are not determined at this point. The stresses above only show the momentary stress profiles in the glass. The rest of the cooling process has an effect on the formation of the final residual stresses in the glass.

In order to obtain a sufficient growth rate an additional heating element may be required between the nAERO processes or between some other coating process, like Chemical Vapor Deposition (CVD) and nAERO process. The requirement is challenging since the glass temperature is well over 500°C and the available space for heating is limited, due to existing process

requirements.

Basically, there are two distinct approaches: heating by convection and heating by means of radiation. Convective heating can be engineered by impinging hot gases on the glass surface. As described earlier, the heat flux depends on the jet momentum and the temperature difference between the jet and the glass surface. To improve the energy efficiency, the gases should be re-circulated in the heating system. The drawback of convective heating is the fact, that in a limited space it can be challenging to prevent it from causing any disturbances to the gas flows in the coating processes. Especially the CVD process is very sensitive to external gas flows as its flows are typically in the laminar flow regime. A radiator, on the other hand, does not cause any additional flows; at least its effect is negligible compared to the flows due to natural convection from the glass plate. Also, the glass absorbtance is high at wavelengths above

2 500 nm, resulting in that a large fraction of incident radiation at long wavelengths is absorbed on the surface or the layers beneath the surface. Glass absorbtance decreases rapidly at wavelengths shorter than 2500 nm. The temperature of a black body that has the peak in emissive power at wavelength 2 500 nm is about 886 °C. The peak emissive power moves to lower black body temperatures at longer wavelengths. The heat flux from the radiator to glass is highly dependent on the temperature difference. Therefore, an optimal radiator temperature depends on the glass absorbtance properties, the desired heat flux and the technical limitations of elevated temperatures. To demonstrate the capability of a radiator to heat the glass, we assume a constant initial glass temperature of 550°C and a heating time of 6 s. For example, by letting the emissivity of the radiator and the absorbtance of the glass be 0.9, the initial heat flux to the glass from a radiator held at 850°C is roughly 50 kW/m². Increasing the radiator temperature results in larger fraction of its emissive power to be in shorter wavelengths. As mentioned above, the glass absorption depends on incident radiation wavelength. Hence, in order to localize the heating effect only to the glass surface it is beneficial to limit radiator temperatures below 900 °C to avoid radiation penetration deeper inside the glass or through it. Process-wise it is beneficial to heat up only the surface of the glass which is to be coated. Higher surface temperatures enable higher growth rate since the deposition efficiency is related to substrate temperature. Moreover, at temperatures above 600 °C glass becomes increasingly softer, eventually increasing the risk of bending between the rollers or roller marks. The gas flows related to the coating process remove heat only from the glass surface that is being coated in the process. As temperature differences within the glass start to decrease after the surface temperature drops in the coating process, conduction from higher

temperatures at inner layers of the glass tend to re-heat the surface. However, temperature differences at typical coating temperatures are crucial in stress formation inside the glass. If the coated surface temperature was higher before the process, smaller temperature differences form inside the glass in the cooling effect during the process. In addition to enhanced deposition efficiency, small temperature differences also decrease the stress levels and hence enable less complicated annealing when the glass is cooled to room temperature in later steps in the glass coating line.

Figure 7 shows the temperature profile of the glass after the heating. The bottom surface is assumed to be insulated, and the only heat transfer method to the top surface is incident radiation.

Thus, additional heating can be accomplished with a radiator. In a preferred embodiment, the radiator can be constructed of pipes heated by electricity or gas. The radiation can be directed towards the glass by inserting a reflector on the topside of the pipes. Convection could be also used, but it incorporates a risk of disturbing the sensitive gas flows in the coating units. Furthermore, an impinging jet apparatus may turn out to be a more expensive solution due to the larger number of components required for the system. While a radiator requires only the heating element and a power source, the gas heating system requires a blower, piping, a pressure manifold, and a re-circulation system. It is possible to produce various embodiments of the invention in accordance with the spirit of the invention. Therefore, the above-presented embodiments must not be interpreted as restrictive to the invention, but the embodiments of the invention can be freely varied within the scope of the inventive features presented in the claims.

Claims

aims

1. Process for coating glass in a glass manufacturing float line, the glass ribbon being coated in at least two coating phases, ch a ra cte ri zed in that a) the glass ribbon is heated between at least two of the coating phases; b) heating comprises heating by radiation; and

c) the surface temperature of the heater is less than 900°C.

2. Process according to claim 1, ch a ra cterized in that the heating is

performed in the AO-section of the float line.

3. The process according to claim 1 or 2, ch a ra cte ri zed in that at least in one of the coating phases precursors are delivered at least in part by means of droplets.

4. The process according to any of the claims 1-3, ch a ra cte rized in that the heating comprises convective heating.

5. Apparatus for manufacturing glass in a glass manufacturing float line, the

apparatus comprising at least two coating means for coating the glass ribbon, ch a ra cte rized in that:

a) the apparatus comprises heating means for heating the glass ribbon between at least two coating means;

b) heating means comprise radiating heating means; and

c) radiating heating means comprise control system for limiting the surface temperature of the radiating heating means below 900°C.

6. Apparatus according to claim 5, ch a ra cterized in that the heating

means are provided in the AO-section of the float line.

7. Apparatus according to claim 5 or 6, ch a ra cteri zed in that at least one of the coating means delivers precursors at least in part by means of droplets.

8. Apparatus according to any of the claims 5-7, ch a ra cte rized in that the heating means comprise convection heating means.