Many people have tried that, and other materials too. The topic of TT belt materials has been discussed here quite often.
While results will vary from rig to rig, mostly depending on the qualities of the motor, materials like the one you suggested do tend to perform better than rubber. To understand why you need to understand what makes this application so challenging.
The belt's role (ideally) is to provide linear transmission of rotational speed from the motor to the platter in the face of a continuously varying load. Stylus drag is always trying to slow the platter, but at unpredictable rates due to the unpredictability of groove modulations. It's the belt's job to link the platter to the motor's more stable rotational speed and prevent variable decelerations and reaccelerations, which slur the music in time.
To provide the most effective linkage a belt material requires two material attributes:
1. Lack of elasticity (which is why rubber is such a poor choice of materials for this purpose). Elasticity anywhere in a power transmission system necessarily produces deceleration when loads increase, followed by acceleration when the load is reduced. Musically, this means dulled or slowed leading edges, reduced peak amplitudes and slurring of trailing edge harmonics.
2. Sufficient friction to prevent slippage between belt and motor capstan (or platter) when transients try to slow the platter. Slippage in a power transmission system causes rapid, though short-lived decelerations when the load presented to a material interface exceeds the ability of the interface to maintain stable contact. Musically this means slowed transients and reduced peak amplitudes, followed by a return to normal speed when the load drops back enough for the two materials to regain good contact.
Dental floss, fishing line and similar materials address #1 so much better than elastic materials that they will normally provide real sonic improvements. However, due to their small contact surface they can fail to meet requirement #2 in some cases. The larger the diameter of your motor capstan and the less slippery its surface, the less likely slippage becomes, so final results will be TT-dependent, as I said above.
When switching to more linear belt materials from elastic ones, you learn that motor distancing and levelling become progressively more critical. I suspect this is the reason mainstream TT manufacturers settle for rubber(y) belts. It makes motor placement both easy and non-critical, since any inaccuracies are accommodated by the stretchiness of the belt. This ease of use makes for happier customers, at least until they hear a rig with a more linear drive system.
Owners of Galibier, Redpoint and Teres tables have experimented with dozens of belt materials (as a group). The current belt material of choice for these tables (which use very low noise, low cogging motors) is a form of 1/2" wide mylar tape. Mylar is inelastic, so it meets requirement #1 exceptionally well. Requirement #2 is trickier. Mylar is typically quite smooth, as smooth as high quality glass, so avoiding slippage is the biggest challenge. This is met partly by choosing the optimal thickness of mylar and partly by choosing a specialized mylar product with a slightly rough surface on the working side. The results are audibly superior to any table using a rubber belt.
I encourage you to experiment with dental floss or fishing line or - if your TT and motor can accommodate it - 1/4" or 1/2" wide mylar. You'll certainly hear differences, and you'll educate your ears and improve your understanding of how a TT works in the critical area of speed control.
While results will vary from rig to rig, mostly depending on the qualities of the motor, materials like the one you suggested do tend to perform better than rubber. To understand why you need to understand what makes this application so challenging.
The belt's role (ideally) is to provide linear transmission of rotational speed from the motor to the platter in the face of a continuously varying load. Stylus drag is always trying to slow the platter, but at unpredictable rates due to the unpredictability of groove modulations. It's the belt's job to link the platter to the motor's more stable rotational speed and prevent variable decelerations and reaccelerations, which slur the music in time.
To provide the most effective linkage a belt material requires two material attributes:
1. Lack of elasticity (which is why rubber is such a poor choice of materials for this purpose). Elasticity anywhere in a power transmission system necessarily produces deceleration when loads increase, followed by acceleration when the load is reduced. Musically, this means dulled or slowed leading edges, reduced peak amplitudes and slurring of trailing edge harmonics.
2. Sufficient friction to prevent slippage between belt and motor capstan (or platter) when transients try to slow the platter. Slippage in a power transmission system causes rapid, though short-lived decelerations when the load presented to a material interface exceeds the ability of the interface to maintain stable contact. Musically this means slowed transients and reduced peak amplitudes, followed by a return to normal speed when the load drops back enough for the two materials to regain good contact.
Dental floss, fishing line and similar materials address #1 so much better than elastic materials that they will normally provide real sonic improvements. However, due to their small contact surface they can fail to meet requirement #2 in some cases. The larger the diameter of your motor capstan and the less slippery its surface, the less likely slippage becomes, so final results will be TT-dependent, as I said above.
When switching to more linear belt materials from elastic ones, you learn that motor distancing and levelling become progressively more critical. I suspect this is the reason mainstream TT manufacturers settle for rubber(y) belts. It makes motor placement both easy and non-critical, since any inaccuracies are accommodated by the stretchiness of the belt. This ease of use makes for happier customers, at least until they hear a rig with a more linear drive system.
Owners of Galibier, Redpoint and Teres tables have experimented with dozens of belt materials (as a group). The current belt material of choice for these tables (which use very low noise, low cogging motors) is a form of 1/2" wide mylar tape. Mylar is inelastic, so it meets requirement #1 exceptionally well. Requirement #2 is trickier. Mylar is typically quite smooth, as smooth as high quality glass, so avoiding slippage is the biggest challenge. This is met partly by choosing the optimal thickness of mylar and partly by choosing a specialized mylar product with a slightly rough surface on the working side. The results are audibly superior to any table using a rubber belt.
I encourage you to experiment with dental floss or fishing line or - if your TT and motor can accommodate it - 1/4" or 1/2" wide mylar. You'll certainly hear differences, and you'll educate your ears and improve your understanding of how a TT works in the critical area of speed control.