# Review: a single-site receptor with a narrow binding window and flat calorimetric data at its centre ## 1. Precise restatement **Claim (existence).** For every integer m ≥ 0 and every width w > 0, there is an equilibrium statistical-mechanical model with the following properties: - **Model.** A receptor that binds at most one ligand, with a finite set of unbound and bound conformations. Each conformation has a temperature-independent enthalpy and a temperature-independent entropy weight. - **Binding constant.** The apparent binding constant is K(T) = Z_bound(T)/Z_unbound(T), up to a temperature-independent affinity factor. - **Peak.** K(T) has a maximum at some T₀, and K is negligible outside a window of width below w around T₀. "Width" here is in β = 1/RT, or equivalently in T near T₀. - **Flat enthalpy.** The van 't Hoff binding enthalpy ΔH = −∂ln K/∂β vanishes at T₀, and so do its first m temperature derivatives (ΔH, ΔC_p, …). **Hidden assumptions:** - Ideal dilute conditions. - The ligand and standard-state contributions are absorbed into a T-independent multiplier. - Conformational entropies are finely tuned: binomial weights times exp(β₀kE). - Large internal enthalpy spacings E, with a total span NE. - "Noncooperative" means a single ligand site, not the absence of conformational coupling. ## 2. Checking the derivation **Partition sums.** With weight C(N,k)e^{β₀kE}·e^{−βkE} = C(N,k)y^k, where y = e^{−x} and x = E(β−β₀): - Even k give Z_u = [(1+y)^N + (1−y)^N]/2. - Odd k give Z_b = A·[(1+y)^N − (1−y)^N]/2. These are standard binomial parity sums, so this step is correct. **Ratio.** Dividing numerator and denominator by (1+y)^N gives K = A(1−z^N)/(1+z^N), with z = (1−y)/(1+y) = (1−e^{−x})/(1+e^{−x}) = tanh(x/2). At β₀ we have z = 0, so K(T₀) = A. The formula is correct. **Maximum at T₀.** Since |z| < 1 and N is even, z^N ≥ 0 with equality only at z = 0. So K ≤ A, with the maximum uniquely at T₀. Correct. **Expansion near T₀.** - ln K − ln A = −2 artanh(z^N) = −2z^N − (2/3)z^{3N} − … - Since z = x/2 − x³/24 + …, ln K − ln A = −2(x/2)^N + O(x^{N+2}). - Hence ΔH = −d ln K/dβ = N·E·(x/2)^{N−1} + O(x^{N+1}). ΔH therefore has a zero of exact order N−1 at β₀. ΔH and its first N−2 β-derivatives vanish, and the (N−1)th does not. Because β(T) is smooth with β′(T₀) ≠ 0, vanishing of the first j derivatives in β is equivalent to vanishing in T (Faà di Bruno). So given m, choosing an even N ≥ m+2 works. **Sign check.** For β > β₀ (below T₀), ΔH > 0. That is correct: K rises with T up to the maximum. **Narrowness.** For fixed N and fixed β ≠ β₀, letting E → ∞ sends z → ±1 and z^N → 1, so K → 0 on both sides. - The half-maximum condition is z^N = 1/3, which gives a full width in β of Δβ = 4·artanh(3^{−1/N})/E. - This scales as 1/E, so the window can be made arbitrarily narrow. - In temperature, ΔT ≈ RT₀²Δβ. **Positivity and physicality.** All weights C(N,k)e^{β₀kE} are positive, so these are legitimate Boltzmann ensembles. ## 3. Attempts to break it, and worked examples **N = 2.** - Z_u = 1 + y², Z_b = 2Ay, so K/A = 2y/(1+y²) = sech(E(β−β₀)). - ΔH = E·tanh(E(β−β₀)), which is zero at T₀, with a full width of ≈ 2.63/E in β. - At T₀ = 300 K, RT₀² ≈ 7.5×10⁵ J·K/mol. So E = 100 kJ/mol gives ΔT ≈ 20 K, and E = 1000 kJ/mol gives ΔT ≈ 2 K. **N = 4.** - Z_u = 1 + 6y² + y⁴, Z_b = 4A(y + y³). - At y = 1 the ratio is 8/8 = 1. - K/A = (1−z⁴)/(1+z⁴), so ΔH, ΔC_p and dΔC_p/dT all vanish at T₀. **Edge cases and limits.** - **Odd N.** If N is odd, the maximum is lost (z^N changes sign). The construction correctly requires even N. - **Higher-order derivatives.** The flatness is only up to order N−2. The (N−1)th derivative of ΔH scales like E^N, so the narrowness is encoded in higher-order calorimetric data. "Locally null" information hides it only up to the prescribed order. This matches the claim's "prescribed finite number" but should be kept in mind. - **Hidden enthalpy scale.** Narrowing requires large E, and the conformations span enthalpies up to NE. Away from T₀, |ΔH| reaches about N·E. The selectivity comes from huge internal enthalpy differences with matched entropies. - **Affinity factor.** The ligand/standard-state factor must be T-independent, which is an idealization. A T-dependent factor would shift ΔH by a smooth term and destroy exact cancellation unless re-tuned. No counterexample to the stated existence claim exists. The construction is a complete, rigorous proof of it. ## 4. Novelty The basic facts are textbook thermodynamics: - ΔH = 0 at an affinity maximum (van 't Hoff). - Higher derivatives (ΔC_p, etc.) need not vanish there. - Local Taylor data does not determine global behaviour. The specific binomial-parity construction giving K ∝ (1−z^N)/(1+z^N) is neat, and I do not know a specific prior source for it. It is, however, an elementary realization of a known general point. I am not confident it is published, so I rate it "new as far as I know" in its specific form. ## 5. Fairness of the one-line summary The claim is phrased as an existence statement ("can have") and is accurate in that sense. It omits several things: - **Fine-tuning.** The entropies must be finely tuned. - **Idealization.** The ligand/standard-state factor is assumed temperature-independent. - **Large enthalpy span.** Narrowness requires a large internal enthalpy span, which grows with both narrowness and N. - **Sense of "noncooperative".** It means single-site only; the response arises from a strongly coupled conformational ensemble. - **Derivative order.** The vanishing derivatives are finite in number by necessity. Analytic K cannot have all derivatives vanish without being constant. The supporting text itself flags most of these points. The summary is fair given the "can". {"verdict": "correct with caveats", "novelty": "new as far as I know", "fair_summary": "yes", "confidence": 4, "one_line": "The binomial-parity construction rigorously gives K/K₀=(1−z^N)/(1+z^N) with ln K−ln K₀ ≈ −2(E(β−β₀)/2)^N and width ∝1/E, so the existence claim holds, though only for a finely tuned idealized ensemble with large hidden enthalpy spans."}