The explicit setting of the last two bytes to 0 is redundant since byte arrays in C# are zero-initialized by default. While this doesn't affect correctness, it can be simplified:

internal static byte[] StringToBytes(string value)
{
 int size = value.Length * 2 + 2;
 byte[] byteBuffer = new byte[size];
 Encoding.Unicode.GetBytes(value.ToCharArray(), 0, value.Length, byteBuffer, 0);
 // Last 2 bytes are already 0 by default
 return byteBuffer;
}

Alternatively, if you want to make the null terminator explicit for clarity, consider adding a comment explaining why.

The StringToBytes method lacks test coverage for this critical bug fix. Consider adding unit tests to verify:

1. Correct 2-byte null termination for Unicode strings
2. Empty string handling
3. Strings with special characters
4. Round-trip consistency with StringFromBytes

This is especially important since this bug affected multiple user-facing APIs (MessageQueue.Label, Message.Label, etc.) and caused data persistence issues with MSMQ.